Article

Who is most at risk of suicide by drug overdose? Has that changed in recent years? Researchers at the National Institute on Drug Abuse analyzed data from 2001 to 2019 from the Centers for Disease Control and Prevention’s National Vital Statistics System to find out.

On the whole, they say, intentional overdose deaths have declined. But suicide rates increased in certain subgroups: young adults (aged 15-24 years), older adults (aged 75-84 years), and non-Hispanic Black women. Rates among women were “consistently higher” than those of men. The highest rates were observed in women aged 45 to 64 years.

Monday was the worst day, and the weekends had the lowest rates. The researchers say social factors, such as more social interactions on the weekend and reluctance about starting the workweek, could be factors.

Seasonally, the numbers ran true to the pattern seen in previous studies: The lowest rates occurred in December and highest in late spring and summer. Perhaps the “collective optimism” of the holiday season and social interactions exert protective effects against suicidality, the researchers suggest.

Factors also may include biological changes. In this study, the researchers found a positive linear relationship between daylength, which varies by latitude, and intentional overdose deaths for both sexes. Daylength is associated with mu opioid receptor (MOR) availability that might underlie seasonal variations in mood, they posit. MORs are the main target of opioid drugs; the researchers cite a study that found altered MOR expression in postmortem brains of suicide victims.

They note some limitations of their study, one being that, in 2019, 5% of overdose deaths had undetermined intent. Improving classifications of overdose deaths is needed, they say.

Moreover, the trends might have changed during the pandemic, as provisional mortality data indicate decreases in deaths by suicides, but also an approximate 30% increase in overall overdose deaths.

“This research underscores the importance of external support structures and environmental factors in determining a person’s suicide risk,” said Emily B. Einstein, PhD, chief of the National Institute on Drug Abuse’s Science Policy Branch and an author on the study. “The risk of intentional overdoses, and suicide risk in general, is not static. This is crucial for clinicians to keep in mind, as they may need to assess patients’ suicide risk frequently rather than at one point in time. It is also important for friends and family members of people who may be at an increased risk of suicide, and for those people themselves, so that they can be aware of the greatest periods of risk and seek help when needed.”

Sources: https://www.nih.gov/news-events/news-releases/suicides-drug-overdose-increased-among-young-people-elderly-people-black-women-despite-overall-downward-tren

Han B, Compton WM, Einstein EB, et al. Intentional drug overdose deaths in the United States.  Am J Psychiatry. doi:10.1176/appi.ajp.2021.21060604

Who is most at risk of suicide by drug overdose? Has that changed in recent years? Researchers at the National Institute on Drug Abuse analyzed data from 2001 to 2019 from the Centers for Disease Control and Prevention’s National Vital Statistics System to find out.

On the whole, they say, intentional overdose deaths have declined. But suicide rates increased in certain subgroups: young adults (aged 15-24 years), older adults (aged 75-84 years), and non-Hispanic Black women. Rates among women were “consistently higher” than those of men. The highest rates were observed in women aged 45 to 64 years.

Monday was the worst day, and the weekends had the lowest rates. The researchers say social factors, such as more social interactions on the weekend and reluctance about starting the workweek, could be factors.

Seasonally, the numbers ran true to the pattern seen in previous studies: The lowest rates occurred in December and highest in late spring and summer. Perhaps the “collective optimism” of the holiday season and social interactions exert protective effects against suicidality, the researchers suggest.

Factors also may include biological changes. In this study, the researchers found a positive linear relationship between daylength, which varies by latitude, and intentional overdose deaths for both sexes. Daylength is associated with mu opioid receptor (MOR) availability that might underlie seasonal variations in mood, they posit. MORs are the main target of opioid drugs; the researchers cite a study that found altered MOR expression in postmortem brains of suicide victims.

They note some limitations of their study, one being that, in 2019, 5% of overdose deaths had undetermined intent. Improving classifications of overdose deaths is needed, they say.

Moreover, the trends might have changed during the pandemic, as provisional mortality data indicate decreases in deaths by suicides, but also an approximate 30% increase in overall overdose deaths.

“This research underscores the importance of external support structures and environmental factors in determining a person’s suicide risk,” said Emily B. Einstein, PhD, chief of the National Institute on Drug Abuse’s Science Policy Branch and an author on the study. “The risk of intentional overdoses, and suicide risk in general, is not static. This is crucial for clinicians to keep in mind, as they may need to assess patients’ suicide risk frequently rather than at one point in time. It is also important for friends and family members of people who may be at an increased risk of suicide, and for those people themselves, so that they can be aware of the greatest periods of risk and seek help when needed.”

Sources: https://www.nih.gov/news-events/news-releases/suicides-drug-overdose-increased-among-young-people-elderly-people-black-women-despite-overall-downward-tren

Han B, Compton WM, Einstein EB, et al. Intentional drug overdose deaths in the United States.  Am J Psychiatry. doi:10.1176/appi.ajp.2021.21060604

Who is most at risk of suicide by drug overdose? Has that changed in recent years? Researchers at the National Institute on Drug Abuse analyzed data from 2001 to 2019 from the Centers for Disease Control and Prevention’s National Vital Statistics System to find out.

On the whole, they say, intentional overdose deaths have declined. But suicide rates increased in certain subgroups: young adults (aged 15-24 years), older adults (aged 75-84 years), and non-Hispanic Black women. Rates among women were “consistently higher” than those of men. The highest rates were observed in women aged 45 to 64 years.

Monday was the worst day, and the weekends had the lowest rates. The researchers say social factors, such as more social interactions on the weekend and reluctance about starting the workweek, could be factors.

Seasonally, the numbers ran true to the pattern seen in previous studies: The lowest rates occurred in December and highest in late spring and summer. Perhaps the “collective optimism” of the holiday season and social interactions exert protective effects against suicidality, the researchers suggest.

Factors also may include biological changes. In this study, the researchers found a positive linear relationship between daylength, which varies by latitude, and intentional overdose deaths for both sexes. Daylength is associated with mu opioid receptor (MOR) availability that might underlie seasonal variations in mood, they posit. MORs are the main target of opioid drugs; the researchers cite a study that found altered MOR expression in postmortem brains of suicide victims.

They note some limitations of their study, one being that, in 2019, 5% of overdose deaths had undetermined intent. Improving classifications of overdose deaths is needed, they say.

Moreover, the trends might have changed during the pandemic, as provisional mortality data indicate decreases in deaths by suicides, but also an approximate 30% increase in overall overdose deaths.

“This research underscores the importance of external support structures and environmental factors in determining a person’s suicide risk,” said Emily B. Einstein, PhD, chief of the National Institute on Drug Abuse’s Science Policy Branch and an author on the study. “The risk of intentional overdoses, and suicide risk in general, is not static. This is crucial for clinicians to keep in mind, as they may need to assess patients’ suicide risk frequently rather than at one point in time. It is also important for friends and family members of people who may be at an increased risk of suicide, and for those people themselves, so that they can be aware of the greatest periods of risk and seek help when needed.”

Sources: https://www.nih.gov/news-events/news-releases/suicides-drug-overdose-increased-among-young-people-elderly-people-black-women-despite-overall-downward-tren

Han B, Compton WM, Einstein EB, et al. Intentional drug overdose deaths in the United States.  Am J Psychiatry. doi:10.1176/appi.ajp.2021.21060604

From environmental tobacco smoke to ultraviolet (UV) radiation, diesel exhaust particulates, lead, and now, chronic infection with Helicobacter pylori (H pylori)—the Report on Carcinogens has regularly updated the list of substances known or “reasonably anticipated” to cause cancer.

The 15th report, which is prepared by the National Toxicology Program (NTP) for the Department of Health and Human Services, has 8 new entries, bringing the number of human carcinogens (eg, metals, pesticides, and drugs) on the list to 256. (The first report, released in 1980, listed 26.) In addition to H pylori infection, this edition adds the flame-retardant chemical antimony trioxide, and 6 haloacetic acids found as water disinfection byproducts.

In 1971, then President Nixon declared “war on cancer” (the second leading cause of death in the US) and signed the National Cancer Act. In 1978, Congress ordered the Report on Carcinogens, to educate the public and health professionals on potential environmental carcinogenic hazards.

Perhaps disheartening to know that even with 256 entries, the list probably understates the number of carcinogens humans and other creatures are exposed to. But things can change with time. Each list goes through a rigorous round of reviews. Sometimes substances are “delisted” after, for instance, litigation or new research. Saccharin, for example, was removed from the ninth edition. It was listed as “reasonably anticipated” in 1981, based on “sufficient evidence of carcinogenicity in experimental animals.” It was removed, however, after extensive review of decades of saccharin use determined that the data were not sufficient to meet current criteria. Further research had revealed, also, that the observed bladder tumors in rats arose from a mechanism not relevant to humans.

Other entries, such as the controversial listing of the cancer drug tamoxifen, walk a fine line between risk and benefit. Tamoxifen, first listed in the ninth report (and still in the 15th report), was included because studies revealed that it could increase the risk of uterine cancer in women. But there also was conclusive evidence that it may prevent or delay breast cancer in women who are at high risk.

Ultimately, the report’s authors make it clear that it is for informative value and guidance, not necessarily a dictate. As one report put it: “Personal decisions concerning voluntary exposures to carcinogenic agents need to be based on additional information that is beyond the scope” of the report.

“As the identification of carcinogens is a key step in cancer prevention,” said Rick Woychik, PhD, director of the National Institute of Environmental Health Sciences and NTP, “publication of the report represents an important government activity towards improving public health.”

From environmental tobacco smoke to ultraviolet (UV) radiation, diesel exhaust particulates, lead, and now, chronic infection with Helicobacter pylori (H pylori)—the Report on Carcinogens has regularly updated the list of substances known or “reasonably anticipated” to cause cancer.

The 15th report, which is prepared by the National Toxicology Program (NTP) for the Department of Health and Human Services, has 8 new entries, bringing the number of human carcinogens (eg, metals, pesticides, and drugs) on the list to 256. (The first report, released in 1980, listed 26.) In addition to H pylori infection, this edition adds the flame-retardant chemical antimony trioxide, and 6 haloacetic acids found as water disinfection byproducts.

In 1971, then President Nixon declared “war on cancer” (the second leading cause of death in the US) and signed the National Cancer Act. In 1978, Congress ordered the Report on Carcinogens, to educate the public and health professionals on potential environmental carcinogenic hazards.

Perhaps disheartening to know that even with 256 entries, the list probably understates the number of carcinogens humans and other creatures are exposed to. But things can change with time. Each list goes through a rigorous round of reviews. Sometimes substances are “delisted” after, for instance, litigation or new research. Saccharin, for example, was removed from the ninth edition. It was listed as “reasonably anticipated” in 1981, based on “sufficient evidence of carcinogenicity in experimental animals.” It was removed, however, after extensive review of decades of saccharin use determined that the data were not sufficient to meet current criteria. Further research had revealed, also, that the observed bladder tumors in rats arose from a mechanism not relevant to humans.

Other entries, such as the controversial listing of the cancer drug tamoxifen, walk a fine line between risk and benefit. Tamoxifen, first listed in the ninth report (and still in the 15th report), was included because studies revealed that it could increase the risk of uterine cancer in women. But there also was conclusive evidence that it may prevent or delay breast cancer in women who are at high risk.

Ultimately, the report’s authors make it clear that it is for informative value and guidance, not necessarily a dictate. As one report put it: “Personal decisions concerning voluntary exposures to carcinogenic agents need to be based on additional information that is beyond the scope” of the report.

“As the identification of carcinogens is a key step in cancer prevention,” said Rick Woychik, PhD, director of the National Institute of Environmental Health Sciences and NTP, “publication of the report represents an important government activity towards improving public health.”

From environmental tobacco smoke to ultraviolet (UV) radiation, diesel exhaust particulates, lead, and now, chronic infection with Helicobacter pylori (H pylori)—the Report on Carcinogens has regularly updated the list of substances known or “reasonably anticipated” to cause cancer.

The 15th report, which is prepared by the National Toxicology Program (NTP) for the Department of Health and Human Services, has 8 new entries, bringing the number of human carcinogens (eg, metals, pesticides, and drugs) on the list to 256. (The first report, released in 1980, listed 26.) In addition to H pylori infection, this edition adds the flame-retardant chemical antimony trioxide, and 6 haloacetic acids found as water disinfection byproducts.

In 1971, then President Nixon declared “war on cancer” (the second leading cause of death in the US) and signed the National Cancer Act. In 1978, Congress ordered the Report on Carcinogens, to educate the public and health professionals on potential environmental carcinogenic hazards.

Perhaps disheartening to know that even with 256 entries, the list probably understates the number of carcinogens humans and other creatures are exposed to. But things can change with time. Each list goes through a rigorous round of reviews. Sometimes substances are “delisted” after, for instance, litigation or new research. Saccharin, for example, was removed from the ninth edition. It was listed as “reasonably anticipated” in 1981, based on “sufficient evidence of carcinogenicity in experimental animals.” It was removed, however, after extensive review of decades of saccharin use determined that the data were not sufficient to meet current criteria. Further research had revealed, also, that the observed bladder tumors in rats arose from a mechanism not relevant to humans.

Other entries, such as the controversial listing of the cancer drug tamoxifen, walk a fine line between risk and benefit. Tamoxifen, first listed in the ninth report (and still in the 15th report), was included because studies revealed that it could increase the risk of uterine cancer in women. But there also was conclusive evidence that it may prevent or delay breast cancer in women who are at high risk.

Ultimately, the report’s authors make it clear that it is for informative value and guidance, not necessarily a dictate. As one report put it: “Personal decisions concerning voluntary exposures to carcinogenic agents need to be based on additional information that is beyond the scope” of the report.

“As the identification of carcinogens is a key step in cancer prevention,” said Rick Woychik, PhD, director of the National Institute of Environmental Health Sciences and NTP, “publication of the report represents an important government activity towards improving public health.”

In a recent letter to the editor, “25-hydroxyvitamin D concentration is key to analyzing vitamin D’s effects” (J Fam Pract. 2021;70:472), Dr. Grant links vitamin D supplementation with important health outcomes. He concludes that the positivity rate of SARS-CoV-2 was only 5.9% in people with higher concentrations of 25(OH)D vs 12.5% in those with lower concentrations. This is a flawed conclusion on the face of it, because the great confabulatory factor is behavior. Is it possible that those more likely to take supplemental vitamin D do so as a result of overall healthier lifestyles and choices (eg, vaccinations)? As health care representatives, we must be very careful about the data we publish and the commentary we attach to it, lest we advertise inadvertent follies. I see so much of that in our “peer-reviewed literature.”

I came to medicine as a chemist, and the rigors of peer review impressed upon the hard (fundamental) sciences are markedly different from those we “claim” adherence to in medicine. I find that some of the medical literature and study designs fall short of what would pass muster in the fundamental science industry. That is a shame! Such statements, as discussed here, have to be served for public consumption, and even to our colleagues, with a generous helping of skepticism and qualification.

RA Segal, MD, MPH
Gainesville, FL

In a recent letter to the editor, “25-hydroxyvitamin D concentration is key to analyzing vitamin D’s effects” (J Fam Pract. 2021;70:472), Dr. Grant links vitamin D supplementation with important health outcomes. He concludes that the positivity rate of SARS-CoV-2 was only 5.9% in people with higher concentrations of 25(OH)D vs 12.5% in those with lower concentrations. This is a flawed conclusion on the face of it, because the great confabulatory factor is behavior. Is it possible that those more likely to take supplemental vitamin D do so as a result of overall healthier lifestyles and choices (eg, vaccinations)? As health care representatives, we must be very careful about the data we publish and the commentary we attach to it, lest we advertise inadvertent follies. I see so much of that in our “peer-reviewed literature.”

I came to medicine as a chemist, and the rigors of peer review impressed upon the hard (fundamental) sciences are markedly different from those we “claim” adherence to in medicine. I find that some of the medical literature and study designs fall short of what would pass muster in the fundamental science industry. That is a shame! Such statements, as discussed here, have to be served for public consumption, and even to our colleagues, with a generous helping of skepticism and qualification.

RA Segal, MD, MPH
Gainesville, FL

In a recent letter to the editor, “25-hydroxyvitamin D concentration is key to analyzing vitamin D’s effects” (J Fam Pract. 2021;70:472), Dr. Grant links vitamin D supplementation with important health outcomes. He concludes that the positivity rate of SARS-CoV-2 was only 5.9% in people with higher concentrations of 25(OH)D vs 12.5% in those with lower concentrations. This is a flawed conclusion on the face of it, because the great confabulatory factor is behavior. Is it possible that those more likely to take supplemental vitamin D do so as a result of overall healthier lifestyles and choices (eg, vaccinations)? As health care representatives, we must be very careful about the data we publish and the commentary we attach to it, lest we advertise inadvertent follies. I see so much of that in our “peer-reviewed literature.”

I came to medicine as a chemist, and the rigors of peer review impressed upon the hard (fundamental) sciences are markedly different from those we “claim” adherence to in medicine. I find that some of the medical literature and study designs fall short of what would pass muster in the fundamental science industry. That is a shame! Such statements, as discussed here, have to be served for public consumption, and even to our colleagues, with a generous helping of skepticism and qualification.

RA Segal, MD, MPH
Gainesville, FL

A 56-year-old woman with a history of a thyroid goiter following a partial thyroidectomy presented to the emergency department with shortness of breath, progressive weakness, and fatigue. She also reported a poor appetite and unintentional weight loss of approximately 40 lbs over the past 2 months but had not sought medical care. She denied having a cough, chest pain, hemoptysis, hematemesis, fevers, chills, or night sweats.

Physical examination revealed a cachectic woman with tachycardia and tachypnea, along with diffuse coarse rales throughout both lungs. The patient’s initial heart rate was 101 beats/min; respiratory rate, 25 breaths/min; and oxygen saturation, 74% on room air. The results of a complete blood count and comprehensive metabolic panel were within normal limits. A chest radiograph was performed, showing innumerable subcentimeter- and ­pericentimeter-sized soft-tissue densities in both lungs (FIGURE 1). Computed tomography (CT) of the chest was subsequently obtained to better characterize the nodules (FIGURES 2A and 2B). The CT revealed innumerable noncalcified pulmonary nodules as well as multiple hilar masses suggestive of malignancy; however, infectious etiologies could not be ruled out.

As the patient was visiting from a tuberculosis-endemic area, she was empirically started on rifampin, isoniazid, pyrazinamide, and ethambutol (RIPE) therapy for treatment of Mycobacterium tuberculosis infection. Subsequently, polymerase chain reaction (PCR) testing and acid-fast bacillus sputum smear came back negative for M tuberculosis and RIPE therapy was stopped. Bacterial culture and fungal screens for Pneumocystis jirovecii, Coccidioides, Cryptococcus, and Histoplasma capsulatum were negative.

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

A biopsy of one of the lung nodules was performed and showed strong positive staining for thyroglobulin and thyroid transcription factor-1, establishing a diagnosis of metastatic follicular thyroid carcinoma. Further imaging revealed additional metastases to the brain, liver, adrenal glands, and spine.

If follicular thyroid carcinoma is caught early (with only local involvement), 5-year relative survival rates are near 100%.

Follicular thyroid carcinoma is the second most common histologic type of thyroid cancer, comprising 10% to 15% of thyroid cancer cases.¹ Distant metastases are not uncommon in this histologic type; about 11% of patients have distant metastases at presentation.² Compared with papillary thyroid cancer, which spreads via the lymphatics, the follicular type can metastasize via vascular invasion, thus leading to its “aggressiveness” and frequent occurrence of metastasis outside the neck (most commonly to the lung).¹ For this reason, an early diagnosis is important.³ In this case, the patient revealed that she hadn’t sought medical care at the onset of her symptoms due to financial limitations and limited access to medical providers; this likely contributed to the patient’s diagnosis at a late stage of disease.

Infectious and septic diseases are part of the differential Dx

The differential diagnosis for diffuse pulmonary nodules includes various infections, silicosis, coal worker’s pneumoconiosis, septic pulmonary emboli, and pulmonary sarcoidosis. All of these diagnoses can cause constitutional, as well as pulmonary, symptoms.

Infections, such as pulmonary tuberculosis, pulmonary coccidioidomycosis, or other cavitating infections, can be ruled out with specific testing for the organism of concern and/or culture.

Silicosis and coal worker’s pneumoconiosis generally cause lesions in the upper lobes and require a significant occupational exposure to silica or coal, respectively.

Continue to: Septic pulmonary emboli

Septic pulmonary emboli are usually seen in the context of sepsis, and there are positive blood cultures at the time of imaging.

Pulmonary sarcoidosis typically manifests with hilar and mediastinal lymphadenopathy and is often accompanied by other organ involvement.

There are options for treatment if diagnosis is made early

Thyroidectomy, radioiodine treatment, and thyroid-stimulating hormone suppressive therapy are the most commonly used therapies for treatment of follicular thyroid carcinoma.⁴ If follicular thyroid carcinoma is caught early (with only local involvement), 5-year relative survival rates are near 100%.⁵ However, outcomes are generally poor in patients who are of older age and those who have macronodular lung metastases or multiple bone metastases.⁶

Our patient was started on levothyroxine. Radioiodine and targeted drug therapy with a tyrosine kinase inhibitor were attempted, but the patient was unable to tolerate them due to her poor functional status. She was thus transitioned to hospice care.

A 56-year-old woman with a history of a thyroid goiter following a partial thyroidectomy presented to the emergency department with shortness of breath, progressive weakness, and fatigue. She also reported a poor appetite and unintentional weight loss of approximately 40 lbs over the past 2 months but had not sought medical care. She denied having a cough, chest pain, hemoptysis, hematemesis, fevers, chills, or night sweats.

Physical examination revealed a cachectic woman with tachycardia and tachypnea, along with diffuse coarse rales throughout both lungs. The patient’s initial heart rate was 101 beats/min; respiratory rate, 25 breaths/min; and oxygen saturation, 74% on room air. The results of a complete blood count and comprehensive metabolic panel were within normal limits. A chest radiograph was performed, showing innumerable subcentimeter- and ­pericentimeter-sized soft-tissue densities in both lungs (FIGURE 1). Computed tomography (CT) of the chest was subsequently obtained to better characterize the nodules (FIGURES 2A and 2B). The CT revealed innumerable noncalcified pulmonary nodules as well as multiple hilar masses suggestive of malignancy; however, infectious etiologies could not be ruled out.

As the patient was visiting from a tuberculosis-endemic area, she was empirically started on rifampin, isoniazid, pyrazinamide, and ethambutol (RIPE) therapy for treatment of Mycobacterium tuberculosis infection. Subsequently, polymerase chain reaction (PCR) testing and acid-fast bacillus sputum smear came back negative for M tuberculosis and RIPE therapy was stopped. Bacterial culture and fungal screens for Pneumocystis jirovecii, Coccidioides, Cryptococcus, and Histoplasma capsulatum were negative.

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

A biopsy of one of the lung nodules was performed and showed strong positive staining for thyroglobulin and thyroid transcription factor-1, establishing a diagnosis of metastatic follicular thyroid carcinoma. Further imaging revealed additional metastases to the brain, liver, adrenal glands, and spine.

If follicular thyroid carcinoma is caught early (with only local involvement), 5-year relative survival rates are near 100%.

Follicular thyroid carcinoma is the second most common histologic type of thyroid cancer, comprising 10% to 15% of thyroid cancer cases.¹ Distant metastases are not uncommon in this histologic type; about 11% of patients have distant metastases at presentation.² Compared with papillary thyroid cancer, which spreads via the lymphatics, the follicular type can metastasize via vascular invasion, thus leading to its “aggressiveness” and frequent occurrence of metastasis outside the neck (most commonly to the lung).¹ For this reason, an early diagnosis is important.³ In this case, the patient revealed that she hadn’t sought medical care at the onset of her symptoms due to financial limitations and limited access to medical providers; this likely contributed to the patient’s diagnosis at a late stage of disease.

Infectious and septic diseases are part of the differential Dx

The differential diagnosis for diffuse pulmonary nodules includes various infections, silicosis, coal worker’s pneumoconiosis, septic pulmonary emboli, and pulmonary sarcoidosis. All of these diagnoses can cause constitutional, as well as pulmonary, symptoms.

Infections, such as pulmonary tuberculosis, pulmonary coccidioidomycosis, or other cavitating infections, can be ruled out with specific testing for the organism of concern and/or culture.

Silicosis and coal worker’s pneumoconiosis generally cause lesions in the upper lobes and require a significant occupational exposure to silica or coal, respectively.

Continue to: Septic pulmonary emboli

Septic pulmonary emboli are usually seen in the context of sepsis, and there are positive blood cultures at the time of imaging.

Pulmonary sarcoidosis typically manifests with hilar and mediastinal lymphadenopathy and is often accompanied by other organ involvement.

There are options for treatment if diagnosis is made early

Thyroidectomy, radioiodine treatment, and thyroid-stimulating hormone suppressive therapy are the most commonly used therapies for treatment of follicular thyroid carcinoma.⁴ If follicular thyroid carcinoma is caught early (with only local involvement), 5-year relative survival rates are near 100%.⁵ However, outcomes are generally poor in patients who are of older age and those who have macronodular lung metastases or multiple bone metastases.⁶

Our patient was started on levothyroxine. Radioiodine and targeted drug therapy with a tyrosine kinase inhibitor were attempted, but the patient was unable to tolerate them due to her poor functional status. She was thus transitioned to hospice care.

A 56-year-old woman with a history of a thyroid goiter following a partial thyroidectomy presented to the emergency department with shortness of breath, progressive weakness, and fatigue. She also reported a poor appetite and unintentional weight loss of approximately 40 lbs over the past 2 months but had not sought medical care. She denied having a cough, chest pain, hemoptysis, hematemesis, fevers, chills, or night sweats.

Physical examination revealed a cachectic woman with tachycardia and tachypnea, along with diffuse coarse rales throughout both lungs. The patient’s initial heart rate was 101 beats/min; respiratory rate, 25 breaths/min; and oxygen saturation, 74% on room air. The results of a complete blood count and comprehensive metabolic panel were within normal limits. A chest radiograph was performed, showing innumerable subcentimeter- and ­pericentimeter-sized soft-tissue densities in both lungs (FIGURE 1). Computed tomography (CT) of the chest was subsequently obtained to better characterize the nodules (FIGURES 2A and 2B). The CT revealed innumerable noncalcified pulmonary nodules as well as multiple hilar masses suggestive of malignancy; however, infectious etiologies could not be ruled out.

As the patient was visiting from a tuberculosis-endemic area, she was empirically started on rifampin, isoniazid, pyrazinamide, and ethambutol (RIPE) therapy for treatment of Mycobacterium tuberculosis infection. Subsequently, polymerase chain reaction (PCR) testing and acid-fast bacillus sputum smear came back negative for M tuberculosis and RIPE therapy was stopped. Bacterial culture and fungal screens for Pneumocystis jirovecii, Coccidioides, Cryptococcus, and Histoplasma capsulatum were negative.

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

A biopsy of one of the lung nodules was performed and showed strong positive staining for thyroglobulin and thyroid transcription factor-1, establishing a diagnosis of metastatic follicular thyroid carcinoma. Further imaging revealed additional metastases to the brain, liver, adrenal glands, and spine.

If follicular thyroid carcinoma is caught early (with only local involvement), 5-year relative survival rates are near 100%.

Follicular thyroid carcinoma is the second most common histologic type of thyroid cancer, comprising 10% to 15% of thyroid cancer cases.¹ Distant metastases are not uncommon in this histologic type; about 11% of patients have distant metastases at presentation.² Compared with papillary thyroid cancer, which spreads via the lymphatics, the follicular type can metastasize via vascular invasion, thus leading to its “aggressiveness” and frequent occurrence of metastasis outside the neck (most commonly to the lung).¹ For this reason, an early diagnosis is important.³ In this case, the patient revealed that she hadn’t sought medical care at the onset of her symptoms due to financial limitations and limited access to medical providers; this likely contributed to the patient’s diagnosis at a late stage of disease.

Infectious and septic diseases are part of the differential Dx

The differential diagnosis for diffuse pulmonary nodules includes various infections, silicosis, coal worker’s pneumoconiosis, septic pulmonary emboli, and pulmonary sarcoidosis. All of these diagnoses can cause constitutional, as well as pulmonary, symptoms.

Infections, such as pulmonary tuberculosis, pulmonary coccidioidomycosis, or other cavitating infections, can be ruled out with specific testing for the organism of concern and/or culture.

Silicosis and coal worker’s pneumoconiosis generally cause lesions in the upper lobes and require a significant occupational exposure to silica or coal, respectively.

Continue to: Septic pulmonary emboli

Septic pulmonary emboli are usually seen in the context of sepsis, and there are positive blood cultures at the time of imaging.

Pulmonary sarcoidosis typically manifests with hilar and mediastinal lymphadenopathy and is often accompanied by other organ involvement.

There are options for treatment if diagnosis is made early

Thyroidectomy, radioiodine treatment, and thyroid-stimulating hormone suppressive therapy are the most commonly used therapies for treatment of follicular thyroid carcinoma.⁴ If follicular thyroid carcinoma is caught early (with only local involvement), 5-year relative survival rates are near 100%.⁵ However, outcomes are generally poor in patients who are of older age and those who have macronodular lung metastases or multiple bone metastases.⁶

Our patient was started on levothyroxine. Radioiodine and targeted drug therapy with a tyrosine kinase inhibitor were attempted, but the patient was unable to tolerate them due to her poor functional status. She was thus transitioned to hospice care.

THE CASE

A previously healthy 3-year-old girl presented to the emergency department with 4 days of fever and 2 days of right-side neck pain. The maximum temperature at home was 103 °F. The patient was irritable and vomited once. There were no other apparent or reported symptoms.

The neck exam was notable for nonfluctuant, swollen, and tender lymph nodes on the right side. Her sclera and conjunctiva were clear, and her oropharynx was unremarkable. Lab work revealed leukocytosis, with a white blood cell (WBC) count of 15.5 × 10³/µL (normal range, 4.0-10.0 × 10³/µL). She was given one 20 cc/kg normal saline bolus, started on intravenous clindamycin for presumed cervical lymphadenitis, and admitted to the hospital.

On Day 2, the patient developed a fine maculopapular rash on her chest, abdomen, and back. She had spiking fevers—as high as 102.2 °F—despite being on antibiotics for more than 24 hours. The erythrocyte sedimentation rate (ESR) was 39 mm/h (0-20 mm/h), and C-reactive protein (CRP) was 71.4 mg/L (0.0-4.9 mg/L). Due to concern for abscess, a neck ultrasound was performed; it showed a chain of enlarged lymph nodes in the right neck (largest, 2.3 × 1.1 × 1.4 cm) and no abscess.

On Day 3, clindamycin was switched to intravenous ampicillin/sulbactam because a nasal swab for methicillin-resistant Staphylococcus aureus was negative. A swab for respiratory viral infections was also negative. The patient then developed notable facial swelling, bilateral bulbar conjunctival injection with limbic sparing, and swelling of her hands and feet.

THE DIAGNOSIS

By the end of Day 3, the patient’s lab studies demonstrated microcytic anemia and low albumin (2.5 g/dL), but no transaminitis, thrombocytosis, or sterile pyuria. An electrocardiogram was unremarkable. A pediatric echocardiogram revealed hyperemic coronaries, indicating inflammation. The coronary arteries were measured in the upper limits of normal, and the patient’s Z-scores were < 2.5. (A Z-score < 2 indicates no involvement, 2 to < 2.5 indicates dilation, and ≥ 2.5 indicates aneurysm abnormality.¹) An ultrasound of the right upper quadrant revealed an enlarged/elongated gallbladder. The patient therefore met clinical criteria for Kawasaki disease.

DISCUSSION

Kawasaki disease is a self-limited vasculitis of childhood and the leading cause of acquired heart disease in children in developed countries.¹ The annual incidence of Kawasaki disease in North America is about 25 cases per 100,000 children < 5 years of age.¹ In the United States, incidence is highest in Asian and Pacific Islander populations (30 per 100,000) and is particularly high among those of Japanese ancestry (~200 per 100,000).² Disease prevalence is also noteworthy in Non-Hispanic African American (17 per 100,000) and Hispanic (16 per 100,000) populations.²

Diagnosis of Kawasaki disease requires presence of fever lasting at least 5 days and at least 4 of the following: bilateral bulbar conjunctival injection, oral mucous membrane changes (erythematous or cracked lips, erythematous pharynx, strawberry tongue), peripheral extremity changes (erythema of palms or soles, edema of hands or feet, and/or periungual desquamation), diffuse maculopapular rash, and cervical lymphadenopathy (≥ 1.5 cm, often unilateral). If ≥ 4 criteria are met, Kawasaki disease can be diagnosed on the fourth day of illness.¹

Continue to: Laboratory findings suggesting...

Laboratory findings suggesting ­Kawasaki disease include a WBC count ≥ 15,000/mcL, normocytic, normochromic anemia, platelets ≥ 450,000/mcL after 7 days of illness, sterile pyuria (≥ 10 WBCs/high-power field), serum alanine aminotransferase level > 50 U/L, and serum albumin ≤ 3 g/dL.

Cardiac abnormalities are not included in the diagnostic criteria for Kawasaki disease but provide evidence in cases of incomplete Kawasaki disease if ≥ 4 criteria are not met and there is strong clinical suspicion.¹ Incomplete Kawasaki disease should be considered in a patient with a CRP level ≥ 3 mg/dL and/or ESR ≥ 40 mm/h, ≥ 3 supplemental laboratory criteria, or a positive echocardiogram.¹

Ultrasound imaging may reveal cervical lymph nodes resembling a “cluster of grapes.”³ The case patient’s imaging showed a “chain of enlarged lymph nodes.” She likely had gallbladder “hydrops” due to its increased longitudinal and horizontal diameter and lack of other anatomic changes.⁴

Prompt treatment is essential

Treatment for complete and incomplete Kawasaki disease is a single high dose of intravenous immunoglobulin (IVIG) along with aspirin. Patients meeting criteria should be treated as soon as the diagnosis is established.⁵ A single high dose of IVIG (2 g/kg), administered over 10 to 12 hours, should be initiated within 5 to 10 days of disease onset. Administering IVIG in the acute phase of Kawasaki disease reduces the prevalence of coronary artery abnormalities.⁶ Corticosteroids may be used as adjunctive therapy for patients with high risk of IVIG resistance.^1,7-9

Our patient was not deemed to be at high risk for IVIG resistance (Non-Japanese patient, age at fever onset > 6 months, absence of coronary artery aneurysm⁹) and was administered IVIG on Day 4. She was also given moderate-dose aspirin, then later transitioned to low-dose aspirin. The patient’s fevers improved, she was less irritable, and she had periods of playfulness. Physical exam then showed erythematous and cracked lips with peeling skin.

Continue to: The patient was discharged...

Untreated children with Kawasaki disease have a 25% chance of developing coronary artery aneurysms.

The patient was discharged home on Day 8, after her fever resolved, with instructions to continue low-dose aspirin and to obtain a repeat echocardiogram, gallbladder ultrasound, and lab work in 2 weeks. At her follow-up appointment, periungual desquamation was noted, and ultrasound showed continued enlarged/elongated gallbladder. A repeat echocardiogram was not available. Overall, the patient recovered from Kawasaki disease after therapeutic intervention.

THE TAKEAWAY

Kawasaki disease can be relatively rare in North American populations, but it is important for physicians to be able to recognize and treat it. Untreated children have a 25% chance of developing coronary artery aneurysms.^1,10,11 Early treatment with IVIG can decrease risk to 5%, resulting in an excellent medium- to long-term prognosis for patients.¹⁰ Thorough physical examination and an appropriate degree of clinical suspicion was key in this case of Kawasaki disease.

CORRESPONDENCE
Taisha Doo, MD, 1401 Madison Street, Suite #100, Seattle, WA 98104; [email protected]

THE CASE

A previously healthy 3-year-old girl presented to the emergency department with 4 days of fever and 2 days of right-side neck pain. The maximum temperature at home was 103 °F. The patient was irritable and vomited once. There were no other apparent or reported symptoms.

The neck exam was notable for nonfluctuant, swollen, and tender lymph nodes on the right side. Her sclera and conjunctiva were clear, and her oropharynx was unremarkable. Lab work revealed leukocytosis, with a white blood cell (WBC) count of 15.5 × 10³/µL (normal range, 4.0-10.0 × 10³/µL). She was given one 20 cc/kg normal saline bolus, started on intravenous clindamycin for presumed cervical lymphadenitis, and admitted to the hospital.

On Day 2, the patient developed a fine maculopapular rash on her chest, abdomen, and back. She had spiking fevers—as high as 102.2 °F—despite being on antibiotics for more than 24 hours. The erythrocyte sedimentation rate (ESR) was 39 mm/h (0-20 mm/h), and C-reactive protein (CRP) was 71.4 mg/L (0.0-4.9 mg/L). Due to concern for abscess, a neck ultrasound was performed; it showed a chain of enlarged lymph nodes in the right neck (largest, 2.3 × 1.1 × 1.4 cm) and no abscess.

On Day 3, clindamycin was switched to intravenous ampicillin/sulbactam because a nasal swab for methicillin-resistant Staphylococcus aureus was negative. A swab for respiratory viral infections was also negative. The patient then developed notable facial swelling, bilateral bulbar conjunctival injection with limbic sparing, and swelling of her hands and feet.

THE DIAGNOSIS

By the end of Day 3, the patient’s lab studies demonstrated microcytic anemia and low albumin (2.5 g/dL), but no transaminitis, thrombocytosis, or sterile pyuria. An electrocardiogram was unremarkable. A pediatric echocardiogram revealed hyperemic coronaries, indicating inflammation. The coronary arteries were measured in the upper limits of normal, and the patient’s Z-scores were < 2.5. (A Z-score < 2 indicates no involvement, 2 to < 2.5 indicates dilation, and ≥ 2.5 indicates aneurysm abnormality.¹) An ultrasound of the right upper quadrant revealed an enlarged/elongated gallbladder. The patient therefore met clinical criteria for Kawasaki disease.

DISCUSSION

Kawasaki disease is a self-limited vasculitis of childhood and the leading cause of acquired heart disease in children in developed countries.¹ The annual incidence of Kawasaki disease in North America is about 25 cases per 100,000 children < 5 years of age.¹ In the United States, incidence is highest in Asian and Pacific Islander populations (30 per 100,000) and is particularly high among those of Japanese ancestry (~200 per 100,000).² Disease prevalence is also noteworthy in Non-Hispanic African American (17 per 100,000) and Hispanic (16 per 100,000) populations.²

Diagnosis of Kawasaki disease requires presence of fever lasting at least 5 days and at least 4 of the following: bilateral bulbar conjunctival injection, oral mucous membrane changes (erythematous or cracked lips, erythematous pharynx, strawberry tongue), peripheral extremity changes (erythema of palms or soles, edema of hands or feet, and/or periungual desquamation), diffuse maculopapular rash, and cervical lymphadenopathy (≥ 1.5 cm, often unilateral). If ≥ 4 criteria are met, Kawasaki disease can be diagnosed on the fourth day of illness.¹

Continue to: Laboratory findings suggesting...

Laboratory findings suggesting ­Kawasaki disease include a WBC count ≥ 15,000/mcL, normocytic, normochromic anemia, platelets ≥ 450,000/mcL after 7 days of illness, sterile pyuria (≥ 10 WBCs/high-power field), serum alanine aminotransferase level > 50 U/L, and serum albumin ≤ 3 g/dL.

Cardiac abnormalities are not included in the diagnostic criteria for Kawasaki disease but provide evidence in cases of incomplete Kawasaki disease if ≥ 4 criteria are not met and there is strong clinical suspicion.¹ Incomplete Kawasaki disease should be considered in a patient with a CRP level ≥ 3 mg/dL and/or ESR ≥ 40 mm/h, ≥ 3 supplemental laboratory criteria, or a positive echocardiogram.¹

Ultrasound imaging may reveal cervical lymph nodes resembling a “cluster of grapes.”³ The case patient’s imaging showed a “chain of enlarged lymph nodes.” She likely had gallbladder “hydrops” due to its increased longitudinal and horizontal diameter and lack of other anatomic changes.⁴

Prompt treatment is essential

Treatment for complete and incomplete Kawasaki disease is a single high dose of intravenous immunoglobulin (IVIG) along with aspirin. Patients meeting criteria should be treated as soon as the diagnosis is established.⁵ A single high dose of IVIG (2 g/kg), administered over 10 to 12 hours, should be initiated within 5 to 10 days of disease onset. Administering IVIG in the acute phase of Kawasaki disease reduces the prevalence of coronary artery abnormalities.⁶ Corticosteroids may be used as adjunctive therapy for patients with high risk of IVIG resistance.^1,7-9

Our patient was not deemed to be at high risk for IVIG resistance (Non-Japanese patient, age at fever onset > 6 months, absence of coronary artery aneurysm⁹) and was administered IVIG on Day 4. She was also given moderate-dose aspirin, then later transitioned to low-dose aspirin. The patient’s fevers improved, she was less irritable, and she had periods of playfulness. Physical exam then showed erythematous and cracked lips with peeling skin.

Continue to: The patient was discharged...

Untreated children with Kawasaki disease have a 25% chance of developing coronary artery aneurysms.

The patient was discharged home on Day 8, after her fever resolved, with instructions to continue low-dose aspirin and to obtain a repeat echocardiogram, gallbladder ultrasound, and lab work in 2 weeks. At her follow-up appointment, periungual desquamation was noted, and ultrasound showed continued enlarged/elongated gallbladder. A repeat echocardiogram was not available. Overall, the patient recovered from Kawasaki disease after therapeutic intervention.

THE TAKEAWAY

Kawasaki disease can be relatively rare in North American populations, but it is important for physicians to be able to recognize and treat it. Untreated children have a 25% chance of developing coronary artery aneurysms.^1,10,11 Early treatment with IVIG can decrease risk to 5%, resulting in an excellent medium- to long-term prognosis for patients.¹⁰ Thorough physical examination and an appropriate degree of clinical suspicion was key in this case of Kawasaki disease.

CORRESPONDENCE
Taisha Doo, MD, 1401 Madison Street, Suite #100, Seattle, WA 98104; [email protected]

THE CASE

A previously healthy 3-year-old girl presented to the emergency department with 4 days of fever and 2 days of right-side neck pain. The maximum temperature at home was 103 °F. The patient was irritable and vomited once. There were no other apparent or reported symptoms.

The neck exam was notable for nonfluctuant, swollen, and tender lymph nodes on the right side. Her sclera and conjunctiva were clear, and her oropharynx was unremarkable. Lab work revealed leukocytosis, with a white blood cell (WBC) count of 15.5 × 10³/µL (normal range, 4.0-10.0 × 10³/µL). She was given one 20 cc/kg normal saline bolus, started on intravenous clindamycin for presumed cervical lymphadenitis, and admitted to the hospital.

On Day 2, the patient developed a fine maculopapular rash on her chest, abdomen, and back. She had spiking fevers—as high as 102.2 °F—despite being on antibiotics for more than 24 hours. The erythrocyte sedimentation rate (ESR) was 39 mm/h (0-20 mm/h), and C-reactive protein (CRP) was 71.4 mg/L (0.0-4.9 mg/L). Due to concern for abscess, a neck ultrasound was performed; it showed a chain of enlarged lymph nodes in the right neck (largest, 2.3 × 1.1 × 1.4 cm) and no abscess.

On Day 3, clindamycin was switched to intravenous ampicillin/sulbactam because a nasal swab for methicillin-resistant Staphylococcus aureus was negative. A swab for respiratory viral infections was also negative. The patient then developed notable facial swelling, bilateral bulbar conjunctival injection with limbic sparing, and swelling of her hands and feet.

THE DIAGNOSIS

By the end of Day 3, the patient’s lab studies demonstrated microcytic anemia and low albumin (2.5 g/dL), but no transaminitis, thrombocytosis, or sterile pyuria. An electrocardiogram was unremarkable. A pediatric echocardiogram revealed hyperemic coronaries, indicating inflammation. The coronary arteries were measured in the upper limits of normal, and the patient’s Z-scores were < 2.5. (A Z-score < 2 indicates no involvement, 2 to < 2.5 indicates dilation, and ≥ 2.5 indicates aneurysm abnormality.¹) An ultrasound of the right upper quadrant revealed an enlarged/elongated gallbladder. The patient therefore met clinical criteria for Kawasaki disease.

DISCUSSION

Kawasaki disease is a self-limited vasculitis of childhood and the leading cause of acquired heart disease in children in developed countries.¹ The annual incidence of Kawasaki disease in North America is about 25 cases per 100,000 children < 5 years of age.¹ In the United States, incidence is highest in Asian and Pacific Islander populations (30 per 100,000) and is particularly high among those of Japanese ancestry (~200 per 100,000).² Disease prevalence is also noteworthy in Non-Hispanic African American (17 per 100,000) and Hispanic (16 per 100,000) populations.²

Diagnosis of Kawasaki disease requires presence of fever lasting at least 5 days and at least 4 of the following: bilateral bulbar conjunctival injection, oral mucous membrane changes (erythematous or cracked lips, erythematous pharynx, strawberry tongue), peripheral extremity changes (erythema of palms or soles, edema of hands or feet, and/or periungual desquamation), diffuse maculopapular rash, and cervical lymphadenopathy (≥ 1.5 cm, often unilateral). If ≥ 4 criteria are met, Kawasaki disease can be diagnosed on the fourth day of illness.¹

Continue to: Laboratory findings suggesting...

Laboratory findings suggesting ­Kawasaki disease include a WBC count ≥ 15,000/mcL, normocytic, normochromic anemia, platelets ≥ 450,000/mcL after 7 days of illness, sterile pyuria (≥ 10 WBCs/high-power field), serum alanine aminotransferase level > 50 U/L, and serum albumin ≤ 3 g/dL.

Cardiac abnormalities are not included in the diagnostic criteria for Kawasaki disease but provide evidence in cases of incomplete Kawasaki disease if ≥ 4 criteria are not met and there is strong clinical suspicion.¹ Incomplete Kawasaki disease should be considered in a patient with a CRP level ≥ 3 mg/dL and/or ESR ≥ 40 mm/h, ≥ 3 supplemental laboratory criteria, or a positive echocardiogram.¹

Ultrasound imaging may reveal cervical lymph nodes resembling a “cluster of grapes.”³ The case patient’s imaging showed a “chain of enlarged lymph nodes.” She likely had gallbladder “hydrops” due to its increased longitudinal and horizontal diameter and lack of other anatomic changes.⁴

Prompt treatment is essential

Treatment for complete and incomplete Kawasaki disease is a single high dose of intravenous immunoglobulin (IVIG) along with aspirin. Patients meeting criteria should be treated as soon as the diagnosis is established.⁵ A single high dose of IVIG (2 g/kg), administered over 10 to 12 hours, should be initiated within 5 to 10 days of disease onset. Administering IVIG in the acute phase of Kawasaki disease reduces the prevalence of coronary artery abnormalities.⁶ Corticosteroids may be used as adjunctive therapy for patients with high risk of IVIG resistance.^1,7-9

Our patient was not deemed to be at high risk for IVIG resistance (Non-Japanese patient, age at fever onset > 6 months, absence of coronary artery aneurysm⁹) and was administered IVIG on Day 4. She was also given moderate-dose aspirin, then later transitioned to low-dose aspirin. The patient’s fevers improved, she was less irritable, and she had periods of playfulness. Physical exam then showed erythematous and cracked lips with peeling skin.

Continue to: The patient was discharged...

Untreated children with Kawasaki disease have a 25% chance of developing coronary artery aneurysms.

The patient was discharged home on Day 8, after her fever resolved, with instructions to continue low-dose aspirin and to obtain a repeat echocardiogram, gallbladder ultrasound, and lab work in 2 weeks. At her follow-up appointment, periungual desquamation was noted, and ultrasound showed continued enlarged/elongated gallbladder. A repeat echocardiogram was not available. Overall, the patient recovered from Kawasaki disease after therapeutic intervention.

THE TAKEAWAY

Kawasaki disease can be relatively rare in North American populations, but it is important for physicians to be able to recognize and treat it. Untreated children have a 25% chance of developing coronary artery aneurysms.^1,10,11 Early treatment with IVIG can decrease risk to 5%, resulting in an excellent medium- to long-term prognosis for patients.¹⁰ Thorough physical examination and an appropriate degree of clinical suspicion was key in this case of Kawasaki disease.

CORRESPONDENCE
Taisha Doo, MD, 1401 Madison Street, Suite #100, Seattle, WA 98104; [email protected]

ILLUSTRATIVE CASE

A 61-year-old man with hypertension and paroxysmal AF presents to your office shortly after experiencing his third episode of AF in the past 6 months. He describes these episodes, which last for several days, as “just awful,” noting that when he experiences AF, he has fatigue, palpitations, and shortness of breath and “can’t stop paying attention to my heart.” The patient, who has a body mass index of 32, consumes more than 15 alcoholic drinks per week. What can you recommend to him that will decrease his likelihood of experiencing more episodes of AF?

AF is the most common sustained cardiac arrhythmia. It is associated with significant morbidity and mortality. Known risk factors include obesity, physical inactivity, sleep apnea, diabetes, and hypertension.²

According to the Centers for Disease Control and Prevention, an estimated 12.1 million people in the United States will have AF by 2030. In 2018, AF was mentioned on more than 183,000 death certificates and was the underlying cause of more than 26,000 of those deaths.³ AF is the primary diagnosis in 450,000 hospitalizations annually,⁴ and the death rate from AF as the primary or contributing cause of death has been rising for more than 2 decades.³

More than 50% of Americans report alcohol consumption within the past month.⁵ Although alcohol use is associated with new and recurrent AF, only limited prospective data show a clear and causal association between abstaining from alcohol and decreasing AF recurrence.

STUDY SUMMARY

Reduction in AF recurrence and total AF burden following alcohol abstinence

This multicenter, prospective, open-label, randomized controlled trial (N = 140) from 6 sites in Australia evaluated the impact of alcohol abstinence on both the recurrence of AF and the amount of time in AF. Study participants were ages 18 to 85 years, consumed 10 or more standard alcohol-containing drinks per week, had paroxysmal or persistent AF, and were in sinus rhythm at the time of enrollment, regardless of antiarrhythmic therapy. Exclusion criteria included alcohol dependence or abuse, severe left ventricular systolic dysfunction (ejection fraction < 35%), clinically significant noncardiac illness, and/or coexisting psychiatric disorder.¹

After a 4-week run-in period, patients were randomized to either an abstinence or a control group in a 1:1 fashion. Patients enrolled in the abstinence group were encouraged to abstain from alcohol consumption for 6 months and were provided with written and oral instructions to assist with abstaining. Control group patients continued their same level of alcohol consumption. Comprehensive rhythm monitoring occurred for all patients after randomization.

The largest challenge to implementation of this intervention is most likely the willingness of patients to cut their alcohol consumption.

Alcohol consumption was reported by both groups using a weekly alcohol diary, supplemented with a visual guide showing pictures of standard alcohol drinks. For the abstinence group, random urine testing for ethyl glucuronide (an alcohol metabolite) was possible if no alcohol intake was reported. Primary outcomes during the 6-month study included recurrence of AF and total AF burden (percentage of time in AF).

Continue to: Secondary outcomes included hospitalizations...

Secondary outcomes included hospitalizations for AF, AF symptom severity, and change in weight. Blood pressure, quality-of-life, and depression scores were missing for > 35% of patients.¹

Patients were randomized evenly to the control and abstinence groups. The typical patient was an overweight male in his early 60s with paroxysmal AF, who was taking an antiarrhythmic agent. Patients in the abstinence group decreased their alcohol consumption from 16.8 to 2.1 drinks per week (87.5% reduction; mean difference = –14.7; 95% CI, –12.7 to –16.7). Patients in the control group reduced their intake from 16.4 to 13.2 drinks per week (19.5% reduction; mean difference = –3.2; 95% CI, –1.9 to –4.4).¹

AF recurred in 53% vs 73% of the abstinence and control groups, respectively, with a longer period before recurrence in the abstinence group than in the control group (hazard ratio = 0.55; 95% CI, 0.36-0.84; P = .005; number needed to treat = 5). The AF burden was also lower in the abstinence group (0.5%; interquartile range [IQR] = 0.0-3.0) than in the control group (1.2%; IQR = 0.0-10.3; P = .01). The abstinence group had a lower percentage of AF hospitalizations compared with the control group (9% vs 20%), and fewer patients reporting moderate or severe AF symptoms (10% vs 32%). In addition, the abstinence group lost 3.7 kg more weight than did the control group at 6 months.¹

WHAT’S NEW

Objective new evidence for effective patient counseling

Alcohol consumption and its association with the onset and recurrence of AF has been documented previously.⁶ This study was the first to prospectively examine if abstaining from alcohol reduces paroxysmal AF episodes in moderate drinkers.

The study identified clinically meaningful findings among those who abstained from alcohol, including decreased AF recurrence rates, increased time to recurrence, and lower overall AF burden. This provides objective evidence that can be used for motivational interviewing in patients with paroxysmal AF who may be receptive to reducing or abstaining from alcohol consumption.

Continue to: CAVEATS

The narrow study population may not be widely applicable

The study population was predominantly male, in their seventh decade of life (mean age, 61), and living in Australia. Rates of AF and symptomatology differ by gender and age, making this information challenging to apply to women or older populations. The study excluded patients with alcohol dependence or abuse, left ventricular systolic dysfunction (ejection fraction < 35%), coexisting psychiatric disorders, and clinically significant noncardiac illnesses, limiting the study’s generalizability to these patient populations. Overall, AF recurrence was low in both groups despite the intervention, and the study did not evaluate the efficacy of the counseling method for abstinence.

Since publication of this article, a prospective cohort study of approximately 3800 Swiss patients with AF evaluated the effect of alcohol consumption on the rate of stroke and embolic events. That study did not find statistically significant correlations between patients who drank no alcohol per day, > 0 to < 1, 1 to < 2, or ≥ 2 drinks per day and their rate of stroke.⁷ However, this study did not specifically evaluate the rate of AF recurrence or time spent in AF among the cohort, which is clinically meaningful for patient morbidity.¹

CHALLENGES TO IMPLEMENTATION

Patient willingness to cut alcohol consumption may be limited

The largest challenge to implementation of this intervention is most likely the willingness of patients to cut their alcohol consumption. In this study population, 697 patients were screened for enrollment and met inclusion criteria; however, 491 patients (70.4%) were not willing to consider abstinence from alcohol, and after the run-in phase, another 17 declined randomization. Many primary care physicians would likely agree that while it is easy to encourage patients to drink less, patient adherence to these recommendations, particularly abstaining, is likely to be limited.

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 61-year-old man with hypertension and paroxysmal AF presents to your office shortly after experiencing his third episode of AF in the past 6 months. He describes these episodes, which last for several days, as “just awful,” noting that when he experiences AF, he has fatigue, palpitations, and shortness of breath and “can’t stop paying attention to my heart.” The patient, who has a body mass index of 32, consumes more than 15 alcoholic drinks per week. What can you recommend to him that will decrease his likelihood of experiencing more episodes of AF?

AF is the most common sustained cardiac arrhythmia. It is associated with significant morbidity and mortality. Known risk factors include obesity, physical inactivity, sleep apnea, diabetes, and hypertension.²

According to the Centers for Disease Control and Prevention, an estimated 12.1 million people in the United States will have AF by 2030. In 2018, AF was mentioned on more than 183,000 death certificates and was the underlying cause of more than 26,000 of those deaths.³ AF is the primary diagnosis in 450,000 hospitalizations annually,⁴ and the death rate from AF as the primary or contributing cause of death has been rising for more than 2 decades.³

More than 50% of Americans report alcohol consumption within the past month.⁵ Although alcohol use is associated with new and recurrent AF, only limited prospective data show a clear and causal association between abstaining from alcohol and decreasing AF recurrence.

STUDY SUMMARY

Reduction in AF recurrence and total AF burden following alcohol abstinence

This multicenter, prospective, open-label, randomized controlled trial (N = 140) from 6 sites in Australia evaluated the impact of alcohol abstinence on both the recurrence of AF and the amount of time in AF. Study participants were ages 18 to 85 years, consumed 10 or more standard alcohol-containing drinks per week, had paroxysmal or persistent AF, and were in sinus rhythm at the time of enrollment, regardless of antiarrhythmic therapy. Exclusion criteria included alcohol dependence or abuse, severe left ventricular systolic dysfunction (ejection fraction < 35%), clinically significant noncardiac illness, and/or coexisting psychiatric disorder.¹

After a 4-week run-in period, patients were randomized to either an abstinence or a control group in a 1:1 fashion. Patients enrolled in the abstinence group were encouraged to abstain from alcohol consumption for 6 months and were provided with written and oral instructions to assist with abstaining. Control group patients continued their same level of alcohol consumption. Comprehensive rhythm monitoring occurred for all patients after randomization.

The largest challenge to implementation of this intervention is most likely the willingness of patients to cut their alcohol consumption.

Alcohol consumption was reported by both groups using a weekly alcohol diary, supplemented with a visual guide showing pictures of standard alcohol drinks. For the abstinence group, random urine testing for ethyl glucuronide (an alcohol metabolite) was possible if no alcohol intake was reported. Primary outcomes during the 6-month study included recurrence of AF and total AF burden (percentage of time in AF).

Continue to: Secondary outcomes included hospitalizations...

Secondary outcomes included hospitalizations for AF, AF symptom severity, and change in weight. Blood pressure, quality-of-life, and depression scores were missing for > 35% of patients.¹

Patients were randomized evenly to the control and abstinence groups. The typical patient was an overweight male in his early 60s with paroxysmal AF, who was taking an antiarrhythmic agent. Patients in the abstinence group decreased their alcohol consumption from 16.8 to 2.1 drinks per week (87.5% reduction; mean difference = –14.7; 95% CI, –12.7 to –16.7). Patients in the control group reduced their intake from 16.4 to 13.2 drinks per week (19.5% reduction; mean difference = –3.2; 95% CI, –1.9 to –4.4).¹

AF recurred in 53% vs 73% of the abstinence and control groups, respectively, with a longer period before recurrence in the abstinence group than in the control group (hazard ratio = 0.55; 95% CI, 0.36-0.84; P = .005; number needed to treat = 5). The AF burden was also lower in the abstinence group (0.5%; interquartile range [IQR] = 0.0-3.0) than in the control group (1.2%; IQR = 0.0-10.3; P = .01). The abstinence group had a lower percentage of AF hospitalizations compared with the control group (9% vs 20%), and fewer patients reporting moderate or severe AF symptoms (10% vs 32%). In addition, the abstinence group lost 3.7 kg more weight than did the control group at 6 months.¹

WHAT’S NEW

Objective new evidence for effective patient counseling

Alcohol consumption and its association with the onset and recurrence of AF has been documented previously.⁶ This study was the first to prospectively examine if abstaining from alcohol reduces paroxysmal AF episodes in moderate drinkers.

The study identified clinically meaningful findings among those who abstained from alcohol, including decreased AF recurrence rates, increased time to recurrence, and lower overall AF burden. This provides objective evidence that can be used for motivational interviewing in patients with paroxysmal AF who may be receptive to reducing or abstaining from alcohol consumption.

Continue to: CAVEATS

The narrow study population may not be widely applicable

The study population was predominantly male, in their seventh decade of life (mean age, 61), and living in Australia. Rates of AF and symptomatology differ by gender and age, making this information challenging to apply to women or older populations. The study excluded patients with alcohol dependence or abuse, left ventricular systolic dysfunction (ejection fraction < 35%), coexisting psychiatric disorders, and clinically significant noncardiac illnesses, limiting the study’s generalizability to these patient populations. Overall, AF recurrence was low in both groups despite the intervention, and the study did not evaluate the efficacy of the counseling method for abstinence.

Since publication of this article, a prospective cohort study of approximately 3800 Swiss patients with AF evaluated the effect of alcohol consumption on the rate of stroke and embolic events. That study did not find statistically significant correlations between patients who drank no alcohol per day, > 0 to < 1, 1 to < 2, or ≥ 2 drinks per day and their rate of stroke.⁷ However, this study did not specifically evaluate the rate of AF recurrence or time spent in AF among the cohort, which is clinically meaningful for patient morbidity.¹

CHALLENGES TO IMPLEMENTATION

Patient willingness to cut alcohol consumption may be limited

The largest challenge to implementation of this intervention is most likely the willingness of patients to cut their alcohol consumption. In this study population, 697 patients were screened for enrollment and met inclusion criteria; however, 491 patients (70.4%) were not willing to consider abstinence from alcohol, and after the run-in phase, another 17 declined randomization. Many primary care physicians would likely agree that while it is easy to encourage patients to drink less, patient adherence to these recommendations, particularly abstaining, is likely to be limited.

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 61-year-old man with hypertension and paroxysmal AF presents to your office shortly after experiencing his third episode of AF in the past 6 months. He describes these episodes, which last for several days, as “just awful,” noting that when he experiences AF, he has fatigue, palpitations, and shortness of breath and “can’t stop paying attention to my heart.” The patient, who has a body mass index of 32, consumes more than 15 alcoholic drinks per week. What can you recommend to him that will decrease his likelihood of experiencing more episodes of AF?

AF is the most common sustained cardiac arrhythmia. It is associated with significant morbidity and mortality. Known risk factors include obesity, physical inactivity, sleep apnea, diabetes, and hypertension.²

According to the Centers for Disease Control and Prevention, an estimated 12.1 million people in the United States will have AF by 2030. In 2018, AF was mentioned on more than 183,000 death certificates and was the underlying cause of more than 26,000 of those deaths.³ AF is the primary diagnosis in 450,000 hospitalizations annually,⁴ and the death rate from AF as the primary or contributing cause of death has been rising for more than 2 decades.³

More than 50% of Americans report alcohol consumption within the past month.⁵ Although alcohol use is associated with new and recurrent AF, only limited prospective data show a clear and causal association between abstaining from alcohol and decreasing AF recurrence.

STUDY SUMMARY

Reduction in AF recurrence and total AF burden following alcohol abstinence

This multicenter, prospective, open-label, randomized controlled trial (N = 140) from 6 sites in Australia evaluated the impact of alcohol abstinence on both the recurrence of AF and the amount of time in AF. Study participants were ages 18 to 85 years, consumed 10 or more standard alcohol-containing drinks per week, had paroxysmal or persistent AF, and were in sinus rhythm at the time of enrollment, regardless of antiarrhythmic therapy. Exclusion criteria included alcohol dependence or abuse, severe left ventricular systolic dysfunction (ejection fraction < 35%), clinically significant noncardiac illness, and/or coexisting psychiatric disorder.¹

After a 4-week run-in period, patients were randomized to either an abstinence or a control group in a 1:1 fashion. Patients enrolled in the abstinence group were encouraged to abstain from alcohol consumption for 6 months and were provided with written and oral instructions to assist with abstaining. Control group patients continued their same level of alcohol consumption. Comprehensive rhythm monitoring occurred for all patients after randomization.

The largest challenge to implementation of this intervention is most likely the willingness of patients to cut their alcohol consumption.

Alcohol consumption was reported by both groups using a weekly alcohol diary, supplemented with a visual guide showing pictures of standard alcohol drinks. For the abstinence group, random urine testing for ethyl glucuronide (an alcohol metabolite) was possible if no alcohol intake was reported. Primary outcomes during the 6-month study included recurrence of AF and total AF burden (percentage of time in AF).

Continue to: Secondary outcomes included hospitalizations...

Secondary outcomes included hospitalizations for AF, AF symptom severity, and change in weight. Blood pressure, quality-of-life, and depression scores were missing for > 35% of patients.¹

Patients were randomized evenly to the control and abstinence groups. The typical patient was an overweight male in his early 60s with paroxysmal AF, who was taking an antiarrhythmic agent. Patients in the abstinence group decreased their alcohol consumption from 16.8 to 2.1 drinks per week (87.5% reduction; mean difference = –14.7; 95% CI, –12.7 to –16.7). Patients in the control group reduced their intake from 16.4 to 13.2 drinks per week (19.5% reduction; mean difference = –3.2; 95% CI, –1.9 to –4.4).¹

AF recurred in 53% vs 73% of the abstinence and control groups, respectively, with a longer period before recurrence in the abstinence group than in the control group (hazard ratio = 0.55; 95% CI, 0.36-0.84; P = .005; number needed to treat = 5). The AF burden was also lower in the abstinence group (0.5%; interquartile range [IQR] = 0.0-3.0) than in the control group (1.2%; IQR = 0.0-10.3; P = .01). The abstinence group had a lower percentage of AF hospitalizations compared with the control group (9% vs 20%), and fewer patients reporting moderate or severe AF symptoms (10% vs 32%). In addition, the abstinence group lost 3.7 kg more weight than did the control group at 6 months.¹

WHAT’S NEW

Objective new evidence for effective patient counseling

Alcohol consumption and its association with the onset and recurrence of AF has been documented previously.⁶ This study was the first to prospectively examine if abstaining from alcohol reduces paroxysmal AF episodes in moderate drinkers.

The study identified clinically meaningful findings among those who abstained from alcohol, including decreased AF recurrence rates, increased time to recurrence, and lower overall AF burden. This provides objective evidence that can be used for motivational interviewing in patients with paroxysmal AF who may be receptive to reducing or abstaining from alcohol consumption.

Continue to: CAVEATS

The narrow study population may not be widely applicable

The study population was predominantly male, in their seventh decade of life (mean age, 61), and living in Australia. Rates of AF and symptomatology differ by gender and age, making this information challenging to apply to women or older populations. The study excluded patients with alcohol dependence or abuse, left ventricular systolic dysfunction (ejection fraction < 35%), coexisting psychiatric disorders, and clinically significant noncardiac illnesses, limiting the study’s generalizability to these patient populations. Overall, AF recurrence was low in both groups despite the intervention, and the study did not evaluate the efficacy of the counseling method for abstinence.

Since publication of this article, a prospective cohort study of approximately 3800 Swiss patients with AF evaluated the effect of alcohol consumption on the rate of stroke and embolic events. That study did not find statistically significant correlations between patients who drank no alcohol per day, > 0 to < 1, 1 to < 2, or ≥ 2 drinks per day and their rate of stroke.⁷ However, this study did not specifically evaluate the rate of AF recurrence or time spent in AF among the cohort, which is clinically meaningful for patient morbidity.¹

CHALLENGES TO IMPLEMENTATION

Patient willingness to cut alcohol consumption may be limited

The largest challenge to implementation of this intervention is most likely the willingness of patients to cut their alcohol consumption. In this study population, 697 patients were screened for enrollment and met inclusion criteria; however, 491 patients (70.4%) were not willing to consider abstinence from alcohol, and after the run-in phase, another 17 declined randomization. Many primary care physicians would likely agree that while it is easy to encourage patients to drink less, patient adherence to these recommendations, particularly abstaining, is likely to be limited.

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.

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

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

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

Moving patients toward completion of advance directives

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

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

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

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

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

Continue to: When ADs are executed...

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

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

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

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

Improve quality of life for patients with any serious illness

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

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

Continue to: Early EOL discussions...

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

Timing and communication are key

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

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

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

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

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

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

How to navigate EOL discussions with patients with dementia

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

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

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

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

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

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

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

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

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

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

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

Facilitate discussions that improve outcomes

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

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

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

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

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

Moving patients toward completion of advance directives

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

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

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

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

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

Continue to: When ADs are executed...

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

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

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

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

Improve quality of life for patients with any serious illness

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

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

Continue to: Early EOL discussions...

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

Timing and communication are key

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

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

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

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

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

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

How to navigate EOL discussions with patients with dementia

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

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

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

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

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

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

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

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

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

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

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

Facilitate discussions that improve outcomes

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

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

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

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

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

Moving patients toward completion of advance directives

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

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

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

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

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

Continue to: When ADs are executed...

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

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

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

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

Improve quality of life for patients with any serious illness

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

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

Continue to: Early EOL discussions...

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

Timing and communication are key

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

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

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

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

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

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

How to navigate EOL discussions with patients with dementia

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

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

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

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

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

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

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

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

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

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

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

Facilitate discussions that improve outcomes

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

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

Gestational diabetes mellitus (GDM), defined as new-onset hyperglycemia detected in a pregnant woman after 24 weeks of gestation, affects 4% to 10% of pregnancies in the United States annually¹ and is a major challenge for health care professionals.² During pregnancy, the body’s physiologic responses are altered to support the growing fetus. One of these changes is an increase in insulin resistance, which suggests that pregnancy alone increases the patient’s risk for type 2 diabetes (T2D). However, several other factors also increase this risk, including maternal age, social barriers to care, obesity, poor weight control, and family history.

Copyright Dave Cutler

If not controlled, GDM results in poor health outcomes for the mother, such as preeclampsia, preterm labor, and maternal T2D.^3-5 For the infant, intrauterine exposure to persistent hyperglycemia is correlated with neonatal macrosomia, hypoglycemia, perinatal complications (eg, preterm delivery, fetal demise), and obesity and insulin resistance later in life.⁴

Primary care physicians (PCPs) are the patient’s main point of contact prior to pregnancy. This relationship makes PCPs a resource for the patient and specialists during and after pregnancy. In this article, we discuss risk factors and how to screen for GDM, provide an update on practice recommendations for treatment and management of GDM in primary care, and describe the effects of uncontrolled GDM.

Know the key risk factors

Prevention begins with identifying the major risk factors that contribute to the development of GDM. These include maternal age, social barriers to care, family history of prediabetes, and obesity and poor weight control.

Older age. A meta-analysis of 24 studies noted strong positive correlation between GDM risk and maternal age.⁶ One of the population-based cohort studies in the meta-analysis examined relationships between maternal age and pregnancy outcomes in women living in British Columbia, Canada (n = 203,414). Data suggested that the relative risk of GDM increased linearly with maternal age to 3.2, 4.2, and 4.4 among women ages ≥ 35, ≥ 40, and ≥ 45 years, respectively.⁷

Social barriers to care. Although the prevalence of GDM has increased over the past few decades,¹ from 2011 to 2019 the increase in GDM in individuals at first live birth was significantly higher in non-Hispanic Asian and Hispanic/Latina women than in non-Hispanic White women.⁸ Data from the Centers for Disease Control and Prevention further suggest that diabetes was more prevalent among individuals with a lower socioeconomic status as indicated by their level of education.⁹ Ogunwole et al¹⁰ suggest that racism is the root cause of these disparities and leads to long-term barriers to care (eg, socioeconomic deprivation, lack of health insurance, limited access to care, and poor health literacy), which ultimately contribute to the development of GDM and progression of diabetes. It is important for PCPs and all health professionals to be aware of these barriers so that they may practice mindfulness and deliver culturally sensitive care to patients from marginalized communities.

Family history of prediabetes. In a population-based cohort study (n = 7020), women with prediabetes (A1C, 5.7%-6.4%) were 2.8 times more likely to develop GDM compared with women with normal A1C (< 5.7%).¹¹ Similar results were seen in a retrospective cohort study (n = 2812), in which women with prediabetes were more likely than women with a normal first trimester A1C to have GDM (29.1% vs 13.7%, respectively; adjusted relative risk = 1.48; 95% CI, 1.15-1.89).¹² In both studies, prediabetes was not associated with a higher risk for adverse maternal or neonatal outcomes.^11,12

Continue to: While there are no current...

Women diagnosed with prediabetes in 1 study were found to have significantly less weight gain during pregnancy compared with patients with normal A1C, suggesting a benefit in early identification and intervention.

While there are no current guidelines for treating prediabetes in pregnancy, women diagnosed with prediabetes in 1 study were found to have significantly less weight gain during pregnancy compared with patients with normal A1C,¹² suggesting there may be a benefit in early identification and intervention, although further research is needed.¹¹ In a separate case-control study (n = 345 women with GDM; n = 800 control), high rates of gestational weight gain (> 0.41 kg/wk) were associated with an increased risk of GDM (odds ratio [OR] = 1.74; 95% CI, 1.16-2.60) compared with women with the lowest rate of gestational weight gain (0.27-0.4 kg/wk [OR = 1.43; 95% CI, 0.96-2.14]).¹³ Thus, it is helpful to have proactive conversations about family planning and adequate weight and glycemic control with high-risk patients to prepare for a healthy pregnancy.

Obesity and weight management. Patients who are overweight (body mass index [BMI], 25-29.9) or obese (BMI > 30) have a substantially increased risk of GDM (adjusted OR = 1.44; 95% CI, 1.04-1.81), as seen in a retrospective cohort study of 1951 pregnant Malaysian women.¹⁴ Several factors have been found to contribute to successful weight control, including calorie prescription, a structured meal plan, high physical activity goals (60-90 min/d), daily weighing and monitoring of food intake, behavior therapy, and continued patient–­provider contact.¹⁵

Most obstetricians use a 2-step method to screen for GDM with an initial 75-g oral glucose tolerance test, followed by a 50-g glucose load test if needed.

The safety, efficacy, and sustainability of weight loss with various dietary plans have been studied in individuals who are overweight and obese.¹⁶ Ultimately, energy expenditure must be greater than energy intake to promote weight loss. Conventional diets with continuous energy restriction (ie, low-fat, low-carbohydrate, and high-protein diets) have proven to be effective for short-term weight loss but data on long-term weight maintenance are limited.¹⁶ The Mediterranean diet, which is comprised mostly of vegetables, fruits, legumes, fish, and grains—with a lower intake of meat and dairy—may reduce gestational weight gain and risk of GDM as suggested by a randomized controlled trial (RCT; n = 1252).¹⁷ Although the choice of diet is up to the patient, it is important to be aware of different diets or refer the patient to a registered dietician who can help the patient if needed.

Reduce risk with adequate weight and glycemic control

Prevention of GDM during pregnancy should focus on weight maintenance and optimal glycemic control. Two systematic reviews, one with 8 RCTs (n = 1792) and another with 5 studies (n = 539), assessed the efficacy and safety of energy-restricted dietary intervention on GDM prevention.¹⁸ The first review found a significant reduction in gestational weight gain and improved glycemic control without increased risk of adverse maternal and fetal outcomes.¹⁸ The second review showed no clear difference between energy-restricted and non–energy-restricted diets on outcomes such as preeclampsia, gestational weight gain, large for gestational age, and macrosomia.¹⁸ These data suggest that while energy-restricted dietary interventions made no difference on maternal and fetal complications, they may still be safely used in pregnancy to reduce gestational weight gain and improve glycemic control.¹⁸

Once a woman is pregnant, it becomes difficult to lose weight because additional calories are needed to support a growing fetus. It is recommended that patients with healthy pregestational BMI consume an extra 200 to 300 calories/d after the first trimester. However, extra caloric intake in a woman with obesity who is pregnant leads to metabolic impairment and increased risk of diabetes for both the mother and fetus.¹⁹ Therefore, it is recommended that patients with obese pregestational BMI not consume additional calories because excess maternal fat is sufficient to support the energy needs of the growing fetus.¹⁹

Continue to: Ultimately, earlier intervention...

Ultimately, earlier intervention—prior to conception—helps patients prepare for a healthier pregnancy, resulting in better long-term outcomes. It is helpful to be familiar with the advantages and disadvantages of common approaches to weight management and to be able to refer patients to nutritionists for optimal planning. When establishing a dietary plan, consider patient-specific factors, such as cultural diets, financial and time constraints, and the patient’s readiness to make and maintain these changes. Consistent ­follow-up and behavioral therapy are necessary to maintain successful weight control.

There are many screening tools, but 1 is preferred in pregnancy

There are several ways to diagnose diabetes in patients who are not pregnant, including A1C, a fasting glucose test, an oral glucose tolerance test (OGTT), or random glucose testing (plus symptoms). However, the preferred method for diagnosing GDM is OGTT because it has a higher sensitivity.²⁰ A1C, while a good measure of hyperglycemic stability, does not register hyperglycemia early enough to diagnose GDM and fasting glucose testing is less sensitive because for most women with GDM, that abnormal postprandial glucose level is the first glycemic abnormality.²¹

When to screen. Blood glucose levels should be checked in all pregnant women as part of their metabolic panel at the first prenatal visit. A reflex A1C for high glucose levels can be ordered based on the physician’s preference. This may help you to identify patients with prediabetes who are at risk for GDM and implement early behavioral and lifestyle changes. However, further research is needed to determine if intervention early in pregnancy can truly reduce the risk of GDM.¹¹

The A1C goal for women with GDM is lower (6.0%) after the first trimester because any rise in A1C is risky and increased red blood cell count turnover may lower A1C.

Screening for GDM should be completed at 24 to 28 weeks of gestation²⁰ because it is likely that this is when the hormonal effects of the placenta that contribute to insulin resistance set the woman up for postprandial hyperglycemia. Currently, there are no evidence-based guidelines for the use of continuous glucose monitoring prior to 24 weeks of gestation to identify GDM.²⁰ If persistent hyperglycemia is present before 24 weeks of gestation, it is considered evidence of a pre-existing metabolic abnormality and is diagnosed as “pregestational diabetes.” Treatment should follow guidelines established for women who had diabetes prior to pregnancy.

How to screen? There is ongoing discussion about what is the optimal screening method for GDM: a 1-step strategy with a fasting 75-g OGTT only, or a 2-step strategy with a 50-g non-fasting glucose load test followed by a fasting 100-g OGTT in women who do not meet the plasma glucose cutoff (TABLE 1).^22-24 Hillier et al²⁵ compared the effectiveness of these strategies in diagnosing GDM and identifying pregnancy complications for the mother and infant. They found that while the 1-step strategy resulted in a 2-fold increase in the diagnosis of GDM, it did not lead to better outcomes for mothers and infants when compared with the 2-step method.²⁵ Currently, the majority of obstetricians (95%) prefer to use the 2-step method.²⁴

Continue to: Manage lifestyle, monitor glucose

Manage lifestyle, monitor glucose

Management of GDM in most women starts with diabetes self-management education and support for therapeutic lifestyle changes, such as nutritional interventions that reduce hyperglycemia and contribute to healthy weight gain during pregnancy.²⁰ This may include medical nutrition therapy that focuses on adequate nutrition for the mother and fetus. Currently, the recommended dietary intake for women who are pregnant (regardless of diabetes) includes a minimum of 175 g of carbohydrates, 71 g of daily protein, and at least 28 g of fiber. Further refinement of dietary intake, including carbohydrate restriction, should be done with guidance from a registered dietitian.²⁰ If the obstetrics team does not include a registered dietitian, a referral to one may be necessary. Regular physical activity should be continued throughout pregnancy as tolerated. Social support, stress reduction, and good sleep hygiene should be encouraged as much as possible.

For successful outcomes, therapeutic lifestyle changes should be coupled with glucose monitoring. The Fifth International Workshop-Conference on Gestational Diabetes Mellitus recommends that women with GDM monitor fasting blood glucose and typically 1-hour postprandial glucose. The glucose goals in GDM are as follows²⁶:

• Fasting glucose < 95 mg/dL (5.3 mmol/L), and either
• 1-hour postprandial glucose < 140 mg/dL (7.8 mmol/L), or
• 2-hour postprandial glucose < 120 mg/dL (6.7 mmol/L).

Importantly, in the second and third trimester, the A1C goal for women with GDM is 6.0%. This is lower than the more traditional A1C goal for 2 reasons: (1) increases in A1C, even within the normal range, increase adverse outcomes; and (2) pregnant women will have an increased red blood cell count turnover, which can lower the A1C.²⁷ In a historical cohort study (n = 27,213), Abell et al²⁸ found that women who have an A1C < 6.0% in the second and third trimester have the lowest risk of giving birth to large-for-gestational-age infants and for having preeclampsia.

Add insulin if glucose targets are not met

Most women who engage in therapeutic lifestyle change (70%-85%) can achieve an A1C < 6% and will not need to take medication to manage GDM.²⁹ If pharmacotherapy is needed to manage glucose, insulin is the preferred treatment for all women with GDM.²⁰ Treatment should be individualized based on the glucose trends the woman is experiencing. Common treatments include bedtime NPH if fasting hyperglycemia is most prominent and analogue insulin at mealtimes for women with prominent postprandial hyperglycemia.

Most women who engage in therapeutic lifestyle change (70%-85%) can achieve an A1C < 6% and will not need to take medication to manage GDM.

Noninsulin agents such as metformin and sulfonylureas are not currently recommended by the American College of Obstetricians and Gynecologists or the American Diabetes Association for use in GDM.^20,24 Despite being used for years in women with pregestational diabetes, metabolic syndrome, and polycystic ovary syndrome, there is evidence that metformin crosses the placenta and fetal safety has not yet been established in RCTs. The Metformin in Gestational Diabetes: The Offspring Follow-Up (MiG TOFU) study was a longitudinal follow-up study that evaluated body composition and metabolic outcomes in children (ages 7-9 years) of women with GDM who had received metformin or insulin while pregnant.³⁰ At age 9 years, children who were exposed to metformin weighed more and had a higher waist-to-height ratio and waist circumference than those exposed to insulin.³⁰

Continue to: Sulfonylureas are no longer recommended...

Sulfonylureas are no longer recommended because of the risk of maternal and fetal hypoglycemia and concerns about this medication crossing the placenta.^24,31,32 Specifically, in a 2015 meta-analysis and systematic review of 15 articles (n = 2509), glyburide had a higher risk of neonatal hypoglycemia and macrosomia than insulin or metformin.³³ For women who cannot manage their glucose with therapeutic lifestyle changes and cannot take insulin, oral therapies may be considered if the risk-benefit ratio is balanced for that person.³⁴

Watch for effects of poor glycemic control on mother, infant

Preeclampsia is defined as new-onset hypertension and proteinuria after 20 weeks of gestation. The correlation between GDM and preeclampsia has partly been explained by their shared overlapping risk factors, including maternal obesity, excessive gestational weight gain, and persistent hyperglycemia.³⁵ On a biochemical level, these risk factors contribute to oxidative stress and systemic vascular dysfunction, which have been hypothesized as the underlying pathophysiology for the development of preeclampsia.³⁵

Neonatal macrosomia, defined as a birth weight ≥ 4000 g, is a common complication that develops in 15% to 45% of infants of mothers with GDM.³⁶ Placental transfer of glucose in mothers with hyperglycemia stimulates the secretion of neonatal insulin and the ultimate storage of the excess glucose as body fat. After delivery, the abrupt discontinuation of placental transfer of glucose to an infant who is actively secreting insulin leads to neonatal hypoglycemia, which if not detected or managed, can lead to long-term neurologic deficits, including recurrent seizures and developmental delays.³⁷ Therefore, it is essential to screen for neonatal hypoglycemia immediately after birth and serially up to 12 hours.³⁸

Postpartum T2D. Poor glycemic control increases the risk of increasing insulin resistance developing into T2D postpartum for mothers.³⁹ It also increases the risk of obesity and insulin resistance later in life for the infant.⁴⁰ A retrospective cohort study (n = 461) found a positive correlation between exposure to maternal GDM and elevated BMI in children ages 6 to 13 years.⁴¹ Kamana et al³⁶ further discussed this correlation and suggested that exposure to maternal hyperglycemia in utero contributes to fetal programming of later adipose deposition. Children may develop without a notable increase in BMI until after puberty.⁴²

Partner with specialists to improve outcomes

Although most women with GDM are managed by specialists (obstetricians, endocrinologists, and maternal-fetal medicine specialists),⁴³ these patients are still seeking care from their family physicians for other complaints. These visits provide key touchpoints during pregnancy and are opportunities for PCPs to identify a pregnancy-related complication or provide additional education or referral to the obstetrician.

Continue to: Also, if you work in an area...

Noninsulin agents, such as metformin and sulfonylureas, are not currently recommended by ACOG or the ADA for use in GDM.

Also, if you work in an area where specialists are less accessible, you may be the clinician providing the majority of care to a patient with GDM. If this is the case, you’ll want to watch for the following risk factors, which should prompt a referral to specialty care:

• a previous pregnancy with GDM²⁰
• a previous birth of an infant weighing > 4000 g⁴⁴
• baseline history of hypertension⁴⁵
• evidence of insulin resistance or polycystic ovary syndrome^46,47
• a history of cardiovascular disease²⁰
• a need to treat GDM with pharmacotherapy.⁴⁸

Ensuring a smooth transition after the birth

Optimal communication and hand-offs throughout pregnancy and after delivery will benefit everyone. When the pregnant patient’s care has been managed by an obstetrician, it is important to address the following issues during the hand-off:

• baseline medical problems
• medical screenings and treatments in pregnancy (retinopathy and nephropathy screening)
• aspirin initiation, if indicated
• management of thyroid abnormalities
• management of mental health conditions
• postpartum glucose management and T2D screening postpartum
• management of complications identified during pregnancy (retinopathy and nephropathy).

Timing and other elements of postpartum care. The first postpartum screen should occur at 4 to 12 weeks postpartum. OGTT is recommended instead of A1C at this time because A1C may still be lowered by the increased red blood cell turnover related to pregnancy and blood loss at delivery. Because women with GDM have a 50% to 75% lifetime risk of T2D,²⁰ patients with normal test results should be re-tested every 1 to 3 years using any of the standard screening methods (A1C, fasting glucose, or OGTT).²⁰

Postpartum visits present another opportunity to screen for diabetes and other postpartum complications, including depression and thyroid abnormalities.

After delivery it may be difficult for women to follow-up with their own personal health care because they are focused on the care of their baby. The increased use of telehealth may make postpartum follow-up visits easier to attend.

Visits present opportunities. Postpartum visits present another opportunity for PCPs to screen for diabetes and other postpartum complications, including depression and thyroid abnormalities. Visits are also an opportunity to discuss timely contraception so as to prevent an early, unplanned pregnancy. Other important aspects of postpartum care are outlined in TABLE 2.^20,49

CORRESPONDENCE
Connie L. Ha, BS, OMS IV, Department of Primary Care, 1310 Club Drive, Touro University California, Vallejo, CA 94592; [email protected]

Gestational diabetes mellitus (GDM), defined as new-onset hyperglycemia detected in a pregnant woman after 24 weeks of gestation, affects 4% to 10% of pregnancies in the United States annually¹ and is a major challenge for health care professionals.² During pregnancy, the body’s physiologic responses are altered to support the growing fetus. One of these changes is an increase in insulin resistance, which suggests that pregnancy alone increases the patient’s risk for type 2 diabetes (T2D). However, several other factors also increase this risk, including maternal age, social barriers to care, obesity, poor weight control, and family history.

Copyright Dave Cutler

If not controlled, GDM results in poor health outcomes for the mother, such as preeclampsia, preterm labor, and maternal T2D.^3-5 For the infant, intrauterine exposure to persistent hyperglycemia is correlated with neonatal macrosomia, hypoglycemia, perinatal complications (eg, preterm delivery, fetal demise), and obesity and insulin resistance later in life.⁴

Primary care physicians (PCPs) are the patient’s main point of contact prior to pregnancy. This relationship makes PCPs a resource for the patient and specialists during and after pregnancy. In this article, we discuss risk factors and how to screen for GDM, provide an update on practice recommendations for treatment and management of GDM in primary care, and describe the effects of uncontrolled GDM.

Know the key risk factors

Prevention begins with identifying the major risk factors that contribute to the development of GDM. These include maternal age, social barriers to care, family history of prediabetes, and obesity and poor weight control.

Older age. A meta-analysis of 24 studies noted strong positive correlation between GDM risk and maternal age.⁶ One of the population-based cohort studies in the meta-analysis examined relationships between maternal age and pregnancy outcomes in women living in British Columbia, Canada (n = 203,414). Data suggested that the relative risk of GDM increased linearly with maternal age to 3.2, 4.2, and 4.4 among women ages ≥ 35, ≥ 40, and ≥ 45 years, respectively.⁷

Social barriers to care. Although the prevalence of GDM has increased over the past few decades,¹ from 2011 to 2019 the increase in GDM in individuals at first live birth was significantly higher in non-Hispanic Asian and Hispanic/Latina women than in non-Hispanic White women.⁸ Data from the Centers for Disease Control and Prevention further suggest that diabetes was more prevalent among individuals with a lower socioeconomic status as indicated by their level of education.⁹ Ogunwole et al¹⁰ suggest that racism is the root cause of these disparities and leads to long-term barriers to care (eg, socioeconomic deprivation, lack of health insurance, limited access to care, and poor health literacy), which ultimately contribute to the development of GDM and progression of diabetes. It is important for PCPs and all health professionals to be aware of these barriers so that they may practice mindfulness and deliver culturally sensitive care to patients from marginalized communities.

Family history of prediabetes. In a population-based cohort study (n = 7020), women with prediabetes (A1C, 5.7%-6.4%) were 2.8 times more likely to develop GDM compared with women with normal A1C (< 5.7%).¹¹ Similar results were seen in a retrospective cohort study (n = 2812), in which women with prediabetes were more likely than women with a normal first trimester A1C to have GDM (29.1% vs 13.7%, respectively; adjusted relative risk = 1.48; 95% CI, 1.15-1.89).¹² In both studies, prediabetes was not associated with a higher risk for adverse maternal or neonatal outcomes.^11,12

Continue to: While there are no current...

Women diagnosed with prediabetes in 1 study were found to have significantly less weight gain during pregnancy compared with patients with normal A1C, suggesting a benefit in early identification and intervention.

While there are no current guidelines for treating prediabetes in pregnancy, women diagnosed with prediabetes in 1 study were found to have significantly less weight gain during pregnancy compared with patients with normal A1C,¹² suggesting there may be a benefit in early identification and intervention, although further research is needed.¹¹ In a separate case-control study (n = 345 women with GDM; n = 800 control), high rates of gestational weight gain (> 0.41 kg/wk) were associated with an increased risk of GDM (odds ratio [OR] = 1.74; 95% CI, 1.16-2.60) compared with women with the lowest rate of gestational weight gain (0.27-0.4 kg/wk [OR = 1.43; 95% CI, 0.96-2.14]).¹³ Thus, it is helpful to have proactive conversations about family planning and adequate weight and glycemic control with high-risk patients to prepare for a healthy pregnancy.

Obesity and weight management. Patients who are overweight (body mass index [BMI], 25-29.9) or obese (BMI > 30) have a substantially increased risk of GDM (adjusted OR = 1.44; 95% CI, 1.04-1.81), as seen in a retrospective cohort study of 1951 pregnant Malaysian women.¹⁴ Several factors have been found to contribute to successful weight control, including calorie prescription, a structured meal plan, high physical activity goals (60-90 min/d), daily weighing and monitoring of food intake, behavior therapy, and continued patient–­provider contact.¹⁵

Most obstetricians use a 2-step method to screen for GDM with an initial 75-g oral glucose tolerance test, followed by a 50-g glucose load test if needed.

The safety, efficacy, and sustainability of weight loss with various dietary plans have been studied in individuals who are overweight and obese.¹⁶ Ultimately, energy expenditure must be greater than energy intake to promote weight loss. Conventional diets with continuous energy restriction (ie, low-fat, low-carbohydrate, and high-protein diets) have proven to be effective for short-term weight loss but data on long-term weight maintenance are limited.¹⁶ The Mediterranean diet, which is comprised mostly of vegetables, fruits, legumes, fish, and grains—with a lower intake of meat and dairy—may reduce gestational weight gain and risk of GDM as suggested by a randomized controlled trial (RCT; n = 1252).¹⁷ Although the choice of diet is up to the patient, it is important to be aware of different diets or refer the patient to a registered dietician who can help the patient if needed.

Reduce risk with adequate weight and glycemic control

Prevention of GDM during pregnancy should focus on weight maintenance and optimal glycemic control. Two systematic reviews, one with 8 RCTs (n = 1792) and another with 5 studies (n = 539), assessed the efficacy and safety of energy-restricted dietary intervention on GDM prevention.¹⁸ The first review found a significant reduction in gestational weight gain and improved glycemic control without increased risk of adverse maternal and fetal outcomes.¹⁸ The second review showed no clear difference between energy-restricted and non–energy-restricted diets on outcomes such as preeclampsia, gestational weight gain, large for gestational age, and macrosomia.¹⁸ These data suggest that while energy-restricted dietary interventions made no difference on maternal and fetal complications, they may still be safely used in pregnancy to reduce gestational weight gain and improve glycemic control.¹⁸

Once a woman is pregnant, it becomes difficult to lose weight because additional calories are needed to support a growing fetus. It is recommended that patients with healthy pregestational BMI consume an extra 200 to 300 calories/d after the first trimester. However, extra caloric intake in a woman with obesity who is pregnant leads to metabolic impairment and increased risk of diabetes for both the mother and fetus.¹⁹ Therefore, it is recommended that patients with obese pregestational BMI not consume additional calories because excess maternal fat is sufficient to support the energy needs of the growing fetus.¹⁹

Continue to: Ultimately, earlier intervention...

Ultimately, earlier intervention—prior to conception—helps patients prepare for a healthier pregnancy, resulting in better long-term outcomes. It is helpful to be familiar with the advantages and disadvantages of common approaches to weight management and to be able to refer patients to nutritionists for optimal planning. When establishing a dietary plan, consider patient-specific factors, such as cultural diets, financial and time constraints, and the patient’s readiness to make and maintain these changes. Consistent ­follow-up and behavioral therapy are necessary to maintain successful weight control.

There are many screening tools, but 1 is preferred in pregnancy

There are several ways to diagnose diabetes in patients who are not pregnant, including A1C, a fasting glucose test, an oral glucose tolerance test (OGTT), or random glucose testing (plus symptoms). However, the preferred method for diagnosing GDM is OGTT because it has a higher sensitivity.²⁰ A1C, while a good measure of hyperglycemic stability, does not register hyperglycemia early enough to diagnose GDM and fasting glucose testing is less sensitive because for most women with GDM, that abnormal postprandial glucose level is the first glycemic abnormality.²¹

When to screen. Blood glucose levels should be checked in all pregnant women as part of their metabolic panel at the first prenatal visit. A reflex A1C for high glucose levels can be ordered based on the physician’s preference. This may help you to identify patients with prediabetes who are at risk for GDM and implement early behavioral and lifestyle changes. However, further research is needed to determine if intervention early in pregnancy can truly reduce the risk of GDM.¹¹

The A1C goal for women with GDM is lower (6.0%) after the first trimester because any rise in A1C is risky and increased red blood cell count turnover may lower A1C.

Screening for GDM should be completed at 24 to 28 weeks of gestation²⁰ because it is likely that this is when the hormonal effects of the placenta that contribute to insulin resistance set the woman up for postprandial hyperglycemia. Currently, there are no evidence-based guidelines for the use of continuous glucose monitoring prior to 24 weeks of gestation to identify GDM.²⁰ If persistent hyperglycemia is present before 24 weeks of gestation, it is considered evidence of a pre-existing metabolic abnormality and is diagnosed as “pregestational diabetes.” Treatment should follow guidelines established for women who had diabetes prior to pregnancy.

How to screen? There is ongoing discussion about what is the optimal screening method for GDM: a 1-step strategy with a fasting 75-g OGTT only, or a 2-step strategy with a 50-g non-fasting glucose load test followed by a fasting 100-g OGTT in women who do not meet the plasma glucose cutoff (TABLE 1).^22-24 Hillier et al²⁵ compared the effectiveness of these strategies in diagnosing GDM and identifying pregnancy complications for the mother and infant. They found that while the 1-step strategy resulted in a 2-fold increase in the diagnosis of GDM, it did not lead to better outcomes for mothers and infants when compared with the 2-step method.²⁵ Currently, the majority of obstetricians (95%) prefer to use the 2-step method.²⁴

Continue to: Manage lifestyle, monitor glucose

Manage lifestyle, monitor glucose

Management of GDM in most women starts with diabetes self-management education and support for therapeutic lifestyle changes, such as nutritional interventions that reduce hyperglycemia and contribute to healthy weight gain during pregnancy.²⁰ This may include medical nutrition therapy that focuses on adequate nutrition for the mother and fetus. Currently, the recommended dietary intake for women who are pregnant (regardless of diabetes) includes a minimum of 175 g of carbohydrates, 71 g of daily protein, and at least 28 g of fiber. Further refinement of dietary intake, including carbohydrate restriction, should be done with guidance from a registered dietitian.²⁰ If the obstetrics team does not include a registered dietitian, a referral to one may be necessary. Regular physical activity should be continued throughout pregnancy as tolerated. Social support, stress reduction, and good sleep hygiene should be encouraged as much as possible.

For successful outcomes, therapeutic lifestyle changes should be coupled with glucose monitoring. The Fifth International Workshop-Conference on Gestational Diabetes Mellitus recommends that women with GDM monitor fasting blood glucose and typically 1-hour postprandial glucose. The glucose goals in GDM are as follows²⁶:

• Fasting glucose < 95 mg/dL (5.3 mmol/L), and either
• 1-hour postprandial glucose < 140 mg/dL (7.8 mmol/L), or
• 2-hour postprandial glucose < 120 mg/dL (6.7 mmol/L).

Importantly, in the second and third trimester, the A1C goal for women with GDM is 6.0%. This is lower than the more traditional A1C goal for 2 reasons: (1) increases in A1C, even within the normal range, increase adverse outcomes; and (2) pregnant women will have an increased red blood cell count turnover, which can lower the A1C.²⁷ In a historical cohort study (n = 27,213), Abell et al²⁸ found that women who have an A1C < 6.0% in the second and third trimester have the lowest risk of giving birth to large-for-gestational-age infants and for having preeclampsia.

Add insulin if glucose targets are not met

Most women who engage in therapeutic lifestyle change (70%-85%) can achieve an A1C < 6% and will not need to take medication to manage GDM.²⁹ If pharmacotherapy is needed to manage glucose, insulin is the preferred treatment for all women with GDM.²⁰ Treatment should be individualized based on the glucose trends the woman is experiencing. Common treatments include bedtime NPH if fasting hyperglycemia is most prominent and analogue insulin at mealtimes for women with prominent postprandial hyperglycemia.

Most women who engage in therapeutic lifestyle change (70%-85%) can achieve an A1C < 6% and will not need to take medication to manage GDM.

Noninsulin agents such as metformin and sulfonylureas are not currently recommended by the American College of Obstetricians and Gynecologists or the American Diabetes Association for use in GDM.^20,24 Despite being used for years in women with pregestational diabetes, metabolic syndrome, and polycystic ovary syndrome, there is evidence that metformin crosses the placenta and fetal safety has not yet been established in RCTs. The Metformin in Gestational Diabetes: The Offspring Follow-Up (MiG TOFU) study was a longitudinal follow-up study that evaluated body composition and metabolic outcomes in children (ages 7-9 years) of women with GDM who had received metformin or insulin while pregnant.³⁰ At age 9 years, children who were exposed to metformin weighed more and had a higher waist-to-height ratio and waist circumference than those exposed to insulin.³⁰

Continue to: Sulfonylureas are no longer recommended...

Sulfonylureas are no longer recommended because of the risk of maternal and fetal hypoglycemia and concerns about this medication crossing the placenta.^24,31,32 Specifically, in a 2015 meta-analysis and systematic review of 15 articles (n = 2509), glyburide had a higher risk of neonatal hypoglycemia and macrosomia than insulin or metformin.³³ For women who cannot manage their glucose with therapeutic lifestyle changes and cannot take insulin, oral therapies may be considered if the risk-benefit ratio is balanced for that person.³⁴

Watch for effects of poor glycemic control on mother, infant

Preeclampsia is defined as new-onset hypertension and proteinuria after 20 weeks of gestation. The correlation between GDM and preeclampsia has partly been explained by their shared overlapping risk factors, including maternal obesity, excessive gestational weight gain, and persistent hyperglycemia.³⁵ On a biochemical level, these risk factors contribute to oxidative stress and systemic vascular dysfunction, which have been hypothesized as the underlying pathophysiology for the development of preeclampsia.³⁵

Neonatal macrosomia, defined as a birth weight ≥ 4000 g, is a common complication that develops in 15% to 45% of infants of mothers with GDM.³⁶ Placental transfer of glucose in mothers with hyperglycemia stimulates the secretion of neonatal insulin and the ultimate storage of the excess glucose as body fat. After delivery, the abrupt discontinuation of placental transfer of glucose to an infant who is actively secreting insulin leads to neonatal hypoglycemia, which if not detected or managed, can lead to long-term neurologic deficits, including recurrent seizures and developmental delays.³⁷ Therefore, it is essential to screen for neonatal hypoglycemia immediately after birth and serially up to 12 hours.³⁸

Postpartum T2D. Poor glycemic control increases the risk of increasing insulin resistance developing into T2D postpartum for mothers.³⁹ It also increases the risk of obesity and insulin resistance later in life for the infant.⁴⁰ A retrospective cohort study (n = 461) found a positive correlation between exposure to maternal GDM and elevated BMI in children ages 6 to 13 years.⁴¹ Kamana et al³⁶ further discussed this correlation and suggested that exposure to maternal hyperglycemia in utero contributes to fetal programming of later adipose deposition. Children may develop without a notable increase in BMI until after puberty.⁴²

Partner with specialists to improve outcomes

Although most women with GDM are managed by specialists (obstetricians, endocrinologists, and maternal-fetal medicine specialists),⁴³ these patients are still seeking care from their family physicians for other complaints. These visits provide key touchpoints during pregnancy and are opportunities for PCPs to identify a pregnancy-related complication or provide additional education or referral to the obstetrician.

Continue to: Also, if you work in an area...

Noninsulin agents, such as metformin and sulfonylureas, are not currently recommended by ACOG or the ADA for use in GDM.

Also, if you work in an area where specialists are less accessible, you may be the clinician providing the majority of care to a patient with GDM. If this is the case, you’ll want to watch for the following risk factors, which should prompt a referral to specialty care:

• a previous pregnancy with GDM²⁰
• a previous birth of an infant weighing > 4000 g⁴⁴
• baseline history of hypertension⁴⁵
• evidence of insulin resistance or polycystic ovary syndrome^46,47
• a history of cardiovascular disease²⁰
• a need to treat GDM with pharmacotherapy.⁴⁸

Ensuring a smooth transition after the birth

Optimal communication and hand-offs throughout pregnancy and after delivery will benefit everyone. When the pregnant patient’s care has been managed by an obstetrician, it is important to address the following issues during the hand-off:

• baseline medical problems
• medical screenings and treatments in pregnancy (retinopathy and nephropathy screening)
• aspirin initiation, if indicated
• management of thyroid abnormalities
• management of mental health conditions
• postpartum glucose management and T2D screening postpartum
• management of complications identified during pregnancy (retinopathy and nephropathy).

Timing and other elements of postpartum care. The first postpartum screen should occur at 4 to 12 weeks postpartum. OGTT is recommended instead of A1C at this time because A1C may still be lowered by the increased red blood cell turnover related to pregnancy and blood loss at delivery. Because women with GDM have a 50% to 75% lifetime risk of T2D,²⁰ patients with normal test results should be re-tested every 1 to 3 years using any of the standard screening methods (A1C, fasting glucose, or OGTT).²⁰

Postpartum visits present another opportunity to screen for diabetes and other postpartum complications, including depression and thyroid abnormalities.

After delivery it may be difficult for women to follow-up with their own personal health care because they are focused on the care of their baby. The increased use of telehealth may make postpartum follow-up visits easier to attend.

Visits present opportunities. Postpartum visits present another opportunity for PCPs to screen for diabetes and other postpartum complications, including depression and thyroid abnormalities. Visits are also an opportunity to discuss timely contraception so as to prevent an early, unplanned pregnancy. Other important aspects of postpartum care are outlined in TABLE 2.^20,49

CORRESPONDENCE
Connie L. Ha, BS, OMS IV, Department of Primary Care, 1310 Club Drive, Touro University California, Vallejo, CA 94592; [email protected]

Gestational diabetes mellitus (GDM), defined as new-onset hyperglycemia detected in a pregnant woman after 24 weeks of gestation, affects 4% to 10% of pregnancies in the United States annually¹ and is a major challenge for health care professionals.² During pregnancy, the body’s physiologic responses are altered to support the growing fetus. One of these changes is an increase in insulin resistance, which suggests that pregnancy alone increases the patient’s risk for type 2 diabetes (T2D). However, several other factors also increase this risk, including maternal age, social barriers to care, obesity, poor weight control, and family history.

Copyright Dave Cutler

If not controlled, GDM results in poor health outcomes for the mother, such as preeclampsia, preterm labor, and maternal T2D.^3-5 For the infant, intrauterine exposure to persistent hyperglycemia is correlated with neonatal macrosomia, hypoglycemia, perinatal complications (eg, preterm delivery, fetal demise), and obesity and insulin resistance later in life.⁴

Primary care physicians (PCPs) are the patient’s main point of contact prior to pregnancy. This relationship makes PCPs a resource for the patient and specialists during and after pregnancy. In this article, we discuss risk factors and how to screen for GDM, provide an update on practice recommendations for treatment and management of GDM in primary care, and describe the effects of uncontrolled GDM.

Know the key risk factors

Prevention begins with identifying the major risk factors that contribute to the development of GDM. These include maternal age, social barriers to care, family history of prediabetes, and obesity and poor weight control.

Older age. A meta-analysis of 24 studies noted strong positive correlation between GDM risk and maternal age.⁶ One of the population-based cohort studies in the meta-analysis examined relationships between maternal age and pregnancy outcomes in women living in British Columbia, Canada (n = 203,414). Data suggested that the relative risk of GDM increased linearly with maternal age to 3.2, 4.2, and 4.4 among women ages ≥ 35, ≥ 40, and ≥ 45 years, respectively.⁷

Social barriers to care. Although the prevalence of GDM has increased over the past few decades,¹ from 2011 to 2019 the increase in GDM in individuals at first live birth was significantly higher in non-Hispanic Asian and Hispanic/Latina women than in non-Hispanic White women.⁸ Data from the Centers for Disease Control and Prevention further suggest that diabetes was more prevalent among individuals with a lower socioeconomic status as indicated by their level of education.⁹ Ogunwole et al¹⁰ suggest that racism is the root cause of these disparities and leads to long-term barriers to care (eg, socioeconomic deprivation, lack of health insurance, limited access to care, and poor health literacy), which ultimately contribute to the development of GDM and progression of diabetes. It is important for PCPs and all health professionals to be aware of these barriers so that they may practice mindfulness and deliver culturally sensitive care to patients from marginalized communities.

Family history of prediabetes. In a population-based cohort study (n = 7020), women with prediabetes (A1C, 5.7%-6.4%) were 2.8 times more likely to develop GDM compared with women with normal A1C (< 5.7%).¹¹ Similar results were seen in a retrospective cohort study (n = 2812), in which women with prediabetes were more likely than women with a normal first trimester A1C to have GDM (29.1% vs 13.7%, respectively; adjusted relative risk = 1.48; 95% CI, 1.15-1.89).¹² In both studies, prediabetes was not associated with a higher risk for adverse maternal or neonatal outcomes.^11,12

Continue to: While there are no current...

Women diagnosed with prediabetes in 1 study were found to have significantly less weight gain during pregnancy compared with patients with normal A1C, suggesting a benefit in early identification and intervention.

While there are no current guidelines for treating prediabetes in pregnancy, women diagnosed with prediabetes in 1 study were found to have significantly less weight gain during pregnancy compared with patients with normal A1C,¹² suggesting there may be a benefit in early identification and intervention, although further research is needed.¹¹ In a separate case-control study (n = 345 women with GDM; n = 800 control), high rates of gestational weight gain (> 0.41 kg/wk) were associated with an increased risk of GDM (odds ratio [OR] = 1.74; 95% CI, 1.16-2.60) compared with women with the lowest rate of gestational weight gain (0.27-0.4 kg/wk [OR = 1.43; 95% CI, 0.96-2.14]).¹³ Thus, it is helpful to have proactive conversations about family planning and adequate weight and glycemic control with high-risk patients to prepare for a healthy pregnancy.

Obesity and weight management. Patients who are overweight (body mass index [BMI], 25-29.9) or obese (BMI > 30) have a substantially increased risk of GDM (adjusted OR = 1.44; 95% CI, 1.04-1.81), as seen in a retrospective cohort study of 1951 pregnant Malaysian women.¹⁴ Several factors have been found to contribute to successful weight control, including calorie prescription, a structured meal plan, high physical activity goals (60-90 min/d), daily weighing and monitoring of food intake, behavior therapy, and continued patient–­provider contact.¹⁵

Most obstetricians use a 2-step method to screen for GDM with an initial 75-g oral glucose tolerance test, followed by a 50-g glucose load test if needed.

The safety, efficacy, and sustainability of weight loss with various dietary plans have been studied in individuals who are overweight and obese.¹⁶ Ultimately, energy expenditure must be greater than energy intake to promote weight loss. Conventional diets with continuous energy restriction (ie, low-fat, low-carbohydrate, and high-protein diets) have proven to be effective for short-term weight loss but data on long-term weight maintenance are limited.¹⁶ The Mediterranean diet, which is comprised mostly of vegetables, fruits, legumes, fish, and grains—with a lower intake of meat and dairy—may reduce gestational weight gain and risk of GDM as suggested by a randomized controlled trial (RCT; n = 1252).¹⁷ Although the choice of diet is up to the patient, it is important to be aware of different diets or refer the patient to a registered dietician who can help the patient if needed.

Reduce risk with adequate weight and glycemic control

Prevention of GDM during pregnancy should focus on weight maintenance and optimal glycemic control. Two systematic reviews, one with 8 RCTs (n = 1792) and another with 5 studies (n = 539), assessed the efficacy and safety of energy-restricted dietary intervention on GDM prevention.¹⁸ The first review found a significant reduction in gestational weight gain and improved glycemic control without increased risk of adverse maternal and fetal outcomes.¹⁸ The second review showed no clear difference between energy-restricted and non–energy-restricted diets on outcomes such as preeclampsia, gestational weight gain, large for gestational age, and macrosomia.¹⁸ These data suggest that while energy-restricted dietary interventions made no difference on maternal and fetal complications, they may still be safely used in pregnancy to reduce gestational weight gain and improve glycemic control.¹⁸

Once a woman is pregnant, it becomes difficult to lose weight because additional calories are needed to support a growing fetus. It is recommended that patients with healthy pregestational BMI consume an extra 200 to 300 calories/d after the first trimester. However, extra caloric intake in a woman with obesity who is pregnant leads to metabolic impairment and increased risk of diabetes for both the mother and fetus.¹⁹ Therefore, it is recommended that patients with obese pregestational BMI not consume additional calories because excess maternal fat is sufficient to support the energy needs of the growing fetus.¹⁹

Continue to: Ultimately, earlier intervention...

Ultimately, earlier intervention—prior to conception—helps patients prepare for a healthier pregnancy, resulting in better long-term outcomes. It is helpful to be familiar with the advantages and disadvantages of common approaches to weight management and to be able to refer patients to nutritionists for optimal planning. When establishing a dietary plan, consider patient-specific factors, such as cultural diets, financial and time constraints, and the patient’s readiness to make and maintain these changes. Consistent ­follow-up and behavioral therapy are necessary to maintain successful weight control.

There are many screening tools, but 1 is preferred in pregnancy

There are several ways to diagnose diabetes in patients who are not pregnant, including A1C, a fasting glucose test, an oral glucose tolerance test (OGTT), or random glucose testing (plus symptoms). However, the preferred method for diagnosing GDM is OGTT because it has a higher sensitivity.²⁰ A1C, while a good measure of hyperglycemic stability, does not register hyperglycemia early enough to diagnose GDM and fasting glucose testing is less sensitive because for most women with GDM, that abnormal postprandial glucose level is the first glycemic abnormality.²¹

When to screen. Blood glucose levels should be checked in all pregnant women as part of their metabolic panel at the first prenatal visit. A reflex A1C for high glucose levels can be ordered based on the physician’s preference. This may help you to identify patients with prediabetes who are at risk for GDM and implement early behavioral and lifestyle changes. However, further research is needed to determine if intervention early in pregnancy can truly reduce the risk of GDM.¹¹

The A1C goal for women with GDM is lower (6.0%) after the first trimester because any rise in A1C is risky and increased red blood cell count turnover may lower A1C.

Screening for GDM should be completed at 24 to 28 weeks of gestation²⁰ because it is likely that this is when the hormonal effects of the placenta that contribute to insulin resistance set the woman up for postprandial hyperglycemia. Currently, there are no evidence-based guidelines for the use of continuous glucose monitoring prior to 24 weeks of gestation to identify GDM.²⁰ If persistent hyperglycemia is present before 24 weeks of gestation, it is considered evidence of a pre-existing metabolic abnormality and is diagnosed as “pregestational diabetes.” Treatment should follow guidelines established for women who had diabetes prior to pregnancy.

How to screen? There is ongoing discussion about what is the optimal screening method for GDM: a 1-step strategy with a fasting 75-g OGTT only, or a 2-step strategy with a 50-g non-fasting glucose load test followed by a fasting 100-g OGTT in women who do not meet the plasma glucose cutoff (TABLE 1).^22-24 Hillier et al²⁵ compared the effectiveness of these strategies in diagnosing GDM and identifying pregnancy complications for the mother and infant. They found that while the 1-step strategy resulted in a 2-fold increase in the diagnosis of GDM, it did not lead to better outcomes for mothers and infants when compared with the 2-step method.²⁵ Currently, the majority of obstetricians (95%) prefer to use the 2-step method.²⁴

Continue to: Manage lifestyle, monitor glucose

Manage lifestyle, monitor glucose

Management of GDM in most women starts with diabetes self-management education and support for therapeutic lifestyle changes, such as nutritional interventions that reduce hyperglycemia and contribute to healthy weight gain during pregnancy.²⁰ This may include medical nutrition therapy that focuses on adequate nutrition for the mother and fetus. Currently, the recommended dietary intake for women who are pregnant (regardless of diabetes) includes a minimum of 175 g of carbohydrates, 71 g of daily protein, and at least 28 g of fiber. Further refinement of dietary intake, including carbohydrate restriction, should be done with guidance from a registered dietitian.²⁰ If the obstetrics team does not include a registered dietitian, a referral to one may be necessary. Regular physical activity should be continued throughout pregnancy as tolerated. Social support, stress reduction, and good sleep hygiene should be encouraged as much as possible.

For successful outcomes, therapeutic lifestyle changes should be coupled with glucose monitoring. The Fifth International Workshop-Conference on Gestational Diabetes Mellitus recommends that women with GDM monitor fasting blood glucose and typically 1-hour postprandial glucose. The glucose goals in GDM are as follows²⁶:

• Fasting glucose < 95 mg/dL (5.3 mmol/L), and either
• 1-hour postprandial glucose < 140 mg/dL (7.8 mmol/L), or
• 2-hour postprandial glucose < 120 mg/dL (6.7 mmol/L).

Importantly, in the second and third trimester, the A1C goal for women with GDM is 6.0%. This is lower than the more traditional A1C goal for 2 reasons: (1) increases in A1C, even within the normal range, increase adverse outcomes; and (2) pregnant women will have an increased red blood cell count turnover, which can lower the A1C.²⁷ In a historical cohort study (n = 27,213), Abell et al²⁸ found that women who have an A1C < 6.0% in the second and third trimester have the lowest risk of giving birth to large-for-gestational-age infants and for having preeclampsia.

Add insulin if glucose targets are not met

Most women who engage in therapeutic lifestyle change (70%-85%) can achieve an A1C < 6% and will not need to take medication to manage GDM.²⁹ If pharmacotherapy is needed to manage glucose, insulin is the preferred treatment for all women with GDM.²⁰ Treatment should be individualized based on the glucose trends the woman is experiencing. Common treatments include bedtime NPH if fasting hyperglycemia is most prominent and analogue insulin at mealtimes for women with prominent postprandial hyperglycemia.

Most women who engage in therapeutic lifestyle change (70%-85%) can achieve an A1C < 6% and will not need to take medication to manage GDM.

Noninsulin agents such as metformin and sulfonylureas are not currently recommended by the American College of Obstetricians and Gynecologists or the American Diabetes Association for use in GDM.^20,24 Despite being used for years in women with pregestational diabetes, metabolic syndrome, and polycystic ovary syndrome, there is evidence that metformin crosses the placenta and fetal safety has not yet been established in RCTs. The Metformin in Gestational Diabetes: The Offspring Follow-Up (MiG TOFU) study was a longitudinal follow-up study that evaluated body composition and metabolic outcomes in children (ages 7-9 years) of women with GDM who had received metformin or insulin while pregnant.³⁰ At age 9 years, children who were exposed to metformin weighed more and had a higher waist-to-height ratio and waist circumference than those exposed to insulin.³⁰

Continue to: Sulfonylureas are no longer recommended...

Sulfonylureas are no longer recommended because of the risk of maternal and fetal hypoglycemia and concerns about this medication crossing the placenta.^24,31,32 Specifically, in a 2015 meta-analysis and systematic review of 15 articles (n = 2509), glyburide had a higher risk of neonatal hypoglycemia and macrosomia than insulin or metformin.³³ For women who cannot manage their glucose with therapeutic lifestyle changes and cannot take insulin, oral therapies may be considered if the risk-benefit ratio is balanced for that person.³⁴

Watch for effects of poor glycemic control on mother, infant

Preeclampsia is defined as new-onset hypertension and proteinuria after 20 weeks of gestation. The correlation between GDM and preeclampsia has partly been explained by their shared overlapping risk factors, including maternal obesity, excessive gestational weight gain, and persistent hyperglycemia.³⁵ On a biochemical level, these risk factors contribute to oxidative stress and systemic vascular dysfunction, which have been hypothesized as the underlying pathophysiology for the development of preeclampsia.³⁵

Neonatal macrosomia, defined as a birth weight ≥ 4000 g, is a common complication that develops in 15% to 45% of infants of mothers with GDM.³⁶ Placental transfer of glucose in mothers with hyperglycemia stimulates the secretion of neonatal insulin and the ultimate storage of the excess glucose as body fat. After delivery, the abrupt discontinuation of placental transfer of glucose to an infant who is actively secreting insulin leads to neonatal hypoglycemia, which if not detected or managed, can lead to long-term neurologic deficits, including recurrent seizures and developmental delays.³⁷ Therefore, it is essential to screen for neonatal hypoglycemia immediately after birth and serially up to 12 hours.³⁸

Postpartum T2D. Poor glycemic control increases the risk of increasing insulin resistance developing into T2D postpartum for mothers.³⁹ It also increases the risk of obesity and insulin resistance later in life for the infant.⁴⁰ A retrospective cohort study (n = 461) found a positive correlation between exposure to maternal GDM and elevated BMI in children ages 6 to 13 years.⁴¹ Kamana et al³⁶ further discussed this correlation and suggested that exposure to maternal hyperglycemia in utero contributes to fetal programming of later adipose deposition. Children may develop without a notable increase in BMI until after puberty.⁴²

Partner with specialists to improve outcomes

Although most women with GDM are managed by specialists (obstetricians, endocrinologists, and maternal-fetal medicine specialists),⁴³ these patients are still seeking care from their family physicians for other complaints. These visits provide key touchpoints during pregnancy and are opportunities for PCPs to identify a pregnancy-related complication or provide additional education or referral to the obstetrician.

Continue to: Also, if you work in an area...

Noninsulin agents, such as metformin and sulfonylureas, are not currently recommended by ACOG or the ADA for use in GDM.

Also, if you work in an area where specialists are less accessible, you may be the clinician providing the majority of care to a patient with GDM. If this is the case, you’ll want to watch for the following risk factors, which should prompt a referral to specialty care:

• a previous pregnancy with GDM²⁰
• a previous birth of an infant weighing > 4000 g⁴⁴
• baseline history of hypertension⁴⁵
• evidence of insulin resistance or polycystic ovary syndrome^46,47
• a history of cardiovascular disease²⁰
• a need to treat GDM with pharmacotherapy.⁴⁸

Ensuring a smooth transition after the birth

Optimal communication and hand-offs throughout pregnancy and after delivery will benefit everyone. When the pregnant patient’s care has been managed by an obstetrician, it is important to address the following issues during the hand-off:

• baseline medical problems
• medical screenings and treatments in pregnancy (retinopathy and nephropathy screening)
• aspirin initiation, if indicated
• management of thyroid abnormalities
• management of mental health conditions
• postpartum glucose management and T2D screening postpartum
• management of complications identified during pregnancy (retinopathy and nephropathy).

Timing and other elements of postpartum care. The first postpartum screen should occur at 4 to 12 weeks postpartum. OGTT is recommended instead of A1C at this time because A1C may still be lowered by the increased red blood cell turnover related to pregnancy and blood loss at delivery. Because women with GDM have a 50% to 75% lifetime risk of T2D,²⁰ patients with normal test results should be re-tested every 1 to 3 years using any of the standard screening methods (A1C, fasting glucose, or OGTT).²⁰

Postpartum visits present another opportunity to screen for diabetes and other postpartum complications, including depression and thyroid abnormalities.

After delivery it may be difficult for women to follow-up with their own personal health care because they are focused on the care of their baby. The increased use of telehealth may make postpartum follow-up visits easier to attend.

Visits present opportunities. Postpartum visits present another opportunity for PCPs to screen for diabetes and other postpartum complications, including depression and thyroid abnormalities. Visits are also an opportunity to discuss timely contraception so as to prevent an early, unplanned pregnancy. Other important aspects of postpartum care are outlined in TABLE 2.^20,49

CORRESPONDENCE
Connie L. Ha, BS, OMS IV, Department of Primary Care, 1310 Club Drive, Touro University California, Vallejo, CA 94592; [email protected]

The Diagnosis: Bullous Pyoderma Gangrenosum

A bone marrow biopsy revealed 60% myeloblasts, leading to a diagnosis of acute myeloid leukemia (AML). A biopsy obtained from the edge of the bullous plaque demonstrated a dense dermal neutrophilic infiltrate with extravasated erythrocytes (Figure). Fite, Gram, and Grocott-Gomori methenamine-silver staining failed to reveal infectious organisms. Tissue and blood cultures were negative. Given the pathologic findings, clinical presentation including recent diagnosis of AML, and exclusion of other underlying disease processes including infection, the diagnosis of bullous pyoderma gangrenosum (PG) was made. The lesion improved with systemic steroids and treatment of the underlying AML with fludarabine and venetoclax chemotherapy.

First recognized in 1916 by French dermatologist Louis Brocq, MD, PG is a sterile neutrophilic dermatosis that predominantly affects women older than 50 years.^1,2 This disorder can develop idiopathically; secondary to trauma; or in association with systemic diseases such as inflammatory bowel disease, rheumatoid arthritis, and hematologic malignancies. The pathogenesis of PG remains unclear; however, overexpression of inflammatory cytokines may mediate its development by stimulating T cells and promoting neutrophilic chemotaxis.³

Pyoderma gangrenosum classically presents as a rapidly enlarging ulcer with cribriform scarring but manifests variably. Four variants of the disorder exist: classic ulcerative, pustular, bullous, and vegetative PG. Ulcerative PG is the most common variant. Bullous PG is associated with hematologic malignancies such as primary myelofibrosis, myelodysplastic disease, and AML. In these patients, hematologic malignancy often exists prior to the development of PG and portends a poorer prognosis. This association underscores the importance of timely diagnosis and thorough hematologic evaluation by obtaining a complete blood cell count with differential, peripheral smear, serum protein electrophoresis with immunofixation, and quantitative immunoglobulins (IgA, IgG, IgM). If any of the results are positive, prompt referral to a hematologist and bone marrow biopsy are paramount.³

The diagnosis of PG remains elusive, as no validated clinical or pathological criteria exist. Histopathologic evaluation may be nonspecific and variable depending on the subtype. Biopsy results for classic ulcerative PG may reveal a neutrophilic infiltrate with leukocytoclasia. Bullous PG may include subepidermal hemorrhagic bullae. Notably, bullous PG appears histologically similar to the superficial bullous variant of Sweet syndrome.

Sweet syndrome (also known as acute febrile neutrophilic dermatosis) is a type of neutrophilic dermatosis characterized by fever, neutrophilia, and the sudden onset of tender erythematous lesions. Variations include idiopathic, subcutaneous, and bullous Sweet syndrome, which present as plaques, nodules, or bullae, respectively.⁴ Similar to PG, Sweet syndrome can manifest in patients with hematologic malignancies. Both PG and Sweet syndrome are thought to exist along a continuum and can be considered intersecting diagnoses in the setting of leukemia or other hematologic malignancies.⁵ There have been reports of the coexistence of distinct PG and Sweet syndrome lesions on a single patient, further supporting the belief that these entities share a common pathologic mechanism.⁶ Sweet syndrome also commonly can be associated with upper respiratory infections; pregnancy; and medications, with culprits including granulocyte colony-stimulating factor, azathioprine, vemurafenib, and isotretinoin.⁷

Other differential diagnoses include brown recluse spider bite, bullous fixed drug eruption (FDE), and necrotizing fasciitis (NF). Venom from the brown recluse spider (Loxosceles reclusa) can trigger toxin-mediated hemolysis, complement-mediated erythrocyte destruction, and basement membrane zone degradation due to the synergistic effects of the toxin’s sphingomyelinase D and protease content.⁸ The inciting bite is painless. After 8 hours, the site becomes painful and pruritic and presents with peripheral erythema and central pallor. After 24 hours, the lesion blisters. The blister ruptures within 3 to 4 days, resulting in eschar formation with the subsequent development of an indurated blue ulcer with a stellate center. Ulcers can take months to heal.⁹ Based on the clinical findings in our patient, this diagnosis was less likely.

Fixed drug eruption is a localized cutaneous reaction that manifests in fixed locations minutes to days after exposure to medications such as trimethoprimsulfamethoxazole, nonsteroidal anti-inflammatory drugs, salicylates, and oral contraceptives. Commonly affected areas include the hands, legs, genitals, and trunk. Lesions initially present as well-demarcated, erythematous to violaceous, round plaques. A rarer variant manifesting as bullae also has been described. Careful consideration of the patient’s history and physical examination findings is sufficient for establishing this diagnosis; however, a punch biopsy can provide clarity. Histopathology reveals a lichenoid tissue reaction with dyskeratosis, broad epidermal necrosis, and damage to the stratum basalis. A lymphocytic perivascular infiltrate also may appear in the dermis.¹⁰ Both the clinical findings and histopathology of our case were not characteristic of FDE.

Necrotizing fasciitis is a fulminant, life-threatening, soft-tissue infection precipitated by polymicrobial flora. Early recognition of NF is difficult, as in its early stages it can mimic cellulitis. As the infection takes its course, necrosis can extend from the skin and into the subcutaneous tissue. Patients also develop fever, leukocytosis, and signs of sepsis. Histopathology demonstrates neutrophilic infiltration with bacterial invasion as well as necrosis of the superficial fascia and subepidermal edema.¹¹ Pyoderma gangrenosum previously has been reported to mimic NF; however, lack of responsiveness to antibiotic therapy would favor a diagnosis of PG over NF.¹²

Treatment of PG is driven by the extent of cutaneous involvement. In mild cases, wound care and topical therapy with corticosteroids and tacrolimus may suffice. Severe cases necessitate systemic therapy with oral corticosteroids or cyclosporine; biologic therapy also may play a role in treatment.⁴ In patients with hematologic malignancy, chemotherapy alone may partially or completely resolve the lesion; however, systemic corticosteroids commonly are included in management.³

The Diagnosis: Bullous Pyoderma Gangrenosum

A bone marrow biopsy revealed 60% myeloblasts, leading to a diagnosis of acute myeloid leukemia (AML). A biopsy obtained from the edge of the bullous plaque demonstrated a dense dermal neutrophilic infiltrate with extravasated erythrocytes (Figure). Fite, Gram, and Grocott-Gomori methenamine-silver staining failed to reveal infectious organisms. Tissue and blood cultures were negative. Given the pathologic findings, clinical presentation including recent diagnosis of AML, and exclusion of other underlying disease processes including infection, the diagnosis of bullous pyoderma gangrenosum (PG) was made. The lesion improved with systemic steroids and treatment of the underlying AML with fludarabine and venetoclax chemotherapy.

First recognized in 1916 by French dermatologist Louis Brocq, MD, PG is a sterile neutrophilic dermatosis that predominantly affects women older than 50 years.^1,2 This disorder can develop idiopathically; secondary to trauma; or in association with systemic diseases such as inflammatory bowel disease, rheumatoid arthritis, and hematologic malignancies. The pathogenesis of PG remains unclear; however, overexpression of inflammatory cytokines may mediate its development by stimulating T cells and promoting neutrophilic chemotaxis.³

Pyoderma gangrenosum classically presents as a rapidly enlarging ulcer with cribriform scarring but manifests variably. Four variants of the disorder exist: classic ulcerative, pustular, bullous, and vegetative PG. Ulcerative PG is the most common variant. Bullous PG is associated with hematologic malignancies such as primary myelofibrosis, myelodysplastic disease, and AML. In these patients, hematologic malignancy often exists prior to the development of PG and portends a poorer prognosis. This association underscores the importance of timely diagnosis and thorough hematologic evaluation by obtaining a complete blood cell count with differential, peripheral smear, serum protein electrophoresis with immunofixation, and quantitative immunoglobulins (IgA, IgG, IgM). If any of the results are positive, prompt referral to a hematologist and bone marrow biopsy are paramount.³

The diagnosis of PG remains elusive, as no validated clinical or pathological criteria exist. Histopathologic evaluation may be nonspecific and variable depending on the subtype. Biopsy results for classic ulcerative PG may reveal a neutrophilic infiltrate with leukocytoclasia. Bullous PG may include subepidermal hemorrhagic bullae. Notably, bullous PG appears histologically similar to the superficial bullous variant of Sweet syndrome.

Sweet syndrome (also known as acute febrile neutrophilic dermatosis) is a type of neutrophilic dermatosis characterized by fever, neutrophilia, and the sudden onset of tender erythematous lesions. Variations include idiopathic, subcutaneous, and bullous Sweet syndrome, which present as plaques, nodules, or bullae, respectively.⁴ Similar to PG, Sweet syndrome can manifest in patients with hematologic malignancies. Both PG and Sweet syndrome are thought to exist along a continuum and can be considered intersecting diagnoses in the setting of leukemia or other hematologic malignancies.⁵ There have been reports of the coexistence of distinct PG and Sweet syndrome lesions on a single patient, further supporting the belief that these entities share a common pathologic mechanism.⁶ Sweet syndrome also commonly can be associated with upper respiratory infections; pregnancy; and medications, with culprits including granulocyte colony-stimulating factor, azathioprine, vemurafenib, and isotretinoin.⁷

Other differential diagnoses include brown recluse spider bite, bullous fixed drug eruption (FDE), and necrotizing fasciitis (NF). Venom from the brown recluse spider (Loxosceles reclusa) can trigger toxin-mediated hemolysis, complement-mediated erythrocyte destruction, and basement membrane zone degradation due to the synergistic effects of the toxin’s sphingomyelinase D and protease content.⁸ The inciting bite is painless. After 8 hours, the site becomes painful and pruritic and presents with peripheral erythema and central pallor. After 24 hours, the lesion blisters. The blister ruptures within 3 to 4 days, resulting in eschar formation with the subsequent development of an indurated blue ulcer with a stellate center. Ulcers can take months to heal.⁹ Based on the clinical findings in our patient, this diagnosis was less likely.

Fixed drug eruption is a localized cutaneous reaction that manifests in fixed locations minutes to days after exposure to medications such as trimethoprimsulfamethoxazole, nonsteroidal anti-inflammatory drugs, salicylates, and oral contraceptives. Commonly affected areas include the hands, legs, genitals, and trunk. Lesions initially present as well-demarcated, erythematous to violaceous, round plaques. A rarer variant manifesting as bullae also has been described. Careful consideration of the patient’s history and physical examination findings is sufficient for establishing this diagnosis; however, a punch biopsy can provide clarity. Histopathology reveals a lichenoid tissue reaction with dyskeratosis, broad epidermal necrosis, and damage to the stratum basalis. A lymphocytic perivascular infiltrate also may appear in the dermis.¹⁰ Both the clinical findings and histopathology of our case were not characteristic of FDE.

Necrotizing fasciitis is a fulminant, life-threatening, soft-tissue infection precipitated by polymicrobial flora. Early recognition of NF is difficult, as in its early stages it can mimic cellulitis. As the infection takes its course, necrosis can extend from the skin and into the subcutaneous tissue. Patients also develop fever, leukocytosis, and signs of sepsis. Histopathology demonstrates neutrophilic infiltration with bacterial invasion as well as necrosis of the superficial fascia and subepidermal edema.¹¹ Pyoderma gangrenosum previously has been reported to mimic NF; however, lack of responsiveness to antibiotic therapy would favor a diagnosis of PG over NF.¹²

Treatment of PG is driven by the extent of cutaneous involvement. In mild cases, wound care and topical therapy with corticosteroids and tacrolimus may suffice. Severe cases necessitate systemic therapy with oral corticosteroids or cyclosporine; biologic therapy also may play a role in treatment.⁴ In patients with hematologic malignancy, chemotherapy alone may partially or completely resolve the lesion; however, systemic corticosteroids commonly are included in management.³

The Diagnosis: Bullous Pyoderma Gangrenosum

A bone marrow biopsy revealed 60% myeloblasts, leading to a diagnosis of acute myeloid leukemia (AML). A biopsy obtained from the edge of the bullous plaque demonstrated a dense dermal neutrophilic infiltrate with extravasated erythrocytes (Figure). Fite, Gram, and Grocott-Gomori methenamine-silver staining failed to reveal infectious organisms. Tissue and blood cultures were negative. Given the pathologic findings, clinical presentation including recent diagnosis of AML, and exclusion of other underlying disease processes including infection, the diagnosis of bullous pyoderma gangrenosum (PG) was made. The lesion improved with systemic steroids and treatment of the underlying AML with fludarabine and venetoclax chemotherapy.

First recognized in 1916 by French dermatologist Louis Brocq, MD, PG is a sterile neutrophilic dermatosis that predominantly affects women older than 50 years.^1,2 This disorder can develop idiopathically; secondary to trauma; or in association with systemic diseases such as inflammatory bowel disease, rheumatoid arthritis, and hematologic malignancies. The pathogenesis of PG remains unclear; however, overexpression of inflammatory cytokines may mediate its development by stimulating T cells and promoting neutrophilic chemotaxis.³

Pyoderma gangrenosum classically presents as a rapidly enlarging ulcer with cribriform scarring but manifests variably. Four variants of the disorder exist: classic ulcerative, pustular, bullous, and vegetative PG. Ulcerative PG is the most common variant. Bullous PG is associated with hematologic malignancies such as primary myelofibrosis, myelodysplastic disease, and AML. In these patients, hematologic malignancy often exists prior to the development of PG and portends a poorer prognosis. This association underscores the importance of timely diagnosis and thorough hematologic evaluation by obtaining a complete blood cell count with differential, peripheral smear, serum protein electrophoresis with immunofixation, and quantitative immunoglobulins (IgA, IgG, IgM). If any of the results are positive, prompt referral to a hematologist and bone marrow biopsy are paramount.³

The diagnosis of PG remains elusive, as no validated clinical or pathological criteria exist. Histopathologic evaluation may be nonspecific and variable depending on the subtype. Biopsy results for classic ulcerative PG may reveal a neutrophilic infiltrate with leukocytoclasia. Bullous PG may include subepidermal hemorrhagic bullae. Notably, bullous PG appears histologically similar to the superficial bullous variant of Sweet syndrome.

Sweet syndrome (also known as acute febrile neutrophilic dermatosis) is a type of neutrophilic dermatosis characterized by fever, neutrophilia, and the sudden onset of tender erythematous lesions. Variations include idiopathic, subcutaneous, and bullous Sweet syndrome, which present as plaques, nodules, or bullae, respectively.⁴ Similar to PG, Sweet syndrome can manifest in patients with hematologic malignancies. Both PG and Sweet syndrome are thought to exist along a continuum and can be considered intersecting diagnoses in the setting of leukemia or other hematologic malignancies.⁵ There have been reports of the coexistence of distinct PG and Sweet syndrome lesions on a single patient, further supporting the belief that these entities share a common pathologic mechanism.⁶ Sweet syndrome also commonly can be associated with upper respiratory infections; pregnancy; and medications, with culprits including granulocyte colony-stimulating factor, azathioprine, vemurafenib, and isotretinoin.⁷

Other differential diagnoses include brown recluse spider bite, bullous fixed drug eruption (FDE), and necrotizing fasciitis (NF). Venom from the brown recluse spider (Loxosceles reclusa) can trigger toxin-mediated hemolysis, complement-mediated erythrocyte destruction, and basement membrane zone degradation due to the synergistic effects of the toxin’s sphingomyelinase D and protease content.⁸ The inciting bite is painless. After 8 hours, the site becomes painful and pruritic and presents with peripheral erythema and central pallor. After 24 hours, the lesion blisters. The blister ruptures within 3 to 4 days, resulting in eschar formation with the subsequent development of an indurated blue ulcer with a stellate center. Ulcers can take months to heal.⁹ Based on the clinical findings in our patient, this diagnosis was less likely.

Fixed drug eruption is a localized cutaneous reaction that manifests in fixed locations minutes to days after exposure to medications such as trimethoprimsulfamethoxazole, nonsteroidal anti-inflammatory drugs, salicylates, and oral contraceptives. Commonly affected areas include the hands, legs, genitals, and trunk. Lesions initially present as well-demarcated, erythematous to violaceous, round plaques. A rarer variant manifesting as bullae also has been described. Careful consideration of the patient’s history and physical examination findings is sufficient for establishing this diagnosis; however, a punch biopsy can provide clarity. Histopathology reveals a lichenoid tissue reaction with dyskeratosis, broad epidermal necrosis, and damage to the stratum basalis. A lymphocytic perivascular infiltrate also may appear in the dermis.¹⁰ Both the clinical findings and histopathology of our case were not characteristic of FDE.

Necrotizing fasciitis is a fulminant, life-threatening, soft-tissue infection precipitated by polymicrobial flora. Early recognition of NF is difficult, as in its early stages it can mimic cellulitis. As the infection takes its course, necrosis can extend from the skin and into the subcutaneous tissue. Patients also develop fever, leukocytosis, and signs of sepsis. Histopathology demonstrates neutrophilic infiltration with bacterial invasion as well as necrosis of the superficial fascia and subepidermal edema.¹¹ Pyoderma gangrenosum previously has been reported to mimic NF; however, lack of responsiveness to antibiotic therapy would favor a diagnosis of PG over NF.¹²

Treatment of PG is driven by the extent of cutaneous involvement. In mild cases, wound care and topical therapy with corticosteroids and tacrolimus may suffice. Severe cases necessitate systemic therapy with oral corticosteroids or cyclosporine; biologic therapy also may play a role in treatment.⁴ In patients with hematologic malignancy, chemotherapy alone may partially or completely resolve the lesion; however, systemic corticosteroids commonly are included in management.³

To the Editor:

Iododerma is a rare dermatologic condition caused by exposure to iodinated contrast media, oral iodine suspensions, or topical povidone-iodine that can manifest as eruptive acneform lesions.^1-3

A 27-year-old woman in septic shock presented for worsening facial lesions that showed no improvement on broad-spectrum antibiotics, antifungals, and antivirals. She initially presented to an outside hospital with abdominal pain and underwent computed tomography (CT) with intravenous (IV) iodinated contrast; 24 hours after this imaging study, the family reported the appearance of “explosive acne overnight.” The lesions first appeared as vegetative and acneform ulcerations on the face. A second abdominal CT scan with IV contrast was performed 4 days after the initial scan, given the concern for spontaneous bacterial peritonitis. Hours after the second study, the lesions progressed to involve the buccal mucosae, tongue, mucosal airway, and distal arms and legs. She became progressively disoriented and developed an altered mentation over the course of the following week. Due to progressive facial edema, she required intubation 5 days after the second CT scan.

The patient had a medical history of end-stage renal disease secondary to crescenteric glomerulonephritis on peritoneal dialysis. Physical examination revealed numerous beefy-red, heaped-up, weepy, crusted nodules clustered on the face (Figure 1) and a few newer bullous-appearing lesions on the hands and feet. She had similar lesions involving the buccal mucosae and tongue with substantial facial edema. Infectious workup was notable for a positive skin culture growing methicillin-susceptible Staphylococcus aureus. All blood and tissue cultures as well as serologies for fungal and viral etiologies were negative. A tissue biopsy revealed necrosis with a neutrophilic infiltrate with mixed cell inflammation (Figure 2), and direct immunofluorescence was negative.

The patient initially was thought to be septic due to viral or bacterial infection. She was transferred from an outside hospital 7 days after the initial appearance of the acneform lesions, having already received IV contrast on 2 occasions within the first 48 hours of illness. Infectious disease was consulted and initiated broad-spectrum antiviral, antimicrobial, and antifungal therapy with acyclovir, linezolid, meropenem, and later micafungin without improvement. The diagnosis of iododerma ultimately was established based on the patient’s elevated urinary iodine levels with preceding iodine exposure in the context of renal failure. The preferential involvement of sebaceous areas and pathology findings were supportive of this diagnosis. Aggressive supportive measures including respiratory support, IV fluids, and dialysis were initiated. Topical iodine solutions, iodine-containing medications, and additional contrast subsequently were avoided. Despite these supportive measures, the patient died within 48 hours of admission from acute respiratory failure. Her autopsy attributed “septic complications of multifocal ulcerative cutaneous disease” as the anatomic cause of death.

Iododerma is an extremely rare neutrophilic dermatosis. The proposed mechanism of action involves a cell-mediated hypersensitivity reaction to iodine with induction of neutrophil degranulation.² There have been documented cases with exposure to oral potassium iodide supplements, amiodarone, topical povidone-iodine, and IV iodinated contrast material.^1-3 Iododerma typically presents 1 to 3 days after exposure to iodine. The most common source is IV radiocontrast. Diagnosis is based on the clinical presentation including acneform to vegetative nodular or bullous eruptions involving sebaceous areas in the context of recent iodine exposure. Elevated urinary iodine levels and histologic findings of neutrophilic infiltrate of the dermis support the diagnosis.^3,4

Although there have been reported cases of iododerma in patients with normal renal function, patients with renal failure are much more susceptible due to the decreased clearance of iodine.⁵ The plasma half-life of radiocontrast is 23 hours in patients with end-stage renal disease vs 2 hours in patients with normal kidney function.³ Dosage adjustments for renal impairment have not been well studied, and no specific guidelines exist for the prevention of iododerma in patients with renal failure.

The first step in treating iododerma is to remove the offending iodine-containing agent. In most cases, cutaneous lesions resolve in 4 to 6 weeks after discontinuation of the source of iodine; however, there have been reported fatalities in the literature secondary to pulmonary edema in patients with iododerma.^6,7 Despite the rarity and diagnostically challenging nature of iododerma, early recognition of this disease is crucial. Although our patient showed symptoms of iododerma after 1 dose of radiocontrast, she was not diagnosed at that time and received a second imaging study with contrast less than 48 hours later. These 2 consecutive exposures to iodine as well as the delayed diagnosis unfortunately resulted in rapid clinical deterioration.

The mainstay of therapy for iododerma includes avoidance of iodine-containing materials as soon as the diagnosis is suspected as well as supportive care. Patients have been successfully treated with systemic corticosteroids, with the addition of cyclosporine and hemodialysis in severe cases.³ Patients with a history of iododerma are advised to avoid iodine in their diet, in topical preparations, and in future imaging studies.⁸

To the Editor:

Iododerma is a rare dermatologic condition caused by exposure to iodinated contrast media, oral iodine suspensions, or topical povidone-iodine that can manifest as eruptive acneform lesions.^1-3

A 27-year-old woman in septic shock presented for worsening facial lesions that showed no improvement on broad-spectrum antibiotics, antifungals, and antivirals. She initially presented to an outside hospital with abdominal pain and underwent computed tomography (CT) with intravenous (IV) iodinated contrast; 24 hours after this imaging study, the family reported the appearance of “explosive acne overnight.” The lesions first appeared as vegetative and acneform ulcerations on the face. A second abdominal CT scan with IV contrast was performed 4 days after the initial scan, given the concern for spontaneous bacterial peritonitis. Hours after the second study, the lesions progressed to involve the buccal mucosae, tongue, mucosal airway, and distal arms and legs. She became progressively disoriented and developed an altered mentation over the course of the following week. Due to progressive facial edema, she required intubation 5 days after the second CT scan.

The patient had a medical history of end-stage renal disease secondary to crescenteric glomerulonephritis on peritoneal dialysis. Physical examination revealed numerous beefy-red, heaped-up, weepy, crusted nodules clustered on the face (Figure 1) and a few newer bullous-appearing lesions on the hands and feet. She had similar lesions involving the buccal mucosae and tongue with substantial facial edema. Infectious workup was notable for a positive skin culture growing methicillin-susceptible Staphylococcus aureus. All blood and tissue cultures as well as serologies for fungal and viral etiologies were negative. A tissue biopsy revealed necrosis with a neutrophilic infiltrate with mixed cell inflammation (Figure 2), and direct immunofluorescence was negative.

The patient initially was thought to be septic due to viral or bacterial infection. She was transferred from an outside hospital 7 days after the initial appearance of the acneform lesions, having already received IV contrast on 2 occasions within the first 48 hours of illness. Infectious disease was consulted and initiated broad-spectrum antiviral, antimicrobial, and antifungal therapy with acyclovir, linezolid, meropenem, and later micafungin without improvement. The diagnosis of iododerma ultimately was established based on the patient’s elevated urinary iodine levels with preceding iodine exposure in the context of renal failure. The preferential involvement of sebaceous areas and pathology findings were supportive of this diagnosis. Aggressive supportive measures including respiratory support, IV fluids, and dialysis were initiated. Topical iodine solutions, iodine-containing medications, and additional contrast subsequently were avoided. Despite these supportive measures, the patient died within 48 hours of admission from acute respiratory failure. Her autopsy attributed “septic complications of multifocal ulcerative cutaneous disease” as the anatomic cause of death.

Iododerma is an extremely rare neutrophilic dermatosis. The proposed mechanism of action involves a cell-mediated hypersensitivity reaction to iodine with induction of neutrophil degranulation.² There have been documented cases with exposure to oral potassium iodide supplements, amiodarone, topical povidone-iodine, and IV iodinated contrast material.^1-3 Iododerma typically presents 1 to 3 days after exposure to iodine. The most common source is IV radiocontrast. Diagnosis is based on the clinical presentation including acneform to vegetative nodular or bullous eruptions involving sebaceous areas in the context of recent iodine exposure. Elevated urinary iodine levels and histologic findings of neutrophilic infiltrate of the dermis support the diagnosis.^3,4

Although there have been reported cases of iododerma in patients with normal renal function, patients with renal failure are much more susceptible due to the decreased clearance of iodine.⁵ The plasma half-life of radiocontrast is 23 hours in patients with end-stage renal disease vs 2 hours in patients with normal kidney function.³ Dosage adjustments for renal impairment have not been well studied, and no specific guidelines exist for the prevention of iododerma in patients with renal failure.

The first step in treating iododerma is to remove the offending iodine-containing agent. In most cases, cutaneous lesions resolve in 4 to 6 weeks after discontinuation of the source of iodine; however, there have been reported fatalities in the literature secondary to pulmonary edema in patients with iododerma.^6,7 Despite the rarity and diagnostically challenging nature of iododerma, early recognition of this disease is crucial. Although our patient showed symptoms of iododerma after 1 dose of radiocontrast, she was not diagnosed at that time and received a second imaging study with contrast less than 48 hours later. These 2 consecutive exposures to iodine as well as the delayed diagnosis unfortunately resulted in rapid clinical deterioration.

The mainstay of therapy for iododerma includes avoidance of iodine-containing materials as soon as the diagnosis is suspected as well as supportive care. Patients have been successfully treated with systemic corticosteroids, with the addition of cyclosporine and hemodialysis in severe cases.³ Patients with a history of iododerma are advised to avoid iodine in their diet, in topical preparations, and in future imaging studies.⁸

To the Editor:

Iododerma is a rare dermatologic condition caused by exposure to iodinated contrast media, oral iodine suspensions, or topical povidone-iodine that can manifest as eruptive acneform lesions.^1-3

A 27-year-old woman in septic shock presented for worsening facial lesions that showed no improvement on broad-spectrum antibiotics, antifungals, and antivirals. She initially presented to an outside hospital with abdominal pain and underwent computed tomography (CT) with intravenous (IV) iodinated contrast; 24 hours after this imaging study, the family reported the appearance of “explosive acne overnight.” The lesions first appeared as vegetative and acneform ulcerations on the face. A second abdominal CT scan with IV contrast was performed 4 days after the initial scan, given the concern for spontaneous bacterial peritonitis. Hours after the second study, the lesions progressed to involve the buccal mucosae, tongue, mucosal airway, and distal arms and legs. She became progressively disoriented and developed an altered mentation over the course of the following week. Due to progressive facial edema, she required intubation 5 days after the second CT scan.

The patient had a medical history of end-stage renal disease secondary to crescenteric glomerulonephritis on peritoneal dialysis. Physical examination revealed numerous beefy-red, heaped-up, weepy, crusted nodules clustered on the face (Figure 1) and a few newer bullous-appearing lesions on the hands and feet. She had similar lesions involving the buccal mucosae and tongue with substantial facial edema. Infectious workup was notable for a positive skin culture growing methicillin-susceptible Staphylococcus aureus. All blood and tissue cultures as well as serologies for fungal and viral etiologies were negative. A tissue biopsy revealed necrosis with a neutrophilic infiltrate with mixed cell inflammation (Figure 2), and direct immunofluorescence was negative.

The patient initially was thought to be septic due to viral or bacterial infection. She was transferred from an outside hospital 7 days after the initial appearance of the acneform lesions, having already received IV contrast on 2 occasions within the first 48 hours of illness. Infectious disease was consulted and initiated broad-spectrum antiviral, antimicrobial, and antifungal therapy with acyclovir, linezolid, meropenem, and later micafungin without improvement. The diagnosis of iododerma ultimately was established based on the patient’s elevated urinary iodine levels with preceding iodine exposure in the context of renal failure. The preferential involvement of sebaceous areas and pathology findings were supportive of this diagnosis. Aggressive supportive measures including respiratory support, IV fluids, and dialysis were initiated. Topical iodine solutions, iodine-containing medications, and additional contrast subsequently were avoided. Despite these supportive measures, the patient died within 48 hours of admission from acute respiratory failure. Her autopsy attributed “septic complications of multifocal ulcerative cutaneous disease” as the anatomic cause of death.

Iododerma is an extremely rare neutrophilic dermatosis. The proposed mechanism of action involves a cell-mediated hypersensitivity reaction to iodine with induction of neutrophil degranulation.² There have been documented cases with exposure to oral potassium iodide supplements, amiodarone, topical povidone-iodine, and IV iodinated contrast material.^1-3 Iododerma typically presents 1 to 3 days after exposure to iodine. The most common source is IV radiocontrast. Diagnosis is based on the clinical presentation including acneform to vegetative nodular or bullous eruptions involving sebaceous areas in the context of recent iodine exposure. Elevated urinary iodine levels and histologic findings of neutrophilic infiltrate of the dermis support the diagnosis.^3,4

Although there have been reported cases of iododerma in patients with normal renal function, patients with renal failure are much more susceptible due to the decreased clearance of iodine.⁵ The plasma half-life of radiocontrast is 23 hours in patients with end-stage renal disease vs 2 hours in patients with normal kidney function.³ Dosage adjustments for renal impairment have not been well studied, and no specific guidelines exist for the prevention of iododerma in patients with renal failure.

The first step in treating iododerma is to remove the offending iodine-containing agent. In most cases, cutaneous lesions resolve in 4 to 6 weeks after discontinuation of the source of iodine; however, there have been reported fatalities in the literature secondary to pulmonary edema in patients with iododerma.^6,7 Despite the rarity and diagnostically challenging nature of iododerma, early recognition of this disease is crucial. Although our patient showed symptoms of iododerma after 1 dose of radiocontrast, she was not diagnosed at that time and received a second imaging study with contrast less than 48 hours later. These 2 consecutive exposures to iodine as well as the delayed diagnosis unfortunately resulted in rapid clinical deterioration.

The mainstay of therapy for iododerma includes avoidance of iodine-containing materials as soon as the diagnosis is suspected as well as supportive care. Patients have been successfully treated with systemic corticosteroids, with the addition of cyclosporine and hemodialysis in severe cases.³ Patients with a history of iododerma are advised to avoid iodine in their diet, in topical preparations, and in future imaging studies.⁸