EVIDENCE SUMMARY

Three meta-analyses that included only randomized controlled trials (RCTs) compared various CBT delivery methods with controls (wait-listed for treatment or general sleep hygiene education) to assess sleep outcomes for adults with insomnia.^1-3 TABLE 1^1-3 summarizes the results.

CBT is comparable to pharmacotherapy

A 2002 comparative meta-analysis of 21 RCTs with a total of 470 patients examined the effectiveness of CBT (stimulus control and/or sleep restriction) compared with pharmacotherapy (benzodiazepines or benzodiazepine agonists) for treating primary insomnia of longer than one month’s duration in adults with no comorbid medical or psychiatric diagnoses.⁴ The CBT group received intervention over an average of 5 weeks, and the pharmacotherapy group received intervention over an average of 2 weeks.

CBT produced greater reductions in sleep onset latency than pharmacotherapy based on mean weighted effect size (1.05 vs 0.45 weighted effect size; 95% confidence interval, 0.17-1.04; P=.01). Although both CBT and pharmacotherapy improved sleep outcomes, no statistical differences were found in wake after sleep onset time, total sleep time, number of awakenings, or sleep quality ratings (TABLE 2⁴).

Continue to: CBT has significant benefit for comorbid insomnia

A 2015 meta-analysis of 23 studies enrolling a total of 1379 adults with a number of illnesses (chronic pain, alcohol dependence, breast cancer, psychiatric disorders, chronic obstructive pulmonary disease, fibromyalgia) and comorbid insomnia investigated the qualitative effectiveness of individual or group CBT therapy.⁵ Subjects received at least 4 face-to-face sessions and at least 2 components of CBT.

The primary outcome showed that sleep quality improved, as measured by a 6.36-point reduction in the Insomnia Severity Index (ISI; a 7-question scale on which 0=no insomnia and 28=severe insomnia) and a 3.3-point reduction in the Pittsburgh Sleep Quality Index (PSQI; a 7-category assessment tool on which 0=perfect quality and 21=poor quality). The effect size was large for both ISI and PSQI, as indicated by standard mean differences greater than 0.8 (1.22 and 0.88, respectively) and was sustained for as long as 18 months.

RECOMMENDATIONS

The American College of Physicians strongly recommends that all adult patients receive CBT as initial treatment for chronic insomnia disorder. It can be performed in multiple settings, including the primary care setting. Compared with hypnotics, CBT is unlikely to have any adverse effects.⁶

EVIDENCE SUMMARY

Three meta-analyses that included only randomized controlled trials (RCTs) compared various CBT delivery methods with controls (wait-listed for treatment or general sleep hygiene education) to assess sleep outcomes for adults with insomnia.^1-3 TABLE 1^1-3 summarizes the results.

CBT is comparable to pharmacotherapy

A 2002 comparative meta-analysis of 21 RCTs with a total of 470 patients examined the effectiveness of CBT (stimulus control and/or sleep restriction) compared with pharmacotherapy (benzodiazepines or benzodiazepine agonists) for treating primary insomnia of longer than one month’s duration in adults with no comorbid medical or psychiatric diagnoses.⁴ The CBT group received intervention over an average of 5 weeks, and the pharmacotherapy group received intervention over an average of 2 weeks.

CBT produced greater reductions in sleep onset latency than pharmacotherapy based on mean weighted effect size (1.05 vs 0.45 weighted effect size; 95% confidence interval, 0.17-1.04; P=.01). Although both CBT and pharmacotherapy improved sleep outcomes, no statistical differences were found in wake after sleep onset time, total sleep time, number of awakenings, or sleep quality ratings (TABLE 2⁴).

Continue to: CBT has significant benefit for comorbid insomnia

A 2015 meta-analysis of 23 studies enrolling a total of 1379 adults with a number of illnesses (chronic pain, alcohol dependence, breast cancer, psychiatric disorders, chronic obstructive pulmonary disease, fibromyalgia) and comorbid insomnia investigated the qualitative effectiveness of individual or group CBT therapy.⁵ Subjects received at least 4 face-to-face sessions and at least 2 components of CBT.

The primary outcome showed that sleep quality improved, as measured by a 6.36-point reduction in the Insomnia Severity Index (ISI; a 7-question scale on which 0=no insomnia and 28=severe insomnia) and a 3.3-point reduction in the Pittsburgh Sleep Quality Index (PSQI; a 7-category assessment tool on which 0=perfect quality and 21=poor quality). The effect size was large for both ISI and PSQI, as indicated by standard mean differences greater than 0.8 (1.22 and 0.88, respectively) and was sustained for as long as 18 months.

RECOMMENDATIONS

The American College of Physicians strongly recommends that all adult patients receive CBT as initial treatment for chronic insomnia disorder. It can be performed in multiple settings, including the primary care setting. Compared with hypnotics, CBT is unlikely to have any adverse effects.⁶

EVIDENCE SUMMARY

Three meta-analyses that included only randomized controlled trials (RCTs) compared various CBT delivery methods with controls (wait-listed for treatment or general sleep hygiene education) to assess sleep outcomes for adults with insomnia.^1-3 TABLE 1^1-3 summarizes the results.

CBT is comparable to pharmacotherapy

A 2002 comparative meta-analysis of 21 RCTs with a total of 470 patients examined the effectiveness of CBT (stimulus control and/or sleep restriction) compared with pharmacotherapy (benzodiazepines or benzodiazepine agonists) for treating primary insomnia of longer than one month’s duration in adults with no comorbid medical or psychiatric diagnoses.⁴ The CBT group received intervention over an average of 5 weeks, and the pharmacotherapy group received intervention over an average of 2 weeks.

CBT produced greater reductions in sleep onset latency than pharmacotherapy based on mean weighted effect size (1.05 vs 0.45 weighted effect size; 95% confidence interval, 0.17-1.04; P=.01). Although both CBT and pharmacotherapy improved sleep outcomes, no statistical differences were found in wake after sleep onset time, total sleep time, number of awakenings, or sleep quality ratings (TABLE 2⁴).

Continue to: CBT has significant benefit for comorbid insomnia

A 2015 meta-analysis of 23 studies enrolling a total of 1379 adults with a number of illnesses (chronic pain, alcohol dependence, breast cancer, psychiatric disorders, chronic obstructive pulmonary disease, fibromyalgia) and comorbid insomnia investigated the qualitative effectiveness of individual or group CBT therapy.⁵ Subjects received at least 4 face-to-face sessions and at least 2 components of CBT.

The primary outcome showed that sleep quality improved, as measured by a 6.36-point reduction in the Insomnia Severity Index (ISI; a 7-question scale on which 0=no insomnia and 28=severe insomnia) and a 3.3-point reduction in the Pittsburgh Sleep Quality Index (PSQI; a 7-category assessment tool on which 0=perfect quality and 21=poor quality). The effect size was large for both ISI and PSQI, as indicated by standard mean differences greater than 0.8 (1.22 and 0.88, respectively) and was sustained for as long as 18 months.

RECOMMENDATIONS

The American College of Physicians strongly recommends that all adult patients receive CBT as initial treatment for chronic insomnia disorder. It can be performed in multiple settings, including the primary care setting. Compared with hypnotics, CBT is unlikely to have any adverse effects.⁶

A 44-year-old African-American woman with sickle cell trait presented to the clinic to establish care. She complained of polyarthralgias that she’d had since adolescence; the pain was worst in her right shoulder. She reported morning stiffness that lasted up to 8 hours, an intermittent facial rash, oral ulcers, joint edema (of which she had pictures on her phone), and photosensitivity. She took ibuprofen and acetaminophen as needed for pain. She once worked as a medical assistant but hadn’t been able to work since 2014 due to pain. She reported having been told as a teenager that she might have juvenile arthritis, but she didn’t recall ever having diagnostic tests performed or receiving treatment other than anti-inflammatories.

A couple of weeks after an initial visit with a rheumatologist, the patient returned to the family medicine clinic. She said she was upset that the specialist had x-rayed her hands, but had not checked her shoulder, which was the primary source of her pain. She also had pain in her hands, hips, feet, and knees.

On physical exam, the patient looked fatigued. A musculoskeletal exam revealed no joint effusions or edema, but was significant for right shoulder pain with reduced abduction to 90°. Gross motor strength was 5/5 in all 4 extremities. Laboratory testing revealed an antinuclear antibody titer of 1:160 and was negative for double-stranded DNA. Bilateral hand and foot x-rays showed no joint erosions. An x-ray of the right shoulder was obtained, which showed evidence of osteopenia and an erosion in the humeral head (FIGURES 1A and 1B).

WHAT IS YOUR DIAGNOSIS?
HOW WOULD YOU TREAT THIS PATIENT?

The patient’s history of morning stiffness and the joint erosion observed on x-ray were highly suggestive of rheumatoid arthritis (RA).

Joint stiffness due to osteoarthritis is worse with activity and generally lasts only a few minutes, while joint stiffness due to rheumatoid arthritis is worse at rest and lasts 30 minutes or more.

RA is a symmetric, inflammatory, peripheral polyarthritis of unknown etiology. It is the most common form of inflammatory arthritis, affecting 1% of the population worldwide.¹ It causes cartilage and bone to erode, leading to the deformation and destruction of joints. If RA is left untreated or is unresponsive to therapy, it can eventually lead to loss of physical function.

Making the diagnosis

The distinctive signs of RA are joint erosions and rheumatoid nodules, which are often absent on initial presentation.

Classification criteria. The 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) classification criteria for RA² are based on the presence of synovitis in at least one joint, the absence of an alternative diagnosis that better explains the synovitis, and a cumulative score of at least 6/10 from the following 4 domains:

1. Number and site of involved joints
   • 2 to 10 large joints (shoulders, elbows, hips, knees, ankles)=1 point
   • 1 to 3 small joints (metacarpophalangeal [MCP] joints, proximal interphalangeal [PIP] joints, 2nd-5th metatarsophalangeal joints, thumb interphalangeal joints, wrists)=2 points
   • 4 to 10 small joints=3 points
   • More than 10 joints (including at least 1 small joint)=5 points
2. Serologic abnormality (rheumatoid factor [RF] and anti-cyclic citrullinated peptide)
   • Low positive=2 points
   • High positive=3 points
3. Elevated acute phase response (erythrocyte sedimentation rate [ESR] or C-reactive protein [CRP])=1 point
4. Symptom duration of at least 6 weeks=1 point.

These criteria are best suited for early disease. For patients with longstanding symptoms, diagnosis is based on an erosive disease with a history of criteria fulfillment, or a currently inactive longstanding disease, with or without treatment, that has previously fulfilled the criteria.³

Continue to: The differential Dx is extensive

The differential includes polyarthralgias such as viral polyarthritis, systemic rheumatic diseases, and osteoarthritis.

Viral polyarthritis is caused by rubella, parvovirus B19⁴, alphaviruses, and hepatitis B. Symptoms can last from 3 days to several weeks, but rarely persist beyond 6 weeks; alphaviruses, however, can last 3 to 6 months.⁵ The common symptom triad includes fever, arthritis, and rash. Chikungunya is an example of an alphavirus that has become a global disease. Alphavirus arthritis can mimic seronegative RA and even satisfy the classification criteria for RA if the initial symptoms of fever and rash and history of travel to endemic regions are not appreciated.⁵

Systemic rheumatic diseases. Early RA may mimic the arthritis of systemic lupus erythematosus (SLE), Sjögren’s syndrome, dermatomyositis, or mixed connective tissue disease.⁶ In contrast to RA, these disorders generally have systemic features, such as rashes, dry mouth and eyes, myositis, or nephritis, and generate autobodies, which are not seen with RA. The CRP is often normal in patients with active SLE, even when the ESR is elevated.

Immediate treatment with disease-modifying antirheumatic drugs is important to achieve disease control and prevent disability.

Osteoarthritis (OA) can be confused with RA, particularly when small joints are involved. A thorough history helps elucidate the diagnosis. For example, OA of the fingers affects distal interphalangeal joints and is associated with Heberden’s nodes, while RA more commonly affects MCP and PIP joints. Swelling from OA is typically firm, while swelling due to RA is warm, boggy, and tender. Joint stiffness due to OA is worse with activity and generally lasts only a few minutes, while joint stiffness due to RA is worse at rest and lasts 30 minutes or more. X-rays show joint-space narrowing with OA, but no erosions or cysts. RF may be present at low levels in older patients with OA, while it is usually associated with high levels in patients with seropositive RA.

Continue to: Treat with disease-modifying antirheumatic drugs

Immediate treatment of RA with disease-modifying antirheumatic drugs (DMARDs) is important to achieve control of the disease and prevent disability.

Our patient. We ordered a purified protein derivative skin test in preparation for the patient to start therapy with a DMARD. Our patient followed up with Rheumatology and was started on indomethacin, with an initial dose of 25 mg bid. (DMARDs are first-line therapy; indomethacin is a second-line choice. In this case, the patient declined DMARDs after hearing they lowered the body’s ability to fight infection.) An RF level measured 6.74 IU/ml, which is within normal limits. The patient was subsequently lost to follow-up.

CORRESPONDENCE
Barbara Kiersz, DO, 6835 Austin Center Blvd., Austin, TX 78731; [email protected].

A 44-year-old African-American woman with sickle cell trait presented to the clinic to establish care. She complained of polyarthralgias that she’d had since adolescence; the pain was worst in her right shoulder. She reported morning stiffness that lasted up to 8 hours, an intermittent facial rash, oral ulcers, joint edema (of which she had pictures on her phone), and photosensitivity. She took ibuprofen and acetaminophen as needed for pain. She once worked as a medical assistant but hadn’t been able to work since 2014 due to pain. She reported having been told as a teenager that she might have juvenile arthritis, but she didn’t recall ever having diagnostic tests performed or receiving treatment other than anti-inflammatories.

A couple of weeks after an initial visit with a rheumatologist, the patient returned to the family medicine clinic. She said she was upset that the specialist had x-rayed her hands, but had not checked her shoulder, which was the primary source of her pain. She also had pain in her hands, hips, feet, and knees.

On physical exam, the patient looked fatigued. A musculoskeletal exam revealed no joint effusions or edema, but was significant for right shoulder pain with reduced abduction to 90°. Gross motor strength was 5/5 in all 4 extremities. Laboratory testing revealed an antinuclear antibody titer of 1:160 and was negative for double-stranded DNA. Bilateral hand and foot x-rays showed no joint erosions. An x-ray of the right shoulder was obtained, which showed evidence of osteopenia and an erosion in the humeral head (FIGURES 1A and 1B).

WHAT IS YOUR DIAGNOSIS?
HOW WOULD YOU TREAT THIS PATIENT?

The patient’s history of morning stiffness and the joint erosion observed on x-ray were highly suggestive of rheumatoid arthritis (RA).

Joint stiffness due to osteoarthritis is worse with activity and generally lasts only a few minutes, while joint stiffness due to rheumatoid arthritis is worse at rest and lasts 30 minutes or more.

RA is a symmetric, inflammatory, peripheral polyarthritis of unknown etiology. It is the most common form of inflammatory arthritis, affecting 1% of the population worldwide.¹ It causes cartilage and bone to erode, leading to the deformation and destruction of joints. If RA is left untreated or is unresponsive to therapy, it can eventually lead to loss of physical function.

Making the diagnosis

The distinctive signs of RA are joint erosions and rheumatoid nodules, which are often absent on initial presentation.

Classification criteria. The 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) classification criteria for RA² are based on the presence of synovitis in at least one joint, the absence of an alternative diagnosis that better explains the synovitis, and a cumulative score of at least 6/10 from the following 4 domains:

1. Number and site of involved joints
   • 2 to 10 large joints (shoulders, elbows, hips, knees, ankles)=1 point
   • 1 to 3 small joints (metacarpophalangeal [MCP] joints, proximal interphalangeal [PIP] joints, 2nd-5th metatarsophalangeal joints, thumb interphalangeal joints, wrists)=2 points
   • 4 to 10 small joints=3 points
   • More than 10 joints (including at least 1 small joint)=5 points
2. Serologic abnormality (rheumatoid factor [RF] and anti-cyclic citrullinated peptide)
   • Low positive=2 points
   • High positive=3 points
3. Elevated acute phase response (erythrocyte sedimentation rate [ESR] or C-reactive protein [CRP])=1 point
4. Symptom duration of at least 6 weeks=1 point.

These criteria are best suited for early disease. For patients with longstanding symptoms, diagnosis is based on an erosive disease with a history of criteria fulfillment, or a currently inactive longstanding disease, with or without treatment, that has previously fulfilled the criteria.³

Continue to: The differential Dx is extensive

The differential includes polyarthralgias such as viral polyarthritis, systemic rheumatic diseases, and osteoarthritis.

Viral polyarthritis is caused by rubella, parvovirus B19⁴, alphaviruses, and hepatitis B. Symptoms can last from 3 days to several weeks, but rarely persist beyond 6 weeks; alphaviruses, however, can last 3 to 6 months.⁵ The common symptom triad includes fever, arthritis, and rash. Chikungunya is an example of an alphavirus that has become a global disease. Alphavirus arthritis can mimic seronegative RA and even satisfy the classification criteria for RA if the initial symptoms of fever and rash and history of travel to endemic regions are not appreciated.⁵

Systemic rheumatic diseases. Early RA may mimic the arthritis of systemic lupus erythematosus (SLE), Sjögren’s syndrome, dermatomyositis, or mixed connective tissue disease.⁶ In contrast to RA, these disorders generally have systemic features, such as rashes, dry mouth and eyes, myositis, or nephritis, and generate autobodies, which are not seen with RA. The CRP is often normal in patients with active SLE, even when the ESR is elevated.

Immediate treatment with disease-modifying antirheumatic drugs is important to achieve disease control and prevent disability.

Osteoarthritis (OA) can be confused with RA, particularly when small joints are involved. A thorough history helps elucidate the diagnosis. For example, OA of the fingers affects distal interphalangeal joints and is associated with Heberden’s nodes, while RA more commonly affects MCP and PIP joints. Swelling from OA is typically firm, while swelling due to RA is warm, boggy, and tender. Joint stiffness due to OA is worse with activity and generally lasts only a few minutes, while joint stiffness due to RA is worse at rest and lasts 30 minutes or more. X-rays show joint-space narrowing with OA, but no erosions or cysts. RF may be present at low levels in older patients with OA, while it is usually associated with high levels in patients with seropositive RA.

Continue to: Treat with disease-modifying antirheumatic drugs

Immediate treatment of RA with disease-modifying antirheumatic drugs (DMARDs) is important to achieve control of the disease and prevent disability.

Our patient. We ordered a purified protein derivative skin test in preparation for the patient to start therapy with a DMARD. Our patient followed up with Rheumatology and was started on indomethacin, with an initial dose of 25 mg bid. (DMARDs are first-line therapy; indomethacin is a second-line choice. In this case, the patient declined DMARDs after hearing they lowered the body’s ability to fight infection.) An RF level measured 6.74 IU/ml, which is within normal limits. The patient was subsequently lost to follow-up.

CORRESPONDENCE
Barbara Kiersz, DO, 6835 Austin Center Blvd., Austin, TX 78731; [email protected].

A 44-year-old African-American woman with sickle cell trait presented to the clinic to establish care. She complained of polyarthralgias that she’d had since adolescence; the pain was worst in her right shoulder. She reported morning stiffness that lasted up to 8 hours, an intermittent facial rash, oral ulcers, joint edema (of which she had pictures on her phone), and photosensitivity. She took ibuprofen and acetaminophen as needed for pain. She once worked as a medical assistant but hadn’t been able to work since 2014 due to pain. She reported having been told as a teenager that she might have juvenile arthritis, but she didn’t recall ever having diagnostic tests performed or receiving treatment other than anti-inflammatories.

A couple of weeks after an initial visit with a rheumatologist, the patient returned to the family medicine clinic. She said she was upset that the specialist had x-rayed her hands, but had not checked her shoulder, which was the primary source of her pain. She also had pain in her hands, hips, feet, and knees.

On physical exam, the patient looked fatigued. A musculoskeletal exam revealed no joint effusions or edema, but was significant for right shoulder pain with reduced abduction to 90°. Gross motor strength was 5/5 in all 4 extremities. Laboratory testing revealed an antinuclear antibody titer of 1:160 and was negative for double-stranded DNA. Bilateral hand and foot x-rays showed no joint erosions. An x-ray of the right shoulder was obtained, which showed evidence of osteopenia and an erosion in the humeral head (FIGURES 1A and 1B).

WHAT IS YOUR DIAGNOSIS?
HOW WOULD YOU TREAT THIS PATIENT?

The patient’s history of morning stiffness and the joint erosion observed on x-ray were highly suggestive of rheumatoid arthritis (RA).

Joint stiffness due to osteoarthritis is worse with activity and generally lasts only a few minutes, while joint stiffness due to rheumatoid arthritis is worse at rest and lasts 30 minutes or more.

RA is a symmetric, inflammatory, peripheral polyarthritis of unknown etiology. It is the most common form of inflammatory arthritis, affecting 1% of the population worldwide.¹ It causes cartilage and bone to erode, leading to the deformation and destruction of joints. If RA is left untreated or is unresponsive to therapy, it can eventually lead to loss of physical function.

Making the diagnosis

The distinctive signs of RA are joint erosions and rheumatoid nodules, which are often absent on initial presentation.

Classification criteria. The 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) classification criteria for RA² are based on the presence of synovitis in at least one joint, the absence of an alternative diagnosis that better explains the synovitis, and a cumulative score of at least 6/10 from the following 4 domains:

1. Number and site of involved joints
   • 2 to 10 large joints (shoulders, elbows, hips, knees, ankles)=1 point
   • 1 to 3 small joints (metacarpophalangeal [MCP] joints, proximal interphalangeal [PIP] joints, 2nd-5th metatarsophalangeal joints, thumb interphalangeal joints, wrists)=2 points
   • 4 to 10 small joints=3 points
   • More than 10 joints (including at least 1 small joint)=5 points
2. Serologic abnormality (rheumatoid factor [RF] and anti-cyclic citrullinated peptide)
   • Low positive=2 points
   • High positive=3 points
3. Elevated acute phase response (erythrocyte sedimentation rate [ESR] or C-reactive protein [CRP])=1 point
4. Symptom duration of at least 6 weeks=1 point.

These criteria are best suited for early disease. For patients with longstanding symptoms, diagnosis is based on an erosive disease with a history of criteria fulfillment, or a currently inactive longstanding disease, with or without treatment, that has previously fulfilled the criteria.³

Continue to: The differential Dx is extensive

The differential includes polyarthralgias such as viral polyarthritis, systemic rheumatic diseases, and osteoarthritis.

Viral polyarthritis is caused by rubella, parvovirus B19⁴, alphaviruses, and hepatitis B. Symptoms can last from 3 days to several weeks, but rarely persist beyond 6 weeks; alphaviruses, however, can last 3 to 6 months.⁵ The common symptom triad includes fever, arthritis, and rash. Chikungunya is an example of an alphavirus that has become a global disease. Alphavirus arthritis can mimic seronegative RA and even satisfy the classification criteria for RA if the initial symptoms of fever and rash and history of travel to endemic regions are not appreciated.⁵

Systemic rheumatic diseases. Early RA may mimic the arthritis of systemic lupus erythematosus (SLE), Sjögren’s syndrome, dermatomyositis, or mixed connective tissue disease.⁶ In contrast to RA, these disorders generally have systemic features, such as rashes, dry mouth and eyes, myositis, or nephritis, and generate autobodies, which are not seen with RA. The CRP is often normal in patients with active SLE, even when the ESR is elevated.

Immediate treatment with disease-modifying antirheumatic drugs is important to achieve disease control and prevent disability.

Osteoarthritis (OA) can be confused with RA, particularly when small joints are involved. A thorough history helps elucidate the diagnosis. For example, OA of the fingers affects distal interphalangeal joints and is associated with Heberden’s nodes, while RA more commonly affects MCP and PIP joints. Swelling from OA is typically firm, while swelling due to RA is warm, boggy, and tender. Joint stiffness due to OA is worse with activity and generally lasts only a few minutes, while joint stiffness due to RA is worse at rest and lasts 30 minutes or more. X-rays show joint-space narrowing with OA, but no erosions or cysts. RF may be present at low levels in older patients with OA, while it is usually associated with high levels in patients with seropositive RA.

Continue to: Treat with disease-modifying antirheumatic drugs

Immediate treatment of RA with disease-modifying antirheumatic drugs (DMARDs) is important to achieve control of the disease and prevent disability.

Our patient. We ordered a purified protein derivative skin test in preparation for the patient to start therapy with a DMARD. Our patient followed up with Rheumatology and was started on indomethacin, with an initial dose of 25 mg bid. (DMARDs are first-line therapy; indomethacin is a second-line choice. In this case, the patient declined DMARDs after hearing they lowered the body’s ability to fight infection.) An RF level measured 6.74 IU/ml, which is within normal limits. The patient was subsequently lost to follow-up.

CORRESPONDENCE
Barbara Kiersz, DO, 6835 Austin Center Blvd., Austin, TX 78731; [email protected].

Thrombocytopenia and neutropenia are commonly encountered laboratory abnormalities. The presence of either requires that you promptly evaluate for life-threatening causes and identify the appropriate etiology. This article identifies key questions to ask. It also includes algorithms and tables that will facilitate your evaluation of patients with isolated thrombocytopenia or isolated neutropenia and speed the way toward appropriate treatment.

Thrombocytopenia: A look at the numbers

Thrombocytopenia is defined as a platelet count <150,000/mcL.¹ The blood abnormality is either suspected based on the patient’s signs or symptoms, such as ecchymoses, petechiae, purpura, epistaxis, gingival bleeding, or melena, or it is incidentally discovered during review of a complete blood count (CBC).

The development of clinical symptoms is closely related to the severity of the thrombocytopenia, with platelet counts <30,000/mcL more likely to result in clinical symptoms with minor trauma and counts <5,000/mcL potentially resulting in spontaneous bleeding. While most patients will have asymptomatic, incidentally-found thrombocytopenia, and likely a benign etiology, those with the signs/symptoms just described, evidence of infection, or thrombosis are more likely to have a serious etiology and require an expedited work-up. Although pregnancy may be associated with thrombocytopenia, this review confines itself to the causes of thrombocytopenia in non-pregnant adults.

Rule out pseudothrombocytopenia

When isolated thrombocytopenia is discovered incidentally in an asymptomatic person, the first step is to perform a repeat CBC with a peripheral smear to confirm the presence of thrombocytopenia, rule out laboratory error, and assess for platelet clumping. If thrombocytopenia is confirmed and platelet clumping is present, it may be due to the calcium chelator in the ethylenediaminetetraacetic anticoagulant contained within the laboratory transport tube; this cause of pseudothrombocytopenia occurs in up to 0.29% of the population.¹ Obtaining a platelet count from a citrated or heparinized tube avoids this phenomenon.

Is the patient’s thrombocytopenia drug induced?

Once true thrombocytopenia is confirmed, the next step is to review the patient’s prescribed medications, as well as any illicit drugs used, for potential causes of drug-induced thrombocytopenia. DITP can be either immune-mediated or nonimmune-mediated.

Immune-mediated drug-induced thrombocytopenia typically occurs within 1 to 2 weeks of medication exposure and begins to improve within 1 to 2 days of stopping the offending drug.

Immune-mediated DITP typically occurs within 1 to 2 weeks of medication exposure and begins to improve within 1 to 2 days of stopping the offending drug.² (See TABLE 1³ for a list of medications that can induce thrombocytopenia.) It should be noted that most patients who take the medications listed in TABLE 1 do not experience thrombocytopenia; nonetheless, it is a potential risk associated with their use.

Heparin-induced thrombocytopenia (HIT) is a unique form of immune-mediated DITP in that it is caused by antibody complexes, resulting in platelet activation, clumping, and thrombotic events.⁴ HIT occurs <1% of patients in intensive care units, but can occur in any patient on long-term heparin therapy. It manifests as a >50% drop in platelet count within 5 to 14 days of the introduction of heparin; however, in those previously exposed to heparin, it can occur within 24 hours.^4,5

Continue to: Non-immune-mediated DITP

Non-immune-mediated DITP, resulting from myelosuppression, chemotherapeutic agents, or valproic acid, is less common.^1,2

Acute and chronic alcohol use. Although alcohol is not a drug per se, it can also result in thrombocytopenia. The mechanism is the direct suppression of bone marrow, although alcohol also causes B12 and folate deficiency, further contributing to the development of the blood abnormality.¹

Is there thrombosis?

In addition to exploring a connection between thrombocytopenia and the drugs a patient is taking, it’s also important to look for evidence of thrombosis. The causes of thrombocytopenia that paradoxically result in thrombosis are: disseminated intravascular coagulation, hemolytic uremic syndrome, thrombotic thrombocytopenic purpura, catastrophic antiphospholipid antibody syndrome, and the previously mentioned HIT. TABLE 2^4,6-9 outlines the clinical settings, laboratory findings, and treatments of thrombocytopenia associated with thrombosis.

Is an infectious cause to blame?

If the patient is ill, consider infectious causes of thrombocytopenia. Thrombocytopenia associated with infection may result from an immune-mediated response to an illness itself, to treatment of an illness, to splenic sequestration, or to bone marrow suppression. TABLE 3^1,9-11 lists common infections that may cause thrombocytopenia.

Of note, infection with Helicobacter pylori can cause asymptomatic thrombocytopenia via an immune-mediated mechanism.¹² Eradication of H pylori results in a variable elevation in platelets, on average 30,000/mcL in 50% of patients with the infection.¹³

Is there pancytopenia?

A review of the peripheral smear, with attention to abnormalities in other cell lines, may assist in arriving at a diagnosis. If the peripheral smear reveals pancytopenia, then, in addition to many of the etiologies described earlier, one should also consider vitamin B12 or folate deficiency, copper deficiency, drug- and viral-induced aplastic anemia, paroxysmal nocturnal hemoglobinuria, leukemias, myelodysplastic disorders, and systemic lupus erythematosis.¹⁴ Pancytopenia is also seen with hypersplenism, which is often associated with cirrhosis.¹⁵ If the etiology isn’t readily apparent, a bone marrow biopsy may be required.

Continue to: Is immune thrombocytopenia to blame?

Immune thrombocytopenia (ITP) is an autoimmune disorder resulting in the destruction of normal platelets and may be primary or secondary to processes described previously (HIT, H pylori infection, etc). Consider ITP if, after a thorough work-up, a cause of isolated thrombocytopenia is not identified.¹⁶ Treatment for ITP is outlined in TABLE 4.¹⁶ FIGURE 1 is an algorithm for the complete evaluation of thrombocytopenia in adults.

Treatment: Platelet transfusions

In general, patients who are not actively bleeding are considered stable and do not require platelet transfusions to minimize their risk of bleeding or prevent bleeding during a planned procedure unless their platelet count falls below the levels specified in TABLE 5.¹⁷ For patients who are actively bleeding, a more aggressive approach may be required. Locally-derived transfusion protocols typically guide transfusions for the actively hemorrhaging patient. The American Association of Blood Banks has put forth evidence-based guidelines for platelet transfusions when a patient is given a diagnosis of thrombocytopenia (see TABLE 5).¹⁷ Single-donor platelets have a shelf life of 3 to 5 days, and one unit will raise platelets 30,000 to 50,000/mcL.

Neutropenia: Prevalence varies by ethnicity

An absolute neutrophil count (ANC) of <1500 cells/mcL traditionally defines neutropenia, with an ANC of 1000 to 1500 cells/mcL constituting mild neutropenia; 500 to 999 cells/mcL, moderate; and <500 cells/mcL, severe.¹⁸ Similar to the evaluation of thrombocytopenia, it is important to repeat the CBC prior to initiating a work-up in order to confirm that the neutropenia is not a laboratory error. Additionally, patients with signs or symptoms of infection should be worked up expeditiously.

Heparin-induced thrombocytopenia occurs in <1% of patients in intensive care units and typically is manifested by a ≥50% drop in platelet count within 5 to 14 days of introducing heparin.

The prevalence of neutropenia varies by ethnicity. According to the National Health and Nutrition Examination Survey 1999 to 2004, the prevalence was 4.5%, 0.79%, and 0.38% in black, white, and Mexican-American participants, respectively.¹⁹ FIGURE 2 outlines the outpatient work-up of adult patients with neutropenia not related to chemotherapy.

Continue to: Is the patient severely ill?

The prognosis of the patient is related both to the etiology of the neutropenia, as well as to the nadir of the neutrophil count. Patients who have an ANC <500 cells/mcL or who have inadequate bone marrow reserves are at highest risk for an overwhelming infection.^20,21 The absence of oral ulcers and gingivitis and/or the presence of purulent material at the site of an infection are signs of adequate bone marrow reserves.

Additionally, neutropenia may be the source—or the result—of a serious life-threatening illness. This distinction may not be readily apparent at the time of the patient’s presentation. If signs or symptoms of a severe illness are apparent (fever, hypotension, tachycardia, ANC <500 cells/mcL), admit the patient to the hospital for evaluation and initiation of antibiotics. 

Is the neutropenia chronic?

A review of previous CBCs will identify whether this condition is new or chronic. A persistent, mild neutropenia (ANC 1000-1500 cells/mcL) in a healthy individual is consistent with benign familial or ethnic neutropenia (see TABLE 6).²⁰ If prior CBCs are unavailable, then a diagnosis of chronic neutropenia may be established by verifying the persistence of mild neutropenia over time.

Cyclic neutropenia is a periodic neutropenia (occurring every 2-5 weeks) associated with mild illnesses that are related to the nadir of the neutrophil count. The diagnosis is established by obtaining serial CBCs twice weekly for 4 to 6 weeks, which reflect cycling of the neutrophil count.^20,22

Are any medications contributing to the neutropenia?

Medications that suppress bone marrow or that interfere with other immune-mediated processes are the most common cause of acquired neutropenia.²³ Drug-induced agranulocytosis is defined as an ANC <500 cells/mcL due to exposure to a drug that results in immunologic or cytotoxic destruction of neutrophils.²⁴

A persistent, mild neutropenia in a healthy individual is consistent with benign familial or ethnic neutropenia.

A systematic review of case reports of drug-induced agranulocytosis (a decrease in peripheral neutrophil count to <500 cells/mcL) revealed that although at least 125 drugs were probably related to agranulocytosis, only 11 drugs were responsible for 50% of cases (carbimazole, clozapine, dapsone, dipyrone, methimazole, penicillin G, procainamide, propylthiouracil, rituximab, sulfasalazine, and ticlopidine), and fatality rates were higher (10% vs 3%) among those patients with a nadir <100 cells/mcL.²⁵ TABLE 7²⁵ lists medications that can be associated with agranulocytosis. Depending on prior exposure to a drug, neutropenia/agranulocytosis can occur within hours to months of exposure to the causal drug and can take a few days to 3 weeks to resolve after cessation.^25,26

Continue to: Has the patient had any recent illnesses?

The usual response to an infection is an increase in neutrophil count. However, certain bacterial, rickettsial, parasitic, and viral infections can result in neutropenia (see TABLE 8^23,27-29). Viral infections may cause transient neutropenia because of either bone marrow suppression or increased peripheral destruction, while neutropenia related to an overwhelming bacterial infection results from the depletion of bone marrow reserves.^23,27

Do you suspect a nutritional deficiency?

Patients with a nutritional deficiency of B12, folate, or copper are likely to exhibit a deficiency in more than just neutrophils.^23,27 In developed countries, people with neutropenia may have a history of malnutrition due to a disease (eg, anorexia nervosa) or surgery (eg, gastric bypass) that causes severe calorie restriction.²⁰

Does your patient have symptoms of a connective tissue disease?

In developed countries, people with neutropenia may have a history of malnutrition due to a disease (eg, anorexia nervosa) or surgery (eg, gastric bypass) that causes severe calorie restriction.

Neutropenia, in association with arthralgias, joint swelling, splenomegaly, or rash may be a manifestation of an underlying collagen vascular disorder, such as rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE).²⁰ If the clinical scenario supports one of these diagnoses, undertake or refer the patient for a rheumatologic evaluation. This may include studies of anti-cyclic citrullinated peptide antibodies, rheumatoid factor to evaluate for RA, and/or antinuclear antibodies to evaluate for SLE.^30,31 While most neutropenias associated with autoimmune disease are mild, neutropenia associated with Felty syndrome (RA, splenomegaly, and neutropenia) may be severe (ANC <100 cells/mcL).^20,23

Is the etiology unclear?

Patients with moderate to severe neutropenia without an apparent etiology, in the setting of aplastic anemia, or in the presence of splenomegaly and/or lymphadenopathy, should undergo a hematologic evaluation and/or bone marrow biopsy, given that hematologic malignancy is a potential cause.^20,27

The treatment of neutropenia hinges on correctly identifying the etiology of the diminished neutrophil count. If the cause is a medication, infection, underlying rheumatologic condition, or nutritional deficiency, then either treating the entity or withdrawing the offending medication should result in resolution of the neutropenia. If the cause is determined to be familial or ethnic, then patient reassurance is all that is required.

CORRESPONDENCE
Richard W. Temple, MD, FAAFP, CDR MC USN, Camp Lejeune Family Medicine Residency, Naval Medical Center Camp Lejeune, 100 Brewster Blvd, Camp Lejeune, NC 28547-2538; [email protected].

Thrombocytopenia and neutropenia are commonly encountered laboratory abnormalities. The presence of either requires that you promptly evaluate for life-threatening causes and identify the appropriate etiology. This article identifies key questions to ask. It also includes algorithms and tables that will facilitate your evaluation of patients with isolated thrombocytopenia or isolated neutropenia and speed the way toward appropriate treatment.

Thrombocytopenia: A look at the numbers

Thrombocytopenia is defined as a platelet count <150,000/mcL.¹ The blood abnormality is either suspected based on the patient’s signs or symptoms, such as ecchymoses, petechiae, purpura, epistaxis, gingival bleeding, or melena, or it is incidentally discovered during review of a complete blood count (CBC).

The development of clinical symptoms is closely related to the severity of the thrombocytopenia, with platelet counts <30,000/mcL more likely to result in clinical symptoms with minor trauma and counts <5,000/mcL potentially resulting in spontaneous bleeding. While most patients will have asymptomatic, incidentally-found thrombocytopenia, and likely a benign etiology, those with the signs/symptoms just described, evidence of infection, or thrombosis are more likely to have a serious etiology and require an expedited work-up. Although pregnancy may be associated with thrombocytopenia, this review confines itself to the causes of thrombocytopenia in non-pregnant adults.

Rule out pseudothrombocytopenia

When isolated thrombocytopenia is discovered incidentally in an asymptomatic person, the first step is to perform a repeat CBC with a peripheral smear to confirm the presence of thrombocytopenia, rule out laboratory error, and assess for platelet clumping. If thrombocytopenia is confirmed and platelet clumping is present, it may be due to the calcium chelator in the ethylenediaminetetraacetic anticoagulant contained within the laboratory transport tube; this cause of pseudothrombocytopenia occurs in up to 0.29% of the population.¹ Obtaining a platelet count from a citrated or heparinized tube avoids this phenomenon.

Is the patient’s thrombocytopenia drug induced?

Once true thrombocytopenia is confirmed, the next step is to review the patient’s prescribed medications, as well as any illicit drugs used, for potential causes of drug-induced thrombocytopenia. DITP can be either immune-mediated or nonimmune-mediated.

Immune-mediated drug-induced thrombocytopenia typically occurs within 1 to 2 weeks of medication exposure and begins to improve within 1 to 2 days of stopping the offending drug.

Immune-mediated DITP typically occurs within 1 to 2 weeks of medication exposure and begins to improve within 1 to 2 days of stopping the offending drug.² (See TABLE 1³ for a list of medications that can induce thrombocytopenia.) It should be noted that most patients who take the medications listed in TABLE 1 do not experience thrombocytopenia; nonetheless, it is a potential risk associated with their use.

Heparin-induced thrombocytopenia (HIT) is a unique form of immune-mediated DITP in that it is caused by antibody complexes, resulting in platelet activation, clumping, and thrombotic events.⁴ HIT occurs <1% of patients in intensive care units, but can occur in any patient on long-term heparin therapy. It manifests as a >50% drop in platelet count within 5 to 14 days of the introduction of heparin; however, in those previously exposed to heparin, it can occur within 24 hours.^4,5

Continue to: Non-immune-mediated DITP

Non-immune-mediated DITP, resulting from myelosuppression, chemotherapeutic agents, or valproic acid, is less common.^1,2

Acute and chronic alcohol use. Although alcohol is not a drug per se, it can also result in thrombocytopenia. The mechanism is the direct suppression of bone marrow, although alcohol also causes B12 and folate deficiency, further contributing to the development of the blood abnormality.¹

Is there thrombosis?

In addition to exploring a connection between thrombocytopenia and the drugs a patient is taking, it’s also important to look for evidence of thrombosis. The causes of thrombocytopenia that paradoxically result in thrombosis are: disseminated intravascular coagulation, hemolytic uremic syndrome, thrombotic thrombocytopenic purpura, catastrophic antiphospholipid antibody syndrome, and the previously mentioned HIT. TABLE 2^4,6-9 outlines the clinical settings, laboratory findings, and treatments of thrombocytopenia associated with thrombosis.

Is an infectious cause to blame?

If the patient is ill, consider infectious causes of thrombocytopenia. Thrombocytopenia associated with infection may result from an immune-mediated response to an illness itself, to treatment of an illness, to splenic sequestration, or to bone marrow suppression. TABLE 3^1,9-11 lists common infections that may cause thrombocytopenia.

Of note, infection with Helicobacter pylori can cause asymptomatic thrombocytopenia via an immune-mediated mechanism.¹² Eradication of H pylori results in a variable elevation in platelets, on average 30,000/mcL in 50% of patients with the infection.¹³

Is there pancytopenia?

A review of the peripheral smear, with attention to abnormalities in other cell lines, may assist in arriving at a diagnosis. If the peripheral smear reveals pancytopenia, then, in addition to many of the etiologies described earlier, one should also consider vitamin B12 or folate deficiency, copper deficiency, drug- and viral-induced aplastic anemia, paroxysmal nocturnal hemoglobinuria, leukemias, myelodysplastic disorders, and systemic lupus erythematosis.¹⁴ Pancytopenia is also seen with hypersplenism, which is often associated with cirrhosis.¹⁵ If the etiology isn’t readily apparent, a bone marrow biopsy may be required.

Continue to: Is immune thrombocytopenia to blame?

Immune thrombocytopenia (ITP) is an autoimmune disorder resulting in the destruction of normal platelets and may be primary or secondary to processes described previously (HIT, H pylori infection, etc). Consider ITP if, after a thorough work-up, a cause of isolated thrombocytopenia is not identified.¹⁶ Treatment for ITP is outlined in TABLE 4.¹⁶ FIGURE 1 is an algorithm for the complete evaluation of thrombocytopenia in adults.

Treatment: Platelet transfusions

In general, patients who are not actively bleeding are considered stable and do not require platelet transfusions to minimize their risk of bleeding or prevent bleeding during a planned procedure unless their platelet count falls below the levels specified in TABLE 5.¹⁷ For patients who are actively bleeding, a more aggressive approach may be required. Locally-derived transfusion protocols typically guide transfusions for the actively hemorrhaging patient. The American Association of Blood Banks has put forth evidence-based guidelines for platelet transfusions when a patient is given a diagnosis of thrombocytopenia (see TABLE 5).¹⁷ Single-donor platelets have a shelf life of 3 to 5 days, and one unit will raise platelets 30,000 to 50,000/mcL.

Neutropenia: Prevalence varies by ethnicity

An absolute neutrophil count (ANC) of <1500 cells/mcL traditionally defines neutropenia, with an ANC of 1000 to 1500 cells/mcL constituting mild neutropenia; 500 to 999 cells/mcL, moderate; and <500 cells/mcL, severe.¹⁸ Similar to the evaluation of thrombocytopenia, it is important to repeat the CBC prior to initiating a work-up in order to confirm that the neutropenia is not a laboratory error. Additionally, patients with signs or symptoms of infection should be worked up expeditiously.

Heparin-induced thrombocytopenia occurs in <1% of patients in intensive care units and typically is manifested by a ≥50% drop in platelet count within 5 to 14 days of introducing heparin.

The prevalence of neutropenia varies by ethnicity. According to the National Health and Nutrition Examination Survey 1999 to 2004, the prevalence was 4.5%, 0.79%, and 0.38% in black, white, and Mexican-American participants, respectively.¹⁹ FIGURE 2 outlines the outpatient work-up of adult patients with neutropenia not related to chemotherapy.

Continue to: Is the patient severely ill?

The prognosis of the patient is related both to the etiology of the neutropenia, as well as to the nadir of the neutrophil count. Patients who have an ANC <500 cells/mcL or who have inadequate bone marrow reserves are at highest risk for an overwhelming infection.^20,21 The absence of oral ulcers and gingivitis and/or the presence of purulent material at the site of an infection are signs of adequate bone marrow reserves.

Additionally, neutropenia may be the source—or the result—of a serious life-threatening illness. This distinction may not be readily apparent at the time of the patient’s presentation. If signs or symptoms of a severe illness are apparent (fever, hypotension, tachycardia, ANC <500 cells/mcL), admit the patient to the hospital for evaluation and initiation of antibiotics. 

Is the neutropenia chronic?

A review of previous CBCs will identify whether this condition is new or chronic. A persistent, mild neutropenia (ANC 1000-1500 cells/mcL) in a healthy individual is consistent with benign familial or ethnic neutropenia (see TABLE 6).²⁰ If prior CBCs are unavailable, then a diagnosis of chronic neutropenia may be established by verifying the persistence of mild neutropenia over time.

Cyclic neutropenia is a periodic neutropenia (occurring every 2-5 weeks) associated with mild illnesses that are related to the nadir of the neutrophil count. The diagnosis is established by obtaining serial CBCs twice weekly for 4 to 6 weeks, which reflect cycling of the neutrophil count.^20,22

Are any medications contributing to the neutropenia?

Medications that suppress bone marrow or that interfere with other immune-mediated processes are the most common cause of acquired neutropenia.²³ Drug-induced agranulocytosis is defined as an ANC <500 cells/mcL due to exposure to a drug that results in immunologic or cytotoxic destruction of neutrophils.²⁴

A persistent, mild neutropenia in a healthy individual is consistent with benign familial or ethnic neutropenia.

A systematic review of case reports of drug-induced agranulocytosis (a decrease in peripheral neutrophil count to <500 cells/mcL) revealed that although at least 125 drugs were probably related to agranulocytosis, only 11 drugs were responsible for 50% of cases (carbimazole, clozapine, dapsone, dipyrone, methimazole, penicillin G, procainamide, propylthiouracil, rituximab, sulfasalazine, and ticlopidine), and fatality rates were higher (10% vs 3%) among those patients with a nadir <100 cells/mcL.²⁵ TABLE 7²⁵ lists medications that can be associated with agranulocytosis. Depending on prior exposure to a drug, neutropenia/agranulocytosis can occur within hours to months of exposure to the causal drug and can take a few days to 3 weeks to resolve after cessation.^25,26

Continue to: Has the patient had any recent illnesses?

The usual response to an infection is an increase in neutrophil count. However, certain bacterial, rickettsial, parasitic, and viral infections can result in neutropenia (see TABLE 8^23,27-29). Viral infections may cause transient neutropenia because of either bone marrow suppression or increased peripheral destruction, while neutropenia related to an overwhelming bacterial infection results from the depletion of bone marrow reserves.^23,27

Do you suspect a nutritional deficiency?

Patients with a nutritional deficiency of B12, folate, or copper are likely to exhibit a deficiency in more than just neutrophils.^23,27 In developed countries, people with neutropenia may have a history of malnutrition due to a disease (eg, anorexia nervosa) or surgery (eg, gastric bypass) that causes severe calorie restriction.²⁰

Does your patient have symptoms of a connective tissue disease?

In developed countries, people with neutropenia may have a history of malnutrition due to a disease (eg, anorexia nervosa) or surgery (eg, gastric bypass) that causes severe calorie restriction.

Neutropenia, in association with arthralgias, joint swelling, splenomegaly, or rash may be a manifestation of an underlying collagen vascular disorder, such as rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE).²⁰ If the clinical scenario supports one of these diagnoses, undertake or refer the patient for a rheumatologic evaluation. This may include studies of anti-cyclic citrullinated peptide antibodies, rheumatoid factor to evaluate for RA, and/or antinuclear antibodies to evaluate for SLE.^30,31 While most neutropenias associated with autoimmune disease are mild, neutropenia associated with Felty syndrome (RA, splenomegaly, and neutropenia) may be severe (ANC <100 cells/mcL).^20,23

Is the etiology unclear?

Patients with moderate to severe neutropenia without an apparent etiology, in the setting of aplastic anemia, or in the presence of splenomegaly and/or lymphadenopathy, should undergo a hematologic evaluation and/or bone marrow biopsy, given that hematologic malignancy is a potential cause.^20,27

The treatment of neutropenia hinges on correctly identifying the etiology of the diminished neutrophil count. If the cause is a medication, infection, underlying rheumatologic condition, or nutritional deficiency, then either treating the entity or withdrawing the offending medication should result in resolution of the neutropenia. If the cause is determined to be familial or ethnic, then patient reassurance is all that is required.

CORRESPONDENCE
Richard W. Temple, MD, FAAFP, CDR MC USN, Camp Lejeune Family Medicine Residency, Naval Medical Center Camp Lejeune, 100 Brewster Blvd, Camp Lejeune, NC 28547-2538; [email protected].

Thrombocytopenia and neutropenia are commonly encountered laboratory abnormalities. The presence of either requires that you promptly evaluate for life-threatening causes and identify the appropriate etiology. This article identifies key questions to ask. It also includes algorithms and tables that will facilitate your evaluation of patients with isolated thrombocytopenia or isolated neutropenia and speed the way toward appropriate treatment.

Thrombocytopenia: A look at the numbers

Thrombocytopenia is defined as a platelet count <150,000/mcL.¹ The blood abnormality is either suspected based on the patient’s signs or symptoms, such as ecchymoses, petechiae, purpura, epistaxis, gingival bleeding, or melena, or it is incidentally discovered during review of a complete blood count (CBC).

The development of clinical symptoms is closely related to the severity of the thrombocytopenia, with platelet counts <30,000/mcL more likely to result in clinical symptoms with minor trauma and counts <5,000/mcL potentially resulting in spontaneous bleeding. While most patients will have asymptomatic, incidentally-found thrombocytopenia, and likely a benign etiology, those with the signs/symptoms just described, evidence of infection, or thrombosis are more likely to have a serious etiology and require an expedited work-up. Although pregnancy may be associated with thrombocytopenia, this review confines itself to the causes of thrombocytopenia in non-pregnant adults.

Rule out pseudothrombocytopenia

When isolated thrombocytopenia is discovered incidentally in an asymptomatic person, the first step is to perform a repeat CBC with a peripheral smear to confirm the presence of thrombocytopenia, rule out laboratory error, and assess for platelet clumping. If thrombocytopenia is confirmed and platelet clumping is present, it may be due to the calcium chelator in the ethylenediaminetetraacetic anticoagulant contained within the laboratory transport tube; this cause of pseudothrombocytopenia occurs in up to 0.29% of the population.¹ Obtaining a platelet count from a citrated or heparinized tube avoids this phenomenon.

Is the patient’s thrombocytopenia drug induced?

Once true thrombocytopenia is confirmed, the next step is to review the patient’s prescribed medications, as well as any illicit drugs used, for potential causes of drug-induced thrombocytopenia. DITP can be either immune-mediated or nonimmune-mediated.

Immune-mediated drug-induced thrombocytopenia typically occurs within 1 to 2 weeks of medication exposure and begins to improve within 1 to 2 days of stopping the offending drug.

Immune-mediated DITP typically occurs within 1 to 2 weeks of medication exposure and begins to improve within 1 to 2 days of stopping the offending drug.² (See TABLE 1³ for a list of medications that can induce thrombocytopenia.) It should be noted that most patients who take the medications listed in TABLE 1 do not experience thrombocytopenia; nonetheless, it is a potential risk associated with their use.

Heparin-induced thrombocytopenia (HIT) is a unique form of immune-mediated DITP in that it is caused by antibody complexes, resulting in platelet activation, clumping, and thrombotic events.⁴ HIT occurs <1% of patients in intensive care units, but can occur in any patient on long-term heparin therapy. It manifests as a >50% drop in platelet count within 5 to 14 days of the introduction of heparin; however, in those previously exposed to heparin, it can occur within 24 hours.^4,5

Continue to: Non-immune-mediated DITP

Non-immune-mediated DITP, resulting from myelosuppression, chemotherapeutic agents, or valproic acid, is less common.^1,2

Acute and chronic alcohol use. Although alcohol is not a drug per se, it can also result in thrombocytopenia. The mechanism is the direct suppression of bone marrow, although alcohol also causes B12 and folate deficiency, further contributing to the development of the blood abnormality.¹

Is there thrombosis?

In addition to exploring a connection between thrombocytopenia and the drugs a patient is taking, it’s also important to look for evidence of thrombosis. The causes of thrombocytopenia that paradoxically result in thrombosis are: disseminated intravascular coagulation, hemolytic uremic syndrome, thrombotic thrombocytopenic purpura, catastrophic antiphospholipid antibody syndrome, and the previously mentioned HIT. TABLE 2^4,6-9 outlines the clinical settings, laboratory findings, and treatments of thrombocytopenia associated with thrombosis.

Is an infectious cause to blame?

If the patient is ill, consider infectious causes of thrombocytopenia. Thrombocytopenia associated with infection may result from an immune-mediated response to an illness itself, to treatment of an illness, to splenic sequestration, or to bone marrow suppression. TABLE 3^1,9-11 lists common infections that may cause thrombocytopenia.

Of note, infection with Helicobacter pylori can cause asymptomatic thrombocytopenia via an immune-mediated mechanism.¹² Eradication of H pylori results in a variable elevation in platelets, on average 30,000/mcL in 50% of patients with the infection.¹³

Is there pancytopenia?

A review of the peripheral smear, with attention to abnormalities in other cell lines, may assist in arriving at a diagnosis. If the peripheral smear reveals pancytopenia, then, in addition to many of the etiologies described earlier, one should also consider vitamin B12 or folate deficiency, copper deficiency, drug- and viral-induced aplastic anemia, paroxysmal nocturnal hemoglobinuria, leukemias, myelodysplastic disorders, and systemic lupus erythematosis.¹⁴ Pancytopenia is also seen with hypersplenism, which is often associated with cirrhosis.¹⁵ If the etiology isn’t readily apparent, a bone marrow biopsy may be required.

Continue to: Is immune thrombocytopenia to blame?

Immune thrombocytopenia (ITP) is an autoimmune disorder resulting in the destruction of normal platelets and may be primary or secondary to processes described previously (HIT, H pylori infection, etc). Consider ITP if, after a thorough work-up, a cause of isolated thrombocytopenia is not identified.¹⁶ Treatment for ITP is outlined in TABLE 4.¹⁶ FIGURE 1 is an algorithm for the complete evaluation of thrombocytopenia in adults.

Treatment: Platelet transfusions

In general, patients who are not actively bleeding are considered stable and do not require platelet transfusions to minimize their risk of bleeding or prevent bleeding during a planned procedure unless their platelet count falls below the levels specified in TABLE 5.¹⁷ For patients who are actively bleeding, a more aggressive approach may be required. Locally-derived transfusion protocols typically guide transfusions for the actively hemorrhaging patient. The American Association of Blood Banks has put forth evidence-based guidelines for platelet transfusions when a patient is given a diagnosis of thrombocytopenia (see TABLE 5).¹⁷ Single-donor platelets have a shelf life of 3 to 5 days, and one unit will raise platelets 30,000 to 50,000/mcL.

Neutropenia: Prevalence varies by ethnicity

An absolute neutrophil count (ANC) of <1500 cells/mcL traditionally defines neutropenia, with an ANC of 1000 to 1500 cells/mcL constituting mild neutropenia; 500 to 999 cells/mcL, moderate; and <500 cells/mcL, severe.¹⁸ Similar to the evaluation of thrombocytopenia, it is important to repeat the CBC prior to initiating a work-up in order to confirm that the neutropenia is not a laboratory error. Additionally, patients with signs or symptoms of infection should be worked up expeditiously.

Heparin-induced thrombocytopenia occurs in <1% of patients in intensive care units and typically is manifested by a ≥50% drop in platelet count within 5 to 14 days of introducing heparin.

The prevalence of neutropenia varies by ethnicity. According to the National Health and Nutrition Examination Survey 1999 to 2004, the prevalence was 4.5%, 0.79%, and 0.38% in black, white, and Mexican-American participants, respectively.¹⁹ FIGURE 2 outlines the outpatient work-up of adult patients with neutropenia not related to chemotherapy.

Continue to: Is the patient severely ill?

The prognosis of the patient is related both to the etiology of the neutropenia, as well as to the nadir of the neutrophil count. Patients who have an ANC <500 cells/mcL or who have inadequate bone marrow reserves are at highest risk for an overwhelming infection.^20,21 The absence of oral ulcers and gingivitis and/or the presence of purulent material at the site of an infection are signs of adequate bone marrow reserves.

Additionally, neutropenia may be the source—or the result—of a serious life-threatening illness. This distinction may not be readily apparent at the time of the patient’s presentation. If signs or symptoms of a severe illness are apparent (fever, hypotension, tachycardia, ANC <500 cells/mcL), admit the patient to the hospital for evaluation and initiation of antibiotics. 

Is the neutropenia chronic?

A review of previous CBCs will identify whether this condition is new or chronic. A persistent, mild neutropenia (ANC 1000-1500 cells/mcL) in a healthy individual is consistent with benign familial or ethnic neutropenia (see TABLE 6).²⁰ If prior CBCs are unavailable, then a diagnosis of chronic neutropenia may be established by verifying the persistence of mild neutropenia over time.

Cyclic neutropenia is a periodic neutropenia (occurring every 2-5 weeks) associated with mild illnesses that are related to the nadir of the neutrophil count. The diagnosis is established by obtaining serial CBCs twice weekly for 4 to 6 weeks, which reflect cycling of the neutrophil count.^20,22

Are any medications contributing to the neutropenia?

Medications that suppress bone marrow or that interfere with other immune-mediated processes are the most common cause of acquired neutropenia.²³ Drug-induced agranulocytosis is defined as an ANC <500 cells/mcL due to exposure to a drug that results in immunologic or cytotoxic destruction of neutrophils.²⁴

A persistent, mild neutropenia in a healthy individual is consistent with benign familial or ethnic neutropenia.

A systematic review of case reports of drug-induced agranulocytosis (a decrease in peripheral neutrophil count to <500 cells/mcL) revealed that although at least 125 drugs were probably related to agranulocytosis, only 11 drugs were responsible for 50% of cases (carbimazole, clozapine, dapsone, dipyrone, methimazole, penicillin G, procainamide, propylthiouracil, rituximab, sulfasalazine, and ticlopidine), and fatality rates were higher (10% vs 3%) among those patients with a nadir <100 cells/mcL.²⁵ TABLE 7²⁵ lists medications that can be associated with agranulocytosis. Depending on prior exposure to a drug, neutropenia/agranulocytosis can occur within hours to months of exposure to the causal drug and can take a few days to 3 weeks to resolve after cessation.^25,26

Continue to: Has the patient had any recent illnesses?

The usual response to an infection is an increase in neutrophil count. However, certain bacterial, rickettsial, parasitic, and viral infections can result in neutropenia (see TABLE 8^23,27-29). Viral infections may cause transient neutropenia because of either bone marrow suppression or increased peripheral destruction, while neutropenia related to an overwhelming bacterial infection results from the depletion of bone marrow reserves.^23,27

Do you suspect a nutritional deficiency?

Patients with a nutritional deficiency of B12, folate, or copper are likely to exhibit a deficiency in more than just neutrophils.^23,27 In developed countries, people with neutropenia may have a history of malnutrition due to a disease (eg, anorexia nervosa) or surgery (eg, gastric bypass) that causes severe calorie restriction.²⁰

Does your patient have symptoms of a connective tissue disease?

In developed countries, people with neutropenia may have a history of malnutrition due to a disease (eg, anorexia nervosa) or surgery (eg, gastric bypass) that causes severe calorie restriction.

Neutropenia, in association with arthralgias, joint swelling, splenomegaly, or rash may be a manifestation of an underlying collagen vascular disorder, such as rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE).²⁰ If the clinical scenario supports one of these diagnoses, undertake or refer the patient for a rheumatologic evaluation. This may include studies of anti-cyclic citrullinated peptide antibodies, rheumatoid factor to evaluate for RA, and/or antinuclear antibodies to evaluate for SLE.^30,31 While most neutropenias associated with autoimmune disease are mild, neutropenia associated with Felty syndrome (RA, splenomegaly, and neutropenia) may be severe (ANC <100 cells/mcL).^20,23

Is the etiology unclear?

Patients with moderate to severe neutropenia without an apparent etiology, in the setting of aplastic anemia, or in the presence of splenomegaly and/or lymphadenopathy, should undergo a hematologic evaluation and/or bone marrow biopsy, given that hematologic malignancy is a potential cause.^20,27

The treatment of neutropenia hinges on correctly identifying the etiology of the diminished neutrophil count. If the cause is a medication, infection, underlying rheumatologic condition, or nutritional deficiency, then either treating the entity or withdrawing the offending medication should result in resolution of the neutropenia. If the cause is determined to be familial or ethnic, then patient reassurance is all that is required.

CORRESPONDENCE
Richard W. Temple, MD, FAAFP, CDR MC USN, Camp Lejeune Family Medicine Residency, Naval Medical Center Camp Lejeune, 100 Brewster Blvd, Camp Lejeune, NC 28547-2538; [email protected].

CASE

A 44-year-old nurse describes persistent fatigue and itching over the last 2 months. She is taking ramipril 5 mg/d for hypertension and has a family history of rheumatic disease. Lab tests reveal a recurrent moderate elevation of gamma glutamyl-transpeptidase (gGT; 75 U/L) associated with, on some occasions, mild elevation of alanine aminotransferase (ALT) levels (100 U/L) of unknown origin. She has no history of hepatitis virus infection, hepatotoxic medications, or alcohol intake. She is overweight with a body mass index of 28.5 kg/m² and a waist circumference of 99 cm (39 inches). Liver ultrasonography detects an enlarged liver with diffuse echostructure dishomogeneity, but no signs of cirrhosis or portal hypertension. The patient’s biliary tree is not dilated.

How would you proceed with the care of this patient?

Cholestasis is characterized by the alteration of bile flow through any part of the biliary system, from the hepatocyte basocellular membrane to the duodenum. The condition is classified as intrahepatic when the cause is a defect of hepatocellular function or obstruction of the biliary tree within the liver. The extrahepatic form includes all conditions obstructing bile flow in the main biliary tract (choledochus, common bile duct).

Suspect intrahepatic cholestasis in a patient with chronic itching, normal transaminases, and mildly elevated gamma glutamyl-transpeptidase.

The key to successfully managing cholestasis lies in the early identification of subtle signs and symptoms before serious complications can arise. In the review that follows, we provide guidance for evaluating laboratory and imaging results that are vital to the accurate diagnosis of intrahepatic and extrahepatic cholestasis. We also detail treatment recommendations.

Clues—subtle and otherwise—of cholestasis

Clinical features of cholestasis include fatigue and itching all over the skin. The latter likely is caused by induction of the enzyme autotaxin, which produces the neuronal activator lysophosphatidic acid. Retention of pruritogenic substances that normally are excreted into bile might contribute to pruritus as well.¹ Jaundice, dark urine, and pale and fatty stools occur with advanced disease. However, a cholestatic condition can be detected in asymptomatic patients with elevated biochemical markers.

Continue to: Mildly elevated gGT and/or alkaline phosphatase (ALP)

Mildly elevated gGT and/or alkaline phosphatase (ALP) (0.5-2.5 times the upper normal limit [UNL] or 19-95 U/L and 60-300 U/L, respectively²) in the presence of normal transaminase levels (<20 U/L) in an asymptomatic patient can indicate chronic liver disease. Signs suggestive of significant liver disease have been reported in many patients with gGT or ALP elevation with good sensitivity (65%) and specificity (83%) for a diagnosis of intrahepatic cholestasis.³ However, because abnormal gGT values are common and often resolve spontaneously, family physicians (FPs) may pay little attention to this finding, thus missing an opportunity for early identification and treatment.

Patient’s history can provide important clues

A thorough patient history is especially important when cholestasis is suspected. Details about the patient’s occupation, environment, and lifestyle are key, as are the specifics of prescribed or over-the-counter medications and supplements that could be hepatotoxic (TABLE 4¹⁰). A number of exogenous substances can cause liver injury, and the use of some herbal products (senna, black cohosh, greater celandine, kava) have been linked to hepatitis and cholestasis.¹¹ Ask patients about alcohol use and history of conditions associated with liver disease, such as diabetes, hyperlipidemia, and thyroid disorders.

Continue to: Indicators pointing to cholestasis? It's time for ultrasonography

Abdominal ultrasonography is a first-line diagnostic tool for cholestasis.

While biopsy is considered the gold standard for diagnosing and staging chronic cholestatic liver disease and can exclude an extrahepatic obstruction, it should be employed only if blood tests have been confirmed, second-level tests have been performed, and ultrasound is inconclusive.¹² (More on biopsy in a bit.)

Ultrasonography is a low-cost, widely available, noninvasive test that allows easy identification of extrahepatic dilatation of the biliary tree and sometimes the underlying cause, as well. Ultrasonography identifies extrahepatic cholestasis by allowing visualization of an enlarged choledochus (>7 mm) or common hepatic duct (>5 mm) and an intrahepatic bile duct diameter that is more than 40% larger than adjacent branches of the portal vein.¹³ However, ultrasonography has a low diagnostic sensitivity for many conditions (eg, 15% to 89% for detecting common bile duct stones),¹⁴ requiring other diagnostic procedures, such as endoscopic retrograde cholangiopancreatography (ERCP) or magnetic resonance cholangiopancreatography (MRCP), before reaching a diagnosis.

For asymptomatic patients with cirrhosis or those at an early stage of liver disease, ultrasound at 6-month intervals combined with serum liver function tests can be useful to track disease progression and screen for hepatocellular carcinoma or cholangiocarcinoma.^15,16

New noninvasive methods. Noninvasive tools for evaluating the presence and severity of liver fibrosis and for differentiating cirrhosis from noncirrhotic conditions have positive predictive values >85% to 90% for some chronic liver diseases.¹⁷ Transient elastography, which assesses liver stiffness, is one such method. Although it is often used successfully, morbid obesity, small intercostal spaces, and ascites limit its diagnostic capability.¹⁸ Recently, some questions about the validity of elastography to assess the extent of fibrosis in patients with chronic cholestatic conditions have been reported.^19,20

Suspect intrahepatic cholestasis? Your next steps

If imaging techniques do not show bile duct obstruction and you suspect the intrahepatic form, second-level tests could have strategic importance. This is where antimitochondrial antibodies (AMAs) come in. AMAs are immunoglobulins (IgG and IgM) directed against mitochondrial antigens. They are important markers for PBC, which is a T-lymphocyte-mediated attack on small intralobular bile ducts resulting in their gradual destruction and eventual disappearance. The sustained loss of intralobular bile ducts leads to signs and symptoms of cholestasis and eventually results in cirrhosis and liver failure.

AMA serum levels show high sensitivity and specificity (90% and 95%, respectively) for PBC.²¹ Some PBC patients (<5%) show histologic confirmation of the disease, but have negative AMA tests (AMA negative PBC or autoimmune cholangitis).²² Therefore, according to the American Association for the Study of Liver Diseases, diagnosis of PBC is guided by the combination of serologic, biochemical, and histologic criteria.²³ Many PBC patients with or without a positive AMA (≥1:40) also have positive circulating antinuclear antibodies (ANA; ≥1:80). The recent availability of lab tests for antibodies (anti-M2, anti-gp120, anti-sp100) has allowed identification of subgroups of patients who have a more aggressive form of PBC. Patients with PBC often have elevated levels of circulating IgM (>280 mg/dL).

Continue to: Other circulating antibodies

Other circulating antibodies can help discriminate among cholestatic disorders. In particular, positive tests for perinuclear anti-neutrophil cytoplasmic antibodies (pANCA) are found in 25% to 95% of patients with primary sclerosing cholangitis (PSC), a chronic progressive disorder of unknown etiology that is characterized by inflammation, fibrosis, and stricturing of medium and large ducts of the intrahepatic and extrahepatic biliary tree.²⁴ Anti-smooth muscle antibodies (SMA) can be observed in both PSC and autoimmune hepatitis.

Alkaline phosphatase levels are useful for monitoring evolution of primary biliary cholangitis disease.

Finally, there are syndromes with serologic and histologic overlap that are characterized by the simultaneous presence of PBC with autoimmune hepatitis or PSC or overlap of PSC with autoimmune hepatitis.

Liver biopsy fills in the rest of the diagnostic picture

Unfortunately, blood tests reveal little about organ integrity and are not useful for disease staging. The decision to perform a liver biopsy should be based on several factors, including the patient’s age, serum parameters, the need to stage the disease, therapy choices, and prognosis.¹² One should also consider that biopsy is a costly procedure with potentially serious adverse effects; it should not be repeated frequently. However, when a biopsy is done, it provides critical information, including damage to medium-sized intrahepatic bile ducts with neoductular formation or bile duct scars and strictures.

Treating intrahepatic cholestasis

Although FPs often can provide most—or even all—of the care for patients with stable conditions, a specialist consultation might recommend further testing to identify the underlying disease, which is essential to establish the most appropriate treatment.

Treatment of patients with PBC is based on administering hydrophilic secondary bile salt ursodeoxycholic acid (UDCA) 15 mg/kg/d, which is used to equilibrate the ratio between hydrophilic and hydrophobic bile salts in the liver and bile,²⁵ and is the only treatment approved by the US Food and Drug Administration (FDA) for PBC.⁴ Tauroursodeoxycholate is better absorbed than UDCA, and, although partially deconjugated and reconjugated with glycine, it undergoes reduced biotransformation to more hydrophobic metabolites and has benefits, including antioxidant, immunomodulation, and neuroprotective effects over UDCA—especially for long-term therapy in PBC.²⁶ However, it is not used often in clinical practice.

Continue to: Bile acid administration counters the cytotoxic effect...

Bile acid administration counters the cytotoxic effect of hydrophobic bile salts. Although it seems that UDCA might improve biochemical and histologic features of the disease at earlier stages (I-II), it fails in patients with more advanced disease.²⁷ In addition, monitoring and defining response to UDCA is inconsistent, partly because of variations in guideline criteria.^28,29

Despite progress in diagnostic techniques, life expectancy and quality of life for patients with advanced cholestatic conditions remain poor.

Recently a new molecule, obeticholic acid (OCA), has been approved by the FDA. A farnesoid X receptor agonist, OCA is indicated for treating patients who do not tolerate UDCA or as an adjunct to UDCA in those with a partial response to UDCA, defined as lowering ALP levels by <1.5 times the baseline value after 12 months of treatment.

Treating PSC is more complex. Combination therapy with prednisone and azathioprine is recommended only when there is an overlap syndrome between PSC and autoimmune hepatitis.⁴ UDCA at a high dosage (15-20 mg/kg/d) is used to facilitate long-lasting biochemical remission. These patients also need to be monitored for inflammatory bowel diseases, which affect up to 75% of patients,³⁰ and for cholangiocarcinoma, which is a life-limiting complication because of a lack of therapy options. Finally, these patients might need endoscopic-guided dilatation of the biliary tree when they have evidence of dominant fibrotic strictures of the greater bile ducts.^14,31

Addressing the systemic effects of intrahepatic cholestasis

Pruritus. A number of potential pruritogens, including bile salts, endogenous opioids, histamine, serotonin, and lisophosphatidic acid (LPA), can be targeted to relieve pruritus.

• Bile acid resin binders such as cholestyramine are the first step for treating pruritus. UDCA also can be useful, mainly for intrahepatic cholestasis during pregnancy. Rifampicin, 300 mg/d, improves cholestatic pruritus, but is associated with hepatotoxicity and a number of severe reactions, such as nausea, loss of appetite, hemolytic anemia, and thrombocytopenia.³¹
• Most evidence favors a role for opioids in relieving itch, and micro-opioid receptor antagonists (naltrexone, naloxone, nalmefene) that exert an antipruritic effect can be effective.
• Sertraline (a selective serotonin reuptake inhibitor), 50 to 75 mg/d, usually is well tolerated in patients with chronic cholestasis and exerts a beneficial effect on pruritus in approximately 40% of patients.³²
• Extracorporeal albumin dialysis removes albumin-bound pruritogens and has been found to be effective in patients with liver failure. Steroids and UV light also can be used in select patients.
• The potent neuronal activator LPA and its converting enzyme autotaxin have been identified in the serum of patients with cholestatic pruritus; experimental modalities using LPA antagonists are ongoing for treating pruritus in patients who do not respond to other medications.³³

Continue to: Malnutrition

Malnutrition. Many patients with cholestasis are at risk for malnutrition, which can be exacerbated in those with cirrhosis. Causes of malnutrition include poor oral intake, malabsorption, or dental problems that prevent the patient from chewing. Assess the nutritional status of every patient with chronic cholestasis, and stress the importance of multivitamin supplementation to reverse systemic alterations caused by malnutrition.³⁴

When the patient has advanced disease

Despite progress in diagnostic techniques, life expectancy and quality of life for patients with advanced cholestatic conditions remain poor. Patients routinely experience fatigue, pruritus, and complications of cirrhosis including ascites, encephalopathy, and bleeding. Cholestasis also carries the risk of life-threatening complications, partly because of comorbidities such as osteoporosis and malabsorption.

ERCP is widely employed for diagnosing and treating pancreatobiliary diseases; however, its use has dropped over the last 10 years because of the risk of complications.

Liver transplantation can improve the life expectancy of patients with advanced disease, but because of long waiting lists, candidates for transplant often die before an organ becomes available. For many patients who are not in end-stage condition, targeted therapy is crucial to slow disease progression and is recommended along with hepatitis A and B vaccinations and nutritional counseling.³⁵

Extrahepatic cholestasis is suspected? How to proceed

Computer tomography (CT) is recommended for better identification of neoplastic causes of biliary obstruction and for staging purposes. MRCP is an excellent noninvasive imaging technique for evaluating biliary ducts.³⁶

MRCP has 92% to 93% sensitivity and 97% to 98% specificity for diagnosing biliary duct stones.³⁷ MRCP also is the first-choice modality for evaluating bile ducts in patients with suspected PSC. If performed in expert centers, the diagnostic accuracy reaches that of ERCP. A meta-analysis of studies from 2000 to 2006 has shown a sensitivity of 86% and specificity of 94% for diagnosing PSC.³⁸

Endoscopic ultrasonography, which uses an ultrasonographic probe, allows clinicians to evaluate the integrity of the biliary and pancreatic ducts and is effective for diagnosing and staging cancer of the ampulla of Vater (sensitivity 93% vs 7% for abdominal ultrasonography and 29% for CT), and identifying biliary stones and biliary tree strictures.

Continue to: ERCP

ERCP is widely employed for diagnosing and treating pancreatobiliary diseases; however, its use has dropped over the last 10 years because of the risk of complications. ERCP is nearly exclusively used as a therapeutic procedure for pancreatic sphincterotomy, biliary dilatations, and removing biliary stones. It also has a diagnostic role in dominant stenosis or suspected biliary malignancy using brushing cytology and sampling biopsies of the bile ducts.

Treating extrahepatic cholestasis

Treatment of the different underlying conditions that cause extrahepatic cholestasis is surgical. Thus, the potential surgical techniques that can resolve or improve an extrahepatic cholestatic condition are guided by the surgeon and beyond the scope of this article.

Treating osteopenia: A concern for intra- and extrahepatic cholestasis

Vitamin D deficiency as a consequence of reduced intestinal absorption (poor availability of bile salts) or decreased hepatic activation to 25,OH-cholecalcipherol in both intrahepatic and extrahepatic cholestasis can lead to reduced bone formation.³⁹ However, osteopenia can occur even in early stages of the disease. Prescribing bisphosphonates, in combination with calcium and vitamin D3_, to improve bone mineral density is a good practice.⁴⁰

CASE

Blood tests and ultrasound imaging suggest the presence of a chronic liver disease. Other lab tests indicate that the patient has an ALP level 3 times normal. This finding, together with the other tests, points to a likely diagnosis of intrahepatic cholestatic liver disease. Serology confirms positivity for ANA (1:160) and AMA (1:640). The clinician suspects PBC, so the patient is referred to a liver specialist for further evaluation and to determine whether a liver biopsy is needed.

The liver specialist confirms the diagnosis of PBC, performs a transient elastographym, which indicates a low-grade liver fibrosis (F1 out of 4), and starts therapy with UDCA.

CORRESPONDENCE
Ignazio Grattagliano, MD, Italian College of General Practitioners and Primary Care, Via del Sansovino 179, 50142, Florence, Italy; [email protected].

CASE

A 44-year-old nurse describes persistent fatigue and itching over the last 2 months. She is taking ramipril 5 mg/d for hypertension and has a family history of rheumatic disease. Lab tests reveal a recurrent moderate elevation of gamma glutamyl-transpeptidase (gGT; 75 U/L) associated with, on some occasions, mild elevation of alanine aminotransferase (ALT) levels (100 U/L) of unknown origin. She has no history of hepatitis virus infection, hepatotoxic medications, or alcohol intake. She is overweight with a body mass index of 28.5 kg/m² and a waist circumference of 99 cm (39 inches). Liver ultrasonography detects an enlarged liver with diffuse echostructure dishomogeneity, but no signs of cirrhosis or portal hypertension. The patient’s biliary tree is not dilated.

How would you proceed with the care of this patient?

Cholestasis is characterized by the alteration of bile flow through any part of the biliary system, from the hepatocyte basocellular membrane to the duodenum. The condition is classified as intrahepatic when the cause is a defect of hepatocellular function or obstruction of the biliary tree within the liver. The extrahepatic form includes all conditions obstructing bile flow in the main biliary tract (choledochus, common bile duct).

Suspect intrahepatic cholestasis in a patient with chronic itching, normal transaminases, and mildly elevated gamma glutamyl-transpeptidase.

The key to successfully managing cholestasis lies in the early identification of subtle signs and symptoms before serious complications can arise. In the review that follows, we provide guidance for evaluating laboratory and imaging results that are vital to the accurate diagnosis of intrahepatic and extrahepatic cholestasis. We also detail treatment recommendations.

Clues—subtle and otherwise—of cholestasis

Clinical features of cholestasis include fatigue and itching all over the skin. The latter likely is caused by induction of the enzyme autotaxin, which produces the neuronal activator lysophosphatidic acid. Retention of pruritogenic substances that normally are excreted into bile might contribute to pruritus as well.¹ Jaundice, dark urine, and pale and fatty stools occur with advanced disease. However, a cholestatic condition can be detected in asymptomatic patients with elevated biochemical markers.

Continue to: Mildly elevated gGT and/or alkaline phosphatase (ALP)

Mildly elevated gGT and/or alkaline phosphatase (ALP) (0.5-2.5 times the upper normal limit [UNL] or 19-95 U/L and 60-300 U/L, respectively²) in the presence of normal transaminase levels (<20 U/L) in an asymptomatic patient can indicate chronic liver disease. Signs suggestive of significant liver disease have been reported in many patients with gGT or ALP elevation with good sensitivity (65%) and specificity (83%) for a diagnosis of intrahepatic cholestasis.³ However, because abnormal gGT values are common and often resolve spontaneously, family physicians (FPs) may pay little attention to this finding, thus missing an opportunity for early identification and treatment.

Patient’s history can provide important clues

A thorough patient history is especially important when cholestasis is suspected. Details about the patient’s occupation, environment, and lifestyle are key, as are the specifics of prescribed or over-the-counter medications and supplements that could be hepatotoxic (TABLE 4¹⁰). A number of exogenous substances can cause liver injury, and the use of some herbal products (senna, black cohosh, greater celandine, kava) have been linked to hepatitis and cholestasis.¹¹ Ask patients about alcohol use and history of conditions associated with liver disease, such as diabetes, hyperlipidemia, and thyroid disorders.

Continue to: Indicators pointing to cholestasis? It's time for ultrasonography

Abdominal ultrasonography is a first-line diagnostic tool for cholestasis.

While biopsy is considered the gold standard for diagnosing and staging chronic cholestatic liver disease and can exclude an extrahepatic obstruction, it should be employed only if blood tests have been confirmed, second-level tests have been performed, and ultrasound is inconclusive.¹² (More on biopsy in a bit.)

Ultrasonography is a low-cost, widely available, noninvasive test that allows easy identification of extrahepatic dilatation of the biliary tree and sometimes the underlying cause, as well. Ultrasonography identifies extrahepatic cholestasis by allowing visualization of an enlarged choledochus (>7 mm) or common hepatic duct (>5 mm) and an intrahepatic bile duct diameter that is more than 40% larger than adjacent branches of the portal vein.¹³ However, ultrasonography has a low diagnostic sensitivity for many conditions (eg, 15% to 89% for detecting common bile duct stones),¹⁴ requiring other diagnostic procedures, such as endoscopic retrograde cholangiopancreatography (ERCP) or magnetic resonance cholangiopancreatography (MRCP), before reaching a diagnosis.

For asymptomatic patients with cirrhosis or those at an early stage of liver disease, ultrasound at 6-month intervals combined with serum liver function tests can be useful to track disease progression and screen for hepatocellular carcinoma or cholangiocarcinoma.^15,16

New noninvasive methods. Noninvasive tools for evaluating the presence and severity of liver fibrosis and for differentiating cirrhosis from noncirrhotic conditions have positive predictive values >85% to 90% for some chronic liver diseases.¹⁷ Transient elastography, which assesses liver stiffness, is one such method. Although it is often used successfully, morbid obesity, small intercostal spaces, and ascites limit its diagnostic capability.¹⁸ Recently, some questions about the validity of elastography to assess the extent of fibrosis in patients with chronic cholestatic conditions have been reported.^19,20

Suspect intrahepatic cholestasis? Your next steps

If imaging techniques do not show bile duct obstruction and you suspect the intrahepatic form, second-level tests could have strategic importance. This is where antimitochondrial antibodies (AMAs) come in. AMAs are immunoglobulins (IgG and IgM) directed against mitochondrial antigens. They are important markers for PBC, which is a T-lymphocyte-mediated attack on small intralobular bile ducts resulting in their gradual destruction and eventual disappearance. The sustained loss of intralobular bile ducts leads to signs and symptoms of cholestasis and eventually results in cirrhosis and liver failure.

AMA serum levels show high sensitivity and specificity (90% and 95%, respectively) for PBC.²¹ Some PBC patients (<5%) show histologic confirmation of the disease, but have negative AMA tests (AMA negative PBC or autoimmune cholangitis).²² Therefore, according to the American Association for the Study of Liver Diseases, diagnosis of PBC is guided by the combination of serologic, biochemical, and histologic criteria.²³ Many PBC patients with or without a positive AMA (≥1:40) also have positive circulating antinuclear antibodies (ANA; ≥1:80). The recent availability of lab tests for antibodies (anti-M2, anti-gp120, anti-sp100) has allowed identification of subgroups of patients who have a more aggressive form of PBC. Patients with PBC often have elevated levels of circulating IgM (>280 mg/dL).

Continue to: Other circulating antibodies

Other circulating antibodies can help discriminate among cholestatic disorders. In particular, positive tests for perinuclear anti-neutrophil cytoplasmic antibodies (pANCA) are found in 25% to 95% of patients with primary sclerosing cholangitis (PSC), a chronic progressive disorder of unknown etiology that is characterized by inflammation, fibrosis, and stricturing of medium and large ducts of the intrahepatic and extrahepatic biliary tree.²⁴ Anti-smooth muscle antibodies (SMA) can be observed in both PSC and autoimmune hepatitis.

Alkaline phosphatase levels are useful for monitoring evolution of primary biliary cholangitis disease.

Finally, there are syndromes with serologic and histologic overlap that are characterized by the simultaneous presence of PBC with autoimmune hepatitis or PSC or overlap of PSC with autoimmune hepatitis.

Liver biopsy fills in the rest of the diagnostic picture

Unfortunately, blood tests reveal little about organ integrity and are not useful for disease staging. The decision to perform a liver biopsy should be based on several factors, including the patient’s age, serum parameters, the need to stage the disease, therapy choices, and prognosis.¹² One should also consider that biopsy is a costly procedure with potentially serious adverse effects; it should not be repeated frequently. However, when a biopsy is done, it provides critical information, including damage to medium-sized intrahepatic bile ducts with neoductular formation or bile duct scars and strictures.

Treating intrahepatic cholestasis

Although FPs often can provide most—or even all—of the care for patients with stable conditions, a specialist consultation might recommend further testing to identify the underlying disease, which is essential to establish the most appropriate treatment.

Treatment of patients with PBC is based on administering hydrophilic secondary bile salt ursodeoxycholic acid (UDCA) 15 mg/kg/d, which is used to equilibrate the ratio between hydrophilic and hydrophobic bile salts in the liver and bile,²⁵ and is the only treatment approved by the US Food and Drug Administration (FDA) for PBC.⁴ Tauroursodeoxycholate is better absorbed than UDCA, and, although partially deconjugated and reconjugated with glycine, it undergoes reduced biotransformation to more hydrophobic metabolites and has benefits, including antioxidant, immunomodulation, and neuroprotective effects over UDCA—especially for long-term therapy in PBC.²⁶ However, it is not used often in clinical practice.

Continue to: Bile acid administration counters the cytotoxic effect...

Bile acid administration counters the cytotoxic effect of hydrophobic bile salts. Although it seems that UDCA might improve biochemical and histologic features of the disease at earlier stages (I-II), it fails in patients with more advanced disease.²⁷ In addition, monitoring and defining response to UDCA is inconsistent, partly because of variations in guideline criteria.^28,29

Despite progress in diagnostic techniques, life expectancy and quality of life for patients with advanced cholestatic conditions remain poor.

Recently a new molecule, obeticholic acid (OCA), has been approved by the FDA. A farnesoid X receptor agonist, OCA is indicated for treating patients who do not tolerate UDCA or as an adjunct to UDCA in those with a partial response to UDCA, defined as lowering ALP levels by <1.5 times the baseline value after 12 months of treatment.

Treating PSC is more complex. Combination therapy with prednisone and azathioprine is recommended only when there is an overlap syndrome between PSC and autoimmune hepatitis.⁴ UDCA at a high dosage (15-20 mg/kg/d) is used to facilitate long-lasting biochemical remission. These patients also need to be monitored for inflammatory bowel diseases, which affect up to 75% of patients,³⁰ and for cholangiocarcinoma, which is a life-limiting complication because of a lack of therapy options. Finally, these patients might need endoscopic-guided dilatation of the biliary tree when they have evidence of dominant fibrotic strictures of the greater bile ducts.^14,31

Addressing the systemic effects of intrahepatic cholestasis

Pruritus. A number of potential pruritogens, including bile salts, endogenous opioids, histamine, serotonin, and lisophosphatidic acid (LPA), can be targeted to relieve pruritus.

• Bile acid resin binders such as cholestyramine are the first step for treating pruritus. UDCA also can be useful, mainly for intrahepatic cholestasis during pregnancy. Rifampicin, 300 mg/d, improves cholestatic pruritus, but is associated with hepatotoxicity and a number of severe reactions, such as nausea, loss of appetite, hemolytic anemia, and thrombocytopenia.³¹
• Most evidence favors a role for opioids in relieving itch, and micro-opioid receptor antagonists (naltrexone, naloxone, nalmefene) that exert an antipruritic effect can be effective.
• Sertraline (a selective serotonin reuptake inhibitor), 50 to 75 mg/d, usually is well tolerated in patients with chronic cholestasis and exerts a beneficial effect on pruritus in approximately 40% of patients.³²
• Extracorporeal albumin dialysis removes albumin-bound pruritogens and has been found to be effective in patients with liver failure. Steroids and UV light also can be used in select patients.
• The potent neuronal activator LPA and its converting enzyme autotaxin have been identified in the serum of patients with cholestatic pruritus; experimental modalities using LPA antagonists are ongoing for treating pruritus in patients who do not respond to other medications.³³

Continue to: Malnutrition

Malnutrition. Many patients with cholestasis are at risk for malnutrition, which can be exacerbated in those with cirrhosis. Causes of malnutrition include poor oral intake, malabsorption, or dental problems that prevent the patient from chewing. Assess the nutritional status of every patient with chronic cholestasis, and stress the importance of multivitamin supplementation to reverse systemic alterations caused by malnutrition.³⁴

When the patient has advanced disease

Despite progress in diagnostic techniques, life expectancy and quality of life for patients with advanced cholestatic conditions remain poor. Patients routinely experience fatigue, pruritus, and complications of cirrhosis including ascites, encephalopathy, and bleeding. Cholestasis also carries the risk of life-threatening complications, partly because of comorbidities such as osteoporosis and malabsorption.

ERCP is widely employed for diagnosing and treating pancreatobiliary diseases; however, its use has dropped over the last 10 years because of the risk of complications.

Liver transplantation can improve the life expectancy of patients with advanced disease, but because of long waiting lists, candidates for transplant often die before an organ becomes available. For many patients who are not in end-stage condition, targeted therapy is crucial to slow disease progression and is recommended along with hepatitis A and B vaccinations and nutritional counseling.³⁵

Extrahepatic cholestasis is suspected? How to proceed

Computer tomography (CT) is recommended for better identification of neoplastic causes of biliary obstruction and for staging purposes. MRCP is an excellent noninvasive imaging technique for evaluating biliary ducts.³⁶

MRCP has 92% to 93% sensitivity and 97% to 98% specificity for diagnosing biliary duct stones.³⁷ MRCP also is the first-choice modality for evaluating bile ducts in patients with suspected PSC. If performed in expert centers, the diagnostic accuracy reaches that of ERCP. A meta-analysis of studies from 2000 to 2006 has shown a sensitivity of 86% and specificity of 94% for diagnosing PSC.³⁸

Endoscopic ultrasonography, which uses an ultrasonographic probe, allows clinicians to evaluate the integrity of the biliary and pancreatic ducts and is effective for diagnosing and staging cancer of the ampulla of Vater (sensitivity 93% vs 7% for abdominal ultrasonography and 29% for CT), and identifying biliary stones and biliary tree strictures.

Continue to: ERCP

ERCP is widely employed for diagnosing and treating pancreatobiliary diseases; however, its use has dropped over the last 10 years because of the risk of complications. ERCP is nearly exclusively used as a therapeutic procedure for pancreatic sphincterotomy, biliary dilatations, and removing biliary stones. It also has a diagnostic role in dominant stenosis or suspected biliary malignancy using brushing cytology and sampling biopsies of the bile ducts.

Treating extrahepatic cholestasis

Treatment of the different underlying conditions that cause extrahepatic cholestasis is surgical. Thus, the potential surgical techniques that can resolve or improve an extrahepatic cholestatic condition are guided by the surgeon and beyond the scope of this article.

Treating osteopenia: A concern for intra- and extrahepatic cholestasis

Vitamin D deficiency as a consequence of reduced intestinal absorption (poor availability of bile salts) or decreased hepatic activation to 25,OH-cholecalcipherol in both intrahepatic and extrahepatic cholestasis can lead to reduced bone formation.³⁹ However, osteopenia can occur even in early stages of the disease. Prescribing bisphosphonates, in combination with calcium and vitamin D3_, to improve bone mineral density is a good practice.⁴⁰

CASE

Blood tests and ultrasound imaging suggest the presence of a chronic liver disease. Other lab tests indicate that the patient has an ALP level 3 times normal. This finding, together with the other tests, points to a likely diagnosis of intrahepatic cholestatic liver disease. Serology confirms positivity for ANA (1:160) and AMA (1:640). The clinician suspects PBC, so the patient is referred to a liver specialist for further evaluation and to determine whether a liver biopsy is needed.

The liver specialist confirms the diagnosis of PBC, performs a transient elastographym, which indicates a low-grade liver fibrosis (F1 out of 4), and starts therapy with UDCA.

CORRESPONDENCE
Ignazio Grattagliano, MD, Italian College of General Practitioners and Primary Care, Via del Sansovino 179, 50142, Florence, Italy; [email protected].

CASE

A 44-year-old nurse describes persistent fatigue and itching over the last 2 months. She is taking ramipril 5 mg/d for hypertension and has a family history of rheumatic disease. Lab tests reveal a recurrent moderate elevation of gamma glutamyl-transpeptidase (gGT; 75 U/L) associated with, on some occasions, mild elevation of alanine aminotransferase (ALT) levels (100 U/L) of unknown origin. She has no history of hepatitis virus infection, hepatotoxic medications, or alcohol intake. She is overweight with a body mass index of 28.5 kg/m² and a waist circumference of 99 cm (39 inches). Liver ultrasonography detects an enlarged liver with diffuse echostructure dishomogeneity, but no signs of cirrhosis or portal hypertension. The patient’s biliary tree is not dilated.

How would you proceed with the care of this patient?

Cholestasis is characterized by the alteration of bile flow through any part of the biliary system, from the hepatocyte basocellular membrane to the duodenum. The condition is classified as intrahepatic when the cause is a defect of hepatocellular function or obstruction of the biliary tree within the liver. The extrahepatic form includes all conditions obstructing bile flow in the main biliary tract (choledochus, common bile duct).

Suspect intrahepatic cholestasis in a patient with chronic itching, normal transaminases, and mildly elevated gamma glutamyl-transpeptidase.

The key to successfully managing cholestasis lies in the early identification of subtle signs and symptoms before serious complications can arise. In the review that follows, we provide guidance for evaluating laboratory and imaging results that are vital to the accurate diagnosis of intrahepatic and extrahepatic cholestasis. We also detail treatment recommendations.

Clues—subtle and otherwise—of cholestasis

Clinical features of cholestasis include fatigue and itching all over the skin. The latter likely is caused by induction of the enzyme autotaxin, which produces the neuronal activator lysophosphatidic acid. Retention of pruritogenic substances that normally are excreted into bile might contribute to pruritus as well.¹ Jaundice, dark urine, and pale and fatty stools occur with advanced disease. However, a cholestatic condition can be detected in asymptomatic patients with elevated biochemical markers.

Continue to: Mildly elevated gGT and/or alkaline phosphatase (ALP)

Mildly elevated gGT and/or alkaline phosphatase (ALP) (0.5-2.5 times the upper normal limit [UNL] or 19-95 U/L and 60-300 U/L, respectively²) in the presence of normal transaminase levels (<20 U/L) in an asymptomatic patient can indicate chronic liver disease. Signs suggestive of significant liver disease have been reported in many patients with gGT or ALP elevation with good sensitivity (65%) and specificity (83%) for a diagnosis of intrahepatic cholestasis.³ However, because abnormal gGT values are common and often resolve spontaneously, family physicians (FPs) may pay little attention to this finding, thus missing an opportunity for early identification and treatment.

Patient’s history can provide important clues

A thorough patient history is especially important when cholestasis is suspected. Details about the patient’s occupation, environment, and lifestyle are key, as are the specifics of prescribed or over-the-counter medications and supplements that could be hepatotoxic (TABLE 4¹⁰). A number of exogenous substances can cause liver injury, and the use of some herbal products (senna, black cohosh, greater celandine, kava) have been linked to hepatitis and cholestasis.¹¹ Ask patients about alcohol use and history of conditions associated with liver disease, such as diabetes, hyperlipidemia, and thyroid disorders.

Continue to: Indicators pointing to cholestasis? It's time for ultrasonography

Abdominal ultrasonography is a first-line diagnostic tool for cholestasis.

While biopsy is considered the gold standard for diagnosing and staging chronic cholestatic liver disease and can exclude an extrahepatic obstruction, it should be employed only if blood tests have been confirmed, second-level tests have been performed, and ultrasound is inconclusive.¹² (More on biopsy in a bit.)

Ultrasonography is a low-cost, widely available, noninvasive test that allows easy identification of extrahepatic dilatation of the biliary tree and sometimes the underlying cause, as well. Ultrasonography identifies extrahepatic cholestasis by allowing visualization of an enlarged choledochus (>7 mm) or common hepatic duct (>5 mm) and an intrahepatic bile duct diameter that is more than 40% larger than adjacent branches of the portal vein.¹³ However, ultrasonography has a low diagnostic sensitivity for many conditions (eg, 15% to 89% for detecting common bile duct stones),¹⁴ requiring other diagnostic procedures, such as endoscopic retrograde cholangiopancreatography (ERCP) or magnetic resonance cholangiopancreatography (MRCP), before reaching a diagnosis.

For asymptomatic patients with cirrhosis or those at an early stage of liver disease, ultrasound at 6-month intervals combined with serum liver function tests can be useful to track disease progression and screen for hepatocellular carcinoma or cholangiocarcinoma.^15,16

New noninvasive methods. Noninvasive tools for evaluating the presence and severity of liver fibrosis and for differentiating cirrhosis from noncirrhotic conditions have positive predictive values >85% to 90% for some chronic liver diseases.¹⁷ Transient elastography, which assesses liver stiffness, is one such method. Although it is often used successfully, morbid obesity, small intercostal spaces, and ascites limit its diagnostic capability.¹⁸ Recently, some questions about the validity of elastography to assess the extent of fibrosis in patients with chronic cholestatic conditions have been reported.^19,20

Suspect intrahepatic cholestasis? Your next steps

If imaging techniques do not show bile duct obstruction and you suspect the intrahepatic form, second-level tests could have strategic importance. This is where antimitochondrial antibodies (AMAs) come in. AMAs are immunoglobulins (IgG and IgM) directed against mitochondrial antigens. They are important markers for PBC, which is a T-lymphocyte-mediated attack on small intralobular bile ducts resulting in their gradual destruction and eventual disappearance. The sustained loss of intralobular bile ducts leads to signs and symptoms of cholestasis and eventually results in cirrhosis and liver failure.

AMA serum levels show high sensitivity and specificity (90% and 95%, respectively) for PBC.²¹ Some PBC patients (<5%) show histologic confirmation of the disease, but have negative AMA tests (AMA negative PBC or autoimmune cholangitis).²² Therefore, according to the American Association for the Study of Liver Diseases, diagnosis of PBC is guided by the combination of serologic, biochemical, and histologic criteria.²³ Many PBC patients with or without a positive AMA (≥1:40) also have positive circulating antinuclear antibodies (ANA; ≥1:80). The recent availability of lab tests for antibodies (anti-M2, anti-gp120, anti-sp100) has allowed identification of subgroups of patients who have a more aggressive form of PBC. Patients with PBC often have elevated levels of circulating IgM (>280 mg/dL).

Continue to: Other circulating antibodies

Other circulating antibodies can help discriminate among cholestatic disorders. In particular, positive tests for perinuclear anti-neutrophil cytoplasmic antibodies (pANCA) are found in 25% to 95% of patients with primary sclerosing cholangitis (PSC), a chronic progressive disorder of unknown etiology that is characterized by inflammation, fibrosis, and stricturing of medium and large ducts of the intrahepatic and extrahepatic biliary tree.²⁴ Anti-smooth muscle antibodies (SMA) can be observed in both PSC and autoimmune hepatitis.

Alkaline phosphatase levels are useful for monitoring evolution of primary biliary cholangitis disease.

Finally, there are syndromes with serologic and histologic overlap that are characterized by the simultaneous presence of PBC with autoimmune hepatitis or PSC or overlap of PSC with autoimmune hepatitis.

Liver biopsy fills in the rest of the diagnostic picture

Unfortunately, blood tests reveal little about organ integrity and are not useful for disease staging. The decision to perform a liver biopsy should be based on several factors, including the patient’s age, serum parameters, the need to stage the disease, therapy choices, and prognosis.¹² One should also consider that biopsy is a costly procedure with potentially serious adverse effects; it should not be repeated frequently. However, when a biopsy is done, it provides critical information, including damage to medium-sized intrahepatic bile ducts with neoductular formation or bile duct scars and strictures.

Treating intrahepatic cholestasis

Although FPs often can provide most—or even all—of the care for patients with stable conditions, a specialist consultation might recommend further testing to identify the underlying disease, which is essential to establish the most appropriate treatment.

Treatment of patients with PBC is based on administering hydrophilic secondary bile salt ursodeoxycholic acid (UDCA) 15 mg/kg/d, which is used to equilibrate the ratio between hydrophilic and hydrophobic bile salts in the liver and bile,²⁵ and is the only treatment approved by the US Food and Drug Administration (FDA) for PBC.⁴ Tauroursodeoxycholate is better absorbed than UDCA, and, although partially deconjugated and reconjugated with glycine, it undergoes reduced biotransformation to more hydrophobic metabolites and has benefits, including antioxidant, immunomodulation, and neuroprotective effects over UDCA—especially for long-term therapy in PBC.²⁶ However, it is not used often in clinical practice.

Continue to: Bile acid administration counters the cytotoxic effect...

Bile acid administration counters the cytotoxic effect of hydrophobic bile salts. Although it seems that UDCA might improve biochemical and histologic features of the disease at earlier stages (I-II), it fails in patients with more advanced disease.²⁷ In addition, monitoring and defining response to UDCA is inconsistent, partly because of variations in guideline criteria.^28,29

Despite progress in diagnostic techniques, life expectancy and quality of life for patients with advanced cholestatic conditions remain poor.

Recently a new molecule, obeticholic acid (OCA), has been approved by the FDA. A farnesoid X receptor agonist, OCA is indicated for treating patients who do not tolerate UDCA or as an adjunct to UDCA in those with a partial response to UDCA, defined as lowering ALP levels by <1.5 times the baseline value after 12 months of treatment.

Treating PSC is more complex. Combination therapy with prednisone and azathioprine is recommended only when there is an overlap syndrome between PSC and autoimmune hepatitis.⁴ UDCA at a high dosage (15-20 mg/kg/d) is used to facilitate long-lasting biochemical remission. These patients also need to be monitored for inflammatory bowel diseases, which affect up to 75% of patients,³⁰ and for cholangiocarcinoma, which is a life-limiting complication because of a lack of therapy options. Finally, these patients might need endoscopic-guided dilatation of the biliary tree when they have evidence of dominant fibrotic strictures of the greater bile ducts.^14,31

Addressing the systemic effects of intrahepatic cholestasis

Pruritus. A number of potential pruritogens, including bile salts, endogenous opioids, histamine, serotonin, and lisophosphatidic acid (LPA), can be targeted to relieve pruritus.

• Bile acid resin binders such as cholestyramine are the first step for treating pruritus. UDCA also can be useful, mainly for intrahepatic cholestasis during pregnancy. Rifampicin, 300 mg/d, improves cholestatic pruritus, but is associated with hepatotoxicity and a number of severe reactions, such as nausea, loss of appetite, hemolytic anemia, and thrombocytopenia.³¹
• Most evidence favors a role for opioids in relieving itch, and micro-opioid receptor antagonists (naltrexone, naloxone, nalmefene) that exert an antipruritic effect can be effective.
• Sertraline (a selective serotonin reuptake inhibitor), 50 to 75 mg/d, usually is well tolerated in patients with chronic cholestasis and exerts a beneficial effect on pruritus in approximately 40% of patients.³²
• Extracorporeal albumin dialysis removes albumin-bound pruritogens and has been found to be effective in patients with liver failure. Steroids and UV light also can be used in select patients.
• The potent neuronal activator LPA and its converting enzyme autotaxin have been identified in the serum of patients with cholestatic pruritus; experimental modalities using LPA antagonists are ongoing for treating pruritus in patients who do not respond to other medications.³³

Continue to: Malnutrition

Malnutrition. Many patients with cholestasis are at risk for malnutrition, which can be exacerbated in those with cirrhosis. Causes of malnutrition include poor oral intake, malabsorption, or dental problems that prevent the patient from chewing. Assess the nutritional status of every patient with chronic cholestasis, and stress the importance of multivitamin supplementation to reverse systemic alterations caused by malnutrition.³⁴

When the patient has advanced disease

Despite progress in diagnostic techniques, life expectancy and quality of life for patients with advanced cholestatic conditions remain poor. Patients routinely experience fatigue, pruritus, and complications of cirrhosis including ascites, encephalopathy, and bleeding. Cholestasis also carries the risk of life-threatening complications, partly because of comorbidities such as osteoporosis and malabsorption.

ERCP is widely employed for diagnosing and treating pancreatobiliary diseases; however, its use has dropped over the last 10 years because of the risk of complications.

Liver transplantation can improve the life expectancy of patients with advanced disease, but because of long waiting lists, candidates for transplant often die before an organ becomes available. For many patients who are not in end-stage condition, targeted therapy is crucial to slow disease progression and is recommended along with hepatitis A and B vaccinations and nutritional counseling.³⁵

Extrahepatic cholestasis is suspected? How to proceed

Computer tomography (CT) is recommended for better identification of neoplastic causes of biliary obstruction and for staging purposes. MRCP is an excellent noninvasive imaging technique for evaluating biliary ducts.³⁶

MRCP has 92% to 93% sensitivity and 97% to 98% specificity for diagnosing biliary duct stones.³⁷ MRCP also is the first-choice modality for evaluating bile ducts in patients with suspected PSC. If performed in expert centers, the diagnostic accuracy reaches that of ERCP. A meta-analysis of studies from 2000 to 2006 has shown a sensitivity of 86% and specificity of 94% for diagnosing PSC.³⁸

Endoscopic ultrasonography, which uses an ultrasonographic probe, allows clinicians to evaluate the integrity of the biliary and pancreatic ducts and is effective for diagnosing and staging cancer of the ampulla of Vater (sensitivity 93% vs 7% for abdominal ultrasonography and 29% for CT), and identifying biliary stones and biliary tree strictures.

Continue to: ERCP

ERCP is widely employed for diagnosing and treating pancreatobiliary diseases; however, its use has dropped over the last 10 years because of the risk of complications. ERCP is nearly exclusively used as a therapeutic procedure for pancreatic sphincterotomy, biliary dilatations, and removing biliary stones. It also has a diagnostic role in dominant stenosis or suspected biliary malignancy using brushing cytology and sampling biopsies of the bile ducts.

Treating extrahepatic cholestasis

Treatment of the different underlying conditions that cause extrahepatic cholestasis is surgical. Thus, the potential surgical techniques that can resolve or improve an extrahepatic cholestatic condition are guided by the surgeon and beyond the scope of this article.

Treating osteopenia: A concern for intra- and extrahepatic cholestasis

Vitamin D deficiency as a consequence of reduced intestinal absorption (poor availability of bile salts) or decreased hepatic activation to 25,OH-cholecalcipherol in both intrahepatic and extrahepatic cholestasis can lead to reduced bone formation.³⁹ However, osteopenia can occur even in early stages of the disease. Prescribing bisphosphonates, in combination with calcium and vitamin D3_, to improve bone mineral density is a good practice.⁴⁰

CASE

Blood tests and ultrasound imaging suggest the presence of a chronic liver disease. Other lab tests indicate that the patient has an ALP level 3 times normal. This finding, together with the other tests, points to a likely diagnosis of intrahepatic cholestatic liver disease. Serology confirms positivity for ANA (1:160) and AMA (1:640). The clinician suspects PBC, so the patient is referred to a liver specialist for further evaluation and to determine whether a liver biopsy is needed.

The liver specialist confirms the diagnosis of PBC, performs a transient elastographym, which indicates a low-grade liver fibrosis (F1 out of 4), and starts therapy with UDCA.

CORRESPONDENCE
Ignazio Grattagliano, MD, Italian College of General Practitioners and Primary Care, Via del Sansovino 179, 50142, Florence, Italy; [email protected].

ILLUSTRATIVE CASE

A 58-year-old man presents to your office with a 2-day history of moderate (6/10) left lower quadrant pain, mild fever (none currently), 2 episodes of vomiting, no diarrhea, and no relief with over-the-counter medications. You suspect diverticulitis and obtain an abdominal computed tomography (CT) scan, which shows mild, uncomplicated (Hinchey stage 1a) diverticulitis.

How would you treat him?

Diverticulitis is common; about 200,000 people per year are admitted to the hospital because of diverticulitis in the United States.^2,3 Health care providers typically treat diverticular disease with antibiotics and bowel rest.^2,3 While severe forms of diverticulitis often require parenteral antibiotics and/or surgery, practitioners are increasingly managing the condition with oral antibiotics.⁴

One previous randomized control trial (RCT; N=623) found that antibiotic treatment (compared with no antibiotic treatment) for acute uncomplicated diverticulitis did not speed recovery or prevent complications (perforation or abscess formation) or recurrence at 12 months.⁵ The study’s strengths included limiting enrollment to people with CT-proven diverticulitis, using a good randomization and concealment process, and employing intention-to-treat analysis. The study was limited by a lack of a standardized antibiotic regimen across centers, previous diverticulitis diagnoses in 40% of patients, non-uniform follow-up processes to confirm anatomic resolution, and the lack of assessment to confirm resolution.⁵

STUDY SUMMARY

RCT finds that watchful waiting is just as effective as antibiotic Tx

This newer study was a single-blind RCT that compared treatment with antibiotics to observation among 528 adult patients in the Netherlands. Patients were enrolled if they had CT-proven, primary, left-sided, uncomplicated acute diverticulitis (Hinchey stage 1a and 1b).¹ (The Hinchey classification is based on radiologic findings, with 0 for clinical diverticulitis only, 1a for confined pericolic inflammation or phlegmon, and 1b for pericolic or mesocolic abscess.⁶) Exclusion criteria included suspicion of colonic cancer by CT or ultrasound (US), previous CT/US-proven diverticulitis, sepsis, pregnancy, or antibiotic use in the previous 4 weeks.¹

Observational vs antibiotic treatment. Enrolled patients were randomized to receive IV administration of amoxicillin-clavulanate 1200 mg 4 times daily for at least 48 hours followed by 625 mg PO 3 times daily for 10 total days of antibiotic treatment (n=266) or to be observed (n=262). Computerized randomization, with a random varying block size and stratified by Hinchey classification and center, was performed, and allocation was concealed. The investigators were masked to the allocation until all analyses were completed.¹

The primary outcome was the time to functional recovery (resumption of pre-illness work activities) during a 6-month follow-up period. Secondary outcomes included hospital readmission rate; complicated, ongoing, and recurrent diverticulitis; sigmoid resection; other nonsurgical intervention; antibiotic treatment adverse effects; and all-cause mortality.

Continue to: Results

This study is the first to look at functional return to work and the only study to gauge long-term outcomes with observational treatment.

Results. Median recovery time for observational treatment was not inferior to antibiotic treatment (14 days vs 12 days; P=.15; hazard ratio [HR] for functional recovery=0.91; lower limit of 1-sided 95% confidence interval, 0.78). Observation was not inferior to antibiotics for any of the secondary endpoints at 6 and 12 months of follow-up (complicated diverticulitis, 3.8% vs 2.6%, respectively; P=.377), recurrent diverticulitis (3.4% vs 3%; P=.494), readmission (17.6% vs 12%; P=.148), or adverse events (48.5% vs 54.5%; P=.221). Initial hospitalization length of stay was shorter in the observation group (2 vs 3 days; P=.006). The researchers conducted a 24-month telephone follow-up, but no differences from the 12-month follow-up were noted.¹

WHAT’S NEW

A study that looks at a true patient-oriented outcome

Previous studies of treatment options for acute uncomplicated diverticulitis looked at short-term outcomes, or at readmission, recurrence, and surgical intervention rate, or requirement for percutaneous drainage.^7,8 This study is the first one to look at functional return to work (a true patient-oriented outcome). And it is the only study to look out to 24 months to gauge long-term outcomes with observational treatment.

CAVEATS

Can’t generalize findings to patients with worse forms of diverticulitis

It is worth noting that the findings of this study apply only to the mildest form of CT-proven acute diverticulitis (those patients classified as having Hinchey 1a disease), and is not generalizable to patients with more severe forms. Not enough patients with Hinchey 1b acute diverticulitis were enrolled in the study to reach any conclusions about treatment.

Various guidelines issued outside the United States recommend antibiotics for uncomplicated diverticulitis; however, the American Gastroenterological Association (AGA) indicates that antibiotics should be used selectively.^1,9,10 This recommendation was based on an emerging understanding that diverticulitis maybe more inflammatory than infectious in nature. The AGA guideline authors acknowledge that their conclusion was based on low-quality evidence.⁹

Continuet to: CHALLENGES TO IMPLEMENTATION

None to speak of

We see no challenges to implementing this recommendation.

ACKNOWLEDGEMENT

The PURLs Surveillance System was supported in part by Grant Number UL1RR024999 from the National Center For Research Resources, a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center For Research Resources or the National Institutes of Health.

ILLUSTRATIVE CASE

A 58-year-old man presents to your office with a 2-day history of moderate (6/10) left lower quadrant pain, mild fever (none currently), 2 episodes of vomiting, no diarrhea, and no relief with over-the-counter medications. You suspect diverticulitis and obtain an abdominal computed tomography (CT) scan, which shows mild, uncomplicated (Hinchey stage 1a) diverticulitis.

How would you treat him?

Diverticulitis is common; about 200,000 people per year are admitted to the hospital because of diverticulitis in the United States.^2,3 Health care providers typically treat diverticular disease with antibiotics and bowel rest.^2,3 While severe forms of diverticulitis often require parenteral antibiotics and/or surgery, practitioners are increasingly managing the condition with oral antibiotics.⁴

One previous randomized control trial (RCT; N=623) found that antibiotic treatment (compared with no antibiotic treatment) for acute uncomplicated diverticulitis did not speed recovery or prevent complications (perforation or abscess formation) or recurrence at 12 months.⁵ The study’s strengths included limiting enrollment to people with CT-proven diverticulitis, using a good randomization and concealment process, and employing intention-to-treat analysis. The study was limited by a lack of a standardized antibiotic regimen across centers, previous diverticulitis diagnoses in 40% of patients, non-uniform follow-up processes to confirm anatomic resolution, and the lack of assessment to confirm resolution.⁵

STUDY SUMMARY

RCT finds that watchful waiting is just as effective as antibiotic Tx

This newer study was a single-blind RCT that compared treatment with antibiotics to observation among 528 adult patients in the Netherlands. Patients were enrolled if they had CT-proven, primary, left-sided, uncomplicated acute diverticulitis (Hinchey stage 1a and 1b).¹ (The Hinchey classification is based on radiologic findings, with 0 for clinical diverticulitis only, 1a for confined pericolic inflammation or phlegmon, and 1b for pericolic or mesocolic abscess.⁶) Exclusion criteria included suspicion of colonic cancer by CT or ultrasound (US), previous CT/US-proven diverticulitis, sepsis, pregnancy, or antibiotic use in the previous 4 weeks.¹

Observational vs antibiotic treatment. Enrolled patients were randomized to receive IV administration of amoxicillin-clavulanate 1200 mg 4 times daily for at least 48 hours followed by 625 mg PO 3 times daily for 10 total days of antibiotic treatment (n=266) or to be observed (n=262). Computerized randomization, with a random varying block size and stratified by Hinchey classification and center, was performed, and allocation was concealed. The investigators were masked to the allocation until all analyses were completed.¹

The primary outcome was the time to functional recovery (resumption of pre-illness work activities) during a 6-month follow-up period. Secondary outcomes included hospital readmission rate; complicated, ongoing, and recurrent diverticulitis; sigmoid resection; other nonsurgical intervention; antibiotic treatment adverse effects; and all-cause mortality.

Continue to: Results

This study is the first to look at functional return to work and the only study to gauge long-term outcomes with observational treatment.

Results. Median recovery time for observational treatment was not inferior to antibiotic treatment (14 days vs 12 days; P=.15; hazard ratio [HR] for functional recovery=0.91; lower limit of 1-sided 95% confidence interval, 0.78). Observation was not inferior to antibiotics for any of the secondary endpoints at 6 and 12 months of follow-up (complicated diverticulitis, 3.8% vs 2.6%, respectively; P=.377), recurrent diverticulitis (3.4% vs 3%; P=.494), readmission (17.6% vs 12%; P=.148), or adverse events (48.5% vs 54.5%; P=.221). Initial hospitalization length of stay was shorter in the observation group (2 vs 3 days; P=.006). The researchers conducted a 24-month telephone follow-up, but no differences from the 12-month follow-up were noted.¹

WHAT’S NEW

A study that looks at a true patient-oriented outcome

Previous studies of treatment options for acute uncomplicated diverticulitis looked at short-term outcomes, or at readmission, recurrence, and surgical intervention rate, or requirement for percutaneous drainage.^7,8 This study is the first one to look at functional return to work (a true patient-oriented outcome). And it is the only study to look out to 24 months to gauge long-term outcomes with observational treatment.

CAVEATS

Can’t generalize findings to patients with worse forms of diverticulitis

It is worth noting that the findings of this study apply only to the mildest form of CT-proven acute diverticulitis (those patients classified as having Hinchey 1a disease), and is not generalizable to patients with more severe forms. Not enough patients with Hinchey 1b acute diverticulitis were enrolled in the study to reach any conclusions about treatment.

Various guidelines issued outside the United States recommend antibiotics for uncomplicated diverticulitis; however, the American Gastroenterological Association (AGA) indicates that antibiotics should be used selectively.^1,9,10 This recommendation was based on an emerging understanding that diverticulitis maybe more inflammatory than infectious in nature. The AGA guideline authors acknowledge that their conclusion was based on low-quality evidence.⁹

Continuet to: CHALLENGES TO IMPLEMENTATION

None to speak of

We see no challenges to implementing this recommendation.

ACKNOWLEDGEMENT

The PURLs Surveillance System was supported in part by Grant Number UL1RR024999 from the National Center For Research Resources, a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center For Research Resources or the National Institutes of Health.

ILLUSTRATIVE CASE

A 58-year-old man presents to your office with a 2-day history of moderate (6/10) left lower quadrant pain, mild fever (none currently), 2 episodes of vomiting, no diarrhea, and no relief with over-the-counter medications. You suspect diverticulitis and obtain an abdominal computed tomography (CT) scan, which shows mild, uncomplicated (Hinchey stage 1a) diverticulitis.

How would you treat him?

Diverticulitis is common; about 200,000 people per year are admitted to the hospital because of diverticulitis in the United States.^2,3 Health care providers typically treat diverticular disease with antibiotics and bowel rest.^2,3 While severe forms of diverticulitis often require parenteral antibiotics and/or surgery, practitioners are increasingly managing the condition with oral antibiotics.⁴

One previous randomized control trial (RCT; N=623) found that antibiotic treatment (compared with no antibiotic treatment) for acute uncomplicated diverticulitis did not speed recovery or prevent complications (perforation or abscess formation) or recurrence at 12 months.⁵ The study’s strengths included limiting enrollment to people with CT-proven diverticulitis, using a good randomization and concealment process, and employing intention-to-treat analysis. The study was limited by a lack of a standardized antibiotic regimen across centers, previous diverticulitis diagnoses in 40% of patients, non-uniform follow-up processes to confirm anatomic resolution, and the lack of assessment to confirm resolution.⁵

STUDY SUMMARY

RCT finds that watchful waiting is just as effective as antibiotic Tx

This newer study was a single-blind RCT that compared treatment with antibiotics to observation among 528 adult patients in the Netherlands. Patients were enrolled if they had CT-proven, primary, left-sided, uncomplicated acute diverticulitis (Hinchey stage 1a and 1b).¹ (The Hinchey classification is based on radiologic findings, with 0 for clinical diverticulitis only, 1a for confined pericolic inflammation or phlegmon, and 1b for pericolic or mesocolic abscess.⁶) Exclusion criteria included suspicion of colonic cancer by CT or ultrasound (US), previous CT/US-proven diverticulitis, sepsis, pregnancy, or antibiotic use in the previous 4 weeks.¹

Observational vs antibiotic treatment. Enrolled patients were randomized to receive IV administration of amoxicillin-clavulanate 1200 mg 4 times daily for at least 48 hours followed by 625 mg PO 3 times daily for 10 total days of antibiotic treatment (n=266) or to be observed (n=262). Computerized randomization, with a random varying block size and stratified by Hinchey classification and center, was performed, and allocation was concealed. The investigators were masked to the allocation until all analyses were completed.¹

The primary outcome was the time to functional recovery (resumption of pre-illness work activities) during a 6-month follow-up period. Secondary outcomes included hospital readmission rate; complicated, ongoing, and recurrent diverticulitis; sigmoid resection; other nonsurgical intervention; antibiotic treatment adverse effects; and all-cause mortality.

Continue to: Results

This study is the first to look at functional return to work and the only study to gauge long-term outcomes with observational treatment.

Results. Median recovery time for observational treatment was not inferior to antibiotic treatment (14 days vs 12 days; P=.15; hazard ratio [HR] for functional recovery=0.91; lower limit of 1-sided 95% confidence interval, 0.78). Observation was not inferior to antibiotics for any of the secondary endpoints at 6 and 12 months of follow-up (complicated diverticulitis, 3.8% vs 2.6%, respectively; P=.377), recurrent diverticulitis (3.4% vs 3%; P=.494), readmission (17.6% vs 12%; P=.148), or adverse events (48.5% vs 54.5%; P=.221). Initial hospitalization length of stay was shorter in the observation group (2 vs 3 days; P=.006). The researchers conducted a 24-month telephone follow-up, but no differences from the 12-month follow-up were noted.¹

WHAT’S NEW

A study that looks at a true patient-oriented outcome

Previous studies of treatment options for acute uncomplicated diverticulitis looked at short-term outcomes, or at readmission, recurrence, and surgical intervention rate, or requirement for percutaneous drainage.^7,8 This study is the first one to look at functional return to work (a true patient-oriented outcome). And it is the only study to look out to 24 months to gauge long-term outcomes with observational treatment.

CAVEATS

Can’t generalize findings to patients with worse forms of diverticulitis

It is worth noting that the findings of this study apply only to the mildest form of CT-proven acute diverticulitis (those patients classified as having Hinchey 1a disease), and is not generalizable to patients with more severe forms. Not enough patients with Hinchey 1b acute diverticulitis were enrolled in the study to reach any conclusions about treatment.

Various guidelines issued outside the United States recommend antibiotics for uncomplicated diverticulitis; however, the American Gastroenterological Association (AGA) indicates that antibiotics should be used selectively.^1,9,10 This recommendation was based on an emerging understanding that diverticulitis maybe more inflammatory than infectious in nature. The AGA guideline authors acknowledge that their conclusion was based on low-quality evidence.⁹

Continuet to: CHALLENGES TO IMPLEMENTATION

None to speak of

We see no challenges to implementing this recommendation.

ACKNOWLEDGEMENT

The PURLs Surveillance System was supported in part by Grant Number UL1RR024999 from the National Center For Research Resources, a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center For Research Resources or the National Institutes of Health.

THE CASE

A 7-year-old boy presented with a one-week history of a pruritic rash, which first appeared on his back and continued to spread across his entire body. The patient’s medical history was significant for a scalp lesion (FIGURE 1) that was being treated with oral griseofulvin (started 3 days earlier). He had no history of seasonal allergies, asthma, recent illness, or recent immunizations.

The physical exam was significant for a body-wide, nonerythematous, papular rash (FIGURE 2). There was evidence of excoriation due to itching. No mucosal involvement was appreciated. The remainder of the examination was unremarkable.

QUESTION

Based on the patient’s history and physical exam, which of the following is the most likely diagnosis?

A. Gianotti-Crosti syndrome

B. Atopic dermatitis

C. Dermatophytid reaction

D. Morbilliform drug eruption.

Continue to: THE DIAGNOSIS

The answer is C, dermatophytid reaction.

DISCUSSION

A dermatophytid reaction is a type of id reaction, or autoeczematization. An id reaction is when a localized dermatitis becomes a generalized pruritic eruption.¹ In this case, the patient’s dermatitis was the result of a dermatophyte infection (tinea capitis), but an id reaction can also occur in response to noninfectious dermatitides and may be of an atopic, contact, or seborrheic nature.¹

Dermatophytid reactions occur in up to 5% of all dermatophyte infections (most commonly tinea pedis) and are proposed to be type IV hypersensitivity reactions to the release of fungal antigens.¹ These reactions can occur either before or after the initiation of antifungal treatment. They manifest as symmetric, pruritic, papulovesicular eruptions with fine scaling and commonly affect the face, trunk, extremities, palms, and interdigital spaces.¹

What about other possible diagnoses?

Gianotti-Crosti syndrome is an asymptomatic, symmetric, papulovesicular dermatosis that involves the face, limbs, and buttocks of children 2 to 6 years of age.² The lesions develop in response to a respiratory or gastrointestinal illness.² They are typically associated with Epstein-Barr virus, hepatitis B, cytomegalovirus, respiratory syncytial virus, and coxsackievirus, but can occur with bacterial infections or following administration of routine immunizations.²

While the initial impulse may be to discontinue oral antifungals, these treatments help resolve the underlying dermatophyte infection and should be continued.

The lesions are self-limited and resolve within 2 months.² Symptomatic lesions may be treated with oral antihistamines or steroids (topical or systemic).²

Continue to: Atopic dermatitis

Atopic dermatitis is characterized by symmetric involvement of the flexural sur­faces of the body with a pruritic, erythematous rash that may have a fine scale.³ It usually manifests prior to 2 years of age, is recurrent, and is commonly associated with allergic rhinitis and asthma.³ Treatment involves trigger avoidance, topical emollients, topical corticosteroids, dilute bleach baths, and topical calcineurin inhibitors.^3,4 For patients with significant nocturnal symptoms and sleep loss, oral antihistamines may be helpful.⁴

Morbilliform drug eruptions are the most common type of dermatologic drug reaction.⁵ These rashes occur approximately one to 2 weeks after exposure to a causative drug; they consist of pruritic, erythematous papules or macules that start centrally and may spread to the proximal extremities.⁵ Treatment involves discontinuation of the offending agent. Symptomatic relief may be achieved with oral antihistamines or topical or systemic corticosteroids.⁵

Treatment of dermatophytid reactions

While the initial impulse in the treatment of a dermatophytid reaction may be to discon­tinue oral antifungals, these treatments actually help resolve the underlying dermatophyte infection and should be continued. For children with tinea capitis, at least 6 weeks of treatment with an oral antifungal agent is warranted. Medications approved by the US Food and Drug Administration include terbinafine (for patients >4 years of age) and griseofulvin (for patients >2 years of age). Dosages are weight-based. (Fluconazole and itraconazole are not approved for this indication.) Lubricants, topical corticosteroids, and oral antihistamines can be used for acute management of pruritus.¹

Our patient was treated successfully with griseofulvin and an oral antihistamine. However, he experienced headaches attrib­uted to griseofulvin and was switched to terbinafine 5 mg/kg/d for 4 weeks. His tinea capitis was resolved at 8 weeks.

CORRESPONDENCE
Richard Temple, MD, CAPT, MC, USN. Department of Family Medicine, Naval Medical Center Camp Lejeune, 100 Brewster Blvd, Camp Lejeune, NC 28547; [email protected].

THE CASE

A 7-year-old boy presented with a one-week history of a pruritic rash, which first appeared on his back and continued to spread across his entire body. The patient’s medical history was significant for a scalp lesion (FIGURE 1) that was being treated with oral griseofulvin (started 3 days earlier). He had no history of seasonal allergies, asthma, recent illness, or recent immunizations.

The physical exam was significant for a body-wide, nonerythematous, papular rash (FIGURE 2). There was evidence of excoriation due to itching. No mucosal involvement was appreciated. The remainder of the examination was unremarkable.

QUESTION

Based on the patient’s history and physical exam, which of the following is the most likely diagnosis?

A. Gianotti-Crosti syndrome

B. Atopic dermatitis

C. Dermatophytid reaction

D. Morbilliform drug eruption.

Continue to: THE DIAGNOSIS

The answer is C, dermatophytid reaction.

DISCUSSION

A dermatophytid reaction is a type of id reaction, or autoeczematization. An id reaction is when a localized dermatitis becomes a generalized pruritic eruption.¹ In this case, the patient’s dermatitis was the result of a dermatophyte infection (tinea capitis), but an id reaction can also occur in response to noninfectious dermatitides and may be of an atopic, contact, or seborrheic nature.¹

Dermatophytid reactions occur in up to 5% of all dermatophyte infections (most commonly tinea pedis) and are proposed to be type IV hypersensitivity reactions to the release of fungal antigens.¹ These reactions can occur either before or after the initiation of antifungal treatment. They manifest as symmetric, pruritic, papulovesicular eruptions with fine scaling and commonly affect the face, trunk, extremities, palms, and interdigital spaces.¹

What about other possible diagnoses?

Gianotti-Crosti syndrome is an asymptomatic, symmetric, papulovesicular dermatosis that involves the face, limbs, and buttocks of children 2 to 6 years of age.² The lesions develop in response to a respiratory or gastrointestinal illness.² They are typically associated with Epstein-Barr virus, hepatitis B, cytomegalovirus, respiratory syncytial virus, and coxsackievirus, but can occur with bacterial infections or following administration of routine immunizations.²

While the initial impulse may be to discontinue oral antifungals, these treatments help resolve the underlying dermatophyte infection and should be continued.

The lesions are self-limited and resolve within 2 months.² Symptomatic lesions may be treated with oral antihistamines or steroids (topical or systemic).²

Continue to: Atopic dermatitis

Atopic dermatitis is characterized by symmetric involvement of the flexural sur­faces of the body with a pruritic, erythematous rash that may have a fine scale.³ It usually manifests prior to 2 years of age, is recurrent, and is commonly associated with allergic rhinitis and asthma.³ Treatment involves trigger avoidance, topical emollients, topical corticosteroids, dilute bleach baths, and topical calcineurin inhibitors.^3,4 For patients with significant nocturnal symptoms and sleep loss, oral antihistamines may be helpful.⁴

Morbilliform drug eruptions are the most common type of dermatologic drug reaction.⁵ These rashes occur approximately one to 2 weeks after exposure to a causative drug; they consist of pruritic, erythematous papules or macules that start centrally and may spread to the proximal extremities.⁵ Treatment involves discontinuation of the offending agent. Symptomatic relief may be achieved with oral antihistamines or topical or systemic corticosteroids.⁵

Treatment of dermatophytid reactions

While the initial impulse in the treatment of a dermatophytid reaction may be to discon­tinue oral antifungals, these treatments actually help resolve the underlying dermatophyte infection and should be continued. For children with tinea capitis, at least 6 weeks of treatment with an oral antifungal agent is warranted. Medications approved by the US Food and Drug Administration include terbinafine (for patients >4 years of age) and griseofulvin (for patients >2 years of age). Dosages are weight-based. (Fluconazole and itraconazole are not approved for this indication.) Lubricants, topical corticosteroids, and oral antihistamines can be used for acute management of pruritus.¹

Our patient was treated successfully with griseofulvin and an oral antihistamine. However, he experienced headaches attrib­uted to griseofulvin and was switched to terbinafine 5 mg/kg/d for 4 weeks. His tinea capitis was resolved at 8 weeks.

CORRESPONDENCE
Richard Temple, MD, CAPT, MC, USN. Department of Family Medicine, Naval Medical Center Camp Lejeune, 100 Brewster Blvd, Camp Lejeune, NC 28547; [email protected].

THE CASE

A 7-year-old boy presented with a one-week history of a pruritic rash, which first appeared on his back and continued to spread across his entire body. The patient’s medical history was significant for a scalp lesion (FIGURE 1) that was being treated with oral griseofulvin (started 3 days earlier). He had no history of seasonal allergies, asthma, recent illness, or recent immunizations.

The physical exam was significant for a body-wide, nonerythematous, papular rash (FIGURE 2). There was evidence of excoriation due to itching. No mucosal involvement was appreciated. The remainder of the examination was unremarkable.

QUESTION

Based on the patient’s history and physical exam, which of the following is the most likely diagnosis?

A. Gianotti-Crosti syndrome

B. Atopic dermatitis

C. Dermatophytid reaction

D. Morbilliform drug eruption.

Continue to: THE DIAGNOSIS

The answer is C, dermatophytid reaction.

DISCUSSION

A dermatophytid reaction is a type of id reaction, or autoeczematization. An id reaction is when a localized dermatitis becomes a generalized pruritic eruption.¹ In this case, the patient’s dermatitis was the result of a dermatophyte infection (tinea capitis), but an id reaction can also occur in response to noninfectious dermatitides and may be of an atopic, contact, or seborrheic nature.¹

Dermatophytid reactions occur in up to 5% of all dermatophyte infections (most commonly tinea pedis) and are proposed to be type IV hypersensitivity reactions to the release of fungal antigens.¹ These reactions can occur either before or after the initiation of antifungal treatment. They manifest as symmetric, pruritic, papulovesicular eruptions with fine scaling and commonly affect the face, trunk, extremities, palms, and interdigital spaces.¹

What about other possible diagnoses?

Gianotti-Crosti syndrome is an asymptomatic, symmetric, papulovesicular dermatosis that involves the face, limbs, and buttocks of children 2 to 6 years of age.² The lesions develop in response to a respiratory or gastrointestinal illness.² They are typically associated with Epstein-Barr virus, hepatitis B, cytomegalovirus, respiratory syncytial virus, and coxsackievirus, but can occur with bacterial infections or following administration of routine immunizations.²

While the initial impulse may be to discontinue oral antifungals, these treatments help resolve the underlying dermatophyte infection and should be continued.

The lesions are self-limited and resolve within 2 months.² Symptomatic lesions may be treated with oral antihistamines or steroids (topical or systemic).²

Continue to: Atopic dermatitis

Atopic dermatitis is characterized by symmetric involvement of the flexural sur­faces of the body with a pruritic, erythematous rash that may have a fine scale.³ It usually manifests prior to 2 years of age, is recurrent, and is commonly associated with allergic rhinitis and asthma.³ Treatment involves trigger avoidance, topical emollients, topical corticosteroids, dilute bleach baths, and topical calcineurin inhibitors.^3,4 For patients with significant nocturnal symptoms and sleep loss, oral antihistamines may be helpful.⁴

Morbilliform drug eruptions are the most common type of dermatologic drug reaction.⁵ These rashes occur approximately one to 2 weeks after exposure to a causative drug; they consist of pruritic, erythematous papules or macules that start centrally and may spread to the proximal extremities.⁵ Treatment involves discontinuation of the offending agent. Symptomatic relief may be achieved with oral antihistamines or topical or systemic corticosteroids.⁵

Treatment of dermatophytid reactions

While the initial impulse in the treatment of a dermatophytid reaction may be to discon­tinue oral antifungals, these treatments actually help resolve the underlying dermatophyte infection and should be continued. For children with tinea capitis, at least 6 weeks of treatment with an oral antifungal agent is warranted. Medications approved by the US Food and Drug Administration include terbinafine (for patients >4 years of age) and griseofulvin (for patients >2 years of age). Dosages are weight-based. (Fluconazole and itraconazole are not approved for this indication.) Lubricants, topical corticosteroids, and oral antihistamines can be used for acute management of pruritus.¹

Our patient was treated successfully with griseofulvin and an oral antihistamine. However, he experienced headaches attrib­uted to griseofulvin and was switched to terbinafine 5 mg/kg/d for 4 weeks. His tinea capitis was resolved at 8 weeks.

CORRESPONDENCE
Richard Temple, MD, CAPT, MC, USN. Department of Family Medicine, Naval Medical Center Camp Lejeune, 100 Brewster Blvd, Camp Lejeune, NC 28547; [email protected].

Diabetes affects approximately 9.4% of the US population (more than 30 million people),¹ and it is one of the most common conditions treated by family physicians. Additionally, more than 80 million Americans meet the criteria for prediabetes.¹ The prevalence of diabetes has increased in adults between the time periods 1988-1994 and 2011-2014, and it varies by race and ethnicity, with the highest prevalence, 18%, among African Americans and Mexican Americans, and the lowest, 9.6%, among non-Hispanic whites (FIGURE).²

Diet is the cornerstone of diabetes treatment

The foundation of a comprehensive management plan for type 2 diabetes mellitus (T2DM) is an appropriate diet. A growing body of evidence shows that a well-structured diet is important in controlling diabetes, delaying or preventing the onset of diabetes, and, in some instances, contributing to its remission. Diabetes UK, the United Kingdom’s equivalent of the American Diabetes Association (ADA), recently updated its clinical guideline for physicians and patients on the role of nutrition in managing and preventing diabetes, and it is consistent with one published by the ADA in 2013.^3,4

The Diabetes UK guideline is the result of an evidence-based process that meets the standards recommended by the National Academy of Medicine (previously the Institute of Medicine): a systematic review and formal assessment of the quality of the evidence, and recommendations based on the highest quality evidence available, with the level of evidence stated for each recommendation.⁵ Assessing the level of evidence and determining the strengths of recommendations were done using the Grades of Recommendation Assessment, Development, and Evaluation (GRADE) system, which uses an approach similar to that of the Strength of Recommendation Taxonomy (SORT).

What, and what not, to focus on. The first set of recommendations states that everyone with, or at risk for, diabetes should receive structured, personalized, and ongoing nutritional advice from a dietician who is coordinated with their clinical care. Nutritional advice should focus on the quality and quantity of food, not on specific nutrients (fat and carbohydrates), since there is no good evidence on what proportion of such nutrients is optimal. And it should be tailored to the culture and eating preferences of the patient.

The type of diet with the strongest evidence base for preventing T2DM is a Mediterranean diet, which is supported by level-4, high-quality evidence. Important aspects of a Mediterranean diet are the regular consumption of nuts, whole grains, fruits, and vegetables; use of olive oil instead of butter; and favoring fish over red meat.⁶ Other dietary patterns associated with reduced risk but supported only by level-2, low-quality evidence, include Dietary Approaches to Stop Hypertension (DASH), vegetarian, vegan, and Nordic healthy diets. Moderate carbohydrate restriction is supported only by level-1, very low-quality evidence.

The UK guideline, too, recommends preferentially eating whole grains, fruits, and green leafy vegetables, as well as yogurt, cheese, tea, and coffee. And it advises reducing consumption of red processed meats, potatoes (especially French fries), sugar-sweetened beverages, and refined carbohydrates. However, these specific food preferences are supported only by low-level evidence. Both Diabetes UK and the ADA recommend minimizing consumption of free sugars and added fructose, in addition to sugar-sweetened drinks, but conclude that artificial sweeteners are safe and can be recommended. Both organizations also recommend against the use of vitamins and minerals to manage or prevent diabetes and against protein restriction for those with diabetic nephropathy.

Plant stanols and plant sterols are found in a variety of plant foods such as cereals, vegetable oils, seeds, and nuts, and are now being added to some food products. (For more on plant stanols and plant sterols.) They have a chemical structure similar to cholesterol and reduce the intestinal absorption of cholesterol, thereby lowering total serum cholesterol and LDL-cholesterol. Both Diabetes UK and the ADA recommend 2 to 3 grams of stanols/sterols per day.

Continue to: Alcohol intake

Alcohol intake. And what about alcohol intake in those with T2DM? Once again, both guidelines are in concert by stating that alcohol use in those with diabetes should be moderate, defined by the ADA as one or fewer drinks/d for women and 2 or fewer for men.

Weight loss and exercise are important, too. Those who are overweight or obese with T2DM can improve glycemic control with a 5% weight loss achieved by reducing caloric intake and by increasing energy expenditure with 150 minutes of moderate physical activity per week over at least 3 days.³ This recommendation is supported by high-quality evidence.

A 15-kg weight loss is recommended for those attempting diabetes remission (supported by moderate-level evidence).³ One small study in the United Kingdom found that more than half of those with T2DM could achieve remission with weight loss of 10 kg or more; 86% with weight loss of 15 kg or more.⁷ The Diabetes UK guideline panel rated this as having moderate-level evidence.

The bottom line. Diet and exercise are key interventions for the prevention and treatment of diabetes and can lead to remission if sufficient weight loss is achieved. To achieve and maintain an optimal diet, patients need individualized professional advice and followup. The evidence base for nutritional advice is growing and can be used to improve the quality of these patient-provider interactions.

Diabetes affects approximately 9.4% of the US population (more than 30 million people),¹ and it is one of the most common conditions treated by family physicians. Additionally, more than 80 million Americans meet the criteria for prediabetes.¹ The prevalence of diabetes has increased in adults between the time periods 1988-1994 and 2011-2014, and it varies by race and ethnicity, with the highest prevalence, 18%, among African Americans and Mexican Americans, and the lowest, 9.6%, among non-Hispanic whites (FIGURE).²

Diet is the cornerstone of diabetes treatment

The foundation of a comprehensive management plan for type 2 diabetes mellitus (T2DM) is an appropriate diet. A growing body of evidence shows that a well-structured diet is important in controlling diabetes, delaying or preventing the onset of diabetes, and, in some instances, contributing to its remission. Diabetes UK, the United Kingdom’s equivalent of the American Diabetes Association (ADA), recently updated its clinical guideline for physicians and patients on the role of nutrition in managing and preventing diabetes, and it is consistent with one published by the ADA in 2013.^3,4

The Diabetes UK guideline is the result of an evidence-based process that meets the standards recommended by the National Academy of Medicine (previously the Institute of Medicine): a systematic review and formal assessment of the quality of the evidence, and recommendations based on the highest quality evidence available, with the level of evidence stated for each recommendation.⁵ Assessing the level of evidence and determining the strengths of recommendations were done using the Grades of Recommendation Assessment, Development, and Evaluation (GRADE) system, which uses an approach similar to that of the Strength of Recommendation Taxonomy (SORT).

What, and what not, to focus on. The first set of recommendations states that everyone with, or at risk for, diabetes should receive structured, personalized, and ongoing nutritional advice from a dietician who is coordinated with their clinical care. Nutritional advice should focus on the quality and quantity of food, not on specific nutrients (fat and carbohydrates), since there is no good evidence on what proportion of such nutrients is optimal. And it should be tailored to the culture and eating preferences of the patient.

The type of diet with the strongest evidence base for preventing T2DM is a Mediterranean diet, which is supported by level-4, high-quality evidence. Important aspects of a Mediterranean diet are the regular consumption of nuts, whole grains, fruits, and vegetables; use of olive oil instead of butter; and favoring fish over red meat.⁶ Other dietary patterns associated with reduced risk but supported only by level-2, low-quality evidence, include Dietary Approaches to Stop Hypertension (DASH), vegetarian, vegan, and Nordic healthy diets. Moderate carbohydrate restriction is supported only by level-1, very low-quality evidence.

The UK guideline, too, recommends preferentially eating whole grains, fruits, and green leafy vegetables, as well as yogurt, cheese, tea, and coffee. And it advises reducing consumption of red processed meats, potatoes (especially French fries), sugar-sweetened beverages, and refined carbohydrates. However, these specific food preferences are supported only by low-level evidence. Both Diabetes UK and the ADA recommend minimizing consumption of free sugars and added fructose, in addition to sugar-sweetened drinks, but conclude that artificial sweeteners are safe and can be recommended. Both organizations also recommend against the use of vitamins and minerals to manage or prevent diabetes and against protein restriction for those with diabetic nephropathy.

Plant stanols and plant sterols are found in a variety of plant foods such as cereals, vegetable oils, seeds, and nuts, and are now being added to some food products. (For more on plant stanols and plant sterols.) They have a chemical structure similar to cholesterol and reduce the intestinal absorption of cholesterol, thereby lowering total serum cholesterol and LDL-cholesterol. Both Diabetes UK and the ADA recommend 2 to 3 grams of stanols/sterols per day.

Continue to: Alcohol intake

Alcohol intake. And what about alcohol intake in those with T2DM? Once again, both guidelines are in concert by stating that alcohol use in those with diabetes should be moderate, defined by the ADA as one or fewer drinks/d for women and 2 or fewer for men.

Weight loss and exercise are important, too. Those who are overweight or obese with T2DM can improve glycemic control with a 5% weight loss achieved by reducing caloric intake and by increasing energy expenditure with 150 minutes of moderate physical activity per week over at least 3 days.³ This recommendation is supported by high-quality evidence.

A 15-kg weight loss is recommended for those attempting diabetes remission (supported by moderate-level evidence).³ One small study in the United Kingdom found that more than half of those with T2DM could achieve remission with weight loss of 10 kg or more; 86% with weight loss of 15 kg or more.⁷ The Diabetes UK guideline panel rated this as having moderate-level evidence.

The bottom line. Diet and exercise are key interventions for the prevention and treatment of diabetes and can lead to remission if sufficient weight loss is achieved. To achieve and maintain an optimal diet, patients need individualized professional advice and followup. The evidence base for nutritional advice is growing and can be used to improve the quality of these patient-provider interactions.

Diabetes affects approximately 9.4% of the US population (more than 30 million people),¹ and it is one of the most common conditions treated by family physicians. Additionally, more than 80 million Americans meet the criteria for prediabetes.¹ The prevalence of diabetes has increased in adults between the time periods 1988-1994 and 2011-2014, and it varies by race and ethnicity, with the highest prevalence, 18%, among African Americans and Mexican Americans, and the lowest, 9.6%, among non-Hispanic whites (FIGURE).²

Diet is the cornerstone of diabetes treatment

The foundation of a comprehensive management plan for type 2 diabetes mellitus (T2DM) is an appropriate diet. A growing body of evidence shows that a well-structured diet is important in controlling diabetes, delaying or preventing the onset of diabetes, and, in some instances, contributing to its remission. Diabetes UK, the United Kingdom’s equivalent of the American Diabetes Association (ADA), recently updated its clinical guideline for physicians and patients on the role of nutrition in managing and preventing diabetes, and it is consistent with one published by the ADA in 2013.^3,4

The Diabetes UK guideline is the result of an evidence-based process that meets the standards recommended by the National Academy of Medicine (previously the Institute of Medicine): a systematic review and formal assessment of the quality of the evidence, and recommendations based on the highest quality evidence available, with the level of evidence stated for each recommendation.⁵ Assessing the level of evidence and determining the strengths of recommendations were done using the Grades of Recommendation Assessment, Development, and Evaluation (GRADE) system, which uses an approach similar to that of the Strength of Recommendation Taxonomy (SORT).

What, and what not, to focus on. The first set of recommendations states that everyone with, or at risk for, diabetes should receive structured, personalized, and ongoing nutritional advice from a dietician who is coordinated with their clinical care. Nutritional advice should focus on the quality and quantity of food, not on specific nutrients (fat and carbohydrates), since there is no good evidence on what proportion of such nutrients is optimal. And it should be tailored to the culture and eating preferences of the patient.

The type of diet with the strongest evidence base for preventing T2DM is a Mediterranean diet, which is supported by level-4, high-quality evidence. Important aspects of a Mediterranean diet are the regular consumption of nuts, whole grains, fruits, and vegetables; use of olive oil instead of butter; and favoring fish over red meat.⁶ Other dietary patterns associated with reduced risk but supported only by level-2, low-quality evidence, include Dietary Approaches to Stop Hypertension (DASH), vegetarian, vegan, and Nordic healthy diets. Moderate carbohydrate restriction is supported only by level-1, very low-quality evidence.

The UK guideline, too, recommends preferentially eating whole grains, fruits, and green leafy vegetables, as well as yogurt, cheese, tea, and coffee. And it advises reducing consumption of red processed meats, potatoes (especially French fries), sugar-sweetened beverages, and refined carbohydrates. However, these specific food preferences are supported only by low-level evidence. Both Diabetes UK and the ADA recommend minimizing consumption of free sugars and added fructose, in addition to sugar-sweetened drinks, but conclude that artificial sweeteners are safe and can be recommended. Both organizations also recommend against the use of vitamins and minerals to manage or prevent diabetes and against protein restriction for those with diabetic nephropathy.

Plant stanols and plant sterols are found in a variety of plant foods such as cereals, vegetable oils, seeds, and nuts, and are now being added to some food products. (For more on plant stanols and plant sterols.) They have a chemical structure similar to cholesterol and reduce the intestinal absorption of cholesterol, thereby lowering total serum cholesterol and LDL-cholesterol. Both Diabetes UK and the ADA recommend 2 to 3 grams of stanols/sterols per day.

Continue to: Alcohol intake

Alcohol intake. And what about alcohol intake in those with T2DM? Once again, both guidelines are in concert by stating that alcohol use in those with diabetes should be moderate, defined by the ADA as one or fewer drinks/d for women and 2 or fewer for men.

Weight loss and exercise are important, too. Those who are overweight or obese with T2DM can improve glycemic control with a 5% weight loss achieved by reducing caloric intake and by increasing energy expenditure with 150 minutes of moderate physical activity per week over at least 3 days.³ This recommendation is supported by high-quality evidence.

A 15-kg weight loss is recommended for those attempting diabetes remission (supported by moderate-level evidence).³ One small study in the United Kingdom found that more than half of those with T2DM could achieve remission with weight loss of 10 kg or more; 86% with weight loss of 15 kg or more.⁷ The Diabetes UK guideline panel rated this as having moderate-level evidence.

The bottom line. Diet and exercise are key interventions for the prevention and treatment of diabetes and can lead to remission if sufficient weight loss is achieved. To achieve and maintain an optimal diet, patients need individualized professional advice and followup. The evidence base for nutritional advice is growing and can be used to improve the quality of these patient-provider interactions.

I’d like to add several comments to Dr. John Hickner’s editorial, “We need to treat gun violence like an epidemic” (J Fam Pract. 2018;67:198).

First, when I hear gun rights advocates (GRAs) discuss the shooting deaths of innocent fellow Americans, I hear the same type of language that I hear from people with substance, tobacco, or alcohol use disorders. They blame everyone else rather than looking at themselves and what their love of weapons does to those around them. When confronted with ever-increasing numbers of gun-related deaths, they say, “Guns don’t kill people; people kill people.” But when used in the way they were intended to be used, guns—especially semi-automatic weapons—will maim or kill people.

And GRAs will not give up their gun rights—even if it means that their friends, neighbors, or relatives might die. GRAs ignore the unalienable rights described in the Declaration of Independence that guarantee “life, liberty, and the pursuit of happiness”—rights that are potentially annihilated when a person is the target of a loaded gun.

Because GRAs will not give up their weapons voluntarily, the only way to escape the continuously widening web of gun violence is to repeal the Second Amendment and allow local communities to vote either in favor or against gun control, setting whatever limits they deem appropriate on gun possession. Repealing the Second Amendment will not eliminate the killing of innocent Americans, but hopefully it will reduce the number of people who die from gun violence.

W.E. Feeman Jr., MD
Bowling Green, Ohio

I’d like to add several comments to Dr. John Hickner’s editorial, “We need to treat gun violence like an epidemic” (J Fam Pract. 2018;67:198).

First, when I hear gun rights advocates (GRAs) discuss the shooting deaths of innocent fellow Americans, I hear the same type of language that I hear from people with substance, tobacco, or alcohol use disorders. They blame everyone else rather than looking at themselves and what their love of weapons does to those around them. When confronted with ever-increasing numbers of gun-related deaths, they say, “Guns don’t kill people; people kill people.” But when used in the way they were intended to be used, guns—especially semi-automatic weapons—will maim or kill people.

And GRAs will not give up their gun rights—even if it means that their friends, neighbors, or relatives might die. GRAs ignore the unalienable rights described in the Declaration of Independence that guarantee “life, liberty, and the pursuit of happiness”—rights that are potentially annihilated when a person is the target of a loaded gun.

Because GRAs will not give up their weapons voluntarily, the only way to escape the continuously widening web of gun violence is to repeal the Second Amendment and allow local communities to vote either in favor or against gun control, setting whatever limits they deem appropriate on gun possession. Repealing the Second Amendment will not eliminate the killing of innocent Americans, but hopefully it will reduce the number of people who die from gun violence.

W.E. Feeman Jr., MD
Bowling Green, Ohio

I’d like to add several comments to Dr. John Hickner’s editorial, “We need to treat gun violence like an epidemic” (J Fam Pract. 2018;67:198).

First, when I hear gun rights advocates (GRAs) discuss the shooting deaths of innocent fellow Americans, I hear the same type of language that I hear from people with substance, tobacco, or alcohol use disorders. They blame everyone else rather than looking at themselves and what their love of weapons does to those around them. When confronted with ever-increasing numbers of gun-related deaths, they say, “Guns don’t kill people; people kill people.” But when used in the way they were intended to be used, guns—especially semi-automatic weapons—will maim or kill people.

And GRAs will not give up their gun rights—even if it means that their friends, neighbors, or relatives might die. GRAs ignore the unalienable rights described in the Declaration of Independence that guarantee “life, liberty, and the pursuit of happiness”—rights that are potentially annihilated when a person is the target of a loaded gun.

Because GRAs will not give up their weapons voluntarily, the only way to escape the continuously widening web of gun violence is to repeal the Second Amendment and allow local communities to vote either in favor or against gun control, setting whatever limits they deem appropriate on gun possession. Repealing the Second Amendment will not eliminate the killing of innocent Americans, but hopefully it will reduce the number of people who die from gun violence.

W.E. Feeman Jr., MD
Bowling Green, Ohio

In this issue of JFP, MacLaughlin and colleagues echo the recommendations of the 2017 American College of Cardiology/American Heart Association (ACC/AHA) guidelines on high blood pressure (BP).¹

This guideline, however, is not endorsed by primary care organizations. Both the American College of Physicians (ACP) and the American Academy of Family Physicians (AAFP) released their own evidence-based guideline in 2017.² (The European Society of Cardiology also declined to endorse the ACC/AHA guideline.³) So how do we make sense of the different recommendations? And how do we decide which guideline is most trustworthy?⁴

Evidence based vs evidence informed

Both guideline writing groups are highly respected and affiliated with influential organizations. Both claim their guidelines are based on scientific evidence and are crafted with the intention to improve health. The 2 guidelines, however, differ in their fidelity to the evidence-based process and in their willingness to generalize disease-centered interventions to non-diseased populations.

Evidence-based guidelines differ from evidence-informed guidelines.

Evidence-based guidelines differ from evidence-informed guidelines. Evidence-based guidelines have an established methodology that includes well-designed specific critical questions, a literature review with clearly defined inclusion and exclusion criteria, an evidence grading system, and a systematic approach to creating recommendations. Evidence-based guidelines are limited in scope and are often controversial because the evidence may not comport with the narrative promulgated by experts. Indeed, the controversy surrounding the 2014 Eighth Joint National Committee (JNC 8) guideline that I co-chaired focused on the one recommendation with the strongest evidence.^5,6

Comprehensive guidelines written by experts are by their very nature evidence-informed guidelines. The ACC/AHA guidelines are comprehensive, providing a panoply of recommendations. When such guidelines are written for primary care, the generalizability of specialized disease-centered knowledge is limited,⁷ and the risk of overdiagnosis and overtreatment rises,⁸ especially when the primary care community is not invited as equal partners in the guideline development process.

Trustworthy guidelines require management of conflicts of interests. A hidden contributor to guideline panel membership and content is organizational sponsorship. Advocacy organizations and specialty societies have governing boards that have fiduciary responsibilities to their organizations. Such responsibilities may supersede the responsibilities of guideline panel members and influence content. JNC 8’s appointed panel members chose to release the 2014 guideline independently, so as not to cede editorial authority to governing boards of associations with potential conflicts of interest.

As Paul Frame said, “An ounce of prevention is a ton of work.”⁹

Dr. Frame, a family medicine pioneer who applied evidence-based medicine to preventive practice, encouraged us to ask critical questions that must be supported by scientific evidence before implementing these practices in healthy populations.¹⁰ The ACC/AHA guidelines advocate recommendations based on untested assumptions: that improved health results from earlier “diagnosis” and disease labeling of individuals with risks (healthy patients), and that such patients should receive aggressive “prevention” with daily and lifelong medications requiring physician monitoring.¹¹ To support their new diagnostic standards, the authors cite similar relative risk (RR) reductions (an outcome-based measure), while discounting the smaller absolute risk (AR) reductions (a population-based measure) in studies supporting lower BP goals.

Continue to: Let's examine what this means

Let’s examine what this means

In 1967, a study of 143 hypertensive patients showed that treating high BP (average diastolic BP between 115 and 129 mm Hg) dramatically improved important health outcomes.¹² The number needed to treat (NNT) after about 1.5 years showed that for every 1.4 people treated, 1 benefited.⁸ This is strong and effective medicine.

We must all advocate for better guideline processes.

Successive randomized controlled trials of lower BP goals showed consistent RR reductions; however, AR reductions were much lower, reflecting a higher NNT.⁸ To prove BP-lowering benefits were not a random effect, higher numbers of participants were needed (SPRINT required over 9300 participants).¹³ The AR reduction in SPRINT was 1.6% (meaning no benefit was seen in 98.4% receiving the intensive intervention). One participant with high cardiovascular disease risk benefited for every 63 subjects given the intensive therapy compared with usual care (BP goal of 120 mm Hg vs 140 mm Hg).^13,14 The researchers noted serious harms in 1 of 22 subjects treated. Treating younger patients to lower BP goals labels healthy people with risk factors as “sick” and commits them to lifelong medications. It exposes them to more frequent harms than benefits. For healthy patients who are unlikely to benefit from taking more antihypertensive medication, these harms matter.

Interpreting the benefits of BP Tx when the benefit to individuals appears small

If only there were a biomarker that could tell us who is most likely to benefit from antihypertensive medication treatment, FPs could ensure that the correct patients are treated. The ACP/AAFP guideline points the way. There is a biomarker, and it is called BP. Systolic BP above 150 mm Hg signals urgency to treat with medications.

A call to advocate. We must all advocate for better guideline processes. The status quo in guideline development and its reliance on special interest funding requires ongoing vigilance to advocate on behalf of our patients. High-value medical care is expensive and hard work. When it is applied to the wrong people at the wrong time, we don’t deliver on our promises.

In this issue of JFP, MacLaughlin and colleagues echo the recommendations of the 2017 American College of Cardiology/American Heart Association (ACC/AHA) guidelines on high blood pressure (BP).¹

This guideline, however, is not endorsed by primary care organizations. Both the American College of Physicians (ACP) and the American Academy of Family Physicians (AAFP) released their own evidence-based guideline in 2017.² (The European Society of Cardiology also declined to endorse the ACC/AHA guideline.³) So how do we make sense of the different recommendations? And how do we decide which guideline is most trustworthy?⁴

Evidence based vs evidence informed

Both guideline writing groups are highly respected and affiliated with influential organizations. Both claim their guidelines are based on scientific evidence and are crafted with the intention to improve health. The 2 guidelines, however, differ in their fidelity to the evidence-based process and in their willingness to generalize disease-centered interventions to non-diseased populations.

Evidence-based guidelines differ from evidence-informed guidelines.

Evidence-based guidelines differ from evidence-informed guidelines. Evidence-based guidelines have an established methodology that includes well-designed specific critical questions, a literature review with clearly defined inclusion and exclusion criteria, an evidence grading system, and a systematic approach to creating recommendations. Evidence-based guidelines are limited in scope and are often controversial because the evidence may not comport with the narrative promulgated by experts. Indeed, the controversy surrounding the 2014 Eighth Joint National Committee (JNC 8) guideline that I co-chaired focused on the one recommendation with the strongest evidence.^5,6

Comprehensive guidelines written by experts are by their very nature evidence-informed guidelines. The ACC/AHA guidelines are comprehensive, providing a panoply of recommendations. When such guidelines are written for primary care, the generalizability of specialized disease-centered knowledge is limited,⁷ and the risk of overdiagnosis and overtreatment rises,⁸ especially when the primary care community is not invited as equal partners in the guideline development process.

Trustworthy guidelines require management of conflicts of interests. A hidden contributor to guideline panel membership and content is organizational sponsorship. Advocacy organizations and specialty societies have governing boards that have fiduciary responsibilities to their organizations. Such responsibilities may supersede the responsibilities of guideline panel members and influence content. JNC 8’s appointed panel members chose to release the 2014 guideline independently, so as not to cede editorial authority to governing boards of associations with potential conflicts of interest.

As Paul Frame said, “An ounce of prevention is a ton of work.”⁹

Dr. Frame, a family medicine pioneer who applied evidence-based medicine to preventive practice, encouraged us to ask critical questions that must be supported by scientific evidence before implementing these practices in healthy populations.¹⁰ The ACC/AHA guidelines advocate recommendations based on untested assumptions: that improved health results from earlier “diagnosis” and disease labeling of individuals with risks (healthy patients), and that such patients should receive aggressive “prevention” with daily and lifelong medications requiring physician monitoring.¹¹ To support their new diagnostic standards, the authors cite similar relative risk (RR) reductions (an outcome-based measure), while discounting the smaller absolute risk (AR) reductions (a population-based measure) in studies supporting lower BP goals.

Continue to: Let's examine what this means

Let’s examine what this means

In 1967, a study of 143 hypertensive patients showed that treating high BP (average diastolic BP between 115 and 129 mm Hg) dramatically improved important health outcomes.¹² The number needed to treat (NNT) after about 1.5 years showed that for every 1.4 people treated, 1 benefited.⁸ This is strong and effective medicine.

We must all advocate for better guideline processes.

Successive randomized controlled trials of lower BP goals showed consistent RR reductions; however, AR reductions were much lower, reflecting a higher NNT.⁸ To prove BP-lowering benefits were not a random effect, higher numbers of participants were needed (SPRINT required over 9300 participants).¹³ The AR reduction in SPRINT was 1.6% (meaning no benefit was seen in 98.4% receiving the intensive intervention). One participant with high cardiovascular disease risk benefited for every 63 subjects given the intensive therapy compared with usual care (BP goal of 120 mm Hg vs 140 mm Hg).^13,14 The researchers noted serious harms in 1 of 22 subjects treated. Treating younger patients to lower BP goals labels healthy people with risk factors as “sick” and commits them to lifelong medications. It exposes them to more frequent harms than benefits. For healthy patients who are unlikely to benefit from taking more antihypertensive medication, these harms matter.

Interpreting the benefits of BP Tx when the benefit to individuals appears small

If only there were a biomarker that could tell us who is most likely to benefit from antihypertensive medication treatment, FPs could ensure that the correct patients are treated. The ACP/AAFP guideline points the way. There is a biomarker, and it is called BP. Systolic BP above 150 mm Hg signals urgency to treat with medications.

A call to advocate. We must all advocate for better guideline processes. The status quo in guideline development and its reliance on special interest funding requires ongoing vigilance to advocate on behalf of our patients. High-value medical care is expensive and hard work. When it is applied to the wrong people at the wrong time, we don’t deliver on our promises.

In this issue of JFP, MacLaughlin and colleagues echo the recommendations of the 2017 American College of Cardiology/American Heart Association (ACC/AHA) guidelines on high blood pressure (BP).¹

This guideline, however, is not endorsed by primary care organizations. Both the American College of Physicians (ACP) and the American Academy of Family Physicians (AAFP) released their own evidence-based guideline in 2017.² (The European Society of Cardiology also declined to endorse the ACC/AHA guideline.³) So how do we make sense of the different recommendations? And how do we decide which guideline is most trustworthy?⁴

Evidence based vs evidence informed

Both guideline writing groups are highly respected and affiliated with influential organizations. Both claim their guidelines are based on scientific evidence and are crafted with the intention to improve health. The 2 guidelines, however, differ in their fidelity to the evidence-based process and in their willingness to generalize disease-centered interventions to non-diseased populations.

Evidence-based guidelines differ from evidence-informed guidelines.

Evidence-based guidelines differ from evidence-informed guidelines. Evidence-based guidelines have an established methodology that includes well-designed specific critical questions, a literature review with clearly defined inclusion and exclusion criteria, an evidence grading system, and a systematic approach to creating recommendations. Evidence-based guidelines are limited in scope and are often controversial because the evidence may not comport with the narrative promulgated by experts. Indeed, the controversy surrounding the 2014 Eighth Joint National Committee (JNC 8) guideline that I co-chaired focused on the one recommendation with the strongest evidence.^5,6

Comprehensive guidelines written by experts are by their very nature evidence-informed guidelines. The ACC/AHA guidelines are comprehensive, providing a panoply of recommendations. When such guidelines are written for primary care, the generalizability of specialized disease-centered knowledge is limited,⁷ and the risk of overdiagnosis and overtreatment rises,⁸ especially when the primary care community is not invited as equal partners in the guideline development process.

Trustworthy guidelines require management of conflicts of interests. A hidden contributor to guideline panel membership and content is organizational sponsorship. Advocacy organizations and specialty societies have governing boards that have fiduciary responsibilities to their organizations. Such responsibilities may supersede the responsibilities of guideline panel members and influence content. JNC 8’s appointed panel members chose to release the 2014 guideline independently, so as not to cede editorial authority to governing boards of associations with potential conflicts of interest.

As Paul Frame said, “An ounce of prevention is a ton of work.”⁹

Dr. Frame, a family medicine pioneer who applied evidence-based medicine to preventive practice, encouraged us to ask critical questions that must be supported by scientific evidence before implementing these practices in healthy populations.¹⁰ The ACC/AHA guidelines advocate recommendations based on untested assumptions: that improved health results from earlier “diagnosis” and disease labeling of individuals with risks (healthy patients), and that such patients should receive aggressive “prevention” with daily and lifelong medications requiring physician monitoring.¹¹ To support their new diagnostic standards, the authors cite similar relative risk (RR) reductions (an outcome-based measure), while discounting the smaller absolute risk (AR) reductions (a population-based measure) in studies supporting lower BP goals.

Continue to: Let's examine what this means

Let’s examine what this means

In 1967, a study of 143 hypertensive patients showed that treating high BP (average diastolic BP between 115 and 129 mm Hg) dramatically improved important health outcomes.¹² The number needed to treat (NNT) after about 1.5 years showed that for every 1.4 people treated, 1 benefited.⁸ This is strong and effective medicine.

We must all advocate for better guideline processes.

Successive randomized controlled trials of lower BP goals showed consistent RR reductions; however, AR reductions were much lower, reflecting a higher NNT.⁸ To prove BP-lowering benefits were not a random effect, higher numbers of participants were needed (SPRINT required over 9300 participants).¹³ The AR reduction in SPRINT was 1.6% (meaning no benefit was seen in 98.4% receiving the intensive intervention). One participant with high cardiovascular disease risk benefited for every 63 subjects given the intensive therapy compared with usual care (BP goal of 120 mm Hg vs 140 mm Hg).^13,14 The researchers noted serious harms in 1 of 22 subjects treated. Treating younger patients to lower BP goals labels healthy people with risk factors as “sick” and commits them to lifelong medications. It exposes them to more frequent harms than benefits. For healthy patients who are unlikely to benefit from taking more antihypertensive medication, these harms matter.

Interpreting the benefits of BP Tx when the benefit to individuals appears small

If only there were a biomarker that could tell us who is most likely to benefit from antihypertensive medication treatment, FPs could ensure that the correct patients are treated. The ACP/AAFP guideline points the way. There is a biomarker, and it is called BP. Systolic BP above 150 mm Hg signals urgency to treat with medications.

A call to advocate. We must all advocate for better guideline processes. The status quo in guideline development and its reliance on special interest funding requires ongoing vigilance to advocate on behalf of our patients. High-value medical care is expensive and hard work. When it is applied to the wrong people at the wrong time, we don’t deliver on our promises.

As members of the baby boomer generation (adults ≥65 years) age, the number of people at risk for diabetes increases. Already nearly one-quarter of people over age 65 have type 2 diabetes (T2DM).¹ With a proliferation of new medications to treat diabetes, deciding which ones to use in older patients is becoming complex.

In this article we review the important issues to consider when prescribing and monitoring diabetes medications in older adults. To provide optimal patient-centered care, it’s necessary to assess comorbid conditions as well as the costs, risks, and benefits of each medication. Determining appropriate goals of therapy and selecting agents that minimize the risk of hypoglycemia will help ensure safe and effective management of older patients with diabetes.

What makes elderly patients unique

The pathophysiology of T2DM in the elderly is unique in that it involves not just insulin resistance but also age-related loss of beta-cell function, leading to reduced insulin secretion and altered effectiveness of pharmacotherapy.² The addition of second and third medications may be needed for those with longstanding T2DM, although these agents often reduce the A1C level to a lesser extent than when used as monotherapy in patients whose beta-cell function is still intact. In addition to physiologic changes, older adults with diabetes have varied general health statuses and care support systems. The goal for glycemic management should be personalized based on an individual’s comorbidities and physical and cognitive functional status (TABLE 1^3,4).²

Higher A1C goals can be acceptable for elderly patients with comorbid conditions such as cognitive dysfunction, dementia, or cardiovascular or renal disease. Evaluate cognition when determining appropriate pharmacotherapy. Assess a patient’s awareness of hypoglycemia and ability to adhere to the regimen prescribed. Visual impairment, decreased dexterity, baseline weight, nutritional and functional status, as well as social support, finances, and formulary restrictions should all be considered when determining the most appropriate regimen for a patient. Also take into account patient and family goals of care.² TABLE 2^2-4 summarizes key risks and benefits of the medications we discuss next.

Metformin

Metformin is recommended as first-line therapy for those with T2DM for a number of reasons, including its potential to reduce cardiovascular events and mortality.^3,5 It also significantly reduces A1C levels by 1% to 1.5%,⁶ while imparting a low risk of hypoglycemia. Metformin is cost effective and well tolerated, making it an excellent choice for use in older patients.

The most common adverse effects are abdominal discomfort, diarrhea, and weight loss. The use of extended-release preparations, as well as slow titration of dosing, can improve gastrointestinal (GI) tolerance. Weight loss may be an attractive side effect in patients who are overweight or obese, but weight loss and diarrhea are concerning effects in frail older adults who may have poor nutritional reserves.⁶

Monitor renal function frequently in older patients receiving metformin.³ Renal failure is a risk factor for adverse events such as lactic acidosis, and metformin is therefore contraindicated in patients with an estimated glomerular filtration rate (eGFR) below 30 mL/min/1.73 m².⁴ With this in mind, metformin should not be started in patients with an eGFR below 45 mL/min/1.73 m². And for patients already taking metformin, reduce the total daily dose if the eGFR falls to between 30 and 45 mL/min/1.73 m².⁴

Metformin can cause a reduction in vitamin B12 levels after long-term use in up to 30% of patients, likely due to decreased absorption from the ileum.⁷ Monitor vitamin B12 serum concentrations periodically with long-term therapy, particularly in patients with peripheral neuropathy or anemia, as these conditions may be exacerbated by vitamin B12 deficiency.^3,4

Continue to: Sulfonylureas

Sulfonylureas

Sulfonylureas increase the secretion of insulin from pancreatic beta cells, significantly lower blood glucose, and reduce A1C levels by 1% to 2%.⁶ Because hypoglycemia is a serious risk with sulfonylureas, they should be used conservatively in the elderly.² Avoid using sulfonylurea formulations with long half-lives or active metabolites, which can cause severe and prolonged hypoglycemia.^8,9

Glyburide is broken down into active metabolites that accumulate in patients who have renal insufficiency; it should be avoided in older adults due to the risk of life-threatening hypoglycemic events.¹⁰ Glipizide has no active metabolites and has the lowest risk of hypoglycemia in the setting of decreased renal function, making it the preferred sulfonylurea for use in the elderly.^3,10

Thiazolidinediones

Thiazolidinediones (TZDs) reduce insulin resistance and decrease hepatic glucose production without increasing the risk of hypoglycemia. These agents effectively lower A1C levels by 1% to 1.5%.¹¹ Despite their efficacy, TZDs have limited benefit because of adverse effects. Serious complications include fluid retention that can exacerbate or lead to worsening heart failure, weight gain, macular edema, and hepatic failure.

Avoid using sulfonylureas with long half-lives or active metabolites, which can cause severe and prolonged hypoglycemia in the elderly.

Specifically, with pioglitazone, there is also a slightly increased risk of bladder cancer.² In one study involving more than 30,000 patients taking pioglitazone, an increase in bladder cancer was noted among those using the medication for more than 2 years.¹² Still, the hazard ratio was only 1.2, with 90 cases diagnosed over the course of the study. A prudent strategy would be to avoid its use in those with high risk of developing bladder cancer. TZDs are contraindicated in patients with New York Heart Association class III or IV heart failure.⁸

Increased fracture risk has been identified in both men and women and is a concerning adverse effect in the elderly.⁸ Fracture risk with TZDs has been approximately twice that of placebo, noted in a study of older women where the fracture rate was 5.1% vs 2.5%, respectively.¹¹ TZDs can be of value in lowering A1C levels without the risk of hypoglycemia. But, due to their adverse effect profile, use TZDs cautiously in older adults at risk for heart failure, falls, or fractures.³

Continue to: DPP-4 inhibitors

Dipeptidyl peptidase-4 (DPP-4) inhibitors work by suppressing the enzyme that degrades 2 incretin hormones, glucagon-like peptide 1 (GLP-1) and gastric inhibitory polypeptide (GIP). The resulting enhancement of incretin activity increases glucose-dependent insulin secretion, decreases glucagon secretion, and promotes satiety.⁶ These agents have modest efficacy with the potential to lower A1C by 0.5% to 0.9%.^8,13 Studies show that DPP-4 inhibitors are well tolerated with a minimal risk of hypoglycemia in the elderly.¹³ These agents are ideal for combin­ation therapy or for monotherapy in older patients who are not good candidates for metformin or a sulfonylurea.

The safety profile, neutral effect on weight, and once-daily dosing make these agents advantageous for use in frail and debilitated elderly patients, as well as in patients with cognitive dysfunction, decreased dexterity, inconsistent meal patterns, or adherence issues. Dose adjustment is required in renal impairment, with the exception of linagliptin. High cost or formulary restrictions may impact use of these agents.

The DPP-4 inhibitors were well tolerated in short-term studies, but long-term safety has yet to be established.⁶ Reported post-marketing adverse effects include acute renal failure, allergic reactions, and acute pancreatitis.^6,14 These agents should be avoided in any patient with a history of pancreatitis.¹⁴ In addition, trials investigating the cardiovascular safety and efficacy of DPP-4 inhibitors point to an increased risk of heart failure with the use of saxagliptin and alogliptin, regardless of age.^15,16 The potential for adverse effects warrants increased patient monitoring when using these agents in older patients.

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are injectable agents that potentiate the actions of the naturally occurring incretin GLP-1, which increases glucose-dependent insulin secretion, inhibits glucagon release, reduces hepatic glucose production, and delays gastric emptying. These agents have a pronounced effect on satiety and promote weight loss. The most common adverse effects are nausea, vomiting, and diarrhea, which occur most commonly during treatment initiation and titration. Studies in elderly patients confirm A1C reductions of 1% to 1.5% and a low risk of hypoglycemia when used alone.^17,18

GLP-1 RAs can be used as monotherapy in older patients at risk for hypoglycemia or in those with hypoglycemic unawareness. They can also be used in combination therapy with other agents, including insulin, though concomitant use with insulin or insulin secretagogues increases the risk of hypoglycemia.³ Weight loss and GI adverse effects may limit the use of these agents in frail or undernourished elderly patients.⁶

Continue to: Since these agents are injected...

Since these agents are injected, they require intact visual, motor, and cognitive skills and thus may not be appropriate in older patients with cognitive or visual impairment or decreased dexterity. In addition, the high cost of these agents may limit their use.

Select a GLP-1 RA based on the frequency of administration, type of glucose control required (fasting or post-prandial), and the patient’s ability to use the administration device. Dose adjustment is required in renal impairment, except with dulaglutide and liraglutide. Use with caution in patients with a history of pancreatitis, and stop GLP-1 RAs if pancreatitis is suspected during treatment.⁴ Avoid GLP-1 RAs in patients with a personal or family history of thyroid-related cancers, as these agents have been associated with medullary thyroid tumors in animals.⁴

A new indication. Recent evidence suggests the GLP-1 RAs may offer additional cardiovascular benefit in patients with diabetes.^18,19 In August 2017, liraglutide gained an additional FDA indication to reduce the risk of major adverse cardiovascular events in adults with T2DM and established cardiovascular disease.

This new indication was based on the Novo Nordisk- and National Institutes of Health-sponsored LEADER trial, in which liraglutide reduced the risk of cardiovascular death, nonfatal heart attack, or nonfatal stroke by 13% vs placebo (P=.01) with an absolute risk reduction (ARR) of 1.9%.¹⁹ Liraglutide demonstrated a 22% reduction in cardiovascular death and a 15% reduction in all-cause death (ARR 1.3%, 1.4% respectively).¹⁹ The new cardiovascular indication may impact the choice of add-on therapy to metformin in patients with preexisting cardiovascular conditions.

Continue to: Sodium glucose cotransporter-2 inhibitors

Sodium glucose cotransporter-2 inhibitors

SGLT-2 inhibitors prevent the reabsorption of renal-filtered glucose, resulting in decreased blood glucose levels and increased urinary excretion of glucose without stimulating insulin secretion, and therefore without increasing the risk of hypoglycemia. Additional effects include decreased blood pressure and weight loss.²⁰ Dose adjustment is required in renal impairment.

SGLT-2 inhibitors can be used as monotherapy or in combination with other agents, including insulin, and the relatively low risk of hypoglycemia and moderate A1C lowering potential of 0.5% to 1% provide an oral option for select older patients.²⁰ Common adverse events include hypotension, hyperkalemia, increased low-density lipoprotein (LDL) levels, acute kidney injury, genital mycotic infections, and hypoglycemia when used in combination with insulin or insulin secretagogues.²⁰

DPP-4 agents are ideal for older patients who are not candidates for metformin or a sulfonylurea.

Additional warnings have been issued by the FDA for the risk of urinary tract infection with sepsis, as well as diabetic ketoacidosis associated with SGLT-2 inhibitor use.²¹ The FDA has reported bone fracture risk and decreased bone mineral density with canagliflozin.²¹ Avoid using SGLT-2 inhibitors in patients with osteopenia or osteoporosis, as the risks outweigh the benefits. Drug-specific warnings may further impact individual use of an agent, with canagliflozin most recently having been associated with increased risk of leg and foot amputations.²¹

Given the adverse effect profile of SGLT-2 inhibitors, assess their risks and benefits in older patients on a case-by-case basis. Before initiating therapy, evaluate each patient’s volume status. A higher incidence of adverse effects related to intravascular volume depletion has been reported in those 65 or older, with a more prominent increase seen in patients 75 or older.²² However, the risk of hypoglycemia does not seem to increase with age.²²

Although many adverse effects have been reported with SGLT-2 inhibitors, empagliflozin was associated with significantly lower rates of all-cause and cardiovascular death and lower risk of hospitalization for heart failure in the only SGLT-2 inhibitor cardiovascular outcomes trial reported to date.²³ If this cardiovascular benefit is replicated in additional trials of the other SGLT-2 inhibitors, use of this drug class may increase.

Continue to: Insulin

Many patients will ultimately require insulin due to the progressive loss of beta-cell function that occurs in advanced diabetes. Starting insulin therapy early on in the disease may actually restore beta-cell function and reduce glucotoxicity.²⁴ In elderly patients with uncontrolled diabetes, early treatment with basal insulin results in better glycemic control and less hypoglycemia than continuing to titrate oral agents.²⁵

Despite these benefits, however, insulin use often is not optimized in the elderly due to concerns about hypoglycemia and difficulty of administration. Safe use of insulin requires careful selection of an appropriate insulin regimen, since insulin use has been identified as an independent predictor of severe hypoglycemia in the elderly.^8,26 Before initiating insulin therapy, evaluate whether an older patient is cognitively and physically able to safely use insulin.

Glipizide has no active metabolites and has the lowest risk of hypoglycemia in the setting of decreased renal function, making it the preferred sulfonylurea for use in the elderly.

Multiple daily injections may be challenging for some older adults. Limit such insulin regimens to use in high-functioning patients. Although all types of insulin can cause hypoglycemia, regimens that mimic insulin’s normal physiologic pattern introduce less hypoglycemic risk. Using basal insulin that mimics the body’s sustained insulin level throughout the day is associated with a lower frequency of hypoglycemia in older people with diabetes than conventional insulin regimens. Long-acting insulins such as glargine, detemir, and degludec offer a lower risk of hypoglycemia, particularly nocturnal hypoglycemia which may contribute to falls.^2,27

Neutral protamine Hagedorn insulin and regular insulin are not recommended for use in the elderly, as they do not mimic the body’s natural basal-bolus insulin production and thus put patients at higher risk of hypoglycemia.⁴ If insulin intensification is needed after optimizing basal insulin, consider adding mealtime insulin with a bolus of rapid-acting insulin (insulin aspart, insulin lispro, or insulin glulisine). It is important to note that the kidneys are responsible for 30% to 80% of insulin clearance from the body.²⁸ Because insulin action is prolonged in renal insufficiency, prevent hypoglycemia by decreasing basal and bolus doses when the eGFR is below 50 mL/min/1.73m².²⁸

Dosing errors. Whenever possible, use insulin preparations that minimize dosing errors. Insulin pen formulations, if financially feasible, allow more accurate dosing and are more acceptable to older patients compared with syringes and vials.²⁹ Pen formulations are particularly preferable for older patients with impaired vision or dexterity.²⁹ In addition, when patients must mix insulins, errors are more likely to occur. The use of premixed insulin vials has been shown to increase dosing accuracy when used by the elderly.³⁰

Continue to: Combining antidiabetes agents

Combining antidiabetes agents

Combination therapy is often needed as T2DM progresses, even though strict glucose control is generally not recommended for the elderly.³¹ The American Geriatrics Society advises avoiding additional medications other than metformin to achieve an A1C level below 7.5% in most older adults.

However, for older patients already taking metformin who are not at their A1C goal, consider adding a second agent, if not contraindicated. Potential agents include a GLP-1 RA, SGLT-2 inhibitor, DDP-4 inhibitor, or short-acting sulfonylurea (glipizide). Alternatively, basal insulin may be added. However, avoid combining a sulfonylurea with insulin, which greatly increases the risk of hypoglycemia.³² Consider adding a GLP-1 RA or basal insulin if the patient is not at his/her target A1C on oral therapy with multiple agents.³

CORRESPONDENCE
Barbara Keber, MD, Glen Cove Hospital, 101 St. Andrews Lane, Glen Cove, NY; [email protected].

As members of the baby boomer generation (adults ≥65 years) age, the number of people at risk for diabetes increases. Already nearly one-quarter of people over age 65 have type 2 diabetes (T2DM).¹ With a proliferation of new medications to treat diabetes, deciding which ones to use in older patients is becoming complex.

In this article we review the important issues to consider when prescribing and monitoring diabetes medications in older adults. To provide optimal patient-centered care, it’s necessary to assess comorbid conditions as well as the costs, risks, and benefits of each medication. Determining appropriate goals of therapy and selecting agents that minimize the risk of hypoglycemia will help ensure safe and effective management of older patients with diabetes.

What makes elderly patients unique

The pathophysiology of T2DM in the elderly is unique in that it involves not just insulin resistance but also age-related loss of beta-cell function, leading to reduced insulin secretion and altered effectiveness of pharmacotherapy.² The addition of second and third medications may be needed for those with longstanding T2DM, although these agents often reduce the A1C level to a lesser extent than when used as monotherapy in patients whose beta-cell function is still intact. In addition to physiologic changes, older adults with diabetes have varied general health statuses and care support systems. The goal for glycemic management should be personalized based on an individual’s comorbidities and physical and cognitive functional status (TABLE 1^3,4).²

Higher A1C goals can be acceptable for elderly patients with comorbid conditions such as cognitive dysfunction, dementia, or cardiovascular or renal disease. Evaluate cognition when determining appropriate pharmacotherapy. Assess a patient’s awareness of hypoglycemia and ability to adhere to the regimen prescribed. Visual impairment, decreased dexterity, baseline weight, nutritional and functional status, as well as social support, finances, and formulary restrictions should all be considered when determining the most appropriate regimen for a patient. Also take into account patient and family goals of care.² TABLE 2^2-4 summarizes key risks and benefits of the medications we discuss next.

Metformin

Metformin is recommended as first-line therapy for those with T2DM for a number of reasons, including its potential to reduce cardiovascular events and mortality.^3,5 It also significantly reduces A1C levels by 1% to 1.5%,⁶ while imparting a low risk of hypoglycemia. Metformin is cost effective and well tolerated, making it an excellent choice for use in older patients.

The most common adverse effects are abdominal discomfort, diarrhea, and weight loss. The use of extended-release preparations, as well as slow titration of dosing, can improve gastrointestinal (GI) tolerance. Weight loss may be an attractive side effect in patients who are overweight or obese, but weight loss and diarrhea are concerning effects in frail older adults who may have poor nutritional reserves.⁶

Monitor renal function frequently in older patients receiving metformin.³ Renal failure is a risk factor for adverse events such as lactic acidosis, and metformin is therefore contraindicated in patients with an estimated glomerular filtration rate (eGFR) below 30 mL/min/1.73 m².⁴ With this in mind, metformin should not be started in patients with an eGFR below 45 mL/min/1.73 m². And for patients already taking metformin, reduce the total daily dose if the eGFR falls to between 30 and 45 mL/min/1.73 m².⁴

Metformin can cause a reduction in vitamin B12 levels after long-term use in up to 30% of patients, likely due to decreased absorption from the ileum.⁷ Monitor vitamin B12 serum concentrations periodically with long-term therapy, particularly in patients with peripheral neuropathy or anemia, as these conditions may be exacerbated by vitamin B12 deficiency.^3,4

Continue to: Sulfonylureas

Sulfonylureas

Sulfonylureas increase the secretion of insulin from pancreatic beta cells, significantly lower blood glucose, and reduce A1C levels by 1% to 2%.⁶ Because hypoglycemia is a serious risk with sulfonylureas, they should be used conservatively in the elderly.² Avoid using sulfonylurea formulations with long half-lives or active metabolites, which can cause severe and prolonged hypoglycemia.^8,9

Glyburide is broken down into active metabolites that accumulate in patients who have renal insufficiency; it should be avoided in older adults due to the risk of life-threatening hypoglycemic events.¹⁰ Glipizide has no active metabolites and has the lowest risk of hypoglycemia in the setting of decreased renal function, making it the preferred sulfonylurea for use in the elderly.^3,10

Thiazolidinediones

Thiazolidinediones (TZDs) reduce insulin resistance and decrease hepatic glucose production without increasing the risk of hypoglycemia. These agents effectively lower A1C levels by 1% to 1.5%.¹¹ Despite their efficacy, TZDs have limited benefit because of adverse effects. Serious complications include fluid retention that can exacerbate or lead to worsening heart failure, weight gain, macular edema, and hepatic failure.

Avoid using sulfonylureas with long half-lives or active metabolites, which can cause severe and prolonged hypoglycemia in the elderly.

Specifically, with pioglitazone, there is also a slightly increased risk of bladder cancer.² In one study involving more than 30,000 patients taking pioglitazone, an increase in bladder cancer was noted among those using the medication for more than 2 years.¹² Still, the hazard ratio was only 1.2, with 90 cases diagnosed over the course of the study. A prudent strategy would be to avoid its use in those with high risk of developing bladder cancer. TZDs are contraindicated in patients with New York Heart Association class III or IV heart failure.⁸

Increased fracture risk has been identified in both men and women and is a concerning adverse effect in the elderly.⁸ Fracture risk with TZDs has been approximately twice that of placebo, noted in a study of older women where the fracture rate was 5.1% vs 2.5%, respectively.¹¹ TZDs can be of value in lowering A1C levels without the risk of hypoglycemia. But, due to their adverse effect profile, use TZDs cautiously in older adults at risk for heart failure, falls, or fractures.³

Continue to: DPP-4 inhibitors

Dipeptidyl peptidase-4 (DPP-4) inhibitors work by suppressing the enzyme that degrades 2 incretin hormones, glucagon-like peptide 1 (GLP-1) and gastric inhibitory polypeptide (GIP). The resulting enhancement of incretin activity increases glucose-dependent insulin secretion, decreases glucagon secretion, and promotes satiety.⁶ These agents have modest efficacy with the potential to lower A1C by 0.5% to 0.9%.^8,13 Studies show that DPP-4 inhibitors are well tolerated with a minimal risk of hypoglycemia in the elderly.¹³ These agents are ideal for combin­ation therapy or for monotherapy in older patients who are not good candidates for metformin or a sulfonylurea.

The safety profile, neutral effect on weight, and once-daily dosing make these agents advantageous for use in frail and debilitated elderly patients, as well as in patients with cognitive dysfunction, decreased dexterity, inconsistent meal patterns, or adherence issues. Dose adjustment is required in renal impairment, with the exception of linagliptin. High cost or formulary restrictions may impact use of these agents.

The DPP-4 inhibitors were well tolerated in short-term studies, but long-term safety has yet to be established.⁶ Reported post-marketing adverse effects include acute renal failure, allergic reactions, and acute pancreatitis.^6,14 These agents should be avoided in any patient with a history of pancreatitis.¹⁴ In addition, trials investigating the cardiovascular safety and efficacy of DPP-4 inhibitors point to an increased risk of heart failure with the use of saxagliptin and alogliptin, regardless of age.^15,16 The potential for adverse effects warrants increased patient monitoring when using these agents in older patients.

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are injectable agents that potentiate the actions of the naturally occurring incretin GLP-1, which increases glucose-dependent insulin secretion, inhibits glucagon release, reduces hepatic glucose production, and delays gastric emptying. These agents have a pronounced effect on satiety and promote weight loss. The most common adverse effects are nausea, vomiting, and diarrhea, which occur most commonly during treatment initiation and titration. Studies in elderly patients confirm A1C reductions of 1% to 1.5% and a low risk of hypoglycemia when used alone.^17,18

GLP-1 RAs can be used as monotherapy in older patients at risk for hypoglycemia or in those with hypoglycemic unawareness. They can also be used in combination therapy with other agents, including insulin, though concomitant use with insulin or insulin secretagogues increases the risk of hypoglycemia.³ Weight loss and GI adverse effects may limit the use of these agents in frail or undernourished elderly patients.⁶

Continue to: Since these agents are injected...

Since these agents are injected, they require intact visual, motor, and cognitive skills and thus may not be appropriate in older patients with cognitive or visual impairment or decreased dexterity. In addition, the high cost of these agents may limit their use.

Select a GLP-1 RA based on the frequency of administration, type of glucose control required (fasting or post-prandial), and the patient’s ability to use the administration device. Dose adjustment is required in renal impairment, except with dulaglutide and liraglutide. Use with caution in patients with a history of pancreatitis, and stop GLP-1 RAs if pancreatitis is suspected during treatment.⁴ Avoid GLP-1 RAs in patients with a personal or family history of thyroid-related cancers, as these agents have been associated with medullary thyroid tumors in animals.⁴

A new indication. Recent evidence suggests the GLP-1 RAs may offer additional cardiovascular benefit in patients with diabetes.^18,19 In August 2017, liraglutide gained an additional FDA indication to reduce the risk of major adverse cardiovascular events in adults with T2DM and established cardiovascular disease.

This new indication was based on the Novo Nordisk- and National Institutes of Health-sponsored LEADER trial, in which liraglutide reduced the risk of cardiovascular death, nonfatal heart attack, or nonfatal stroke by 13% vs placebo (P=.01) with an absolute risk reduction (ARR) of 1.9%.¹⁹ Liraglutide demonstrated a 22% reduction in cardiovascular death and a 15% reduction in all-cause death (ARR 1.3%, 1.4% respectively).¹⁹ The new cardiovascular indication may impact the choice of add-on therapy to metformin in patients with preexisting cardiovascular conditions.

Continue to: Sodium glucose cotransporter-2 inhibitors

Sodium glucose cotransporter-2 inhibitors

SGLT-2 inhibitors prevent the reabsorption of renal-filtered glucose, resulting in decreased blood glucose levels and increased urinary excretion of glucose without stimulating insulin secretion, and therefore without increasing the risk of hypoglycemia. Additional effects include decreased blood pressure and weight loss.²⁰ Dose adjustment is required in renal impairment.

SGLT-2 inhibitors can be used as monotherapy or in combination with other agents, including insulin, and the relatively low risk of hypoglycemia and moderate A1C lowering potential of 0.5% to 1% provide an oral option for select older patients.²⁰ Common adverse events include hypotension, hyperkalemia, increased low-density lipoprotein (LDL) levels, acute kidney injury, genital mycotic infections, and hypoglycemia when used in combination with insulin or insulin secretagogues.²⁰

DPP-4 agents are ideal for older patients who are not candidates for metformin or a sulfonylurea.

Additional warnings have been issued by the FDA for the risk of urinary tract infection with sepsis, as well as diabetic ketoacidosis associated with SGLT-2 inhibitor use.²¹ The FDA has reported bone fracture risk and decreased bone mineral density with canagliflozin.²¹ Avoid using SGLT-2 inhibitors in patients with osteopenia or osteoporosis, as the risks outweigh the benefits. Drug-specific warnings may further impact individual use of an agent, with canagliflozin most recently having been associated with increased risk of leg and foot amputations.²¹

Given the adverse effect profile of SGLT-2 inhibitors, assess their risks and benefits in older patients on a case-by-case basis. Before initiating therapy, evaluate each patient’s volume status. A higher incidence of adverse effects related to intravascular volume depletion has been reported in those 65 or older, with a more prominent increase seen in patients 75 or older.²² However, the risk of hypoglycemia does not seem to increase with age.²²

Although many adverse effects have been reported with SGLT-2 inhibitors, empagliflozin was associated with significantly lower rates of all-cause and cardiovascular death and lower risk of hospitalization for heart failure in the only SGLT-2 inhibitor cardiovascular outcomes trial reported to date.²³ If this cardiovascular benefit is replicated in additional trials of the other SGLT-2 inhibitors, use of this drug class may increase.

Continue to: Insulin

Many patients will ultimately require insulin due to the progressive loss of beta-cell function that occurs in advanced diabetes. Starting insulin therapy early on in the disease may actually restore beta-cell function and reduce glucotoxicity.²⁴ In elderly patients with uncontrolled diabetes, early treatment with basal insulin results in better glycemic control and less hypoglycemia than continuing to titrate oral agents.²⁵

Despite these benefits, however, insulin use often is not optimized in the elderly due to concerns about hypoglycemia and difficulty of administration. Safe use of insulin requires careful selection of an appropriate insulin regimen, since insulin use has been identified as an independent predictor of severe hypoglycemia in the elderly.^8,26 Before initiating insulin therapy, evaluate whether an older patient is cognitively and physically able to safely use insulin.

Glipizide has no active metabolites and has the lowest risk of hypoglycemia in the setting of decreased renal function, making it the preferred sulfonylurea for use in the elderly.

Multiple daily injections may be challenging for some older adults. Limit such insulin regimens to use in high-functioning patients. Although all types of insulin can cause hypoglycemia, regimens that mimic insulin’s normal physiologic pattern introduce less hypoglycemic risk. Using basal insulin that mimics the body’s sustained insulin level throughout the day is associated with a lower frequency of hypoglycemia in older people with diabetes than conventional insulin regimens. Long-acting insulins such as glargine, detemir, and degludec offer a lower risk of hypoglycemia, particularly nocturnal hypoglycemia which may contribute to falls.^2,27

Neutral protamine Hagedorn insulin and regular insulin are not recommended for use in the elderly, as they do not mimic the body’s natural basal-bolus insulin production and thus put patients at higher risk of hypoglycemia.⁴ If insulin intensification is needed after optimizing basal insulin, consider adding mealtime insulin with a bolus of rapid-acting insulin (insulin aspart, insulin lispro, or insulin glulisine). It is important to note that the kidneys are responsible for 30% to 80% of insulin clearance from the body.²⁸ Because insulin action is prolonged in renal insufficiency, prevent hypoglycemia by decreasing basal and bolus doses when the eGFR is below 50 mL/min/1.73m².²⁸

Dosing errors. Whenever possible, use insulin preparations that minimize dosing errors. Insulin pen formulations, if financially feasible, allow more accurate dosing and are more acceptable to older patients compared with syringes and vials.²⁹ Pen formulations are particularly preferable for older patients with impaired vision or dexterity.²⁹ In addition, when patients must mix insulins, errors are more likely to occur. The use of premixed insulin vials has been shown to increase dosing accuracy when used by the elderly.³⁰

Continue to: Combining antidiabetes agents

Combining antidiabetes agents

Combination therapy is often needed as T2DM progresses, even though strict glucose control is generally not recommended for the elderly.³¹ The American Geriatrics Society advises avoiding additional medications other than metformin to achieve an A1C level below 7.5% in most older adults.

However, for older patients already taking metformin who are not at their A1C goal, consider adding a second agent, if not contraindicated. Potential agents include a GLP-1 RA, SGLT-2 inhibitor, DDP-4 inhibitor, or short-acting sulfonylurea (glipizide). Alternatively, basal insulin may be added. However, avoid combining a sulfonylurea with insulin, which greatly increases the risk of hypoglycemia.³² Consider adding a GLP-1 RA or basal insulin if the patient is not at his/her target A1C on oral therapy with multiple agents.³

CORRESPONDENCE
Barbara Keber, MD, Glen Cove Hospital, 101 St. Andrews Lane, Glen Cove, NY; [email protected].

As members of the baby boomer generation (adults ≥65 years) age, the number of people at risk for diabetes increases. Already nearly one-quarter of people over age 65 have type 2 diabetes (T2DM).¹ With a proliferation of new medications to treat diabetes, deciding which ones to use in older patients is becoming complex.

In this article we review the important issues to consider when prescribing and monitoring diabetes medications in older adults. To provide optimal patient-centered care, it’s necessary to assess comorbid conditions as well as the costs, risks, and benefits of each medication. Determining appropriate goals of therapy and selecting agents that minimize the risk of hypoglycemia will help ensure safe and effective management of older patients with diabetes.

What makes elderly patients unique

The pathophysiology of T2DM in the elderly is unique in that it involves not just insulin resistance but also age-related loss of beta-cell function, leading to reduced insulin secretion and altered effectiveness of pharmacotherapy.² The addition of second and third medications may be needed for those with longstanding T2DM, although these agents often reduce the A1C level to a lesser extent than when used as monotherapy in patients whose beta-cell function is still intact. In addition to physiologic changes, older adults with diabetes have varied general health statuses and care support systems. The goal for glycemic management should be personalized based on an individual’s comorbidities and physical and cognitive functional status (TABLE 1^3,4).²

Higher A1C goals can be acceptable for elderly patients with comorbid conditions such as cognitive dysfunction, dementia, or cardiovascular or renal disease. Evaluate cognition when determining appropriate pharmacotherapy. Assess a patient’s awareness of hypoglycemia and ability to adhere to the regimen prescribed. Visual impairment, decreased dexterity, baseline weight, nutritional and functional status, as well as social support, finances, and formulary restrictions should all be considered when determining the most appropriate regimen for a patient. Also take into account patient and family goals of care.² TABLE 2^2-4 summarizes key risks and benefits of the medications we discuss next.

Metformin

Metformin is recommended as first-line therapy for those with T2DM for a number of reasons, including its potential to reduce cardiovascular events and mortality.^3,5 It also significantly reduces A1C levels by 1% to 1.5%,⁶ while imparting a low risk of hypoglycemia. Metformin is cost effective and well tolerated, making it an excellent choice for use in older patients.

The most common adverse effects are abdominal discomfort, diarrhea, and weight loss. The use of extended-release preparations, as well as slow titration of dosing, can improve gastrointestinal (GI) tolerance. Weight loss may be an attractive side effect in patients who are overweight or obese, but weight loss and diarrhea are concerning effects in frail older adults who may have poor nutritional reserves.⁶

Monitor renal function frequently in older patients receiving metformin.³ Renal failure is a risk factor for adverse events such as lactic acidosis, and metformin is therefore contraindicated in patients with an estimated glomerular filtration rate (eGFR) below 30 mL/min/1.73 m².⁴ With this in mind, metformin should not be started in patients with an eGFR below 45 mL/min/1.73 m². And for patients already taking metformin, reduce the total daily dose if the eGFR falls to between 30 and 45 mL/min/1.73 m².⁴

Metformin can cause a reduction in vitamin B12 levels after long-term use in up to 30% of patients, likely due to decreased absorption from the ileum.⁷ Monitor vitamin B12 serum concentrations periodically with long-term therapy, particularly in patients with peripheral neuropathy or anemia, as these conditions may be exacerbated by vitamin B12 deficiency.^3,4

Continue to: Sulfonylureas

Sulfonylureas

Sulfonylureas increase the secretion of insulin from pancreatic beta cells, significantly lower blood glucose, and reduce A1C levels by 1% to 2%.⁶ Because hypoglycemia is a serious risk with sulfonylureas, they should be used conservatively in the elderly.² Avoid using sulfonylurea formulations with long half-lives or active metabolites, which can cause severe and prolonged hypoglycemia.^8,9

Glyburide is broken down into active metabolites that accumulate in patients who have renal insufficiency; it should be avoided in older adults due to the risk of life-threatening hypoglycemic events.¹⁰ Glipizide has no active metabolites and has the lowest risk of hypoglycemia in the setting of decreased renal function, making it the preferred sulfonylurea for use in the elderly.^3,10

Thiazolidinediones

Thiazolidinediones (TZDs) reduce insulin resistance and decrease hepatic glucose production without increasing the risk of hypoglycemia. These agents effectively lower A1C levels by 1% to 1.5%.¹¹ Despite their efficacy, TZDs have limited benefit because of adverse effects. Serious complications include fluid retention that can exacerbate or lead to worsening heart failure, weight gain, macular edema, and hepatic failure.

Avoid using sulfonylureas with long half-lives or active metabolites, which can cause severe and prolonged hypoglycemia in the elderly.

Specifically, with pioglitazone, there is also a slightly increased risk of bladder cancer.² In one study involving more than 30,000 patients taking pioglitazone, an increase in bladder cancer was noted among those using the medication for more than 2 years.¹² Still, the hazard ratio was only 1.2, with 90 cases diagnosed over the course of the study. A prudent strategy would be to avoid its use in those with high risk of developing bladder cancer. TZDs are contraindicated in patients with New York Heart Association class III or IV heart failure.⁸

Increased fracture risk has been identified in both men and women and is a concerning adverse effect in the elderly.⁸ Fracture risk with TZDs has been approximately twice that of placebo, noted in a study of older women where the fracture rate was 5.1% vs 2.5%, respectively.¹¹ TZDs can be of value in lowering A1C levels without the risk of hypoglycemia. But, due to their adverse effect profile, use TZDs cautiously in older adults at risk for heart failure, falls, or fractures.³

Continue to: DPP-4 inhibitors

Dipeptidyl peptidase-4 (DPP-4) inhibitors work by suppressing the enzyme that degrades 2 incretin hormones, glucagon-like peptide 1 (GLP-1) and gastric inhibitory polypeptide (GIP). The resulting enhancement of incretin activity increases glucose-dependent insulin secretion, decreases glucagon secretion, and promotes satiety.⁶ These agents have modest efficacy with the potential to lower A1C by 0.5% to 0.9%.^8,13 Studies show that DPP-4 inhibitors are well tolerated with a minimal risk of hypoglycemia in the elderly.¹³ These agents are ideal for combin­ation therapy or for monotherapy in older patients who are not good candidates for metformin or a sulfonylurea.

The safety profile, neutral effect on weight, and once-daily dosing make these agents advantageous for use in frail and debilitated elderly patients, as well as in patients with cognitive dysfunction, decreased dexterity, inconsistent meal patterns, or adherence issues. Dose adjustment is required in renal impairment, with the exception of linagliptin. High cost or formulary restrictions may impact use of these agents.

The DPP-4 inhibitors were well tolerated in short-term studies, but long-term safety has yet to be established.⁶ Reported post-marketing adverse effects include acute renal failure, allergic reactions, and acute pancreatitis.^6,14 These agents should be avoided in any patient with a history of pancreatitis.¹⁴ In addition, trials investigating the cardiovascular safety and efficacy of DPP-4 inhibitors point to an increased risk of heart failure with the use of saxagliptin and alogliptin, regardless of age.^15,16 The potential for adverse effects warrants increased patient monitoring when using these agents in older patients.

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are injectable agents that potentiate the actions of the naturally occurring incretin GLP-1, which increases glucose-dependent insulin secretion, inhibits glucagon release, reduces hepatic glucose production, and delays gastric emptying. These agents have a pronounced effect on satiety and promote weight loss. The most common adverse effects are nausea, vomiting, and diarrhea, which occur most commonly during treatment initiation and titration. Studies in elderly patients confirm A1C reductions of 1% to 1.5% and a low risk of hypoglycemia when used alone.^17,18

GLP-1 RAs can be used as monotherapy in older patients at risk for hypoglycemia or in those with hypoglycemic unawareness. They can also be used in combination therapy with other agents, including insulin, though concomitant use with insulin or insulin secretagogues increases the risk of hypoglycemia.³ Weight loss and GI adverse effects may limit the use of these agents in frail or undernourished elderly patients.⁶

Continue to: Since these agents are injected...

Since these agents are injected, they require intact visual, motor, and cognitive skills and thus may not be appropriate in older patients with cognitive or visual impairment or decreased dexterity. In addition, the high cost of these agents may limit their use.

Select a GLP-1 RA based on the frequency of administration, type of glucose control required (fasting or post-prandial), and the patient’s ability to use the administration device. Dose adjustment is required in renal impairment, except with dulaglutide and liraglutide. Use with caution in patients with a history of pancreatitis, and stop GLP-1 RAs if pancreatitis is suspected during treatment.⁴ Avoid GLP-1 RAs in patients with a personal or family history of thyroid-related cancers, as these agents have been associated with medullary thyroid tumors in animals.⁴

A new indication. Recent evidence suggests the GLP-1 RAs may offer additional cardiovascular benefit in patients with diabetes.^18,19 In August 2017, liraglutide gained an additional FDA indication to reduce the risk of major adverse cardiovascular events in adults with T2DM and established cardiovascular disease.

This new indication was based on the Novo Nordisk- and National Institutes of Health-sponsored LEADER trial, in which liraglutide reduced the risk of cardiovascular death, nonfatal heart attack, or nonfatal stroke by 13% vs placebo (P=.01) with an absolute risk reduction (ARR) of 1.9%.¹⁹ Liraglutide demonstrated a 22% reduction in cardiovascular death and a 15% reduction in all-cause death (ARR 1.3%, 1.4% respectively).¹⁹ The new cardiovascular indication may impact the choice of add-on therapy to metformin in patients with preexisting cardiovascular conditions.

Continue to: Sodium glucose cotransporter-2 inhibitors

Sodium glucose cotransporter-2 inhibitors

SGLT-2 inhibitors prevent the reabsorption of renal-filtered glucose, resulting in decreased blood glucose levels and increased urinary excretion of glucose without stimulating insulin secretion, and therefore without increasing the risk of hypoglycemia. Additional effects include decreased blood pressure and weight loss.²⁰ Dose adjustment is required in renal impairment.

SGLT-2 inhibitors can be used as monotherapy or in combination with other agents, including insulin, and the relatively low risk of hypoglycemia and moderate A1C lowering potential of 0.5% to 1% provide an oral option for select older patients.²⁰ Common adverse events include hypotension, hyperkalemia, increased low-density lipoprotein (LDL) levels, acute kidney injury, genital mycotic infections, and hypoglycemia when used in combination with insulin or insulin secretagogues.²⁰

DPP-4 agents are ideal for older patients who are not candidates for metformin or a sulfonylurea.

Additional warnings have been issued by the FDA for the risk of urinary tract infection with sepsis, as well as diabetic ketoacidosis associated with SGLT-2 inhibitor use.²¹ The FDA has reported bone fracture risk and decreased bone mineral density with canagliflozin.²¹ Avoid using SGLT-2 inhibitors in patients with osteopenia or osteoporosis, as the risks outweigh the benefits. Drug-specific warnings may further impact individual use of an agent, with canagliflozin most recently having been associated with increased risk of leg and foot amputations.²¹

Given the adverse effect profile of SGLT-2 inhibitors, assess their risks and benefits in older patients on a case-by-case basis. Before initiating therapy, evaluate each patient’s volume status. A higher incidence of adverse effects related to intravascular volume depletion has been reported in those 65 or older, with a more prominent increase seen in patients 75 or older.²² However, the risk of hypoglycemia does not seem to increase with age.²²

Although many adverse effects have been reported with SGLT-2 inhibitors, empagliflozin was associated with significantly lower rates of all-cause and cardiovascular death and lower risk of hospitalization for heart failure in the only SGLT-2 inhibitor cardiovascular outcomes trial reported to date.²³ If this cardiovascular benefit is replicated in additional trials of the other SGLT-2 inhibitors, use of this drug class may increase.

Continue to: Insulin

Many patients will ultimately require insulin due to the progressive loss of beta-cell function that occurs in advanced diabetes. Starting insulin therapy early on in the disease may actually restore beta-cell function and reduce glucotoxicity.²⁴ In elderly patients with uncontrolled diabetes, early treatment with basal insulin results in better glycemic control and less hypoglycemia than continuing to titrate oral agents.²⁵

Despite these benefits, however, insulin use often is not optimized in the elderly due to concerns about hypoglycemia and difficulty of administration. Safe use of insulin requires careful selection of an appropriate insulin regimen, since insulin use has been identified as an independent predictor of severe hypoglycemia in the elderly.^8,26 Before initiating insulin therapy, evaluate whether an older patient is cognitively and physically able to safely use insulin.

Glipizide has no active metabolites and has the lowest risk of hypoglycemia in the setting of decreased renal function, making it the preferred sulfonylurea for use in the elderly.

Multiple daily injections may be challenging for some older adults. Limit such insulin regimens to use in high-functioning patients. Although all types of insulin can cause hypoglycemia, regimens that mimic insulin’s normal physiologic pattern introduce less hypoglycemic risk. Using basal insulin that mimics the body’s sustained insulin level throughout the day is associated with a lower frequency of hypoglycemia in older people with diabetes than conventional insulin regimens. Long-acting insulins such as glargine, detemir, and degludec offer a lower risk of hypoglycemia, particularly nocturnal hypoglycemia which may contribute to falls.^2,27

Neutral protamine Hagedorn insulin and regular insulin are not recommended for use in the elderly, as they do not mimic the body’s natural basal-bolus insulin production and thus put patients at higher risk of hypoglycemia.⁴ If insulin intensification is needed after optimizing basal insulin, consider adding mealtime insulin with a bolus of rapid-acting insulin (insulin aspart, insulin lispro, or insulin glulisine). It is important to note that the kidneys are responsible for 30% to 80% of insulin clearance from the body.²⁸ Because insulin action is prolonged in renal insufficiency, prevent hypoglycemia by decreasing basal and bolus doses when the eGFR is below 50 mL/min/1.73m².²⁸

Dosing errors. Whenever possible, use insulin preparations that minimize dosing errors. Insulin pen formulations, if financially feasible, allow more accurate dosing and are more acceptable to older patients compared with syringes and vials.²⁹ Pen formulations are particularly preferable for older patients with impaired vision or dexterity.²⁹ In addition, when patients must mix insulins, errors are more likely to occur. The use of premixed insulin vials has been shown to increase dosing accuracy when used by the elderly.³⁰

Continue to: Combining antidiabetes agents

Combining antidiabetes agents

Combination therapy is often needed as T2DM progresses, even though strict glucose control is generally not recommended for the elderly.³¹ The American Geriatrics Society advises avoiding additional medications other than metformin to achieve an A1C level below 7.5% in most older adults.

However, for older patients already taking metformin who are not at their A1C goal, consider adding a second agent, if not contraindicated. Potential agents include a GLP-1 RA, SGLT-2 inhibitor, DDP-4 inhibitor, or short-acting sulfonylurea (glipizide). Alternatively, basal insulin may be added. However, avoid combining a sulfonylurea with insulin, which greatly increases the risk of hypoglycemia.³² Consider adding a GLP-1 RA or basal insulin if the patient is not at his/her target A1C on oral therapy with multiple agents.³

CORRESPONDENCE
Barbara Keber, MD, Glen Cove Hospital, 101 St. Andrews Lane, Glen Cove, NY; [email protected].