Cleveland Clinic Journal of Medicine

Laennec’s stethoscope has survived more than 200 years, much longer than some of his contemporaries predicted. But will it survive the challenge of bedside ultrasonography and other technologic advances?

The physical examination, with its roots extending at least as far back as Hippocrates, may be at a crossroads as the mainstay of diagnosis. Physical signs can be subjective and lack sensitivity and specificity. Modern imaging and laboratory studies may already be more trusted.

If the physical examination is to survive, it must be accurate, reproducible, and efficient. Needed is a simple, evidence-based approach to the physical examination that enhances its diagnostic accuracy while maintaining bedside efficiency.

Here, we analyze the accuracy of the physical signs that are most effective in the clinical diagnosis of 4 common cardiopulmonary conditions that often present with dyspnea: pneumonia, pleural effusion, chronic obstructive pulmonary disease (COPD), and congestive heart failure.

LIKELIHOOD RATIOS

To grasp the significance of physical findings, it is necessary to understand the concept of likelihood ratios, which are widely accepted measures of the accuracy of a test or clinical finding.^1,2 The positive likelihood ratio is the probability of a disease being present when the test is positive or the clinical finding is present, while the negative likelihood ratio is the probability that the disease is present when the test is negative or the clinical finding is absent. They are calculated as follows¹:

Positive likelihood ratio = sensitivity / (1 – specificity)

Negative likelihood ratio = (1 – sensitivity) / specificity

Table 1 shows how the likelihood ratio of a test changes the posttest probability that a condition is present or absent, according to an analysis by McGee.²

STANDARDIZED TERMINOLOGY

PNEUMONIA

Pneumonia is a common disease, with more than 2 million cases annually in the United States. It is most often diagnosed by standard chest radiography, although computed tomography can identify it earlier and with higher sensitivity and specificity.⁵ The amount of published data on physical examination findings in pneumonia is surprisingly small.

Asymmetry in chest expansion: Specific, reproducible, but not sensitive

The physical finding with the highest positive likelihood ratio for diagnosing pneumonia is asymmetry in chest expansion.^6,7

From Diaz-Guzman E, Budev MM. Accuracy of the physical examination in evaluating pleural effusion. Cleve Clin J Med 2008; 75:297–303.

In a 1984 study of 1,819 patients presenting to an emergency department with acute cough, Diehr et al⁶ evaluated several physical signs of pneumonia. Asymmetric chest expansion had a specificity and positive predictive value of 100%, but its sensitivity was only 4.3%. Thus, it is not a good screening test, but it is a good diagnostic or confirmatory test. From these numbers, Metlay et al⁸ calculated that the positive likelihood ratio was infinity and the negative likelihood ratio was 0.96.

McGee,⁷ on the other hand, calculated the positive likelihood ratio of asymmetric chest expansion at 44.1. McGee also found chest expansion to be a highly reproducible finding, with an interobserver agreement kappa score of 0.85.⁷ (A kappa score of 1.0 would indicate perfect interobserver agreement.) Interestingly, chest radiographs interpreted for pulmonary infiltrates have an interobserver kappa score of only 0.38.⁷ Further studies of this physical sign could shed more light upon this area of uncertainty.

Other signs of pneumonia

None of the other physical signs studied for the diagnosis of pneumonia has as high a positive likelihood ratio as asymmetric chest expansion.^6–12

Egophony is a high-pitched or nasal quality of the patient’s voice heard on auscultation over lung tissue that is consolidated or fibrosed, due to enhanced transmission of high-frequency sound across fluid. It is often described as the “E-to-A change.” Although listening for egophony is widely done and easy to do, we calculate that this sign has a positive likelihood ratio of only 6.8 based on pooled data from 3 trials with a total of 3,245 patients.^6,10,11

Faring less favorably, in descending order of diagnostic accuracy, are:

Percussion dullness (positive likelihood ratio 5.7 based on 4 studies with 3,653 patients)^6,10–12

Bronchophony or bronchial breath sounds (positive likelihood ratio 3.3 based on 1,118 patients)¹⁰

Crackles have long been taught as a common physical finding in pneumonia. Bohadana et al pointed out that “crackle” can be defined acoustically but does not suggest any means or site of generation.⁴ Pooled data from 4 studies in 3,647 patients^6,10–12 result in a positive likelihood ratio for crackles in the diagnosis of pneumonia of only 3.2.

Diminished breath sounds (positive likelihood ratio 2.5 based on 3 studies with 1,828 patients).^10–12

These physical examination maneuvers are time-honored and part of the rite of training for medical students and residents. As we have shown, they are not extremely helpful as individual tests in diagnosing pneumonia; however, they may be useful when used in combination as a clinical prediction rule or diagnostic algorithm. These rules often have higher diagnostic accuracy but drawbacks of taking more time and not being easily reproducible.

PLEURAL EFFUSION

Pleural effusion commonly occurs in patients with congestive heart failure, pneumonia, and malignancies. The following are signs of effusion.

Dullness to percussion had a positive likelihood ratio of 5.7 from pooled data from 3 studies analyzed by Wong et al.¹³

Asymmetric chest expansion, in a study by Kalantri et al,¹⁴ had a positive likelihood ratio of 8.1 and a negative likelihood ratio of 0.29, the latter making it a reasonably good test to help rule out a pleural effusion.

Negative signs. Since a pleural effusion is an abnormal fluid collection in the pleural space and not the lung parenchyma, one would not expect it to cause loud breath sounds, adventitious sounds, or vocal resonance. Since these 3 findings emanate from the lung, their absence would be expected to support the presence of a pleural effusion.

Tactile fremitus, also known as vocal fremitus, is the vibration felt on the chest wall while the patient is speaking. Traditionally, the patient says “ninety-nine” as the examiner feels for asymmetry in vibration. A consolidation such as pneumonia increases the vibration, while fluid in a pleural effusion diminishes it.

DIAGNOSTIC ALGORITHM FOR PNEUMONIA OR PLEURAL EFFUSION

Patients presenting with cough or dyspnea will most likely be evaluated for pneumonia and pleural effusion, among other diagnoses. We propose the following physical examination strategy in this setting.

First, evaluate the patient for asymmetric chest expansion. The positive likelihood ratio for this sign is excellent for pneumonia (44.1) and moderate for pleural effusion (8.1); therefore, both conditions are possible with a positive test.

Second, percuss the chest. Dullness to percussion has a low positive likelihood ratio for pneumonia but a moderate one for pleural effusion.¹³ The absence of this sign is only modest in excluding a pleural effusion (negative likelihood ratio 0.31 in pooled data analyzed by Wong et al).¹³

Third, auscultate the chest to elicit normal, diminished, or adventitious breath sounds. Diminished breath sounds may be noted in both conditions, but vocal resonance (egophony or bronchophony) and tactile fremitus should not be present directly over a pleural effusion. Either vocal resonance or tactile fremitus in a patient with asymmetric chest expansion would strongly support the diagnosis of pneumonia.

Figure 2 summarizes our proposed diagnostic algorithm for pneumonia and pleural effusion.

CHRONIC OBSTRUCTIVE PULMONARY DISEASE

COPD imposes a heavy burden on public health worldwide in terms of cost and mortality. It is the third leading cause of death in the United States, after heart disease and cancer.¹⁵

Spirometry remains the gold standard for diagnosis. The Global Initiative for Chronic Obstructive Lung Disease standard for diagnosing COPD was the better of 2 spirometry test results, showing a forced expiratory volume in 1 second (FEV₁) and FEV₁/forced vital capacity ratio less than 70%.¹⁶

Unfortunately, there is little evidence that physical signs aid in the early diagnosis of COPD, as physical signs of airflow limitation may not manifest until lung function is substantially impaired.^17,18

Early inspiratory crackles had a positive likelihood ratio of 14.6 based on 2 small studies.^19,20

Percussion dullness over the left sternal border in the fifth intercostal space should be present in the normal situation and is known as cardiac dullness. Absent cardiac dullness had a positive likelihood ratio of 16 and a negative likelihood ratio of 0.8 for diagnosing COPD in a study in 92 patients with a history of smoking or self-reported COPD.²¹ The kappa score was 0.49, signifying moderate interobserver  agreement.

A combined strategy using the history and physical examination may have the highest diagnostic accuracy. Many of these combinations are too cumbersome for practical clinical use. However, 1 of them is based on only 3 questions²¹:

• Has the patient smoked for more than 70-pack years?
• Has the patient been previously diagnosed with chronic bronchitis or emphysema?
• Are breath sounds diminished in intensity?

Answering yes to 2 of these questions gives a positive likelihood ratio of a diagnosis of COPD of 33.5.

Early detection of COPD may improve outcomes and lower healthcare costs and thus would be clinically useful. Unfortunately, a diagnostic approach using the history and physical in the early diagnosis of COPD remains uncertain at this time.

The clinical presentation of acute congestive heart failure has much in common with pneumonia, pleural effusion, and COPD.

Echocardiography, the gold standard for diagnosis, is costly and may not be immediately available for most patients evaluated for cardiorespiratory complaints. The American College of Cardiology reports the cost of standard echocardiography to be between $1,000 and $2,000.²² A physical examination approach in the assessment of dyspnea can be very useful.

Height of jugular venous distention approximates central venous pressure

Assessing the central venous pressure by estimating the vertical height of distention of the right internal or external jugular vein is validated and easily reproducible.^23,24 The use of the external jugular vein is supported by correlation with catheter-measured central venous pressure in critically ill patients.^25,26 The central venous pressure reflects the right atrial pressure, and in the absence of tricuspid stenosis, the right ventricular end-diastolic pressure. An elevation in central venous pressure can be seen in patients with congestive heart failure, pulmonary hypertension, and pulmonary valve stenosis.

The right side is preferred due to its anatomically direct route to the heart. In contrast, the left internal jugular vein crosses the mediastinum and can be compressed by the aorta, causing a false elevation.

In summary, an elevated jugular venous pressure on examination is a good test to rule in an elevated central venous pressure, and its absence is a good sign in ruling out an elevated central venous pressure. When using jugular venous pressure specifically for the diagnosis of congestive heart failure with reduced ejection fraction (ie, ejection fraction < 50%), the positive likelihood ratio is 6.3 based on 3 studies.^25–27

Heart failure with preserved ejection fraction has not been well studied for physical examination. The Irbesartan in Heart Failure with Preserved Ejection Fraction Trial (I-Preserve)²⁸ looked only at the sensitivity of elevated jugular venous pressure in 4,128 patients, which was 8%. Specificity was not reported.

The abdominojugular reflux

Another way to gauge the jugular venous pressure is to examine the neck veins while firmly pressing on the mid-abdomen for 10 to 15 seconds to look for the abdominojugular reflux, also known as the hepatojugular reflux. An increase in the jugular venous pressure of 3 cm from baseline constitutes a positive abdominojugular reflux. It has a positive likelihood ratio of 8.0 and a negative likelihood ratio of 0.3 for the diagnosis of congestive heart failure by the assessment of end-diastolic pressure of the left ventricle (Table 5).^29–31

The abdominojugular reflux is a much more reliable test than examination of neck veins for jugular venous pressure. The interobserver agreement for examining neck veins has a wide range of kappa scores (0.08–0.81), whereas the abdominojugular reflux has a very high kappa score of 0.92.⁷ Interestingly, chest radiography showing interstitial edema has a kappa of 0.83.⁷

Displaced apical impulse

An evaluation of the apical impulse of the heart is also a very good and quick test in the examination of patients suspected of having congestive heart failure. An abnormal finding is defined by an apical impulse displaced laterally (to the left of the midclavicular line).

Using data from several studies,^32–35 a displaced apical impulse has a positive likelihood ratio of 10.3. The absence of this finding, however, is not very good for ruling out congestive heart failure, with a negative likelihood ratio of 0.7. Interobserver agreement is moderate to excellent (kappa score 0.43–0.86).⁷

A third heart sound

Auscultation to assess the third heart sound is much more difficult. A systematic review found that likelihood ratios vary widely and confidence intervals are wide.³⁶ Interobserver agreement also varies widely (kappa scores –0.17 to 0.84).⁷ In a primary care study,³⁷ a third heart sound had a very low sensitivity (4.3%) but a specificity of 99.8%.

Therefore, we are uncertain about a conclusion for this physical finding based on the concern for wide ranges in likelihood ratio and poor interobserver reliability.

PHYSICAL EXAMINATION STILL HAS A FUTURE

The physical examination has a long and distinguished place in the history of medicine. Technologic advances have changed the manner in which clinicians practice the art of healing. Modern technology in US healthcare has become a double-edged sword, with many benefits as well as detriments.³ Reproducibility and accuracy are paramount for the physical examination to remain a core component of medical diagnosis. Advances in the diagnostic accuracy of laboratory and imaging studies challenge the importance of the physical examination. However, we firmly believe that the traditional techniques have stood the test of time and have a future in the clinical practice of medicine.

Acknowledgments: The authors thank Ruby Marr, MD, Mohammed Nabhan, MD, Rajiv Doddamani, MD, and Sohaib Galani, MD, for their important contributions to this article, which included research assistance and editorial advice.

Laennec’s stethoscope has survived more than 200 years, much longer than some of his contemporaries predicted. But will it survive the challenge of bedside ultrasonography and other technologic advances?

The physical examination, with its roots extending at least as far back as Hippocrates, may be at a crossroads as the mainstay of diagnosis. Physical signs can be subjective and lack sensitivity and specificity. Modern imaging and laboratory studies may already be more trusted.

If the physical examination is to survive, it must be accurate, reproducible, and efficient. Needed is a simple, evidence-based approach to the physical examination that enhances its diagnostic accuracy while maintaining bedside efficiency.

Here, we analyze the accuracy of the physical signs that are most effective in the clinical diagnosis of 4 common cardiopulmonary conditions that often present with dyspnea: pneumonia, pleural effusion, chronic obstructive pulmonary disease (COPD), and congestive heart failure.

LIKELIHOOD RATIOS

To grasp the significance of physical findings, it is necessary to understand the concept of likelihood ratios, which are widely accepted measures of the accuracy of a test or clinical finding.^1,2 The positive likelihood ratio is the probability of a disease being present when the test is positive or the clinical finding is present, while the negative likelihood ratio is the probability that the disease is present when the test is negative or the clinical finding is absent. They are calculated as follows¹:

Positive likelihood ratio = sensitivity / (1 – specificity)

Negative likelihood ratio = (1 – sensitivity) / specificity

Table 1 shows how the likelihood ratio of a test changes the posttest probability that a condition is present or absent, according to an analysis by McGee.²

STANDARDIZED TERMINOLOGY

PNEUMONIA

Pneumonia is a common disease, with more than 2 million cases annually in the United States. It is most often diagnosed by standard chest radiography, although computed tomography can identify it earlier and with higher sensitivity and specificity.⁵ The amount of published data on physical examination findings in pneumonia is surprisingly small.

Asymmetry in chest expansion: Specific, reproducible, but not sensitive

The physical finding with the highest positive likelihood ratio for diagnosing pneumonia is asymmetry in chest expansion.^6,7

From Diaz-Guzman E, Budev MM. Accuracy of the physical examination in evaluating pleural effusion. Cleve Clin J Med 2008; 75:297–303.

In a 1984 study of 1,819 patients presenting to an emergency department with acute cough, Diehr et al⁶ evaluated several physical signs of pneumonia. Asymmetric chest expansion had a specificity and positive predictive value of 100%, but its sensitivity was only 4.3%. Thus, it is not a good screening test, but it is a good diagnostic or confirmatory test. From these numbers, Metlay et al⁸ calculated that the positive likelihood ratio was infinity and the negative likelihood ratio was 0.96.

McGee,⁷ on the other hand, calculated the positive likelihood ratio of asymmetric chest expansion at 44.1. McGee also found chest expansion to be a highly reproducible finding, with an interobserver agreement kappa score of 0.85.⁷ (A kappa score of 1.0 would indicate perfect interobserver agreement.) Interestingly, chest radiographs interpreted for pulmonary infiltrates have an interobserver kappa score of only 0.38.⁷ Further studies of this physical sign could shed more light upon this area of uncertainty.

Other signs of pneumonia

None of the other physical signs studied for the diagnosis of pneumonia has as high a positive likelihood ratio as asymmetric chest expansion.^6–12

Egophony is a high-pitched or nasal quality of the patient’s voice heard on auscultation over lung tissue that is consolidated or fibrosed, due to enhanced transmission of high-frequency sound across fluid. It is often described as the “E-to-A change.” Although listening for egophony is widely done and easy to do, we calculate that this sign has a positive likelihood ratio of only 6.8 based on pooled data from 3 trials with a total of 3,245 patients.^6,10,11

Faring less favorably, in descending order of diagnostic accuracy, are:

Percussion dullness (positive likelihood ratio 5.7 based on 4 studies with 3,653 patients)^6,10–12

Bronchophony or bronchial breath sounds (positive likelihood ratio 3.3 based on 1,118 patients)¹⁰

Crackles have long been taught as a common physical finding in pneumonia. Bohadana et al pointed out that “crackle” can be defined acoustically but does not suggest any means or site of generation.⁴ Pooled data from 4 studies in 3,647 patients^6,10–12 result in a positive likelihood ratio for crackles in the diagnosis of pneumonia of only 3.2.

Diminished breath sounds (positive likelihood ratio 2.5 based on 3 studies with 1,828 patients).^10–12

These physical examination maneuvers are time-honored and part of the rite of training for medical students and residents. As we have shown, they are not extremely helpful as individual tests in diagnosing pneumonia; however, they may be useful when used in combination as a clinical prediction rule or diagnostic algorithm. These rules often have higher diagnostic accuracy but drawbacks of taking more time and not being easily reproducible.

PLEURAL EFFUSION

Pleural effusion commonly occurs in patients with congestive heart failure, pneumonia, and malignancies. The following are signs of effusion.

Dullness to percussion had a positive likelihood ratio of 5.7 from pooled data from 3 studies analyzed by Wong et al.¹³

Asymmetric chest expansion, in a study by Kalantri et al,¹⁴ had a positive likelihood ratio of 8.1 and a negative likelihood ratio of 0.29, the latter making it a reasonably good test to help rule out a pleural effusion.

Negative signs. Since a pleural effusion is an abnormal fluid collection in the pleural space and not the lung parenchyma, one would not expect it to cause loud breath sounds, adventitious sounds, or vocal resonance. Since these 3 findings emanate from the lung, their absence would be expected to support the presence of a pleural effusion.

Tactile fremitus, also known as vocal fremitus, is the vibration felt on the chest wall while the patient is speaking. Traditionally, the patient says “ninety-nine” as the examiner feels for asymmetry in vibration. A consolidation such as pneumonia increases the vibration, while fluid in a pleural effusion diminishes it.

DIAGNOSTIC ALGORITHM FOR PNEUMONIA OR PLEURAL EFFUSION

Patients presenting with cough or dyspnea will most likely be evaluated for pneumonia and pleural effusion, among other diagnoses. We propose the following physical examination strategy in this setting.

First, evaluate the patient for asymmetric chest expansion. The positive likelihood ratio for this sign is excellent for pneumonia (44.1) and moderate for pleural effusion (8.1); therefore, both conditions are possible with a positive test.

Second, percuss the chest. Dullness to percussion has a low positive likelihood ratio for pneumonia but a moderate one for pleural effusion.¹³ The absence of this sign is only modest in excluding a pleural effusion (negative likelihood ratio 0.31 in pooled data analyzed by Wong et al).¹³

Third, auscultate the chest to elicit normal, diminished, or adventitious breath sounds. Diminished breath sounds may be noted in both conditions, but vocal resonance (egophony or bronchophony) and tactile fremitus should not be present directly over a pleural effusion. Either vocal resonance or tactile fremitus in a patient with asymmetric chest expansion would strongly support the diagnosis of pneumonia.

Figure 2 summarizes our proposed diagnostic algorithm for pneumonia and pleural effusion.

CHRONIC OBSTRUCTIVE PULMONARY DISEASE

COPD imposes a heavy burden on public health worldwide in terms of cost and mortality. It is the third leading cause of death in the United States, after heart disease and cancer.¹⁵

Spirometry remains the gold standard for diagnosis. The Global Initiative for Chronic Obstructive Lung Disease standard for diagnosing COPD was the better of 2 spirometry test results, showing a forced expiratory volume in 1 second (FEV₁) and FEV₁/forced vital capacity ratio less than 70%.¹⁶

Unfortunately, there is little evidence that physical signs aid in the early diagnosis of COPD, as physical signs of airflow limitation may not manifest until lung function is substantially impaired.^17,18

Early inspiratory crackles had a positive likelihood ratio of 14.6 based on 2 small studies.^19,20

Percussion dullness over the left sternal border in the fifth intercostal space should be present in the normal situation and is known as cardiac dullness. Absent cardiac dullness had a positive likelihood ratio of 16 and a negative likelihood ratio of 0.8 for diagnosing COPD in a study in 92 patients with a history of smoking or self-reported COPD.²¹ The kappa score was 0.49, signifying moderate interobserver  agreement.

A combined strategy using the history and physical examination may have the highest diagnostic accuracy. Many of these combinations are too cumbersome for practical clinical use. However, 1 of them is based on only 3 questions²¹:

• Has the patient smoked for more than 70-pack years?
• Has the patient been previously diagnosed with chronic bronchitis or emphysema?
• Are breath sounds diminished in intensity?

Answering yes to 2 of these questions gives a positive likelihood ratio of a diagnosis of COPD of 33.5.

Early detection of COPD may improve outcomes and lower healthcare costs and thus would be clinically useful. Unfortunately, a diagnostic approach using the history and physical in the early diagnosis of COPD remains uncertain at this time.

The clinical presentation of acute congestive heart failure has much in common with pneumonia, pleural effusion, and COPD.

Echocardiography, the gold standard for diagnosis, is costly and may not be immediately available for most patients evaluated for cardiorespiratory complaints. The American College of Cardiology reports the cost of standard echocardiography to be between $1,000 and $2,000.²² A physical examination approach in the assessment of dyspnea can be very useful.

Height of jugular venous distention approximates central venous pressure

Assessing the central venous pressure by estimating the vertical height of distention of the right internal or external jugular vein is validated and easily reproducible.^23,24 The use of the external jugular vein is supported by correlation with catheter-measured central venous pressure in critically ill patients.^25,26 The central venous pressure reflects the right atrial pressure, and in the absence of tricuspid stenosis, the right ventricular end-diastolic pressure. An elevation in central venous pressure can be seen in patients with congestive heart failure, pulmonary hypertension, and pulmonary valve stenosis.

The right side is preferred due to its anatomically direct route to the heart. In contrast, the left internal jugular vein crosses the mediastinum and can be compressed by the aorta, causing a false elevation.

In summary, an elevated jugular venous pressure on examination is a good test to rule in an elevated central venous pressure, and its absence is a good sign in ruling out an elevated central venous pressure. When using jugular venous pressure specifically for the diagnosis of congestive heart failure with reduced ejection fraction (ie, ejection fraction < 50%), the positive likelihood ratio is 6.3 based on 3 studies.^25–27

Heart failure with preserved ejection fraction has not been well studied for physical examination. The Irbesartan in Heart Failure with Preserved Ejection Fraction Trial (I-Preserve)²⁸ looked only at the sensitivity of elevated jugular venous pressure in 4,128 patients, which was 8%. Specificity was not reported.

The abdominojugular reflux

Another way to gauge the jugular venous pressure is to examine the neck veins while firmly pressing on the mid-abdomen for 10 to 15 seconds to look for the abdominojugular reflux, also known as the hepatojugular reflux. An increase in the jugular venous pressure of 3 cm from baseline constitutes a positive abdominojugular reflux. It has a positive likelihood ratio of 8.0 and a negative likelihood ratio of 0.3 for the diagnosis of congestive heart failure by the assessment of end-diastolic pressure of the left ventricle (Table 5).^29–31

The abdominojugular reflux is a much more reliable test than examination of neck veins for jugular venous pressure. The interobserver agreement for examining neck veins has a wide range of kappa scores (0.08–0.81), whereas the abdominojugular reflux has a very high kappa score of 0.92.⁷ Interestingly, chest radiography showing interstitial edema has a kappa of 0.83.⁷

Displaced apical impulse

An evaluation of the apical impulse of the heart is also a very good and quick test in the examination of patients suspected of having congestive heart failure. An abnormal finding is defined by an apical impulse displaced laterally (to the left of the midclavicular line).

Using data from several studies,^32–35 a displaced apical impulse has a positive likelihood ratio of 10.3. The absence of this finding, however, is not very good for ruling out congestive heart failure, with a negative likelihood ratio of 0.7. Interobserver agreement is moderate to excellent (kappa score 0.43–0.86).⁷

A third heart sound

Auscultation to assess the third heart sound is much more difficult. A systematic review found that likelihood ratios vary widely and confidence intervals are wide.³⁶ Interobserver agreement also varies widely (kappa scores –0.17 to 0.84).⁷ In a primary care study,³⁷ a third heart sound had a very low sensitivity (4.3%) but a specificity of 99.8%.

Therefore, we are uncertain about a conclusion for this physical finding based on the concern for wide ranges in likelihood ratio and poor interobserver reliability.

PHYSICAL EXAMINATION STILL HAS A FUTURE

The physical examination has a long and distinguished place in the history of medicine. Technologic advances have changed the manner in which clinicians practice the art of healing. Modern technology in US healthcare has become a double-edged sword, with many benefits as well as detriments.³ Reproducibility and accuracy are paramount for the physical examination to remain a core component of medical diagnosis. Advances in the diagnostic accuracy of laboratory and imaging studies challenge the importance of the physical examination. However, we firmly believe that the traditional techniques have stood the test of time and have a future in the clinical practice of medicine.

Acknowledgments: The authors thank Ruby Marr, MD, Mohammed Nabhan, MD, Rajiv Doddamani, MD, and Sohaib Galani, MD, for their important contributions to this article, which included research assistance and editorial advice.

Laennec’s stethoscope has survived more than 200 years, much longer than some of his contemporaries predicted. But will it survive the challenge of bedside ultrasonography and other technologic advances?

The physical examination, with its roots extending at least as far back as Hippocrates, may be at a crossroads as the mainstay of diagnosis. Physical signs can be subjective and lack sensitivity and specificity. Modern imaging and laboratory studies may already be more trusted.

If the physical examination is to survive, it must be accurate, reproducible, and efficient. Needed is a simple, evidence-based approach to the physical examination that enhances its diagnostic accuracy while maintaining bedside efficiency.

Here, we analyze the accuracy of the physical signs that are most effective in the clinical diagnosis of 4 common cardiopulmonary conditions that often present with dyspnea: pneumonia, pleural effusion, chronic obstructive pulmonary disease (COPD), and congestive heart failure.

LIKELIHOOD RATIOS

To grasp the significance of physical findings, it is necessary to understand the concept of likelihood ratios, which are widely accepted measures of the accuracy of a test or clinical finding.^1,2 The positive likelihood ratio is the probability of a disease being present when the test is positive or the clinical finding is present, while the negative likelihood ratio is the probability that the disease is present when the test is negative or the clinical finding is absent. They are calculated as follows¹:

Positive likelihood ratio = sensitivity / (1 – specificity)

Negative likelihood ratio = (1 – sensitivity) / specificity

Table 1 shows how the likelihood ratio of a test changes the posttest probability that a condition is present or absent, according to an analysis by McGee.²

STANDARDIZED TERMINOLOGY

PNEUMONIA

Pneumonia is a common disease, with more than 2 million cases annually in the United States. It is most often diagnosed by standard chest radiography, although computed tomography can identify it earlier and with higher sensitivity and specificity.⁵ The amount of published data on physical examination findings in pneumonia is surprisingly small.

Asymmetry in chest expansion: Specific, reproducible, but not sensitive

The physical finding with the highest positive likelihood ratio for diagnosing pneumonia is asymmetry in chest expansion.^6,7

From Diaz-Guzman E, Budev MM. Accuracy of the physical examination in evaluating pleural effusion. Cleve Clin J Med 2008; 75:297–303.

In a 1984 study of 1,819 patients presenting to an emergency department with acute cough, Diehr et al⁶ evaluated several physical signs of pneumonia. Asymmetric chest expansion had a specificity and positive predictive value of 100%, but its sensitivity was only 4.3%. Thus, it is not a good screening test, but it is a good diagnostic or confirmatory test. From these numbers, Metlay et al⁸ calculated that the positive likelihood ratio was infinity and the negative likelihood ratio was 0.96.

McGee,⁷ on the other hand, calculated the positive likelihood ratio of asymmetric chest expansion at 44.1. McGee also found chest expansion to be a highly reproducible finding, with an interobserver agreement kappa score of 0.85.⁷ (A kappa score of 1.0 would indicate perfect interobserver agreement.) Interestingly, chest radiographs interpreted for pulmonary infiltrates have an interobserver kappa score of only 0.38.⁷ Further studies of this physical sign could shed more light upon this area of uncertainty.

Other signs of pneumonia

None of the other physical signs studied for the diagnosis of pneumonia has as high a positive likelihood ratio as asymmetric chest expansion.^6–12

Egophony is a high-pitched or nasal quality of the patient’s voice heard on auscultation over lung tissue that is consolidated or fibrosed, due to enhanced transmission of high-frequency sound across fluid. It is often described as the “E-to-A change.” Although listening for egophony is widely done and easy to do, we calculate that this sign has a positive likelihood ratio of only 6.8 based on pooled data from 3 trials with a total of 3,245 patients.^6,10,11

Faring less favorably, in descending order of diagnostic accuracy, are:

Percussion dullness (positive likelihood ratio 5.7 based on 4 studies with 3,653 patients)^6,10–12

Bronchophony or bronchial breath sounds (positive likelihood ratio 3.3 based on 1,118 patients)¹⁰

Crackles have long been taught as a common physical finding in pneumonia. Bohadana et al pointed out that “crackle” can be defined acoustically but does not suggest any means or site of generation.⁴ Pooled data from 4 studies in 3,647 patients^6,10–12 result in a positive likelihood ratio for crackles in the diagnosis of pneumonia of only 3.2.

Diminished breath sounds (positive likelihood ratio 2.5 based on 3 studies with 1,828 patients).^10–12

These physical examination maneuvers are time-honored and part of the rite of training for medical students and residents. As we have shown, they are not extremely helpful as individual tests in diagnosing pneumonia; however, they may be useful when used in combination as a clinical prediction rule or diagnostic algorithm. These rules often have higher diagnostic accuracy but drawbacks of taking more time and not being easily reproducible.

PLEURAL EFFUSION

Pleural effusion commonly occurs in patients with congestive heart failure, pneumonia, and malignancies. The following are signs of effusion.

Dullness to percussion had a positive likelihood ratio of 5.7 from pooled data from 3 studies analyzed by Wong et al.¹³

Asymmetric chest expansion, in a study by Kalantri et al,¹⁴ had a positive likelihood ratio of 8.1 and a negative likelihood ratio of 0.29, the latter making it a reasonably good test to help rule out a pleural effusion.

Negative signs. Since a pleural effusion is an abnormal fluid collection in the pleural space and not the lung parenchyma, one would not expect it to cause loud breath sounds, adventitious sounds, or vocal resonance. Since these 3 findings emanate from the lung, their absence would be expected to support the presence of a pleural effusion.

Tactile fremitus, also known as vocal fremitus, is the vibration felt on the chest wall while the patient is speaking. Traditionally, the patient says “ninety-nine” as the examiner feels for asymmetry in vibration. A consolidation such as pneumonia increases the vibration, while fluid in a pleural effusion diminishes it.

DIAGNOSTIC ALGORITHM FOR PNEUMONIA OR PLEURAL EFFUSION

Patients presenting with cough or dyspnea will most likely be evaluated for pneumonia and pleural effusion, among other diagnoses. We propose the following physical examination strategy in this setting.

First, evaluate the patient for asymmetric chest expansion. The positive likelihood ratio for this sign is excellent for pneumonia (44.1) and moderate for pleural effusion (8.1); therefore, both conditions are possible with a positive test.

Second, percuss the chest. Dullness to percussion has a low positive likelihood ratio for pneumonia but a moderate one for pleural effusion.¹³ The absence of this sign is only modest in excluding a pleural effusion (negative likelihood ratio 0.31 in pooled data analyzed by Wong et al).¹³

Third, auscultate the chest to elicit normal, diminished, or adventitious breath sounds. Diminished breath sounds may be noted in both conditions, but vocal resonance (egophony or bronchophony) and tactile fremitus should not be present directly over a pleural effusion. Either vocal resonance or tactile fremitus in a patient with asymmetric chest expansion would strongly support the diagnosis of pneumonia.

Figure 2 summarizes our proposed diagnostic algorithm for pneumonia and pleural effusion.

CHRONIC OBSTRUCTIVE PULMONARY DISEASE

COPD imposes a heavy burden on public health worldwide in terms of cost and mortality. It is the third leading cause of death in the United States, after heart disease and cancer.¹⁵

Spirometry remains the gold standard for diagnosis. The Global Initiative for Chronic Obstructive Lung Disease standard for diagnosing COPD was the better of 2 spirometry test results, showing a forced expiratory volume in 1 second (FEV₁) and FEV₁/forced vital capacity ratio less than 70%.¹⁶

Unfortunately, there is little evidence that physical signs aid in the early diagnosis of COPD, as physical signs of airflow limitation may not manifest until lung function is substantially impaired.^17,18

Early inspiratory crackles had a positive likelihood ratio of 14.6 based on 2 small studies.^19,20

Percussion dullness over the left sternal border in the fifth intercostal space should be present in the normal situation and is known as cardiac dullness. Absent cardiac dullness had a positive likelihood ratio of 16 and a negative likelihood ratio of 0.8 for diagnosing COPD in a study in 92 patients with a history of smoking or self-reported COPD.²¹ The kappa score was 0.49, signifying moderate interobserver  agreement.

A combined strategy using the history and physical examination may have the highest diagnostic accuracy. Many of these combinations are too cumbersome for practical clinical use. However, 1 of them is based on only 3 questions²¹:

• Has the patient smoked for more than 70-pack years?
• Has the patient been previously diagnosed with chronic bronchitis or emphysema?
• Are breath sounds diminished in intensity?

Answering yes to 2 of these questions gives a positive likelihood ratio of a diagnosis of COPD of 33.5.

Early detection of COPD may improve outcomes and lower healthcare costs and thus would be clinically useful. Unfortunately, a diagnostic approach using the history and physical in the early diagnosis of COPD remains uncertain at this time.

The clinical presentation of acute congestive heart failure has much in common with pneumonia, pleural effusion, and COPD.

Echocardiography, the gold standard for diagnosis, is costly and may not be immediately available for most patients evaluated for cardiorespiratory complaints. The American College of Cardiology reports the cost of standard echocardiography to be between $1,000 and $2,000.²² A physical examination approach in the assessment of dyspnea can be very useful.

Height of jugular venous distention approximates central venous pressure

Assessing the central venous pressure by estimating the vertical height of distention of the right internal or external jugular vein is validated and easily reproducible.^23,24 The use of the external jugular vein is supported by correlation with catheter-measured central venous pressure in critically ill patients.^25,26 The central venous pressure reflects the right atrial pressure, and in the absence of tricuspid stenosis, the right ventricular end-diastolic pressure. An elevation in central venous pressure can be seen in patients with congestive heart failure, pulmonary hypertension, and pulmonary valve stenosis.

The right side is preferred due to its anatomically direct route to the heart. In contrast, the left internal jugular vein crosses the mediastinum and can be compressed by the aorta, causing a false elevation.

In summary, an elevated jugular venous pressure on examination is a good test to rule in an elevated central venous pressure, and its absence is a good sign in ruling out an elevated central venous pressure. When using jugular venous pressure specifically for the diagnosis of congestive heart failure with reduced ejection fraction (ie, ejection fraction < 50%), the positive likelihood ratio is 6.3 based on 3 studies.^25–27

Heart failure with preserved ejection fraction has not been well studied for physical examination. The Irbesartan in Heart Failure with Preserved Ejection Fraction Trial (I-Preserve)²⁸ looked only at the sensitivity of elevated jugular venous pressure in 4,128 patients, which was 8%. Specificity was not reported.

The abdominojugular reflux

Another way to gauge the jugular venous pressure is to examine the neck veins while firmly pressing on the mid-abdomen for 10 to 15 seconds to look for the abdominojugular reflux, also known as the hepatojugular reflux. An increase in the jugular venous pressure of 3 cm from baseline constitutes a positive abdominojugular reflux. It has a positive likelihood ratio of 8.0 and a negative likelihood ratio of 0.3 for the diagnosis of congestive heart failure by the assessment of end-diastolic pressure of the left ventricle (Table 5).^29–31

The abdominojugular reflux is a much more reliable test than examination of neck veins for jugular venous pressure. The interobserver agreement for examining neck veins has a wide range of kappa scores (0.08–0.81), whereas the abdominojugular reflux has a very high kappa score of 0.92.⁷ Interestingly, chest radiography showing interstitial edema has a kappa of 0.83.⁷

Displaced apical impulse

An evaluation of the apical impulse of the heart is also a very good and quick test in the examination of patients suspected of having congestive heart failure. An abnormal finding is defined by an apical impulse displaced laterally (to the left of the midclavicular line).

Using data from several studies,^32–35 a displaced apical impulse has a positive likelihood ratio of 10.3. The absence of this finding, however, is not very good for ruling out congestive heart failure, with a negative likelihood ratio of 0.7. Interobserver agreement is moderate to excellent (kappa score 0.43–0.86).⁷

A third heart sound

Auscultation to assess the third heart sound is much more difficult. A systematic review found that likelihood ratios vary widely and confidence intervals are wide.³⁶ Interobserver agreement also varies widely (kappa scores –0.17 to 0.84).⁷ In a primary care study,³⁷ a third heart sound had a very low sensitivity (4.3%) but a specificity of 99.8%.

Therefore, we are uncertain about a conclusion for this physical finding based on the concern for wide ranges in likelihood ratio and poor interobserver reliability.

PHYSICAL EXAMINATION STILL HAS A FUTURE

The physical examination has a long and distinguished place in the history of medicine. Technologic advances have changed the manner in which clinicians practice the art of healing. Modern technology in US healthcare has become a double-edged sword, with many benefits as well as detriments.³ Reproducibility and accuracy are paramount for the physical examination to remain a core component of medical diagnosis. Advances in the diagnostic accuracy of laboratory and imaging studies challenge the importance of the physical examination. However, we firmly believe that the traditional techniques have stood the test of time and have a future in the clinical practice of medicine.

Acknowledgments: The authors thank Ruby Marr, MD, Mohammed Nabhan, MD, Rajiv Doddamani, MD, and Sohaib Galani, MD, for their important contributions to this article, which included research assistance and editorial advice.

Yes. The time has come to change the way we think about fasting before routine lipid testing. We now have robust evidence supporting the routine use of nonfasting lipid testing. Fasting lipid testing is rarely needed, but may be considered for patients with very high triglycerides or before starting treatment in patients with genetic lipid disorders. For most patients, nonfasting lipid testing is appropriate: it is evidence-based, safe, valid, and convenient. More widespread adoption of this strategy by US healthcare providers would improve quality of care and patient and clinician satisfaction.

GUIDELINES HAVE CHANGED

In 2014, the US Department of Veterans Affairs practice guidelines recommended nonfasting lipid testing for cardiovascular risk assessment.¹ Other recent clinical guidelines and expert consensus statements from Europe and Canada now also recommend nonfasting lipid testing for most routine clinical evaluations.

Physiologically, we spend most of our lives in the nonfasting state, yet fasting lipid testing was standard practice advocated by earlier clinical guidelines. The rationale for fasting before measuring lipids was to reduce variability and to allow for a more accurate derivation of the low-density lipoprotein cholesterol (LDL-C) concentration using the Friedewald formula. There was also concern that an increase in triglyceride concentrations after consuming a fatty meal would reduce the validity of the results. However, numerous studies have found that the increase in plasma triglycerides after normal food intake is much less than that during a fat-tolerance test, making this less of a concern for most patients.^2,3

In addition, recent studies suggest that postprandial effects do not diminish and may even strengthen the risk associations of lipids with cardiovascular disease, in particular for triglycerides.⁴ Moreover, in certain patients, such as those with metabolic syndrome, diabetes mellitus, or certain genetic abnormalities, fasting can mask abnormalities in triglyceride-rich lipid metabolism, which may only be detected when triglycerides are measured in a nonfasting state. Nonfasting measurements may identify patients with elevated residual risk despite optimal guideline-based treatment.

Recently published recommendations for nonfasting lipid testing for routine assessments are summarized in Table 1.^1,5–11

EFFECTS OF THE POSTPRANDIAL STATE ON LIPID LEVELS AND RISK ASSESSMENT

A common concern for clinicians who do not routinely order nonfasting lipid testing is the potential variability in lipid levels and interpretation of these values for treatment decisions. But in most circumstances the differences between fasting and nonfasting measurements are small and are not clinically relevant. Differences in high-density lipoprotein  cholesterol (HDL-C) are negligible; slightly lower levels are seen (up to −8 mg/dL) for nonfasting total cholesterol, LDL-C, and non-HDL-C compared with fasting levels; and differences are modest (up to 25 mg/dL higher) for triglycerides.⁵ These data should reassure clinicians who rely on lipid levels to guide management decisions.⁹

Cardiovascular risk assessment

Current algorithms for assessing risk of cardiovascular disease use total cholesterol and HDL-C, not triglycerides or LDL-C. Hence, eating has no effect on the risk estimates.

For clinicians who prefer an absolute lipid target for managing risk in patients on lipid-modifying therapy, a nonfasting LDL-C or non-HDL-C (or apolipoprotein B) may be used. The non-HDL-C level is a better risk marker than LDL-C, particularly in patients with low LDL-C or with triglyceride levels of 200 mg/dL or higher.¹² Treatment goals for non-HDL-C are 30 mg/dL higher than for LDL-C (fasting or nonfasting). In addition, for these patients with low LDL-C or high triglycerides, a new LDL-C calculation method has more consistent results for fasting and nonfasting values than the commonly used Friedewald calculation.¹²

The adequacy of nonfasting lipid testing for general screening for cardiovascular disease has been verified in large prospective studies over the past several decades.^2,13,14 These studies evaluated cardiovascular event and mortality rates and found consistent associations of nonfasting lipid levels with cardiovascular disease. Studies that included both fasting and nonfasting patient populations found similar or occasionally even greater cardiovascular risk associations for nonfasting lipid measurements (including for LDL-C and triglycerides) compared with fasting lipid measurements.

The Emerging Risk Factors Collaboration¹⁴ reviewed the data from 68 studies in more than 300,000 people and found that the relationship between lipid levels and incident cardiovascular events was just as strong when nonfasting lipid values were used. In fact, at least 3 large statin trials reviewed (a total of 43,000 people) used nonfasting lipids.¹⁴

Genetic studies using mendelian randomization have also linked nonfasting triglyceride levels (and remnant cholesterol) to an increased risk of cardiovascular events and of death from any cause.^15,16

Therefore, the evidence overall suggests that nonfasting lipid measurements are acceptable with respect to risk assessment, and indeed may be preferred in most instances, especially in patients with an atherogenic metabolic milieu that may otherwise be masked by the fasting state.

OTHER BENEFITS OF NONFASTING LIPID TESTING

Nonfasting lipid panels are more economical and safer for certain groups, such as elderly or diabetic patients. A pilot study¹⁷ found that up to 27.1% of patients with diabetes reported experiencing a fasting-evoked hypoglycemic event en route to testing because of fasting for blood work. These events are vastly underreported and add to patient morbidity that can easily be avoided by adopting nonfasting lipid testing.

No study has assessed the cost-effectiveness of fasting vs nonfasting lipid testing. It is common for patients to present for their office appointment without having obtained a fasting lipid panel simply because they forgot to fast and were turned away by the laboratory. Thus, management decisions during the visit are often deferred, and patients must return to the laboratory and reschedule follow-up visits. This is inefficient, increases outpatient waiting times, and also potentially deprives others of access to needed care. Laboratory workflow can also suffer from an influx of early morning visits for fasting tests, decreasing system efficiency. Decreased efficiency in multiple levels of the healthcare system leads to increased costs, burden on healthcare providers, and decreased patient and physician satisfaction.

GETTING WITH THE GUIDELINES

The 2002 National Cholesterol Education Program expert panel report¹⁸ and the 2013 joint cholesterol guidelines of the American College of Cardiology and the American Heart Association⁹ both recommended that initial screening should involve fasting lipid testing, but they also allowed measuring nonfasting total cholesterol, HDL-C, and non-HDL-C.¹⁸ And internationally, there has been a shift in practice recommendations toward nonfasting lipids over the past 10 years (Table 1).

In 2014, the US Department of Veterans Affairs, the UK National Clinical Guideline Centre, and the Joint British Societies said that fasting is no longer needed for routine testing.¹⁰ In 2016, the European Atherosclerosis Society and the European Federation of Clinical Chemistry and Laboratory Medicine recommended nonfasting lipid testing as the standard of care and provided clinically useful cut points for both fasting and nonfasting lipid measurements.⁵

In most guidelines, the threshold for elevated nonfasting triglycerides was defined as 175 mg/dL (≥ 2 mmol/L) or greater, and this level has been validated prospectively in a large study of US women.^5,19 Repeat measurement of fasting triglycerides may be considered when nonfasting levels are greater than  400 mg/dL,⁵ although there is no consensus in the guidelines regarding when or if fasting triglycerides should be remeasured. (In the Danish experience,⁵ only 10% of patients have required repeat fasting values). In addition, the 2016 Canadian Hypertension Education Program guidelines⁶ removed fasting as a requirement. The 2016 Canadian Cardiovascular Society dyslipidemia guidelines⁷ reported that nonfasting lipid testing is a suitable alternative to fasting. Furthermore, the most recent revision of the European Society of Cardiology dyslipidemia guidelines⁸ acknowledged that nonfasting lipid panels are acceptable for screening and management of patients without severe hypertriglyceridemia or those with extremely low LDL-C levels.

LIMITATIONS OF THE EVIDENCE

To date, no study has assessed the predictive value of fasting vs nonfasting lipid measurements in the same individuals, and there have been no randomized outcomes trials or cost-effectiveness analyses. Ethnic variations in lipoproteins and nonfasting status also need to be investigated as nonfasting lipid testing becomes more universally accepted.

TAKE-HOME POINTS

• Robust evidence supports the routine use of nonfasting lipid testing, with fasting panels reserved potentially for patients with very high triglycerides and before starting treatment in those with genetic lipid disorders.
• For most patients, nonfasting tests are evidence-based, safe, valid, and convenient.
• More widespread adoption of this strategy by US healthcare providers would improve both quality of care and patient-clinician satisfaction.

Yes. The time has come to change the way we think about fasting before routine lipid testing. We now have robust evidence supporting the routine use of nonfasting lipid testing. Fasting lipid testing is rarely needed, but may be considered for patients with very high triglycerides or before starting treatment in patients with genetic lipid disorders. For most patients, nonfasting lipid testing is appropriate: it is evidence-based, safe, valid, and convenient. More widespread adoption of this strategy by US healthcare providers would improve quality of care and patient and clinician satisfaction.

GUIDELINES HAVE CHANGED

In 2014, the US Department of Veterans Affairs practice guidelines recommended nonfasting lipid testing for cardiovascular risk assessment.¹ Other recent clinical guidelines and expert consensus statements from Europe and Canada now also recommend nonfasting lipid testing for most routine clinical evaluations.

Physiologically, we spend most of our lives in the nonfasting state, yet fasting lipid testing was standard practice advocated by earlier clinical guidelines. The rationale for fasting before measuring lipids was to reduce variability and to allow for a more accurate derivation of the low-density lipoprotein cholesterol (LDL-C) concentration using the Friedewald formula. There was also concern that an increase in triglyceride concentrations after consuming a fatty meal would reduce the validity of the results. However, numerous studies have found that the increase in plasma triglycerides after normal food intake is much less than that during a fat-tolerance test, making this less of a concern for most patients.^2,3

In addition, recent studies suggest that postprandial effects do not diminish and may even strengthen the risk associations of lipids with cardiovascular disease, in particular for triglycerides.⁴ Moreover, in certain patients, such as those with metabolic syndrome, diabetes mellitus, or certain genetic abnormalities, fasting can mask abnormalities in triglyceride-rich lipid metabolism, which may only be detected when triglycerides are measured in a nonfasting state. Nonfasting measurements may identify patients with elevated residual risk despite optimal guideline-based treatment.

Recently published recommendations for nonfasting lipid testing for routine assessments are summarized in Table 1.^1,5–11

EFFECTS OF THE POSTPRANDIAL STATE ON LIPID LEVELS AND RISK ASSESSMENT

A common concern for clinicians who do not routinely order nonfasting lipid testing is the potential variability in lipid levels and interpretation of these values for treatment decisions. But in most circumstances the differences between fasting and nonfasting measurements are small and are not clinically relevant. Differences in high-density lipoprotein  cholesterol (HDL-C) are negligible; slightly lower levels are seen (up to −8 mg/dL) for nonfasting total cholesterol, LDL-C, and non-HDL-C compared with fasting levels; and differences are modest (up to 25 mg/dL higher) for triglycerides.⁵ These data should reassure clinicians who rely on lipid levels to guide management decisions.⁹

Cardiovascular risk assessment

Current algorithms for assessing risk of cardiovascular disease use total cholesterol and HDL-C, not triglycerides or LDL-C. Hence, eating has no effect on the risk estimates.

For clinicians who prefer an absolute lipid target for managing risk in patients on lipid-modifying therapy, a nonfasting LDL-C or non-HDL-C (or apolipoprotein B) may be used. The non-HDL-C level is a better risk marker than LDL-C, particularly in patients with low LDL-C or with triglyceride levels of 200 mg/dL or higher.¹² Treatment goals for non-HDL-C are 30 mg/dL higher than for LDL-C (fasting or nonfasting). In addition, for these patients with low LDL-C or high triglycerides, a new LDL-C calculation method has more consistent results for fasting and nonfasting values than the commonly used Friedewald calculation.¹²

The adequacy of nonfasting lipid testing for general screening for cardiovascular disease has been verified in large prospective studies over the past several decades.^2,13,14 These studies evaluated cardiovascular event and mortality rates and found consistent associations of nonfasting lipid levels with cardiovascular disease. Studies that included both fasting and nonfasting patient populations found similar or occasionally even greater cardiovascular risk associations for nonfasting lipid measurements (including for LDL-C and triglycerides) compared with fasting lipid measurements.

The Emerging Risk Factors Collaboration¹⁴ reviewed the data from 68 studies in more than 300,000 people and found that the relationship between lipid levels and incident cardiovascular events was just as strong when nonfasting lipid values were used. In fact, at least 3 large statin trials reviewed (a total of 43,000 people) used nonfasting lipids.¹⁴

Genetic studies using mendelian randomization have also linked nonfasting triglyceride levels (and remnant cholesterol) to an increased risk of cardiovascular events and of death from any cause.^15,16

Therefore, the evidence overall suggests that nonfasting lipid measurements are acceptable with respect to risk assessment, and indeed may be preferred in most instances, especially in patients with an atherogenic metabolic milieu that may otherwise be masked by the fasting state.

OTHER BENEFITS OF NONFASTING LIPID TESTING

Nonfasting lipid panels are more economical and safer for certain groups, such as elderly or diabetic patients. A pilot study¹⁷ found that up to 27.1% of patients with diabetes reported experiencing a fasting-evoked hypoglycemic event en route to testing because of fasting for blood work. These events are vastly underreported and add to patient morbidity that can easily be avoided by adopting nonfasting lipid testing.

No study has assessed the cost-effectiveness of fasting vs nonfasting lipid testing. It is common for patients to present for their office appointment without having obtained a fasting lipid panel simply because they forgot to fast and were turned away by the laboratory. Thus, management decisions during the visit are often deferred, and patients must return to the laboratory and reschedule follow-up visits. This is inefficient, increases outpatient waiting times, and also potentially deprives others of access to needed care. Laboratory workflow can also suffer from an influx of early morning visits for fasting tests, decreasing system efficiency. Decreased efficiency in multiple levels of the healthcare system leads to increased costs, burden on healthcare providers, and decreased patient and physician satisfaction.

GETTING WITH THE GUIDELINES

The 2002 National Cholesterol Education Program expert panel report¹⁸ and the 2013 joint cholesterol guidelines of the American College of Cardiology and the American Heart Association⁹ both recommended that initial screening should involve fasting lipid testing, but they also allowed measuring nonfasting total cholesterol, HDL-C, and non-HDL-C.¹⁸ And internationally, there has been a shift in practice recommendations toward nonfasting lipids over the past 10 years (Table 1).

In 2014, the US Department of Veterans Affairs, the UK National Clinical Guideline Centre, and the Joint British Societies said that fasting is no longer needed for routine testing.¹⁰ In 2016, the European Atherosclerosis Society and the European Federation of Clinical Chemistry and Laboratory Medicine recommended nonfasting lipid testing as the standard of care and provided clinically useful cut points for both fasting and nonfasting lipid measurements.⁵

In most guidelines, the threshold for elevated nonfasting triglycerides was defined as 175 mg/dL (≥ 2 mmol/L) or greater, and this level has been validated prospectively in a large study of US women.^5,19 Repeat measurement of fasting triglycerides may be considered when nonfasting levels are greater than  400 mg/dL,⁵ although there is no consensus in the guidelines regarding when or if fasting triglycerides should be remeasured. (In the Danish experience,⁵ only 10% of patients have required repeat fasting values). In addition, the 2016 Canadian Hypertension Education Program guidelines⁶ removed fasting as a requirement. The 2016 Canadian Cardiovascular Society dyslipidemia guidelines⁷ reported that nonfasting lipid testing is a suitable alternative to fasting. Furthermore, the most recent revision of the European Society of Cardiology dyslipidemia guidelines⁸ acknowledged that nonfasting lipid panels are acceptable for screening and management of patients without severe hypertriglyceridemia or those with extremely low LDL-C levels.

LIMITATIONS OF THE EVIDENCE

To date, no study has assessed the predictive value of fasting vs nonfasting lipid measurements in the same individuals, and there have been no randomized outcomes trials or cost-effectiveness analyses. Ethnic variations in lipoproteins and nonfasting status also need to be investigated as nonfasting lipid testing becomes more universally accepted.

TAKE-HOME POINTS

• Robust evidence supports the routine use of nonfasting lipid testing, with fasting panels reserved potentially for patients with very high triglycerides and before starting treatment in those with genetic lipid disorders.
• For most patients, nonfasting tests are evidence-based, safe, valid, and convenient.
• More widespread adoption of this strategy by US healthcare providers would improve both quality of care and patient-clinician satisfaction.

Yes. The time has come to change the way we think about fasting before routine lipid testing. We now have robust evidence supporting the routine use of nonfasting lipid testing. Fasting lipid testing is rarely needed, but may be considered for patients with very high triglycerides or before starting treatment in patients with genetic lipid disorders. For most patients, nonfasting lipid testing is appropriate: it is evidence-based, safe, valid, and convenient. More widespread adoption of this strategy by US healthcare providers would improve quality of care and patient and clinician satisfaction.

GUIDELINES HAVE CHANGED

In 2014, the US Department of Veterans Affairs practice guidelines recommended nonfasting lipid testing for cardiovascular risk assessment.¹ Other recent clinical guidelines and expert consensus statements from Europe and Canada now also recommend nonfasting lipid testing for most routine clinical evaluations.

Physiologically, we spend most of our lives in the nonfasting state, yet fasting lipid testing was standard practice advocated by earlier clinical guidelines. The rationale for fasting before measuring lipids was to reduce variability and to allow for a more accurate derivation of the low-density lipoprotein cholesterol (LDL-C) concentration using the Friedewald formula. There was also concern that an increase in triglyceride concentrations after consuming a fatty meal would reduce the validity of the results. However, numerous studies have found that the increase in plasma triglycerides after normal food intake is much less than that during a fat-tolerance test, making this less of a concern for most patients.^2,3

In addition, recent studies suggest that postprandial effects do not diminish and may even strengthen the risk associations of lipids with cardiovascular disease, in particular for triglycerides.⁴ Moreover, in certain patients, such as those with metabolic syndrome, diabetes mellitus, or certain genetic abnormalities, fasting can mask abnormalities in triglyceride-rich lipid metabolism, which may only be detected when triglycerides are measured in a nonfasting state. Nonfasting measurements may identify patients with elevated residual risk despite optimal guideline-based treatment.

Recently published recommendations for nonfasting lipid testing for routine assessments are summarized in Table 1.^1,5–11

EFFECTS OF THE POSTPRANDIAL STATE ON LIPID LEVELS AND RISK ASSESSMENT

A common concern for clinicians who do not routinely order nonfasting lipid testing is the potential variability in lipid levels and interpretation of these values for treatment decisions. But in most circumstances the differences between fasting and nonfasting measurements are small and are not clinically relevant. Differences in high-density lipoprotein  cholesterol (HDL-C) are negligible; slightly lower levels are seen (up to −8 mg/dL) for nonfasting total cholesterol, LDL-C, and non-HDL-C compared with fasting levels; and differences are modest (up to 25 mg/dL higher) for triglycerides.⁵ These data should reassure clinicians who rely on lipid levels to guide management decisions.⁹

Cardiovascular risk assessment

Current algorithms for assessing risk of cardiovascular disease use total cholesterol and HDL-C, not triglycerides or LDL-C. Hence, eating has no effect on the risk estimates.

For clinicians who prefer an absolute lipid target for managing risk in patients on lipid-modifying therapy, a nonfasting LDL-C or non-HDL-C (or apolipoprotein B) may be used. The non-HDL-C level is a better risk marker than LDL-C, particularly in patients with low LDL-C or with triglyceride levels of 200 mg/dL or higher.¹² Treatment goals for non-HDL-C are 30 mg/dL higher than for LDL-C (fasting or nonfasting). In addition, for these patients with low LDL-C or high triglycerides, a new LDL-C calculation method has more consistent results for fasting and nonfasting values than the commonly used Friedewald calculation.¹²

The adequacy of nonfasting lipid testing for general screening for cardiovascular disease has been verified in large prospective studies over the past several decades.^2,13,14 These studies evaluated cardiovascular event and mortality rates and found consistent associations of nonfasting lipid levels with cardiovascular disease. Studies that included both fasting and nonfasting patient populations found similar or occasionally even greater cardiovascular risk associations for nonfasting lipid measurements (including for LDL-C and triglycerides) compared with fasting lipid measurements.

The Emerging Risk Factors Collaboration¹⁴ reviewed the data from 68 studies in more than 300,000 people and found that the relationship between lipid levels and incident cardiovascular events was just as strong when nonfasting lipid values were used. In fact, at least 3 large statin trials reviewed (a total of 43,000 people) used nonfasting lipids.¹⁴

Genetic studies using mendelian randomization have also linked nonfasting triglyceride levels (and remnant cholesterol) to an increased risk of cardiovascular events and of death from any cause.^15,16

Therefore, the evidence overall suggests that nonfasting lipid measurements are acceptable with respect to risk assessment, and indeed may be preferred in most instances, especially in patients with an atherogenic metabolic milieu that may otherwise be masked by the fasting state.

OTHER BENEFITS OF NONFASTING LIPID TESTING

Nonfasting lipid panels are more economical and safer for certain groups, such as elderly or diabetic patients. A pilot study¹⁷ found that up to 27.1% of patients with diabetes reported experiencing a fasting-evoked hypoglycemic event en route to testing because of fasting for blood work. These events are vastly underreported and add to patient morbidity that can easily be avoided by adopting nonfasting lipid testing.

No study has assessed the cost-effectiveness of fasting vs nonfasting lipid testing. It is common for patients to present for their office appointment without having obtained a fasting lipid panel simply because they forgot to fast and were turned away by the laboratory. Thus, management decisions during the visit are often deferred, and patients must return to the laboratory and reschedule follow-up visits. This is inefficient, increases outpatient waiting times, and also potentially deprives others of access to needed care. Laboratory workflow can also suffer from an influx of early morning visits for fasting tests, decreasing system efficiency. Decreased efficiency in multiple levels of the healthcare system leads to increased costs, burden on healthcare providers, and decreased patient and physician satisfaction.

GETTING WITH THE GUIDELINES

The 2002 National Cholesterol Education Program expert panel report¹⁸ and the 2013 joint cholesterol guidelines of the American College of Cardiology and the American Heart Association⁹ both recommended that initial screening should involve fasting lipid testing, but they also allowed measuring nonfasting total cholesterol, HDL-C, and non-HDL-C.¹⁸ And internationally, there has been a shift in practice recommendations toward nonfasting lipids over the past 10 years (Table 1).

In 2014, the US Department of Veterans Affairs, the UK National Clinical Guideline Centre, and the Joint British Societies said that fasting is no longer needed for routine testing.¹⁰ In 2016, the European Atherosclerosis Society and the European Federation of Clinical Chemistry and Laboratory Medicine recommended nonfasting lipid testing as the standard of care and provided clinically useful cut points for both fasting and nonfasting lipid measurements.⁵

In most guidelines, the threshold for elevated nonfasting triglycerides was defined as 175 mg/dL (≥ 2 mmol/L) or greater, and this level has been validated prospectively in a large study of US women.^5,19 Repeat measurement of fasting triglycerides may be considered when nonfasting levels are greater than  400 mg/dL,⁵ although there is no consensus in the guidelines regarding when or if fasting triglycerides should be remeasured. (In the Danish experience,⁵ only 10% of patients have required repeat fasting values). In addition, the 2016 Canadian Hypertension Education Program guidelines⁶ removed fasting as a requirement. The 2016 Canadian Cardiovascular Society dyslipidemia guidelines⁷ reported that nonfasting lipid testing is a suitable alternative to fasting. Furthermore, the most recent revision of the European Society of Cardiology dyslipidemia guidelines⁸ acknowledged that nonfasting lipid panels are acceptable for screening and management of patients without severe hypertriglyceridemia or those with extremely low LDL-C levels.

LIMITATIONS OF THE EVIDENCE

To date, no study has assessed the predictive value of fasting vs nonfasting lipid measurements in the same individuals, and there have been no randomized outcomes trials or cost-effectiveness analyses. Ethnic variations in lipoproteins and nonfasting status also need to be investigated as nonfasting lipid testing becomes more universally accepted.

TAKE-HOME POINTS

• Robust evidence supports the routine use of nonfasting lipid testing, with fasting panels reserved potentially for patients with very high triglycerides and before starting treatment in those with genetic lipid disorders.
• For most patients, nonfasting tests are evidence-based, safe, valid, and convenient.
• More widespread adoption of this strategy by US healthcare providers would improve both quality of care and patient-clinician satisfaction.

The article in this issue by Bersoux et al on pharmacotherapy to manage obesity¹ is apropos in light of a recent study² showing that patients are filling 15 times more prescriptions for antidiabetic medications (excluding insulin) than for antiobesity drugs. What makes this finding significant is that nearly 3 times more adults meet the criteria for use of antiobesity drugs than for antidiabetic drugs—116 million vs 30 million, respectively.

See related article

This underuse of antiobesity medications has been noted in other studies. In 1 study,³ only about 2% of adults eligible for weight-loss drug therapy received a prescription. Conversely, about 86% of adults diagnosed with diabetes received antidiabetic medications.³

WEIGHT LOSS: IT'S IMPORTANT

This underuse of weight-loss drugs occurs despite our understanding that obesity is a risk factor for developing diabetes and that weight loss in obese patients reduces the risk.

The landmark Diabetes Prevention Program study found that even modest weight loss of 7% reduced the risk of developing diabetes by 58% in overweight and prediabetic individuals.⁴ Additionally, a 5% to 10% weight loss can lead to significant improvements in many comorbidities, including diabetes, hyperlipidemia, hypertension, sleep apnea, and fatty liver disease.

Antiobesity medications can help patients achieve weight-loss goals, especially if lifestyle and behavioral modifications alone have been unsuccessful. Data show that these drugs result in an average weight loss of 5% to 15% when added to diet and exercise.

BARRIERS TO PRESCRIBING WEIGHT-LOSS DRUGS

Why are practitioners reluctant to prescribe these drugs despite the worsening obesity epidemic and despite knowing that obesity is a risk factor for diabetes? Many of us who practice obesity medicine believe there are several reasons.

One barrier is the misconception that obesity does not warrant treatment with weight-loss medications, even though most practitioners will readily admit that patients cannot achieve effective, durable, and meaningful weight loss with behavioral changes and lifestyle modifications alone.

Other barriers stem from issues such as time constraints in the office, lack of training to treat this condition, and not enough data on the newer chronic weight-loss medications. And there are stringent requirements for patient follow-up once a medication has been initiated. Finally, it’s often difficult to obtain insurance coverage.

Addressing the barriers

Of these, I believe the biggest barrier for busy practitioners is finding the time and effort they need to devote to prescribing weight-loss medications. There are ways to address these issues.

Regarding time constraints, practitioners can discuss weight loss at follow-up visits and refer patients to obesity specialists. Regarding gaps in training and knowledge of obesity management, there are consensus guidelines for the identification, evaluation, and treatment of the overweight or obese individual.^5–7 Guidelines provide extensive information on the pharmacologic treatment of obesity. These resources provide valuable evidence-based recommendations on how to manage this chronic disease.

ARMED WITH INFORMATION, PHARMACOLOGIC OPTIONS

Bersoux et al provide another valuable resource for clinical use of weight-loss drugs.¹ They accurately review the available medications, their mechanisms of action, dosing, efficacy, side effect profiles, and clinical indications. Their review is comprehensive in every aspect of this drug class.

This is important information for practitioners to have when considering prescribing antiobesity medications. It is especially important for primary care practitioners because of the large number of obese or overweight patients they treat.

Drug options have expanded

We did not always have this many drugs to choose from. As Bersoux et al note, practitioners had limited options for weight-loss medications during the 1990s and early 2000s, and several of those had to be taken off the market because of serious side effects. Then between 2012 and 2014, the US Food and Drug Administration approved 4 new medications, giving us a total of 6 weight-loss drugs. Those approvals greatly increased the available drug treatments, giving us much-needed options beyond lifestyle and behavioral modifications.

Although it is widely accepted that antiobesity drugs are underused, the study by Thomas et al was the first to quantify the extent of underuse, especially for the newer chronic weight-loss drugs.² Their data show that only about 19% of antiobesity prescriptions were for the newer drugs while 74% were for the older but short-term medication phentermine.

Bersoux et al seem to encourage primary care physicians, or anyone caring for overweight or obese patients, to consider prescribing these treatments if nonpharmacologic options are unsuccessful. I agree with this concept because there are not enough specialists to care for the more than 116 million individuals who are potential candidates for antiobesity medications.

THE TIME HAS COME

This new class of medications has been strongly endorsed by the most prestigious organizations and societies involved in developing treatment guidelines for the overweight or obese patient. It is time for everyone who sees overweight or obese patients in daily practice to consider adopting chronic weight-loss medications as adjunctive therapy if lifestyle and behavioral strategies are ineffective.

The article in this issue by Bersoux et al on pharmacotherapy to manage obesity¹ is apropos in light of a recent study² showing that patients are filling 15 times more prescriptions for antidiabetic medications (excluding insulin) than for antiobesity drugs. What makes this finding significant is that nearly 3 times more adults meet the criteria for use of antiobesity drugs than for antidiabetic drugs—116 million vs 30 million, respectively.

See related article

This underuse of antiobesity medications has been noted in other studies. In 1 study,³ only about 2% of adults eligible for weight-loss drug therapy received a prescription. Conversely, about 86% of adults diagnosed with diabetes received antidiabetic medications.³

WEIGHT LOSS: IT'S IMPORTANT

This underuse of weight-loss drugs occurs despite our understanding that obesity is a risk factor for developing diabetes and that weight loss in obese patients reduces the risk.

The landmark Diabetes Prevention Program study found that even modest weight loss of 7% reduced the risk of developing diabetes by 58% in overweight and prediabetic individuals.⁴ Additionally, a 5% to 10% weight loss can lead to significant improvements in many comorbidities, including diabetes, hyperlipidemia, hypertension, sleep apnea, and fatty liver disease.

Antiobesity medications can help patients achieve weight-loss goals, especially if lifestyle and behavioral modifications alone have been unsuccessful. Data show that these drugs result in an average weight loss of 5% to 15% when added to diet and exercise.

BARRIERS TO PRESCRIBING WEIGHT-LOSS DRUGS

Why are practitioners reluctant to prescribe these drugs despite the worsening obesity epidemic and despite knowing that obesity is a risk factor for diabetes? Many of us who practice obesity medicine believe there are several reasons.

One barrier is the misconception that obesity does not warrant treatment with weight-loss medications, even though most practitioners will readily admit that patients cannot achieve effective, durable, and meaningful weight loss with behavioral changes and lifestyle modifications alone.

Other barriers stem from issues such as time constraints in the office, lack of training to treat this condition, and not enough data on the newer chronic weight-loss medications. And there are stringent requirements for patient follow-up once a medication has been initiated. Finally, it’s often difficult to obtain insurance coverage.

Addressing the barriers

Of these, I believe the biggest barrier for busy practitioners is finding the time and effort they need to devote to prescribing weight-loss medications. There are ways to address these issues.

Regarding time constraints, practitioners can discuss weight loss at follow-up visits and refer patients to obesity specialists. Regarding gaps in training and knowledge of obesity management, there are consensus guidelines for the identification, evaluation, and treatment of the overweight or obese individual.^5–7 Guidelines provide extensive information on the pharmacologic treatment of obesity. These resources provide valuable evidence-based recommendations on how to manage this chronic disease.

ARMED WITH INFORMATION, PHARMACOLOGIC OPTIONS

Bersoux et al provide another valuable resource for clinical use of weight-loss drugs.¹ They accurately review the available medications, their mechanisms of action, dosing, efficacy, side effect profiles, and clinical indications. Their review is comprehensive in every aspect of this drug class.

This is important information for practitioners to have when considering prescribing antiobesity medications. It is especially important for primary care practitioners because of the large number of obese or overweight patients they treat.

Drug options have expanded

We did not always have this many drugs to choose from. As Bersoux et al note, practitioners had limited options for weight-loss medications during the 1990s and early 2000s, and several of those had to be taken off the market because of serious side effects. Then between 2012 and 2014, the US Food and Drug Administration approved 4 new medications, giving us a total of 6 weight-loss drugs. Those approvals greatly increased the available drug treatments, giving us much-needed options beyond lifestyle and behavioral modifications.

Although it is widely accepted that antiobesity drugs are underused, the study by Thomas et al was the first to quantify the extent of underuse, especially for the newer chronic weight-loss drugs.² Their data show that only about 19% of antiobesity prescriptions were for the newer drugs while 74% were for the older but short-term medication phentermine.

Bersoux et al seem to encourage primary care physicians, or anyone caring for overweight or obese patients, to consider prescribing these treatments if nonpharmacologic options are unsuccessful. I agree with this concept because there are not enough specialists to care for the more than 116 million individuals who are potential candidates for antiobesity medications.

THE TIME HAS COME

This new class of medications has been strongly endorsed by the most prestigious organizations and societies involved in developing treatment guidelines for the overweight or obese patient. It is time for everyone who sees overweight or obese patients in daily practice to consider adopting chronic weight-loss medications as adjunctive therapy if lifestyle and behavioral strategies are ineffective.

The article in this issue by Bersoux et al on pharmacotherapy to manage obesity¹ is apropos in light of a recent study² showing that patients are filling 15 times more prescriptions for antidiabetic medications (excluding insulin) than for antiobesity drugs. What makes this finding significant is that nearly 3 times more adults meet the criteria for use of antiobesity drugs than for antidiabetic drugs—116 million vs 30 million, respectively.

See related article

This underuse of antiobesity medications has been noted in other studies. In 1 study,³ only about 2% of adults eligible for weight-loss drug therapy received a prescription. Conversely, about 86% of adults diagnosed with diabetes received antidiabetic medications.³

WEIGHT LOSS: IT'S IMPORTANT

This underuse of weight-loss drugs occurs despite our understanding that obesity is a risk factor for developing diabetes and that weight loss in obese patients reduces the risk.

The landmark Diabetes Prevention Program study found that even modest weight loss of 7% reduced the risk of developing diabetes by 58% in overweight and prediabetic individuals.⁴ Additionally, a 5% to 10% weight loss can lead to significant improvements in many comorbidities, including diabetes, hyperlipidemia, hypertension, sleep apnea, and fatty liver disease.

Antiobesity medications can help patients achieve weight-loss goals, especially if lifestyle and behavioral modifications alone have been unsuccessful. Data show that these drugs result in an average weight loss of 5% to 15% when added to diet and exercise.

BARRIERS TO PRESCRIBING WEIGHT-LOSS DRUGS

Why are practitioners reluctant to prescribe these drugs despite the worsening obesity epidemic and despite knowing that obesity is a risk factor for diabetes? Many of us who practice obesity medicine believe there are several reasons.

One barrier is the misconception that obesity does not warrant treatment with weight-loss medications, even though most practitioners will readily admit that patients cannot achieve effective, durable, and meaningful weight loss with behavioral changes and lifestyle modifications alone.

Other barriers stem from issues such as time constraints in the office, lack of training to treat this condition, and not enough data on the newer chronic weight-loss medications. And there are stringent requirements for patient follow-up once a medication has been initiated. Finally, it’s often difficult to obtain insurance coverage.

Addressing the barriers

Of these, I believe the biggest barrier for busy practitioners is finding the time and effort they need to devote to prescribing weight-loss medications. There are ways to address these issues.

Regarding time constraints, practitioners can discuss weight loss at follow-up visits and refer patients to obesity specialists. Regarding gaps in training and knowledge of obesity management, there are consensus guidelines for the identification, evaluation, and treatment of the overweight or obese individual.^5–7 Guidelines provide extensive information on the pharmacologic treatment of obesity. These resources provide valuable evidence-based recommendations on how to manage this chronic disease.

ARMED WITH INFORMATION, PHARMACOLOGIC OPTIONS

Bersoux et al provide another valuable resource for clinical use of weight-loss drugs.¹ They accurately review the available medications, their mechanisms of action, dosing, efficacy, side effect profiles, and clinical indications. Their review is comprehensive in every aspect of this drug class.

This is important information for practitioners to have when considering prescribing antiobesity medications. It is especially important for primary care practitioners because of the large number of obese or overweight patients they treat.

Drug options have expanded

We did not always have this many drugs to choose from. As Bersoux et al note, practitioners had limited options for weight-loss medications during the 1990s and early 2000s, and several of those had to be taken off the market because of serious side effects. Then between 2012 and 2014, the US Food and Drug Administration approved 4 new medications, giving us a total of 6 weight-loss drugs. Those approvals greatly increased the available drug treatments, giving us much-needed options beyond lifestyle and behavioral modifications.

Although it is widely accepted that antiobesity drugs are underused, the study by Thomas et al was the first to quantify the extent of underuse, especially for the newer chronic weight-loss drugs.² Their data show that only about 19% of antiobesity prescriptions were for the newer drugs while 74% were for the older but short-term medication phentermine.

Bersoux et al seem to encourage primary care physicians, or anyone caring for overweight or obese patients, to consider prescribing these treatments if nonpharmacologic options are unsuccessful. I agree with this concept because there are not enough specialists to care for the more than 116 million individuals who are potential candidates for antiobesity medications.

THE TIME HAS COME

This new class of medications has been strongly endorsed by the most prestigious organizations and societies involved in developing treatment guidelines for the overweight or obese patient. It is time for everyone who sees overweight or obese patients in daily practice to consider adopting chronic weight-loss medications as adjunctive therapy if lifestyle and behavioral strategies are ineffective.

The successful interplay between the host defense system and infectious invaders depends on controlling the tissue damage that ensues from both the infection and the resultant inflammatory response. Even though an underactive immune system predisposes to unusual and potentially severe infections, an overly vigorous host response to infection can be as destructive as the infection itself. We can improve the outcome of some infections by introducing potent anti-inflammatory and immunosuppressive therapy concurrent with appropriate anti-infective therapy. What initially seemed counterintuitive has become the standard of care in the treatment of bacterial and mycobacterial meningitis and severe Pneumocystis and bacterial pneumonias, and favorable data are accruing in other infections such as bacterial arthritis.

A twist on the above scenario can occur when an immunosuppressed patient with a partially controlled indolent infection has his or her immune system suddenly normalized due to successful treatment of the underlying cause of their immunodeficiency. This treatment may be the introduction of successful antiretroviral therapy against human immunodeficiency virus (HIV), effective therapy of an immunosuppressing infection like tuberculosis, or withdrawal of an immunosuppressive anti-tumor necrosis factor (anti-TNF) drug. In this scenario, where the immune system is rapidly reconstituted and concurrently activated by the presence of persistent antigenic challenge or immunostimulatory molecules, a vigorous and clinically counterproductive inflammatory response may ensue, causing “collateral damage” to normal tissue. This immune reactivation syndrome may include fever, sweats, adenitis, and local tissue destruction at the site of infectious agents and associated phlogistic breakdown products. The result of this robust, tissue-injurious inflammatory response can be particularly devastating if it occurs in the brain or the retina, and may cause diagnostic confusion.

The trigger for this regional and systemic inflammatory response is multifactorial. It includes the newly recovered responsiveness to high levels of circulating cytokines, reaction to immune-stimulating fatty acids and other molecules released from dying mycobacteria (perhaps akin to the Jarisch-Herxheimer reaction to rapidly dying spirochetes), and possibly an over-vigorous “rebooting” immune system if an appropriate regulatory cell network is yet to be reconstituted.

In this issue of the Journal, Hara et al provide images from a patient appropriately treated for tuberculosis who experienced continued systemic symptoms of infection with the appearance of new pulmonary lesions. The trigger was the withdrawal of the infliximab (anti-TNF) therapy he was taking for ulcerative colitis, which at face value might be expected to facilitate the successful treatment of his tuberculosis. This seemingly paradoxical reaction has been well described with the successful treatment of HIV-infected patients coinfected with mycobacteria (tuberculous or nontuberculous), cytomegalovirus, and herpes-associated Kaposi sarcoma and zoster. But as in this instructive description of a patient with an immune reactivation syndrome, it also occurs in the setting of non-HIV reversibly immunosuppressed patients.^1,2 The syndrome is often recognized 1 to 2 months after immune reconstitution and the initiation of anti-infective therapy.

The treatment of this paradoxical reaction is (not so paradoxically) the administration of corticosteroids or other immunosuppressive drugs. The efficacy of corticosteroids has been demonstrated in a small placebo-controlled trial³ as well as in clinical practice. The mechanism driving this reaction may not be the same for all infections, and thus steroids may not be ideal treatment for all patients. There are reports of using infliximab to temper the immune reactivation syndrome in some patients who did not respond to corticosteroids.

There is no definitive confirmatory test for immune reactivation syndrome. And certainly in the case of known mycobacterial infection, we must ensure the absence of drug resistance and that the appropriate antibiotics are being used, and that no additional infection is present and untreated by the antimycobacterial therapy. While lymphocytosis and an overly robust tuberculin skin test response have been described in patients with tuberculosis experiencing an immune reactivation syndrome, this “paradoxical reaction” remains a clinical diagnosis, worth considering in the appropriate setting.

The successful interplay between the host defense system and infectious invaders depends on controlling the tissue damage that ensues from both the infection and the resultant inflammatory response. Even though an underactive immune system predisposes to unusual and potentially severe infections, an overly vigorous host response to infection can be as destructive as the infection itself. We can improve the outcome of some infections by introducing potent anti-inflammatory and immunosuppressive therapy concurrent with appropriate anti-infective therapy. What initially seemed counterintuitive has become the standard of care in the treatment of bacterial and mycobacterial meningitis and severe Pneumocystis and bacterial pneumonias, and favorable data are accruing in other infections such as bacterial arthritis.

A twist on the above scenario can occur when an immunosuppressed patient with a partially controlled indolent infection has his or her immune system suddenly normalized due to successful treatment of the underlying cause of their immunodeficiency. This treatment may be the introduction of successful antiretroviral therapy against human immunodeficiency virus (HIV), effective therapy of an immunosuppressing infection like tuberculosis, or withdrawal of an immunosuppressive anti-tumor necrosis factor (anti-TNF) drug. In this scenario, where the immune system is rapidly reconstituted and concurrently activated by the presence of persistent antigenic challenge or immunostimulatory molecules, a vigorous and clinically counterproductive inflammatory response may ensue, causing “collateral damage” to normal tissue. This immune reactivation syndrome may include fever, sweats, adenitis, and local tissue destruction at the site of infectious agents and associated phlogistic breakdown products. The result of this robust, tissue-injurious inflammatory response can be particularly devastating if it occurs in the brain or the retina, and may cause diagnostic confusion.

The trigger for this regional and systemic inflammatory response is multifactorial. It includes the newly recovered responsiveness to high levels of circulating cytokines, reaction to immune-stimulating fatty acids and other molecules released from dying mycobacteria (perhaps akin to the Jarisch-Herxheimer reaction to rapidly dying spirochetes), and possibly an over-vigorous “rebooting” immune system if an appropriate regulatory cell network is yet to be reconstituted.

In this issue of the Journal, Hara et al provide images from a patient appropriately treated for tuberculosis who experienced continued systemic symptoms of infection with the appearance of new pulmonary lesions. The trigger was the withdrawal of the infliximab (anti-TNF) therapy he was taking for ulcerative colitis, which at face value might be expected to facilitate the successful treatment of his tuberculosis. This seemingly paradoxical reaction has been well described with the successful treatment of HIV-infected patients coinfected with mycobacteria (tuberculous or nontuberculous), cytomegalovirus, and herpes-associated Kaposi sarcoma and zoster. But as in this instructive description of a patient with an immune reactivation syndrome, it also occurs in the setting of non-HIV reversibly immunosuppressed patients.^1,2 The syndrome is often recognized 1 to 2 months after immune reconstitution and the initiation of anti-infective therapy.

The treatment of this paradoxical reaction is (not so paradoxically) the administration of corticosteroids or other immunosuppressive drugs. The efficacy of corticosteroids has been demonstrated in a small placebo-controlled trial³ as well as in clinical practice. The mechanism driving this reaction may not be the same for all infections, and thus steroids may not be ideal treatment for all patients. There are reports of using infliximab to temper the immune reactivation syndrome in some patients who did not respond to corticosteroids.

There is no definitive confirmatory test for immune reactivation syndrome. And certainly in the case of known mycobacterial infection, we must ensure the absence of drug resistance and that the appropriate antibiotics are being used, and that no additional infection is present and untreated by the antimycobacterial therapy. While lymphocytosis and an overly robust tuberculin skin test response have been described in patients with tuberculosis experiencing an immune reactivation syndrome, this “paradoxical reaction” remains a clinical diagnosis, worth considering in the appropriate setting.

The successful interplay between the host defense system and infectious invaders depends on controlling the tissue damage that ensues from both the infection and the resultant inflammatory response. Even though an underactive immune system predisposes to unusual and potentially severe infections, an overly vigorous host response to infection can be as destructive as the infection itself. We can improve the outcome of some infections by introducing potent anti-inflammatory and immunosuppressive therapy concurrent with appropriate anti-infective therapy. What initially seemed counterintuitive has become the standard of care in the treatment of bacterial and mycobacterial meningitis and severe Pneumocystis and bacterial pneumonias, and favorable data are accruing in other infections such as bacterial arthritis.

A twist on the above scenario can occur when an immunosuppressed patient with a partially controlled indolent infection has his or her immune system suddenly normalized due to successful treatment of the underlying cause of their immunodeficiency. This treatment may be the introduction of successful antiretroviral therapy against human immunodeficiency virus (HIV), effective therapy of an immunosuppressing infection like tuberculosis, or withdrawal of an immunosuppressive anti-tumor necrosis factor (anti-TNF) drug. In this scenario, where the immune system is rapidly reconstituted and concurrently activated by the presence of persistent antigenic challenge or immunostimulatory molecules, a vigorous and clinically counterproductive inflammatory response may ensue, causing “collateral damage” to normal tissue. This immune reactivation syndrome may include fever, sweats, adenitis, and local tissue destruction at the site of infectious agents and associated phlogistic breakdown products. The result of this robust, tissue-injurious inflammatory response can be particularly devastating if it occurs in the brain or the retina, and may cause diagnostic confusion.

The trigger for this regional and systemic inflammatory response is multifactorial. It includes the newly recovered responsiveness to high levels of circulating cytokines, reaction to immune-stimulating fatty acids and other molecules released from dying mycobacteria (perhaps akin to the Jarisch-Herxheimer reaction to rapidly dying spirochetes), and possibly an over-vigorous “rebooting” immune system if an appropriate regulatory cell network is yet to be reconstituted.

In this issue of the Journal, Hara et al provide images from a patient appropriately treated for tuberculosis who experienced continued systemic symptoms of infection with the appearance of new pulmonary lesions. The trigger was the withdrawal of the infliximab (anti-TNF) therapy he was taking for ulcerative colitis, which at face value might be expected to facilitate the successful treatment of his tuberculosis. This seemingly paradoxical reaction has been well described with the successful treatment of HIV-infected patients coinfected with mycobacteria (tuberculous or nontuberculous), cytomegalovirus, and herpes-associated Kaposi sarcoma and zoster. But as in this instructive description of a patient with an immune reactivation syndrome, it also occurs in the setting of non-HIV reversibly immunosuppressed patients.^1,2 The syndrome is often recognized 1 to 2 months after immune reconstitution and the initiation of anti-infective therapy.

The treatment of this paradoxical reaction is (not so paradoxically) the administration of corticosteroids or other immunosuppressive drugs. The efficacy of corticosteroids has been demonstrated in a small placebo-controlled trial³ as well as in clinical practice. The mechanism driving this reaction may not be the same for all infections, and thus steroids may not be ideal treatment for all patients. There are reports of using infliximab to temper the immune reactivation syndrome in some patients who did not respond to corticosteroids.

There is no definitive confirmatory test for immune reactivation syndrome. And certainly in the case of known mycobacterial infection, we must ensure the absence of drug resistance and that the appropriate antibiotics are being used, and that no additional infection is present and untreated by the antimycobacterial therapy. While lymphocytosis and an overly robust tuberculin skin test response have been described in patients with tuberculosis experiencing an immune reactivation syndrome, this “paradoxical reaction” remains a clinical diagnosis, worth considering in the appropriate setting.

Weight-loss drugs are not magic pills, but they can help patients lose about 10 to 25 more pounds than they otherwise could, when used in a program that includes diet, exercise, and other lifestyle changes.

See related editorial

HALF OF ADULTS MAY BE OBESE BY 2030

Obesity is a major public health challenge in the United States, with nearly 37% of adults classified as obese.¹ The prevalence has increased more than 75% since 1980,² and it is estimated that 51% of US adults will be obese by 2030.³ Obesity is the second-leading cause of preventable deaths, after smoking.⁴

Obesity increases the risk of many chronic medical conditions, including type 2 diabetes mellitus, heart disease, hypertension, stroke, nonalcoholic fatty liver disease, osteoarthritis, and cancers of the breast, colon, endometrium, and kidney.⁵

WHEN IS DRUG THERAPY INDICATED?

Guidelines from the major obesity societies recommend that all weight-loss programs have a lifestyle component that includes a low-calorie diet, increased physical activity, and behavioral therapy, to which pharmacotherapy may be added as an adjunct.^6–8

Weight-loss medications are indicated for patients who have a body mass index (BMI) of at least 30 kg/m² or who have obesity-associated comorbidities and a BMI of at least 27 kg/m². However, the best results are achieved when pharmacotherapy is combined with lifestyle modification.⁹

HISTORY OF WEIGHT-LOSS DRUGS: NOT A PRETTY PICTURE

The earliest drugs to induce weight loss, which worked mainly by increasing metabolism, included thyroid hormone, amphetamines (which also suppress appetite), and dinitrophenol (a pesticide). Adverse reactions limited their usefulness: cardiovascular effects with thyroid hormones, abuse potential with amphetamines, and neuropathy and cataracts with dinitrophenol.

Researchers then looked to drugs that could suppress appetite like amphetamines do, but without the potential for abuse. Medications that increased levels of norepinephrine and serotonin, both by increasing release and decreasing reuptake of these neuromodulators, had some success. But again, serious adverse effects occurred, and several drugs had to be withdrawn from the market.

The most publicized of these withdrawals was for the combination fenfluramine and phentermine (“fen-phen”) and its cousin dexfenfluramine (Redux). Up to 30% of patients taking fenfluramine-phentermine developed echocardiographic evidence of valvular heart disease.¹¹ Fenfluramine also increased the risk of pulmonary hypertension. These findings led to the 1997 withdrawal of these drugs from the US market.

Sibutramine (Meridia), a norepinephrine and serotonin reuptake inhibitor, was approved for weight loss in 1997. Increases in blood pressure and heart rate were noted in the initial trial,¹² and then a postmarketing study found increased rates of nonfatal myocardial infarction and stroke in patients with preexisting cardiovascular disease or diabetes mellitus.¹³ Based on these results, sibutramine was withdrawn from both US and European markets.

Rimonabant (Acomplia, Zimulti), a cannabinoid-receptor inhibitor, was approved in Europe in 2006, but its approval was withdrawn just 2 years later because of increased suicidality in a postmarketing study.¹⁴ It was never approved for use in the United States.

NORADRENERGIC SYMPATHOMIMETICS: FOR SHORT-TERM USE

Several noradrenergic sympathomimetic drugs are FDA-approved for short-term weight loss, but phentermine is by far the most commonly prescribed drug in this class. In fact, it is the most commonly prescribed drug for obesity in the United States.¹⁵

Phentermine

Phentermine is an atypical amphetamine analogue that suppresses appetite by norepinephrine agonism in the central nervous system. The FDA approved it for short-term weight management in 1959, and its use became widespread in the 1960s, followed by decades of popularity.

Dosage. Phentermine is prescribed at an oral dose of 15, 30, or 37.5 mg daily, either before breakfast or 1 to 2 hours after. It is a schedule IV controlled substance, based on its similarity to amphetamine. (The 5 US controlled substance schedules range from schedule I, which includes heroin, amphetamine, and cannabis, to schedule V, which includes cough syrups containing no more than 200 mg of codeine per 100 mL.) However, concerns about addiction and dependence with phentermine are largely unfounded, and abrupt cessation of the drug has not been shown to cause amphetamine-like withdrawal.¹⁶

Adverse effects. Common adverse reactions include nervousness, insomnia, and dry mouth, but these effects tend to wane with continued use.

Contraindications. Cardiovascular disease is a contraindication to phentermine because of concerns about increased blood pressure and pulse rate, although these concerns seem to be more theoretic than observed.¹⁶ Other contraindications include hyperthyroidism, glaucoma, agitation, a history of drug abuse, pregnancy, breastfeeding, and current or recent use of a monoamine oxidase inhibitor. No serious adverse events have been reported in trials of phentermine.

Efficacy. In a pooled analysis of 6 trials lasting 2 to 24 weeks completed between 1975 and 1999, phentermine-treated patients lost an average of 3.6 kg more weight than placebo recipients.¹⁷ More than 80% of study participants were women.

In a 36-week study in 108 women,¹⁸ participants lost a mean of 12.2 kg with continuous phentermine use, 13.0 kg with intermittent use (4 weeks on, 4 weeks off; the difference was not significant), and 4.8 kg with placebo.

Minimal data exist on long-term efficacy of phentermine monotherapy.

DRUGS FOR LONG-TERM THERAPY

Orlistat

Orlistat was approved as a prescription drug (Xenical, 120 mg) in 1999 and as an over-the-counter medication (Alli, 60 mg) in 2007.

Orlistat works by inhibiting pancreatic and gastric lipase, causing incomplete hydrol­ysis of ingested fat, thereby increasing fecal fat excretion in a dose-dependent manner. It is a good choice for weight-loss drug therapy because of its safe cardiovascular risk profile and beneficial effects on lipid levels. However, its long-term effect on weight is only modest.^19,20

Dosage. The dosage for prescription orlistat is 120 mg 3 times per day, in addition to a low-fat diet (< 30% of daily calories from fat). To prevent potential deficiencies of fat-soluble vitamins, a daily multivitamin supplement is recommended, but it should not be taken with meals.

Efficacy. In a 2014 systematic review, 35% to 73% of patients treated with orlistat 120 mg had lost at least 5% of their body weight at 1 year, and 14% to 41% had lost at least 10%.²¹ At the end of the second year, orlistat-treated patients had lost about 3.3 kg more than placebo recipients.

In a randomized trial,²² 4 years of treatment with orlistat vs placebo led to a significant (37.3%) risk reduction in the incidence of type 2 diabetes mellitus in obese participants, as well as significant improvements in cardiovascular risk factors. Mean weight loss at 1 year was significantly greater with orlistat than with placebo (10.6 vs 6.2 kg), and it remained greater at 4 years (5.8 vs 3.0 kg; P < .001).

Adverse effects. Long-term orlistat use is hampered by adverse reactions. A population-based, retrospective cohort analysis showed that fewer than 10% of patients were still using it at 1 year, and only 2% were using it at 2 years, although reasons for discontinuation were not reported.²³

Adverse reactions are predominantly gastrointestinal, attributed to the high content of undigested fat in stools. Patients who do not limit their dietary fat intake are affected the most. Other reported adverse reactions include hepatotoxicity and oxalate-induced nephropathy.

Orlistat has been reported to interfere with some drugs, particularly those that are lipophilic. Drugs that should be closely monitored with orlistat are warfarin, amiodarone, cyclosporine, certain antiepileptic drugs, and levothyroxine.

Phentermine-topiramate

The combination of phentermine and topiramate was approved by the FDA in 2012 and is available under the brand name Qsymia.

Topiramate had been approved for treating seizure disorder in 1996 and as migraine prophylaxis in 2004. It is not approved as monotherapy for obesity; however, patients taking it for seizures or for psychiatric disorders (eg, binge eating, borderline personality disorder) have reported weight loss during treatment.

How topiramate promotes weight loss is not known. Proposed mechanisms include taste inhibition by carbonic anhydrase, influences on gamma-aminobutyric acid transmission causing appetite suppression, sensitization of insulin activity, and adiponectin secretion in the peripheral tissues.^24,25

Phentermine-topiramate therapy has an advantage over monotherapy because lower doses of each medication can be used to achieve the same benefit, thus avoiding dose-related adverse reactions.

Dosage. Phentermine-topiramate is available in capsules containing 3.75/23, 7.5/46, 11.25/69, and 15/92 mg. The recommended starting dosage is 3.75/23 mg/day for 14 days, increasing to 7.5/46 mg/day. If patients do not lose at least 3% of their body weight after 12 weeks, the dose can be increased to 11.25/69 mg daily for 14 days, followed by 15/92 mg daily.²⁶ Phentermine-topiramate is a schedule IV controlled substance with a low potential for abuse and dependence.

Efficacy. Approval of phentermine-topiramate for treating obesity was primarily based on 3 clinical trials.^27–29 In 1 of these trials,²⁸ at 1 year, patients had lost 9.9 kg with the medium dose and 12.9 kg with the high dose.

Adverse effects. Phentermine-topiramate was well tolerated in the trials. The most commonly reported adverse reactions were dry mouth, dizziness, constipation, insomnia, dysgeusia, paresthesia, and increased resting heart rate.^28,29 Acute myopia and angle-closure glaucoma also have been reported with topiramate.³⁰ Topiramate monotherapy has been associated with dose-dependent neuropsychiatric adverse effects, including memory symptoms and depression. However, across all 3 trials of phentermine-topiramate therapy, symptoms of depression improved over time, and no significant increase in suicide risk was identified.^27–29

Recommended monitoring for patients on phentermine-topiramate includes a blood chemistry panel, resting heart rate, blood pressure, and depression screening.

Because topiramate has teratogenic potential (craniofacial abnormalities), it is labeled as pregnancy category X (contraindicated). A negative pregnancy test is needed before women of childbearing age take the drug and monthly thereafter. Women should be counseled to use effective birth control. A home pregnancy test is an alternative to laboratory testing, but this option should be left to the prescribing clinician’s judgment and be based on reliability of the test and patient compliance.

Lorcaserin

Lorcaserin (Belviq) was approved by the FDA in 2012 for chronic weight management. It suppresses appetite by activating the serotonin 2C receptor in the brain. Because it is selective for the 2C receptor, it does not appear to have the same detrimental effects on heart valves as occurred with less-selective serotonergic agents such as fenfluramine and dexfenfluramine.³¹

Dosage. The recommended dosage for lorcaserin is 10 mg twice daily. Lorcaserin is a schedule IV controlled substance because of studies that showed increases in positive subjective measures such as euphoria in patients taking the drug. The incidence of euphoria was similar to that seen with zolpidem.³²

Efficacy. Lorcaserin was approved on the basis of 2 trials in nondiabetic obese and overweight adults who did not have diabetes but who had a weight-related condition,^33,34 and in a third trial in obese and overweight adults with type 2 diabetes mellitus who were taking oral hypoglycemic agents.³⁵ In these trials, lorcaserin use resulted in a modest 4.7- to 5.8-kg weight loss compared with 1.6 to 2.2 kg in the placebo group.^33–35 There was a high dropout rate in all 3 of these studies (33% to 45% of participants).

A pilot study that added phentermine to lorcaserin yielded double the weight loss from lorcaserin alone.³⁶ This drug combination warrants further investigation.

Contraindications. Lorcaserin should not be given to patients who have severe renal insufficiency (creatinine clearance < 30 mL/min) or severe hepatic impairment, or who are pregnant.

Adverse effects. Common adverse reactions include dry mouth, dizziness, somnolence, headache, and gastrointestinal disturbances (nausea, constipation, or diarrhea).³⁷

Patients with type 2 diabetes mellitus should be monitored for hypoglycemia.

Lorcaserin should be used with extreme caution in patients taking other serotonergic agents because of the risk of the serotonin syndrome.

A theoretic potential for increased risk of breast cancer also exists with lorcaserin. When rats were given supraphysiologic doses of lorcaserin (more than 50 times higher than recommended in humans), fibroadenomas and adenocarcinomas occurred at higher rates.³⁸ Breast cancer data were not reported in the 3 randomized trials discussed above.^33–35

Naltrexone-bupropion

The combination of naltrexone and bupropion was approved by the FDA in 2014 under the brand name Contrave. Both drugs are approved for monotherapy in conditions other than obesity.

Naltrexone is a mu opioid receptor antagonist approved to treat alcohol and opioid dependency. Bupropion is a dopamine-norepinephrine reuptake inhibitor approved to treat depression and to help with smoking cessation. Combining the drugs produces weight loss and metabolic benefits through effects on 2 areas of the brain that regulate food intake: the hypothalamus (appetite) and the mesolimbic dopamine circuit (reward system).

Dosage. Naltrexone-bupropion comes as an extended-release tablet of 8/90 mg. The maintenance dose of 2 tablets twice daily is reached at week 4 through a specific dose-titration regimen (Table 1). The dose should be adjusted if patients have renal or hepatic impairment or if they are also taking a CYP2B6 inhibitor.

Efficacy. FDA approval was based on the results of 4 clinical trials.^39–42 Using a modified intention-to-treat analysis, Yanovski and Yanovski⁴³ calculated that at 1 year, placebo-subtracted mean weight loss was 4.6% (4.9 kg), and mean total weight loss was 6.8% (7.3 kg) across the studies. Attrition rates, however, were high, ranging from 42% to 50%.

Cardiometabolic effects in 2 of the trials^40,41 included decreased waist circumference, triglyceride levels, and C-reactive protein levels, and increased high-density lipoprotein levels at the initial dose. At the maintenance dose, additional lowering of fasting plasma insulin and glucose levels occurred along with lower levels of the homeostatic model assessment of insulin resistance. In the COR-Diabetes Study Group trial, patients with type 2 diabetes mellitus had decreased hemoglobin A_1c levels without an increase in hypoglycemia and an increased likelihood of reaching the target hemoglobin A_1c level below 7%.³⁹

Contraindications. Naltrexone-bupropion is contraindicated for patients who have uncontrolled hypertension, seizure disorder, eating disorder, or end-stage renal failure; who are pregnant; or who have been treated with a monoamine oxidase inhibitor within 14 days. It should not be used with other bupropion-containing products or in patients who have taken opioids chronically or have acute opiate withdrawal.

Because of its bupropion component, this product carries an FDA black-box warning about possible suicidal thoughts and behaviors and neuropsychiatric reactions.

Adverse effects. The adverse reactions most commonly associated with naltrexone-bupropion were nausea (32.5%), constipation (19.2%), headache (17.6%), vomiting (10.7%), dizziness (9.9%), insomnia (9.2%), dry mouth (8.1%), and diarrhea (7.1%).⁴⁴

Liraglutide

Liraglutide, previously FDA-approved to treat type 2 diabetes mellitus under the brand name Victoza, received approval in 2014 in a higher-dose formulation (Saxenda) to treat obesity.

Liraglutide is a glucagon-like peptide-1 receptor agonist that stimulates glucose-dependent insulin release from the pancreatic islet cells, slows gastric emptying, regulates postprandial glucagon, and reduces food intake.

Dosage. Liraglutide is given as a once-daily injection in the abdomen, thigh, or arm. The initial dosage is 0.6 mg daily for the first week and can be titrated up by 0.6 mg weekly to a target dose of 3 mg daily. If a patient does not lose 4% of baseline body weight after 16 weeks on the target dose, the drug should be discontinued because it is unlikely to lead to clinically significant weight loss.

Efficacy. Liraglutide for weight management (3 mg once daily) was evaluated in a large (N = 3,731), randomized, double-blind, placebo-controlled international trial.⁴⁵ Participants did not have diabetes mellitus, but 60% had prediabetes. Liraglutide or placebo was given for 56 weeks, along with lifestyle counseling. At the end of the study, the liraglutide group had lost a mean of 8.4 kg vs 2.8 kg in the placebo group. Additionally, 63% of the liraglutide group lost at least 5% of body weight vs 27% in the placebo group, and 33% lost at least 10% of body weight vs 10% in the placebo group.

A 2-year extension found systolic blood pressure decreased with no change in pulse, and the prevalence of prediabetes and metabolic syndrome decreased by 52% and 59%, respectively.⁴⁶ At 2 years, mean scores for physical function, self-esteem, and work had improved more in the liraglutide group than the placebo group.⁴⁷

Adverse effects. The most common adverse reactions with liraglutide were nausea, vomiting, diarrhea, constipation, hypoglycemia, and loss of appetite. In most cases, nausea and vomiting were tolerable, transient, and associated with greater weight loss but not with decreased quality-of-life scores. Serious adverse reactions included pancreatitis, gallbladder disease, renal impairment, and suicidal thoughts.

For obese patients, when lifestyle modifications do not result in the desired weight loss, pharmacotherapy is an option. Practitioners have several FDA-approved options for weight management. Because of evidence that these drugs can postpone the onset of other complications and improve metabolic and cardiovascular parameters, they should be considered.

In phase 3 trials, these drugs caused modest weight loss of 5% to 10% of body weight. More weight was lost with the combination of phentermine-topiramate than with the other drugs.

In a 2016 meta-analysis, these drugs were associated with at least 5% weight reduction compared with placebo.⁴⁸ Phentermine-topiramate and liraglutide were most likely to produce at least a 5% weight loss, while liraglutide and naltrexone-bupropion were most likely to be discontinued because of adverse events. Combination drugs may have the advantages of synergistic effects on weight loss and fewer adverse reactions because lower doses of the individual drug components are used.

Response to therapy with most of these drugs should be evaluated at 12 weeks on the maintenance dose. If less than 5% weight loss has been achieved, the medication should be discontinued.

Adverse-effect profiles, drug interactions, abuse, misuse, and overdose potential should be considered when prescribing these drugs. Weight-loss drugs are contraindicated in pregnancy because they offer no potential benefit to a pregnant woman and may harm the fetus.

The development of new drugs and better drug combinations is expected to provide more effective therapeutic strategies, which are essential for combating the obesity epidemic.

Weight-loss drugs are not magic pills, but they can help patients lose about 10 to 25 more pounds than they otherwise could, when used in a program that includes diet, exercise, and other lifestyle changes.

See related editorial

HALF OF ADULTS MAY BE OBESE BY 2030

Obesity is a major public health challenge in the United States, with nearly 37% of adults classified as obese.¹ The prevalence has increased more than 75% since 1980,² and it is estimated that 51% of US adults will be obese by 2030.³ Obesity is the second-leading cause of preventable deaths, after smoking.⁴

Obesity increases the risk of many chronic medical conditions, including type 2 diabetes mellitus, heart disease, hypertension, stroke, nonalcoholic fatty liver disease, osteoarthritis, and cancers of the breast, colon, endometrium, and kidney.⁵

WHEN IS DRUG THERAPY INDICATED?

Guidelines from the major obesity societies recommend that all weight-loss programs have a lifestyle component that includes a low-calorie diet, increased physical activity, and behavioral therapy, to which pharmacotherapy may be added as an adjunct.^6–8

Weight-loss medications are indicated for patients who have a body mass index (BMI) of at least 30 kg/m² or who have obesity-associated comorbidities and a BMI of at least 27 kg/m². However, the best results are achieved when pharmacotherapy is combined with lifestyle modification.⁹

HISTORY OF WEIGHT-LOSS DRUGS: NOT A PRETTY PICTURE

The earliest drugs to induce weight loss, which worked mainly by increasing metabolism, included thyroid hormone, amphetamines (which also suppress appetite), and dinitrophenol (a pesticide). Adverse reactions limited their usefulness: cardiovascular effects with thyroid hormones, abuse potential with amphetamines, and neuropathy and cataracts with dinitrophenol.

Researchers then looked to drugs that could suppress appetite like amphetamines do, but without the potential for abuse. Medications that increased levels of norepinephrine and serotonin, both by increasing release and decreasing reuptake of these neuromodulators, had some success. But again, serious adverse effects occurred, and several drugs had to be withdrawn from the market.

The most publicized of these withdrawals was for the combination fenfluramine and phentermine (“fen-phen”) and its cousin dexfenfluramine (Redux). Up to 30% of patients taking fenfluramine-phentermine developed echocardiographic evidence of valvular heart disease.¹¹ Fenfluramine also increased the risk of pulmonary hypertension. These findings led to the 1997 withdrawal of these drugs from the US market.

Sibutramine (Meridia), a norepinephrine and serotonin reuptake inhibitor, was approved for weight loss in 1997. Increases in blood pressure and heart rate were noted in the initial trial,¹² and then a postmarketing study found increased rates of nonfatal myocardial infarction and stroke in patients with preexisting cardiovascular disease or diabetes mellitus.¹³ Based on these results, sibutramine was withdrawn from both US and European markets.

Rimonabant (Acomplia, Zimulti), a cannabinoid-receptor inhibitor, was approved in Europe in 2006, but its approval was withdrawn just 2 years later because of increased suicidality in a postmarketing study.¹⁴ It was never approved for use in the United States.

NORADRENERGIC SYMPATHOMIMETICS: FOR SHORT-TERM USE

Several noradrenergic sympathomimetic drugs are FDA-approved for short-term weight loss, but phentermine is by far the most commonly prescribed drug in this class. In fact, it is the most commonly prescribed drug for obesity in the United States.¹⁵

Phentermine

Phentermine is an atypical amphetamine analogue that suppresses appetite by norepinephrine agonism in the central nervous system. The FDA approved it for short-term weight management in 1959, and its use became widespread in the 1960s, followed by decades of popularity.

Dosage. Phentermine is prescribed at an oral dose of 15, 30, or 37.5 mg daily, either before breakfast or 1 to 2 hours after. It is a schedule IV controlled substance, based on its similarity to amphetamine. (The 5 US controlled substance schedules range from schedule I, which includes heroin, amphetamine, and cannabis, to schedule V, which includes cough syrups containing no more than 200 mg of codeine per 100 mL.) However, concerns about addiction and dependence with phentermine are largely unfounded, and abrupt cessation of the drug has not been shown to cause amphetamine-like withdrawal.¹⁶

Adverse effects. Common adverse reactions include nervousness, insomnia, and dry mouth, but these effects tend to wane with continued use.

Contraindications. Cardiovascular disease is a contraindication to phentermine because of concerns about increased blood pressure and pulse rate, although these concerns seem to be more theoretic than observed.¹⁶ Other contraindications include hyperthyroidism, glaucoma, agitation, a history of drug abuse, pregnancy, breastfeeding, and current or recent use of a monoamine oxidase inhibitor. No serious adverse events have been reported in trials of phentermine.

Efficacy. In a pooled analysis of 6 trials lasting 2 to 24 weeks completed between 1975 and 1999, phentermine-treated patients lost an average of 3.6 kg more weight than placebo recipients.¹⁷ More than 80% of study participants were women.

In a 36-week study in 108 women,¹⁸ participants lost a mean of 12.2 kg with continuous phentermine use, 13.0 kg with intermittent use (4 weeks on, 4 weeks off; the difference was not significant), and 4.8 kg with placebo.

Minimal data exist on long-term efficacy of phentermine monotherapy.

DRUGS FOR LONG-TERM THERAPY

Orlistat

Orlistat was approved as a prescription drug (Xenical, 120 mg) in 1999 and as an over-the-counter medication (Alli, 60 mg) in 2007.

Orlistat works by inhibiting pancreatic and gastric lipase, causing incomplete hydrol­ysis of ingested fat, thereby increasing fecal fat excretion in a dose-dependent manner. It is a good choice for weight-loss drug therapy because of its safe cardiovascular risk profile and beneficial effects on lipid levels. However, its long-term effect on weight is only modest.^19,20

Dosage. The dosage for prescription orlistat is 120 mg 3 times per day, in addition to a low-fat diet (< 30% of daily calories from fat). To prevent potential deficiencies of fat-soluble vitamins, a daily multivitamin supplement is recommended, but it should not be taken with meals.

Efficacy. In a 2014 systematic review, 35% to 73% of patients treated with orlistat 120 mg had lost at least 5% of their body weight at 1 year, and 14% to 41% had lost at least 10%.²¹ At the end of the second year, orlistat-treated patients had lost about 3.3 kg more than placebo recipients.

In a randomized trial,²² 4 years of treatment with orlistat vs placebo led to a significant (37.3%) risk reduction in the incidence of type 2 diabetes mellitus in obese participants, as well as significant improvements in cardiovascular risk factors. Mean weight loss at 1 year was significantly greater with orlistat than with placebo (10.6 vs 6.2 kg), and it remained greater at 4 years (5.8 vs 3.0 kg; P < .001).

Adverse effects. Long-term orlistat use is hampered by adverse reactions. A population-based, retrospective cohort analysis showed that fewer than 10% of patients were still using it at 1 year, and only 2% were using it at 2 years, although reasons for discontinuation were not reported.²³

Adverse reactions are predominantly gastrointestinal, attributed to the high content of undigested fat in stools. Patients who do not limit their dietary fat intake are affected the most. Other reported adverse reactions include hepatotoxicity and oxalate-induced nephropathy.

Orlistat has been reported to interfere with some drugs, particularly those that are lipophilic. Drugs that should be closely monitored with orlistat are warfarin, amiodarone, cyclosporine, certain antiepileptic drugs, and levothyroxine.

Phentermine-topiramate

The combination of phentermine and topiramate was approved by the FDA in 2012 and is available under the brand name Qsymia.

Topiramate had been approved for treating seizure disorder in 1996 and as migraine prophylaxis in 2004. It is not approved as monotherapy for obesity; however, patients taking it for seizures or for psychiatric disorders (eg, binge eating, borderline personality disorder) have reported weight loss during treatment.

How topiramate promotes weight loss is not known. Proposed mechanisms include taste inhibition by carbonic anhydrase, influences on gamma-aminobutyric acid transmission causing appetite suppression, sensitization of insulin activity, and adiponectin secretion in the peripheral tissues.^24,25

Phentermine-topiramate therapy has an advantage over monotherapy because lower doses of each medication can be used to achieve the same benefit, thus avoiding dose-related adverse reactions.

Dosage. Phentermine-topiramate is available in capsules containing 3.75/23, 7.5/46, 11.25/69, and 15/92 mg. The recommended starting dosage is 3.75/23 mg/day for 14 days, increasing to 7.5/46 mg/day. If patients do not lose at least 3% of their body weight after 12 weeks, the dose can be increased to 11.25/69 mg daily for 14 days, followed by 15/92 mg daily.²⁶ Phentermine-topiramate is a schedule IV controlled substance with a low potential for abuse and dependence.

Efficacy. Approval of phentermine-topiramate for treating obesity was primarily based on 3 clinical trials.^27–29 In 1 of these trials,²⁸ at 1 year, patients had lost 9.9 kg with the medium dose and 12.9 kg with the high dose.

Adverse effects. Phentermine-topiramate was well tolerated in the trials. The most commonly reported adverse reactions were dry mouth, dizziness, constipation, insomnia, dysgeusia, paresthesia, and increased resting heart rate.^28,29 Acute myopia and angle-closure glaucoma also have been reported with topiramate.³⁰ Topiramate monotherapy has been associated with dose-dependent neuropsychiatric adverse effects, including memory symptoms and depression. However, across all 3 trials of phentermine-topiramate therapy, symptoms of depression improved over time, and no significant increase in suicide risk was identified.^27–29

Recommended monitoring for patients on phentermine-topiramate includes a blood chemistry panel, resting heart rate, blood pressure, and depression screening.

Because topiramate has teratogenic potential (craniofacial abnormalities), it is labeled as pregnancy category X (contraindicated). A negative pregnancy test is needed before women of childbearing age take the drug and monthly thereafter. Women should be counseled to use effective birth control. A home pregnancy test is an alternative to laboratory testing, but this option should be left to the prescribing clinician’s judgment and be based on reliability of the test and patient compliance.

Lorcaserin

Lorcaserin (Belviq) was approved by the FDA in 2012 for chronic weight management. It suppresses appetite by activating the serotonin 2C receptor in the brain. Because it is selective for the 2C receptor, it does not appear to have the same detrimental effects on heart valves as occurred with less-selective serotonergic agents such as fenfluramine and dexfenfluramine.³¹

Dosage. The recommended dosage for lorcaserin is 10 mg twice daily. Lorcaserin is a schedule IV controlled substance because of studies that showed increases in positive subjective measures such as euphoria in patients taking the drug. The incidence of euphoria was similar to that seen with zolpidem.³²

Efficacy. Lorcaserin was approved on the basis of 2 trials in nondiabetic obese and overweight adults who did not have diabetes but who had a weight-related condition,^33,34 and in a third trial in obese and overweight adults with type 2 diabetes mellitus who were taking oral hypoglycemic agents.³⁵ In these trials, lorcaserin use resulted in a modest 4.7- to 5.8-kg weight loss compared with 1.6 to 2.2 kg in the placebo group.^33–35 There was a high dropout rate in all 3 of these studies (33% to 45% of participants).

A pilot study that added phentermine to lorcaserin yielded double the weight loss from lorcaserin alone.³⁶ This drug combination warrants further investigation.

Contraindications. Lorcaserin should not be given to patients who have severe renal insufficiency (creatinine clearance < 30 mL/min) or severe hepatic impairment, or who are pregnant.

Adverse effects. Common adverse reactions include dry mouth, dizziness, somnolence, headache, and gastrointestinal disturbances (nausea, constipation, or diarrhea).³⁷

Patients with type 2 diabetes mellitus should be monitored for hypoglycemia.

Lorcaserin should be used with extreme caution in patients taking other serotonergic agents because of the risk of the serotonin syndrome.

A theoretic potential for increased risk of breast cancer also exists with lorcaserin. When rats were given supraphysiologic doses of lorcaserin (more than 50 times higher than recommended in humans), fibroadenomas and adenocarcinomas occurred at higher rates.³⁸ Breast cancer data were not reported in the 3 randomized trials discussed above.^33–35

Naltrexone-bupropion

The combination of naltrexone and bupropion was approved by the FDA in 2014 under the brand name Contrave. Both drugs are approved for monotherapy in conditions other than obesity.

Naltrexone is a mu opioid receptor antagonist approved to treat alcohol and opioid dependency. Bupropion is a dopamine-norepinephrine reuptake inhibitor approved to treat depression and to help with smoking cessation. Combining the drugs produces weight loss and metabolic benefits through effects on 2 areas of the brain that regulate food intake: the hypothalamus (appetite) and the mesolimbic dopamine circuit (reward system).

Dosage. Naltrexone-bupropion comes as an extended-release tablet of 8/90 mg. The maintenance dose of 2 tablets twice daily is reached at week 4 through a specific dose-titration regimen (Table 1). The dose should be adjusted if patients have renal or hepatic impairment or if they are also taking a CYP2B6 inhibitor.

Efficacy. FDA approval was based on the results of 4 clinical trials.^39–42 Using a modified intention-to-treat analysis, Yanovski and Yanovski⁴³ calculated that at 1 year, placebo-subtracted mean weight loss was 4.6% (4.9 kg), and mean total weight loss was 6.8% (7.3 kg) across the studies. Attrition rates, however, were high, ranging from 42% to 50%.

Cardiometabolic effects in 2 of the trials^40,41 included decreased waist circumference, triglyceride levels, and C-reactive protein levels, and increased high-density lipoprotein levels at the initial dose. At the maintenance dose, additional lowering of fasting plasma insulin and glucose levels occurred along with lower levels of the homeostatic model assessment of insulin resistance. In the COR-Diabetes Study Group trial, patients with type 2 diabetes mellitus had decreased hemoglobin A_1c levels without an increase in hypoglycemia and an increased likelihood of reaching the target hemoglobin A_1c level below 7%.³⁹

Contraindications. Naltrexone-bupropion is contraindicated for patients who have uncontrolled hypertension, seizure disorder, eating disorder, or end-stage renal failure; who are pregnant; or who have been treated with a monoamine oxidase inhibitor within 14 days. It should not be used with other bupropion-containing products or in patients who have taken opioids chronically or have acute opiate withdrawal.

Because of its bupropion component, this product carries an FDA black-box warning about possible suicidal thoughts and behaviors and neuropsychiatric reactions.

Adverse effects. The adverse reactions most commonly associated with naltrexone-bupropion were nausea (32.5%), constipation (19.2%), headache (17.6%), vomiting (10.7%), dizziness (9.9%), insomnia (9.2%), dry mouth (8.1%), and diarrhea (7.1%).⁴⁴

Liraglutide

Liraglutide, previously FDA-approved to treat type 2 diabetes mellitus under the brand name Victoza, received approval in 2014 in a higher-dose formulation (Saxenda) to treat obesity.

Liraglutide is a glucagon-like peptide-1 receptor agonist that stimulates glucose-dependent insulin release from the pancreatic islet cells, slows gastric emptying, regulates postprandial glucagon, and reduces food intake.

Dosage. Liraglutide is given as a once-daily injection in the abdomen, thigh, or arm. The initial dosage is 0.6 mg daily for the first week and can be titrated up by 0.6 mg weekly to a target dose of 3 mg daily. If a patient does not lose 4% of baseline body weight after 16 weeks on the target dose, the drug should be discontinued because it is unlikely to lead to clinically significant weight loss.

Efficacy. Liraglutide for weight management (3 mg once daily) was evaluated in a large (N = 3,731), randomized, double-blind, placebo-controlled international trial.⁴⁵ Participants did not have diabetes mellitus, but 60% had prediabetes. Liraglutide or placebo was given for 56 weeks, along with lifestyle counseling. At the end of the study, the liraglutide group had lost a mean of 8.4 kg vs 2.8 kg in the placebo group. Additionally, 63% of the liraglutide group lost at least 5% of body weight vs 27% in the placebo group, and 33% lost at least 10% of body weight vs 10% in the placebo group.

A 2-year extension found systolic blood pressure decreased with no change in pulse, and the prevalence of prediabetes and metabolic syndrome decreased by 52% and 59%, respectively.⁴⁶ At 2 years, mean scores for physical function, self-esteem, and work had improved more in the liraglutide group than the placebo group.⁴⁷

Adverse effects. The most common adverse reactions with liraglutide were nausea, vomiting, diarrhea, constipation, hypoglycemia, and loss of appetite. In most cases, nausea and vomiting were tolerable, transient, and associated with greater weight loss but not with decreased quality-of-life scores. Serious adverse reactions included pancreatitis, gallbladder disease, renal impairment, and suicidal thoughts.

For obese patients, when lifestyle modifications do not result in the desired weight loss, pharmacotherapy is an option. Practitioners have several FDA-approved options for weight management. Because of evidence that these drugs can postpone the onset of other complications and improve metabolic and cardiovascular parameters, they should be considered.

In phase 3 trials, these drugs caused modest weight loss of 5% to 10% of body weight. More weight was lost with the combination of phentermine-topiramate than with the other drugs.

In a 2016 meta-analysis, these drugs were associated with at least 5% weight reduction compared with placebo.⁴⁸ Phentermine-topiramate and liraglutide were most likely to produce at least a 5% weight loss, while liraglutide and naltrexone-bupropion were most likely to be discontinued because of adverse events. Combination drugs may have the advantages of synergistic effects on weight loss and fewer adverse reactions because lower doses of the individual drug components are used.

Response to therapy with most of these drugs should be evaluated at 12 weeks on the maintenance dose. If less than 5% weight loss has been achieved, the medication should be discontinued.

Adverse-effect profiles, drug interactions, abuse, misuse, and overdose potential should be considered when prescribing these drugs. Weight-loss drugs are contraindicated in pregnancy because they offer no potential benefit to a pregnant woman and may harm the fetus.

The development of new drugs and better drug combinations is expected to provide more effective therapeutic strategies, which are essential for combating the obesity epidemic.

Weight-loss drugs are not magic pills, but they can help patients lose about 10 to 25 more pounds than they otherwise could, when used in a program that includes diet, exercise, and other lifestyle changes.

See related editorial

HALF OF ADULTS MAY BE OBESE BY 2030

Obesity is a major public health challenge in the United States, with nearly 37% of adults classified as obese.¹ The prevalence has increased more than 75% since 1980,² and it is estimated that 51% of US adults will be obese by 2030.³ Obesity is the second-leading cause of preventable deaths, after smoking.⁴

Obesity increases the risk of many chronic medical conditions, including type 2 diabetes mellitus, heart disease, hypertension, stroke, nonalcoholic fatty liver disease, osteoarthritis, and cancers of the breast, colon, endometrium, and kidney.⁵

WHEN IS DRUG THERAPY INDICATED?

Guidelines from the major obesity societies recommend that all weight-loss programs have a lifestyle component that includes a low-calorie diet, increased physical activity, and behavioral therapy, to which pharmacotherapy may be added as an adjunct.^6–8

Weight-loss medications are indicated for patients who have a body mass index (BMI) of at least 30 kg/m² or who have obesity-associated comorbidities and a BMI of at least 27 kg/m². However, the best results are achieved when pharmacotherapy is combined with lifestyle modification.⁹

HISTORY OF WEIGHT-LOSS DRUGS: NOT A PRETTY PICTURE

The earliest drugs to induce weight loss, which worked mainly by increasing metabolism, included thyroid hormone, amphetamines (which also suppress appetite), and dinitrophenol (a pesticide). Adverse reactions limited their usefulness: cardiovascular effects with thyroid hormones, abuse potential with amphetamines, and neuropathy and cataracts with dinitrophenol.

Researchers then looked to drugs that could suppress appetite like amphetamines do, but without the potential for abuse. Medications that increased levels of norepinephrine and serotonin, both by increasing release and decreasing reuptake of these neuromodulators, had some success. But again, serious adverse effects occurred, and several drugs had to be withdrawn from the market.

The most publicized of these withdrawals was for the combination fenfluramine and phentermine (“fen-phen”) and its cousin dexfenfluramine (Redux). Up to 30% of patients taking fenfluramine-phentermine developed echocardiographic evidence of valvular heart disease.¹¹ Fenfluramine also increased the risk of pulmonary hypertension. These findings led to the 1997 withdrawal of these drugs from the US market.

Sibutramine (Meridia), a norepinephrine and serotonin reuptake inhibitor, was approved for weight loss in 1997. Increases in blood pressure and heart rate were noted in the initial trial,¹² and then a postmarketing study found increased rates of nonfatal myocardial infarction and stroke in patients with preexisting cardiovascular disease or diabetes mellitus.¹³ Based on these results, sibutramine was withdrawn from both US and European markets.

Rimonabant (Acomplia, Zimulti), a cannabinoid-receptor inhibitor, was approved in Europe in 2006, but its approval was withdrawn just 2 years later because of increased suicidality in a postmarketing study.¹⁴ It was never approved for use in the United States.

NORADRENERGIC SYMPATHOMIMETICS: FOR SHORT-TERM USE

Several noradrenergic sympathomimetic drugs are FDA-approved for short-term weight loss, but phentermine is by far the most commonly prescribed drug in this class. In fact, it is the most commonly prescribed drug for obesity in the United States.¹⁵

Phentermine

Phentermine is an atypical amphetamine analogue that suppresses appetite by norepinephrine agonism in the central nervous system. The FDA approved it for short-term weight management in 1959, and its use became widespread in the 1960s, followed by decades of popularity.

Dosage. Phentermine is prescribed at an oral dose of 15, 30, or 37.5 mg daily, either before breakfast or 1 to 2 hours after. It is a schedule IV controlled substance, based on its similarity to amphetamine. (The 5 US controlled substance schedules range from schedule I, which includes heroin, amphetamine, and cannabis, to schedule V, which includes cough syrups containing no more than 200 mg of codeine per 100 mL.) However, concerns about addiction and dependence with phentermine are largely unfounded, and abrupt cessation of the drug has not been shown to cause amphetamine-like withdrawal.¹⁶

Adverse effects. Common adverse reactions include nervousness, insomnia, and dry mouth, but these effects tend to wane with continued use.

Contraindications. Cardiovascular disease is a contraindication to phentermine because of concerns about increased blood pressure and pulse rate, although these concerns seem to be more theoretic than observed.¹⁶ Other contraindications include hyperthyroidism, glaucoma, agitation, a history of drug abuse, pregnancy, breastfeeding, and current or recent use of a monoamine oxidase inhibitor. No serious adverse events have been reported in trials of phentermine.

Efficacy. In a pooled analysis of 6 trials lasting 2 to 24 weeks completed between 1975 and 1999, phentermine-treated patients lost an average of 3.6 kg more weight than placebo recipients.¹⁷ More than 80% of study participants were women.

In a 36-week study in 108 women,¹⁸ participants lost a mean of 12.2 kg with continuous phentermine use, 13.0 kg with intermittent use (4 weeks on, 4 weeks off; the difference was not significant), and 4.8 kg with placebo.

Minimal data exist on long-term efficacy of phentermine monotherapy.

DRUGS FOR LONG-TERM THERAPY

Orlistat

Orlistat was approved as a prescription drug (Xenical, 120 mg) in 1999 and as an over-the-counter medication (Alli, 60 mg) in 2007.

Orlistat works by inhibiting pancreatic and gastric lipase, causing incomplete hydrol­ysis of ingested fat, thereby increasing fecal fat excretion in a dose-dependent manner. It is a good choice for weight-loss drug therapy because of its safe cardiovascular risk profile and beneficial effects on lipid levels. However, its long-term effect on weight is only modest.^19,20

Dosage. The dosage for prescription orlistat is 120 mg 3 times per day, in addition to a low-fat diet (< 30% of daily calories from fat). To prevent potential deficiencies of fat-soluble vitamins, a daily multivitamin supplement is recommended, but it should not be taken with meals.

Efficacy. In a 2014 systematic review, 35% to 73% of patients treated with orlistat 120 mg had lost at least 5% of their body weight at 1 year, and 14% to 41% had lost at least 10%.²¹ At the end of the second year, orlistat-treated patients had lost about 3.3 kg more than placebo recipients.

In a randomized trial,²² 4 years of treatment with orlistat vs placebo led to a significant (37.3%) risk reduction in the incidence of type 2 diabetes mellitus in obese participants, as well as significant improvements in cardiovascular risk factors. Mean weight loss at 1 year was significantly greater with orlistat than with placebo (10.6 vs 6.2 kg), and it remained greater at 4 years (5.8 vs 3.0 kg; P < .001).

Adverse effects. Long-term orlistat use is hampered by adverse reactions. A population-based, retrospective cohort analysis showed that fewer than 10% of patients were still using it at 1 year, and only 2% were using it at 2 years, although reasons for discontinuation were not reported.²³

Adverse reactions are predominantly gastrointestinal, attributed to the high content of undigested fat in stools. Patients who do not limit their dietary fat intake are affected the most. Other reported adverse reactions include hepatotoxicity and oxalate-induced nephropathy.

Orlistat has been reported to interfere with some drugs, particularly those that are lipophilic. Drugs that should be closely monitored with orlistat are warfarin, amiodarone, cyclosporine, certain antiepileptic drugs, and levothyroxine.

Phentermine-topiramate

The combination of phentermine and topiramate was approved by the FDA in 2012 and is available under the brand name Qsymia.

Topiramate had been approved for treating seizure disorder in 1996 and as migraine prophylaxis in 2004. It is not approved as monotherapy for obesity; however, patients taking it for seizures or for psychiatric disorders (eg, binge eating, borderline personality disorder) have reported weight loss during treatment.

How topiramate promotes weight loss is not known. Proposed mechanisms include taste inhibition by carbonic anhydrase, influences on gamma-aminobutyric acid transmission causing appetite suppression, sensitization of insulin activity, and adiponectin secretion in the peripheral tissues.^24,25

Phentermine-topiramate therapy has an advantage over monotherapy because lower doses of each medication can be used to achieve the same benefit, thus avoiding dose-related adverse reactions.

Dosage. Phentermine-topiramate is available in capsules containing 3.75/23, 7.5/46, 11.25/69, and 15/92 mg. The recommended starting dosage is 3.75/23 mg/day for 14 days, increasing to 7.5/46 mg/day. If patients do not lose at least 3% of their body weight after 12 weeks, the dose can be increased to 11.25/69 mg daily for 14 days, followed by 15/92 mg daily.²⁶ Phentermine-topiramate is a schedule IV controlled substance with a low potential for abuse and dependence.

Efficacy. Approval of phentermine-topiramate for treating obesity was primarily based on 3 clinical trials.^27–29 In 1 of these trials,²⁸ at 1 year, patients had lost 9.9 kg with the medium dose and 12.9 kg with the high dose.

Adverse effects. Phentermine-topiramate was well tolerated in the trials. The most commonly reported adverse reactions were dry mouth, dizziness, constipation, insomnia, dysgeusia, paresthesia, and increased resting heart rate.^28,29 Acute myopia and angle-closure glaucoma also have been reported with topiramate.³⁰ Topiramate monotherapy has been associated with dose-dependent neuropsychiatric adverse effects, including memory symptoms and depression. However, across all 3 trials of phentermine-topiramate therapy, symptoms of depression improved over time, and no significant increase in suicide risk was identified.^27–29

Recommended monitoring for patients on phentermine-topiramate includes a blood chemistry panel, resting heart rate, blood pressure, and depression screening.

Because topiramate has teratogenic potential (craniofacial abnormalities), it is labeled as pregnancy category X (contraindicated). A negative pregnancy test is needed before women of childbearing age take the drug and monthly thereafter. Women should be counseled to use effective birth control. A home pregnancy test is an alternative to laboratory testing, but this option should be left to the prescribing clinician’s judgment and be based on reliability of the test and patient compliance.

Lorcaserin

Lorcaserin (Belviq) was approved by the FDA in 2012 for chronic weight management. It suppresses appetite by activating the serotonin 2C receptor in the brain. Because it is selective for the 2C receptor, it does not appear to have the same detrimental effects on heart valves as occurred with less-selective serotonergic agents such as fenfluramine and dexfenfluramine.³¹

Dosage. The recommended dosage for lorcaserin is 10 mg twice daily. Lorcaserin is a schedule IV controlled substance because of studies that showed increases in positive subjective measures such as euphoria in patients taking the drug. The incidence of euphoria was similar to that seen with zolpidem.³²

Efficacy. Lorcaserin was approved on the basis of 2 trials in nondiabetic obese and overweight adults who did not have diabetes but who had a weight-related condition,^33,34 and in a third trial in obese and overweight adults with type 2 diabetes mellitus who were taking oral hypoglycemic agents.³⁵ In these trials, lorcaserin use resulted in a modest 4.7- to 5.8-kg weight loss compared with 1.6 to 2.2 kg in the placebo group.^33–35 There was a high dropout rate in all 3 of these studies (33% to 45% of participants).

A pilot study that added phentermine to lorcaserin yielded double the weight loss from lorcaserin alone.³⁶ This drug combination warrants further investigation.

Contraindications. Lorcaserin should not be given to patients who have severe renal insufficiency (creatinine clearance < 30 mL/min) or severe hepatic impairment, or who are pregnant.

Adverse effects. Common adverse reactions include dry mouth, dizziness, somnolence, headache, and gastrointestinal disturbances (nausea, constipation, or diarrhea).³⁷

Patients with type 2 diabetes mellitus should be monitored for hypoglycemia.

Lorcaserin should be used with extreme caution in patients taking other serotonergic agents because of the risk of the serotonin syndrome.

A theoretic potential for increased risk of breast cancer also exists with lorcaserin. When rats were given supraphysiologic doses of lorcaserin (more than 50 times higher than recommended in humans), fibroadenomas and adenocarcinomas occurred at higher rates.³⁸ Breast cancer data were not reported in the 3 randomized trials discussed above.^33–35

Naltrexone-bupropion

The combination of naltrexone and bupropion was approved by the FDA in 2014 under the brand name Contrave. Both drugs are approved for monotherapy in conditions other than obesity.

Naltrexone is a mu opioid receptor antagonist approved to treat alcohol and opioid dependency. Bupropion is a dopamine-norepinephrine reuptake inhibitor approved to treat depression and to help with smoking cessation. Combining the drugs produces weight loss and metabolic benefits through effects on 2 areas of the brain that regulate food intake: the hypothalamus (appetite) and the mesolimbic dopamine circuit (reward system).

Dosage. Naltrexone-bupropion comes as an extended-release tablet of 8/90 mg. The maintenance dose of 2 tablets twice daily is reached at week 4 through a specific dose-titration regimen (Table 1). The dose should be adjusted if patients have renal or hepatic impairment or if they are also taking a CYP2B6 inhibitor.

Efficacy. FDA approval was based on the results of 4 clinical trials.^39–42 Using a modified intention-to-treat analysis, Yanovski and Yanovski⁴³ calculated that at 1 year, placebo-subtracted mean weight loss was 4.6% (4.9 kg), and mean total weight loss was 6.8% (7.3 kg) across the studies. Attrition rates, however, were high, ranging from 42% to 50%.

Cardiometabolic effects in 2 of the trials^40,41 included decreased waist circumference, triglyceride levels, and C-reactive protein levels, and increased high-density lipoprotein levels at the initial dose. At the maintenance dose, additional lowering of fasting plasma insulin and glucose levels occurred along with lower levels of the homeostatic model assessment of insulin resistance. In the COR-Diabetes Study Group trial, patients with type 2 diabetes mellitus had decreased hemoglobin A_1c levels without an increase in hypoglycemia and an increased likelihood of reaching the target hemoglobin A_1c level below 7%.³⁹

Contraindications. Naltrexone-bupropion is contraindicated for patients who have uncontrolled hypertension, seizure disorder, eating disorder, or end-stage renal failure; who are pregnant; or who have been treated with a monoamine oxidase inhibitor within 14 days. It should not be used with other bupropion-containing products or in patients who have taken opioids chronically or have acute opiate withdrawal.

Because of its bupropion component, this product carries an FDA black-box warning about possible suicidal thoughts and behaviors and neuropsychiatric reactions.

Adverse effects. The adverse reactions most commonly associated with naltrexone-bupropion were nausea (32.5%), constipation (19.2%), headache (17.6%), vomiting (10.7%), dizziness (9.9%), insomnia (9.2%), dry mouth (8.1%), and diarrhea (7.1%).⁴⁴

Liraglutide

Liraglutide, previously FDA-approved to treat type 2 diabetes mellitus under the brand name Victoza, received approval in 2014 in a higher-dose formulation (Saxenda) to treat obesity.

Liraglutide is a glucagon-like peptide-1 receptor agonist that stimulates glucose-dependent insulin release from the pancreatic islet cells, slows gastric emptying, regulates postprandial glucagon, and reduces food intake.

Dosage. Liraglutide is given as a once-daily injection in the abdomen, thigh, or arm. The initial dosage is 0.6 mg daily for the first week and can be titrated up by 0.6 mg weekly to a target dose of 3 mg daily. If a patient does not lose 4% of baseline body weight after 16 weeks on the target dose, the drug should be discontinued because it is unlikely to lead to clinically significant weight loss.

Efficacy. Liraglutide for weight management (3 mg once daily) was evaluated in a large (N = 3,731), randomized, double-blind, placebo-controlled international trial.⁴⁵ Participants did not have diabetes mellitus, but 60% had prediabetes. Liraglutide or placebo was given for 56 weeks, along with lifestyle counseling. At the end of the study, the liraglutide group had lost a mean of 8.4 kg vs 2.8 kg in the placebo group. Additionally, 63% of the liraglutide group lost at least 5% of body weight vs 27% in the placebo group, and 33% lost at least 10% of body weight vs 10% in the placebo group.

A 2-year extension found systolic blood pressure decreased with no change in pulse, and the prevalence of prediabetes and metabolic syndrome decreased by 52% and 59%, respectively.⁴⁶ At 2 years, mean scores for physical function, self-esteem, and work had improved more in the liraglutide group than the placebo group.⁴⁷

Adverse effects. The most common adverse reactions with liraglutide were nausea, vomiting, diarrhea, constipation, hypoglycemia, and loss of appetite. In most cases, nausea and vomiting were tolerable, transient, and associated with greater weight loss but not with decreased quality-of-life scores. Serious adverse reactions included pancreatitis, gallbladder disease, renal impairment, and suicidal thoughts.

For obese patients, when lifestyle modifications do not result in the desired weight loss, pharmacotherapy is an option. Practitioners have several FDA-approved options for weight management. Because of evidence that these drugs can postpone the onset of other complications and improve metabolic and cardiovascular parameters, they should be considered.

In phase 3 trials, these drugs caused modest weight loss of 5% to 10% of body weight. More weight was lost with the combination of phentermine-topiramate than with the other drugs.

In a 2016 meta-analysis, these drugs were associated with at least 5% weight reduction compared with placebo.⁴⁸ Phentermine-topiramate and liraglutide were most likely to produce at least a 5% weight loss, while liraglutide and naltrexone-bupropion were most likely to be discontinued because of adverse events. Combination drugs may have the advantages of synergistic effects on weight loss and fewer adverse reactions because lower doses of the individual drug components are used.

Response to therapy with most of these drugs should be evaluated at 12 weeks on the maintenance dose. If less than 5% weight loss has been achieved, the medication should be discontinued.

Adverse-effect profiles, drug interactions, abuse, misuse, and overdose potential should be considered when prescribing these drugs. Weight-loss drugs are contraindicated in pregnancy because they offer no potential benefit to a pregnant woman and may harm the fetus.

The development of new drugs and better drug combinations is expected to provide more effective therapeutic strategies, which are essential for combating the obesity epidemic.

A 76-year-old man with ulcerative colitis presented with a 1-week history of low-grade fever and progressive dyspnea. He was taking infliximab for the ulcerative colitis. He was known to be negative for human immunodeficiency virus.

Since the M tuberculosis cultured from his lung proved to be sensitive to the antituberculosis drugs, we suspected that the nodules were a paradoxical reaction to the drug therapy, and thus we continued the treatment because of the continued low-grade fever. After 9 months of therapy, the fever had resolved and the nodules had disappeared, confirming our suspicion of a paradoxical reaction. The number of lymphocytes gradually increased during drug therapy.

Paradoxical reaction during tuberculosis treatment is defined as a worsening of pre-existing lesions or as the emergence of new lesions during appropriate therapy.^1,2 The diagnosis is sometimes difficult, since new lesions can resemble other lung diseases. However, a paradoxical reaction involving randomly distributed nodules is rare and radiographically resembles metastatic lung cancer. Clinicians should be aware of this type of reaction in patients on tuberculosis therapy.