Applied Evidence

Medical errors in all settings contributed to as many as 250,000 deaths per year in the United States between 2000 and 2008, according to a 2016 study.¹ Diagnostic error, in particular, remains a leading cause of morbidity and mortality in the United States and worldwide. In 2017, 12 million patients (roughly 5% of all US adults) who sought outpatient care experienced missed, delayed, or incorrect diagnosis at least once.²

In his classic work, How Doctors Think, Jerome Groopman, MD, explored the diagnostic process with a focus on the role of cognitive bias in clinical decision-making. Groopman examined how physicians can become sidetracked in their thinking and “blinded” to potential alternative diagnoses.³ Medical error is not necessarily because of a deficiency in medical knowledge; rather, physicians become susceptible to medical error when defective and faulty reasoning distort their diagnostic ability.⁴

Cognitive bias in the diagnostic process has been extensively studied, and a full review is beyond the scope of this article.⁵ However, here we will examine pathways leading to diagnostic errors in the primary care setting, specifically the role of cognitive bias in the work-up of polymyalgia rheumatica (PMR), ovarian cancer (OC), Lewy body dementia (LBD), and fibromyalgia (FM). As these 4 disease states are seen with low-to-moderate frequency in primary care, cognitive bias can complicate accurate diagnosis. But first, a word about how to understand clinical reasoning.

There are 2 types of reasoning (and 1 is more prone to error)

Physician clinical reasoning can be divided into 2 different cognitive approaches.

Type 1 reasoning employs intuition and heuristics; this type is automatic, reflexive, and quick.⁵ While the use of mental shortcuts in type 1 increases the speed with which decisions are made, it also makes this form of reasoning more prone to error.

Type 2 reasoning requires conscious effort. It is goal directed and rigorous and therefore slower than type 1 reasoning. Extrapolated to the clinical context, clinicians transition from type 2 to type 1 reasoning as they gain experience and training throughout their careers and develop their own conscious and subconscious heuristics. Deviations from accurate decision-making occur in a systematic manner due to cognitive biases and result in medical error.⁶table 1⁷ lists common types of cognitive bias.

An important question to ask. Physicians tend to fall into a pattern of quick, type 1 reasoning. However, it’s important to strive to maintain a broad differential diagnosis and avoid premature closure of the diagnostic process. It’s critical that we consider alternative diagnoses (ie, consciously move from type 1 to type 2 thinking) and continue to ask ourselves, “What else?” while working through differential diagnoses. This can be a powerful debiasing technique.

Continue to: The discussion...

The discussion of the following 4 disease states demonstrates how cognitive bias can lead to diagnostic error.

Case 1

An 82-year-old woman with a history of hypertension; wide-angle glaucoma; stage 2 chronic kidney disease; osteopenia; severe osteoarthritis (OA) affecting the hips, shoulders, and knees; insomnia; and depression is transferred to a new family medicine practice for evaluation. She has been taking nonsteroidal anti-­inflammatory drugs (NSAIDs) for chronic pain secondary to OA for 6 months, with no improvement in symptoms.

The patient is barely able to ambulate and appears to be in considerable pain. She is relying heavily on her walker and is assisted by her granddaughter. The primary care physician (PCP) obtains a detailed history that includes chronic shoulder and hip pain. Given that the patient has not responded to NSAID treatment over the previous 6 months, the PCP takes a moment to reconsider the diagnosis of OA and considers other options.

It’s critical that we consider alternative diagnoses and continue to ask ourselves, “What else?” while working through differential diagnoses.

In light of the high prevalence of PMR in older women, the physician pursues a more specific physical examination tailored to ferret out PMR. He had learned this diagnostic shortcut as a resident, remembered it, and adeptly applied it whenever circumstances warranted. He asks the patient to raise her arms above her head (goalpost sign). She is unable to perform this task and experiences severe bilateral shoulder pain on trial. The PCP then places the patient on the examining table and attempts to assist her in rolling toward him. The patient is also unable to perform this maneuver and experiences significant bilateral hip pain on trial.

Based primarily on the patient’s history and physical exam findings, the PCP makes a presumptive diagnosis of PMR vs OA vs combined PMR with OA, orders an erythrocyte sedimentation rate (ESR) and basic rheumatologic marker panel, and starts the patient on prednisone 10 mg/d. Lab work comes back and reveals mildly elevated ESR with all other findings within normal limits. Two weeks later, the patient returns for her follow-up visit, walking without a walker for the first time in years.

PMR can be mistaken for OA

PMR is the most common inflammatory rheumatic disease in older patients.⁸ It is a debilitating illness with simple, effective treatment but has devastating consequences if missed or left untreated.⁹ PMR typically manifests in patients older than age 50, with a peak incidence at 80 years of age. It is also far more common in women.¹⁰

Approximately 80% of patients with PMR initially present to their PCP, often posing a diagnostic challenge to many clinicians.¹¹ Due to overlap in symptoms, the condition is often misdiagnosed as OA, a more common condition seen by PCPs. Also, there are no specific diagnostic tests for PMR. An elevated ESR can help confirm the diagnosis, but one-third of patients with PMR have a normal ESR.¹² Therefore, the diagnostic conundrum the physician faces is OA vs rheumatoid arthritis (RA), PMR, or another condition.

Continue to: The consequences...

The consequences of a missed and delayed PMR diagnosis range from seriously impaired quality of life to significantly increased risk of vascular events (eg, blindness, stroke) due to temporal arteritis.¹³ Early diagnosis is even more critical as the risk of a vascular event and death is highest during initial phases of the disease course.¹⁴

FPs often miss this Dx. A timely diagnosis relies almost exclusively on an accurate, thorough history and physical exam. However, PCPs often struggle to correctly diagnose PMR. According to a study by Bahlas and colleagues,¹⁵ the accuracy rate for correctly diagnosing PMR was 24% among a cohort of family physicians.

The differential diagnosis for PMR is broad and includes seronegative spondyloarthropathies, malignancy, Lyme disease, hypothyroidism, and both RA and OA.¹⁶ 

PCPs are extremely adept at correctly diagnosing RA, but not PMR. A study by Blaauw and colleagues¹⁷ comparing PCPs and rheumatologists found PCPs correctly identified 92% of RA cases but only 55% of PMR cases. When rheumatologists reviewed these same cases, they correctly identified PMR and RA almost 100% of the time.¹⁷ The difference in diagnostic accuracy between rheumatologists and PCPs suggests limited experience and gaps in fund of knowledge. 

Making the diagnosis. The diagnosis of PMR is often made on empiric response to corticosteroid treatment, but doing so based solely on a patient’s response is controversial.¹⁸ There are rare instances in which patients with PMR fail to respond to treatment. On the other hand, some inflammatory conditions that mimic or share symptoms with PMR also respond to corticosteroids, potentially resulting in erroneous confirmation bias.

Some classification criteria use rapid response to low-dose prednisone/­prednisolone (≤ 20 mg) to confirm the diagnosis,¹⁹ while other more recent guidelines no longer include this approach.²⁰ If PMR continues to be suspected after a trial of steroids is unsuccessful, the PCP can try another course of higher dose steroids or consult with Rheumatology.

Case 2 

A woman in her mid-40s presented to a PCP’s office with a chief complaint of dyspepsia and bloating.^a The patient was attending a meeting in New York City, and this was her first visit to this physician. The patient previously had been treated for these symptoms by her hometown PCP and gastroenterologist.

Continue to: A full history...

A full history and physical exam revealed a myriad of gastrointestinal (GI) complaints, such as diarrhea. But the PCP recalled a recent roundtable discussion on debiasing techniques specifically related to gynecologic disorders, including OC. Therefore, he decided to include OC in the differential diagnosis—something he would not routinely have done given the preponderance of GI symptoms. Despite the patient’s reluctance and time constraints, the PCP ordered a transvaginal ultrasound. Findings from the ultrasound study revealed stage II OC, which carries a good prognosis. The patient is currently undergoing treatment and was last reported as doing well.

Early signs of ovarian cancer can be chalked up to a “GI issue”

OC is the second most common gynecologic cancer²¹ and the fifth leading cause of cancer-related death²² in US women. Compared to other cancers, the prognosis for localized ­early-stage OC is surprisingly good, with a 5-year survival rate approaching 93%.²³ However, most disease is detected in later stages, and the 5-year survival rate drops to a low of 29%.²⁴

There remains no established screening protocol for OC. Fewer than a quarter of all cases are diagnosed in stage I, and detection of OC relies heavily on the physician’s ability to decipher vague symptomatology that overlaps with other, more common maladies. This poses an obvious diagnostic challenge and, not surprisingly, a high level of susceptibility to cognitive bias. 

Most PCPs correctly identified bloating as a key symptom of ovarian cancer; however, they weren’t as good at identifying less common symptoms, such as inability to finish a meal.

More than 90% of patients with OC present with some combination of the following symptoms prior to diagnosis: abdominal (77%), GI (70%), pain (58%), constitutional (50%), urinary (34%), and pelvic (26%).²⁵ The 3 most common isolated symptoms in patients with OC are abdominal bloating, decrease in appetite, and frank abdominal pain.²⁶ Patients with biopsy-confirmed OC experience these symptoms an average of 6 months prior to actual diagnosis.²⁷

Knowledge gaps play a role. Studies assessing the ability of health care providers to identify presenting symptoms of OC reveal specific knowledge gaps. For instance, in a survey by Gajjar and colleagues,²⁸ most PCPs correctly identified bloating as a key symptom of OC; however, they weren’t as good at identifying less common symptoms, such as inability to finish a meal and early satiety. Moreover, survey participants misinterpreted or missed GI symptoms as an important manifestation of early OC disease.²⁸ These specific knowledge gaps combine with physician errors in thinking, further obscuring and extending the diagnostic process. The point prevalence for OC is relatively low, and many PCPs only encounter a few cases during their entire career.²⁹ This low pre-test probability may also fuel the delay in diagnosis.

Watch for these forms of bias. Since nonspecific symptoms of early-stage OC resemble those of other more benign conditions, a form of anchoring error known as multiple alternatives bias can arise. In this scenario, clinicians investigate only 1 potential plausible diagnosis and remain focused on that single, often faulty, conclusion. This persists despite other equally plausible alternatives that arise as the investigation proceeds.²⁸

Affective error may also play a role in missed or delayed diagnosis. For example, a physician would prefer to diagnose and treat a common GI illness than consider OC. Another distortion involves outcome bias wherein the physician gives more significance to benign conditions such as irritable bowel syndrome because they have a more favorable outcome and clear treatment path. Physicians also favor these benign conditions because they encounter them more frequently than OC in the clinic setting. (This is known as availability bias.) Outcome bias and multiple alternatives bias can result in noninvestigation of symptoms and inefficient or improper management, leading to a delay in arriving at the correct diagnosis or anchoring on a plausible but incorrect diagnosis.

Continue to: An incorrect initial diagnostic...

An incorrect initial diagnostic path often triggers a cascade of subsequent errors. The physician orders additional unhelpful and expensive tests in an effort to characterize the suspected GI pathology. This then leads the physician to prematurely terminate the work-up and accept the most favored diagnosis. Lastly, sunk-cost fallacy comes into play: The physician has “invested” time and energy investigating a particular diagnosis and rather than abandon the presumed diagnosis, continues to put more time and effort in going down an incorrect diagnostic path. 

A series of failures. These biases and miscues have been observed in several studies. For example, a survey of 1725 women by Goff and colleagues³⁰ sought to identify factors related to delayed OC diagnosis. The authors found that the following factors were significantly associated with a delayed diagnosis: omission of a pelvic exam at initial presentation, a separate exploration of a multitude of collateral symptoms, a failure to order ultrasound/computed tomography/CA-125 test, and a failure to consider age as a factor (especially if the patient was outside the norm). 

Responses from the survey also revealed that physicians initially ordered work-ups related to GI etiology and only later considered a pelvic work-up. This suggests that well-known presenting signs and symptoms or a constellation of typical and atypical symptoms of OC often failed to trigger physician recognition. Understandably, patients presenting with menorrhagia or gynecologic complaints are more likely to have OC detected at an earlier stage than patients who present with GI or abdominal signs alone.³¹ table 2⁷ summarizes some of the cognitive biases seen in the diagnostic path of OC.

Case 3

A 56-year-old man is brought to the ED by his wife and children for evaluation of odd behavior and episodes of confusion. The patient recently had a negative neurologic work-up for transient ischemic attack and cerebrovascular accident and is admitted for further work-up. He reports visual hallucinations to nursing staff. Screening for memory problems shows no significant deficits. The patient in fact scored a 27 on the Mini–Mental State Examination, well within the normal range. The family notes that the patient has had difficulty with planning over the previous year and has not seemed like his “old self.” The patient has no history of psychosis, schizophrenia, bipolar disorder, or any other psychiatric illness.

While in the hospital, he becomes acutely upset by the hallucinations and is given haloperidol and lorazepam by house staff. In the morning, the patient exhibits severe signs of Parkinson disease that include rigidity and masked facies.

Given the patient’s poor response to haloperidol and continued confusion, the team consulted Neurology and Psychiatry. Gathering a more detailed history from the patient and family, the patient is given a diagnosis of classic LBD. The antipsychotic medications are stopped. The patient and his family receive education about LBD treatment and management, and the patient is discharged to outpatient care.

Psychiatric symptoms can be an early “misdirect” in cases of Lewy body disease

LBD, the second leading neurodegenerative dementia after Alzheimer disease (AD), affects 1.5 million Americans,³² representing about 10% of all dementia cases. LBD and AD overlap in 25% of dementia cases.³³ In patients older than 85 years, the prevalence jumps to 5% of the general population and 22% of all cases of dementia.³³ Despite its prevalence, a recent study showed that only 6% of PCPs correctly identified LBD as the primary diagnosis when presented with typical case examples.³² 

Continue to: 3 stages of presentation

3 stages of presentation. Unlike other forms of dementia, LBD typically presents first with psychiatric symptoms, then with cognitive impairment, and last with parkinsonian symptoms. Additionally, rapid eye movement sleep behavior disorder and often subtle elements of nonmemory cognitive impairment distinguish LBD from both AD and vascular dementia.³² The primary cognitive deficit in LBD is not in memory but in attention, executive functioning, and visuospatial ability.³⁴ Only in the later stages of the disease do patients exhibit gradual and progressive memory loss. 

Mistaken for many things. When evaluating patients exhibiting signs of dementia, it’s important to include LBD in the differential, with increased suspicion for patients experiencing episodes of psychosis or delirium. The uniqueness of LBD lies in its psychotic symptomatology, particularly during earlier stages of the disease. This feature helps distinguish LBD from both AD and vascular dementia. As seen in the case, LBD can also be confused with acute delirium.

The uniqueness of Lewy body dementia lies in its psychotic symptomatology, particularly during earlier stages of the disease.

Older adult patients presenting to the ED or clinic with visual hallucinations, delirium, and mental confusion may receive a false diagnosis of schizophrenia, medication- or substance-induced psychosis, Parkinson disease, or delirium of unknown etiology.³⁵ Unfortunately, LBD is often overlooked and not considered in the differential diagnosis. Due to underrecognition, patients may receive treatment with typical antipsychotics. The addition of a neuroleptic to help control the psychotic symptoms causes patients with LBD to develop severe extrapyramidal symptoms and worsening mental status,³⁶ leading to severe parkinsonian signs, which further muddies the diagnostic process. In addition, treatment for suspected Parkinson disease, including carbidopa-levodopa, has no benefit for patients with LBD and may increase psychotic symptoms.³⁷

First-line treatment for LBD includes psychoeducation for the patient and family, cholinesterase inhibitors (eg, rivastigmine), and avoidance of high-potency antipsychotics, such as haloperidol. Although persistent hallucinations and psychosis remain difficult to treat in LBD, low-dose quetiapine is 1 option. Incorrectly diagnosing and prescribing treatment for another condition exacerbates symptoms in this patient population.

Case 4

A 36-year-old Hispanic woman presents to the PCP for her annual physical exam. The patient’s medical record shows 2 previous office visits over the past 2 years—an annual physical exam and an office visit for the flu vaccine. The patient is highly accomplished in her profession, working as a certified public accountant for a major corporation. She is a nonsmoker and reports only casual social drinking and no recreational drug use. The patient is slightly overweight for her height but is otherwise healthy. Previous lab studies are within normal limits.

The patient has been experiencing chronic pain for the past few years after a motor vehicle accident. She has seen a physiatrist and another provider, both of whom found no “objective” causes of her chronic pain. They started the patient on sertraline for depression and an analgesic, both of which were ineffective.

The patient likes to exercise at a gym twice a week by doing light cardio (treadmill) exercise and light weightlifting. Lately, however, she has been unable to exercise due to the pain. At this visit, she mentions having low energy, poor sleep, frequent fatigue, and generalized soreness and pain in multiple areas of her body. The PCP recognizes the patient’s presenting symptoms as significant for FM and starts her on pregabalin and hydrotherapy, with positive results.

Continue to: Fibromyalgia skepticism may lead to a Dx of depression

FM, the second most common disorder seen in rheumatologic practice after OA, is estimated to affect approximately 1 in 20 patients (approximately 5 million Americans) in the primary care setting.^38,39 The condition has a high female-to-male preponderance (3.4% vs 0.5%).⁴⁰ While the primary symptom of FM is chronic pain, patients commonly present with fatigue and sleep disturbance.⁴¹ Comorbid conditions include headaches, irritable bowel syndrome, and mood disturbances (most commonly anxiety and depression). 

Patients with fibromyalgia, who are often otherwise healthy, often present multiple times to the same PCP with a chief complaint of chronic pain.

Several studies have explored reasons for the misdiagnosis and underdiagnosis of FM. One important factor is ongoing skepticism among some physicians and the public, in general, as to whether FM is a real disease. This issue was addressed by a study by White and colleagues,⁴² who estimated and compared the point prevalence of FM and related disorders in Amish vs non-Amish adults. The authors hypothesized that if litigation and/or compensation availability have a major impact on FM prevalence, then there would be a near zero prevalence of FM in the Amish community. And yet, researchers found an overall age- and sex-adjusted FM prevalence of 7.3% (95% CI; 5.3%-9.7%); this was both statistically greater than zero (P < .0001) and greater than 2 control populations of non-Amish adults (both P < .05).

Many physicians consider FM fundamentally an emotional disturbance, and the high preponderance of FM in female patients may contribute to this misconception as reports of pain and emotional distress by women are often dismissed as hysteria.⁴³ Physicians often explore the emotional aspects of FM, incorrectly diagnosing patients with depression and subsequently treating them with a psychotropic drug.³⁹ Alternatively, they may focus on the musculoskeletal presentations of FM and prescribe analgesics or physical therapy, both of which do little to alleviate FM. 

To make the correct diagnosis of FM, the American College of Rheumatology created a specific set of criteria in 1990, which was updated in 2010.⁴⁴ For a diagnosis of FM, a patient must have at least a 3-month history of bilateral pain above and below the waist and along the axial skeletal spine. Although not included in the updated 2010 criteria, many clinicians continue to check for tender points, following the 1990 criteria requiring the presence of 11 of 18 points to make the diagnosis.

At least 3 cognitive biases relating to FM apply: anchoring, availability, and fundamental attribution error (see table 3).⁷ Anchoring occurs when the PCP settles on a psychiatric diagnosis of exaggerated pain syndrome, muscle overuse, or OA and fails to explore alternative etiology. Availability bias may obscure the true diagnosis of FM. Since PCPs see many patients with RA or OA, they may overlook or dismiss the possibility of FM. Attribution error happens when physicians dismiss the complaints of patients with FM as merely due to psychological distress, hysteria, or acting out.⁴³

Patients with FM, who are often otherwise healthy, often present multiple times to the same PCP with a chief complaint of chronic pain. These repeat presentations can result in compassion fatigue and impact care. As Aloush and colleagues⁴⁰ noted in their study, “FM patients were perceived as more difficult than RA patients, with a high level of concern and emotional response. A high proportion of physicians were reluctant to accept them because they feel emotional/psychological difficulties meeting and coping with these patients.”In response, patients with undiagnosed FM or inadequately treated FM may visit other PCPs, which may or may not result in a correct diagnosis and treatment.

We can do better

Primary care physicians face the daunting task of diagnosing and treating a wide range of common conditions while also trying to recognize less-common conditions with atypical presentations—all during a busy clinic workday. Nonetheless, we should strive to overcome internal (eg, cognitive bias and fund-of-knowledge deficits) and external (eg, time constraints, limited resources) pressures to improve diagnostic accuracy and care.

Each of the 4 disease states we’ve discussed have high rates of missed and/or delayed diagnosis. Each presents a unique set of confounders: PMR with its overlapping symptoms of many other rheumatologic diseases; OC with its often vague and misleading GI symptomatology; LBD with overlapping features of AD and Parkinson disease; and FM with skepticism. As gatekeepers to health care, it falls on PCPs to sort out these diagnostic dilemmas to avoid medical errors. Fundamental knowledge of each disease, its unique pathophysiology and symptoms, and varying presentations can be learned, internalized, and subsequently put into clinical practice to improve patient outcomes.

CORRESPONDENCE
Paul D. Rosen MD, Brooklyn Hospital Center, Department of Family Medicine, 121 Dekalb Avenue, Brooklyn, New York 11201; [email protected]

Medical errors in all settings contributed to as many as 250,000 deaths per year in the United States between 2000 and 2008, according to a 2016 study.¹ Diagnostic error, in particular, remains a leading cause of morbidity and mortality in the United States and worldwide. In 2017, 12 million patients (roughly 5% of all US adults) who sought outpatient care experienced missed, delayed, or incorrect diagnosis at least once.²

In his classic work, How Doctors Think, Jerome Groopman, MD, explored the diagnostic process with a focus on the role of cognitive bias in clinical decision-making. Groopman examined how physicians can become sidetracked in their thinking and “blinded” to potential alternative diagnoses.³ Medical error is not necessarily because of a deficiency in medical knowledge; rather, physicians become susceptible to medical error when defective and faulty reasoning distort their diagnostic ability.⁴

Cognitive bias in the diagnostic process has been extensively studied, and a full review is beyond the scope of this article.⁵ However, here we will examine pathways leading to diagnostic errors in the primary care setting, specifically the role of cognitive bias in the work-up of polymyalgia rheumatica (PMR), ovarian cancer (OC), Lewy body dementia (LBD), and fibromyalgia (FM). As these 4 disease states are seen with low-to-moderate frequency in primary care, cognitive bias can complicate accurate diagnosis. But first, a word about how to understand clinical reasoning.

There are 2 types of reasoning (and 1 is more prone to error)

Physician clinical reasoning can be divided into 2 different cognitive approaches.

Type 1 reasoning employs intuition and heuristics; this type is automatic, reflexive, and quick.⁵ While the use of mental shortcuts in type 1 increases the speed with which decisions are made, it also makes this form of reasoning more prone to error.

Type 2 reasoning requires conscious effort. It is goal directed and rigorous and therefore slower than type 1 reasoning. Extrapolated to the clinical context, clinicians transition from type 2 to type 1 reasoning as they gain experience and training throughout their careers and develop their own conscious and subconscious heuristics. Deviations from accurate decision-making occur in a systematic manner due to cognitive biases and result in medical error.⁶table 1⁷ lists common types of cognitive bias.

An important question to ask. Physicians tend to fall into a pattern of quick, type 1 reasoning. However, it’s important to strive to maintain a broad differential diagnosis and avoid premature closure of the diagnostic process. It’s critical that we consider alternative diagnoses (ie, consciously move from type 1 to type 2 thinking) and continue to ask ourselves, “What else?” while working through differential diagnoses. This can be a powerful debiasing technique.

Continue to: The discussion...

The discussion of the following 4 disease states demonstrates how cognitive bias can lead to diagnostic error.

Case 1

An 82-year-old woman with a history of hypertension; wide-angle glaucoma; stage 2 chronic kidney disease; osteopenia; severe osteoarthritis (OA) affecting the hips, shoulders, and knees; insomnia; and depression is transferred to a new family medicine practice for evaluation. She has been taking nonsteroidal anti-­inflammatory drugs (NSAIDs) for chronic pain secondary to OA for 6 months, with no improvement in symptoms.

The patient is barely able to ambulate and appears to be in considerable pain. She is relying heavily on her walker and is assisted by her granddaughter. The primary care physician (PCP) obtains a detailed history that includes chronic shoulder and hip pain. Given that the patient has not responded to NSAID treatment over the previous 6 months, the PCP takes a moment to reconsider the diagnosis of OA and considers other options.

It’s critical that we consider alternative diagnoses and continue to ask ourselves, “What else?” while working through differential diagnoses.

In light of the high prevalence of PMR in older women, the physician pursues a more specific physical examination tailored to ferret out PMR. He had learned this diagnostic shortcut as a resident, remembered it, and adeptly applied it whenever circumstances warranted. He asks the patient to raise her arms above her head (goalpost sign). She is unable to perform this task and experiences severe bilateral shoulder pain on trial. The PCP then places the patient on the examining table and attempts to assist her in rolling toward him. The patient is also unable to perform this maneuver and experiences significant bilateral hip pain on trial.

Based primarily on the patient’s history and physical exam findings, the PCP makes a presumptive diagnosis of PMR vs OA vs combined PMR with OA, orders an erythrocyte sedimentation rate (ESR) and basic rheumatologic marker panel, and starts the patient on prednisone 10 mg/d. Lab work comes back and reveals mildly elevated ESR with all other findings within normal limits. Two weeks later, the patient returns for her follow-up visit, walking without a walker for the first time in years.

PMR can be mistaken for OA

PMR is the most common inflammatory rheumatic disease in older patients.⁸ It is a debilitating illness with simple, effective treatment but has devastating consequences if missed or left untreated.⁹ PMR typically manifests in patients older than age 50, with a peak incidence at 80 years of age. It is also far more common in women.¹⁰

Approximately 80% of patients with PMR initially present to their PCP, often posing a diagnostic challenge to many clinicians.¹¹ Due to overlap in symptoms, the condition is often misdiagnosed as OA, a more common condition seen by PCPs. Also, there are no specific diagnostic tests for PMR. An elevated ESR can help confirm the diagnosis, but one-third of patients with PMR have a normal ESR.¹² Therefore, the diagnostic conundrum the physician faces is OA vs rheumatoid arthritis (RA), PMR, or another condition.

Continue to: The consequences...

The consequences of a missed and delayed PMR diagnosis range from seriously impaired quality of life to significantly increased risk of vascular events (eg, blindness, stroke) due to temporal arteritis.¹³ Early diagnosis is even more critical as the risk of a vascular event and death is highest during initial phases of the disease course.¹⁴

FPs often miss this Dx. A timely diagnosis relies almost exclusively on an accurate, thorough history and physical exam. However, PCPs often struggle to correctly diagnose PMR. According to a study by Bahlas and colleagues,¹⁵ the accuracy rate for correctly diagnosing PMR was 24% among a cohort of family physicians.

The differential diagnosis for PMR is broad and includes seronegative spondyloarthropathies, malignancy, Lyme disease, hypothyroidism, and both RA and OA.¹⁶ 

PCPs are extremely adept at correctly diagnosing RA, but not PMR. A study by Blaauw and colleagues¹⁷ comparing PCPs and rheumatologists found PCPs correctly identified 92% of RA cases but only 55% of PMR cases. When rheumatologists reviewed these same cases, they correctly identified PMR and RA almost 100% of the time.¹⁷ The difference in diagnostic accuracy between rheumatologists and PCPs suggests limited experience and gaps in fund of knowledge. 

Making the diagnosis. The diagnosis of PMR is often made on empiric response to corticosteroid treatment, but doing so based solely on a patient’s response is controversial.¹⁸ There are rare instances in which patients with PMR fail to respond to treatment. On the other hand, some inflammatory conditions that mimic or share symptoms with PMR also respond to corticosteroids, potentially resulting in erroneous confirmation bias.

Some classification criteria use rapid response to low-dose prednisone/­prednisolone (≤ 20 mg) to confirm the diagnosis,¹⁹ while other more recent guidelines no longer include this approach.²⁰ If PMR continues to be suspected after a trial of steroids is unsuccessful, the PCP can try another course of higher dose steroids or consult with Rheumatology.

Case 2 

A woman in her mid-40s presented to a PCP’s office with a chief complaint of dyspepsia and bloating.^a The patient was attending a meeting in New York City, and this was her first visit to this physician. The patient previously had been treated for these symptoms by her hometown PCP and gastroenterologist.

Continue to: A full history...

A full history and physical exam revealed a myriad of gastrointestinal (GI) complaints, such as diarrhea. But the PCP recalled a recent roundtable discussion on debiasing techniques specifically related to gynecologic disorders, including OC. Therefore, he decided to include OC in the differential diagnosis—something he would not routinely have done given the preponderance of GI symptoms. Despite the patient’s reluctance and time constraints, the PCP ordered a transvaginal ultrasound. Findings from the ultrasound study revealed stage II OC, which carries a good prognosis. The patient is currently undergoing treatment and was last reported as doing well.

Early signs of ovarian cancer can be chalked up to a “GI issue”

OC is the second most common gynecologic cancer²¹ and the fifth leading cause of cancer-related death²² in US women. Compared to other cancers, the prognosis for localized ­early-stage OC is surprisingly good, with a 5-year survival rate approaching 93%.²³ However, most disease is detected in later stages, and the 5-year survival rate drops to a low of 29%.²⁴

There remains no established screening protocol for OC. Fewer than a quarter of all cases are diagnosed in stage I, and detection of OC relies heavily on the physician’s ability to decipher vague symptomatology that overlaps with other, more common maladies. This poses an obvious diagnostic challenge and, not surprisingly, a high level of susceptibility to cognitive bias. 

Most PCPs correctly identified bloating as a key symptom of ovarian cancer; however, they weren’t as good at identifying less common symptoms, such as inability to finish a meal.

More than 90% of patients with OC present with some combination of the following symptoms prior to diagnosis: abdominal (77%), GI (70%), pain (58%), constitutional (50%), urinary (34%), and pelvic (26%).²⁵ The 3 most common isolated symptoms in patients with OC are abdominal bloating, decrease in appetite, and frank abdominal pain.²⁶ Patients with biopsy-confirmed OC experience these symptoms an average of 6 months prior to actual diagnosis.²⁷

Knowledge gaps play a role. Studies assessing the ability of health care providers to identify presenting symptoms of OC reveal specific knowledge gaps. For instance, in a survey by Gajjar and colleagues,²⁸ most PCPs correctly identified bloating as a key symptom of OC; however, they weren’t as good at identifying less common symptoms, such as inability to finish a meal and early satiety. Moreover, survey participants misinterpreted or missed GI symptoms as an important manifestation of early OC disease.²⁸ These specific knowledge gaps combine with physician errors in thinking, further obscuring and extending the diagnostic process. The point prevalence for OC is relatively low, and many PCPs only encounter a few cases during their entire career.²⁹ This low pre-test probability may also fuel the delay in diagnosis.

Watch for these forms of bias. Since nonspecific symptoms of early-stage OC resemble those of other more benign conditions, a form of anchoring error known as multiple alternatives bias can arise. In this scenario, clinicians investigate only 1 potential plausible diagnosis and remain focused on that single, often faulty, conclusion. This persists despite other equally plausible alternatives that arise as the investigation proceeds.²⁸

Affective error may also play a role in missed or delayed diagnosis. For example, a physician would prefer to diagnose and treat a common GI illness than consider OC. Another distortion involves outcome bias wherein the physician gives more significance to benign conditions such as irritable bowel syndrome because they have a more favorable outcome and clear treatment path. Physicians also favor these benign conditions because they encounter them more frequently than OC in the clinic setting. (This is known as availability bias.) Outcome bias and multiple alternatives bias can result in noninvestigation of symptoms and inefficient or improper management, leading to a delay in arriving at the correct diagnosis or anchoring on a plausible but incorrect diagnosis.

Continue to: An incorrect initial diagnostic...

An incorrect initial diagnostic path often triggers a cascade of subsequent errors. The physician orders additional unhelpful and expensive tests in an effort to characterize the suspected GI pathology. This then leads the physician to prematurely terminate the work-up and accept the most favored diagnosis. Lastly, sunk-cost fallacy comes into play: The physician has “invested” time and energy investigating a particular diagnosis and rather than abandon the presumed diagnosis, continues to put more time and effort in going down an incorrect diagnostic path. 

A series of failures. These biases and miscues have been observed in several studies. For example, a survey of 1725 women by Goff and colleagues³⁰ sought to identify factors related to delayed OC diagnosis. The authors found that the following factors were significantly associated with a delayed diagnosis: omission of a pelvic exam at initial presentation, a separate exploration of a multitude of collateral symptoms, a failure to order ultrasound/computed tomography/CA-125 test, and a failure to consider age as a factor (especially if the patient was outside the norm). 

Responses from the survey also revealed that physicians initially ordered work-ups related to GI etiology and only later considered a pelvic work-up. This suggests that well-known presenting signs and symptoms or a constellation of typical and atypical symptoms of OC often failed to trigger physician recognition. Understandably, patients presenting with menorrhagia or gynecologic complaints are more likely to have OC detected at an earlier stage than patients who present with GI or abdominal signs alone.³¹ table 2⁷ summarizes some of the cognitive biases seen in the diagnostic path of OC.

Case 3

A 56-year-old man is brought to the ED by his wife and children for evaluation of odd behavior and episodes of confusion. The patient recently had a negative neurologic work-up for transient ischemic attack and cerebrovascular accident and is admitted for further work-up. He reports visual hallucinations to nursing staff. Screening for memory problems shows no significant deficits. The patient in fact scored a 27 on the Mini–Mental State Examination, well within the normal range. The family notes that the patient has had difficulty with planning over the previous year and has not seemed like his “old self.” The patient has no history of psychosis, schizophrenia, bipolar disorder, or any other psychiatric illness.

While in the hospital, he becomes acutely upset by the hallucinations and is given haloperidol and lorazepam by house staff. In the morning, the patient exhibits severe signs of Parkinson disease that include rigidity and masked facies.

Given the patient’s poor response to haloperidol and continued confusion, the team consulted Neurology and Psychiatry. Gathering a more detailed history from the patient and family, the patient is given a diagnosis of classic LBD. The antipsychotic medications are stopped. The patient and his family receive education about LBD treatment and management, and the patient is discharged to outpatient care.

Psychiatric symptoms can be an early “misdirect” in cases of Lewy body disease

LBD, the second leading neurodegenerative dementia after Alzheimer disease (AD), affects 1.5 million Americans,³² representing about 10% of all dementia cases. LBD and AD overlap in 25% of dementia cases.³³ In patients older than 85 years, the prevalence jumps to 5% of the general population and 22% of all cases of dementia.³³ Despite its prevalence, a recent study showed that only 6% of PCPs correctly identified LBD as the primary diagnosis when presented with typical case examples.³² 

Continue to: 3 stages of presentation

3 stages of presentation. Unlike other forms of dementia, LBD typically presents first with psychiatric symptoms, then with cognitive impairment, and last with parkinsonian symptoms. Additionally, rapid eye movement sleep behavior disorder and often subtle elements of nonmemory cognitive impairment distinguish LBD from both AD and vascular dementia.³² The primary cognitive deficit in LBD is not in memory but in attention, executive functioning, and visuospatial ability.³⁴ Only in the later stages of the disease do patients exhibit gradual and progressive memory loss. 

Mistaken for many things. When evaluating patients exhibiting signs of dementia, it’s important to include LBD in the differential, with increased suspicion for patients experiencing episodes of psychosis or delirium. The uniqueness of LBD lies in its psychotic symptomatology, particularly during earlier stages of the disease. This feature helps distinguish LBD from both AD and vascular dementia. As seen in the case, LBD can also be confused with acute delirium.

The uniqueness of Lewy body dementia lies in its psychotic symptomatology, particularly during earlier stages of the disease.

Older adult patients presenting to the ED or clinic with visual hallucinations, delirium, and mental confusion may receive a false diagnosis of schizophrenia, medication- or substance-induced psychosis, Parkinson disease, or delirium of unknown etiology.³⁵ Unfortunately, LBD is often overlooked and not considered in the differential diagnosis. Due to underrecognition, patients may receive treatment with typical antipsychotics. The addition of a neuroleptic to help control the psychotic symptoms causes patients with LBD to develop severe extrapyramidal symptoms and worsening mental status,³⁶ leading to severe parkinsonian signs, which further muddies the diagnostic process. In addition, treatment for suspected Parkinson disease, including carbidopa-levodopa, has no benefit for patients with LBD and may increase psychotic symptoms.³⁷

First-line treatment for LBD includes psychoeducation for the patient and family, cholinesterase inhibitors (eg, rivastigmine), and avoidance of high-potency antipsychotics, such as haloperidol. Although persistent hallucinations and psychosis remain difficult to treat in LBD, low-dose quetiapine is 1 option. Incorrectly diagnosing and prescribing treatment for another condition exacerbates symptoms in this patient population.

Case 4

A 36-year-old Hispanic woman presents to the PCP for her annual physical exam. The patient’s medical record shows 2 previous office visits over the past 2 years—an annual physical exam and an office visit for the flu vaccine. The patient is highly accomplished in her profession, working as a certified public accountant for a major corporation. She is a nonsmoker and reports only casual social drinking and no recreational drug use. The patient is slightly overweight for her height but is otherwise healthy. Previous lab studies are within normal limits.

The patient has been experiencing chronic pain for the past few years after a motor vehicle accident. She has seen a physiatrist and another provider, both of whom found no “objective” causes of her chronic pain. They started the patient on sertraline for depression and an analgesic, both of which were ineffective.

The patient likes to exercise at a gym twice a week by doing light cardio (treadmill) exercise and light weightlifting. Lately, however, she has been unable to exercise due to the pain. At this visit, she mentions having low energy, poor sleep, frequent fatigue, and generalized soreness and pain in multiple areas of her body. The PCP recognizes the patient’s presenting symptoms as significant for FM and starts her on pregabalin and hydrotherapy, with positive results.

Continue to: Fibromyalgia skepticism may lead to a Dx of depression

FM, the second most common disorder seen in rheumatologic practice after OA, is estimated to affect approximately 1 in 20 patients (approximately 5 million Americans) in the primary care setting.^38,39 The condition has a high female-to-male preponderance (3.4% vs 0.5%).⁴⁰ While the primary symptom of FM is chronic pain, patients commonly present with fatigue and sleep disturbance.⁴¹ Comorbid conditions include headaches, irritable bowel syndrome, and mood disturbances (most commonly anxiety and depression). 

Patients with fibromyalgia, who are often otherwise healthy, often present multiple times to the same PCP with a chief complaint of chronic pain.

Several studies have explored reasons for the misdiagnosis and underdiagnosis of FM. One important factor is ongoing skepticism among some physicians and the public, in general, as to whether FM is a real disease. This issue was addressed by a study by White and colleagues,⁴² who estimated and compared the point prevalence of FM and related disorders in Amish vs non-Amish adults. The authors hypothesized that if litigation and/or compensation availability have a major impact on FM prevalence, then there would be a near zero prevalence of FM in the Amish community. And yet, researchers found an overall age- and sex-adjusted FM prevalence of 7.3% (95% CI; 5.3%-9.7%); this was both statistically greater than zero (P < .0001) and greater than 2 control populations of non-Amish adults (both P < .05).

Many physicians consider FM fundamentally an emotional disturbance, and the high preponderance of FM in female patients may contribute to this misconception as reports of pain and emotional distress by women are often dismissed as hysteria.⁴³ Physicians often explore the emotional aspects of FM, incorrectly diagnosing patients with depression and subsequently treating them with a psychotropic drug.³⁹ Alternatively, they may focus on the musculoskeletal presentations of FM and prescribe analgesics or physical therapy, both of which do little to alleviate FM. 

To make the correct diagnosis of FM, the American College of Rheumatology created a specific set of criteria in 1990, which was updated in 2010.⁴⁴ For a diagnosis of FM, a patient must have at least a 3-month history of bilateral pain above and below the waist and along the axial skeletal spine. Although not included in the updated 2010 criteria, many clinicians continue to check for tender points, following the 1990 criteria requiring the presence of 11 of 18 points to make the diagnosis.

At least 3 cognitive biases relating to FM apply: anchoring, availability, and fundamental attribution error (see table 3).⁷ Anchoring occurs when the PCP settles on a psychiatric diagnosis of exaggerated pain syndrome, muscle overuse, or OA and fails to explore alternative etiology. Availability bias may obscure the true diagnosis of FM. Since PCPs see many patients with RA or OA, they may overlook or dismiss the possibility of FM. Attribution error happens when physicians dismiss the complaints of patients with FM as merely due to psychological distress, hysteria, or acting out.⁴³

Patients with FM, who are often otherwise healthy, often present multiple times to the same PCP with a chief complaint of chronic pain. These repeat presentations can result in compassion fatigue and impact care. As Aloush and colleagues⁴⁰ noted in their study, “FM patients were perceived as more difficult than RA patients, with a high level of concern and emotional response. A high proportion of physicians were reluctant to accept them because they feel emotional/psychological difficulties meeting and coping with these patients.”In response, patients with undiagnosed FM or inadequately treated FM may visit other PCPs, which may or may not result in a correct diagnosis and treatment.

We can do better

Primary care physicians face the daunting task of diagnosing and treating a wide range of common conditions while also trying to recognize less-common conditions with atypical presentations—all during a busy clinic workday. Nonetheless, we should strive to overcome internal (eg, cognitive bias and fund-of-knowledge deficits) and external (eg, time constraints, limited resources) pressures to improve diagnostic accuracy and care.

Each of the 4 disease states we’ve discussed have high rates of missed and/or delayed diagnosis. Each presents a unique set of confounders: PMR with its overlapping symptoms of many other rheumatologic diseases; OC with its often vague and misleading GI symptomatology; LBD with overlapping features of AD and Parkinson disease; and FM with skepticism. As gatekeepers to health care, it falls on PCPs to sort out these diagnostic dilemmas to avoid medical errors. Fundamental knowledge of each disease, its unique pathophysiology and symptoms, and varying presentations can be learned, internalized, and subsequently put into clinical practice to improve patient outcomes.

CORRESPONDENCE
Paul D. Rosen MD, Brooklyn Hospital Center, Department of Family Medicine, 121 Dekalb Avenue, Brooklyn, New York 11201; [email protected]

Medical errors in all settings contributed to as many as 250,000 deaths per year in the United States between 2000 and 2008, according to a 2016 study.¹ Diagnostic error, in particular, remains a leading cause of morbidity and mortality in the United States and worldwide. In 2017, 12 million patients (roughly 5% of all US adults) who sought outpatient care experienced missed, delayed, or incorrect diagnosis at least once.²

In his classic work, How Doctors Think, Jerome Groopman, MD, explored the diagnostic process with a focus on the role of cognitive bias in clinical decision-making. Groopman examined how physicians can become sidetracked in their thinking and “blinded” to potential alternative diagnoses.³ Medical error is not necessarily because of a deficiency in medical knowledge; rather, physicians become susceptible to medical error when defective and faulty reasoning distort their diagnostic ability.⁴

Cognitive bias in the diagnostic process has been extensively studied, and a full review is beyond the scope of this article.⁵ However, here we will examine pathways leading to diagnostic errors in the primary care setting, specifically the role of cognitive bias in the work-up of polymyalgia rheumatica (PMR), ovarian cancer (OC), Lewy body dementia (LBD), and fibromyalgia (FM). As these 4 disease states are seen with low-to-moderate frequency in primary care, cognitive bias can complicate accurate diagnosis. But first, a word about how to understand clinical reasoning.

There are 2 types of reasoning (and 1 is more prone to error)

Physician clinical reasoning can be divided into 2 different cognitive approaches.

Type 1 reasoning employs intuition and heuristics; this type is automatic, reflexive, and quick.⁵ While the use of mental shortcuts in type 1 increases the speed with which decisions are made, it also makes this form of reasoning more prone to error.

Type 2 reasoning requires conscious effort. It is goal directed and rigorous and therefore slower than type 1 reasoning. Extrapolated to the clinical context, clinicians transition from type 2 to type 1 reasoning as they gain experience and training throughout their careers and develop their own conscious and subconscious heuristics. Deviations from accurate decision-making occur in a systematic manner due to cognitive biases and result in medical error.⁶table 1⁷ lists common types of cognitive bias.

An important question to ask. Physicians tend to fall into a pattern of quick, type 1 reasoning. However, it’s important to strive to maintain a broad differential diagnosis and avoid premature closure of the diagnostic process. It’s critical that we consider alternative diagnoses (ie, consciously move from type 1 to type 2 thinking) and continue to ask ourselves, “What else?” while working through differential diagnoses. This can be a powerful debiasing technique.

Continue to: The discussion...

The discussion of the following 4 disease states demonstrates how cognitive bias can lead to diagnostic error.

Case 1

An 82-year-old woman with a history of hypertension; wide-angle glaucoma; stage 2 chronic kidney disease; osteopenia; severe osteoarthritis (OA) affecting the hips, shoulders, and knees; insomnia; and depression is transferred to a new family medicine practice for evaluation. She has been taking nonsteroidal anti-­inflammatory drugs (NSAIDs) for chronic pain secondary to OA for 6 months, with no improvement in symptoms.

The patient is barely able to ambulate and appears to be in considerable pain. She is relying heavily on her walker and is assisted by her granddaughter. The primary care physician (PCP) obtains a detailed history that includes chronic shoulder and hip pain. Given that the patient has not responded to NSAID treatment over the previous 6 months, the PCP takes a moment to reconsider the diagnosis of OA and considers other options.

It’s critical that we consider alternative diagnoses and continue to ask ourselves, “What else?” while working through differential diagnoses.

In light of the high prevalence of PMR in older women, the physician pursues a more specific physical examination tailored to ferret out PMR. He had learned this diagnostic shortcut as a resident, remembered it, and adeptly applied it whenever circumstances warranted. He asks the patient to raise her arms above her head (goalpost sign). She is unable to perform this task and experiences severe bilateral shoulder pain on trial. The PCP then places the patient on the examining table and attempts to assist her in rolling toward him. The patient is also unable to perform this maneuver and experiences significant bilateral hip pain on trial.

Based primarily on the patient’s history and physical exam findings, the PCP makes a presumptive diagnosis of PMR vs OA vs combined PMR with OA, orders an erythrocyte sedimentation rate (ESR) and basic rheumatologic marker panel, and starts the patient on prednisone 10 mg/d. Lab work comes back and reveals mildly elevated ESR with all other findings within normal limits. Two weeks later, the patient returns for her follow-up visit, walking without a walker for the first time in years.

PMR can be mistaken for OA

PMR is the most common inflammatory rheumatic disease in older patients.⁸ It is a debilitating illness with simple, effective treatment but has devastating consequences if missed or left untreated.⁹ PMR typically manifests in patients older than age 50, with a peak incidence at 80 years of age. It is also far more common in women.¹⁰

Approximately 80% of patients with PMR initially present to their PCP, often posing a diagnostic challenge to many clinicians.¹¹ Due to overlap in symptoms, the condition is often misdiagnosed as OA, a more common condition seen by PCPs. Also, there are no specific diagnostic tests for PMR. An elevated ESR can help confirm the diagnosis, but one-third of patients with PMR have a normal ESR.¹² Therefore, the diagnostic conundrum the physician faces is OA vs rheumatoid arthritis (RA), PMR, or another condition.

Continue to: The consequences...

The consequences of a missed and delayed PMR diagnosis range from seriously impaired quality of life to significantly increased risk of vascular events (eg, blindness, stroke) due to temporal arteritis.¹³ Early diagnosis is even more critical as the risk of a vascular event and death is highest during initial phases of the disease course.¹⁴

FPs often miss this Dx. A timely diagnosis relies almost exclusively on an accurate, thorough history and physical exam. However, PCPs often struggle to correctly diagnose PMR. According to a study by Bahlas and colleagues,¹⁵ the accuracy rate for correctly diagnosing PMR was 24% among a cohort of family physicians.

The differential diagnosis for PMR is broad and includes seronegative spondyloarthropathies, malignancy, Lyme disease, hypothyroidism, and both RA and OA.¹⁶ 

PCPs are extremely adept at correctly diagnosing RA, but not PMR. A study by Blaauw and colleagues¹⁷ comparing PCPs and rheumatologists found PCPs correctly identified 92% of RA cases but only 55% of PMR cases. When rheumatologists reviewed these same cases, they correctly identified PMR and RA almost 100% of the time.¹⁷ The difference in diagnostic accuracy between rheumatologists and PCPs suggests limited experience and gaps in fund of knowledge. 

Making the diagnosis. The diagnosis of PMR is often made on empiric response to corticosteroid treatment, but doing so based solely on a patient’s response is controversial.¹⁸ There are rare instances in which patients with PMR fail to respond to treatment. On the other hand, some inflammatory conditions that mimic or share symptoms with PMR also respond to corticosteroids, potentially resulting in erroneous confirmation bias.

Some classification criteria use rapid response to low-dose prednisone/­prednisolone (≤ 20 mg) to confirm the diagnosis,¹⁹ while other more recent guidelines no longer include this approach.²⁰ If PMR continues to be suspected after a trial of steroids is unsuccessful, the PCP can try another course of higher dose steroids or consult with Rheumatology.

Case 2 

A woman in her mid-40s presented to a PCP’s office with a chief complaint of dyspepsia and bloating.^a The patient was attending a meeting in New York City, and this was her first visit to this physician. The patient previously had been treated for these symptoms by her hometown PCP and gastroenterologist.

Continue to: A full history...

A full history and physical exam revealed a myriad of gastrointestinal (GI) complaints, such as diarrhea. But the PCP recalled a recent roundtable discussion on debiasing techniques specifically related to gynecologic disorders, including OC. Therefore, he decided to include OC in the differential diagnosis—something he would not routinely have done given the preponderance of GI symptoms. Despite the patient’s reluctance and time constraints, the PCP ordered a transvaginal ultrasound. Findings from the ultrasound study revealed stage II OC, which carries a good prognosis. The patient is currently undergoing treatment and was last reported as doing well.

Early signs of ovarian cancer can be chalked up to a “GI issue”

OC is the second most common gynecologic cancer²¹ and the fifth leading cause of cancer-related death²² in US women. Compared to other cancers, the prognosis for localized ­early-stage OC is surprisingly good, with a 5-year survival rate approaching 93%.²³ However, most disease is detected in later stages, and the 5-year survival rate drops to a low of 29%.²⁴

There remains no established screening protocol for OC. Fewer than a quarter of all cases are diagnosed in stage I, and detection of OC relies heavily on the physician’s ability to decipher vague symptomatology that overlaps with other, more common maladies. This poses an obvious diagnostic challenge and, not surprisingly, a high level of susceptibility to cognitive bias. 

Most PCPs correctly identified bloating as a key symptom of ovarian cancer; however, they weren’t as good at identifying less common symptoms, such as inability to finish a meal.

More than 90% of patients with OC present with some combination of the following symptoms prior to diagnosis: abdominal (77%), GI (70%), pain (58%), constitutional (50%), urinary (34%), and pelvic (26%).²⁵ The 3 most common isolated symptoms in patients with OC are abdominal bloating, decrease in appetite, and frank abdominal pain.²⁶ Patients with biopsy-confirmed OC experience these symptoms an average of 6 months prior to actual diagnosis.²⁷

Knowledge gaps play a role. Studies assessing the ability of health care providers to identify presenting symptoms of OC reveal specific knowledge gaps. For instance, in a survey by Gajjar and colleagues,²⁸ most PCPs correctly identified bloating as a key symptom of OC; however, they weren’t as good at identifying less common symptoms, such as inability to finish a meal and early satiety. Moreover, survey participants misinterpreted or missed GI symptoms as an important manifestation of early OC disease.²⁸ These specific knowledge gaps combine with physician errors in thinking, further obscuring and extending the diagnostic process. The point prevalence for OC is relatively low, and many PCPs only encounter a few cases during their entire career.²⁹ This low pre-test probability may also fuel the delay in diagnosis.

Watch for these forms of bias. Since nonspecific symptoms of early-stage OC resemble those of other more benign conditions, a form of anchoring error known as multiple alternatives bias can arise. In this scenario, clinicians investigate only 1 potential plausible diagnosis and remain focused on that single, often faulty, conclusion. This persists despite other equally plausible alternatives that arise as the investigation proceeds.²⁸

Affective error may also play a role in missed or delayed diagnosis. For example, a physician would prefer to diagnose and treat a common GI illness than consider OC. Another distortion involves outcome bias wherein the physician gives more significance to benign conditions such as irritable bowel syndrome because they have a more favorable outcome and clear treatment path. Physicians also favor these benign conditions because they encounter them more frequently than OC in the clinic setting. (This is known as availability bias.) Outcome bias and multiple alternatives bias can result in noninvestigation of symptoms and inefficient or improper management, leading to a delay in arriving at the correct diagnosis or anchoring on a plausible but incorrect diagnosis.

Continue to: An incorrect initial diagnostic...

An incorrect initial diagnostic path often triggers a cascade of subsequent errors. The physician orders additional unhelpful and expensive tests in an effort to characterize the suspected GI pathology. This then leads the physician to prematurely terminate the work-up and accept the most favored diagnosis. Lastly, sunk-cost fallacy comes into play: The physician has “invested” time and energy investigating a particular diagnosis and rather than abandon the presumed diagnosis, continues to put more time and effort in going down an incorrect diagnostic path. 

A series of failures. These biases and miscues have been observed in several studies. For example, a survey of 1725 women by Goff and colleagues³⁰ sought to identify factors related to delayed OC diagnosis. The authors found that the following factors were significantly associated with a delayed diagnosis: omission of a pelvic exam at initial presentation, a separate exploration of a multitude of collateral symptoms, a failure to order ultrasound/computed tomography/CA-125 test, and a failure to consider age as a factor (especially if the patient was outside the norm). 

Responses from the survey also revealed that physicians initially ordered work-ups related to GI etiology and only later considered a pelvic work-up. This suggests that well-known presenting signs and symptoms or a constellation of typical and atypical symptoms of OC often failed to trigger physician recognition. Understandably, patients presenting with menorrhagia or gynecologic complaints are more likely to have OC detected at an earlier stage than patients who present with GI or abdominal signs alone.³¹ table 2⁷ summarizes some of the cognitive biases seen in the diagnostic path of OC.

Case 3

A 56-year-old man is brought to the ED by his wife and children for evaluation of odd behavior and episodes of confusion. The patient recently had a negative neurologic work-up for transient ischemic attack and cerebrovascular accident and is admitted for further work-up. He reports visual hallucinations to nursing staff. Screening for memory problems shows no significant deficits. The patient in fact scored a 27 on the Mini–Mental State Examination, well within the normal range. The family notes that the patient has had difficulty with planning over the previous year and has not seemed like his “old self.” The patient has no history of psychosis, schizophrenia, bipolar disorder, or any other psychiatric illness.

While in the hospital, he becomes acutely upset by the hallucinations and is given haloperidol and lorazepam by house staff. In the morning, the patient exhibits severe signs of Parkinson disease that include rigidity and masked facies.

Given the patient’s poor response to haloperidol and continued confusion, the team consulted Neurology and Psychiatry. Gathering a more detailed history from the patient and family, the patient is given a diagnosis of classic LBD. The antipsychotic medications are stopped. The patient and his family receive education about LBD treatment and management, and the patient is discharged to outpatient care.

Psychiatric symptoms can be an early “misdirect” in cases of Lewy body disease

LBD, the second leading neurodegenerative dementia after Alzheimer disease (AD), affects 1.5 million Americans,³² representing about 10% of all dementia cases. LBD and AD overlap in 25% of dementia cases.³³ In patients older than 85 years, the prevalence jumps to 5% of the general population and 22% of all cases of dementia.³³ Despite its prevalence, a recent study showed that only 6% of PCPs correctly identified LBD as the primary diagnosis when presented with typical case examples.³² 

Continue to: 3 stages of presentation

3 stages of presentation. Unlike other forms of dementia, LBD typically presents first with psychiatric symptoms, then with cognitive impairment, and last with parkinsonian symptoms. Additionally, rapid eye movement sleep behavior disorder and often subtle elements of nonmemory cognitive impairment distinguish LBD from both AD and vascular dementia.³² The primary cognitive deficit in LBD is not in memory but in attention, executive functioning, and visuospatial ability.³⁴ Only in the later stages of the disease do patients exhibit gradual and progressive memory loss. 

Mistaken for many things. When evaluating patients exhibiting signs of dementia, it’s important to include LBD in the differential, with increased suspicion for patients experiencing episodes of psychosis or delirium. The uniqueness of LBD lies in its psychotic symptomatology, particularly during earlier stages of the disease. This feature helps distinguish LBD from both AD and vascular dementia. As seen in the case, LBD can also be confused with acute delirium.

The uniqueness of Lewy body dementia lies in its psychotic symptomatology, particularly during earlier stages of the disease.

Older adult patients presenting to the ED or clinic with visual hallucinations, delirium, and mental confusion may receive a false diagnosis of schizophrenia, medication- or substance-induced psychosis, Parkinson disease, or delirium of unknown etiology.³⁵ Unfortunately, LBD is often overlooked and not considered in the differential diagnosis. Due to underrecognition, patients may receive treatment with typical antipsychotics. The addition of a neuroleptic to help control the psychotic symptoms causes patients with LBD to develop severe extrapyramidal symptoms and worsening mental status,³⁶ leading to severe parkinsonian signs, which further muddies the diagnostic process. In addition, treatment for suspected Parkinson disease, including carbidopa-levodopa, has no benefit for patients with LBD and may increase psychotic symptoms.³⁷

First-line treatment for LBD includes psychoeducation for the patient and family, cholinesterase inhibitors (eg, rivastigmine), and avoidance of high-potency antipsychotics, such as haloperidol. Although persistent hallucinations and psychosis remain difficult to treat in LBD, low-dose quetiapine is 1 option. Incorrectly diagnosing and prescribing treatment for another condition exacerbates symptoms in this patient population.

Case 4

A 36-year-old Hispanic woman presents to the PCP for her annual physical exam. The patient’s medical record shows 2 previous office visits over the past 2 years—an annual physical exam and an office visit for the flu vaccine. The patient is highly accomplished in her profession, working as a certified public accountant for a major corporation. She is a nonsmoker and reports only casual social drinking and no recreational drug use. The patient is slightly overweight for her height but is otherwise healthy. Previous lab studies are within normal limits.

The patient has been experiencing chronic pain for the past few years after a motor vehicle accident. She has seen a physiatrist and another provider, both of whom found no “objective” causes of her chronic pain. They started the patient on sertraline for depression and an analgesic, both of which were ineffective.

The patient likes to exercise at a gym twice a week by doing light cardio (treadmill) exercise and light weightlifting. Lately, however, she has been unable to exercise due to the pain. At this visit, she mentions having low energy, poor sleep, frequent fatigue, and generalized soreness and pain in multiple areas of her body. The PCP recognizes the patient’s presenting symptoms as significant for FM and starts her on pregabalin and hydrotherapy, with positive results.

Continue to: Fibromyalgia skepticism may lead to a Dx of depression

FM, the second most common disorder seen in rheumatologic practice after OA, is estimated to affect approximately 1 in 20 patients (approximately 5 million Americans) in the primary care setting.^38,39 The condition has a high female-to-male preponderance (3.4% vs 0.5%).⁴⁰ While the primary symptom of FM is chronic pain, patients commonly present with fatigue and sleep disturbance.⁴¹ Comorbid conditions include headaches, irritable bowel syndrome, and mood disturbances (most commonly anxiety and depression). 

Patients with fibromyalgia, who are often otherwise healthy, often present multiple times to the same PCP with a chief complaint of chronic pain.

Several studies have explored reasons for the misdiagnosis and underdiagnosis of FM. One important factor is ongoing skepticism among some physicians and the public, in general, as to whether FM is a real disease. This issue was addressed by a study by White and colleagues,⁴² who estimated and compared the point prevalence of FM and related disorders in Amish vs non-Amish adults. The authors hypothesized that if litigation and/or compensation availability have a major impact on FM prevalence, then there would be a near zero prevalence of FM in the Amish community. And yet, researchers found an overall age- and sex-adjusted FM prevalence of 7.3% (95% CI; 5.3%-9.7%); this was both statistically greater than zero (P < .0001) and greater than 2 control populations of non-Amish adults (both P < .05).

Many physicians consider FM fundamentally an emotional disturbance, and the high preponderance of FM in female patients may contribute to this misconception as reports of pain and emotional distress by women are often dismissed as hysteria.⁴³ Physicians often explore the emotional aspects of FM, incorrectly diagnosing patients with depression and subsequently treating them with a psychotropic drug.³⁹ Alternatively, they may focus on the musculoskeletal presentations of FM and prescribe analgesics or physical therapy, both of which do little to alleviate FM. 

To make the correct diagnosis of FM, the American College of Rheumatology created a specific set of criteria in 1990, which was updated in 2010.⁴⁴ For a diagnosis of FM, a patient must have at least a 3-month history of bilateral pain above and below the waist and along the axial skeletal spine. Although not included in the updated 2010 criteria, many clinicians continue to check for tender points, following the 1990 criteria requiring the presence of 11 of 18 points to make the diagnosis.

At least 3 cognitive biases relating to FM apply: anchoring, availability, and fundamental attribution error (see table 3).⁷ Anchoring occurs when the PCP settles on a psychiatric diagnosis of exaggerated pain syndrome, muscle overuse, or OA and fails to explore alternative etiology. Availability bias may obscure the true diagnosis of FM. Since PCPs see many patients with RA or OA, they may overlook or dismiss the possibility of FM. Attribution error happens when physicians dismiss the complaints of patients with FM as merely due to psychological distress, hysteria, or acting out.⁴³

Patients with FM, who are often otherwise healthy, often present multiple times to the same PCP with a chief complaint of chronic pain. These repeat presentations can result in compassion fatigue and impact care. As Aloush and colleagues⁴⁰ noted in their study, “FM patients were perceived as more difficult than RA patients, with a high level of concern and emotional response. A high proportion of physicians were reluctant to accept them because they feel emotional/psychological difficulties meeting and coping with these patients.”In response, patients with undiagnosed FM or inadequately treated FM may visit other PCPs, which may or may not result in a correct diagnosis and treatment.

We can do better

Primary care physicians face the daunting task of diagnosing and treating a wide range of common conditions while also trying to recognize less-common conditions with atypical presentations—all during a busy clinic workday. Nonetheless, we should strive to overcome internal (eg, cognitive bias and fund-of-knowledge deficits) and external (eg, time constraints, limited resources) pressures to improve diagnostic accuracy and care.

Each of the 4 disease states we’ve discussed have high rates of missed and/or delayed diagnosis. Each presents a unique set of confounders: PMR with its overlapping symptoms of many other rheumatologic diseases; OC with its often vague and misleading GI symptomatology; LBD with overlapping features of AD and Parkinson disease; and FM with skepticism. As gatekeepers to health care, it falls on PCPs to sort out these diagnostic dilemmas to avoid medical errors. Fundamental knowledge of each disease, its unique pathophysiology and symptoms, and varying presentations can be learned, internalized, and subsequently put into clinical practice to improve patient outcomes.

CORRESPONDENCE
Paul D. Rosen MD, Brooklyn Hospital Center, Department of Family Medicine, 121 Dekalb Avenue, Brooklyn, New York 11201; [email protected]

The hormones thyroxine (T4) and triiodothyronine (T3), produced by the thyroid gland, are crucial for maintaining metabolism. A deficit of thyroid hormone production—hypothyroidism—is a common endocrine disorder seen in primary care.

Many patients with hypothyroidism do not achieve optimal treatment goals or see an improvement in their quality of life.

Although the diagnosis and management of hypothyroidism are considered straightforward, many patients with hypothyroidism do not achieve optimal treatment goals or see an improvement in their quality of life. In this article, we address the questionable utility of screening; outline the diagnostic approach, including the central role of laboratory testing; and explain why treatment requires a precise approach to address the range of patient types.

Epidemiology and classification

Estimates are that almost 5% of Americans 12 years or older have hypothyroidism; older people and women are more likely to develop the condition. ¹ In the US National Health and Nutrition Examination Survey (NHANES III) of 13,344 people without known thyroid disease or a family history, hypothyroidism was found in 4.6% (overt [clinical] in 0.3% and subclinical in 4.3%); 11% had a high serum thyroid peroxidase antibody level, which increases their risk of hypothyroidism, and is treated the same as hypothyroidism of other causes; and, overall, lower serum thyroid-stimulating hormone (TSH) levels were seen in Blacks, compared to Whites and Mexican Americans.¹

Copyright Steve and Myriam Oh

Primary hypothyroidism accounts for > 95% of cases of hypothyroidism, representing a failure of the thyroid gland to produce sufficient hormone. It has been shown that, in iodine-replete countries such as the United States, the prevalence of spontaneous hypothyroidism is 1% to 2%, and it is 10 times more common in women.^2,3

Central hypothyroidism is caused by insufficient stimulation of the thyroid gland by TSH, due to pituitary (secondary hypothyroidism) or hypothalamic (tertiary hypothyroidism) disease and is estimated to occur in 1 in every 20,000 to 80,000 people in the general population.⁴

How does hypothyroidism manifest?

Signs and symptoms. Manifestations of hypothyroidism range from life-threatening to minimal or no clinical signs and symptoms (TABLE W1). Signs and symptoms of low thyroid function vary by the degree of hypothyroidism at presentation.

Common signs and symptoms of low thyroid function include fatigue, weight gain, dry skin, brittle hair, hair loss, morning stiffness, muscle aches, joint pain, cold intolerance, diffuse headache, constipation, difficulty concentrating, low libido, depression, and menstrual irregularities. On physical examination, a patient might present with bradycardia, hypotension, hypothermia with slow speech or movement, coarse facial appearance, goiter, diffuse hair loss, cold hands and feet, and a prolonged Achilles tendon reflex.⁵ Skin findings, such as keratosis pilaris, palmoplantar keratoderma (thickening of the skin), and pityriasis rubra pilar, can be associated with autoimmune hypothyroidism.^6,7

Continue to: Carpal tunnel syndrome...

Carpal tunnel syndrome, plantar fasciitis, infertility or miscarriage, dyspepsia, and small intestinal bacterial overgrowth can be associated with hypothyroidism; thyroid function should therefore be assessed in patients who have any of these conditions, along with other signs and symptoms of low thyroid function.^8,9 A patient with severe hypothyroidism might present with hemodynamic instability, pericardial or pleural effusion, and myxedema coma.¹⁰

Clues in the history and from the lab. A history of radiation to the head, neck, or chest area and a history or family history of autoimmune disorders are risk factors for autoimmune thyroid disease.^11,12 Laboratory findings can include markers of oxidative stress, such as elevated levels of low-density lipoprotein cholesterol and serum malondialdehyde.^13-15

Screening and diagnosis

Screening. The US Preventive Services Task Force has asserted that evidence is insufficient by which to evaluate the benefits and risks of routine screening for thyroid dysfunction in nonpregnant, asymptomatic adults.¹⁶ According to the American Thyroid Association and the American Association of Clinical Endocrinologists, screening should be considered in high-risk patients, including those who take medication that affects thyroid function or the results of thyroid hormone assays (TABLE W2, available at mdedge.com/familymedicine).^17-20

Screening inpatients is challenging and usually not recommended unless thyroid disease is strongly suspected. This is because changes in the levels of thyroid hormones, binding proteins, and the TSH concentration can occur in severe nonthyroidal illness; in addition, assay interference by antibodies and other substances can affect thyroid hormone measurement.²¹

Testing strategy. Generally, screening and diagnosis of hypothyroidism are based primarily on laboratory testing, because signs and symptoms are nonspecific (FIGURE 1⁵). A serum TSH level is usually the initial test when screening for thyroid dysfunction. A normal serum TSH value ranges from 0.5-5.0 mIU/L.

A patient with severe hypothyroidism might present with hemodynamic instability, pericardial or pleural effusion, or myxedema coma.

When an abnormal serum TSH level is found, further tests can be performed to investigate, including a serum free thyroxine (FT4) test. (Our preference is to order TSH and FT4 assays simultaneously to facilitate and confirm the diagnosis.) An FT4 test measures the amount of unattached, or free, thyroxine in blood by immunoassay. A normal FT4 value usually ranges from 0.7-1.9 ng/dL.

The combination of a high TSH level and a low FT4 level could be an indication of an underactive thyroid gland (ie, clinical or overt hypothyroidism). Milder, subclinical hypothyroidism is characterized by a ­higher-than-normal TSH level but a normal FT4 level.²² Central (secondary) hypothyroidism is characterized by a low serum FT4 level and a serum TSH level that can be below the reference range, low normal, or even slightly high.⁴

Continue to: These measurements...

These measurements must be interpreted within the context of the laboratory-specific normal range for each test. Third-generation serum TSH assays are more sensitive and specific than serum FT4 measurements for hypothyroidism. FT4 is usually measured by automated analogue immunoassay, which generally provides reliable results; abnormal binding proteins or other interferences occur in some patients, however, resulting in reporting of a falsely high, or falsely low, FT4 level. In such cases, FT4 by direct dialysis, or total T4, can be measured for further evaluation. In primary care, you are most likely to encounter primary hypothyroidism; secondary (central) hypothyroidism is much rarer (< 5% of cases).⁴

The ins and outs of treatment

For most patients, hypothyroidism is a permanent disorder requiring lifelong thyroid hormone replacement therapy—unless the disease is transient (ie, painless or subacute thyroiditis); reversible, because it is caused by medication; or responsive to medical intervention that addresses the underlying autoimmune condition.¹⁹ Goals of treatment (Figure 2^5,23) are to:

• normalize the TSH level to 0.5-5.0 mIU/L (the main goal), with an age-related shift toward a higher TSH goal in older patients (and an upper limit of normal of 7.5 mIU/L in patients who are ≥ 80 years of age)²⁰
• restore the euthyroid state
• relieve symptoms
• reduce any goiter
• avoid overtreatment (iatrogenic thyrotoxicosis).

Desiccated thyroid extract (DTE), developed in the late 1880s and made from the dried thyroid gland of pigs, sheep, or cows, was the earliest treatment for hypothyroidism. The use of DTE has declined since the introduction of synthetic thyroxine (T4, or levothyroxine [here, referred to as LT4]), which is now the standard treatment.^20-22 LT4 is deiodinated in peripheral tissues to form T3, the active thyroid hormone; this process accounts for 80% of total T3 production daily.²⁴

LT4 formulations. LT4 is commercially available in tablet, soft-gel, and liquid preparations. Most patients are treated with the tablet; the soft-gel capsule or liquid is an option for patients who absorb the tablet poorly (because of atrophic gastritis, celiac disease, or gluten sensitivity or because they are post bariatric surgery). Increasing the dosage of the tablet form of LT4, with ongoing TSH monitoring, is more cost effective than moving to an alternative preparation.

If a switch of LT4 formulation is made (ie, from one manufacturer to another), test the serum TSH level to ensure that the therapeutic goal is being reached. Also, in our experience, it is best to prescribe a brand-name preparation of levothyroxine, not a generic, whenever possible, due to the variability in generic formulations and the potential presence of other (inert) ingredients.²⁵

Dosing (TABLE 3^20,23). The average full replacement dosage of LT4 for a young, healthy adult is approximately 1.6 mcg/kg/d. Older patients (> 65 years) or those with coronary artery disease (CAD) should be started on a lower dosage (25-50 mcg/d) and titrated to goal accordingly.

LT4 (tablets, soft-gel capsules, or liquid) should be administered on an empty stomach, with water only, 30 to 60 minutes before breakfast. Medications that interfere with LT4 absorption (eg, bile acid resins, calcium carbonate, ferrous sulfate) should be taken several hours after LT4. For patients who cannot take LT4 in the morning, taking it at bedtime (≥ 2-3 hours after the last meal) is acceptable.

Continue to: Monitoring and titrating

Monitoring and titrating. Hypothyroid symptoms usually improve after 2 or 3 weeks of LT4 treatment; in severe hypothyroidism, complete recovery might take months. Approximately 6 weeks after LT4 therapy is initiated, serum TSH should be measured. After assessing whether administration of LT4 at the starting dosage is appropriate, that dosage can be increased, or decreased, every 4 to 6 weeks until the TSH goal is reached. Once the patient is maintained at a given dosage, measure serum TSH once a year—more often if there is an abnormal result or a change in the patient’s health status.²³

Adverse effects of LT4 therapy are rare, unless over-replacement occurs. Rarely, patients have an allergy to the dye or an excipient (filler) in the tablet.^26-28 The white, 50-mcg tablets can be given safely to patients with dye sensitivity. For those who have an allergy to an excipient (except gelatin) or gluten intolerance, the LT4 soft-gel capsule or liquid preparation (Tirosint) can be prescribed.

Pure LT4, in a capsule made from vegetable sources, can be ordered through a compounding pharmacy for patients who are allergic to animal products.

Anemia, especially iron-deficiency anemia, can cause intolerance to LT4 therapy; in such patients, lowering the starting dosage and treating anemia are indicated.²⁹

Persistent symptoms (despite a normal TSH level). Because many hypothyroid symptoms are nonspecific, patients might come to think that their LT4 dosage is inadequate if they feel tired or gain weight. Persistent hypothyroid symptoms despite a normal serum TSH level might be due to (1) the inability of LT4 therapy to restore tissue thyroid hormone levels to normal or (2) other variables unrelated to hypothyroidism, including disorders associated with inflammation or autoimmune disease, certain medications, diet, lifestyle, and environmental toxins.

These patients might benefit from a detailed history to identify other causes and a switch to either LT4 + liothyronine (LT3; synthetic T3) combination therapy or DTE^26,30-33 (TABLE 4³⁴), although a beneficial effect of LT4 + LT3 therapy was not seen in several studies.^35,36 Over-replacement with LT4 should be discouraged, due to concerns about thyrotoxicosis and its complications (eg, atrial fibrillation, accelerated bone loss).

DTE and LT4 + LT3. Use of DTE has decreased since the 1970s, when LT4 became the therapy of choice. Subsequently, anecdotal evidence emerged that some patients did not feel well on LT4 and preferred to return to DTE.^32,33

Continue to: Several clinical trials...

Several clinical trials addressed the question of whether residual symptoms could be resolved through LT4 + LT3 combination therapy^31-39 (TABLE 4³⁴), but evidence of any consistent superiority of combination therapy was not demonstrated.^35-39 In selected cases, patients might prefer the combination approach.^31,33,39 The quality of life of hypothyroid patients was found to be similarly improved with LT4 or DTE, but the latter was associated with modest weight loss (approximately 4 lbs); nearly 50% of study patients preferred treatment with DTE over LT4.³³ A follow-up study did not confirm weight loss with DTE, however.³⁴

When LT4 monotherapy and LT4 + LT3 combination therapy were compared, results were mixed^31-39; responsiveness to therapy containing LT3 might therefore depend on multiple variables, including genetic background, nutritional and lifestyle factors, stress, presence of comorbidities and autoimmune disorders, and other unidentified or poorly defined variables.^40-48

Consider screening for hypothyroidism in patients who take medication that affects either their thyroid function or the results of thyroid hormone assays.

Although combination therapy and DTE are not generally recommended over LT4 monotherapy, they might offer better options for patients who are still symptomatic when being treated with LT4 only: In a randomized, double-blind, crossover study that compared LT4 with DTE and with LT4 + LT3, one-third of the most highly symptomatic patients who had low scores on mood, cognitive, and quality-of-life assessments improved significantly after they were switched to combination therapy or DTE.³⁴

The 2014 American Thyroid Association guidelines²⁴ do not support routine use of LT4 + LT3 in hypothyroid patients who have residual symptoms after LT4 monotherapy; however, a therapeutic trial of LT4 + LT3, while maintaining a normal serum TSH, is reasonable in selected patients. Candidates for DTE or LT4 + LT3 might include patients who do not feel well on LT4 monotherapy, are post thyroidectomy or post radioiodine therapy, or have a low serum T3 level. DTE and combination therapy are discouraged in older patients, patients who have underlying CAD, and pregnant patients.

Special treatment circumstances

A number of patient variables have the potential to alter management strategies for hypothyroidism.^{18,20,23,40,49-53}

Age, comorbidity. Older patients (> 65 years) and patients with cardiopulmonary disease or CAD should be treated with LT4, 25 to 50 mcg/d, initially; that dosage can be titrated upward by 12.5 to 25 mcg/d every 4 to 6 weeks until the TSH goal is reached—preferably, in the range of 4 to 8 mIU/L. An increase in the dosage of LT4 might be required in the presence of malabsorption (eg, gastrointestinal disorders, celiac disease) and in nephrotic syndrome.^18,20,23

Body weight. A decrease in the dosage of LT4 might be indicated in the setting of significant weight loss (> 10% body weight).²³

Continue to: Co-pharmacy

Co-pharmacy. An increase in the dosage of LT4 might be required when other drugs (eg, phenytoin, phenobarbital, rifampin, and carbamazepine) have led to an increased rate of thyroid hormone metabolism. A decrease in the dosage of LT4 might be necessary after initiation of androgen therapy.²³

Pregnancy. Women with pre-e­xisting hypothyroidism require an increase of 25% to 50% in their LT4 dosage during pregnancy to maintain a TSH level in the recommended pregnancy reference range. Thyroid function should be monitored every 4 to 6 weeks to ensure that the TSH target for each trimester is reached (first trimester, 0.1-4 mIU/L; second trimester, 0.2-4 mIU/L; third trimester, 0.3-4 mIU/L). Postpartum, LT4 can be reduced to the prepartum dosage; TSH should be checked every 4 to 6 weeks to maintain the TSH goal.²³

Estrogen therapy. Hypothyroid women who are receiving estrogen therapy might require an increase in their LT4 dosage because serum thyroxine-binding globulin levels are increased by estrogens or through other mechanisms that have not been identified.²³

Surgical candidacy. Observational studies show few adverse outcomes in surgical patients with mild (subclinical) hypothyroidism or moderate hypothyroidism; however, the risk of adverse surgical outcome might be increased in patients with severe hypothyroidism. For patients in whom surgery is planned and who have:

• subclinical hypothyroidism (elevated TSH and normal FT4), we recommend that surgery—urgent or elective—not be posptoned but proceed.
• moderate (overt) hypothyroidism who require urgent surgery, we recommend not postponing surgery, even though minor perioperative complications might develop. Such patients should be treated with LT4 as soon as the diagnosis for which surgery is required has been made. Alternatively, when moderate hypothyroidism is discovered in a patient who is being evaluated for elective surgery, we recommend postponing surgery until the euthyroid state is restored.
• severe hypothyroidism (myxedema coma [discussed in a bit]; severe clinical symptoms of chronic hypothyroidism, such as altered mental status, pericardial effusion, or heart failure; or a very low level of T4), surgery should be delayed until hypothyroidism has been treated. When emergency surgery is required for a severely hypothyroid patient, they should be treated with LT4 as soon as the diagnosis for which surgery is indicated has been made. When emergency surgery must be performed in a patient with myxedema coma, we recommend treatment with LT4 + LT3, rather than LT4 alone, often administered intravenously because LT4 is poorly absorbed in these patients.

Nonadherence. For patients who do not take LT4 regularly or do not respond to efforts to improve adherence, LT4 can be given weekly, instead of daily, although this interval is not ideal. Weekly dosing should not be used in older patients with CAD.²³

Thyroid cancer. Patients who are post total thyroidectomy for thyroid cancer need to take LT4 to treat hypothyroidism and to prevent recurrence of thyroid cancer. The goal TSH level should be based on the cancer stage and risk of recurrence and should be monitored by an endocrinologist.

Myxedema coma. This medical emergency has high mortality. Myxedema coma occurs when severe hypothyroidism leads to any, or a combination, of the following: diminished mentation; hypothermia; bradycardia; hyponatremia; hypotension; cardiovascular, respiratory, and gastrointestinal dysfunction; and renal insufficiency. LT4, LT3, and glucocorticoids should be administered intravenously and the patient monitored closely—preferably in consultation with an endocrinologist.

Continue to: When to seek consultation

A patient with hypothyroidism should be referred to Endocrinology if they are < 18 years of age, pregnant, unresponsive to therapy, or have cardiac disease, coexisting endocrine disease, suspected myxedema coma, goiter or thyroid nodules, or a structural thyroid abnormality.

What we know about nutrition and hypothyroidism

Although it is commonly recognized that iodine is essential for production of thyroid hormone, other nutritional factors might contribute to proper production of thyroid hormones, including:

• adequate intake of iron, tyrosine, selenium, zinc, and vitamins E, B₂, B₃, B₆, C, and D^44,45
• selenium and zinc, which increase conversion of T4 to T3 and might be important in the management of hypothyroid patients^40,46
• vitamin A, zinc, and regular exercise, which have been shown to improve cellular sensitivity to thyroid hormones.

Low iron stores can contribute to persistent symptoms and poor quality of life in patients with hypothyroidism, despite their being treated according to guidelines.^29,47

Despite what is known about these nutritional connections, there is insufficient evidence that improving nutrition can reverse hypothyroidism.

Prevention

Prevention of hypothyroidism should take into account variables that affect or inhibit thyroid function, such as stress, infection (eg, Epstein-Barr virus), excessive fluoride intake, toxins (eg, pesticides, solvents, mercury, cadmium, and lead), autoimmune disease (eg, celiac disease), and food sensitivity.^54,55 Oxidative stress can also cause thyroid impairment.^40-48,54-58

Otherwise, there are, at present, no effective strategies for preventing thyroid disorders.

Subclinical hypothyroidism: Elusive management target

Subclinical hypothyroidism is defined as a normal serum FT4 level in the presence of an elevated serum TSH level. The prevalence of subclinical hypothyroidism varies from 3% to 15%, depending on the population studied; a higher incidence has been noted in women and older people.⁵⁹ In the NHANES III,¹ which excluded people with previously diagnosed thyroid disease, the incidence of subclinical hypothyroidism was 4.3%.

Continue to: Causes of subclinical...

Causes of subclinical hypothyroidism are the same as those of overt hypothyroidism, and include Hashimoto disease. The combination of an elevated TSH level and a normal FT4 level is associated with disorders characterized by protein-binding variations (eg, pregnancy, genetic disorders, drugs), TSH-secreting pituitary adenoma, class II and III obesity (respectively, body mass index, ≥ 35 but < 40 and ≥ 40), and assay variability.^49,51

Lab diagnosis: Fraught with difficulty

The serum TSH level and either the total T4 level or the FT4 level should be measured to make a diagnosis of subclinical hypothyroidism. Most laboratories use a 1-step analogue immunoassay to determine free thyroid hormones; protein-binding variations can thus affect measurement of FT4.

Increasing the dosage of the tablet form of LT4 (while monitoring the TSH level), is more cost effective than moving to an alternative preparation, such as a softgel or liquid.

Several scenarios that can result in inaccurate measurement of FT4 by radioimmunoassay include genetic disorders that affect binding proteins; pregnancy; use of certain drugs, including heparin, furosemide, antiepileptic agents, salicylate, ferrous sulfate, and cholesterol-binding resins; and some medical conditions, including cardiac surgery, critical illness, and renal failure. Variables that inhibit proper production of thyroid hormones—stress, infection, fluoride (an iodine antagonist), toxins (pesticides, mercury, cadmium, lead) and autoimmune conditions, such as celiac disease—should be considered when attempting to determine the cause of subclinical hypothyroidism.

Liquid chromatography–mass spectrometry measurement of thyroid hormones might be more accurate than immunoassay.⁵³ Measuring serum total T4 and FT4 by dialysis, free from interfering proteins, might also be useful when measurement of FT4 by immunoassay is affected by binding-protein variations.

Prescribe a brand-name preparation of levothyroxine whenever possible; generic formulations might have variable potency or contain other ingredients.

Features of subclinical hypothyroidism

Most patients who have subclinical hypothyroidism and a serum TSH level < 10 mIU/L are asymptomatic. Some might have nonspecific symptoms of hypothyroidism, however, such as reduced quality of life, poor cognitive function, and poor memory—symptoms that do not typically correlate with the serum TSH level.

An increase in the dosage of LT4 might be required in pregnancy or when weight gain is significant (> 10% of body weight).

It has been suggested that some elderly people normally have a higher level of serum TSH, and that they might have even a prolonged lifespan.⁵¹ Additionally, it has been shown that, in nonpregnant adult patients with subclinical hypothyroidism and a serum TSH level of 4.5 to 10 mIU/L, treatment with LT4 was not associated with improvement in thyroid-related symptoms or general quality of life.⁵²

Treat, or don't treat, subclinical hypothyroidism?

It is well accepted that the goal of therapy in hypothyroid patients is to normalize the serum TSH level; however, the American Thyroid Association and the American Association of Clinical Endocrinology recommend starting LT4 in patients with a serum TSH level ≥ 10 mIU/L (TABLE 5).^59-62 The principal reason for not treating subclinical hypothyroidism is the lack of benefit in reducing the risk of cardiovascular morbidity and mortality when the TSH level is between 7.5 and 10 mIU/L.⁶²

Continue to: Routine treatment

Routine treatment of patients with a serum TSH level of 4.5 to 10 mIU/L remains controversial. When TSH is 7.0 to 9.9 mIU/L, treatment is recommended for (1) patients < 65 years and (2) for older patients (> 65 years) only when there are convincing hypothyroid symptoms because of concern about unintended overtreatment.

When the TSH level is anywhere above the upper limit of normal to 6.9 mIU/L, treatment is recommended for patients < 65 years old, patients who have a high titer of thyroid peroxidase antibodies, and patients with goiter—but not for patients > 65 years (and, especially, not for octogenarians) because their upper limit of normal could be as high as 6 to 8 mIU/L, especially if they are otherwise healthy.

Treatment should be considered for women with subclinical hypothyroidism who are trying to conceive or experiencing an infertility problem.

For patients with subclinical hypothyroidism who are not being treated, monitor thyroid function every 6 to 12 months by testing TSH and FT4.

CORRESPONDENCE
Thanh D. Hoang, DO, Division of Endocrinology, Walter Reed National Military Medical Center, 8901 Wisconsin Avenue, Bethesda, MD 20889; [email protected]

The hormones thyroxine (T4) and triiodothyronine (T3), produced by the thyroid gland, are crucial for maintaining metabolism. A deficit of thyroid hormone production—hypothyroidism—is a common endocrine disorder seen in primary care.

Many patients with hypothyroidism do not achieve optimal treatment goals or see an improvement in their quality of life.

Although the diagnosis and management of hypothyroidism are considered straightforward, many patients with hypothyroidism do not achieve optimal treatment goals or see an improvement in their quality of life. In this article, we address the questionable utility of screening; outline the diagnostic approach, including the central role of laboratory testing; and explain why treatment requires a precise approach to address the range of patient types.

Epidemiology and classification

Estimates are that almost 5% of Americans 12 years or older have hypothyroidism; older people and women are more likely to develop the condition. ¹ In the US National Health and Nutrition Examination Survey (NHANES III) of 13,344 people without known thyroid disease or a family history, hypothyroidism was found in 4.6% (overt [clinical] in 0.3% and subclinical in 4.3%); 11% had a high serum thyroid peroxidase antibody level, which increases their risk of hypothyroidism, and is treated the same as hypothyroidism of other causes; and, overall, lower serum thyroid-stimulating hormone (TSH) levels were seen in Blacks, compared to Whites and Mexican Americans.¹

Copyright Steve and Myriam Oh

Primary hypothyroidism accounts for > 95% of cases of hypothyroidism, representing a failure of the thyroid gland to produce sufficient hormone. It has been shown that, in iodine-replete countries such as the United States, the prevalence of spontaneous hypothyroidism is 1% to 2%, and it is 10 times more common in women.^2,3

Central hypothyroidism is caused by insufficient stimulation of the thyroid gland by TSH, due to pituitary (secondary hypothyroidism) or hypothalamic (tertiary hypothyroidism) disease and is estimated to occur in 1 in every 20,000 to 80,000 people in the general population.⁴

How does hypothyroidism manifest?

Signs and symptoms. Manifestations of hypothyroidism range from life-threatening to minimal or no clinical signs and symptoms (TABLE W1). Signs and symptoms of low thyroid function vary by the degree of hypothyroidism at presentation.

Common signs and symptoms of low thyroid function include fatigue, weight gain, dry skin, brittle hair, hair loss, morning stiffness, muscle aches, joint pain, cold intolerance, diffuse headache, constipation, difficulty concentrating, low libido, depression, and menstrual irregularities. On physical examination, a patient might present with bradycardia, hypotension, hypothermia with slow speech or movement, coarse facial appearance, goiter, diffuse hair loss, cold hands and feet, and a prolonged Achilles tendon reflex.⁵ Skin findings, such as keratosis pilaris, palmoplantar keratoderma (thickening of the skin), and pityriasis rubra pilar, can be associated with autoimmune hypothyroidism.^6,7

Continue to: Carpal tunnel syndrome...

Carpal tunnel syndrome, plantar fasciitis, infertility or miscarriage, dyspepsia, and small intestinal bacterial overgrowth can be associated with hypothyroidism; thyroid function should therefore be assessed in patients who have any of these conditions, along with other signs and symptoms of low thyroid function.^8,9 A patient with severe hypothyroidism might present with hemodynamic instability, pericardial or pleural effusion, and myxedema coma.¹⁰

Clues in the history and from the lab. A history of radiation to the head, neck, or chest area and a history or family history of autoimmune disorders are risk factors for autoimmune thyroid disease.^11,12 Laboratory findings can include markers of oxidative stress, such as elevated levels of low-density lipoprotein cholesterol and serum malondialdehyde.^13-15

Screening and diagnosis

Screening. The US Preventive Services Task Force has asserted that evidence is insufficient by which to evaluate the benefits and risks of routine screening for thyroid dysfunction in nonpregnant, asymptomatic adults.¹⁶ According to the American Thyroid Association and the American Association of Clinical Endocrinologists, screening should be considered in high-risk patients, including those who take medication that affects thyroid function or the results of thyroid hormone assays (TABLE W2, available at mdedge.com/familymedicine).^17-20

Screening inpatients is challenging and usually not recommended unless thyroid disease is strongly suspected. This is because changes in the levels of thyroid hormones, binding proteins, and the TSH concentration can occur in severe nonthyroidal illness; in addition, assay interference by antibodies and other substances can affect thyroid hormone measurement.²¹

Testing strategy. Generally, screening and diagnosis of hypothyroidism are based primarily on laboratory testing, because signs and symptoms are nonspecific (FIGURE 1⁵). A serum TSH level is usually the initial test when screening for thyroid dysfunction. A normal serum TSH value ranges from 0.5-5.0 mIU/L.

A patient with severe hypothyroidism might present with hemodynamic instability, pericardial or pleural effusion, or myxedema coma.

When an abnormal serum TSH level is found, further tests can be performed to investigate, including a serum free thyroxine (FT4) test. (Our preference is to order TSH and FT4 assays simultaneously to facilitate and confirm the diagnosis.) An FT4 test measures the amount of unattached, or free, thyroxine in blood by immunoassay. A normal FT4 value usually ranges from 0.7-1.9 ng/dL.

The combination of a high TSH level and a low FT4 level could be an indication of an underactive thyroid gland (ie, clinical or overt hypothyroidism). Milder, subclinical hypothyroidism is characterized by a ­higher-than-normal TSH level but a normal FT4 level.²² Central (secondary) hypothyroidism is characterized by a low serum FT4 level and a serum TSH level that can be below the reference range, low normal, or even slightly high.⁴

Continue to: These measurements...

These measurements must be interpreted within the context of the laboratory-specific normal range for each test. Third-generation serum TSH assays are more sensitive and specific than serum FT4 measurements for hypothyroidism. FT4 is usually measured by automated analogue immunoassay, which generally provides reliable results; abnormal binding proteins or other interferences occur in some patients, however, resulting in reporting of a falsely high, or falsely low, FT4 level. In such cases, FT4 by direct dialysis, or total T4, can be measured for further evaluation. In primary care, you are most likely to encounter primary hypothyroidism; secondary (central) hypothyroidism is much rarer (< 5% of cases).⁴

The ins and outs of treatment

For most patients, hypothyroidism is a permanent disorder requiring lifelong thyroid hormone replacement therapy—unless the disease is transient (ie, painless or subacute thyroiditis); reversible, because it is caused by medication; or responsive to medical intervention that addresses the underlying autoimmune condition.¹⁹ Goals of treatment (Figure 2^5,23) are to:

• normalize the TSH level to 0.5-5.0 mIU/L (the main goal), with an age-related shift toward a higher TSH goal in older patients (and an upper limit of normal of 7.5 mIU/L in patients who are ≥ 80 years of age)²⁰
• restore the euthyroid state
• relieve symptoms
• reduce any goiter
• avoid overtreatment (iatrogenic thyrotoxicosis).

Desiccated thyroid extract (DTE), developed in the late 1880s and made from the dried thyroid gland of pigs, sheep, or cows, was the earliest treatment for hypothyroidism. The use of DTE has declined since the introduction of synthetic thyroxine (T4, or levothyroxine [here, referred to as LT4]), which is now the standard treatment.^20-22 LT4 is deiodinated in peripheral tissues to form T3, the active thyroid hormone; this process accounts for 80% of total T3 production daily.²⁴

LT4 formulations. LT4 is commercially available in tablet, soft-gel, and liquid preparations. Most patients are treated with the tablet; the soft-gel capsule or liquid is an option for patients who absorb the tablet poorly (because of atrophic gastritis, celiac disease, or gluten sensitivity or because they are post bariatric surgery). Increasing the dosage of the tablet form of LT4, with ongoing TSH monitoring, is more cost effective than moving to an alternative preparation.

If a switch of LT4 formulation is made (ie, from one manufacturer to another), test the serum TSH level to ensure that the therapeutic goal is being reached. Also, in our experience, it is best to prescribe a brand-name preparation of levothyroxine, not a generic, whenever possible, due to the variability in generic formulations and the potential presence of other (inert) ingredients.²⁵

Dosing (TABLE 3^20,23). The average full replacement dosage of LT4 for a young, healthy adult is approximately 1.6 mcg/kg/d. Older patients (> 65 years) or those with coronary artery disease (CAD) should be started on a lower dosage (25-50 mcg/d) and titrated to goal accordingly.

LT4 (tablets, soft-gel capsules, or liquid) should be administered on an empty stomach, with water only, 30 to 60 minutes before breakfast. Medications that interfere with LT4 absorption (eg, bile acid resins, calcium carbonate, ferrous sulfate) should be taken several hours after LT4. For patients who cannot take LT4 in the morning, taking it at bedtime (≥ 2-3 hours after the last meal) is acceptable.

Continue to: Monitoring and titrating

Monitoring and titrating. Hypothyroid symptoms usually improve after 2 or 3 weeks of LT4 treatment; in severe hypothyroidism, complete recovery might take months. Approximately 6 weeks after LT4 therapy is initiated, serum TSH should be measured. After assessing whether administration of LT4 at the starting dosage is appropriate, that dosage can be increased, or decreased, every 4 to 6 weeks until the TSH goal is reached. Once the patient is maintained at a given dosage, measure serum TSH once a year—more often if there is an abnormal result or a change in the patient’s health status.²³

Adverse effects of LT4 therapy are rare, unless over-replacement occurs. Rarely, patients have an allergy to the dye or an excipient (filler) in the tablet.^26-28 The white, 50-mcg tablets can be given safely to patients with dye sensitivity. For those who have an allergy to an excipient (except gelatin) or gluten intolerance, the LT4 soft-gel capsule or liquid preparation (Tirosint) can be prescribed.

Pure LT4, in a capsule made from vegetable sources, can be ordered through a compounding pharmacy for patients who are allergic to animal products.

Anemia, especially iron-deficiency anemia, can cause intolerance to LT4 therapy; in such patients, lowering the starting dosage and treating anemia are indicated.²⁹

Persistent symptoms (despite a normal TSH level). Because many hypothyroid symptoms are nonspecific, patients might come to think that their LT4 dosage is inadequate if they feel tired or gain weight. Persistent hypothyroid symptoms despite a normal serum TSH level might be due to (1) the inability of LT4 therapy to restore tissue thyroid hormone levels to normal or (2) other variables unrelated to hypothyroidism, including disorders associated with inflammation or autoimmune disease, certain medications, diet, lifestyle, and environmental toxins.

These patients might benefit from a detailed history to identify other causes and a switch to either LT4 + liothyronine (LT3; synthetic T3) combination therapy or DTE^26,30-33 (TABLE 4³⁴), although a beneficial effect of LT4 + LT3 therapy was not seen in several studies.^35,36 Over-replacement with LT4 should be discouraged, due to concerns about thyrotoxicosis and its complications (eg, atrial fibrillation, accelerated bone loss).

DTE and LT4 + LT3. Use of DTE has decreased since the 1970s, when LT4 became the therapy of choice. Subsequently, anecdotal evidence emerged that some patients did not feel well on LT4 and preferred to return to DTE.^32,33

Continue to: Several clinical trials...

Several clinical trials addressed the question of whether residual symptoms could be resolved through LT4 + LT3 combination therapy^31-39 (TABLE 4³⁴), but evidence of any consistent superiority of combination therapy was not demonstrated.^35-39 In selected cases, patients might prefer the combination approach.^31,33,39 The quality of life of hypothyroid patients was found to be similarly improved with LT4 or DTE, but the latter was associated with modest weight loss (approximately 4 lbs); nearly 50% of study patients preferred treatment with DTE over LT4.³³ A follow-up study did not confirm weight loss with DTE, however.³⁴

When LT4 monotherapy and LT4 + LT3 combination therapy were compared, results were mixed^31-39; responsiveness to therapy containing LT3 might therefore depend on multiple variables, including genetic background, nutritional and lifestyle factors, stress, presence of comorbidities and autoimmune disorders, and other unidentified or poorly defined variables.^40-48

Consider screening for hypothyroidism in patients who take medication that affects either their thyroid function or the results of thyroid hormone assays.

Although combination therapy and DTE are not generally recommended over LT4 monotherapy, they might offer better options for patients who are still symptomatic when being treated with LT4 only: In a randomized, double-blind, crossover study that compared LT4 with DTE and with LT4 + LT3, one-third of the most highly symptomatic patients who had low scores on mood, cognitive, and quality-of-life assessments improved significantly after they were switched to combination therapy or DTE.³⁴

The 2014 American Thyroid Association guidelines²⁴ do not support routine use of LT4 + LT3 in hypothyroid patients who have residual symptoms after LT4 monotherapy; however, a therapeutic trial of LT4 + LT3, while maintaining a normal serum TSH, is reasonable in selected patients. Candidates for DTE or LT4 + LT3 might include patients who do not feel well on LT4 monotherapy, are post thyroidectomy or post radioiodine therapy, or have a low serum T3 level. DTE and combination therapy are discouraged in older patients, patients who have underlying CAD, and pregnant patients.

Special treatment circumstances

A number of patient variables have the potential to alter management strategies for hypothyroidism.^{18,20,23,40,49-53}

Age, comorbidity. Older patients (> 65 years) and patients with cardiopulmonary disease or CAD should be treated with LT4, 25 to 50 mcg/d, initially; that dosage can be titrated upward by 12.5 to 25 mcg/d every 4 to 6 weeks until the TSH goal is reached—preferably, in the range of 4 to 8 mIU/L. An increase in the dosage of LT4 might be required in the presence of malabsorption (eg, gastrointestinal disorders, celiac disease) and in nephrotic syndrome.^18,20,23

Body weight. A decrease in the dosage of LT4 might be indicated in the setting of significant weight loss (> 10% body weight).²³

Continue to: Co-pharmacy

Co-pharmacy. An increase in the dosage of LT4 might be required when other drugs (eg, phenytoin, phenobarbital, rifampin, and carbamazepine) have led to an increased rate of thyroid hormone metabolism. A decrease in the dosage of LT4 might be necessary after initiation of androgen therapy.²³

Pregnancy. Women with pre-e­xisting hypothyroidism require an increase of 25% to 50% in their LT4 dosage during pregnancy to maintain a TSH level in the recommended pregnancy reference range. Thyroid function should be monitored every 4 to 6 weeks to ensure that the TSH target for each trimester is reached (first trimester, 0.1-4 mIU/L; second trimester, 0.2-4 mIU/L; third trimester, 0.3-4 mIU/L). Postpartum, LT4 can be reduced to the prepartum dosage; TSH should be checked every 4 to 6 weeks to maintain the TSH goal.²³

Estrogen therapy. Hypothyroid women who are receiving estrogen therapy might require an increase in their LT4 dosage because serum thyroxine-binding globulin levels are increased by estrogens or through other mechanisms that have not been identified.²³

Surgical candidacy. Observational studies show few adverse outcomes in surgical patients with mild (subclinical) hypothyroidism or moderate hypothyroidism; however, the risk of adverse surgical outcome might be increased in patients with severe hypothyroidism. For patients in whom surgery is planned and who have:

• subclinical hypothyroidism (elevated TSH and normal FT4), we recommend that surgery—urgent or elective—not be posptoned but proceed.
• moderate (overt) hypothyroidism who require urgent surgery, we recommend not postponing surgery, even though minor perioperative complications might develop. Such patients should be treated with LT4 as soon as the diagnosis for which surgery is required has been made. Alternatively, when moderate hypothyroidism is discovered in a patient who is being evaluated for elective surgery, we recommend postponing surgery until the euthyroid state is restored.
• severe hypothyroidism (myxedema coma [discussed in a bit]; severe clinical symptoms of chronic hypothyroidism, such as altered mental status, pericardial effusion, or heart failure; or a very low level of T4), surgery should be delayed until hypothyroidism has been treated. When emergency surgery is required for a severely hypothyroid patient, they should be treated with LT4 as soon as the diagnosis for which surgery is indicated has been made. When emergency surgery must be performed in a patient with myxedema coma, we recommend treatment with LT4 + LT3, rather than LT4 alone, often administered intravenously because LT4 is poorly absorbed in these patients.

Nonadherence. For patients who do not take LT4 regularly or do not respond to efforts to improve adherence, LT4 can be given weekly, instead of daily, although this interval is not ideal. Weekly dosing should not be used in older patients with CAD.²³

Thyroid cancer. Patients who are post total thyroidectomy for thyroid cancer need to take LT4 to treat hypothyroidism and to prevent recurrence of thyroid cancer. The goal TSH level should be based on the cancer stage and risk of recurrence and should be monitored by an endocrinologist.

Myxedema coma. This medical emergency has high mortality. Myxedema coma occurs when severe hypothyroidism leads to any, or a combination, of the following: diminished mentation; hypothermia; bradycardia; hyponatremia; hypotension; cardiovascular, respiratory, and gastrointestinal dysfunction; and renal insufficiency. LT4, LT3, and glucocorticoids should be administered intravenously and the patient monitored closely—preferably in consultation with an endocrinologist.

Continue to: When to seek consultation

A patient with hypothyroidism should be referred to Endocrinology if they are < 18 years of age, pregnant, unresponsive to therapy, or have cardiac disease, coexisting endocrine disease, suspected myxedema coma, goiter or thyroid nodules, or a structural thyroid abnormality.

What we know about nutrition and hypothyroidism

Although it is commonly recognized that iodine is essential for production of thyroid hormone, other nutritional factors might contribute to proper production of thyroid hormones, including:

• adequate intake of iron, tyrosine, selenium, zinc, and vitamins E, B₂, B₃, B₆, C, and D^44,45
• selenium and zinc, which increase conversion of T4 to T3 and might be important in the management of hypothyroid patients^40,46
• vitamin A, zinc, and regular exercise, which have been shown to improve cellular sensitivity to thyroid hormones.

Low iron stores can contribute to persistent symptoms and poor quality of life in patients with hypothyroidism, despite their being treated according to guidelines.^29,47

Despite what is known about these nutritional connections, there is insufficient evidence that improving nutrition can reverse hypothyroidism.

Prevention

Prevention of hypothyroidism should take into account variables that affect or inhibit thyroid function, such as stress, infection (eg, Epstein-Barr virus), excessive fluoride intake, toxins (eg, pesticides, solvents, mercury, cadmium, and lead), autoimmune disease (eg, celiac disease), and food sensitivity.^54,55 Oxidative stress can also cause thyroid impairment.^40-48,54-58

Otherwise, there are, at present, no effective strategies for preventing thyroid disorders.

Subclinical hypothyroidism: Elusive management target

Subclinical hypothyroidism is defined as a normal serum FT4 level in the presence of an elevated serum TSH level. The prevalence of subclinical hypothyroidism varies from 3% to 15%, depending on the population studied; a higher incidence has been noted in women and older people.⁵⁹ In the NHANES III,¹ which excluded people with previously diagnosed thyroid disease, the incidence of subclinical hypothyroidism was 4.3%.

Continue to: Causes of subclinical...

Causes of subclinical hypothyroidism are the same as those of overt hypothyroidism, and include Hashimoto disease. The combination of an elevated TSH level and a normal FT4 level is associated with disorders characterized by protein-binding variations (eg, pregnancy, genetic disorders, drugs), TSH-secreting pituitary adenoma, class II and III obesity (respectively, body mass index, ≥ 35 but < 40 and ≥ 40), and assay variability.^49,51

Lab diagnosis: Fraught with difficulty

The serum TSH level and either the total T4 level or the FT4 level should be measured to make a diagnosis of subclinical hypothyroidism. Most laboratories use a 1-step analogue immunoassay to determine free thyroid hormones; protein-binding variations can thus affect measurement of FT4.

Increasing the dosage of the tablet form of LT4 (while monitoring the TSH level), is more cost effective than moving to an alternative preparation, such as a softgel or liquid.

Several scenarios that can result in inaccurate measurement of FT4 by radioimmunoassay include genetic disorders that affect binding proteins; pregnancy; use of certain drugs, including heparin, furosemide, antiepileptic agents, salicylate, ferrous sulfate, and cholesterol-binding resins; and some medical conditions, including cardiac surgery, critical illness, and renal failure. Variables that inhibit proper production of thyroid hormones—stress, infection, fluoride (an iodine antagonist), toxins (pesticides, mercury, cadmium, lead) and autoimmune conditions, such as celiac disease—should be considered when attempting to determine the cause of subclinical hypothyroidism.

Liquid chromatography–mass spectrometry measurement of thyroid hormones might be more accurate than immunoassay.⁵³ Measuring serum total T4 and FT4 by dialysis, free from interfering proteins, might also be useful when measurement of FT4 by immunoassay is affected by binding-protein variations.

Prescribe a brand-name preparation of levothyroxine whenever possible; generic formulations might have variable potency or contain other ingredients.

Features of subclinical hypothyroidism

Most patients who have subclinical hypothyroidism and a serum TSH level < 10 mIU/L are asymptomatic. Some might have nonspecific symptoms of hypothyroidism, however, such as reduced quality of life, poor cognitive function, and poor memory—symptoms that do not typically correlate with the serum TSH level.

An increase in the dosage of LT4 might be required in pregnancy or when weight gain is significant (> 10% of body weight).

It has been suggested that some elderly people normally have a higher level of serum TSH, and that they might have even a prolonged lifespan.⁵¹ Additionally, it has been shown that, in nonpregnant adult patients with subclinical hypothyroidism and a serum TSH level of 4.5 to 10 mIU/L, treatment with LT4 was not associated with improvement in thyroid-related symptoms or general quality of life.⁵²

Treat, or don't treat, subclinical hypothyroidism?

It is well accepted that the goal of therapy in hypothyroid patients is to normalize the serum TSH level; however, the American Thyroid Association and the American Association of Clinical Endocrinology recommend starting LT4 in patients with a serum TSH level ≥ 10 mIU/L (TABLE 5).^59-62 The principal reason for not treating subclinical hypothyroidism is the lack of benefit in reducing the risk of cardiovascular morbidity and mortality when the TSH level is between 7.5 and 10 mIU/L.⁶²

Continue to: Routine treatment

Routine treatment of patients with a serum TSH level of 4.5 to 10 mIU/L remains controversial. When TSH is 7.0 to 9.9 mIU/L, treatment is recommended for (1) patients < 65 years and (2) for older patients (> 65 years) only when there are convincing hypothyroid symptoms because of concern about unintended overtreatment.

When the TSH level is anywhere above the upper limit of normal to 6.9 mIU/L, treatment is recommended for patients < 65 years old, patients who have a high titer of thyroid peroxidase antibodies, and patients with goiter—but not for patients > 65 years (and, especially, not for octogenarians) because their upper limit of normal could be as high as 6 to 8 mIU/L, especially if they are otherwise healthy.

Treatment should be considered for women with subclinical hypothyroidism who are trying to conceive or experiencing an infertility problem.

For patients with subclinical hypothyroidism who are not being treated, monitor thyroid function every 6 to 12 months by testing TSH and FT4.

CORRESPONDENCE
Thanh D. Hoang, DO, Division of Endocrinology, Walter Reed National Military Medical Center, 8901 Wisconsin Avenue, Bethesda, MD 20889; [email protected]

The hormones thyroxine (T4) and triiodothyronine (T3), produced by the thyroid gland, are crucial for maintaining metabolism. A deficit of thyroid hormone production—hypothyroidism—is a common endocrine disorder seen in primary care.

Many patients with hypothyroidism do not achieve optimal treatment goals or see an improvement in their quality of life.

Although the diagnosis and management of hypothyroidism are considered straightforward, many patients with hypothyroidism do not achieve optimal treatment goals or see an improvement in their quality of life. In this article, we address the questionable utility of screening; outline the diagnostic approach, including the central role of laboratory testing; and explain why treatment requires a precise approach to address the range of patient types.

Epidemiology and classification

Estimates are that almost 5% of Americans 12 years or older have hypothyroidism; older people and women are more likely to develop the condition. ¹ In the US National Health and Nutrition Examination Survey (NHANES III) of 13,344 people without known thyroid disease or a family history, hypothyroidism was found in 4.6% (overt [clinical] in 0.3% and subclinical in 4.3%); 11% had a high serum thyroid peroxidase antibody level, which increases their risk of hypothyroidism, and is treated the same as hypothyroidism of other causes; and, overall, lower serum thyroid-stimulating hormone (TSH) levels were seen in Blacks, compared to Whites and Mexican Americans.¹

Copyright Steve and Myriam Oh

Primary hypothyroidism accounts for > 95% of cases of hypothyroidism, representing a failure of the thyroid gland to produce sufficient hormone. It has been shown that, in iodine-replete countries such as the United States, the prevalence of spontaneous hypothyroidism is 1% to 2%, and it is 10 times more common in women.^2,3

Central hypothyroidism is caused by insufficient stimulation of the thyroid gland by TSH, due to pituitary (secondary hypothyroidism) or hypothalamic (tertiary hypothyroidism) disease and is estimated to occur in 1 in every 20,000 to 80,000 people in the general population.⁴

How does hypothyroidism manifest?

Signs and symptoms. Manifestations of hypothyroidism range from life-threatening to minimal or no clinical signs and symptoms (TABLE W1). Signs and symptoms of low thyroid function vary by the degree of hypothyroidism at presentation.

Common signs and symptoms of low thyroid function include fatigue, weight gain, dry skin, brittle hair, hair loss, morning stiffness, muscle aches, joint pain, cold intolerance, diffuse headache, constipation, difficulty concentrating, low libido, depression, and menstrual irregularities. On physical examination, a patient might present with bradycardia, hypotension, hypothermia with slow speech or movement, coarse facial appearance, goiter, diffuse hair loss, cold hands and feet, and a prolonged Achilles tendon reflex.⁵ Skin findings, such as keratosis pilaris, palmoplantar keratoderma (thickening of the skin), and pityriasis rubra pilar, can be associated with autoimmune hypothyroidism.^6,7

Continue to: Carpal tunnel syndrome...

Carpal tunnel syndrome, plantar fasciitis, infertility or miscarriage, dyspepsia, and small intestinal bacterial overgrowth can be associated with hypothyroidism; thyroid function should therefore be assessed in patients who have any of these conditions, along with other signs and symptoms of low thyroid function.^8,9 A patient with severe hypothyroidism might present with hemodynamic instability, pericardial or pleural effusion, and myxedema coma.¹⁰

Clues in the history and from the lab. A history of radiation to the head, neck, or chest area and a history or family history of autoimmune disorders are risk factors for autoimmune thyroid disease.^11,12 Laboratory findings can include markers of oxidative stress, such as elevated levels of low-density lipoprotein cholesterol and serum malondialdehyde.^13-15

Screening and diagnosis

Screening. The US Preventive Services Task Force has asserted that evidence is insufficient by which to evaluate the benefits and risks of routine screening for thyroid dysfunction in nonpregnant, asymptomatic adults.¹⁶ According to the American Thyroid Association and the American Association of Clinical Endocrinologists, screening should be considered in high-risk patients, including those who take medication that affects thyroid function or the results of thyroid hormone assays (TABLE W2, available at mdedge.com/familymedicine).^17-20

Screening inpatients is challenging and usually not recommended unless thyroid disease is strongly suspected. This is because changes in the levels of thyroid hormones, binding proteins, and the TSH concentration can occur in severe nonthyroidal illness; in addition, assay interference by antibodies and other substances can affect thyroid hormone measurement.²¹

Testing strategy. Generally, screening and diagnosis of hypothyroidism are based primarily on laboratory testing, because signs and symptoms are nonspecific (FIGURE 1⁵). A serum TSH level is usually the initial test when screening for thyroid dysfunction. A normal serum TSH value ranges from 0.5-5.0 mIU/L.

A patient with severe hypothyroidism might present with hemodynamic instability, pericardial or pleural effusion, or myxedema coma.

When an abnormal serum TSH level is found, further tests can be performed to investigate, including a serum free thyroxine (FT4) test. (Our preference is to order TSH and FT4 assays simultaneously to facilitate and confirm the diagnosis.) An FT4 test measures the amount of unattached, or free, thyroxine in blood by immunoassay. A normal FT4 value usually ranges from 0.7-1.9 ng/dL.

The combination of a high TSH level and a low FT4 level could be an indication of an underactive thyroid gland (ie, clinical or overt hypothyroidism). Milder, subclinical hypothyroidism is characterized by a ­higher-than-normal TSH level but a normal FT4 level.²² Central (secondary) hypothyroidism is characterized by a low serum FT4 level and a serum TSH level that can be below the reference range, low normal, or even slightly high.⁴

Continue to: These measurements...

These measurements must be interpreted within the context of the laboratory-specific normal range for each test. Third-generation serum TSH assays are more sensitive and specific than serum FT4 measurements for hypothyroidism. FT4 is usually measured by automated analogue immunoassay, which generally provides reliable results; abnormal binding proteins or other interferences occur in some patients, however, resulting in reporting of a falsely high, or falsely low, FT4 level. In such cases, FT4 by direct dialysis, or total T4, can be measured for further evaluation. In primary care, you are most likely to encounter primary hypothyroidism; secondary (central) hypothyroidism is much rarer (< 5% of cases).⁴

The ins and outs of treatment

For most patients, hypothyroidism is a permanent disorder requiring lifelong thyroid hormone replacement therapy—unless the disease is transient (ie, painless or subacute thyroiditis); reversible, because it is caused by medication; or responsive to medical intervention that addresses the underlying autoimmune condition.¹⁹ Goals of treatment (Figure 2^5,23) are to:

• normalize the TSH level to 0.5-5.0 mIU/L (the main goal), with an age-related shift toward a higher TSH goal in older patients (and an upper limit of normal of 7.5 mIU/L in patients who are ≥ 80 years of age)²⁰
• restore the euthyroid state
• relieve symptoms
• reduce any goiter
• avoid overtreatment (iatrogenic thyrotoxicosis).

Desiccated thyroid extract (DTE), developed in the late 1880s and made from the dried thyroid gland of pigs, sheep, or cows, was the earliest treatment for hypothyroidism. The use of DTE has declined since the introduction of synthetic thyroxine (T4, or levothyroxine [here, referred to as LT4]), which is now the standard treatment.^20-22 LT4 is deiodinated in peripheral tissues to form T3, the active thyroid hormone; this process accounts for 80% of total T3 production daily.²⁴

LT4 formulations. LT4 is commercially available in tablet, soft-gel, and liquid preparations. Most patients are treated with the tablet; the soft-gel capsule or liquid is an option for patients who absorb the tablet poorly (because of atrophic gastritis, celiac disease, or gluten sensitivity or because they are post bariatric surgery). Increasing the dosage of the tablet form of LT4, with ongoing TSH monitoring, is more cost effective than moving to an alternative preparation.

If a switch of LT4 formulation is made (ie, from one manufacturer to another), test the serum TSH level to ensure that the therapeutic goal is being reached. Also, in our experience, it is best to prescribe a brand-name preparation of levothyroxine, not a generic, whenever possible, due to the variability in generic formulations and the potential presence of other (inert) ingredients.²⁵

Dosing (TABLE 3^20,23). The average full replacement dosage of LT4 for a young, healthy adult is approximately 1.6 mcg/kg/d. Older patients (> 65 years) or those with coronary artery disease (CAD) should be started on a lower dosage (25-50 mcg/d) and titrated to goal accordingly.

LT4 (tablets, soft-gel capsules, or liquid) should be administered on an empty stomach, with water only, 30 to 60 minutes before breakfast. Medications that interfere with LT4 absorption (eg, bile acid resins, calcium carbonate, ferrous sulfate) should be taken several hours after LT4. For patients who cannot take LT4 in the morning, taking it at bedtime (≥ 2-3 hours after the last meal) is acceptable.

Continue to: Monitoring and titrating

Monitoring and titrating. Hypothyroid symptoms usually improve after 2 or 3 weeks of LT4 treatment; in severe hypothyroidism, complete recovery might take months. Approximately 6 weeks after LT4 therapy is initiated, serum TSH should be measured. After assessing whether administration of LT4 at the starting dosage is appropriate, that dosage can be increased, or decreased, every 4 to 6 weeks until the TSH goal is reached. Once the patient is maintained at a given dosage, measure serum TSH once a year—more often if there is an abnormal result or a change in the patient’s health status.²³

Adverse effects of LT4 therapy are rare, unless over-replacement occurs. Rarely, patients have an allergy to the dye or an excipient (filler) in the tablet.^26-28 The white, 50-mcg tablets can be given safely to patients with dye sensitivity. For those who have an allergy to an excipient (except gelatin) or gluten intolerance, the LT4 soft-gel capsule or liquid preparation (Tirosint) can be prescribed.

Pure LT4, in a capsule made from vegetable sources, can be ordered through a compounding pharmacy for patients who are allergic to animal products.

Anemia, especially iron-deficiency anemia, can cause intolerance to LT4 therapy; in such patients, lowering the starting dosage and treating anemia are indicated.²⁹

Persistent symptoms (despite a normal TSH level). Because many hypothyroid symptoms are nonspecific, patients might come to think that their LT4 dosage is inadequate if they feel tired or gain weight. Persistent hypothyroid symptoms despite a normal serum TSH level might be due to (1) the inability of LT4 therapy to restore tissue thyroid hormone levels to normal or (2) other variables unrelated to hypothyroidism, including disorders associated with inflammation or autoimmune disease, certain medications, diet, lifestyle, and environmental toxins.

These patients might benefit from a detailed history to identify other causes and a switch to either LT4 + liothyronine (LT3; synthetic T3) combination therapy or DTE^26,30-33 (TABLE 4³⁴), although a beneficial effect of LT4 + LT3 therapy was not seen in several studies.^35,36 Over-replacement with LT4 should be discouraged, due to concerns about thyrotoxicosis and its complications (eg, atrial fibrillation, accelerated bone loss).

DTE and LT4 + LT3. Use of DTE has decreased since the 1970s, when LT4 became the therapy of choice. Subsequently, anecdotal evidence emerged that some patients did not feel well on LT4 and preferred to return to DTE.^32,33

Continue to: Several clinical trials...

Several clinical trials addressed the question of whether residual symptoms could be resolved through LT4 + LT3 combination therapy^31-39 (TABLE 4³⁴), but evidence of any consistent superiority of combination therapy was not demonstrated.^35-39 In selected cases, patients might prefer the combination approach.^31,33,39 The quality of life of hypothyroid patients was found to be similarly improved with LT4 or DTE, but the latter was associated with modest weight loss (approximately 4 lbs); nearly 50% of study patients preferred treatment with DTE over LT4.³³ A follow-up study did not confirm weight loss with DTE, however.³⁴

When LT4 monotherapy and LT4 + LT3 combination therapy were compared, results were mixed^31-39; responsiveness to therapy containing LT3 might therefore depend on multiple variables, including genetic background, nutritional and lifestyle factors, stress, presence of comorbidities and autoimmune disorders, and other unidentified or poorly defined variables.^40-48

Consider screening for hypothyroidism in patients who take medication that affects either their thyroid function or the results of thyroid hormone assays.

Although combination therapy and DTE are not generally recommended over LT4 monotherapy, they might offer better options for patients who are still symptomatic when being treated with LT4 only: In a randomized, double-blind, crossover study that compared LT4 with DTE and with LT4 + LT3, one-third of the most highly symptomatic patients who had low scores on mood, cognitive, and quality-of-life assessments improved significantly after they were switched to combination therapy or DTE.³⁴

The 2014 American Thyroid Association guidelines²⁴ do not support routine use of LT4 + LT3 in hypothyroid patients who have residual symptoms after LT4 monotherapy; however, a therapeutic trial of LT4 + LT3, while maintaining a normal serum TSH, is reasonable in selected patients. Candidates for DTE or LT4 + LT3 might include patients who do not feel well on LT4 monotherapy, are post thyroidectomy or post radioiodine therapy, or have a low serum T3 level. DTE and combination therapy are discouraged in older patients, patients who have underlying CAD, and pregnant patients.

Special treatment circumstances

A number of patient variables have the potential to alter management strategies for hypothyroidism.^{18,20,23,40,49-53}

Age, comorbidity. Older patients (> 65 years) and patients with cardiopulmonary disease or CAD should be treated with LT4, 25 to 50 mcg/d, initially; that dosage can be titrated upward by 12.5 to 25 mcg/d every 4 to 6 weeks until the TSH goal is reached—preferably, in the range of 4 to 8 mIU/L. An increase in the dosage of LT4 might be required in the presence of malabsorption (eg, gastrointestinal disorders, celiac disease) and in nephrotic syndrome.^18,20,23

Body weight. A decrease in the dosage of LT4 might be indicated in the setting of significant weight loss (> 10% body weight).²³

Continue to: Co-pharmacy

Co-pharmacy. An increase in the dosage of LT4 might be required when other drugs (eg, phenytoin, phenobarbital, rifampin, and carbamazepine) have led to an increased rate of thyroid hormone metabolism. A decrease in the dosage of LT4 might be necessary after initiation of androgen therapy.²³

Pregnancy. Women with pre-e­xisting hypothyroidism require an increase of 25% to 50% in their LT4 dosage during pregnancy to maintain a TSH level in the recommended pregnancy reference range. Thyroid function should be monitored every 4 to 6 weeks to ensure that the TSH target for each trimester is reached (first trimester, 0.1-4 mIU/L; second trimester, 0.2-4 mIU/L; third trimester, 0.3-4 mIU/L). Postpartum, LT4 can be reduced to the prepartum dosage; TSH should be checked every 4 to 6 weeks to maintain the TSH goal.²³

Estrogen therapy. Hypothyroid women who are receiving estrogen therapy might require an increase in their LT4 dosage because serum thyroxine-binding globulin levels are increased by estrogens or through other mechanisms that have not been identified.²³

Surgical candidacy. Observational studies show few adverse outcomes in surgical patients with mild (subclinical) hypothyroidism or moderate hypothyroidism; however, the risk of adverse surgical outcome might be increased in patients with severe hypothyroidism. For patients in whom surgery is planned and who have:

• subclinical hypothyroidism (elevated TSH and normal FT4), we recommend that surgery—urgent or elective—not be posptoned but proceed.
• moderate (overt) hypothyroidism who require urgent surgery, we recommend not postponing surgery, even though minor perioperative complications might develop. Such patients should be treated with LT4 as soon as the diagnosis for which surgery is required has been made. Alternatively, when moderate hypothyroidism is discovered in a patient who is being evaluated for elective surgery, we recommend postponing surgery until the euthyroid state is restored.
• severe hypothyroidism (myxedema coma [discussed in a bit]; severe clinical symptoms of chronic hypothyroidism, such as altered mental status, pericardial effusion, or heart failure; or a very low level of T4), surgery should be delayed until hypothyroidism has been treated. When emergency surgery is required for a severely hypothyroid patient, they should be treated with LT4 as soon as the diagnosis for which surgery is indicated has been made. When emergency surgery must be performed in a patient with myxedema coma, we recommend treatment with LT4 + LT3, rather than LT4 alone, often administered intravenously because LT4 is poorly absorbed in these patients.

Nonadherence. For patients who do not take LT4 regularly or do not respond to efforts to improve adherence, LT4 can be given weekly, instead of daily, although this interval is not ideal. Weekly dosing should not be used in older patients with CAD.²³

Thyroid cancer. Patients who are post total thyroidectomy for thyroid cancer need to take LT4 to treat hypothyroidism and to prevent recurrence of thyroid cancer. The goal TSH level should be based on the cancer stage and risk of recurrence and should be monitored by an endocrinologist.

Myxedema coma. This medical emergency has high mortality. Myxedema coma occurs when severe hypothyroidism leads to any, or a combination, of the following: diminished mentation; hypothermia; bradycardia; hyponatremia; hypotension; cardiovascular, respiratory, and gastrointestinal dysfunction; and renal insufficiency. LT4, LT3, and glucocorticoids should be administered intravenously and the patient monitored closely—preferably in consultation with an endocrinologist.

Continue to: When to seek consultation

A patient with hypothyroidism should be referred to Endocrinology if they are < 18 years of age, pregnant, unresponsive to therapy, or have cardiac disease, coexisting endocrine disease, suspected myxedema coma, goiter or thyroid nodules, or a structural thyroid abnormality.

What we know about nutrition and hypothyroidism

Although it is commonly recognized that iodine is essential for production of thyroid hormone, other nutritional factors might contribute to proper production of thyroid hormones, including:

• adequate intake of iron, tyrosine, selenium, zinc, and vitamins E, B₂, B₃, B₆, C, and D^44,45
• selenium and zinc, which increase conversion of T4 to T3 and might be important in the management of hypothyroid patients^40,46
• vitamin A, zinc, and regular exercise, which have been shown to improve cellular sensitivity to thyroid hormones.

Low iron stores can contribute to persistent symptoms and poor quality of life in patients with hypothyroidism, despite their being treated according to guidelines.^29,47

Despite what is known about these nutritional connections, there is insufficient evidence that improving nutrition can reverse hypothyroidism.

Prevention

Prevention of hypothyroidism should take into account variables that affect or inhibit thyroid function, such as stress, infection (eg, Epstein-Barr virus), excessive fluoride intake, toxins (eg, pesticides, solvents, mercury, cadmium, and lead), autoimmune disease (eg, celiac disease), and food sensitivity.^54,55 Oxidative stress can also cause thyroid impairment.^40-48,54-58

Otherwise, there are, at present, no effective strategies for preventing thyroid disorders.

Subclinical hypothyroidism: Elusive management target

Subclinical hypothyroidism is defined as a normal serum FT4 level in the presence of an elevated serum TSH level. The prevalence of subclinical hypothyroidism varies from 3% to 15%, depending on the population studied; a higher incidence has been noted in women and older people.⁵⁹ In the NHANES III,¹ which excluded people with previously diagnosed thyroid disease, the incidence of subclinical hypothyroidism was 4.3%.

Continue to: Causes of subclinical...

Causes of subclinical hypothyroidism are the same as those of overt hypothyroidism, and include Hashimoto disease. The combination of an elevated TSH level and a normal FT4 level is associated with disorders characterized by protein-binding variations (eg, pregnancy, genetic disorders, drugs), TSH-secreting pituitary adenoma, class II and III obesity (respectively, body mass index, ≥ 35 but < 40 and ≥ 40), and assay variability.^49,51

Lab diagnosis: Fraught with difficulty

The serum TSH level and either the total T4 level or the FT4 level should be measured to make a diagnosis of subclinical hypothyroidism. Most laboratories use a 1-step analogue immunoassay to determine free thyroid hormones; protein-binding variations can thus affect measurement of FT4.

Increasing the dosage of the tablet form of LT4 (while monitoring the TSH level), is more cost effective than moving to an alternative preparation, such as a softgel or liquid.

Several scenarios that can result in inaccurate measurement of FT4 by radioimmunoassay include genetic disorders that affect binding proteins; pregnancy; use of certain drugs, including heparin, furosemide, antiepileptic agents, salicylate, ferrous sulfate, and cholesterol-binding resins; and some medical conditions, including cardiac surgery, critical illness, and renal failure. Variables that inhibit proper production of thyroid hormones—stress, infection, fluoride (an iodine antagonist), toxins (pesticides, mercury, cadmium, lead) and autoimmune conditions, such as celiac disease—should be considered when attempting to determine the cause of subclinical hypothyroidism.

Liquid chromatography–mass spectrometry measurement of thyroid hormones might be more accurate than immunoassay.⁵³ Measuring serum total T4 and FT4 by dialysis, free from interfering proteins, might also be useful when measurement of FT4 by immunoassay is affected by binding-protein variations.

Prescribe a brand-name preparation of levothyroxine whenever possible; generic formulations might have variable potency or contain other ingredients.

Features of subclinical hypothyroidism

Most patients who have subclinical hypothyroidism and a serum TSH level < 10 mIU/L are asymptomatic. Some might have nonspecific symptoms of hypothyroidism, however, such as reduced quality of life, poor cognitive function, and poor memory—symptoms that do not typically correlate with the serum TSH level.

An increase in the dosage of LT4 might be required in pregnancy or when weight gain is significant (> 10% of body weight).

It has been suggested that some elderly people normally have a higher level of serum TSH, and that they might have even a prolonged lifespan.⁵¹ Additionally, it has been shown that, in nonpregnant adult patients with subclinical hypothyroidism and a serum TSH level of 4.5 to 10 mIU/L, treatment with LT4 was not associated with improvement in thyroid-related symptoms or general quality of life.⁵²

Treat, or don't treat, subclinical hypothyroidism?

It is well accepted that the goal of therapy in hypothyroid patients is to normalize the serum TSH level; however, the American Thyroid Association and the American Association of Clinical Endocrinology recommend starting LT4 in patients with a serum TSH level ≥ 10 mIU/L (TABLE 5).^59-62 The principal reason for not treating subclinical hypothyroidism is the lack of benefit in reducing the risk of cardiovascular morbidity and mortality when the TSH level is between 7.5 and 10 mIU/L.⁶²

Continue to: Routine treatment

Routine treatment of patients with a serum TSH level of 4.5 to 10 mIU/L remains controversial. When TSH is 7.0 to 9.9 mIU/L, treatment is recommended for (1) patients < 65 years and (2) for older patients (> 65 years) only when there are convincing hypothyroid symptoms because of concern about unintended overtreatment.

When the TSH level is anywhere above the upper limit of normal to 6.9 mIU/L, treatment is recommended for patients < 65 years old, patients who have a high titer of thyroid peroxidase antibodies, and patients with goiter—but not for patients > 65 years (and, especially, not for octogenarians) because their upper limit of normal could be as high as 6 to 8 mIU/L, especially if they are otherwise healthy.

Treatment should be considered for women with subclinical hypothyroidism who are trying to conceive or experiencing an infertility problem.

For patients with subclinical hypothyroidism who are not being treated, monitor thyroid function every 6 to 12 months by testing TSH and FT4.

CORRESPONDENCE
Thanh D. Hoang, DO, Division of Endocrinology, Walter Reed National Military Medical Center, 8901 Wisconsin Avenue, Bethesda, MD 20889; [email protected]

Dyspareunia is persistent or recurrent pain before, during, or after sexual contact and is not limited to cisgender individuals or vaginal intercourse.^1-3 With a prevalence as high as 45% in the United States,^2-5 it is one of the most common complaints in gynecologic practices.^5,6

Causes and contributing factors

There are many possible causes of dyspareunia.^2,4,6 While some patients have a single cause, most cases are complex, with multiple overlapping causes and maintaining factors.^4,6 Identifying each contributing factor can help you appropriately address all components.

Physical conditions. The range of physical contributors to dyspareunia includes inflammatory processes, structural abnormalities, musculoskeletal dysfunctions, pelvic organ disorders, injuries, iatrogenic effects, infections, allergic reactions, sensitization, hormonal changes, medication effects, adhesions, autoimmune disorders, and other pain syndromes (TABLE 1^2-4,6-11).

Inadequate arousal. One of the primary causes of pain during vaginal penetration is inadequate arousal and lubrication.^1,2,9-11 Arousal is the phase of the sexual response cycle that leads to genital tumescence and prepares the genitals for sexual contact through penile/clitoral erection, vaginal engorgement, and lubrication, which prevents pain and enhances pleasurable sensation.^9-11

While some physical conditions can lead to an inability to lubricate, the most common causes of inadequate lubrication are psychosocial-behavioral, wherein patients have the same physical ability to lubricate as patients without genital pain but do not progress through the arousal phase.^9-11 Behavioral factors such as inadequate or ineffective foreplay can fail to produce engorgement and lubrication, while psychosocial factors such as low attraction to partner, relationship stressors, anxiety, or low self-esteem can have an inhibitory effect on sexual arousal.^1,2,9-11 Psychosocial and behavioral factors may also be maintaining factors or consequences of dyspareunia, and need to be assessed and treated.^1,2,9-11

Psychological trauma. Exposure to psychological traumas and the development of posttraumatic stress disorder (PTSD) have been linked with the development of pain disorders in general and dyspareunia specifically. Most patients seeking treatment for chronic pain disorders have a history of physical or sexual abuse.¹² Changes in physiologic processes (eg, neurochemical, endocrine) that occur with PTSD interfere with the sexual response cycle, and sexual traumas specifically have been linked with pelvic floor dysfunction.^13,14 Additionally, when PTSD is caused by a sexual trauma, even consensual sexual encounters can trigger flashbacks, intrusive memories, hyperarousal, and muscle tension that interfere with the sexual response cycle and contribute to genital pain.¹³

Vaginismus is both a physiologic and psychological contributor to dyspareunia.^1,2,4 Patients experiencing pain can develop anxiety about repeated pain and involuntarily contract their pelvic muscles, thereby creating more pain, increasing anxiety, decreasing lubrication, and causing pelvic floor dysfunction.^1-4,6 Consequently, all patients with dyspareunia should be assessed and continually monitored for symptoms of vaginismus.

Continue to: Anxiety

Anxiety. As with other pain disorders, anxiety develops around pain triggers.^10,15 When expecting sexual activity, patients can experience extreme worry and panic attacks.^10,15,16 The distress of sexual encounters can interfere with physiologic arousal and sexual desire, impacting all phases of the sexual response cycle.^1,2

Relationship issues. Difficulty engaging in or avoidance of sexual activity can interfere with romantic relationships.^2,10,16 Severe pain or vaginismus contractions can prevent penetration, leading to unconsummated marriages and an inability to conceive through intercourse.¹⁰ The distress surrounding sexual encounters can precipitate erectile dysfunction in male partners, or partners may continue to demand sexual encounters despite the patient’s pain, further impacting the relationship and heightening sexual distress.¹⁰ These stressors have led to relationships ending, patients reluctantly agreeing to nonmonogamy to appease their partners, and patients avoiding relationships altogether.^10,16

Devalued self-image. Difficulties with sexuality and relationships impact the self-image of patients with dyspareunia. Diminished self-image may include feeling “inadequate” as a woman and as a sexual partner, or feeling like a “failure.”¹⁶ Women with dyspareunia often have more distress related to their body image, physical appearance, and genital self-image than do women without genital pain.¹⁷ Feeling resentment toward their body, or feeling “ugly,” embarrassed, shamed, “broken,” and “useless” also contribute to increased depressive symptoms found in patients with dyspareunia.^16,18

Making the diagnosis

Most patients do not report symptoms unless directly asked^2,7; therefore, it is recommended that all patients be screened as a part of an initial intake and before any genital exam (TABLE 2^{2-4,6,7,9,11,19,20}).^4,7,21 If this screen is positive, a separate appointment may be needed for a thorough evaluation and before any attempt is made at a genital exam.^4,7

Items to include in the clinical interview

Given the range of possible causes of dyspareunia and its contributing factors and symptoms, a thorough clinical interview is essential. Begin with a review of the patient’s complete medical and surgical history to identify possible known contributors to genital pain.⁴ Pregnancy history is of particular importance as the prevalence of postpartum dyspareunia is 35%, with risk being greater for patients who experienced dyspareunia symptoms before pregnancy.²²

Consider using a measure such as the Female Sexual Function Index or the McGill Pain Questionnaire to help patients more thoroughly describe their symptoms.

Knowing the location and quality of pain is important for differentiating between possible diagnoses, as is specifying dyspareunia as lifelong or acquired, superficial or deep, and primary or secondary.^1-4,6 Confirm the specific location(s) of pain—eg, at the introitus, in the vestibule, on the labia, in the perineum, or near the clitoris.^2,4,6 A diagram or model may be needed to help patients to localize pain.⁴

To help narrow the differential, include the following elements in your assessment: pain quality, timing (eg, initial onset, episode onset, episode duration, situational triggers), alleviating factors, symptoms in surrounding structures (eg, bladder, bowel, muscles, bones), sexual history, other areas of sexual functioning, history of psychological trauma, relationship effects, and mental health (TABLE 2^{2-4,6,7,9,11,19,20} and Table 3^23-28). Screening for a history of sexual trauma is particularly important, as a recent systematic review and meta-analysis found that women with a history of sexual assault had a 42% higher risk of gynecologic problems overall, a 74% higher risk of dyspareunia, and a 71% higher risk of vaginismus than women without a history of sexual assault.²⁹ Using measures such as the Female Sexual Function Index or the McGill Pain Questionnaire can help patients more thoroughly describe their symptoms (TABLE 3^23-28).³

Continue to: Guidelines for the physical exam

Before the exam, ensure the patient has not used any topical genital treatment in the past 2 weeks that may interfere with sensitivity to the exam.⁴ To decrease patients’ anxiety about the exam, remind them that they can stop the exam at any time.⁷ Also consider offering the use of a mirror to better pinpoint the location of pain, and to possibly help the patient learn more about her anatomy.^2,7

Begin the exam by palpating surrounding areas that may be involved in pain, including the abdomen and musculoskeletal features.^3,6,19 Next visually inspect the external genitalia for lesions, abrasions, discoloration, erythema, or other abnormal findings.^2,3,6 Ask the patient for permission before contacting the genitals. Because the labia may be a site of pain, apply gentle pressure in retracting it to fully examine the vestibule.^6,7 Contraction of the pelvic floor muscles during approach or initial palpation could signal possible vaginismus.⁴

After visual inspection of external genitalia, use a cotton swab to map the vulva and vestibule in a clockwise fashion to precisely identify any painful locations.^2-4,6 If the patient’s history of pain has been intermittent, it’s possible that the cotton swab will not elicit pain on the day of the initial exam, but it may on other days.⁴

Begin the internal exam by inserting a single finger into the first inch of the vagina and have the patient squeeze and release to assess tenderness, muscle tightness, and control.^2,6 Advance the finger further into the vagina and palpate clockwise, examining the levator muscles, obturator muscles, rectum, urethra, and bladder for abnormal tightness or reproduction of pain.^2,4,6 Complete a bimanual exam to evaluate the pelvic organs and adnexa.^2,4 If indicated, a more thorough evaluation of pelvic floor musculature can be performed by a physical therapist or gynecologist who specializes in pelvic pain.^2-4

If the patient consents to further evaluation, consider using a small speculum, advanced slowly, for further internal examination, noting any lesions, abrasions, discharge, ectropion, or tenderness.^2-4,7 A rectal exam may also be needed in cases of deep dyspareunia.⁶ Initial work-up may include a potassium hydroxide wet prep, sexually transmitted infection testing, and pelvic ultrasound.^2,4 In some cases, laparoscopy or biopsy may be needed.^2,4

Treatments for common causes

Treatment often begins with education about anatomy, to help patients communicate about symptoms and engage more fully in their care.³ Additional education may be needed on genital functioning and the necessity of adequate stimulation and lubrication prior to penetration.^1,2,9-11 A discussion of treatments for the wide range of possible causes of dyspareunia is outside the scope of this article. However, some basic behavioral changes may help patients address some of the more common contributing factors.

A recent systematic review and meta-analysis found that women with a history of sexual assault had a 74% higher risk of dyspareunia than women without such a history.

For example, if vaginal infection is suspected, advise patients to discontinue the use of harsh soaps, known vaginal irritants (eg, perfumed products, bath additives), and douches.³ Recommend using only ­preservative- and alcohol-free lubricants for sexual contact, and avoiding lubricants with added functions (eg, warming).³ It’s worth noting that avoidance of tight clothing and thong underwear due to possible risk for infections may not be necessary. A recent study found that women who frequently wore thong underwear (more than half of the time) were no more likely to develop urinary tract infections, yeast vaginitis, or bacterial vaginosis than those who avoid such items.³⁰ However, noncotton underwear fabric, rather than tightness, was associated with yeast vaginitis³⁰; therefore, patients may want to consider using only breathable underwear.³

Continue to: Medication

Medication. Medication may be used to treat the underlying contributing conditions or the symptom of pain directly. Some common options are particularly important for patients whose dyspareunia does not have an identifiable cause. These medications include anti-inflammatory agents, topical anesthetics, tricyclic antidepressants, and hormonal treatments.^2-4 Since effectiveness varies based on subtypes of pain, select a medication according to the location, timing, and hypothesized mechanism of pain.^3,31,32

Medication for deep pain. A meta-analysis and systematic review found that patients with some types of chronic pelvic pain with pain deep in the vagina or pelvis experienced greater than 50% reduction in pain using medroxyprogesterone acetate compared with placebo.³³ Other treatments for deep pain depend on physical exam findings.

Medication for superficial pain. Many remedies have been tried, with at least 26 different treatments for vulvodynia pain alone.¹⁶ Only some of these treatments have supporting evidence. For patients with vulvar pain, an intent-to-treat RCT found that patients using a topical steroid experienced a 23% reduction in pain from pre-treatment to 6-month follow-up.³²

Surgery is also effective for vulvar pain.^34,35 For provoked vestibulodynia (in which pain is localized to the vestibule and triggered by contact with the vulva), or vulvar vestibulitis, RCTs have found that vestibulectomy has stronger effects on pain than other treatments,^31,35 with a 53% reduction in pain during intercourse and a 70% reduction in vestibular pain overall.³⁵ However, while vestibulectomy is effective for provoked vestibulodynia, it is not recommended for generalized vulvodynia, in which pain is diffuse across the vulva and occurs without vulvar contact.³⁴

Unsupported treatments. A number of other treatments have not yet been found effective. Although lidocaine for vulvar pain is often used, RCTs have not found any significant reduction in symptoms, and a ­double-blind RCT found that lidocaine ointment actually performed worse than placebo.^31,34 Similarly, oral tricyclics have not been found to decrease vulvar pain more than placebo in double-blind studies.^31,34 Furthermore, a meta-analysis of RCTs comparing treatments with placebo for vestibular pain found no significant decrease in dyspareunia for topical conjugated estrogen, topical lidocaine, oral desipramine, oral desipramine with topical lidocaine, laser therapy, or transcranial direct current.³²

Tx risks to consider. Risks and benefits of dyspareunia treatment options should be thoroughly weighed and discussed with the patient.^2-4 Vestibulectomy, despite reducing pain for many patients, has led to increased pain for 9% of patients who underwent the procedure.³⁵ Topical treatments may lead to allergic reactions, inflammation, and worsening of symptoms,⁴ and hormonal treatments have been found to increase the risk of weight gain and bloating and are not appropriate for patients trying to conceive.³³

Coordinate care with other providers

While medications and surgery can reduce pain, they have not been shown to improve other aspects of sexual functioning such as sexual satisfaction, frequency of sexual intercourse, or overall sense of sexual functioning.³⁵ Additionally, pain reduction does not address muscle tension, anxiety, self-­esteem, and relationship problems. As a result, a multidisciplinary approach is generally needed.^3,4,32,33

Continue to: Physical therapists

Physical therapists. Pelvic floor physical therapists are often members of the dyspareunia treatment team and can provide a thorough evaluation and treatment of pelvic floor disorders.^2-4 An RCT with intent-to-treat analysis found that pain was reduced by 71% following pelvic floor physical therapy.³⁶ Another RCT found that 90% of patients reported a clinically meaningful decrease in pain with pelvic floor physical therapy.³⁷ In addition to addressing pain, pelvic floor physical therapy has also been found to improve sexual functioning, sexual satisfaction, distress, and patient perception of improvement.^34,36,37

Behavioral health specialists. Psychotherapists, especially those trained in sex therapy, couples therapy, or cognitive behavioral therapy (CBT), are also typically on the treatment team. Multiple RCTs have found evidence of CBT’s effectiveness in the direct treatment of dyspareunia pain. Bergeron et al³⁵ found a 37.5% reduction in vulvar vestibulitis pain intensity during intercourse after patients completed group CBT. Another intent-to-treat RCT found that patients receiving CBT experienced more pain reduction (~ 30%) than patients who were treated with a topical steroid.³⁸

In addition to having a direct impact on pain, CBT has also been found to have a clinically and statistically significant positive impact on other aspects of sexual experience, such as overall sexuality, self-efficacy, overall sexual functioning, frequency of intercourse, and catastrophizing.^34,38 A recent meta-­analysis of RCTs found that about 80% of vaginismus patients were able to achieve penetrative intercourse after treatment with behavioral sex therapy or CBT.³⁹ This success rate was not exceeded by physical or surgical treatments.³⁹

When PTSD is thought to be a contributing factor, trauma therapy will likely be needed in addition to treatments for dyspareunia. First-line treatments for PTSD include cognitive processing therapy, prolonged exposure, trauma-focused CBT, and cognitive therapy.⁴⁰

Chronic pelvic pain with pain deep in the vagina or pelvis has been reduced by > 50%, compared with placebo, using medroxyprogesterone acetate.

Psychotherapists can also help patients reduce anxiety, reintroduce sexual contact without triggering pain or anxiety, address emotional and self-esteem effects of dyspareunia, address relationship issues, and refocus sexual encounters on pleasure rather than pain avoidance.^2-4 Despite patient reports of high treatment satisfaction following therapy,³⁸ many patients may initially lack confidence in psychotherapy as a treatment for pain³⁵ and may need to be educated on its effectiveness and multidimensional benefits.

Gynecologists. Often a gynecologist with specialization in pelvic pain is an essential member of the team for diagnostic clarification, recommendation of treatment options, and performance of more advanced treatments.^2,3 If pain has become chronic, the patient may also benefit from a pain management team and support groups.^2,3

Follow-up steps

Patients who screen negative for dyspareunia should be re-screened periodically. Continue to assess patients diagnosed with dyspareunia for vaginismus symptoms (if they are not initially present) to ensure that the treatment plan is appropriately adjusted. Once treatment has begun, ask about adverse effects and confidence in the treatment plan to minimize negative impacts on treatment adherence and to anticipate a need for a change in the treatment approach.^31,35 In addition to tracking treatment effects on pain, continue to assess for patient-centered outcomes such as emotional functioning, self-esteem, and sexual and relationship satisfaction.³⁴ The Female Sexual Function Index can be a useful tool to track symptoms.^27,34

Finally, patients who do not experience sufficient improvement in symptoms and functioning with initial treatment may need continued support and encouragement. Given the broad range of contributing factors and the high number of potential treatments, patients may find hope in learning that multiple other treatment options may be available.

CORRESPONDENCE
Adrienne A. Williams, PhD, Department of Family and Community Medicine, University of Illinois at Chicago College of Medicine, 1919 W Taylor Street, MC 663, Chicago, IL 60612; [email protected]

Dyspareunia is persistent or recurrent pain before, during, or after sexual contact and is not limited to cisgender individuals or vaginal intercourse.^1-3 With a prevalence as high as 45% in the United States,^2-5 it is one of the most common complaints in gynecologic practices.^5,6

Causes and contributing factors

There are many possible causes of dyspareunia.^2,4,6 While some patients have a single cause, most cases are complex, with multiple overlapping causes and maintaining factors.^4,6 Identifying each contributing factor can help you appropriately address all components.

Physical conditions. The range of physical contributors to dyspareunia includes inflammatory processes, structural abnormalities, musculoskeletal dysfunctions, pelvic organ disorders, injuries, iatrogenic effects, infections, allergic reactions, sensitization, hormonal changes, medication effects, adhesions, autoimmune disorders, and other pain syndromes (TABLE 1^2-4,6-11).

Inadequate arousal. One of the primary causes of pain during vaginal penetration is inadequate arousal and lubrication.^1,2,9-11 Arousal is the phase of the sexual response cycle that leads to genital tumescence and prepares the genitals for sexual contact through penile/clitoral erection, vaginal engorgement, and lubrication, which prevents pain and enhances pleasurable sensation.^9-11

While some physical conditions can lead to an inability to lubricate, the most common causes of inadequate lubrication are psychosocial-behavioral, wherein patients have the same physical ability to lubricate as patients without genital pain but do not progress through the arousal phase.^9-11 Behavioral factors such as inadequate or ineffective foreplay can fail to produce engorgement and lubrication, while psychosocial factors such as low attraction to partner, relationship stressors, anxiety, or low self-esteem can have an inhibitory effect on sexual arousal.^1,2,9-11 Psychosocial and behavioral factors may also be maintaining factors or consequences of dyspareunia, and need to be assessed and treated.^1,2,9-11

Psychological trauma. Exposure to psychological traumas and the development of posttraumatic stress disorder (PTSD) have been linked with the development of pain disorders in general and dyspareunia specifically. Most patients seeking treatment for chronic pain disorders have a history of physical or sexual abuse.¹² Changes in physiologic processes (eg, neurochemical, endocrine) that occur with PTSD interfere with the sexual response cycle, and sexual traumas specifically have been linked with pelvic floor dysfunction.^13,14 Additionally, when PTSD is caused by a sexual trauma, even consensual sexual encounters can trigger flashbacks, intrusive memories, hyperarousal, and muscle tension that interfere with the sexual response cycle and contribute to genital pain.¹³

Vaginismus is both a physiologic and psychological contributor to dyspareunia.^1,2,4 Patients experiencing pain can develop anxiety about repeated pain and involuntarily contract their pelvic muscles, thereby creating more pain, increasing anxiety, decreasing lubrication, and causing pelvic floor dysfunction.^1-4,6 Consequently, all patients with dyspareunia should be assessed and continually monitored for symptoms of vaginismus.

Continue to: Anxiety

Anxiety. As with other pain disorders, anxiety develops around pain triggers.^10,15 When expecting sexual activity, patients can experience extreme worry and panic attacks.^10,15,16 The distress of sexual encounters can interfere with physiologic arousal and sexual desire, impacting all phases of the sexual response cycle.^1,2

Relationship issues. Difficulty engaging in or avoidance of sexual activity can interfere with romantic relationships.^2,10,16 Severe pain or vaginismus contractions can prevent penetration, leading to unconsummated marriages and an inability to conceive through intercourse.¹⁰ The distress surrounding sexual encounters can precipitate erectile dysfunction in male partners, or partners may continue to demand sexual encounters despite the patient’s pain, further impacting the relationship and heightening sexual distress.¹⁰ These stressors have led to relationships ending, patients reluctantly agreeing to nonmonogamy to appease their partners, and patients avoiding relationships altogether.^10,16

Devalued self-image. Difficulties with sexuality and relationships impact the self-image of patients with dyspareunia. Diminished self-image may include feeling “inadequate” as a woman and as a sexual partner, or feeling like a “failure.”¹⁶ Women with dyspareunia often have more distress related to their body image, physical appearance, and genital self-image than do women without genital pain.¹⁷ Feeling resentment toward their body, or feeling “ugly,” embarrassed, shamed, “broken,” and “useless” also contribute to increased depressive symptoms found in patients with dyspareunia.^16,18

Making the diagnosis

Most patients do not report symptoms unless directly asked^2,7; therefore, it is recommended that all patients be screened as a part of an initial intake and before any genital exam (TABLE 2^{2-4,6,7,9,11,19,20}).^4,7,21 If this screen is positive, a separate appointment may be needed for a thorough evaluation and before any attempt is made at a genital exam.^4,7

Items to include in the clinical interview

Given the range of possible causes of dyspareunia and its contributing factors and symptoms, a thorough clinical interview is essential. Begin with a review of the patient’s complete medical and surgical history to identify possible known contributors to genital pain.⁴ Pregnancy history is of particular importance as the prevalence of postpartum dyspareunia is 35%, with risk being greater for patients who experienced dyspareunia symptoms before pregnancy.²²

Consider using a measure such as the Female Sexual Function Index or the McGill Pain Questionnaire to help patients more thoroughly describe their symptoms.

Knowing the location and quality of pain is important for differentiating between possible diagnoses, as is specifying dyspareunia as lifelong or acquired, superficial or deep, and primary or secondary.^1-4,6 Confirm the specific location(s) of pain—eg, at the introitus, in the vestibule, on the labia, in the perineum, or near the clitoris.^2,4,6 A diagram or model may be needed to help patients to localize pain.⁴

To help narrow the differential, include the following elements in your assessment: pain quality, timing (eg, initial onset, episode onset, episode duration, situational triggers), alleviating factors, symptoms in surrounding structures (eg, bladder, bowel, muscles, bones), sexual history, other areas of sexual functioning, history of psychological trauma, relationship effects, and mental health (TABLE 2^{2-4,6,7,9,11,19,20} and Table 3^23-28). Screening for a history of sexual trauma is particularly important, as a recent systematic review and meta-analysis found that women with a history of sexual assault had a 42% higher risk of gynecologic problems overall, a 74% higher risk of dyspareunia, and a 71% higher risk of vaginismus than women without a history of sexual assault.²⁹ Using measures such as the Female Sexual Function Index or the McGill Pain Questionnaire can help patients more thoroughly describe their symptoms (TABLE 3^23-28).³

Continue to: Guidelines for the physical exam

Before the exam, ensure the patient has not used any topical genital treatment in the past 2 weeks that may interfere with sensitivity to the exam.⁴ To decrease patients’ anxiety about the exam, remind them that they can stop the exam at any time.⁷ Also consider offering the use of a mirror to better pinpoint the location of pain, and to possibly help the patient learn more about her anatomy.^2,7

Begin the exam by palpating surrounding areas that may be involved in pain, including the abdomen and musculoskeletal features.^3,6,19 Next visually inspect the external genitalia for lesions, abrasions, discoloration, erythema, or other abnormal findings.^2,3,6 Ask the patient for permission before contacting the genitals. Because the labia may be a site of pain, apply gentle pressure in retracting it to fully examine the vestibule.^6,7 Contraction of the pelvic floor muscles during approach or initial palpation could signal possible vaginismus.⁴

After visual inspection of external genitalia, use a cotton swab to map the vulva and vestibule in a clockwise fashion to precisely identify any painful locations.^2-4,6 If the patient’s history of pain has been intermittent, it’s possible that the cotton swab will not elicit pain on the day of the initial exam, but it may on other days.⁴

Begin the internal exam by inserting a single finger into the first inch of the vagina and have the patient squeeze and release to assess tenderness, muscle tightness, and control.^2,6 Advance the finger further into the vagina and palpate clockwise, examining the levator muscles, obturator muscles, rectum, urethra, and bladder for abnormal tightness or reproduction of pain.^2,4,6 Complete a bimanual exam to evaluate the pelvic organs and adnexa.^2,4 If indicated, a more thorough evaluation of pelvic floor musculature can be performed by a physical therapist or gynecologist who specializes in pelvic pain.^2-4

If the patient consents to further evaluation, consider using a small speculum, advanced slowly, for further internal examination, noting any lesions, abrasions, discharge, ectropion, or tenderness.^2-4,7 A rectal exam may also be needed in cases of deep dyspareunia.⁶ Initial work-up may include a potassium hydroxide wet prep, sexually transmitted infection testing, and pelvic ultrasound.^2,4 In some cases, laparoscopy or biopsy may be needed.^2,4

Treatments for common causes

Treatment often begins with education about anatomy, to help patients communicate about symptoms and engage more fully in their care.³ Additional education may be needed on genital functioning and the necessity of adequate stimulation and lubrication prior to penetration.^1,2,9-11 A discussion of treatments for the wide range of possible causes of dyspareunia is outside the scope of this article. However, some basic behavioral changes may help patients address some of the more common contributing factors.

A recent systematic review and meta-analysis found that women with a history of sexual assault had a 74% higher risk of dyspareunia than women without such a history.

For example, if vaginal infection is suspected, advise patients to discontinue the use of harsh soaps, known vaginal irritants (eg, perfumed products, bath additives), and douches.³ Recommend using only ­preservative- and alcohol-free lubricants for sexual contact, and avoiding lubricants with added functions (eg, warming).³ It’s worth noting that avoidance of tight clothing and thong underwear due to possible risk for infections may not be necessary. A recent study found that women who frequently wore thong underwear (more than half of the time) were no more likely to develop urinary tract infections, yeast vaginitis, or bacterial vaginosis than those who avoid such items.³⁰ However, noncotton underwear fabric, rather than tightness, was associated with yeast vaginitis³⁰; therefore, patients may want to consider using only breathable underwear.³

Continue to: Medication

Medication. Medication may be used to treat the underlying contributing conditions or the symptom of pain directly. Some common options are particularly important for patients whose dyspareunia does not have an identifiable cause. These medications include anti-inflammatory agents, topical anesthetics, tricyclic antidepressants, and hormonal treatments.^2-4 Since effectiveness varies based on subtypes of pain, select a medication according to the location, timing, and hypothesized mechanism of pain.^3,31,32

Medication for deep pain. A meta-analysis and systematic review found that patients with some types of chronic pelvic pain with pain deep in the vagina or pelvis experienced greater than 50% reduction in pain using medroxyprogesterone acetate compared with placebo.³³ Other treatments for deep pain depend on physical exam findings.

Medication for superficial pain. Many remedies have been tried, with at least 26 different treatments for vulvodynia pain alone.¹⁶ Only some of these treatments have supporting evidence. For patients with vulvar pain, an intent-to-treat RCT found that patients using a topical steroid experienced a 23% reduction in pain from pre-treatment to 6-month follow-up.³²

Surgery is also effective for vulvar pain.^34,35 For provoked vestibulodynia (in which pain is localized to the vestibule and triggered by contact with the vulva), or vulvar vestibulitis, RCTs have found that vestibulectomy has stronger effects on pain than other treatments,^31,35 with a 53% reduction in pain during intercourse and a 70% reduction in vestibular pain overall.³⁵ However, while vestibulectomy is effective for provoked vestibulodynia, it is not recommended for generalized vulvodynia, in which pain is diffuse across the vulva and occurs without vulvar contact.³⁴

Unsupported treatments. A number of other treatments have not yet been found effective. Although lidocaine for vulvar pain is often used, RCTs have not found any significant reduction in symptoms, and a ­double-blind RCT found that lidocaine ointment actually performed worse than placebo.^31,34 Similarly, oral tricyclics have not been found to decrease vulvar pain more than placebo in double-blind studies.^31,34 Furthermore, a meta-analysis of RCTs comparing treatments with placebo for vestibular pain found no significant decrease in dyspareunia for topical conjugated estrogen, topical lidocaine, oral desipramine, oral desipramine with topical lidocaine, laser therapy, or transcranial direct current.³²

Tx risks to consider. Risks and benefits of dyspareunia treatment options should be thoroughly weighed and discussed with the patient.^2-4 Vestibulectomy, despite reducing pain for many patients, has led to increased pain for 9% of patients who underwent the procedure.³⁵ Topical treatments may lead to allergic reactions, inflammation, and worsening of symptoms,⁴ and hormonal treatments have been found to increase the risk of weight gain and bloating and are not appropriate for patients trying to conceive.³³

Coordinate care with other providers

While medications and surgery can reduce pain, they have not been shown to improve other aspects of sexual functioning such as sexual satisfaction, frequency of sexual intercourse, or overall sense of sexual functioning.³⁵ Additionally, pain reduction does not address muscle tension, anxiety, self-­esteem, and relationship problems. As a result, a multidisciplinary approach is generally needed.^3,4,32,33

Continue to: Physical therapists

Physical therapists. Pelvic floor physical therapists are often members of the dyspareunia treatment team and can provide a thorough evaluation and treatment of pelvic floor disorders.^2-4 An RCT with intent-to-treat analysis found that pain was reduced by 71% following pelvic floor physical therapy.³⁶ Another RCT found that 90% of patients reported a clinically meaningful decrease in pain with pelvic floor physical therapy.³⁷ In addition to addressing pain, pelvic floor physical therapy has also been found to improve sexual functioning, sexual satisfaction, distress, and patient perception of improvement.^34,36,37

Behavioral health specialists. Psychotherapists, especially those trained in sex therapy, couples therapy, or cognitive behavioral therapy (CBT), are also typically on the treatment team. Multiple RCTs have found evidence of CBT’s effectiveness in the direct treatment of dyspareunia pain. Bergeron et al³⁵ found a 37.5% reduction in vulvar vestibulitis pain intensity during intercourse after patients completed group CBT. Another intent-to-treat RCT found that patients receiving CBT experienced more pain reduction (~ 30%) than patients who were treated with a topical steroid.³⁸

In addition to having a direct impact on pain, CBT has also been found to have a clinically and statistically significant positive impact on other aspects of sexual experience, such as overall sexuality, self-efficacy, overall sexual functioning, frequency of intercourse, and catastrophizing.^34,38 A recent meta-­analysis of RCTs found that about 80% of vaginismus patients were able to achieve penetrative intercourse after treatment with behavioral sex therapy or CBT.³⁹ This success rate was not exceeded by physical or surgical treatments.³⁹

When PTSD is thought to be a contributing factor, trauma therapy will likely be needed in addition to treatments for dyspareunia. First-line treatments for PTSD include cognitive processing therapy, prolonged exposure, trauma-focused CBT, and cognitive therapy.⁴⁰

Chronic pelvic pain with pain deep in the vagina or pelvis has been reduced by > 50%, compared with placebo, using medroxyprogesterone acetate.

Psychotherapists can also help patients reduce anxiety, reintroduce sexual contact without triggering pain or anxiety, address emotional and self-esteem effects of dyspareunia, address relationship issues, and refocus sexual encounters on pleasure rather than pain avoidance.^2-4 Despite patient reports of high treatment satisfaction following therapy,³⁸ many patients may initially lack confidence in psychotherapy as a treatment for pain³⁵ and may need to be educated on its effectiveness and multidimensional benefits.

Gynecologists. Often a gynecologist with specialization in pelvic pain is an essential member of the team for diagnostic clarification, recommendation of treatment options, and performance of more advanced treatments.^2,3 If pain has become chronic, the patient may also benefit from a pain management team and support groups.^2,3

Follow-up steps

Patients who screen negative for dyspareunia should be re-screened periodically. Continue to assess patients diagnosed with dyspareunia for vaginismus symptoms (if they are not initially present) to ensure that the treatment plan is appropriately adjusted. Once treatment has begun, ask about adverse effects and confidence in the treatment plan to minimize negative impacts on treatment adherence and to anticipate a need for a change in the treatment approach.^31,35 In addition to tracking treatment effects on pain, continue to assess for patient-centered outcomes such as emotional functioning, self-esteem, and sexual and relationship satisfaction.³⁴ The Female Sexual Function Index can be a useful tool to track symptoms.^27,34

Finally, patients who do not experience sufficient improvement in symptoms and functioning with initial treatment may need continued support and encouragement. Given the broad range of contributing factors and the high number of potential treatments, patients may find hope in learning that multiple other treatment options may be available.

CORRESPONDENCE
Adrienne A. Williams, PhD, Department of Family and Community Medicine, University of Illinois at Chicago College of Medicine, 1919 W Taylor Street, MC 663, Chicago, IL 60612; [email protected]

Dyspareunia is persistent or recurrent pain before, during, or after sexual contact and is not limited to cisgender individuals or vaginal intercourse.^1-3 With a prevalence as high as 45% in the United States,^2-5 it is one of the most common complaints in gynecologic practices.^5,6

Causes and contributing factors

There are many possible causes of dyspareunia.^2,4,6 While some patients have a single cause, most cases are complex, with multiple overlapping causes and maintaining factors.^4,6 Identifying each contributing factor can help you appropriately address all components.

Physical conditions. The range of physical contributors to dyspareunia includes inflammatory processes, structural abnormalities, musculoskeletal dysfunctions, pelvic organ disorders, injuries, iatrogenic effects, infections, allergic reactions, sensitization, hormonal changes, medication effects, adhesions, autoimmune disorders, and other pain syndromes (TABLE 1^2-4,6-11).

Inadequate arousal. One of the primary causes of pain during vaginal penetration is inadequate arousal and lubrication.^1,2,9-11 Arousal is the phase of the sexual response cycle that leads to genital tumescence and prepares the genitals for sexual contact through penile/clitoral erection, vaginal engorgement, and lubrication, which prevents pain and enhances pleasurable sensation.^9-11

While some physical conditions can lead to an inability to lubricate, the most common causes of inadequate lubrication are psychosocial-behavioral, wherein patients have the same physical ability to lubricate as patients without genital pain but do not progress through the arousal phase.^9-11 Behavioral factors such as inadequate or ineffective foreplay can fail to produce engorgement and lubrication, while psychosocial factors such as low attraction to partner, relationship stressors, anxiety, or low self-esteem can have an inhibitory effect on sexual arousal.^1,2,9-11 Psychosocial and behavioral factors may also be maintaining factors or consequences of dyspareunia, and need to be assessed and treated.^1,2,9-11

Psychological trauma. Exposure to psychological traumas and the development of posttraumatic stress disorder (PTSD) have been linked with the development of pain disorders in general and dyspareunia specifically. Most patients seeking treatment for chronic pain disorders have a history of physical or sexual abuse.¹² Changes in physiologic processes (eg, neurochemical, endocrine) that occur with PTSD interfere with the sexual response cycle, and sexual traumas specifically have been linked with pelvic floor dysfunction.^13,14 Additionally, when PTSD is caused by a sexual trauma, even consensual sexual encounters can trigger flashbacks, intrusive memories, hyperarousal, and muscle tension that interfere with the sexual response cycle and contribute to genital pain.¹³

Vaginismus is both a physiologic and psychological contributor to dyspareunia.^1,2,4 Patients experiencing pain can develop anxiety about repeated pain and involuntarily contract their pelvic muscles, thereby creating more pain, increasing anxiety, decreasing lubrication, and causing pelvic floor dysfunction.^1-4,6 Consequently, all patients with dyspareunia should be assessed and continually monitored for symptoms of vaginismus.

Continue to: Anxiety

Anxiety. As with other pain disorders, anxiety develops around pain triggers.^10,15 When expecting sexual activity, patients can experience extreme worry and panic attacks.^10,15,16 The distress of sexual encounters can interfere with physiologic arousal and sexual desire, impacting all phases of the sexual response cycle.^1,2

Relationship issues. Difficulty engaging in or avoidance of sexual activity can interfere with romantic relationships.^2,10,16 Severe pain or vaginismus contractions can prevent penetration, leading to unconsummated marriages and an inability to conceive through intercourse.¹⁰ The distress surrounding sexual encounters can precipitate erectile dysfunction in male partners, or partners may continue to demand sexual encounters despite the patient’s pain, further impacting the relationship and heightening sexual distress.¹⁰ These stressors have led to relationships ending, patients reluctantly agreeing to nonmonogamy to appease their partners, and patients avoiding relationships altogether.^10,16

Devalued self-image. Difficulties with sexuality and relationships impact the self-image of patients with dyspareunia. Diminished self-image may include feeling “inadequate” as a woman and as a sexual partner, or feeling like a “failure.”¹⁶ Women with dyspareunia often have more distress related to their body image, physical appearance, and genital self-image than do women without genital pain.¹⁷ Feeling resentment toward their body, or feeling “ugly,” embarrassed, shamed, “broken,” and “useless” also contribute to increased depressive symptoms found in patients with dyspareunia.^16,18

Making the diagnosis

Most patients do not report symptoms unless directly asked^2,7; therefore, it is recommended that all patients be screened as a part of an initial intake and before any genital exam (TABLE 2^{2-4,6,7,9,11,19,20}).^4,7,21 If this screen is positive, a separate appointment may be needed for a thorough evaluation and before any attempt is made at a genital exam.^4,7

Items to include in the clinical interview

Given the range of possible causes of dyspareunia and its contributing factors and symptoms, a thorough clinical interview is essential. Begin with a review of the patient’s complete medical and surgical history to identify possible known contributors to genital pain.⁴ Pregnancy history is of particular importance as the prevalence of postpartum dyspareunia is 35%, with risk being greater for patients who experienced dyspareunia symptoms before pregnancy.²²

Consider using a measure such as the Female Sexual Function Index or the McGill Pain Questionnaire to help patients more thoroughly describe their symptoms.

Knowing the location and quality of pain is important for differentiating between possible diagnoses, as is specifying dyspareunia as lifelong or acquired, superficial or deep, and primary or secondary.^1-4,6 Confirm the specific location(s) of pain—eg, at the introitus, in the vestibule, on the labia, in the perineum, or near the clitoris.^2,4,6 A diagram or model may be needed to help patients to localize pain.⁴

To help narrow the differential, include the following elements in your assessment: pain quality, timing (eg, initial onset, episode onset, episode duration, situational triggers), alleviating factors, symptoms in surrounding structures (eg, bladder, bowel, muscles, bones), sexual history, other areas of sexual functioning, history of psychological trauma, relationship effects, and mental health (TABLE 2^{2-4,6,7,9,11,19,20} and Table 3^23-28). Screening for a history of sexual trauma is particularly important, as a recent systematic review and meta-analysis found that women with a history of sexual assault had a 42% higher risk of gynecologic problems overall, a 74% higher risk of dyspareunia, and a 71% higher risk of vaginismus than women without a history of sexual assault.²⁹ Using measures such as the Female Sexual Function Index or the McGill Pain Questionnaire can help patients more thoroughly describe their symptoms (TABLE 3^23-28).³

Continue to: Guidelines for the physical exam

Before the exam, ensure the patient has not used any topical genital treatment in the past 2 weeks that may interfere with sensitivity to the exam.⁴ To decrease patients’ anxiety about the exam, remind them that they can stop the exam at any time.⁷ Also consider offering the use of a mirror to better pinpoint the location of pain, and to possibly help the patient learn more about her anatomy.^2,7

Begin the exam by palpating surrounding areas that may be involved in pain, including the abdomen and musculoskeletal features.^3,6,19 Next visually inspect the external genitalia for lesions, abrasions, discoloration, erythema, or other abnormal findings.^2,3,6 Ask the patient for permission before contacting the genitals. Because the labia may be a site of pain, apply gentle pressure in retracting it to fully examine the vestibule.^6,7 Contraction of the pelvic floor muscles during approach or initial palpation could signal possible vaginismus.⁴

After visual inspection of external genitalia, use a cotton swab to map the vulva and vestibule in a clockwise fashion to precisely identify any painful locations.^2-4,6 If the patient’s history of pain has been intermittent, it’s possible that the cotton swab will not elicit pain on the day of the initial exam, but it may on other days.⁴

Begin the internal exam by inserting a single finger into the first inch of the vagina and have the patient squeeze and release to assess tenderness, muscle tightness, and control.^2,6 Advance the finger further into the vagina and palpate clockwise, examining the levator muscles, obturator muscles, rectum, urethra, and bladder for abnormal tightness or reproduction of pain.^2,4,6 Complete a bimanual exam to evaluate the pelvic organs and adnexa.^2,4 If indicated, a more thorough evaluation of pelvic floor musculature can be performed by a physical therapist or gynecologist who specializes in pelvic pain.^2-4

If the patient consents to further evaluation, consider using a small speculum, advanced slowly, for further internal examination, noting any lesions, abrasions, discharge, ectropion, or tenderness.^2-4,7 A rectal exam may also be needed in cases of deep dyspareunia.⁶ Initial work-up may include a potassium hydroxide wet prep, sexually transmitted infection testing, and pelvic ultrasound.^2,4 In some cases, laparoscopy or biopsy may be needed.^2,4

Treatments for common causes

Treatment often begins with education about anatomy, to help patients communicate about symptoms and engage more fully in their care.³ Additional education may be needed on genital functioning and the necessity of adequate stimulation and lubrication prior to penetration.^1,2,9-11 A discussion of treatments for the wide range of possible causes of dyspareunia is outside the scope of this article. However, some basic behavioral changes may help patients address some of the more common contributing factors.

A recent systematic review and meta-analysis found that women with a history of sexual assault had a 74% higher risk of dyspareunia than women without such a history.

For example, if vaginal infection is suspected, advise patients to discontinue the use of harsh soaps, known vaginal irritants (eg, perfumed products, bath additives), and douches.³ Recommend using only ­preservative- and alcohol-free lubricants for sexual contact, and avoiding lubricants with added functions (eg, warming).³ It’s worth noting that avoidance of tight clothing and thong underwear due to possible risk for infections may not be necessary. A recent study found that women who frequently wore thong underwear (more than half of the time) were no more likely to develop urinary tract infections, yeast vaginitis, or bacterial vaginosis than those who avoid such items.³⁰ However, noncotton underwear fabric, rather than tightness, was associated with yeast vaginitis³⁰; therefore, patients may want to consider using only breathable underwear.³

Continue to: Medication

Medication. Medication may be used to treat the underlying contributing conditions or the symptom of pain directly. Some common options are particularly important for patients whose dyspareunia does not have an identifiable cause. These medications include anti-inflammatory agents, topical anesthetics, tricyclic antidepressants, and hormonal treatments.^2-4 Since effectiveness varies based on subtypes of pain, select a medication according to the location, timing, and hypothesized mechanism of pain.^3,31,32

Medication for deep pain. A meta-analysis and systematic review found that patients with some types of chronic pelvic pain with pain deep in the vagina or pelvis experienced greater than 50% reduction in pain using medroxyprogesterone acetate compared with placebo.³³ Other treatments for deep pain depend on physical exam findings.

Medication for superficial pain. Many remedies have been tried, with at least 26 different treatments for vulvodynia pain alone.¹⁶ Only some of these treatments have supporting evidence. For patients with vulvar pain, an intent-to-treat RCT found that patients using a topical steroid experienced a 23% reduction in pain from pre-treatment to 6-month follow-up.³²

Surgery is also effective for vulvar pain.^34,35 For provoked vestibulodynia (in which pain is localized to the vestibule and triggered by contact with the vulva), or vulvar vestibulitis, RCTs have found that vestibulectomy has stronger effects on pain than other treatments,^31,35 with a 53% reduction in pain during intercourse and a 70% reduction in vestibular pain overall.³⁵ However, while vestibulectomy is effective for provoked vestibulodynia, it is not recommended for generalized vulvodynia, in which pain is diffuse across the vulva and occurs without vulvar contact.³⁴

Unsupported treatments. A number of other treatments have not yet been found effective. Although lidocaine for vulvar pain is often used, RCTs have not found any significant reduction in symptoms, and a ­double-blind RCT found that lidocaine ointment actually performed worse than placebo.^31,34 Similarly, oral tricyclics have not been found to decrease vulvar pain more than placebo in double-blind studies.^31,34 Furthermore, a meta-analysis of RCTs comparing treatments with placebo for vestibular pain found no significant decrease in dyspareunia for topical conjugated estrogen, topical lidocaine, oral desipramine, oral desipramine with topical lidocaine, laser therapy, or transcranial direct current.³²

Tx risks to consider. Risks and benefits of dyspareunia treatment options should be thoroughly weighed and discussed with the patient.^2-4 Vestibulectomy, despite reducing pain for many patients, has led to increased pain for 9% of patients who underwent the procedure.³⁵ Topical treatments may lead to allergic reactions, inflammation, and worsening of symptoms,⁴ and hormonal treatments have been found to increase the risk of weight gain and bloating and are not appropriate for patients trying to conceive.³³

Coordinate care with other providers

While medications and surgery can reduce pain, they have not been shown to improve other aspects of sexual functioning such as sexual satisfaction, frequency of sexual intercourse, or overall sense of sexual functioning.³⁵ Additionally, pain reduction does not address muscle tension, anxiety, self-­esteem, and relationship problems. As a result, a multidisciplinary approach is generally needed.^3,4,32,33

Continue to: Physical therapists

Physical therapists. Pelvic floor physical therapists are often members of the dyspareunia treatment team and can provide a thorough evaluation and treatment of pelvic floor disorders.^2-4 An RCT with intent-to-treat analysis found that pain was reduced by 71% following pelvic floor physical therapy.³⁶ Another RCT found that 90% of patients reported a clinically meaningful decrease in pain with pelvic floor physical therapy.³⁷ In addition to addressing pain, pelvic floor physical therapy has also been found to improve sexual functioning, sexual satisfaction, distress, and patient perception of improvement.^34,36,37

Behavioral health specialists. Psychotherapists, especially those trained in sex therapy, couples therapy, or cognitive behavioral therapy (CBT), are also typically on the treatment team. Multiple RCTs have found evidence of CBT’s effectiveness in the direct treatment of dyspareunia pain. Bergeron et al³⁵ found a 37.5% reduction in vulvar vestibulitis pain intensity during intercourse after patients completed group CBT. Another intent-to-treat RCT found that patients receiving CBT experienced more pain reduction (~ 30%) than patients who were treated with a topical steroid.³⁸

In addition to having a direct impact on pain, CBT has also been found to have a clinically and statistically significant positive impact on other aspects of sexual experience, such as overall sexuality, self-efficacy, overall sexual functioning, frequency of intercourse, and catastrophizing.^34,38 A recent meta-­analysis of RCTs found that about 80% of vaginismus patients were able to achieve penetrative intercourse after treatment with behavioral sex therapy or CBT.³⁹ This success rate was not exceeded by physical or surgical treatments.³⁹

When PTSD is thought to be a contributing factor, trauma therapy will likely be needed in addition to treatments for dyspareunia. First-line treatments for PTSD include cognitive processing therapy, prolonged exposure, trauma-focused CBT, and cognitive therapy.⁴⁰

Chronic pelvic pain with pain deep in the vagina or pelvis has been reduced by > 50%, compared with placebo, using medroxyprogesterone acetate.

Psychotherapists can also help patients reduce anxiety, reintroduce sexual contact without triggering pain or anxiety, address emotional and self-esteem effects of dyspareunia, address relationship issues, and refocus sexual encounters on pleasure rather than pain avoidance.^2-4 Despite patient reports of high treatment satisfaction following therapy,³⁸ many patients may initially lack confidence in psychotherapy as a treatment for pain³⁵ and may need to be educated on its effectiveness and multidimensional benefits.

Gynecologists. Often a gynecologist with specialization in pelvic pain is an essential member of the team for diagnostic clarification, recommendation of treatment options, and performance of more advanced treatments.^2,3 If pain has become chronic, the patient may also benefit from a pain management team and support groups.^2,3

Follow-up steps

Patients who screen negative for dyspareunia should be re-screened periodically. Continue to assess patients diagnosed with dyspareunia for vaginismus symptoms (if they are not initially present) to ensure that the treatment plan is appropriately adjusted. Once treatment has begun, ask about adverse effects and confidence in the treatment plan to minimize negative impacts on treatment adherence and to anticipate a need for a change in the treatment approach.^31,35 In addition to tracking treatment effects on pain, continue to assess for patient-centered outcomes such as emotional functioning, self-esteem, and sexual and relationship satisfaction.³⁴ The Female Sexual Function Index can be a useful tool to track symptoms.^27,34

Finally, patients who do not experience sufficient improvement in symptoms and functioning with initial treatment may need continued support and encouragement. Given the broad range of contributing factors and the high number of potential treatments, patients may find hope in learning that multiple other treatment options may be available.

CORRESPONDENCE
Adrienne A. Williams, PhD, Department of Family and Community Medicine, University of Illinois at Chicago College of Medicine, 1919 W Taylor Street, MC 663, Chicago, IL 60612; [email protected]

CASE

Jack Z, a 75-year-old man with well-controlled hypertension, diabetes controlled by diet, and atrial fibrillation (AF) presents to the family medicine clinic to establish care with you after moving to the community from out of town.

The patient describes a 1-year history of AF. He provides you with an echocardiography report from 6 months ago that shows no evidence of structural heart disease. He takes lisinopril, to control blood pressure; an anticoagulant; a beta-blocker; and amiodarone for rhythm control. Initially, he took flecainide, which was ineffective for rhythm control, before being switched to amiodarone. He had 2 cardioversion procedures, each time after episodes of symptoms. He does not smoke or drink alcohol.

Mr. Z describes worsening palpitations and shortness of breath over the past 9 months. Symptoms now include episodes of exertional fatigue, even when he is not having palpitations. Prior to the episodes of worsening symptoms, he tells you that he lived a “fairly active” life, golfing twice a week.

The patient’s previous primary care physician had encouraged him to talk to his cardiologist about “other options” for managing AF, because levels of his liver enzymes had started to rise (a known adverse effect of amiodarone¹) when measured 3 months ago. He did not undertake that conversation, but asks you now about other treatments for AF.

Atrial fibrillation is the most common sustained cardiac arrhythmia, characterized by discordant electrical activation of the atria due to structural or electrophysiological abnormalities, or both. The disorder is associated with an increased rate of stroke and heart failure and is independently associated with a 1.5- to 2-fold risk of all-cause mortality.²

In this article, we review the pathophysiology of AF; management, including the role of, and indications for, catheter ablation; and patient- and disease-related factors associated with ablation (including odds of success, complications, risk of recurrence, and continuing need for thromboprophylaxis) that family physicians should consider when contemplating referral to a cardiologist or electrophysiologist for catheter ablation for AF.

What provokes AF?

AF is thought to occur as a result of an interaction among 3 phenomena:

• enhanced automaticity of abnormal atrial tissue
• triggered activity of ectopic foci within 1 or more pulmonary veins, lying within the left atrium
• re-entry, in which there is propagation of electrical impulses from an ectopic beat through another pathway.

Continue to: In patients who progress...

In patients who progress from paroxysmal to persistent AF (see “Subtypes,” below), 2 distinct pathways, facilitated by the presence of abnormal tissue, continuously activate one another, thus maintaining the arrhythmia. Myocardial tissue in the pulmonary veins is responsible for most ectopic electrical impulses in patients with drug-refractory AF (see “Rhythm control”).

Subtypes. For the purpose of planning treatment, AF is classified as:

• Paroxysmal. Terminates spontaneously or with intervention ≤ 7 days after onset.
• Persistent. Continuous and sustained for > 7 days.
• Longstanding persistent. Continuous for > 12 months.
• Permanent. The patient and physician accept that there will be no further attempt to restore or maintain sinus rhythm.

Goals of treatment

Primary management goals in patients with AF are 2-fold: control of symptoms and prevention of thromboembolism. A patient with new-onset AF who presents acutely with inadequate rate control and hemodynamic compromise requires urgent assessment to determine the cause of the arrhythmia and need for cardioversion.³ A symptomatic patient with AF who does not have high-risk features (eg, valvular heart disease, mechanical valves) might be a candidate for rhythm control in addition to rate control.^3,4

Rate control. After evaluation in the hospital, a patient who has a rapid ventricular response but remains hemodynamically stable, without evidence of heart failure, should be initiated on a rate-controlling medication, such as a beta-blocker or nondihydropyridine calcium-channel blocker. A resting heart rate goal of < 80 beats per minute (bpm) is recommended for a symptomatic patient with AF. The heart rate goal can be relaxed, to < 110 bpm, in an asymptomatic patient with preserved left ventricular function.^5,6

Rhythm control, indicated in patients who remain symptomatic on rate-controlling medication, can be achieved either with an antiarrhythmic drug (AAD) or by catheter ablation.^4,5 In stable patients, rhythm control should be considered only after a thorough work-up for a reversible cause of AF, and can be achieved with an oral AAD or, in select patients, through catheter ablation (TABLE 1^3,6). Other indications for chronic rhythm control include treatment of patients with ­tachycardia-induced cardiomyopathy.⁵

Continuing symptoms when taking rate-controlling medication for atrial fibrillation is one indication for rhythm control with an antiarrhythmic drug or by catheter ablation.

A major study that documented the benefit of early rhythm control evaluated long-term outcomes in 2789 patients with AF who were undergoing catheter ablation.⁷ Patients were randomized to early rhythm control (catheter ablation or AAD) or “usual care”—ie, in this study, rhythm control limited to symptomatic patients. Primary outcomes were death from cardiovascular causes, stroke, and hospitalization with worsening heart failure or acute coronary syndrome. A first primary outcome event occurred in 249 patients (3.9/100 person-years) assigned to early rhythm control, compared to 316 (5.0 per 100 person-years) in the group assigned to usual care.

The study was terminated early (after 5.1 years) because of overwhelming evidence of efficacy (number need to treat = 7). Although early rhythm control was obtained through both catheter ablation and AAD (hazard ratio [HR] = 0.79; 96% CI, 0.66-0.94; P = .005), success was attributed to the use of catheter ablation for a rhythm-control strategy and its use among patients whose AF was present for < 1 year. Most patients in both treatment groups continued to receive anticoagulation, rate control, and optimization of cardiovascular risk.⁷

Continue to: Notably, direct studies...

Notably, direct studies comparing ablation and AAD have not confirmed the benefit of ablation over AAD in outcomes of all-cause mortality, bleeding, stroke, or cardiac arrest over a 5-year period.⁸

Adverse effects and mortality outcomes with AAD. Concern over using AAD for rhythm control is based mostly on adverse effects and long-term (1-year) mortality outcomes. Long-term AAD therapy has been shown to decrease the recurrence of AF—but without evidence to suggest other mortality benefits.

A meta-analysis of 59 randomized controlled trials reviewed 20,981 patients receiving AAD (including quinidine, disopyramide, propafenone, flecainide, metoprolol, amiodarone, dofetilide, dronedarone, and sotalol) for long-term effects on death, stroke, adverse reactions, and recurrence of AF.⁹ Findings at 10 months suggest that:

• Compared to placebo, amiodarone and sotalol increased the risk of all-cause mortality during the study period.
• There was minimal difference in mortality among patients taking dofetilide or dronedarone, compared to placebo.
• There were insufficient data to draw conclusions about the effect of disopyramide, flecainide, and propafenone on mortality.

Before starting a patient on AAD, the risk of arrhythmias and the potential for these agents to cause toxicity and adverse events should always be discussed.

CASE

You tell Mr. Z that you need to know the status of his comorbidities to make a recommendation about “other” management options, and proceed to take a detailed history.

Recent history. Mr. Z reveals that “today is a good day”: He has had “only 1” episode of palpitations, which resolved on its own. The previous episode, he explains, was 3 days ago, when palpitations were associated with lightheadedness and shortness of breath. He denies chest pains or swelling of the legs.

Physical exam. The patient appears spry, comfortable, and in no acute distress. Vital signs are within normal limits. A body mass index of 28.4 puts him in the “overweight” category. His blood pressure is 118/75 mm Hg.

Continue to: Cardiac examination...

Cardiac examination is significant for an irregular rhythm without murmurs, rubs, or gallops. His lungs are clear bilaterally; his abdomen is soft and nondistended. His extremities show no edema.

Testing. You obtain an electrocardiogram, which demonstrates a controlled ventricular rate of 88 bpm and AF. You order a complete blood count, comprehensive metabolic panel, tests of hemoglobin A1C and thyroid-stimulating hormone, lipid panel, echocardiogram, and a chest radiograph.

Results. The chest radiograph is negative for an acute cardiopulmonary process; cardiac size is normal. Aspartate aminotransferase and alanine aminotransferase levels are higher than twice the normal limit. The echocardiogram reveals an estimated left ventricular ejection fraction of 55% to 60%; no structural abnormalities are noted.

In which AF patients is catheter ablation indicated?

Ablation is recommended for select patients (TABLE 1^3,6) with symptomatic paroxysmal AF that is refractory to AAD or who are intolerant of AAD.^3,6 It is a reasonable first-line therapy for high-performing athletes in whom AAD would affect athletic performance.^3,10 It is also a reasonable option in select patients > 75 years and as an alternative to AAD therapy.³ Finally, catheter ablation should be considered in symptomatic patients with longstanding persistent AF and congestive heart failure, with or without reduced left ventricular ejection fraction.³

CASE

You inform Mr. Z that his symptoms are likely a result of symptomatic paroxysmal AF, which was refractory to flecainide and amiodarone, and that his abnormal liver function test results preclude continued use of amiodarone. You propose Holter monitoring to correlate timing of symptoms with the arrhythmia, but he reports this has been done, and the correlation confirmed, by his previous physician.

You explain that, because the diagnosis of symptomatic paroxysmal AF refractory to AADs has been confirmed, he is categorized as a patient who might benefit from catheter ablation, based on:

• the type of AF (ie, paroxysmal AF is associated with better ablation outcomes)
• persistent symptoms that are refractory to AADs
• his intolerance of AAD
• the length of time since onset of symptoms.

Mr. Z agrees to consider your recommendation.

Continue to: What are the benefits of catheter ablation?

Ablation can be achieved through radiofrequency (RF) ablation, cryoablation, or newer, laser-based balloon ablation. Primary outcomes used to determine the success of any options for performing ablation include mortality, stroke, and hospitalization. Other endpoints include maintenance of sinus rhythm, freedom from AF, reduction in AF burden (estimated through patients’ report of symptoms, recurrence rate, need for a second ablation procedure, and serial long-term monitoring through an implantable cardiac monitoring device), quality of life, and prevention of AF progression.³

Patient and disease variables (TABLE 2^11-13). The success rate of catheter ablation, defined as freedom from either symptomatic or asymptomatic episodes of AF, is dependent on several factors,^3,14 including:

• type of AF (paroxysmal or persistent)
• duration and degree of symptoms
• age
• sex
• comorbidities, including heart failure and structural heart or lung disease.

Overall, in patients with paroxysmal AF, an estimated 75% are symptom free 1 year after ablation.¹⁵ Patients with persistent and longstanding persistent AF experience a lower success rate.

Ablation is recommended in patients with symptomatic paroxysmal atrial fibrillation that is refractory to antiarrhythmics or when they are intolerant of these drugs.

RF catheter ablation has demonstrated superiority to AAD in reducing the need for cardioversion (relative risk [RR] = 0.62; 95% CI, 0.47-0.82) and cardiac-related hospitalization (RR = 0.27; 95% CI, 0.10-0.72;) at 12 months in patients with nonparoxysmal AF (persistent or longstanding persistent).¹⁶

Effect on mortality. Among patients with heart failure with reduced ejection fraction, long-term studies of cardiovascular outcomes 5 years post ablation concluded that ablation is associated with a decrease in all-cause mortality (RR = 0.69; 95% CI, 0.54-0.88; P = .003) and a reduction in hospitalization (RR = 0.62; 95% CI, 0.47-0.82; P = .0006); younger (< 65 years) and male patients derive greater benefit.^6,17 Indications for ablation in patients with heart failure are similar to those in patients without heart failure; ablation can therefore be considered for select heart failure patients who remain symptomatic or for whom AAD has failed.³

Older patients. Ablation can be considered for patients > 75 years with symptomatic paroxysmal AF refractory to AAD or who are intolerant of AAD.³ A study that assessed the benefit of catheter ablation reviewed 587 older (> 75 years) patients with AF, of whom 324 were eligible for ablation. Endpoints were maintenance of sinus rhythm, stroke, death, and major bleeding. Return to normal sinus rhythm was an independent factor, associated with a decrease in the risk of mortality among all patient groups that underwent ablation (HR = 0.36; 95% CI, 0.2-0.63; P = .0005). Age > 75 years (HR = 1.09; 95% CI, 1.01-1.16; P > .02) and depressed ejection fraction < 40% (HR = 2.38; 95% CI, 1.28-4.4; P = .006) were determined to be unfavorable parameters for survival.¹⁸

Complications and risks

Complications of catheter ablation for AF, although infrequent, can be severe (TABLE 3³). Early mortality, defined as death during initial admission or 30-day readmission, occurs in approximately 0.5% of cases; half of deaths take place during readmission.¹¹

Continue to: Complications vary...

Complications vary, based on the type and site of ablation.^19,20 Cardiac tamponade or perforation, the most life-threatening complications, taken together occur in an estimated 1.9% of patients (odds ratio [OR] = 2.98; 95% CI, 1.36-6.56; P = .007).¹¹ Other in-hospital complications independently predictive of death include any cardiac complications (OR = 12.8; 95% CI, 6.86 to 23.8; P < .001) and neurologic complications (cerebrovascular accident and transient ischemic attack) (OR = 8.72; 95% CI, 2.71-28.1; P < .001).

Other complications that do not cause death but might prolong the hospital stay include pericarditis without effusion, anesthesia-related complications, and vascular-access complications. Patients whose ablation is performed at an institution where the volume of ablations is low are also at higher risk of early mortality (OR = 2.35; 95% CI, 1.33-4.15; P = .003).¹⁶

Recurrence is common (TABLE 2^11-13). Risk of recurrence following ablation is significant; early (within 3 months after ablation) recurrence is seen in 50% of patients.^21,22 However, this is a so-called "blanking period"—ie, a temporary period of inflammatory and proarrhythmic changes that are not predictors of later recurrence. The 5-year post-ablation recurrence rate is approximately 25.5%; longstanding persistent and persistent AF and the presence of comorbidities are major risk factors for recurrence.^13,23

Recurrence is also associated with the type of procedure; pulmonary vein isolation, alone or in combination with another type of procedure, results in higher long-term success.^21,23

Other variables affect outcome (TABLE 2^11-13). Following AF ablation, patients with nonparoxysmal AF at baseline, advanced age, sleep apnea and obesity, left atrial enlargement, and any structural heart disease tend to have a poorer long-term (5-year) outcome (ie, freedom from extended episodes of AF).^3,13,23,24

Patients who undergo repeat procedures have higher arrhythmia-free survival; the highest ablation success rate is for patients with paroxysmal AF.^13,23

Exposure to ionizing radiation. Fluoroscopy is required for multiple components of atrial mapping and ablation during RF ablation, including navigation, visualization, and monitoring of catheter placement. Patients undergoing this particular procedure therefore receive significant exposure to ionizing radiation. A reduction in, even complete elimination of, fluoroscopy has been achieved with:

• nonfluoroscopic 3-dimensional mapping systems²⁵
• intracardiac echocardiography, which utilizes ultrasonographic imaging as the primary visual mode for tracking and manipulating the catheter
• robotic guided navigation.^26-28

Continue to: CASE

CASE

At his return visit, Mr. Z says that he is concerned about, first, undergoing catheter ablation at his age and, second, the risks associated with the procedure. You explain that it is true that ablation is ideal in younger patients who have minimal comorbidities and that the risk of complications increases with age—but that there is no cutoff or absolute age contraindication to ablation.

You tell Mr. Z that you will work with him on risk-factor modification in anticipation of ablation. You also assure him that the decision whether to ablate must be a joint one—between him and a cardiologist experienced both in electrophysiology and in performing this highly technical procedure. And you explain that a highly practiced specialist can identify Mr. Z’s risk factors that might make ablation more difficult to perform and affect the long-term outcome.

With Mr. Z’s agreement, you screen for sleep apnea and start him on a lifestyle modification plan to achieve a more ideal weight, explaining that the risk of recurrence of AF after catheter ablation is increased by obesity and sleep apnea, in addition to age. You explain that, based on his CHA₂DS₂–VASc (congestive heart failure; hypertension; age, ≥ 75 years; diabetes; prior stroke, transient ischemic attack, or thromboembolism; vascular disease; age, 65 to 74 years; sex category) score of 3, he will remain on anticoagulation whether or not he has the ablation.

You refer the patient to the nearest high-volume cardiac ablation center.

Last, you caution Mr. Z that, based on his lipid levels, his 10-year risk of heart disease or stroke is elevated. You recommend treatment with a statin agent while he continues his other medications.

Delivering energy to myocardium

Myocardial tissue in pulmonary veins is responsible for most ectopic electrical impulses in patients with drug-refractory AF. The goal of catheter ablation in AF is destruction (scarring) of tissue that is the source of abnormal vein potentials.¹⁵

How RF ablation works. Ablation is most commonly performed using RF energy, a high-frequency form of electrical energy. Electrophysiology studies are carried out at the time of ablation by percutaneous, fluoroscopically guided insertion of 2 to 5 catheters, usually through the femoral or internal jugular vein, which are then positioned within several areas of the heart—usually, the right atrium, bundle of His, right ventricle, and coronary sinus.

Continue to: Electrical current...

Successful ablation is defined as freedom from symptomatic or asymptomatic AF episodes, and depends on AF type; duration and degree of symptoms; and age, sex, and comorbidities.

Electrical current is applied through the catheters from an external generator to stimulate the myocardium and thus determine its electrophysiologic properties. The anatomic and electrical activity of the left atrium and pulmonary veins is then identified, a technique known as electro-anatomical mapping (FIGURE). After arrhythmogenic myocardial tissue is mapped, ablation is carried out with RF energy through the catheter to the pathogenic myocardium from which arrhythmias are initiated or conducted. The result is thermal destruction of tissue and creation of small, shallow lesions that vary in size with the type of catheter and the force of contact pressure applied.^3,29

Other energy sources used in catheter ablation include cryothermal energy, which utilizes liquid nitrous oxide under pressure through a cryocatheter or cryoballoon catheter. Application of cryothermal energy freezes tissue and disrupts cell membranes and any electrical activity. Cryoballoon ablation has been shown to be similarly safe and efficacious as RF ablation in patients with paroxysmal AF.^30,31

Newer laser-based balloon ablations are performed under ultrasonographic guidance and utilize arcs of laser energy delivered to the pathogenic myocardium.³

Thromboembolism prophylaxis

Oral anticoagulation to decrease the risk of stroke is initiated in all patients with AF, based on a thromboembolic risk profile determined by their CHA₂DS₂–VASc score, with anticoagulation recommended when the score is ≥ 2 in men and ≥ 3 in women. Options for anticoagulation include warfarin and one of the novel oral anticoagulants dabigatran, apixaban, rivaroxaban, and edoxaban.⁴ Recommendations are as follows³:

• For patients with a CHA₂DS₂–VASc score of ≥ 2 (men) or ≥ 3 (women), anticoagulation should be continued indefinitely, regardless of how successful the ablation procedure is.
• When patients choose to discontinue anticoagulation, they should be counseled in detail about the risk of doing so. The continued need for frequent arrhythmia monitoring should be emphasized.

The route from primary careto catheter ablation

Perform a thorough evaluation. Patients who present to you with palpitations should first undergo a routine workup for AF, followed by confirmation of the diagnosis. Exclude structural heart disease with echocardiography. Undertake monitoring, which is essential to determine whether symptoms are a reflection of the arrhythmia, using noncontinuous or continuous electrocardiographic (EKG) monitoring. Noncontinuous detection devices include:

• scheduled or symptom-initiated EKG
• a Holter monitor, worn for at least 24 hours and as long as 7 days
• trans-telephonic recordings and patient- or automatically activated devices
• an external loop recorder.³²

Continuous EKG monitoring is more permanent (≥ 12 months). This is usually achieved through an implantable loop powered by a battery that lasts as long as 3 years.³

Ablation: Yes or no? Ablation is not recommended to avoid anticoagulation or when anticoagulation is contraindicated.⁵ With regard to specific patient criteria, the ideal patient:

• is symptomatic
• has failed AAD therapy
• does not have pulmonary disease
• has a normal or mildly dilated left atrium or normal or mildly reduced left ventricular ejection fraction.⁵

Continue to: There is no absolute age...

There is no absolute age or comorbidity contraindication to ablation. The patient should be referred to a cardiologist who has received appropriate training in electrophysiology, to identify comorbidities that (1) increase the technical difficulty of the procedure and baseline risk and (2) affect long-term outcome,¹² and who performs the procedure in a center that has considerable experience with catheter ablation.³³

Once the decision is made to perform ablation, you can provide strategies that optimize the outcome (freedom from AF episodes). Those tactics include weight loss and screening evaluation and, if indicated, treatment for sleep apnea.³

Protocol. Prior to the procedure, the patient fasts overnight; they might be asked to taper or discontinue cardiac medications that have electrophysiologic effects. Studies suggest a low risk of bleeding associated with catheter ablation; anticoagulation should therefore continue uninterrupted for patients undergoing catheter ablation for AF^3,4,34,35; however, this practice varies with the cardiologist or electrophysiologist performing ablation.

Patients whose ablation is performed at an institution where the volume of such procedures is low are at higher risk of early mortality.

Because of the length and complexity of the procedure, electro-anatomical mapping and ablation are conducted with the patient under general anesthesia.³ The patient is kept supine, and remains so for 2 to 4 hours afterward to allow for hemostasis at puncture sites.³

Patients might be monitored overnight, although same-day catheter ablation has been shown to be safe and cost-effective in select patients.^36,37 Post ablation, patients follow up with the cardiologist and electrophysiologist. Long-term arrhythmia monitoring is required.³ Anticoagulation is continued for at least 2 months, and is discontinued based on the patient’s risk for stroke, utilizing their CHA₂DS₂–VASc score.^3,4

CASE

At Mr. Z’s 6-month primary care follow-up, he confirms what has been reported to you as the referring physician: He had a successful catheter ablation and continues to have regular follow-up monitoring with the cardiologist. He is no longer taking amiodarone.

At this visit, he reports no recurrence of AF-associated symptoms or detectable AF on cardiac monitoring. He has lost 8 lbs. You counsel to him to continue to maintain a healthy lifestyle.

CORRESPONDENCE
Amimi S. Osayande MD, FAAFP, Northside-Gwinnett Family Medicine Residency Program, Strickland Family Medicine Center, 665 Duluth Highway, Suite 501, Lawrenceville, GA 30046; [email protected]

CASE

Jack Z, a 75-year-old man with well-controlled hypertension, diabetes controlled by diet, and atrial fibrillation (AF) presents to the family medicine clinic to establish care with you after moving to the community from out of town.

The patient describes a 1-year history of AF. He provides you with an echocardiography report from 6 months ago that shows no evidence of structural heart disease. He takes lisinopril, to control blood pressure; an anticoagulant; a beta-blocker; and amiodarone for rhythm control. Initially, he took flecainide, which was ineffective for rhythm control, before being switched to amiodarone. He had 2 cardioversion procedures, each time after episodes of symptoms. He does not smoke or drink alcohol.

Mr. Z describes worsening palpitations and shortness of breath over the past 9 months. Symptoms now include episodes of exertional fatigue, even when he is not having palpitations. Prior to the episodes of worsening symptoms, he tells you that he lived a “fairly active” life, golfing twice a week.

The patient’s previous primary care physician had encouraged him to talk to his cardiologist about “other options” for managing AF, because levels of his liver enzymes had started to rise (a known adverse effect of amiodarone¹) when measured 3 months ago. He did not undertake that conversation, but asks you now about other treatments for AF.

Atrial fibrillation is the most common sustained cardiac arrhythmia, characterized by discordant electrical activation of the atria due to structural or electrophysiological abnormalities, or both. The disorder is associated with an increased rate of stroke and heart failure and is independently associated with a 1.5- to 2-fold risk of all-cause mortality.²

In this article, we review the pathophysiology of AF; management, including the role of, and indications for, catheter ablation; and patient- and disease-related factors associated with ablation (including odds of success, complications, risk of recurrence, and continuing need for thromboprophylaxis) that family physicians should consider when contemplating referral to a cardiologist or electrophysiologist for catheter ablation for AF.

What provokes AF?

AF is thought to occur as a result of an interaction among 3 phenomena:

• enhanced automaticity of abnormal atrial tissue
• triggered activity of ectopic foci within 1 or more pulmonary veins, lying within the left atrium
• re-entry, in which there is propagation of electrical impulses from an ectopic beat through another pathway.

Continue to: In patients who progress...

In patients who progress from paroxysmal to persistent AF (see “Subtypes,” below), 2 distinct pathways, facilitated by the presence of abnormal tissue, continuously activate one another, thus maintaining the arrhythmia. Myocardial tissue in the pulmonary veins is responsible for most ectopic electrical impulses in patients with drug-refractory AF (see “Rhythm control”).

Subtypes. For the purpose of planning treatment, AF is classified as:

• Paroxysmal. Terminates spontaneously or with intervention ≤ 7 days after onset.
• Persistent. Continuous and sustained for > 7 days.
• Longstanding persistent. Continuous for > 12 months.
• Permanent. The patient and physician accept that there will be no further attempt to restore or maintain sinus rhythm.

Goals of treatment

Primary management goals in patients with AF are 2-fold: control of symptoms and prevention of thromboembolism. A patient with new-onset AF who presents acutely with inadequate rate control and hemodynamic compromise requires urgent assessment to determine the cause of the arrhythmia and need for cardioversion.³ A symptomatic patient with AF who does not have high-risk features (eg, valvular heart disease, mechanical valves) might be a candidate for rhythm control in addition to rate control.^3,4

Rate control. After evaluation in the hospital, a patient who has a rapid ventricular response but remains hemodynamically stable, without evidence of heart failure, should be initiated on a rate-controlling medication, such as a beta-blocker or nondihydropyridine calcium-channel blocker. A resting heart rate goal of < 80 beats per minute (bpm) is recommended for a symptomatic patient with AF. The heart rate goal can be relaxed, to < 110 bpm, in an asymptomatic patient with preserved left ventricular function.^5,6

Rhythm control, indicated in patients who remain symptomatic on rate-controlling medication, can be achieved either with an antiarrhythmic drug (AAD) or by catheter ablation.^4,5 In stable patients, rhythm control should be considered only after a thorough work-up for a reversible cause of AF, and can be achieved with an oral AAD or, in select patients, through catheter ablation (TABLE 1^3,6). Other indications for chronic rhythm control include treatment of patients with ­tachycardia-induced cardiomyopathy.⁵

Continuing symptoms when taking rate-controlling medication for atrial fibrillation is one indication for rhythm control with an antiarrhythmic drug or by catheter ablation.

A major study that documented the benefit of early rhythm control evaluated long-term outcomes in 2789 patients with AF who were undergoing catheter ablation.⁷ Patients were randomized to early rhythm control (catheter ablation or AAD) or “usual care”—ie, in this study, rhythm control limited to symptomatic patients. Primary outcomes were death from cardiovascular causes, stroke, and hospitalization with worsening heart failure or acute coronary syndrome. A first primary outcome event occurred in 249 patients (3.9/100 person-years) assigned to early rhythm control, compared to 316 (5.0 per 100 person-years) in the group assigned to usual care.

The study was terminated early (after 5.1 years) because of overwhelming evidence of efficacy (number need to treat = 7). Although early rhythm control was obtained through both catheter ablation and AAD (hazard ratio [HR] = 0.79; 96% CI, 0.66-0.94; P = .005), success was attributed to the use of catheter ablation for a rhythm-control strategy and its use among patients whose AF was present for < 1 year. Most patients in both treatment groups continued to receive anticoagulation, rate control, and optimization of cardiovascular risk.⁷

Continue to: Notably, direct studies...

Notably, direct studies comparing ablation and AAD have not confirmed the benefit of ablation over AAD in outcomes of all-cause mortality, bleeding, stroke, or cardiac arrest over a 5-year period.⁸

Adverse effects and mortality outcomes with AAD. Concern over using AAD for rhythm control is based mostly on adverse effects and long-term (1-year) mortality outcomes. Long-term AAD therapy has been shown to decrease the recurrence of AF—but without evidence to suggest other mortality benefits.

A meta-analysis of 59 randomized controlled trials reviewed 20,981 patients receiving AAD (including quinidine, disopyramide, propafenone, flecainide, metoprolol, amiodarone, dofetilide, dronedarone, and sotalol) for long-term effects on death, stroke, adverse reactions, and recurrence of AF.⁹ Findings at 10 months suggest that:

• Compared to placebo, amiodarone and sotalol increased the risk of all-cause mortality during the study period.
• There was minimal difference in mortality among patients taking dofetilide or dronedarone, compared to placebo.
• There were insufficient data to draw conclusions about the effect of disopyramide, flecainide, and propafenone on mortality.

Before starting a patient on AAD, the risk of arrhythmias and the potential for these agents to cause toxicity and adverse events should always be discussed.

CASE

You tell Mr. Z that you need to know the status of his comorbidities to make a recommendation about “other” management options, and proceed to take a detailed history.

Recent history. Mr. Z reveals that “today is a good day”: He has had “only 1” episode of palpitations, which resolved on its own. The previous episode, he explains, was 3 days ago, when palpitations were associated with lightheadedness and shortness of breath. He denies chest pains or swelling of the legs.

Physical exam. The patient appears spry, comfortable, and in no acute distress. Vital signs are within normal limits. A body mass index of 28.4 puts him in the “overweight” category. His blood pressure is 118/75 mm Hg.

Continue to: Cardiac examination...

Cardiac examination is significant for an irregular rhythm without murmurs, rubs, or gallops. His lungs are clear bilaterally; his abdomen is soft and nondistended. His extremities show no edema.

Testing. You obtain an electrocardiogram, which demonstrates a controlled ventricular rate of 88 bpm and AF. You order a complete blood count, comprehensive metabolic panel, tests of hemoglobin A1C and thyroid-stimulating hormone, lipid panel, echocardiogram, and a chest radiograph.

Results. The chest radiograph is negative for an acute cardiopulmonary process; cardiac size is normal. Aspartate aminotransferase and alanine aminotransferase levels are higher than twice the normal limit. The echocardiogram reveals an estimated left ventricular ejection fraction of 55% to 60%; no structural abnormalities are noted.

In which AF patients is catheter ablation indicated?

Ablation is recommended for select patients (TABLE 1^3,6) with symptomatic paroxysmal AF that is refractory to AAD or who are intolerant of AAD.^3,6 It is a reasonable first-line therapy for high-performing athletes in whom AAD would affect athletic performance.^3,10 It is also a reasonable option in select patients > 75 years and as an alternative to AAD therapy.³ Finally, catheter ablation should be considered in symptomatic patients with longstanding persistent AF and congestive heart failure, with or without reduced left ventricular ejection fraction.³

CASE

You inform Mr. Z that his symptoms are likely a result of symptomatic paroxysmal AF, which was refractory to flecainide and amiodarone, and that his abnormal liver function test results preclude continued use of amiodarone. You propose Holter monitoring to correlate timing of symptoms with the arrhythmia, but he reports this has been done, and the correlation confirmed, by his previous physician.

You explain that, because the diagnosis of symptomatic paroxysmal AF refractory to AADs has been confirmed, he is categorized as a patient who might benefit from catheter ablation, based on:

• the type of AF (ie, paroxysmal AF is associated with better ablation outcomes)
• persistent symptoms that are refractory to AADs
• his intolerance of AAD
• the length of time since onset of symptoms.

Mr. Z agrees to consider your recommendation.

Continue to: What are the benefits of catheter ablation?

Ablation can be achieved through radiofrequency (RF) ablation, cryoablation, or newer, laser-based balloon ablation. Primary outcomes used to determine the success of any options for performing ablation include mortality, stroke, and hospitalization. Other endpoints include maintenance of sinus rhythm, freedom from AF, reduction in AF burden (estimated through patients’ report of symptoms, recurrence rate, need for a second ablation procedure, and serial long-term monitoring through an implantable cardiac monitoring device), quality of life, and prevention of AF progression.³

Patient and disease variables (TABLE 2^11-13). The success rate of catheter ablation, defined as freedom from either symptomatic or asymptomatic episodes of AF, is dependent on several factors,^3,14 including:

• type of AF (paroxysmal or persistent)
• duration and degree of symptoms
• age
• sex
• comorbidities, including heart failure and structural heart or lung disease.

Overall, in patients with paroxysmal AF, an estimated 75% are symptom free 1 year after ablation.¹⁵ Patients with persistent and longstanding persistent AF experience a lower success rate.

Ablation is recommended in patients with symptomatic paroxysmal atrial fibrillation that is refractory to antiarrhythmics or when they are intolerant of these drugs.

RF catheter ablation has demonstrated superiority to AAD in reducing the need for cardioversion (relative risk [RR] = 0.62; 95% CI, 0.47-0.82) and cardiac-related hospitalization (RR = 0.27; 95% CI, 0.10-0.72;) at 12 months in patients with nonparoxysmal AF (persistent or longstanding persistent).¹⁶

Effect on mortality. Among patients with heart failure with reduced ejection fraction, long-term studies of cardiovascular outcomes 5 years post ablation concluded that ablation is associated with a decrease in all-cause mortality (RR = 0.69; 95% CI, 0.54-0.88; P = .003) and a reduction in hospitalization (RR = 0.62; 95% CI, 0.47-0.82; P = .0006); younger (< 65 years) and male patients derive greater benefit.^6,17 Indications for ablation in patients with heart failure are similar to those in patients without heart failure; ablation can therefore be considered for select heart failure patients who remain symptomatic or for whom AAD has failed.³

Older patients. Ablation can be considered for patients > 75 years with symptomatic paroxysmal AF refractory to AAD or who are intolerant of AAD.³ A study that assessed the benefit of catheter ablation reviewed 587 older (> 75 years) patients with AF, of whom 324 were eligible for ablation. Endpoints were maintenance of sinus rhythm, stroke, death, and major bleeding. Return to normal sinus rhythm was an independent factor, associated with a decrease in the risk of mortality among all patient groups that underwent ablation (HR = 0.36; 95% CI, 0.2-0.63; P = .0005). Age > 75 years (HR = 1.09; 95% CI, 1.01-1.16; P > .02) and depressed ejection fraction < 40% (HR = 2.38; 95% CI, 1.28-4.4; P = .006) were determined to be unfavorable parameters for survival.¹⁸

Complications and risks

Complications of catheter ablation for AF, although infrequent, can be severe (TABLE 3³). Early mortality, defined as death during initial admission or 30-day readmission, occurs in approximately 0.5% of cases; half of deaths take place during readmission.¹¹

Continue to: Complications vary...

Complications vary, based on the type and site of ablation.^19,20 Cardiac tamponade or perforation, the most life-threatening complications, taken together occur in an estimated 1.9% of patients (odds ratio [OR] = 2.98; 95% CI, 1.36-6.56; P = .007).¹¹ Other in-hospital complications independently predictive of death include any cardiac complications (OR = 12.8; 95% CI, 6.86 to 23.8; P < .001) and neurologic complications (cerebrovascular accident and transient ischemic attack) (OR = 8.72; 95% CI, 2.71-28.1; P < .001).

Other complications that do not cause death but might prolong the hospital stay include pericarditis without effusion, anesthesia-related complications, and vascular-access complications. Patients whose ablation is performed at an institution where the volume of ablations is low are also at higher risk of early mortality (OR = 2.35; 95% CI, 1.33-4.15; P = .003).¹⁶

Recurrence is common (TABLE 2^11-13). Risk of recurrence following ablation is significant; early (within 3 months after ablation) recurrence is seen in 50% of patients.^21,22 However, this is a so-called "blanking period"—ie, a temporary period of inflammatory and proarrhythmic changes that are not predictors of later recurrence. The 5-year post-ablation recurrence rate is approximately 25.5%; longstanding persistent and persistent AF and the presence of comorbidities are major risk factors for recurrence.^13,23

Recurrence is also associated with the type of procedure; pulmonary vein isolation, alone or in combination with another type of procedure, results in higher long-term success.^21,23

Other variables affect outcome (TABLE 2^11-13). Following AF ablation, patients with nonparoxysmal AF at baseline, advanced age, sleep apnea and obesity, left atrial enlargement, and any structural heart disease tend to have a poorer long-term (5-year) outcome (ie, freedom from extended episodes of AF).^3,13,23,24

Patients who undergo repeat procedures have higher arrhythmia-free survival; the highest ablation success rate is for patients with paroxysmal AF.^13,23

Exposure to ionizing radiation. Fluoroscopy is required for multiple components of atrial mapping and ablation during RF ablation, including navigation, visualization, and monitoring of catheter placement. Patients undergoing this particular procedure therefore receive significant exposure to ionizing radiation. A reduction in, even complete elimination of, fluoroscopy has been achieved with:

• nonfluoroscopic 3-dimensional mapping systems²⁵
• intracardiac echocardiography, which utilizes ultrasonographic imaging as the primary visual mode for tracking and manipulating the catheter
• robotic guided navigation.^26-28

Continue to: CASE

CASE

At his return visit, Mr. Z says that he is concerned about, first, undergoing catheter ablation at his age and, second, the risks associated with the procedure. You explain that it is true that ablation is ideal in younger patients who have minimal comorbidities and that the risk of complications increases with age—but that there is no cutoff or absolute age contraindication to ablation.

You tell Mr. Z that you will work with him on risk-factor modification in anticipation of ablation. You also assure him that the decision whether to ablate must be a joint one—between him and a cardiologist experienced both in electrophysiology and in performing this highly technical procedure. And you explain that a highly practiced specialist can identify Mr. Z’s risk factors that might make ablation more difficult to perform and affect the long-term outcome.

With Mr. Z’s agreement, you screen for sleep apnea and start him on a lifestyle modification plan to achieve a more ideal weight, explaining that the risk of recurrence of AF after catheter ablation is increased by obesity and sleep apnea, in addition to age. You explain that, based on his CHA₂DS₂–VASc (congestive heart failure; hypertension; age, ≥ 75 years; diabetes; prior stroke, transient ischemic attack, or thromboembolism; vascular disease; age, 65 to 74 years; sex category) score of 3, he will remain on anticoagulation whether or not he has the ablation.

You refer the patient to the nearest high-volume cardiac ablation center.

Last, you caution Mr. Z that, based on his lipid levels, his 10-year risk of heart disease or stroke is elevated. You recommend treatment with a statin agent while he continues his other medications.

Delivering energy to myocardium

Myocardial tissue in pulmonary veins is responsible for most ectopic electrical impulses in patients with drug-refractory AF. The goal of catheter ablation in AF is destruction (scarring) of tissue that is the source of abnormal vein potentials.¹⁵

How RF ablation works. Ablation is most commonly performed using RF energy, a high-frequency form of electrical energy. Electrophysiology studies are carried out at the time of ablation by percutaneous, fluoroscopically guided insertion of 2 to 5 catheters, usually through the femoral or internal jugular vein, which are then positioned within several areas of the heart—usually, the right atrium, bundle of His, right ventricle, and coronary sinus.

Continue to: Electrical current...

Successful ablation is defined as freedom from symptomatic or asymptomatic AF episodes, and depends on AF type; duration and degree of symptoms; and age, sex, and comorbidities.

Electrical current is applied through the catheters from an external generator to stimulate the myocardium and thus determine its electrophysiologic properties. The anatomic and electrical activity of the left atrium and pulmonary veins is then identified, a technique known as electro-anatomical mapping (FIGURE). After arrhythmogenic myocardial tissue is mapped, ablation is carried out with RF energy through the catheter to the pathogenic myocardium from which arrhythmias are initiated or conducted. The result is thermal destruction of tissue and creation of small, shallow lesions that vary in size with the type of catheter and the force of contact pressure applied.^3,29

Other energy sources used in catheter ablation include cryothermal energy, which utilizes liquid nitrous oxide under pressure through a cryocatheter or cryoballoon catheter. Application of cryothermal energy freezes tissue and disrupts cell membranes and any electrical activity. Cryoballoon ablation has been shown to be similarly safe and efficacious as RF ablation in patients with paroxysmal AF.^30,31

Newer laser-based balloon ablations are performed under ultrasonographic guidance and utilize arcs of laser energy delivered to the pathogenic myocardium.³

Thromboembolism prophylaxis

Oral anticoagulation to decrease the risk of stroke is initiated in all patients with AF, based on a thromboembolic risk profile determined by their CHA₂DS₂–VASc score, with anticoagulation recommended when the score is ≥ 2 in men and ≥ 3 in women. Options for anticoagulation include warfarin and one of the novel oral anticoagulants dabigatran, apixaban, rivaroxaban, and edoxaban.⁴ Recommendations are as follows³:

• For patients with a CHA₂DS₂–VASc score of ≥ 2 (men) or ≥ 3 (women), anticoagulation should be continued indefinitely, regardless of how successful the ablation procedure is.
• When patients choose to discontinue anticoagulation, they should be counseled in detail about the risk of doing so. The continued need for frequent arrhythmia monitoring should be emphasized.

The route from primary careto catheter ablation

Perform a thorough evaluation. Patients who present to you with palpitations should first undergo a routine workup for AF, followed by confirmation of the diagnosis. Exclude structural heart disease with echocardiography. Undertake monitoring, which is essential to determine whether symptoms are a reflection of the arrhythmia, using noncontinuous or continuous electrocardiographic (EKG) monitoring. Noncontinuous detection devices include:

• scheduled or symptom-initiated EKG
• a Holter monitor, worn for at least 24 hours and as long as 7 days
• trans-telephonic recordings and patient- or automatically activated devices
• an external loop recorder.³²

Continuous EKG monitoring is more permanent (≥ 12 months). This is usually achieved through an implantable loop powered by a battery that lasts as long as 3 years.³

Ablation: Yes or no? Ablation is not recommended to avoid anticoagulation or when anticoagulation is contraindicated.⁵ With regard to specific patient criteria, the ideal patient:

• is symptomatic
• has failed AAD therapy
• does not have pulmonary disease
• has a normal or mildly dilated left atrium or normal or mildly reduced left ventricular ejection fraction.⁵

Continue to: There is no absolute age...

There is no absolute age or comorbidity contraindication to ablation. The patient should be referred to a cardiologist who has received appropriate training in electrophysiology, to identify comorbidities that (1) increase the technical difficulty of the procedure and baseline risk and (2) affect long-term outcome,¹² and who performs the procedure in a center that has considerable experience with catheter ablation.³³

Once the decision is made to perform ablation, you can provide strategies that optimize the outcome (freedom from AF episodes). Those tactics include weight loss and screening evaluation and, if indicated, treatment for sleep apnea.³

Protocol. Prior to the procedure, the patient fasts overnight; they might be asked to taper or discontinue cardiac medications that have electrophysiologic effects. Studies suggest a low risk of bleeding associated with catheter ablation; anticoagulation should therefore continue uninterrupted for patients undergoing catheter ablation for AF^3,4,34,35; however, this practice varies with the cardiologist or electrophysiologist performing ablation.

Patients whose ablation is performed at an institution where the volume of such procedures is low are at higher risk of early mortality.

Because of the length and complexity of the procedure, electro-anatomical mapping and ablation are conducted with the patient under general anesthesia.³ The patient is kept supine, and remains so for 2 to 4 hours afterward to allow for hemostasis at puncture sites.³

Patients might be monitored overnight, although same-day catheter ablation has been shown to be safe and cost-effective in select patients.^36,37 Post ablation, patients follow up with the cardiologist and electrophysiologist. Long-term arrhythmia monitoring is required.³ Anticoagulation is continued for at least 2 months, and is discontinued based on the patient’s risk for stroke, utilizing their CHA₂DS₂–VASc score.^3,4

CASE

At Mr. Z’s 6-month primary care follow-up, he confirms what has been reported to you as the referring physician: He had a successful catheter ablation and continues to have regular follow-up monitoring with the cardiologist. He is no longer taking amiodarone.

At this visit, he reports no recurrence of AF-associated symptoms or detectable AF on cardiac monitoring. He has lost 8 lbs. You counsel to him to continue to maintain a healthy lifestyle.

CORRESPONDENCE
Amimi S. Osayande MD, FAAFP, Northside-Gwinnett Family Medicine Residency Program, Strickland Family Medicine Center, 665 Duluth Highway, Suite 501, Lawrenceville, GA 30046; [email protected]

CASE

Jack Z, a 75-year-old man with well-controlled hypertension, diabetes controlled by diet, and atrial fibrillation (AF) presents to the family medicine clinic to establish care with you after moving to the community from out of town.

The patient describes a 1-year history of AF. He provides you with an echocardiography report from 6 months ago that shows no evidence of structural heart disease. He takes lisinopril, to control blood pressure; an anticoagulant; a beta-blocker; and amiodarone for rhythm control. Initially, he took flecainide, which was ineffective for rhythm control, before being switched to amiodarone. He had 2 cardioversion procedures, each time after episodes of symptoms. He does not smoke or drink alcohol.

Mr. Z describes worsening palpitations and shortness of breath over the past 9 months. Symptoms now include episodes of exertional fatigue, even when he is not having palpitations. Prior to the episodes of worsening symptoms, he tells you that he lived a “fairly active” life, golfing twice a week.

The patient’s previous primary care physician had encouraged him to talk to his cardiologist about “other options” for managing AF, because levels of his liver enzymes had started to rise (a known adverse effect of amiodarone¹) when measured 3 months ago. He did not undertake that conversation, but asks you now about other treatments for AF.

Atrial fibrillation is the most common sustained cardiac arrhythmia, characterized by discordant electrical activation of the atria due to structural or electrophysiological abnormalities, or both. The disorder is associated with an increased rate of stroke and heart failure and is independently associated with a 1.5- to 2-fold risk of all-cause mortality.²

In this article, we review the pathophysiology of AF; management, including the role of, and indications for, catheter ablation; and patient- and disease-related factors associated with ablation (including odds of success, complications, risk of recurrence, and continuing need for thromboprophylaxis) that family physicians should consider when contemplating referral to a cardiologist or electrophysiologist for catheter ablation for AF.

What provokes AF?

AF is thought to occur as a result of an interaction among 3 phenomena:

• enhanced automaticity of abnormal atrial tissue
• triggered activity of ectopic foci within 1 or more pulmonary veins, lying within the left atrium
• re-entry, in which there is propagation of electrical impulses from an ectopic beat through another pathway.

Continue to: In patients who progress...

In patients who progress from paroxysmal to persistent AF (see “Subtypes,” below), 2 distinct pathways, facilitated by the presence of abnormal tissue, continuously activate one another, thus maintaining the arrhythmia. Myocardial tissue in the pulmonary veins is responsible for most ectopic electrical impulses in patients with drug-refractory AF (see “Rhythm control”).

Subtypes. For the purpose of planning treatment, AF is classified as:

• Paroxysmal. Terminates spontaneously or with intervention ≤ 7 days after onset.
• Persistent. Continuous and sustained for > 7 days.
• Longstanding persistent. Continuous for > 12 months.
• Permanent. The patient and physician accept that there will be no further attempt to restore or maintain sinus rhythm.

Goals of treatment

Primary management goals in patients with AF are 2-fold: control of symptoms and prevention of thromboembolism. A patient with new-onset AF who presents acutely with inadequate rate control and hemodynamic compromise requires urgent assessment to determine the cause of the arrhythmia and need for cardioversion.³ A symptomatic patient with AF who does not have high-risk features (eg, valvular heart disease, mechanical valves) might be a candidate for rhythm control in addition to rate control.^3,4

Rate control. After evaluation in the hospital, a patient who has a rapid ventricular response but remains hemodynamically stable, without evidence of heart failure, should be initiated on a rate-controlling medication, such as a beta-blocker or nondihydropyridine calcium-channel blocker. A resting heart rate goal of < 80 beats per minute (bpm) is recommended for a symptomatic patient with AF. The heart rate goal can be relaxed, to < 110 bpm, in an asymptomatic patient with preserved left ventricular function.^5,6

Rhythm control, indicated in patients who remain symptomatic on rate-controlling medication, can be achieved either with an antiarrhythmic drug (AAD) or by catheter ablation.^4,5 In stable patients, rhythm control should be considered only after a thorough work-up for a reversible cause of AF, and can be achieved with an oral AAD or, in select patients, through catheter ablation (TABLE 1^3,6). Other indications for chronic rhythm control include treatment of patients with ­tachycardia-induced cardiomyopathy.⁵

Continuing symptoms when taking rate-controlling medication for atrial fibrillation is one indication for rhythm control with an antiarrhythmic drug or by catheter ablation.

A major study that documented the benefit of early rhythm control evaluated long-term outcomes in 2789 patients with AF who were undergoing catheter ablation.⁷ Patients were randomized to early rhythm control (catheter ablation or AAD) or “usual care”—ie, in this study, rhythm control limited to symptomatic patients. Primary outcomes were death from cardiovascular causes, stroke, and hospitalization with worsening heart failure or acute coronary syndrome. A first primary outcome event occurred in 249 patients (3.9/100 person-years) assigned to early rhythm control, compared to 316 (5.0 per 100 person-years) in the group assigned to usual care.

The study was terminated early (after 5.1 years) because of overwhelming evidence of efficacy (number need to treat = 7). Although early rhythm control was obtained through both catheter ablation and AAD (hazard ratio [HR] = 0.79; 96% CI, 0.66-0.94; P = .005), success was attributed to the use of catheter ablation for a rhythm-control strategy and its use among patients whose AF was present for < 1 year. Most patients in both treatment groups continued to receive anticoagulation, rate control, and optimization of cardiovascular risk.⁷

Continue to: Notably, direct studies...

Notably, direct studies comparing ablation and AAD have not confirmed the benefit of ablation over AAD in outcomes of all-cause mortality, bleeding, stroke, or cardiac arrest over a 5-year period.⁸

Adverse effects and mortality outcomes with AAD. Concern over using AAD for rhythm control is based mostly on adverse effects and long-term (1-year) mortality outcomes. Long-term AAD therapy has been shown to decrease the recurrence of AF—but without evidence to suggest other mortality benefits.

A meta-analysis of 59 randomized controlled trials reviewed 20,981 patients receiving AAD (including quinidine, disopyramide, propafenone, flecainide, metoprolol, amiodarone, dofetilide, dronedarone, and sotalol) for long-term effects on death, stroke, adverse reactions, and recurrence of AF.⁹ Findings at 10 months suggest that:

• Compared to placebo, amiodarone and sotalol increased the risk of all-cause mortality during the study period.
• There was minimal difference in mortality among patients taking dofetilide or dronedarone, compared to placebo.
• There were insufficient data to draw conclusions about the effect of disopyramide, flecainide, and propafenone on mortality.

Before starting a patient on AAD, the risk of arrhythmias and the potential for these agents to cause toxicity and adverse events should always be discussed.

CASE

You tell Mr. Z that you need to know the status of his comorbidities to make a recommendation about “other” management options, and proceed to take a detailed history.

Recent history. Mr. Z reveals that “today is a good day”: He has had “only 1” episode of palpitations, which resolved on its own. The previous episode, he explains, was 3 days ago, when palpitations were associated with lightheadedness and shortness of breath. He denies chest pains or swelling of the legs.

Physical exam. The patient appears spry, comfortable, and in no acute distress. Vital signs are within normal limits. A body mass index of 28.4 puts him in the “overweight” category. His blood pressure is 118/75 mm Hg.

Continue to: Cardiac examination...

Cardiac examination is significant for an irregular rhythm without murmurs, rubs, or gallops. His lungs are clear bilaterally; his abdomen is soft and nondistended. His extremities show no edema.

Testing. You obtain an electrocardiogram, which demonstrates a controlled ventricular rate of 88 bpm and AF. You order a complete blood count, comprehensive metabolic panel, tests of hemoglobin A1C and thyroid-stimulating hormone, lipid panel, echocardiogram, and a chest radiograph.

Results. The chest radiograph is negative for an acute cardiopulmonary process; cardiac size is normal. Aspartate aminotransferase and alanine aminotransferase levels are higher than twice the normal limit. The echocardiogram reveals an estimated left ventricular ejection fraction of 55% to 60%; no structural abnormalities are noted.

In which AF patients is catheter ablation indicated?

Ablation is recommended for select patients (TABLE 1^3,6) with symptomatic paroxysmal AF that is refractory to AAD or who are intolerant of AAD.^3,6 It is a reasonable first-line therapy for high-performing athletes in whom AAD would affect athletic performance.^3,10 It is also a reasonable option in select patients > 75 years and as an alternative to AAD therapy.³ Finally, catheter ablation should be considered in symptomatic patients with longstanding persistent AF and congestive heart failure, with or without reduced left ventricular ejection fraction.³

CASE

You inform Mr. Z that his symptoms are likely a result of symptomatic paroxysmal AF, which was refractory to flecainide and amiodarone, and that his abnormal liver function test results preclude continued use of amiodarone. You propose Holter monitoring to correlate timing of symptoms with the arrhythmia, but he reports this has been done, and the correlation confirmed, by his previous physician.

You explain that, because the diagnosis of symptomatic paroxysmal AF refractory to AADs has been confirmed, he is categorized as a patient who might benefit from catheter ablation, based on:

• the type of AF (ie, paroxysmal AF is associated with better ablation outcomes)
• persistent symptoms that are refractory to AADs
• his intolerance of AAD
• the length of time since onset of symptoms.

Mr. Z agrees to consider your recommendation.

Continue to: What are the benefits of catheter ablation?

Ablation can be achieved through radiofrequency (RF) ablation, cryoablation, or newer, laser-based balloon ablation. Primary outcomes used to determine the success of any options for performing ablation include mortality, stroke, and hospitalization. Other endpoints include maintenance of sinus rhythm, freedom from AF, reduction in AF burden (estimated through patients’ report of symptoms, recurrence rate, need for a second ablation procedure, and serial long-term monitoring through an implantable cardiac monitoring device), quality of life, and prevention of AF progression.³

Patient and disease variables (TABLE 2^11-13). The success rate of catheter ablation, defined as freedom from either symptomatic or asymptomatic episodes of AF, is dependent on several factors,^3,14 including:

• type of AF (paroxysmal or persistent)
• duration and degree of symptoms
• age
• sex
• comorbidities, including heart failure and structural heart or lung disease.

Overall, in patients with paroxysmal AF, an estimated 75% are symptom free 1 year after ablation.¹⁵ Patients with persistent and longstanding persistent AF experience a lower success rate.

Ablation is recommended in patients with symptomatic paroxysmal atrial fibrillation that is refractory to antiarrhythmics or when they are intolerant of these drugs.

RF catheter ablation has demonstrated superiority to AAD in reducing the need for cardioversion (relative risk [RR] = 0.62; 95% CI, 0.47-0.82) and cardiac-related hospitalization (RR = 0.27; 95% CI, 0.10-0.72;) at 12 months in patients with nonparoxysmal AF (persistent or longstanding persistent).¹⁶

Effect on mortality. Among patients with heart failure with reduced ejection fraction, long-term studies of cardiovascular outcomes 5 years post ablation concluded that ablation is associated with a decrease in all-cause mortality (RR = 0.69; 95% CI, 0.54-0.88; P = .003) and a reduction in hospitalization (RR = 0.62; 95% CI, 0.47-0.82; P = .0006); younger (< 65 years) and male patients derive greater benefit.^6,17 Indications for ablation in patients with heart failure are similar to those in patients without heart failure; ablation can therefore be considered for select heart failure patients who remain symptomatic or for whom AAD has failed.³

Older patients. Ablation can be considered for patients > 75 years with symptomatic paroxysmal AF refractory to AAD or who are intolerant of AAD.³ A study that assessed the benefit of catheter ablation reviewed 587 older (> 75 years) patients with AF, of whom 324 were eligible for ablation. Endpoints were maintenance of sinus rhythm, stroke, death, and major bleeding. Return to normal sinus rhythm was an independent factor, associated with a decrease in the risk of mortality among all patient groups that underwent ablation (HR = 0.36; 95% CI, 0.2-0.63; P = .0005). Age > 75 years (HR = 1.09; 95% CI, 1.01-1.16; P > .02) and depressed ejection fraction < 40% (HR = 2.38; 95% CI, 1.28-4.4; P = .006) were determined to be unfavorable parameters for survival.¹⁸

Complications and risks

Complications of catheter ablation for AF, although infrequent, can be severe (TABLE 3³). Early mortality, defined as death during initial admission or 30-day readmission, occurs in approximately 0.5% of cases; half of deaths take place during readmission.¹¹

Continue to: Complications vary...

Complications vary, based on the type and site of ablation.^19,20 Cardiac tamponade or perforation, the most life-threatening complications, taken together occur in an estimated 1.9% of patients (odds ratio [OR] = 2.98; 95% CI, 1.36-6.56; P = .007).¹¹ Other in-hospital complications independently predictive of death include any cardiac complications (OR = 12.8; 95% CI, 6.86 to 23.8; P < .001) and neurologic complications (cerebrovascular accident and transient ischemic attack) (OR = 8.72; 95% CI, 2.71-28.1; P < .001).

Other complications that do not cause death but might prolong the hospital stay include pericarditis without effusion, anesthesia-related complications, and vascular-access complications. Patients whose ablation is performed at an institution where the volume of ablations is low are also at higher risk of early mortality (OR = 2.35; 95% CI, 1.33-4.15; P = .003).¹⁶

Recurrence is common (TABLE 2^11-13). Risk of recurrence following ablation is significant; early (within 3 months after ablation) recurrence is seen in 50% of patients.^21,22 However, this is a so-called "blanking period"—ie, a temporary period of inflammatory and proarrhythmic changes that are not predictors of later recurrence. The 5-year post-ablation recurrence rate is approximately 25.5%; longstanding persistent and persistent AF and the presence of comorbidities are major risk factors for recurrence.^13,23

Recurrence is also associated with the type of procedure; pulmonary vein isolation, alone or in combination with another type of procedure, results in higher long-term success.^21,23

Other variables affect outcome (TABLE 2^11-13). Following AF ablation, patients with nonparoxysmal AF at baseline, advanced age, sleep apnea and obesity, left atrial enlargement, and any structural heart disease tend to have a poorer long-term (5-year) outcome (ie, freedom from extended episodes of AF).^3,13,23,24

Patients who undergo repeat procedures have higher arrhythmia-free survival; the highest ablation success rate is for patients with paroxysmal AF.^13,23

Exposure to ionizing radiation. Fluoroscopy is required for multiple components of atrial mapping and ablation during RF ablation, including navigation, visualization, and monitoring of catheter placement. Patients undergoing this particular procedure therefore receive significant exposure to ionizing radiation. A reduction in, even complete elimination of, fluoroscopy has been achieved with:

• nonfluoroscopic 3-dimensional mapping systems²⁵
• intracardiac echocardiography, which utilizes ultrasonographic imaging as the primary visual mode for tracking and manipulating the catheter
• robotic guided navigation.^26-28

Continue to: CASE

CASE

At his return visit, Mr. Z says that he is concerned about, first, undergoing catheter ablation at his age and, second, the risks associated with the procedure. You explain that it is true that ablation is ideal in younger patients who have minimal comorbidities and that the risk of complications increases with age—but that there is no cutoff or absolute age contraindication to ablation.

You tell Mr. Z that you will work with him on risk-factor modification in anticipation of ablation. You also assure him that the decision whether to ablate must be a joint one—between him and a cardiologist experienced both in electrophysiology and in performing this highly technical procedure. And you explain that a highly practiced specialist can identify Mr. Z’s risk factors that might make ablation more difficult to perform and affect the long-term outcome.

With Mr. Z’s agreement, you screen for sleep apnea and start him on a lifestyle modification plan to achieve a more ideal weight, explaining that the risk of recurrence of AF after catheter ablation is increased by obesity and sleep apnea, in addition to age. You explain that, based on his CHA₂DS₂–VASc (congestive heart failure; hypertension; age, ≥ 75 years; diabetes; prior stroke, transient ischemic attack, or thromboembolism; vascular disease; age, 65 to 74 years; sex category) score of 3, he will remain on anticoagulation whether or not he has the ablation.

You refer the patient to the nearest high-volume cardiac ablation center.

Last, you caution Mr. Z that, based on his lipid levels, his 10-year risk of heart disease or stroke is elevated. You recommend treatment with a statin agent while he continues his other medications.

Delivering energy to myocardium

Myocardial tissue in pulmonary veins is responsible for most ectopic electrical impulses in patients with drug-refractory AF. The goal of catheter ablation in AF is destruction (scarring) of tissue that is the source of abnormal vein potentials.¹⁵

How RF ablation works. Ablation is most commonly performed using RF energy, a high-frequency form of electrical energy. Electrophysiology studies are carried out at the time of ablation by percutaneous, fluoroscopically guided insertion of 2 to 5 catheters, usually through the femoral or internal jugular vein, which are then positioned within several areas of the heart—usually, the right atrium, bundle of His, right ventricle, and coronary sinus.

Continue to: Electrical current...

Successful ablation is defined as freedom from symptomatic or asymptomatic AF episodes, and depends on AF type; duration and degree of symptoms; and age, sex, and comorbidities.

Electrical current is applied through the catheters from an external generator to stimulate the myocardium and thus determine its electrophysiologic properties. The anatomic and electrical activity of the left atrium and pulmonary veins is then identified, a technique known as electro-anatomical mapping (FIGURE). After arrhythmogenic myocardial tissue is mapped, ablation is carried out with RF energy through the catheter to the pathogenic myocardium from which arrhythmias are initiated or conducted. The result is thermal destruction of tissue and creation of small, shallow lesions that vary in size with the type of catheter and the force of contact pressure applied.^3,29

Other energy sources used in catheter ablation include cryothermal energy, which utilizes liquid nitrous oxide under pressure through a cryocatheter or cryoballoon catheter. Application of cryothermal energy freezes tissue and disrupts cell membranes and any electrical activity. Cryoballoon ablation has been shown to be similarly safe and efficacious as RF ablation in patients with paroxysmal AF.^30,31

Newer laser-based balloon ablations are performed under ultrasonographic guidance and utilize arcs of laser energy delivered to the pathogenic myocardium.³

Thromboembolism prophylaxis

Oral anticoagulation to decrease the risk of stroke is initiated in all patients with AF, based on a thromboembolic risk profile determined by their CHA₂DS₂–VASc score, with anticoagulation recommended when the score is ≥ 2 in men and ≥ 3 in women. Options for anticoagulation include warfarin and one of the novel oral anticoagulants dabigatran, apixaban, rivaroxaban, and edoxaban.⁴ Recommendations are as follows³:

• For patients with a CHA₂DS₂–VASc score of ≥ 2 (men) or ≥ 3 (women), anticoagulation should be continued indefinitely, regardless of how successful the ablation procedure is.
• When patients choose to discontinue anticoagulation, they should be counseled in detail about the risk of doing so. The continued need for frequent arrhythmia monitoring should be emphasized.

The route from primary careto catheter ablation

Perform a thorough evaluation. Patients who present to you with palpitations should first undergo a routine workup for AF, followed by confirmation of the diagnosis. Exclude structural heart disease with echocardiography. Undertake monitoring, which is essential to determine whether symptoms are a reflection of the arrhythmia, using noncontinuous or continuous electrocardiographic (EKG) monitoring. Noncontinuous detection devices include:

• scheduled or symptom-initiated EKG
• a Holter monitor, worn for at least 24 hours and as long as 7 days
• trans-telephonic recordings and patient- or automatically activated devices
• an external loop recorder.³²

Continuous EKG monitoring is more permanent (≥ 12 months). This is usually achieved through an implantable loop powered by a battery that lasts as long as 3 years.³

Ablation: Yes or no? Ablation is not recommended to avoid anticoagulation or when anticoagulation is contraindicated.⁵ With regard to specific patient criteria, the ideal patient:

• is symptomatic
• has failed AAD therapy
• does not have pulmonary disease
• has a normal or mildly dilated left atrium or normal or mildly reduced left ventricular ejection fraction.⁵

Continue to: There is no absolute age...

There is no absolute age or comorbidity contraindication to ablation. The patient should be referred to a cardiologist who has received appropriate training in electrophysiology, to identify comorbidities that (1) increase the technical difficulty of the procedure and baseline risk and (2) affect long-term outcome,¹² and who performs the procedure in a center that has considerable experience with catheter ablation.³³

Once the decision is made to perform ablation, you can provide strategies that optimize the outcome (freedom from AF episodes). Those tactics include weight loss and screening evaluation and, if indicated, treatment for sleep apnea.³

Protocol. Prior to the procedure, the patient fasts overnight; they might be asked to taper or discontinue cardiac medications that have electrophysiologic effects. Studies suggest a low risk of bleeding associated with catheter ablation; anticoagulation should therefore continue uninterrupted for patients undergoing catheter ablation for AF^3,4,34,35; however, this practice varies with the cardiologist or electrophysiologist performing ablation.

Patients whose ablation is performed at an institution where the volume of such procedures is low are at higher risk of early mortality.

Because of the length and complexity of the procedure, electro-anatomical mapping and ablation are conducted with the patient under general anesthesia.³ The patient is kept supine, and remains so for 2 to 4 hours afterward to allow for hemostasis at puncture sites.³

Patients might be monitored overnight, although same-day catheter ablation has been shown to be safe and cost-effective in select patients.^36,37 Post ablation, patients follow up with the cardiologist and electrophysiologist. Long-term arrhythmia monitoring is required.³ Anticoagulation is continued for at least 2 months, and is discontinued based on the patient’s risk for stroke, utilizing their CHA₂DS₂–VASc score.^3,4

CASE

At Mr. Z’s 6-month primary care follow-up, he confirms what has been reported to you as the referring physician: He had a successful catheter ablation and continues to have regular follow-up monitoring with the cardiologist. He is no longer taking amiodarone.

At this visit, he reports no recurrence of AF-associated symptoms or detectable AF on cardiac monitoring. He has lost 8 lbs. You counsel to him to continue to maintain a healthy lifestyle.

CORRESPONDENCE
Amimi S. Osayande MD, FAAFP, Northside-Gwinnett Family Medicine Residency Program, Strickland Family Medicine Center, 665 Duluth Highway, Suite 501, Lawrenceville, GA 30046; [email protected]

As the geriatric population continues to grow and treatment advances blur the lines between improving the length of life vs improving its quality, end-of-life (EOL) conversations are becoming increasingly important. These discussions are a crucial part of the advance care planning (ACP) process, in which patients discuss their treatment preferences and values with their caregiver/surrogate decision maker and health care provider to ultimately improve EOL decision-making and care. ^1,2

EOL conversations are most helpful when incorporated in the outpatient setting as part of the patient’s ongoing health care plan or when initiating treatment for a chronic or life-threatening disease. Because family physicians promote general wellness, understand the patient’s health status and medical history, and have an ongoing—and often longstanding—relationship with patients and their families, we are ideally positioned to engage patients in EOL discussions. However, these conversations can be challenging in the outpatient setting, and often clinicians struggle not only to find ways to raise the subject, but also to find the time to have these supportive, meaningful conversations.³

In this article, we will address the importance of having EOL discussions in the outpatient setting, specifically about advance directives (ADs), and the reasons why patients and physicians might avoid these discussions. The role of palliative care in EOL care, along with its benefits and methods for overcoming patient and physician barriers to its successful use, are reviewed. Finally, we examine specific challenges associated with discussing EOL care with patients with decreased mental capacity, such as those with dementia, and provide strategies to successfully facilitate EOL discussions in these populations. 

Moving patients toward completion of advance directives

Although many older patients express a desire to document their wishes before EOL situations arise, they may not fully understand the benefits of an AD or how to complete one. ⁴ Often the family physician is best equipped to address the patient’s concerns and discuss their goals for EOL care, as well as the potential situations that might arise.

Managing an aging population. Projections suggest that primary care physicians will encounter increasing numbers of geriatric patients in the next 2 decades. Thus it is essential for those in primary care to receive proper training during their residency for the care of this group of patients. According to a group of academic educators and geriatricians from internal medicine and family medicine whose goal was to define a set of minimal and essential competencies in the care of older adults, this includes training on how to discuss and document “advance care planning and goals of care with all patients with chronic or complex illness,” as well as how to differentiate among “types of code status, health care proxies, and advanced directives” within the state in which training occurs. ⁵

Educate patients and ease fears. Patients often avoid EOL conversations or wait for their family physician to start the conversation. They may not understand how ADs can help guide care or they may believe they are “too healthy” to have these conversations at this time. ⁴ Simply asking about existing ADs or providing forms to patients during an outpatient visit can open the door to more in-depth discussions. Some examples of opening phrases include:

• Do you have a living will or durable power of attorney for health care?
• Have you ever discussed your health care wishes with your loved ones?
• Who would you want to speak for you regarding your health care if you could not speak for yourself? Have you discussed your health care wishes with that person?

By normalizing the conversation as a routine part of comprehensive, patient-centered care, the family physician can allay patient fears, foster open and honest conversations, and encourage ongoing discussions with loved ones as situations arise.⁶

Continue to: When ADs are executed...

As many as 90% of patients with a life-threatening illness report never having discussed EOL care issues with their clinician.

When ADs are executed, patients often fail to have meaningful conversations with their surrogates about specific treatment wishes or EOL scenarios. As a result, the surrogate may not feel prepared to serve as a proxy decision maker or may find the role extremely stressful.⁷ Physicians should encourage open conversations between patients and their surrogates about potential EOL scenarios when possible. When possible and appropriate, it is also important to encourage the patient to include the surrogate in future outpatient visits so that the surrogate can understand the patient’s health status and potential decisions they may need to make.

Don’t overlook clinician barriers. Family physicians also might avoid AD discussions because they do not understand laws that govern ADs, which vary from state to state. Various online resources for patients and physicians exist that clarify state-specific regulations and provide state-specific forms (TABLE).

Time constraints present another challenge for family physicians. This can be addressed by establishing workflows that include EOL elements. Also, the Centers for Medicare and Medicaid Services (CMS) has provided separate billing codes for AD discussion based on time spent explaining and discussing how to complete forms.⁸ CPT codes 99497 and 99498 are time-based codes that cover the first 30 minutes and each additional 30 minutes, respectively, of time spent explaining and discussing how to complete standard forms in a face-to-face setting (TABLE).⁹ CMS also includes discussion of AD documents as an optional element of the annual Medicare wellness visit.⁸ 

Improve quality of life for patients with any serious illness

Unlike hospice, which focuses on providing comfort rather than cure in the final months of a patient’s life, palliative care strives to prevent and relieve the patient’s suffering from a serious illness that is not immediately life-threatening. Palliative care focuses on the early identification, careful assessment, and treatment of the physical, psychosocial, and spiritual symptoms associated with a patient’s condition(s).^10,11 It has been well established that palliative care has a positive effect on many clinical outcomes including symptom burden, quality of life, satisfaction with care, and survival.^12-14 Patients who receive palliative care consultation also tend to perceive a higher quality of care.¹⁵

Conversations lead to better outcomes. Palliative care consultation is being increasingly used in the outpatient setting and can be introduced early in a disease process. Doing so provides an additional opportunity for the family physician to introduce an EOL discussion. A comparison of outcomes between patients who had initial inpatient palliative care consultation vs outpatient palliative care referral found that outpatient referral improved quality EOL care and was associated with significantly fewer emergency department visits (68% vs 48%; P < .001) and hospital admissions (86% vs 52%; P < .001), as well as shorter hospital stays in the last 30 days of life (3-11 vs 5-14 days; P = .01).¹⁴ Despite these benefits, 60% to 90% of patients with a serious illness report never having discussed EOL care issues with their clinician.^16,17

Continue to: Early EOL discussions...

Early EOL discussions have also been shown to have a positive impact on families. In a US study, family members stated that timely EOL care discussions allowed them to make use of hospice and palliative care services sooner and to make the most of their time with the patient.¹⁸

Timing and communication are key

Logistically it can be difficult to gather the right people (patient, family, etc) in the right place and at the right time. For physicians, the most often cited barriers include inadequate time to conduct an EOL discussion, ¹⁹ a perceived lack of competence in EOL conversations, ^1,20 difficulty navigating patient readiness, ²¹ and a fear of destroying hope due to prognostic uncertainty. ^19,20

A prospective, observational study used the Quality of Communication (QOC) questionnaire to assess life-sustaining treatment preferences, ACP, and the quality of EOL care communication in Dutch outpatients with clinically stable but severe chronic obstructive pulmonary disease (n = 105) or congestive heart failure (n = 80). The QOC questionnaire is a validated instrument that asks patients to rate their physician on several communication skills from 0 (“the very worst” or “My doctor didn’t do this”) to 10 (“the very best”). In this study, quality communication was identified by patients as one of the most important skills for physicians to provide adequate EOL care. ²² While QOC ratings were high for general communication skills (median, 8.0 points), quality EOL care communication was rated very low (median, 0.0 points). Researchers say that this was primarily because most EOL topics were not discussed—especially spirituality, prognosis, and what dying might be like. ²² In a secondary analysis that evaluated quality of EOL care communication during 1-year follow-up of patients with advanced chronic organ failure (n = 265) with the QOC questionnaire, patient ratings improved to moderate to good (medians, 6-8 points) when these topics were addressed. ²³  

Pick a strategy and prepare. As the older population continues to grow, the demands of palliative care management cannot be met by specialists alone and the responsibility of discussing EOL care with patients and their families will increasingly fall to family physicians as well. ²⁴ Several strategies and approaches have evolved to assist family physicians with acquiring the skills to conduct productive EOL discussions. These include widely referenced resources, such as VitalTalk ²⁵ and the ABCDE Plan. ²⁶ VitalTalk teaches skills to help clinicians navigate difficult conversations, ²⁵ and the “ABCDE” method provides a pneumonic for recommendations for how to deliver bad news ( A dvance preparation; B uild a therapeutic environment/relationship; C ommunicate well; D eal with patient and family reactions; E ncourage and validate emotions). ²⁶

Researchers found that timely EOL care discussions allowed family members to make use of hospice and palliative care services sooner and maximize their time with the patient.

Other strategies include familiarizing oneself with the patient’s medical history and present situation (eg, What are the patient’s symptoms? What do other involved clinicians think and recommend? What therapies have been attempted? What are the relevant social and emotional dynamics?); asking the patient who they want present for the EOL conversation; scheduling the conversation for when you can set aside an appropriate amount of time and in a private place where there will be no interruptions; and going into the meeting with your goal in mind, whether it is to deliver bad news, clarify the prognosis, establish goals of care, or communicate the patient’s goals and wishes for the EOL to those in attendance. ²⁷ It can be very helpful to begin the conversation by clarifying what the patient and their family/surrogate understand about the current diagnosis and prognosis. From there, the family physician can present a “headline” that prepares them for the current conversation (eg, “I have your latest test results, and I need to share some serious news”). This can facilitate a more detailed discussion of the patient’s and surrogate’s goals of care. Using these strategies, family physicians can lead a productive EOL discussion that benefits everyone.  

Continue to: How to navigate EOL discussions with patients with dementia

How to navigate EOL discussions with patients with dementia

EOL discussions with patients with dementia become even more complex and warrant specific discussion because one must consider the timing of such discussions, ^2,28,29 the trajectory of the disease and how that affects the patient’s capacity for EOL conversations, and the critical importance of engaging caregivers/surrogate decision makers in these discussions. ² ACP provides an opportunity for the physician, patient, and caregiver/­surrogate to jointly explore the patient’s values, beliefs, and preferences for care through the EOL as the disease progresses and the patient’s decisional capacity declines.

Ensure meaningful participation with timing. EOL discussions should occur while the patient has the cognitive capacity to actively participate in the planning process. A National Institutes of Health stage I behavioral intervention development trial evaluated a structured psychoeducational intervention, known as SPIRIT (Sharing Patient’s Illness Representation to Increase Trust), that aimed to promote cognitive and emotional preparation for EOL decisions for patients and their surrogates.²⁸ It was found to be effective in patients, including those with end-stage renal disease and advanced heart failure, and their surrogates.²⁸ Preliminary results from the trial confirmed that people with mild-to-­moderate dementia (recent Montreal Cognitive Assessment score ≥ 13) are able to participate meaningfully in EOL discussions and ACP.²⁸

Song et al²⁹ adapted SPIRIT for use with patients with dementia and conducted a feasibility study with 23 patient-surrogate dyads.The mixed-methods study involved an expert panel review of the adapted SPIRIT, followed by a randomized trial with qualitative interviews. All 23 patients with dementia, including 14 with moderate dementia, were able to articulate their values and EOL preferences somewhat or very coherently (91.3% inter-rater reliability).²⁹ In addition, dyad care goal congruence (agreement between patient’s EOL preferences and surrogate’s understanding of those preferences) and surrogate decision-making confidence (comfort in performing as a surrogate) were high and patient decisional conflict (patient difficulty in weighing the benefits and burdens of life-sustaining treatments and decision-making) was low, both at baseline as well as post ­intervention.²⁹ Although preparedness for EOL decision-making outcome measures did not change, people with dementia and their surrogates perceived SPIRIT to be beneficial, particularly in helping them be on the same page.²⁹

Patient ratings of physician communication improved when EOL topics such as spirituality, prognosis, and what dying might be like were discussed.

The randomized trial portion of the study (phase 2) continues to recruit 120 patient-surrogate dyads. Patient and surrogate self-reported preparedness for EOL decision-making are the primary outcomes, measured at baseline and 2 to 3 days post intervention. The estimated study completion date is May 31, 2022.³⁰

Evidence-based clinical guidance can improve the process. Following the Belgian Centre for Evidence-Based Medicine’s procedures as a sample methodology, Piers et al² developed evidence-based clinical recommendations for providers to use in the practical application of ACP in their care of patients with dementia.The researchers searched the literature; developed recommendations based on the evidence obtained, as well as their collective expert opinion; and performed validation using expert and end-user feedback and peer review. The study resulted in 32 recommendations focused on 8 domains that ranged from the beginning of the process (preconditions for optimal implementation of ACP) to later stages (ACP when it is difficult/no longer possible to communicate).²Specific guidance for ACP in dementia care include the following: 

• ACP initiation. Begin conversations around the time of diagnosis, continue them throughout ongoing care, and revisit them when changes occur in the patient’s health, financial, or residential status.
• ACP conversations. Use conversations to identify significant others in the patient’s life (potential caregivers and/or surrogate decision makers) and explore the patient’s awareness of the disease and its trajectory. Discuss the patient’s values and beliefs, as well as their fears about, and preferences for, future care and the EOL.  
• Role of significant others in the ACP process. Involve a patient’s significant others early in the ACP process, educate them about the surrogate ­decision-maker role, assess their disease awareness, and inform them about the disease trajectory and anticipated EOL decisions. ²

Continue to: Incorporate and document patients' values and preferences with LEAD

Incorporate and document patients’ values and preferences with LEAD. Dassel et al³¹ developed the Life-planning in Early Alzheimer’s and Dementia (LEAD) tool, which is a validated dementia-focused EOL planning tool that can be used to promote discussion and document a patient’s care preferences and values within the context of their changing cognitive ability.Dassel et al³¹ used a 4-phase mixed-method design that included (1) focus groups of patients with early-stage dementia and family caregivers, (2) clinical utility evaluation by content experts, (3) instrument completion sampling to evaluate its psychometric properties, and (4) additional focus groups to inform how the instrument should be used by families and in clinical practice.Six scales with high internal consistency and high test-retest reliability were identified: 3 scales represented patient values (concern about being a burden, the importance of quality [vs length] of life, and the preference for autonomy in decision-making) and 3 scales represented patient preferences (use of life-prolonging measures, controlling the timing of death, and the location of EOL care).³¹

When having EOL discussions with patients with dementia, one needs to consider the timing of such discussions and the trajectory of the disease.

The LEAD Guide can be used as a self-­assessment tool that is completed individually and then shared with the surrogate decision maker and health care provider.³² It also can be used to guide conversations with the surrogate and physician, as well as with trusted family and friends. Using this framework, family physicians can facilitate EOL planning with the patient and their surrogate that is based on the patient’s values and preferences for EOL care prior to, and in anticipation of, the patient’s loss of decisional capacity.³¹

Facilitate discussions that improve outcomes

Conversations about EOL care are taking on increased importance as the population ages and treatments advance. Understanding the concerns of patients and their surrogate decision makers, as well as the resources available to guide these difficult discussions ( TABLE ), will help family physicians conduct effective conversations that enhance care, reduce the burden on surrogate decision makers, and have a positive impact on many clinical outcomes.

CORRESPONDENCE
Shirley Bodi, MD, 3000 Arlington Avenue, Department of Family Medicine, Dowling Hall, Suite 2200, University of Toledo College of Medicine and Life Sciences, Toledo, OH 43614; [email protected]

As the geriatric population continues to grow and treatment advances blur the lines between improving the length of life vs improving its quality, end-of-life (EOL) conversations are becoming increasingly important. These discussions are a crucial part of the advance care planning (ACP) process, in which patients discuss their treatment preferences and values with their caregiver/surrogate decision maker and health care provider to ultimately improve EOL decision-making and care. ^1,2

EOL conversations are most helpful when incorporated in the outpatient setting as part of the patient’s ongoing health care plan or when initiating treatment for a chronic or life-threatening disease. Because family physicians promote general wellness, understand the patient’s health status and medical history, and have an ongoing—and often longstanding—relationship with patients and their families, we are ideally positioned to engage patients in EOL discussions. However, these conversations can be challenging in the outpatient setting, and often clinicians struggle not only to find ways to raise the subject, but also to find the time to have these supportive, meaningful conversations.³

In this article, we will address the importance of having EOL discussions in the outpatient setting, specifically about advance directives (ADs), and the reasons why patients and physicians might avoid these discussions. The role of palliative care in EOL care, along with its benefits and methods for overcoming patient and physician barriers to its successful use, are reviewed. Finally, we examine specific challenges associated with discussing EOL care with patients with decreased mental capacity, such as those with dementia, and provide strategies to successfully facilitate EOL discussions in these populations. 

Moving patients toward completion of advance directives

Although many older patients express a desire to document their wishes before EOL situations arise, they may not fully understand the benefits of an AD or how to complete one. ⁴ Often the family physician is best equipped to address the patient’s concerns and discuss their goals for EOL care, as well as the potential situations that might arise.

Managing an aging population. Projections suggest that primary care physicians will encounter increasing numbers of geriatric patients in the next 2 decades. Thus it is essential for those in primary care to receive proper training during their residency for the care of this group of patients. According to a group of academic educators and geriatricians from internal medicine and family medicine whose goal was to define a set of minimal and essential competencies in the care of older adults, this includes training on how to discuss and document “advance care planning and goals of care with all patients with chronic or complex illness,” as well as how to differentiate among “types of code status, health care proxies, and advanced directives” within the state in which training occurs. ⁵

Educate patients and ease fears. Patients often avoid EOL conversations or wait for their family physician to start the conversation. They may not understand how ADs can help guide care or they may believe they are “too healthy” to have these conversations at this time. ⁴ Simply asking about existing ADs or providing forms to patients during an outpatient visit can open the door to more in-depth discussions. Some examples of opening phrases include:

• Do you have a living will or durable power of attorney for health care?
• Have you ever discussed your health care wishes with your loved ones?
• Who would you want to speak for you regarding your health care if you could not speak for yourself? Have you discussed your health care wishes with that person?

By normalizing the conversation as a routine part of comprehensive, patient-centered care, the family physician can allay patient fears, foster open and honest conversations, and encourage ongoing discussions with loved ones as situations arise.⁶

Continue to: When ADs are executed...

As many as 90% of patients with a life-threatening illness report never having discussed EOL care issues with their clinician.

When ADs are executed, patients often fail to have meaningful conversations with their surrogates about specific treatment wishes or EOL scenarios. As a result, the surrogate may not feel prepared to serve as a proxy decision maker or may find the role extremely stressful.⁷ Physicians should encourage open conversations between patients and their surrogates about potential EOL scenarios when possible. When possible and appropriate, it is also important to encourage the patient to include the surrogate in future outpatient visits so that the surrogate can understand the patient’s health status and potential decisions they may need to make.

Don’t overlook clinician barriers. Family physicians also might avoid AD discussions because they do not understand laws that govern ADs, which vary from state to state. Various online resources for patients and physicians exist that clarify state-specific regulations and provide state-specific forms (TABLE).

Time constraints present another challenge for family physicians. This can be addressed by establishing workflows that include EOL elements. Also, the Centers for Medicare and Medicaid Services (CMS) has provided separate billing codes for AD discussion based on time spent explaining and discussing how to complete forms.⁸ CPT codes 99497 and 99498 are time-based codes that cover the first 30 minutes and each additional 30 minutes, respectively, of time spent explaining and discussing how to complete standard forms in a face-to-face setting (TABLE).⁹ CMS also includes discussion of AD documents as an optional element of the annual Medicare wellness visit.⁸ 

Improve quality of life for patients with any serious illness

Unlike hospice, which focuses on providing comfort rather than cure in the final months of a patient’s life, palliative care strives to prevent and relieve the patient’s suffering from a serious illness that is not immediately life-threatening. Palliative care focuses on the early identification, careful assessment, and treatment of the physical, psychosocial, and spiritual symptoms associated with a patient’s condition(s).^10,11 It has been well established that palliative care has a positive effect on many clinical outcomes including symptom burden, quality of life, satisfaction with care, and survival.^12-14 Patients who receive palliative care consultation also tend to perceive a higher quality of care.¹⁵

Conversations lead to better outcomes. Palliative care consultation is being increasingly used in the outpatient setting and can be introduced early in a disease process. Doing so provides an additional opportunity for the family physician to introduce an EOL discussion. A comparison of outcomes between patients who had initial inpatient palliative care consultation vs outpatient palliative care referral found that outpatient referral improved quality EOL care and was associated with significantly fewer emergency department visits (68% vs 48%; P < .001) and hospital admissions (86% vs 52%; P < .001), as well as shorter hospital stays in the last 30 days of life (3-11 vs 5-14 days; P = .01).¹⁴ Despite these benefits, 60% to 90% of patients with a serious illness report never having discussed EOL care issues with their clinician.^16,17

Continue to: Early EOL discussions...

Early EOL discussions have also been shown to have a positive impact on families. In a US study, family members stated that timely EOL care discussions allowed them to make use of hospice and palliative care services sooner and to make the most of their time with the patient.¹⁸

Timing and communication are key

Logistically it can be difficult to gather the right people (patient, family, etc) in the right place and at the right time. For physicians, the most often cited barriers include inadequate time to conduct an EOL discussion, ¹⁹ a perceived lack of competence in EOL conversations, ^1,20 difficulty navigating patient readiness, ²¹ and a fear of destroying hope due to prognostic uncertainty. ^19,20

A prospective, observational study used the Quality of Communication (QOC) questionnaire to assess life-sustaining treatment preferences, ACP, and the quality of EOL care communication in Dutch outpatients with clinically stable but severe chronic obstructive pulmonary disease (n = 105) or congestive heart failure (n = 80). The QOC questionnaire is a validated instrument that asks patients to rate their physician on several communication skills from 0 (“the very worst” or “My doctor didn’t do this”) to 10 (“the very best”). In this study, quality communication was identified by patients as one of the most important skills for physicians to provide adequate EOL care. ²² While QOC ratings were high for general communication skills (median, 8.0 points), quality EOL care communication was rated very low (median, 0.0 points). Researchers say that this was primarily because most EOL topics were not discussed—especially spirituality, prognosis, and what dying might be like. ²² In a secondary analysis that evaluated quality of EOL care communication during 1-year follow-up of patients with advanced chronic organ failure (n = 265) with the QOC questionnaire, patient ratings improved to moderate to good (medians, 6-8 points) when these topics were addressed. ²³  

Pick a strategy and prepare. As the older population continues to grow, the demands of palliative care management cannot be met by specialists alone and the responsibility of discussing EOL care with patients and their families will increasingly fall to family physicians as well. ²⁴ Several strategies and approaches have evolved to assist family physicians with acquiring the skills to conduct productive EOL discussions. These include widely referenced resources, such as VitalTalk ²⁵ and the ABCDE Plan. ²⁶ VitalTalk teaches skills to help clinicians navigate difficult conversations, ²⁵ and the “ABCDE” method provides a pneumonic for recommendations for how to deliver bad news ( A dvance preparation; B uild a therapeutic environment/relationship; C ommunicate well; D eal with patient and family reactions; E ncourage and validate emotions). ²⁶

Researchers found that timely EOL care discussions allowed family members to make use of hospice and palliative care services sooner and maximize their time with the patient.

Other strategies include familiarizing oneself with the patient’s medical history and present situation (eg, What are the patient’s symptoms? What do other involved clinicians think and recommend? What therapies have been attempted? What are the relevant social and emotional dynamics?); asking the patient who they want present for the EOL conversation; scheduling the conversation for when you can set aside an appropriate amount of time and in a private place where there will be no interruptions; and going into the meeting with your goal in mind, whether it is to deliver bad news, clarify the prognosis, establish goals of care, or communicate the patient’s goals and wishes for the EOL to those in attendance. ²⁷ It can be very helpful to begin the conversation by clarifying what the patient and their family/surrogate understand about the current diagnosis and prognosis. From there, the family physician can present a “headline” that prepares them for the current conversation (eg, “I have your latest test results, and I need to share some serious news”). This can facilitate a more detailed discussion of the patient’s and surrogate’s goals of care. Using these strategies, family physicians can lead a productive EOL discussion that benefits everyone.  

Continue to: How to navigate EOL discussions with patients with dementia

How to navigate EOL discussions with patients with dementia

EOL discussions with patients with dementia become even more complex and warrant specific discussion because one must consider the timing of such discussions, ^2,28,29 the trajectory of the disease and how that affects the patient’s capacity for EOL conversations, and the critical importance of engaging caregivers/surrogate decision makers in these discussions. ² ACP provides an opportunity for the physician, patient, and caregiver/­surrogate to jointly explore the patient’s values, beliefs, and preferences for care through the EOL as the disease progresses and the patient’s decisional capacity declines.

Ensure meaningful participation with timing. EOL discussions should occur while the patient has the cognitive capacity to actively participate in the planning process. A National Institutes of Health stage I behavioral intervention development trial evaluated a structured psychoeducational intervention, known as SPIRIT (Sharing Patient’s Illness Representation to Increase Trust), that aimed to promote cognitive and emotional preparation for EOL decisions for patients and their surrogates.²⁸ It was found to be effective in patients, including those with end-stage renal disease and advanced heart failure, and their surrogates.²⁸ Preliminary results from the trial confirmed that people with mild-to-­moderate dementia (recent Montreal Cognitive Assessment score ≥ 13) are able to participate meaningfully in EOL discussions and ACP.²⁸

Song et al²⁹ adapted SPIRIT for use with patients with dementia and conducted a feasibility study with 23 patient-surrogate dyads.The mixed-methods study involved an expert panel review of the adapted SPIRIT, followed by a randomized trial with qualitative interviews. All 23 patients with dementia, including 14 with moderate dementia, were able to articulate their values and EOL preferences somewhat or very coherently (91.3% inter-rater reliability).²⁹ In addition, dyad care goal congruence (agreement between patient’s EOL preferences and surrogate’s understanding of those preferences) and surrogate decision-making confidence (comfort in performing as a surrogate) were high and patient decisional conflict (patient difficulty in weighing the benefits and burdens of life-sustaining treatments and decision-making) was low, both at baseline as well as post ­intervention.²⁹ Although preparedness for EOL decision-making outcome measures did not change, people with dementia and their surrogates perceived SPIRIT to be beneficial, particularly in helping them be on the same page.²⁹

Patient ratings of physician communication improved when EOL topics such as spirituality, prognosis, and what dying might be like were discussed.

The randomized trial portion of the study (phase 2) continues to recruit 120 patient-surrogate dyads. Patient and surrogate self-reported preparedness for EOL decision-making are the primary outcomes, measured at baseline and 2 to 3 days post intervention. The estimated study completion date is May 31, 2022.³⁰

Evidence-based clinical guidance can improve the process. Following the Belgian Centre for Evidence-Based Medicine’s procedures as a sample methodology, Piers et al² developed evidence-based clinical recommendations for providers to use in the practical application of ACP in their care of patients with dementia.The researchers searched the literature; developed recommendations based on the evidence obtained, as well as their collective expert opinion; and performed validation using expert and end-user feedback and peer review. The study resulted in 32 recommendations focused on 8 domains that ranged from the beginning of the process (preconditions for optimal implementation of ACP) to later stages (ACP when it is difficult/no longer possible to communicate).²Specific guidance for ACP in dementia care include the following: 

• ACP initiation. Begin conversations around the time of diagnosis, continue them throughout ongoing care, and revisit them when changes occur in the patient’s health, financial, or residential status.
• ACP conversations. Use conversations to identify significant others in the patient’s life (potential caregivers and/or surrogate decision makers) and explore the patient’s awareness of the disease and its trajectory. Discuss the patient’s values and beliefs, as well as their fears about, and preferences for, future care and the EOL.  
• Role of significant others in the ACP process. Involve a patient’s significant others early in the ACP process, educate them about the surrogate ­decision-maker role, assess their disease awareness, and inform them about the disease trajectory and anticipated EOL decisions. ²

Continue to: Incorporate and document patients' values and preferences with LEAD

Incorporate and document patients’ values and preferences with LEAD. Dassel et al³¹ developed the Life-planning in Early Alzheimer’s and Dementia (LEAD) tool, which is a validated dementia-focused EOL planning tool that can be used to promote discussion and document a patient’s care preferences and values within the context of their changing cognitive ability.Dassel et al³¹ used a 4-phase mixed-method design that included (1) focus groups of patients with early-stage dementia and family caregivers, (2) clinical utility evaluation by content experts, (3) instrument completion sampling to evaluate its psychometric properties, and (4) additional focus groups to inform how the instrument should be used by families and in clinical practice.Six scales with high internal consistency and high test-retest reliability were identified: 3 scales represented patient values (concern about being a burden, the importance of quality [vs length] of life, and the preference for autonomy in decision-making) and 3 scales represented patient preferences (use of life-prolonging measures, controlling the timing of death, and the location of EOL care).³¹

When having EOL discussions with patients with dementia, one needs to consider the timing of such discussions and the trajectory of the disease.

The LEAD Guide can be used as a self-­assessment tool that is completed individually and then shared with the surrogate decision maker and health care provider.³² It also can be used to guide conversations with the surrogate and physician, as well as with trusted family and friends. Using this framework, family physicians can facilitate EOL planning with the patient and their surrogate that is based on the patient’s values and preferences for EOL care prior to, and in anticipation of, the patient’s loss of decisional capacity.³¹

Facilitate discussions that improve outcomes

Conversations about EOL care are taking on increased importance as the population ages and treatments advance. Understanding the concerns of patients and their surrogate decision makers, as well as the resources available to guide these difficult discussions ( TABLE ), will help family physicians conduct effective conversations that enhance care, reduce the burden on surrogate decision makers, and have a positive impact on many clinical outcomes.

CORRESPONDENCE
Shirley Bodi, MD, 3000 Arlington Avenue, Department of Family Medicine, Dowling Hall, Suite 2200, University of Toledo College of Medicine and Life Sciences, Toledo, OH 43614; [email protected]

As the geriatric population continues to grow and treatment advances blur the lines between improving the length of life vs improving its quality, end-of-life (EOL) conversations are becoming increasingly important. These discussions are a crucial part of the advance care planning (ACP) process, in which patients discuss their treatment preferences and values with their caregiver/surrogate decision maker and health care provider to ultimately improve EOL decision-making and care. ^1,2

EOL conversations are most helpful when incorporated in the outpatient setting as part of the patient’s ongoing health care plan or when initiating treatment for a chronic or life-threatening disease. Because family physicians promote general wellness, understand the patient’s health status and medical history, and have an ongoing—and often longstanding—relationship with patients and their families, we are ideally positioned to engage patients in EOL discussions. However, these conversations can be challenging in the outpatient setting, and often clinicians struggle not only to find ways to raise the subject, but also to find the time to have these supportive, meaningful conversations.³

In this article, we will address the importance of having EOL discussions in the outpatient setting, specifically about advance directives (ADs), and the reasons why patients and physicians might avoid these discussions. The role of palliative care in EOL care, along with its benefits and methods for overcoming patient and physician barriers to its successful use, are reviewed. Finally, we examine specific challenges associated with discussing EOL care with patients with decreased mental capacity, such as those with dementia, and provide strategies to successfully facilitate EOL discussions in these populations. 

Moving patients toward completion of advance directives

Although many older patients express a desire to document their wishes before EOL situations arise, they may not fully understand the benefits of an AD or how to complete one. ⁴ Often the family physician is best equipped to address the patient’s concerns and discuss their goals for EOL care, as well as the potential situations that might arise.

Managing an aging population. Projections suggest that primary care physicians will encounter increasing numbers of geriatric patients in the next 2 decades. Thus it is essential for those in primary care to receive proper training during their residency for the care of this group of patients. According to a group of academic educators and geriatricians from internal medicine and family medicine whose goal was to define a set of minimal and essential competencies in the care of older adults, this includes training on how to discuss and document “advance care planning and goals of care with all patients with chronic or complex illness,” as well as how to differentiate among “types of code status, health care proxies, and advanced directives” within the state in which training occurs. ⁵

Educate patients and ease fears. Patients often avoid EOL conversations or wait for their family physician to start the conversation. They may not understand how ADs can help guide care or they may believe they are “too healthy” to have these conversations at this time. ⁴ Simply asking about existing ADs or providing forms to patients during an outpatient visit can open the door to more in-depth discussions. Some examples of opening phrases include:

• Do you have a living will or durable power of attorney for health care?
• Have you ever discussed your health care wishes with your loved ones?
• Who would you want to speak for you regarding your health care if you could not speak for yourself? Have you discussed your health care wishes with that person?

By normalizing the conversation as a routine part of comprehensive, patient-centered care, the family physician can allay patient fears, foster open and honest conversations, and encourage ongoing discussions with loved ones as situations arise.⁶

Continue to: When ADs are executed...

As many as 90% of patients with a life-threatening illness report never having discussed EOL care issues with their clinician.

When ADs are executed, patients often fail to have meaningful conversations with their surrogates about specific treatment wishes or EOL scenarios. As a result, the surrogate may not feel prepared to serve as a proxy decision maker or may find the role extremely stressful.⁷ Physicians should encourage open conversations between patients and their surrogates about potential EOL scenarios when possible. When possible and appropriate, it is also important to encourage the patient to include the surrogate in future outpatient visits so that the surrogate can understand the patient’s health status and potential decisions they may need to make.

Don’t overlook clinician barriers. Family physicians also might avoid AD discussions because they do not understand laws that govern ADs, which vary from state to state. Various online resources for patients and physicians exist that clarify state-specific regulations and provide state-specific forms (TABLE).

Time constraints present another challenge for family physicians. This can be addressed by establishing workflows that include EOL elements. Also, the Centers for Medicare and Medicaid Services (CMS) has provided separate billing codes for AD discussion based on time spent explaining and discussing how to complete forms.⁸ CPT codes 99497 and 99498 are time-based codes that cover the first 30 minutes and each additional 30 minutes, respectively, of time spent explaining and discussing how to complete standard forms in a face-to-face setting (TABLE).⁹ CMS also includes discussion of AD documents as an optional element of the annual Medicare wellness visit.⁸ 

Improve quality of life for patients with any serious illness

Unlike hospice, which focuses on providing comfort rather than cure in the final months of a patient’s life, palliative care strives to prevent and relieve the patient’s suffering from a serious illness that is not immediately life-threatening. Palliative care focuses on the early identification, careful assessment, and treatment of the physical, psychosocial, and spiritual symptoms associated with a patient’s condition(s).^10,11 It has been well established that palliative care has a positive effect on many clinical outcomes including symptom burden, quality of life, satisfaction with care, and survival.^12-14 Patients who receive palliative care consultation also tend to perceive a higher quality of care.¹⁵

Conversations lead to better outcomes. Palliative care consultation is being increasingly used in the outpatient setting and can be introduced early in a disease process. Doing so provides an additional opportunity for the family physician to introduce an EOL discussion. A comparison of outcomes between patients who had initial inpatient palliative care consultation vs outpatient palliative care referral found that outpatient referral improved quality EOL care and was associated with significantly fewer emergency department visits (68% vs 48%; P < .001) and hospital admissions (86% vs 52%; P < .001), as well as shorter hospital stays in the last 30 days of life (3-11 vs 5-14 days; P = .01).¹⁴ Despite these benefits, 60% to 90% of patients with a serious illness report never having discussed EOL care issues with their clinician.^16,17

Continue to: Early EOL discussions...

Early EOL discussions have also been shown to have a positive impact on families. In a US study, family members stated that timely EOL care discussions allowed them to make use of hospice and palliative care services sooner and to make the most of their time with the patient.¹⁸

Timing and communication are key

Logistically it can be difficult to gather the right people (patient, family, etc) in the right place and at the right time. For physicians, the most often cited barriers include inadequate time to conduct an EOL discussion, ¹⁹ a perceived lack of competence in EOL conversations, ^1,20 difficulty navigating patient readiness, ²¹ and a fear of destroying hope due to prognostic uncertainty. ^19,20

A prospective, observational study used the Quality of Communication (QOC) questionnaire to assess life-sustaining treatment preferences, ACP, and the quality of EOL care communication in Dutch outpatients with clinically stable but severe chronic obstructive pulmonary disease (n = 105) or congestive heart failure (n = 80). The QOC questionnaire is a validated instrument that asks patients to rate their physician on several communication skills from 0 (“the very worst” or “My doctor didn’t do this”) to 10 (“the very best”). In this study, quality communication was identified by patients as one of the most important skills for physicians to provide adequate EOL care. ²² While QOC ratings were high for general communication skills (median, 8.0 points), quality EOL care communication was rated very low (median, 0.0 points). Researchers say that this was primarily because most EOL topics were not discussed—especially spirituality, prognosis, and what dying might be like. ²² In a secondary analysis that evaluated quality of EOL care communication during 1-year follow-up of patients with advanced chronic organ failure (n = 265) with the QOC questionnaire, patient ratings improved to moderate to good (medians, 6-8 points) when these topics were addressed. ²³  

Pick a strategy and prepare. As the older population continues to grow, the demands of palliative care management cannot be met by specialists alone and the responsibility of discussing EOL care with patients and their families will increasingly fall to family physicians as well. ²⁴ Several strategies and approaches have evolved to assist family physicians with acquiring the skills to conduct productive EOL discussions. These include widely referenced resources, such as VitalTalk ²⁵ and the ABCDE Plan. ²⁶ VitalTalk teaches skills to help clinicians navigate difficult conversations, ²⁵ and the “ABCDE” method provides a pneumonic for recommendations for how to deliver bad news ( A dvance preparation; B uild a therapeutic environment/relationship; C ommunicate well; D eal with patient and family reactions; E ncourage and validate emotions). ²⁶

Researchers found that timely EOL care discussions allowed family members to make use of hospice and palliative care services sooner and maximize their time with the patient.

Other strategies include familiarizing oneself with the patient’s medical history and present situation (eg, What are the patient’s symptoms? What do other involved clinicians think and recommend? What therapies have been attempted? What are the relevant social and emotional dynamics?); asking the patient who they want present for the EOL conversation; scheduling the conversation for when you can set aside an appropriate amount of time and in a private place where there will be no interruptions; and going into the meeting with your goal in mind, whether it is to deliver bad news, clarify the prognosis, establish goals of care, or communicate the patient’s goals and wishes for the EOL to those in attendance. ²⁷ It can be very helpful to begin the conversation by clarifying what the patient and their family/surrogate understand about the current diagnosis and prognosis. From there, the family physician can present a “headline” that prepares them for the current conversation (eg, “I have your latest test results, and I need to share some serious news”). This can facilitate a more detailed discussion of the patient’s and surrogate’s goals of care. Using these strategies, family physicians can lead a productive EOL discussion that benefits everyone.  

Continue to: How to navigate EOL discussions with patients with dementia

How to navigate EOL discussions with patients with dementia

EOL discussions with patients with dementia become even more complex and warrant specific discussion because one must consider the timing of such discussions, ^2,28,29 the trajectory of the disease and how that affects the patient’s capacity for EOL conversations, and the critical importance of engaging caregivers/surrogate decision makers in these discussions. ² ACP provides an opportunity for the physician, patient, and caregiver/­surrogate to jointly explore the patient’s values, beliefs, and preferences for care through the EOL as the disease progresses and the patient’s decisional capacity declines.

Ensure meaningful participation with timing. EOL discussions should occur while the patient has the cognitive capacity to actively participate in the planning process. A National Institutes of Health stage I behavioral intervention development trial evaluated a structured psychoeducational intervention, known as SPIRIT (Sharing Patient’s Illness Representation to Increase Trust), that aimed to promote cognitive and emotional preparation for EOL decisions for patients and their surrogates.²⁸ It was found to be effective in patients, including those with end-stage renal disease and advanced heart failure, and their surrogates.²⁸ Preliminary results from the trial confirmed that people with mild-to-­moderate dementia (recent Montreal Cognitive Assessment score ≥ 13) are able to participate meaningfully in EOL discussions and ACP.²⁸

Song et al²⁹ adapted SPIRIT for use with patients with dementia and conducted a feasibility study with 23 patient-surrogate dyads.The mixed-methods study involved an expert panel review of the adapted SPIRIT, followed by a randomized trial with qualitative interviews. All 23 patients with dementia, including 14 with moderate dementia, were able to articulate their values and EOL preferences somewhat or very coherently (91.3% inter-rater reliability).²⁹ In addition, dyad care goal congruence (agreement between patient’s EOL preferences and surrogate’s understanding of those preferences) and surrogate decision-making confidence (comfort in performing as a surrogate) were high and patient decisional conflict (patient difficulty in weighing the benefits and burdens of life-sustaining treatments and decision-making) was low, both at baseline as well as post ­intervention.²⁹ Although preparedness for EOL decision-making outcome measures did not change, people with dementia and their surrogates perceived SPIRIT to be beneficial, particularly in helping them be on the same page.²⁹

Patient ratings of physician communication improved when EOL topics such as spirituality, prognosis, and what dying might be like were discussed.

The randomized trial portion of the study (phase 2) continues to recruit 120 patient-surrogate dyads. Patient and surrogate self-reported preparedness for EOL decision-making are the primary outcomes, measured at baseline and 2 to 3 days post intervention. The estimated study completion date is May 31, 2022.³⁰

Evidence-based clinical guidance can improve the process. Following the Belgian Centre for Evidence-Based Medicine’s procedures as a sample methodology, Piers et al² developed evidence-based clinical recommendations for providers to use in the practical application of ACP in their care of patients with dementia.The researchers searched the literature; developed recommendations based on the evidence obtained, as well as their collective expert opinion; and performed validation using expert and end-user feedback and peer review. The study resulted in 32 recommendations focused on 8 domains that ranged from the beginning of the process (preconditions for optimal implementation of ACP) to later stages (ACP when it is difficult/no longer possible to communicate).²Specific guidance for ACP in dementia care include the following: 

• ACP initiation. Begin conversations around the time of diagnosis, continue them throughout ongoing care, and revisit them when changes occur in the patient’s health, financial, or residential status.
• ACP conversations. Use conversations to identify significant others in the patient’s life (potential caregivers and/or surrogate decision makers) and explore the patient’s awareness of the disease and its trajectory. Discuss the patient’s values and beliefs, as well as their fears about, and preferences for, future care and the EOL.  
• Role of significant others in the ACP process. Involve a patient’s significant others early in the ACP process, educate them about the surrogate ­decision-maker role, assess their disease awareness, and inform them about the disease trajectory and anticipated EOL decisions. ²

Continue to: Incorporate and document patients' values and preferences with LEAD

Incorporate and document patients’ values and preferences with LEAD. Dassel et al³¹ developed the Life-planning in Early Alzheimer’s and Dementia (LEAD) tool, which is a validated dementia-focused EOL planning tool that can be used to promote discussion and document a patient’s care preferences and values within the context of their changing cognitive ability.Dassel et al³¹ used a 4-phase mixed-method design that included (1) focus groups of patients with early-stage dementia and family caregivers, (2) clinical utility evaluation by content experts, (3) instrument completion sampling to evaluate its psychometric properties, and (4) additional focus groups to inform how the instrument should be used by families and in clinical practice.Six scales with high internal consistency and high test-retest reliability were identified: 3 scales represented patient values (concern about being a burden, the importance of quality [vs length] of life, and the preference for autonomy in decision-making) and 3 scales represented patient preferences (use of life-prolonging measures, controlling the timing of death, and the location of EOL care).³¹

When having EOL discussions with patients with dementia, one needs to consider the timing of such discussions and the trajectory of the disease.

The LEAD Guide can be used as a self-­assessment tool that is completed individually and then shared with the surrogate decision maker and health care provider.³² It also can be used to guide conversations with the surrogate and physician, as well as with trusted family and friends. Using this framework, family physicians can facilitate EOL planning with the patient and their surrogate that is based on the patient’s values and preferences for EOL care prior to, and in anticipation of, the patient’s loss of decisional capacity.³¹

Facilitate discussions that improve outcomes

Conversations about EOL care are taking on increased importance as the population ages and treatments advance. Understanding the concerns of patients and their surrogate decision makers, as well as the resources available to guide these difficult discussions ( TABLE ), will help family physicians conduct effective conversations that enhance care, reduce the burden on surrogate decision makers, and have a positive impact on many clinical outcomes.

CORRESPONDENCE
Shirley Bodi, MD, 3000 Arlington Avenue, Department of Family Medicine, Dowling Hall, Suite 2200, University of Toledo College of Medicine and Life Sciences, Toledo, OH 43614; [email protected]

Gestational diabetes mellitus (GDM), defined as new-onset hyperglycemia detected in a pregnant woman after 24 weeks of gestation, affects 4% to 10% of pregnancies in the United States annually¹ and is a major challenge for health care professionals.² During pregnancy, the body’s physiologic responses are altered to support the growing fetus. One of these changes is an increase in insulin resistance, which suggests that pregnancy alone increases the patient’s risk for type 2 diabetes (T2D). However, several other factors also increase this risk, including maternal age, social barriers to care, obesity, poor weight control, and family history.

Copyright Dave Cutler

If not controlled, GDM results in poor health outcomes for the mother, such as preeclampsia, preterm labor, and maternal T2D.^3-5 For the infant, intrauterine exposure to persistent hyperglycemia is correlated with neonatal macrosomia, hypoglycemia, perinatal complications (eg, preterm delivery, fetal demise), and obesity and insulin resistance later in life.⁴

Primary care physicians (PCPs) are the patient’s main point of contact prior to pregnancy. This relationship makes PCPs a resource for the patient and specialists during and after pregnancy. In this article, we discuss risk factors and how to screen for GDM, provide an update on practice recommendations for treatment and management of GDM in primary care, and describe the effects of uncontrolled GDM.

Know the key risk factors

Prevention begins with identifying the major risk factors that contribute to the development of GDM. These include maternal age, social barriers to care, family history of prediabetes, and obesity and poor weight control.

Older age. A meta-analysis of 24 studies noted strong positive correlation between GDM risk and maternal age.⁶ One of the population-based cohort studies in the meta-analysis examined relationships between maternal age and pregnancy outcomes in women living in British Columbia, Canada (n = 203,414). Data suggested that the relative risk of GDM increased linearly with maternal age to 3.2, 4.2, and 4.4 among women ages ≥ 35, ≥ 40, and ≥ 45 years, respectively.⁷

Social barriers to care. Although the prevalence of GDM has increased over the past few decades,¹ from 2011 to 2019 the increase in GDM in individuals at first live birth was significantly higher in non-Hispanic Asian and Hispanic/Latina women than in non-Hispanic White women.⁸ Data from the Centers for Disease Control and Prevention further suggest that diabetes was more prevalent among individuals with a lower socioeconomic status as indicated by their level of education.⁹ Ogunwole et al¹⁰ suggest that racism is the root cause of these disparities and leads to long-term barriers to care (eg, socioeconomic deprivation, lack of health insurance, limited access to care, and poor health literacy), which ultimately contribute to the development of GDM and progression of diabetes. It is important for PCPs and all health professionals to be aware of these barriers so that they may practice mindfulness and deliver culturally sensitive care to patients from marginalized communities.

Family history of prediabetes. In a population-based cohort study (n = 7020), women with prediabetes (A1C, 5.7%-6.4%) were 2.8 times more likely to develop GDM compared with women with normal A1C (< 5.7%).¹¹ Similar results were seen in a retrospective cohort study (n = 2812), in which women with prediabetes were more likely than women with a normal first trimester A1C to have GDM (29.1% vs 13.7%, respectively; adjusted relative risk = 1.48; 95% CI, 1.15-1.89).¹² In both studies, prediabetes was not associated with a higher risk for adverse maternal or neonatal outcomes.^11,12

Continue to: While there are no current...

Women diagnosed with prediabetes in 1 study were found to have significantly less weight gain during pregnancy compared with patients with normal A1C, suggesting a benefit in early identification and intervention.

While there are no current guidelines for treating prediabetes in pregnancy, women diagnosed with prediabetes in 1 study were found to have significantly less weight gain during pregnancy compared with patients with normal A1C,¹² suggesting there may be a benefit in early identification and intervention, although further research is needed.¹¹ In a separate case-control study (n = 345 women with GDM; n = 800 control), high rates of gestational weight gain (> 0.41 kg/wk) were associated with an increased risk of GDM (odds ratio [OR] = 1.74; 95% CI, 1.16-2.60) compared with women with the lowest rate of gestational weight gain (0.27-0.4 kg/wk [OR = 1.43; 95% CI, 0.96-2.14]).¹³ Thus, it is helpful to have proactive conversations about family planning and adequate weight and glycemic control with high-risk patients to prepare for a healthy pregnancy.

Obesity and weight management. Patients who are overweight (body mass index [BMI], 25-29.9) or obese (BMI > 30) have a substantially increased risk of GDM (adjusted OR = 1.44; 95% CI, 1.04-1.81), as seen in a retrospective cohort study of 1951 pregnant Malaysian women.¹⁴ Several factors have been found to contribute to successful weight control, including calorie prescription, a structured meal plan, high physical activity goals (60-90 min/d), daily weighing and monitoring of food intake, behavior therapy, and continued patient–­provider contact.¹⁵

Most obstetricians use a 2-step method to screen for GDM with an initial 75-g oral glucose tolerance test, followed by a 50-g glucose load test if needed.

The safety, efficacy, and sustainability of weight loss with various dietary plans have been studied in individuals who are overweight and obese.¹⁶ Ultimately, energy expenditure must be greater than energy intake to promote weight loss. Conventional diets with continuous energy restriction (ie, low-fat, low-carbohydrate, and high-protein diets) have proven to be effective for short-term weight loss but data on long-term weight maintenance are limited.¹⁶ The Mediterranean diet, which is comprised mostly of vegetables, fruits, legumes, fish, and grains—with a lower intake of meat and dairy—may reduce gestational weight gain and risk of GDM as suggested by a randomized controlled trial (RCT; n = 1252).¹⁷ Although the choice of diet is up to the patient, it is important to be aware of different diets or refer the patient to a registered dietician who can help the patient if needed.

Reduce risk with adequate weight and glycemic control

Prevention of GDM during pregnancy should focus on weight maintenance and optimal glycemic control. Two systematic reviews, one with 8 RCTs (n = 1792) and another with 5 studies (n = 539), assessed the efficacy and safety of energy-restricted dietary intervention on GDM prevention.¹⁸ The first review found a significant reduction in gestational weight gain and improved glycemic control without increased risk of adverse maternal and fetal outcomes.¹⁸ The second review showed no clear difference between energy-restricted and non–energy-restricted diets on outcomes such as preeclampsia, gestational weight gain, large for gestational age, and macrosomia.¹⁸ These data suggest that while energy-restricted dietary interventions made no difference on maternal and fetal complications, they may still be safely used in pregnancy to reduce gestational weight gain and improve glycemic control.¹⁸

Once a woman is pregnant, it becomes difficult to lose weight because additional calories are needed to support a growing fetus. It is recommended that patients with healthy pregestational BMI consume an extra 200 to 300 calories/d after the first trimester. However, extra caloric intake in a woman with obesity who is pregnant leads to metabolic impairment and increased risk of diabetes for both the mother and fetus.¹⁹ Therefore, it is recommended that patients with obese pregestational BMI not consume additional calories because excess maternal fat is sufficient to support the energy needs of the growing fetus.¹⁹

Continue to: Ultimately, earlier intervention...

Ultimately, earlier intervention—prior to conception—helps patients prepare for a healthier pregnancy, resulting in better long-term outcomes. It is helpful to be familiar with the advantages and disadvantages of common approaches to weight management and to be able to refer patients to nutritionists for optimal planning. When establishing a dietary plan, consider patient-specific factors, such as cultural diets, financial and time constraints, and the patient’s readiness to make and maintain these changes. Consistent ­follow-up and behavioral therapy are necessary to maintain successful weight control.

There are many screening tools, but 1 is preferred in pregnancy

There are several ways to diagnose diabetes in patients who are not pregnant, including A1C, a fasting glucose test, an oral glucose tolerance test (OGTT), or random glucose testing (plus symptoms). However, the preferred method for diagnosing GDM is OGTT because it has a higher sensitivity.²⁰ A1C, while a good measure of hyperglycemic stability, does not register hyperglycemia early enough to diagnose GDM and fasting glucose testing is less sensitive because for most women with GDM, that abnormal postprandial glucose level is the first glycemic abnormality.²¹

When to screen. Blood glucose levels should be checked in all pregnant women as part of their metabolic panel at the first prenatal visit. A reflex A1C for high glucose levels can be ordered based on the physician’s preference. This may help you to identify patients with prediabetes who are at risk for GDM and implement early behavioral and lifestyle changes. However, further research is needed to determine if intervention early in pregnancy can truly reduce the risk of GDM.¹¹

The A1C goal for women with GDM is lower (6.0%) after the first trimester because any rise in A1C is risky and increased red blood cell count turnover may lower A1C.

Screening for GDM should be completed at 24 to 28 weeks of gestation²⁰ because it is likely that this is when the hormonal effects of the placenta that contribute to insulin resistance set the woman up for postprandial hyperglycemia. Currently, there are no evidence-based guidelines for the use of continuous glucose monitoring prior to 24 weeks of gestation to identify GDM.²⁰ If persistent hyperglycemia is present before 24 weeks of gestation, it is considered evidence of a pre-existing metabolic abnormality and is diagnosed as “pregestational diabetes.” Treatment should follow guidelines established for women who had diabetes prior to pregnancy.

How to screen? There is ongoing discussion about what is the optimal screening method for GDM: a 1-step strategy with a fasting 75-g OGTT only, or a 2-step strategy with a 50-g non-fasting glucose load test followed by a fasting 100-g OGTT in women who do not meet the plasma glucose cutoff (TABLE 1).^22-24 Hillier et al²⁵ compared the effectiveness of these strategies in diagnosing GDM and identifying pregnancy complications for the mother and infant. They found that while the 1-step strategy resulted in a 2-fold increase in the diagnosis of GDM, it did not lead to better outcomes for mothers and infants when compared with the 2-step method.²⁵ Currently, the majority of obstetricians (95%) prefer to use the 2-step method.²⁴

Continue to: Manage lifestyle, monitor glucose

Manage lifestyle, monitor glucose

Management of GDM in most women starts with diabetes self-management education and support for therapeutic lifestyle changes, such as nutritional interventions that reduce hyperglycemia and contribute to healthy weight gain during pregnancy.²⁰ This may include medical nutrition therapy that focuses on adequate nutrition for the mother and fetus. Currently, the recommended dietary intake for women who are pregnant (regardless of diabetes) includes a minimum of 175 g of carbohydrates, 71 g of daily protein, and at least 28 g of fiber. Further refinement of dietary intake, including carbohydrate restriction, should be done with guidance from a registered dietitian.²⁰ If the obstetrics team does not include a registered dietitian, a referral to one may be necessary. Regular physical activity should be continued throughout pregnancy as tolerated. Social support, stress reduction, and good sleep hygiene should be encouraged as much as possible.

For successful outcomes, therapeutic lifestyle changes should be coupled with glucose monitoring. The Fifth International Workshop-Conference on Gestational Diabetes Mellitus recommends that women with GDM monitor fasting blood glucose and typically 1-hour postprandial glucose. The glucose goals in GDM are as follows²⁶:

• Fasting glucose < 95 mg/dL (5.3 mmol/L), and either
• 1-hour postprandial glucose < 140 mg/dL (7.8 mmol/L), or
• 2-hour postprandial glucose < 120 mg/dL (6.7 mmol/L).

Importantly, in the second and third trimester, the A1C goal for women with GDM is 6.0%. This is lower than the more traditional A1C goal for 2 reasons: (1) increases in A1C, even within the normal range, increase adverse outcomes; and (2) pregnant women will have an increased red blood cell count turnover, which can lower the A1C.²⁷ In a historical cohort study (n = 27,213), Abell et al²⁸ found that women who have an A1C < 6.0% in the second and third trimester have the lowest risk of giving birth to large-for-gestational-age infants and for having preeclampsia.

Add insulin if glucose targets are not met

Most women who engage in therapeutic lifestyle change (70%-85%) can achieve an A1C < 6% and will not need to take medication to manage GDM.²⁹ If pharmacotherapy is needed to manage glucose, insulin is the preferred treatment for all women with GDM.²⁰ Treatment should be individualized based on the glucose trends the woman is experiencing. Common treatments include bedtime NPH if fasting hyperglycemia is most prominent and analogue insulin at mealtimes for women with prominent postprandial hyperglycemia.

Most women who engage in therapeutic lifestyle change (70%-85%) can achieve an A1C < 6% and will not need to take medication to manage GDM.

Noninsulin agents such as metformin and sulfonylureas are not currently recommended by the American College of Obstetricians and Gynecologists or the American Diabetes Association for use in GDM.^20,24 Despite being used for years in women with pregestational diabetes, metabolic syndrome, and polycystic ovary syndrome, there is evidence that metformin crosses the placenta and fetal safety has not yet been established in RCTs. The Metformin in Gestational Diabetes: The Offspring Follow-Up (MiG TOFU) study was a longitudinal follow-up study that evaluated body composition and metabolic outcomes in children (ages 7-9 years) of women with GDM who had received metformin or insulin while pregnant.³⁰ At age 9 years, children who were exposed to metformin weighed more and had a higher waist-to-height ratio and waist circumference than those exposed to insulin.³⁰

Continue to: Sulfonylureas are no longer recommended...

Sulfonylureas are no longer recommended because of the risk of maternal and fetal hypoglycemia and concerns about this medication crossing the placenta.^24,31,32 Specifically, in a 2015 meta-analysis and systematic review of 15 articles (n = 2509), glyburide had a higher risk of neonatal hypoglycemia and macrosomia than insulin or metformin.³³ For women who cannot manage their glucose with therapeutic lifestyle changes and cannot take insulin, oral therapies may be considered if the risk-benefit ratio is balanced for that person.³⁴

Watch for effects of poor glycemic control on mother, infant

Preeclampsia is defined as new-onset hypertension and proteinuria after 20 weeks of gestation. The correlation between GDM and preeclampsia has partly been explained by their shared overlapping risk factors, including maternal obesity, excessive gestational weight gain, and persistent hyperglycemia.³⁵ On a biochemical level, these risk factors contribute to oxidative stress and systemic vascular dysfunction, which have been hypothesized as the underlying pathophysiology for the development of preeclampsia.³⁵

Neonatal macrosomia, defined as a birth weight ≥ 4000 g, is a common complication that develops in 15% to 45% of infants of mothers with GDM.³⁶ Placental transfer of glucose in mothers with hyperglycemia stimulates the secretion of neonatal insulin and the ultimate storage of the excess glucose as body fat. After delivery, the abrupt discontinuation of placental transfer of glucose to an infant who is actively secreting insulin leads to neonatal hypoglycemia, which if not detected or managed, can lead to long-term neurologic deficits, including recurrent seizures and developmental delays.³⁷ Therefore, it is essential to screen for neonatal hypoglycemia immediately after birth and serially up to 12 hours.³⁸

Postpartum T2D. Poor glycemic control increases the risk of increasing insulin resistance developing into T2D postpartum for mothers.³⁹ It also increases the risk of obesity and insulin resistance later in life for the infant.⁴⁰ A retrospective cohort study (n = 461) found a positive correlation between exposure to maternal GDM and elevated BMI in children ages 6 to 13 years.⁴¹ Kamana et al³⁶ further discussed this correlation and suggested that exposure to maternal hyperglycemia in utero contributes to fetal programming of later adipose deposition. Children may develop without a notable increase in BMI until after puberty.⁴²

Partner with specialists to improve outcomes

Although most women with GDM are managed by specialists (obstetricians, endocrinologists, and maternal-fetal medicine specialists),⁴³ these patients are still seeking care from their family physicians for other complaints. These visits provide key touchpoints during pregnancy and are opportunities for PCPs to identify a pregnancy-related complication or provide additional education or referral to the obstetrician.

Continue to: Also, if you work in an area...

Noninsulin agents, such as metformin and sulfonylureas, are not currently recommended by ACOG or the ADA for use in GDM.

Also, if you work in an area where specialists are less accessible, you may be the clinician providing the majority of care to a patient with GDM. If this is the case, you’ll want to watch for the following risk factors, which should prompt a referral to specialty care:

• a previous pregnancy with GDM²⁰
• a previous birth of an infant weighing > 4000 g⁴⁴
• baseline history of hypertension⁴⁵
• evidence of insulin resistance or polycystic ovary syndrome^46,47
• a history of cardiovascular disease²⁰
• a need to treat GDM with pharmacotherapy.⁴⁸

Ensuring a smooth transition after the birth

Optimal communication and hand-offs throughout pregnancy and after delivery will benefit everyone. When the pregnant patient’s care has been managed by an obstetrician, it is important to address the following issues during the hand-off:

• baseline medical problems
• medical screenings and treatments in pregnancy (retinopathy and nephropathy screening)
• aspirin initiation, if indicated
• management of thyroid abnormalities
• management of mental health conditions
• postpartum glucose management and T2D screening postpartum
• management of complications identified during pregnancy (retinopathy and nephropathy).

Timing and other elements of postpartum care. The first postpartum screen should occur at 4 to 12 weeks postpartum. OGTT is recommended instead of A1C at this time because A1C may still be lowered by the increased red blood cell turnover related to pregnancy and blood loss at delivery. Because women with GDM have a 50% to 75% lifetime risk of T2D,²⁰ patients with normal test results should be re-tested every 1 to 3 years using any of the standard screening methods (A1C, fasting glucose, or OGTT).²⁰

Postpartum visits present another opportunity to screen for diabetes and other postpartum complications, including depression and thyroid abnormalities.

After delivery it may be difficult for women to follow-up with their own personal health care because they are focused on the care of their baby. The increased use of telehealth may make postpartum follow-up visits easier to attend.

Visits present opportunities. Postpartum visits present another opportunity for PCPs to screen for diabetes and other postpartum complications, including depression and thyroid abnormalities. Visits are also an opportunity to discuss timely contraception so as to prevent an early, unplanned pregnancy. Other important aspects of postpartum care are outlined in TABLE 2.^20,49

CORRESPONDENCE
Connie L. Ha, BS, OMS IV, Department of Primary Care, 1310 Club Drive, Touro University California, Vallejo, CA 94592; [email protected]