Cleveland Clinic Journal of Medicine

A 35-year-old woman is admitted to the hospital with a 5-day history of abdominal distention and jaundice. She reports no history of fever, chills, night sweats, abdominal pain, nausea, vomiting, diarrhea, changes in urine color, change in stool color, weight loss, weight gain, or loss of appetite.

She is petite, with a body mass index of 19.4 kg/m². She has no known history of medical conditions or surgery and is not taking any medications. Her family history is unremarkable, and she denies current or past tobacco, alcohol, or illicit drug use.

RECENT TRAVEL

She says that during a trip to Central America several months ago, she had suffered a seizure and was taken to a local hospital, where laboratory testing revealed elevated aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels. She says that the rest of the workup at that time was normal.

About 1 week after that incident, she returned home and saw her primary care physician, who ordered further testing, which showed mild hyperbilirubinemia and mild elevation of AST and ALT levels. Her physician attributed the elevations to atovaquone, which she had been taking for malaria prophylaxis, as repeat testing 2 weeks later showed improvement in AST and ALT levels.

The patient says she returned to her normal state of health until about 5 days ago, when she noticed jaundice and abdominal distention, but without abdominal pain, dark urine, or clay-colored stools. She became concerned and went to her local hospital. Testing there noted mild elevation of AST and ALT, as well as an elevated international normalized ratio (INR) and hyperbilirubinemia. Computed tomography of the abdomen and pelvis showed hepatomegaly with possible fatty liver. Because of these results, the patient was transferred to our institution for further evaluation.

EVALUATION AT OUR INSTITUTION

On examination at our institution, she is afebrile, and vital signs are within normal ranges. She has bilateral scleral icterus and diffuse jaundice, but no other skin finding such as rash or spider angioma. She has no lymphadenopathy. Her abdomen is distended, with tense ascites, and her liver is tender to palpation. The tip of the spleen is not palpable.

The cardiovascular examination reveals no murmurs, rubs, or gallops, but she has jugular venous distention and +2 pitting edema of both lower extremities.

On respiratory examination, there is dullness to percussion, with slight crackles on auscultation at the right lung base. The neurologic examination is normal.

Table 1 shows the results of initial laboratory testing.

1. Which study would provide the most information on the cause of ascites?

• Abdominal ultrasonography
• Abdominal paracentesis with ascitic fluid analysis
• Chest radiography
• Echocardiography
• Urine protein-to-creatinine ratio

Abdominal paracentesis with ascitic fluid analysis is the essential study for any patient with clinically apparent new-onset ascites.^1–3 It is the study that provides the most information on the cause of ascites.

In our patient, abdominal paracentesis yields 1,000 mL of straw-colored ascitic fluid, and analysis shows 86 nucleated cells, 28 of which are polymorphonuclear cells, and 0 red blood cells, with negative Gram stain and culture. The ascitic albumin level is 0.85 g/dL, with an ascitic protein of 1.1 g/dL.

Abdominal ultrasonography shows a diffusely echogenic liver, no focal lesions, moderate ascites, normal portal vein flow, no intrahepatic or extrahepatic biliary duct dilation, normal kidney sizes, no hydronephrosis, and no intra-abdominal mass. Chest radiography is clear with no sign of consolidation, edema, or effusion. Echocardiography shows a normal left ventricular ejection fraction with no valvular disease or pericardial effusion. A random urine protein-creatinine ratio is normal at 0.1 (reference range < 0.2).

2. What is the most likely cause of her ascites based on the workup to this point?

• Cirrhosis
• Heart failure
• Nephrotic syndrome
• Portal vein thrombus
• Abdominal malignancy
• Malaria

An initial approach to ascitic fluid analysis is to calculate the serum-ascites albumin gradient (SAAG). The SAAG is calculated as the serum albumin level minus the ascitic fluid albumin level.^4,5 This is useful in determining the cause of the ascites (Figure 1).^4,5 A gradient of 1.1 g/dL or higher indicates portal hypertension.^4,5

Common causes of portal hypertension include cirrhosis, alcoholic hepatitis, heart failure, vascular occlusion syndromes (eg, Budd-Chiari syndrome, portal vein thrombosis), idiopathic portal fibrosis, and metastatic liver disease.^5,6

If portal hypertension is present based on the SAAG, the next step is to review the ascitic protein level to help distinguish between a hepatic and a cardiac etiology of the ascites. An ascitic protein level less than 2.5 g/dL indicates a primary liver pathology (eg, cirrhosis). An ascitic protein level of 2.5 g/dL or greater typically indicates a cardiac condition (eg, heart failure, pericardial disease) with secondary congestive hepatopathy.^5,6

If the SAAG is less than 1.1 g/dL, the ascites is likely not from portal hypertension. Typical causes of a low SAAG include infection, malignancy, pancreatic ascites, and nephrotic syndrome.^5,6

In our patient, the SAAG is 1.35 g/dL (2.2 g/dL minus 0.85 g/dL), ie, elevated and due to portal hypertension. With an SAAG of 1.1 g/dL or greater and an ascitic fluid protein level less than 2.5 g/dL, as in our patient, the most likely cause is cirrhosis.

Heart failure is unlikely based on her normal brain natriuretic peptide level, an ascitic fluid protein level below 2.5 g/dL, and normal results on echocardiography. Nephrotic syndrome is also very unlikely based on the patient’s normal random urine protein-creatinine ratio. Portal vein thrombus and abdominal malignancy are essentially ruled out by the negative results of Doppler abdominal ultrasonography, with normal venous flow and no intra-abdominal mass and coupled with an elevated SAAG.

Although the patient has a history of travel, the incubation period for malaria would not fit the time frame of presentation. Also, she did not have typical malarial symptoms, her rapid malaria test was negative, and a peripheral blood smear for blood parasites was negative. It should be noted, however, that Plasmodium malariae infection classically presents with flulike symptoms and can resemble nephrotic syndrome, including peripheral edema, ascites, heavy proteinuria, hypoalbuminemia, and hyperlipidemia.⁷

3. In which patients is antibiotic prophylaxis against spontaneous bacterial peritonitis (SBP) appropriate?

• Any patient with cirrhosis
• Any patient with cirrhosis who is hospitalized
• Any patient with cirrhosis and an ascitic fluid protein level below 2.0 g/dL
• Any patient with cirrhosis and a history of SBP

Any patient with cirrhosis and a history of SBP should receive prophylactic antibiotics,⁸ as should any patient deemed at high risk of SBP. It is indicated in the following patients:

• Patients with cirrhosis and gastrointestinal bleeding^9,10
• Patients with cirrhosis and a previous episode of SBP⁸
• Patients with cirrhosis and an ascitic fluid protein level less than 1.5 g/dL with either impaired renal function (creatinine ≥ 1.2 mg/dL, blood urea nitrogen level ≥ 25 mg/dL, or serum sodium ≤ 130 mmol/L) or liver failure (Child-Pugh score ≥ 9 and a bilirubin ≥ 3 mg/dL)⁹
• Patients with cirrhosis who are hospitalized for other reasons and have an ascitic protein level < 1.0 g/dL.⁹

Our patient has no signs or symptoms of gastrointestinal bleeding and no history of SBP. Her ascitic fluid protein level is 1.1 g/dL, and she has normal renal function. However, her Child-Pugh score is 12 (3 points for total bilirubin > 3 mg/dL, 3 points for serum albumin < 2.8 g/dL, 2 points for an INR 1.7 to 2.2, 3 points for moderate ascites, and 1 point for no encephalopathy), with a bilirubin of 17.0 mg/dL. Based on this, she is placed on antibiotic prophylaxis for SBP.

Our patient then undergoes an extensive workup for liver disease. Results of tests for toxins, autoimmune diseases, and inheritable diseases are all within normal limits. At this point, despite the patient’s reported negative alcohol history, our leading diagnosis is alcoholic hepatitis.

To confirm this diagnosis, she subsequently undergoes transjugular liver biopsy, considered the gold standard for the diagnosis of alcoholic hepatitis. During the procedure, the hepatic venous pressure gradient is measured at 18 mm Hg (reference range 1–5 mm Hg), suggestive of portal hypertension. The pathology study shows severe fatty change, active steatohepatitis with ballooning degeneration, easily identifiable Mallory-Denk bodies, and prominent neutrophilic infiltration, as well as extensive bridging fibrosis (Figure 2). These findings point to alcoholic hepatitis.

After the biopsy results, we speak with the patient further about her alcohol habits. At this point, she informs us that she has consumed significant amounts of alcohol since the age of 18 (6 to 12 alcoholic beverages per day, including beer and hard liquor). Therefore, based on this new information, on her jaundice and ascites, and on results of laboratory testing and biopsy, we confirmed our diagnosis of alcoholic hepatitis.

4. When is drug treatment appropriate for alcoholic hepatitis?

• Model for End-stage Liver Disease (MELD) score greater than 12
• MELD score greater than 15
• Maddrey Discriminant Function score greater than 25
• Maddrey Discriminant Function score greater than 32
• Glasgow score greater than 5
• Glasgow score greater than 7

The best answer is a Maddrey Discriminant Function score greater than 32. A variety of scoring systems have been used to assess the severity of alcoholic hepatitis and to guide treatment, including the Maddrey Discriminant Function score, the MELD score, and the Glasgow score.^11–16 They share similar laboratory values in their calculations, including prothrombin time (or INR) and total bilirubin.^11–16 Typically, a Maddrey Discriminant Function score greater than 32, a Glasgow score of greater than 9, or a MELD score greater than 21 is used to determine whether pharmacologic treatment is indicated.^11–16

The typical treatment is prednisolone or pentoxifylline.^11,17–21 The Lille score is designed to help decide whether to stop corticosteroids after 1 week of administration due to lack of treatment response.²² It predicts mortality rates within 6 months; a score of 0.45 or less indicates a good prognosis, and corticosteroid therapy should continue for 28 days (Figure 3).²²

Our patient’s discriminant function score is 50, her Glasgow score is 10, and her MELD score is 28; thus, she begins treatment with oral prednisolone. Her Lille score at 1 week is 0.119, indicating a good prognosis, and her corticosteroids are continued for a total of 28 days.

It should be highlighted that the most important treatment is abstinence from alcohol.¹¹ Recent literature suggests that any benefit of prednisolone or pentoxifylline in terms of mortality rates is questionable,^19–20 and there is evidence that giving both drugs simultaneously may improve mortality rates,^11,21 but the evidence remains conflicting at this time.

ALCOHOLIC HEPATITIS

Alcoholic hepatitis is a clinical syndrome of jaundice and liver failure, often in the setting of heavy alcohol use for decades.^11,12 The incidence is unknown, but the typical age of presentation is between 40 and 50.^11,12 The chief sign is a rapid onset of jaundice (< 3 months); common signs and symptoms include fever, ascites, proximal muscle loss, and an enlarged, tender liver.¹² Encephalopathy may be seen in severe alcoholic hepatitis.¹²

Our patient is 35 years old. She has jaundice with rapid onset, as well as ascites and a tender liver.

The diagnosis of alcoholic hepatitis must take into account the patient’s history, physical examination, and laboratory findings. Until proven otherwise, the diagnosis should be presumed in the following scenario: ascites and jaundice on examination (usually with a duration < 3 months); a history of heavy alcohol use; neutrophilic leukocytosis; an AST level that is elevated but below 300 U/L; an ALT level above the normal range but below 300 U/L; an AST-ALT ratio greater than 2; a total serum bilirubin level above 5 mg/dL; and an elevated INR.^11,12 Liver biopsy is the gold standard for diagnosis. Though not routinely done because of risks associated with the procedure, it may help confirm the diagnosis if it is in question.

CASE CONCLUDED

We start our patient on oral prednisolone 40 mg daily for alcoholic hepatitis. Her symptoms and laboratory testing results including bilirubin improve. Her Lille score at 7 days indicates a good prognosis, prompting continuation of corticosteroid treatment for the full 28 days.

She is referred to an outpatient alcohol rehabilitation program and has remained sober as of the last outpatient note.

Alcoholic hepatitis is extremely difficult to diagnose, and no single blood test or imaging study confirms the diagnosis. The history, physical examination findings, and laboratory findings are crucial. If the diagnosis is still in doubt, liver biopsy may help confirm the diagnosis.

A 35-year-old woman is admitted to the hospital with a 5-day history of abdominal distention and jaundice. She reports no history of fever, chills, night sweats, abdominal pain, nausea, vomiting, diarrhea, changes in urine color, change in stool color, weight loss, weight gain, or loss of appetite.

She is petite, with a body mass index of 19.4 kg/m². She has no known history of medical conditions or surgery and is not taking any medications. Her family history is unremarkable, and she denies current or past tobacco, alcohol, or illicit drug use.

RECENT TRAVEL

She says that during a trip to Central America several months ago, she had suffered a seizure and was taken to a local hospital, where laboratory testing revealed elevated aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels. She says that the rest of the workup at that time was normal.

About 1 week after that incident, she returned home and saw her primary care physician, who ordered further testing, which showed mild hyperbilirubinemia and mild elevation of AST and ALT levels. Her physician attributed the elevations to atovaquone, which she had been taking for malaria prophylaxis, as repeat testing 2 weeks later showed improvement in AST and ALT levels.

The patient says she returned to her normal state of health until about 5 days ago, when she noticed jaundice and abdominal distention, but without abdominal pain, dark urine, or clay-colored stools. She became concerned and went to her local hospital. Testing there noted mild elevation of AST and ALT, as well as an elevated international normalized ratio (INR) and hyperbilirubinemia. Computed tomography of the abdomen and pelvis showed hepatomegaly with possible fatty liver. Because of these results, the patient was transferred to our institution for further evaluation.

EVALUATION AT OUR INSTITUTION

On examination at our institution, she is afebrile, and vital signs are within normal ranges. She has bilateral scleral icterus and diffuse jaundice, but no other skin finding such as rash or spider angioma. She has no lymphadenopathy. Her abdomen is distended, with tense ascites, and her liver is tender to palpation. The tip of the spleen is not palpable.

The cardiovascular examination reveals no murmurs, rubs, or gallops, but she has jugular venous distention and +2 pitting edema of both lower extremities.

On respiratory examination, there is dullness to percussion, with slight crackles on auscultation at the right lung base. The neurologic examination is normal.

Table 1 shows the results of initial laboratory testing.

1. Which study would provide the most information on the cause of ascites?

• Abdominal ultrasonography
• Abdominal paracentesis with ascitic fluid analysis
• Chest radiography
• Echocardiography
• Urine protein-to-creatinine ratio

Abdominal paracentesis with ascitic fluid analysis is the essential study for any patient with clinically apparent new-onset ascites.^1–3 It is the study that provides the most information on the cause of ascites.

In our patient, abdominal paracentesis yields 1,000 mL of straw-colored ascitic fluid, and analysis shows 86 nucleated cells, 28 of which are polymorphonuclear cells, and 0 red blood cells, with negative Gram stain and culture. The ascitic albumin level is 0.85 g/dL, with an ascitic protein of 1.1 g/dL.

Abdominal ultrasonography shows a diffusely echogenic liver, no focal lesions, moderate ascites, normal portal vein flow, no intrahepatic or extrahepatic biliary duct dilation, normal kidney sizes, no hydronephrosis, and no intra-abdominal mass. Chest radiography is clear with no sign of consolidation, edema, or effusion. Echocardiography shows a normal left ventricular ejection fraction with no valvular disease or pericardial effusion. A random urine protein-creatinine ratio is normal at 0.1 (reference range < 0.2).

2. What is the most likely cause of her ascites based on the workup to this point?

• Cirrhosis
• Heart failure
• Nephrotic syndrome
• Portal vein thrombus
• Abdominal malignancy
• Malaria

An initial approach to ascitic fluid analysis is to calculate the serum-ascites albumin gradient (SAAG). The SAAG is calculated as the serum albumin level minus the ascitic fluid albumin level.^4,5 This is useful in determining the cause of the ascites (Figure 1).^4,5 A gradient of 1.1 g/dL or higher indicates portal hypertension.^4,5

Common causes of portal hypertension include cirrhosis, alcoholic hepatitis, heart failure, vascular occlusion syndromes (eg, Budd-Chiari syndrome, portal vein thrombosis), idiopathic portal fibrosis, and metastatic liver disease.^5,6

If portal hypertension is present based on the SAAG, the next step is to review the ascitic protein level to help distinguish between a hepatic and a cardiac etiology of the ascites. An ascitic protein level less than 2.5 g/dL indicates a primary liver pathology (eg, cirrhosis). An ascitic protein level of 2.5 g/dL or greater typically indicates a cardiac condition (eg, heart failure, pericardial disease) with secondary congestive hepatopathy.^5,6

If the SAAG is less than 1.1 g/dL, the ascites is likely not from portal hypertension. Typical causes of a low SAAG include infection, malignancy, pancreatic ascites, and nephrotic syndrome.^5,6

In our patient, the SAAG is 1.35 g/dL (2.2 g/dL minus 0.85 g/dL), ie, elevated and due to portal hypertension. With an SAAG of 1.1 g/dL or greater and an ascitic fluid protein level less than 2.5 g/dL, as in our patient, the most likely cause is cirrhosis.

Heart failure is unlikely based on her normal brain natriuretic peptide level, an ascitic fluid protein level below 2.5 g/dL, and normal results on echocardiography. Nephrotic syndrome is also very unlikely based on the patient’s normal random urine protein-creatinine ratio. Portal vein thrombus and abdominal malignancy are essentially ruled out by the negative results of Doppler abdominal ultrasonography, with normal venous flow and no intra-abdominal mass and coupled with an elevated SAAG.

Although the patient has a history of travel, the incubation period for malaria would not fit the time frame of presentation. Also, she did not have typical malarial symptoms, her rapid malaria test was negative, and a peripheral blood smear for blood parasites was negative. It should be noted, however, that Plasmodium malariae infection classically presents with flulike symptoms and can resemble nephrotic syndrome, including peripheral edema, ascites, heavy proteinuria, hypoalbuminemia, and hyperlipidemia.⁷

3. In which patients is antibiotic prophylaxis against spontaneous bacterial peritonitis (SBP) appropriate?

• Any patient with cirrhosis
• Any patient with cirrhosis who is hospitalized
• Any patient with cirrhosis and an ascitic fluid protein level below 2.0 g/dL
• Any patient with cirrhosis and a history of SBP

Any patient with cirrhosis and a history of SBP should receive prophylactic antibiotics,⁸ as should any patient deemed at high risk of SBP. It is indicated in the following patients:

• Patients with cirrhosis and gastrointestinal bleeding^9,10
• Patients with cirrhosis and a previous episode of SBP⁸
• Patients with cirrhosis and an ascitic fluid protein level less than 1.5 g/dL with either impaired renal function (creatinine ≥ 1.2 mg/dL, blood urea nitrogen level ≥ 25 mg/dL, or serum sodium ≤ 130 mmol/L) or liver failure (Child-Pugh score ≥ 9 and a bilirubin ≥ 3 mg/dL)⁹
• Patients with cirrhosis who are hospitalized for other reasons and have an ascitic protein level < 1.0 g/dL.⁹

Our patient has no signs or symptoms of gastrointestinal bleeding and no history of SBP. Her ascitic fluid protein level is 1.1 g/dL, and she has normal renal function. However, her Child-Pugh score is 12 (3 points for total bilirubin > 3 mg/dL, 3 points for serum albumin < 2.8 g/dL, 2 points for an INR 1.7 to 2.2, 3 points for moderate ascites, and 1 point for no encephalopathy), with a bilirubin of 17.0 mg/dL. Based on this, she is placed on antibiotic prophylaxis for SBP.

Our patient then undergoes an extensive workup for liver disease. Results of tests for toxins, autoimmune diseases, and inheritable diseases are all within normal limits. At this point, despite the patient’s reported negative alcohol history, our leading diagnosis is alcoholic hepatitis.

To confirm this diagnosis, she subsequently undergoes transjugular liver biopsy, considered the gold standard for the diagnosis of alcoholic hepatitis. During the procedure, the hepatic venous pressure gradient is measured at 18 mm Hg (reference range 1–5 mm Hg), suggestive of portal hypertension. The pathology study shows severe fatty change, active steatohepatitis with ballooning degeneration, easily identifiable Mallory-Denk bodies, and prominent neutrophilic infiltration, as well as extensive bridging fibrosis (Figure 2). These findings point to alcoholic hepatitis.

After the biopsy results, we speak with the patient further about her alcohol habits. At this point, she informs us that she has consumed significant amounts of alcohol since the age of 18 (6 to 12 alcoholic beverages per day, including beer and hard liquor). Therefore, based on this new information, on her jaundice and ascites, and on results of laboratory testing and biopsy, we confirmed our diagnosis of alcoholic hepatitis.

4. When is drug treatment appropriate for alcoholic hepatitis?

• Model for End-stage Liver Disease (MELD) score greater than 12
• MELD score greater than 15
• Maddrey Discriminant Function score greater than 25
• Maddrey Discriminant Function score greater than 32
• Glasgow score greater than 5
• Glasgow score greater than 7

The best answer is a Maddrey Discriminant Function score greater than 32. A variety of scoring systems have been used to assess the severity of alcoholic hepatitis and to guide treatment, including the Maddrey Discriminant Function score, the MELD score, and the Glasgow score.^11–16 They share similar laboratory values in their calculations, including prothrombin time (or INR) and total bilirubin.^11–16 Typically, a Maddrey Discriminant Function score greater than 32, a Glasgow score of greater than 9, or a MELD score greater than 21 is used to determine whether pharmacologic treatment is indicated.^11–16

The typical treatment is prednisolone or pentoxifylline.^11,17–21 The Lille score is designed to help decide whether to stop corticosteroids after 1 week of administration due to lack of treatment response.²² It predicts mortality rates within 6 months; a score of 0.45 or less indicates a good prognosis, and corticosteroid therapy should continue for 28 days (Figure 3).²²

Our patient’s discriminant function score is 50, her Glasgow score is 10, and her MELD score is 28; thus, she begins treatment with oral prednisolone. Her Lille score at 1 week is 0.119, indicating a good prognosis, and her corticosteroids are continued for a total of 28 days.

It should be highlighted that the most important treatment is abstinence from alcohol.¹¹ Recent literature suggests that any benefit of prednisolone or pentoxifylline in terms of mortality rates is questionable,^19–20 and there is evidence that giving both drugs simultaneously may improve mortality rates,^11,21 but the evidence remains conflicting at this time.

ALCOHOLIC HEPATITIS

Alcoholic hepatitis is a clinical syndrome of jaundice and liver failure, often in the setting of heavy alcohol use for decades.^11,12 The incidence is unknown, but the typical age of presentation is between 40 and 50.^11,12 The chief sign is a rapid onset of jaundice (< 3 months); common signs and symptoms include fever, ascites, proximal muscle loss, and an enlarged, tender liver.¹² Encephalopathy may be seen in severe alcoholic hepatitis.¹²

Our patient is 35 years old. She has jaundice with rapid onset, as well as ascites and a tender liver.

The diagnosis of alcoholic hepatitis must take into account the patient’s history, physical examination, and laboratory findings. Until proven otherwise, the diagnosis should be presumed in the following scenario: ascites and jaundice on examination (usually with a duration < 3 months); a history of heavy alcohol use; neutrophilic leukocytosis; an AST level that is elevated but below 300 U/L; an ALT level above the normal range but below 300 U/L; an AST-ALT ratio greater than 2; a total serum bilirubin level above 5 mg/dL; and an elevated INR.^11,12 Liver biopsy is the gold standard for diagnosis. Though not routinely done because of risks associated with the procedure, it may help confirm the diagnosis if it is in question.

CASE CONCLUDED

We start our patient on oral prednisolone 40 mg daily for alcoholic hepatitis. Her symptoms and laboratory testing results including bilirubin improve. Her Lille score at 7 days indicates a good prognosis, prompting continuation of corticosteroid treatment for the full 28 days.

She is referred to an outpatient alcohol rehabilitation program and has remained sober as of the last outpatient note.

Alcoholic hepatitis is extremely difficult to diagnose, and no single blood test or imaging study confirms the diagnosis. The history, physical examination findings, and laboratory findings are crucial. If the diagnosis is still in doubt, liver biopsy may help confirm the diagnosis.

A 35-year-old woman is admitted to the hospital with a 5-day history of abdominal distention and jaundice. She reports no history of fever, chills, night sweats, abdominal pain, nausea, vomiting, diarrhea, changes in urine color, change in stool color, weight loss, weight gain, or loss of appetite.

She is petite, with a body mass index of 19.4 kg/m². She has no known history of medical conditions or surgery and is not taking any medications. Her family history is unremarkable, and she denies current or past tobacco, alcohol, or illicit drug use.

RECENT TRAVEL

She says that during a trip to Central America several months ago, she had suffered a seizure and was taken to a local hospital, where laboratory testing revealed elevated aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels. She says that the rest of the workup at that time was normal.

About 1 week after that incident, she returned home and saw her primary care physician, who ordered further testing, which showed mild hyperbilirubinemia and mild elevation of AST and ALT levels. Her physician attributed the elevations to atovaquone, which she had been taking for malaria prophylaxis, as repeat testing 2 weeks later showed improvement in AST and ALT levels.

The patient says she returned to her normal state of health until about 5 days ago, when she noticed jaundice and abdominal distention, but without abdominal pain, dark urine, or clay-colored stools. She became concerned and went to her local hospital. Testing there noted mild elevation of AST and ALT, as well as an elevated international normalized ratio (INR) and hyperbilirubinemia. Computed tomography of the abdomen and pelvis showed hepatomegaly with possible fatty liver. Because of these results, the patient was transferred to our institution for further evaluation.

EVALUATION AT OUR INSTITUTION

On examination at our institution, she is afebrile, and vital signs are within normal ranges. She has bilateral scleral icterus and diffuse jaundice, but no other skin finding such as rash or spider angioma. She has no lymphadenopathy. Her abdomen is distended, with tense ascites, and her liver is tender to palpation. The tip of the spleen is not palpable.

The cardiovascular examination reveals no murmurs, rubs, or gallops, but she has jugular venous distention and +2 pitting edema of both lower extremities.

On respiratory examination, there is dullness to percussion, with slight crackles on auscultation at the right lung base. The neurologic examination is normal.

Table 1 shows the results of initial laboratory testing.

1. Which study would provide the most information on the cause of ascites?

• Abdominal ultrasonography
• Abdominal paracentesis with ascitic fluid analysis
• Chest radiography
• Echocardiography
• Urine protein-to-creatinine ratio

Abdominal paracentesis with ascitic fluid analysis is the essential study for any patient with clinically apparent new-onset ascites.^1–3 It is the study that provides the most information on the cause of ascites.

In our patient, abdominal paracentesis yields 1,000 mL of straw-colored ascitic fluid, and analysis shows 86 nucleated cells, 28 of which are polymorphonuclear cells, and 0 red blood cells, with negative Gram stain and culture. The ascitic albumin level is 0.85 g/dL, with an ascitic protein of 1.1 g/dL.

Abdominal ultrasonography shows a diffusely echogenic liver, no focal lesions, moderate ascites, normal portal vein flow, no intrahepatic or extrahepatic biliary duct dilation, normal kidney sizes, no hydronephrosis, and no intra-abdominal mass. Chest radiography is clear with no sign of consolidation, edema, or effusion. Echocardiography shows a normal left ventricular ejection fraction with no valvular disease or pericardial effusion. A random urine protein-creatinine ratio is normal at 0.1 (reference range < 0.2).

2. What is the most likely cause of her ascites based on the workup to this point?

• Cirrhosis
• Heart failure
• Nephrotic syndrome
• Portal vein thrombus
• Abdominal malignancy
• Malaria

An initial approach to ascitic fluid analysis is to calculate the serum-ascites albumin gradient (SAAG). The SAAG is calculated as the serum albumin level minus the ascitic fluid albumin level.^4,5 This is useful in determining the cause of the ascites (Figure 1).^4,5 A gradient of 1.1 g/dL or higher indicates portal hypertension.^4,5

Common causes of portal hypertension include cirrhosis, alcoholic hepatitis, heart failure, vascular occlusion syndromes (eg, Budd-Chiari syndrome, portal vein thrombosis), idiopathic portal fibrosis, and metastatic liver disease.^5,6

If portal hypertension is present based on the SAAG, the next step is to review the ascitic protein level to help distinguish between a hepatic and a cardiac etiology of the ascites. An ascitic protein level less than 2.5 g/dL indicates a primary liver pathology (eg, cirrhosis). An ascitic protein level of 2.5 g/dL or greater typically indicates a cardiac condition (eg, heart failure, pericardial disease) with secondary congestive hepatopathy.^5,6

If the SAAG is less than 1.1 g/dL, the ascites is likely not from portal hypertension. Typical causes of a low SAAG include infection, malignancy, pancreatic ascites, and nephrotic syndrome.^5,6

In our patient, the SAAG is 1.35 g/dL (2.2 g/dL minus 0.85 g/dL), ie, elevated and due to portal hypertension. With an SAAG of 1.1 g/dL or greater and an ascitic fluid protein level less than 2.5 g/dL, as in our patient, the most likely cause is cirrhosis.

Heart failure is unlikely based on her normal brain natriuretic peptide level, an ascitic fluid protein level below 2.5 g/dL, and normal results on echocardiography. Nephrotic syndrome is also very unlikely based on the patient’s normal random urine protein-creatinine ratio. Portal vein thrombus and abdominal malignancy are essentially ruled out by the negative results of Doppler abdominal ultrasonography, with normal venous flow and no intra-abdominal mass and coupled with an elevated SAAG.

Although the patient has a history of travel, the incubation period for malaria would not fit the time frame of presentation. Also, she did not have typical malarial symptoms, her rapid malaria test was negative, and a peripheral blood smear for blood parasites was negative. It should be noted, however, that Plasmodium malariae infection classically presents with flulike symptoms and can resemble nephrotic syndrome, including peripheral edema, ascites, heavy proteinuria, hypoalbuminemia, and hyperlipidemia.⁷

3. In which patients is antibiotic prophylaxis against spontaneous bacterial peritonitis (SBP) appropriate?

• Any patient with cirrhosis
• Any patient with cirrhosis who is hospitalized
• Any patient with cirrhosis and an ascitic fluid protein level below 2.0 g/dL
• Any patient with cirrhosis and a history of SBP

Any patient with cirrhosis and a history of SBP should receive prophylactic antibiotics,⁸ as should any patient deemed at high risk of SBP. It is indicated in the following patients:

• Patients with cirrhosis and gastrointestinal bleeding^9,10
• Patients with cirrhosis and a previous episode of SBP⁸
• Patients with cirrhosis and an ascitic fluid protein level less than 1.5 g/dL with either impaired renal function (creatinine ≥ 1.2 mg/dL, blood urea nitrogen level ≥ 25 mg/dL, or serum sodium ≤ 130 mmol/L) or liver failure (Child-Pugh score ≥ 9 and a bilirubin ≥ 3 mg/dL)⁹
• Patients with cirrhosis who are hospitalized for other reasons and have an ascitic protein level < 1.0 g/dL.⁹

Our patient has no signs or symptoms of gastrointestinal bleeding and no history of SBP. Her ascitic fluid protein level is 1.1 g/dL, and she has normal renal function. However, her Child-Pugh score is 12 (3 points for total bilirubin > 3 mg/dL, 3 points for serum albumin < 2.8 g/dL, 2 points for an INR 1.7 to 2.2, 3 points for moderate ascites, and 1 point for no encephalopathy), with a bilirubin of 17.0 mg/dL. Based on this, she is placed on antibiotic prophylaxis for SBP.

Our patient then undergoes an extensive workup for liver disease. Results of tests for toxins, autoimmune diseases, and inheritable diseases are all within normal limits. At this point, despite the patient’s reported negative alcohol history, our leading diagnosis is alcoholic hepatitis.

To confirm this diagnosis, she subsequently undergoes transjugular liver biopsy, considered the gold standard for the diagnosis of alcoholic hepatitis. During the procedure, the hepatic venous pressure gradient is measured at 18 mm Hg (reference range 1–5 mm Hg), suggestive of portal hypertension. The pathology study shows severe fatty change, active steatohepatitis with ballooning degeneration, easily identifiable Mallory-Denk bodies, and prominent neutrophilic infiltration, as well as extensive bridging fibrosis (Figure 2). These findings point to alcoholic hepatitis.

After the biopsy results, we speak with the patient further about her alcohol habits. At this point, she informs us that she has consumed significant amounts of alcohol since the age of 18 (6 to 12 alcoholic beverages per day, including beer and hard liquor). Therefore, based on this new information, on her jaundice and ascites, and on results of laboratory testing and biopsy, we confirmed our diagnosis of alcoholic hepatitis.

4. When is drug treatment appropriate for alcoholic hepatitis?

• Model for End-stage Liver Disease (MELD) score greater than 12
• MELD score greater than 15
• Maddrey Discriminant Function score greater than 25
• Maddrey Discriminant Function score greater than 32
• Glasgow score greater than 5
• Glasgow score greater than 7

The best answer is a Maddrey Discriminant Function score greater than 32. A variety of scoring systems have been used to assess the severity of alcoholic hepatitis and to guide treatment, including the Maddrey Discriminant Function score, the MELD score, and the Glasgow score.^11–16 They share similar laboratory values in their calculations, including prothrombin time (or INR) and total bilirubin.^11–16 Typically, a Maddrey Discriminant Function score greater than 32, a Glasgow score of greater than 9, or a MELD score greater than 21 is used to determine whether pharmacologic treatment is indicated.^11–16

The typical treatment is prednisolone or pentoxifylline.^11,17–21 The Lille score is designed to help decide whether to stop corticosteroids after 1 week of administration due to lack of treatment response.²² It predicts mortality rates within 6 months; a score of 0.45 or less indicates a good prognosis, and corticosteroid therapy should continue for 28 days (Figure 3).²²

Our patient’s discriminant function score is 50, her Glasgow score is 10, and her MELD score is 28; thus, she begins treatment with oral prednisolone. Her Lille score at 1 week is 0.119, indicating a good prognosis, and her corticosteroids are continued for a total of 28 days.

It should be highlighted that the most important treatment is abstinence from alcohol.¹¹ Recent literature suggests that any benefit of prednisolone or pentoxifylline in terms of mortality rates is questionable,^19–20 and there is evidence that giving both drugs simultaneously may improve mortality rates,^11,21 but the evidence remains conflicting at this time.

ALCOHOLIC HEPATITIS

Alcoholic hepatitis is a clinical syndrome of jaundice and liver failure, often in the setting of heavy alcohol use for decades.^11,12 The incidence is unknown, but the typical age of presentation is between 40 and 50.^11,12 The chief sign is a rapid onset of jaundice (< 3 months); common signs and symptoms include fever, ascites, proximal muscle loss, and an enlarged, tender liver.¹² Encephalopathy may be seen in severe alcoholic hepatitis.¹²

Our patient is 35 years old. She has jaundice with rapid onset, as well as ascites and a tender liver.

The diagnosis of alcoholic hepatitis must take into account the patient’s history, physical examination, and laboratory findings. Until proven otherwise, the diagnosis should be presumed in the following scenario: ascites and jaundice on examination (usually with a duration < 3 months); a history of heavy alcohol use; neutrophilic leukocytosis; an AST level that is elevated but below 300 U/L; an ALT level above the normal range but below 300 U/L; an AST-ALT ratio greater than 2; a total serum bilirubin level above 5 mg/dL; and an elevated INR.^11,12 Liver biopsy is the gold standard for diagnosis. Though not routinely done because of risks associated with the procedure, it may help confirm the diagnosis if it is in question.

CASE CONCLUDED

We start our patient on oral prednisolone 40 mg daily for alcoholic hepatitis. Her symptoms and laboratory testing results including bilirubin improve. Her Lille score at 7 days indicates a good prognosis, prompting continuation of corticosteroid treatment for the full 28 days.

She is referred to an outpatient alcohol rehabilitation program and has remained sober as of the last outpatient note.

Alcoholic hepatitis is extremely difficult to diagnose, and no single blood test or imaging study confirms the diagnosis. The history, physical examination findings, and laboratory findings are crucial. If the diagnosis is still in doubt, liver biopsy may help confirm the diagnosis.

A 40-year-old woman with a history of hypertension, who was recently started on a diuretic, presents to the emergency department after a witnessed syncopal event. She reports a prodrome of lightheadedness, nausea, and darkening of her vision that occurred a few seconds after standing, followed by loss of consciousness. She had a complete, spontaneous recovery after 10 seconds, but upon arousal she noticed she had lost bladder control.

Her blood pressure is 120/80 mm Hg supine, 110/70 mm Hg sitting, and 90/60 mm Hg standing. She has no focal neurologic deficits. The cardiac examination is normal, without murmurs, and electrocardiography shows sinus tachycardia (heart rate 110 bpm) without other abnormalities. Results of laboratory testing are unremarkable.

Should you order neuroimaging to evaluate for syncope?

DEFINITIONS, CLASSIFICATIONS

Syncope is an abrupt loss of consciousness due to transient global cerebral hypoperfusion, with a concomitant loss of postural tone and rapid, spontaneous recovery.¹ Recovery from syncope is characterized by immediate restoration of orientation and normal behavior, although the period after recovery may be accompanied by fatigue.²

The European Society of Cardiology² has classified syncope into 3 main categories: reflex (neurally mediated) syncope, syncope due to orthostatic hypotension, and cardiac syncope. Determining the cause is critical, as this determines the prognosis.

KEYS TO THE EVALUATION

According to the 2017 American College of Cardiology/American Heart Association (ACC/AHA) and the 2009 European Society of Cardiology guidelines, the evaluation of syncope should include a thorough history, taken from the patient and witnesses, and a complete physical examination. This can identify the cause of syncope in up to 50% of cases and differentiate between cardiac and noncardiac causes. Features that point to cardiac syncope include age older than 60, male sex, known heart disease, brief prodrome, syncope during exertion or when supine, first syncopal event, family history of sudden cardiac death, and abnormal physical examination.¹

Features that suggest noncardiac syncope are young age; syncope only when standing; recurrent syncope; a prodrome of nausea, vomiting, and a warm sensation; and triggers such as dehydration, pain, distressful stimulus, cough, laugh micturition, defecation, and swallowing.¹

Electrocardiography should follow the history and physical examination. When done at presentation, electrocardiography is diagnostic in only about 5% of cases. However, given the importance of the diagnosis, it remains an essential part of the initial evaluation of syncope.³

If a clear cause of syncope is identified at this point, no further workup is needed, and the cause of syncope should be addressed.¹ If the cause is still unclear, the ACC/AHA guidelines recommend further evaluation based on the clinical presentation and risk stratification.

Routine use of additional testing is costly; tests should be ordered on the basis of their potential diagnostic and prognostic value. Additional evaluation should follow a stepwise approach and can include targeted blood work, autonomic nerve evaluation, tilt-table testing, transthoracic echocardiography, stress testing, electrocardiographic monitoring, and electrophysiologic testing.¹

Syncope is rarely a manifestation of neurologic disease, yet 11% to 58% of patients with a first episode of uncomplicated syncope undergo extensive neuroimaging with magnetic resonance imaging, computed tomography, electroencephalography (EEG), and carotid ultrasonography.⁴ Evidence suggests that routine neurologic testing is of limited value given its low diagnostic yield and high cost.

Epilepsy is the most common neurologic cause of loss of consciousness but is estimated to account for less than 5% of patients with syncope.⁵ A thorough and thoughtful neurologic history and examination is often enough to distinguish between syncope, convulsive syncope, epileptic convulsions, and pseudosyncope.

In syncope, the loss of consciousness usually occurs 30 seconds to several minutes after standing. It presents with or without a prodrome (warmth, palpitations, and diaphoresis) and can be relieved with supine positioning. True loss of consciousness usually lasts less than a minute and is accompanied by loss of postural tone, with little or no fatigue in the recovery period.⁶

Conversely, in convulsive syncope, the prodrome can include pallor and diaphoresis. Loss of consciousness lasts about 30 seconds but is accompanied by fixed gaze, upward eye deviation, nuchal rigidity, tonic spasms, myoclonic jerks, tonic-clonic convulsions, and oral automatisms.⁶

Pseudosyncope is characterized by a prodrome of lightheadedness, shortness of breath, chest pain, and tingling sensations, followed by episodes of apparent loss of consciousness that last longer than several minutes and occur multiple times a day. During these episodes, patients purposefully try to avoid trauma when they lose consciousness, and almost always keep their eyes closed, in contrast to syncopal episodes, when the eyes are open and glassy.⁷

ROLE OF ELECTROENCEPHALOGRAPHY

If the diagnosis remains unclear after the history and neurologic examination, EEG is recommended (class IIa, ie, reasonable, can be useful) during tilt-table testing, as it can help differentiate syncope, pseudosyncope, and epilepsy.¹

In an epileptic convulsion, EEG shows epileptiform discharges, whereas in syncope, it shows diffuse brainwave slowing with delta waves and a flatline pattern. In pseudosyncope and psychogenic nonepileptic seizures, EEG shows normal activity.⁸

Routine EEG is not recommended if there are no specific neurologic signs of epilepsy or if the history and neurologic examination indicate syncope or pseudosyncope.¹

Structural brain disease does not typically present with transient global cerebral hypoperfusion resulting in syncope, so magnetic resonance imaging and computed tomography have a low diagnostic yield. Studies have revealed that for the 11% to 58% of patients who undergo neuroimaging, it establishes a diagnosis in only 0.2% to 1%.⁹ For this reason and in view of their high cost, these imaging tests should not be routinely ordered in the evaluation of syncope.^4,10 Similarly, carotid artery imaging should not be routinely ordered if there is no focal neurologic finding suggesting unilateral ischemia.¹⁰

CASE CONTINUED

In our 40-year-old patient, the history suggests dehydration, as she recently started taking a diuretic. Thus, laboratory testing is reasonable.

Loss of bladder control is often interpreted as a red flag for neurologic disease, but syncope can often present with urinary incontinence. Urinary incontinence may also occur in epileptic seizure and in nonepileptic events such as syncope. A pooled analysis by Brigo et al¹¹ determined that urinary incontinence had no value in distinguishing between epilepsy and syncope. Therefore, this physical finding should not incline the clinician to one diagnosis or the other.


Given our patient’s presentation, findings on physical examination, and absence of focal neurologic deficits, she should not undergo neuroimaging for syncope evaluation. The more likely cause of her syncope is orthostatic intolerance (orthostatic hypotension or vasovagal syncope) in the setting of intravascular volume depletion, likely secondary to diuretic use. Obtaining orthostatic vital signs is mandatory, and this confirms the diagnosis.

A 40-year-old woman with a history of hypertension, who was recently started on a diuretic, presents to the emergency department after a witnessed syncopal event. She reports a prodrome of lightheadedness, nausea, and darkening of her vision that occurred a few seconds after standing, followed by loss of consciousness. She had a complete, spontaneous recovery after 10 seconds, but upon arousal she noticed she had lost bladder control.

Her blood pressure is 120/80 mm Hg supine, 110/70 mm Hg sitting, and 90/60 mm Hg standing. She has no focal neurologic deficits. The cardiac examination is normal, without murmurs, and electrocardiography shows sinus tachycardia (heart rate 110 bpm) without other abnormalities. Results of laboratory testing are unremarkable.

Should you order neuroimaging to evaluate for syncope?

DEFINITIONS, CLASSIFICATIONS

Syncope is an abrupt loss of consciousness due to transient global cerebral hypoperfusion, with a concomitant loss of postural tone and rapid, spontaneous recovery.¹ Recovery from syncope is characterized by immediate restoration of orientation and normal behavior, although the period after recovery may be accompanied by fatigue.²

The European Society of Cardiology² has classified syncope into 3 main categories: reflex (neurally mediated) syncope, syncope due to orthostatic hypotension, and cardiac syncope. Determining the cause is critical, as this determines the prognosis.

KEYS TO THE EVALUATION

According to the 2017 American College of Cardiology/American Heart Association (ACC/AHA) and the 2009 European Society of Cardiology guidelines, the evaluation of syncope should include a thorough history, taken from the patient and witnesses, and a complete physical examination. This can identify the cause of syncope in up to 50% of cases and differentiate between cardiac and noncardiac causes. Features that point to cardiac syncope include age older than 60, male sex, known heart disease, brief prodrome, syncope during exertion or when supine, first syncopal event, family history of sudden cardiac death, and abnormal physical examination.¹

Features that suggest noncardiac syncope are young age; syncope only when standing; recurrent syncope; a prodrome of nausea, vomiting, and a warm sensation; and triggers such as dehydration, pain, distressful stimulus, cough, laugh micturition, defecation, and swallowing.¹

Electrocardiography should follow the history and physical examination. When done at presentation, electrocardiography is diagnostic in only about 5% of cases. However, given the importance of the diagnosis, it remains an essential part of the initial evaluation of syncope.³

If a clear cause of syncope is identified at this point, no further workup is needed, and the cause of syncope should be addressed.¹ If the cause is still unclear, the ACC/AHA guidelines recommend further evaluation based on the clinical presentation and risk stratification.

Routine use of additional testing is costly; tests should be ordered on the basis of their potential diagnostic and prognostic value. Additional evaluation should follow a stepwise approach and can include targeted blood work, autonomic nerve evaluation, tilt-table testing, transthoracic echocardiography, stress testing, electrocardiographic monitoring, and electrophysiologic testing.¹

Syncope is rarely a manifestation of neurologic disease, yet 11% to 58% of patients with a first episode of uncomplicated syncope undergo extensive neuroimaging with magnetic resonance imaging, computed tomography, electroencephalography (EEG), and carotid ultrasonography.⁴ Evidence suggests that routine neurologic testing is of limited value given its low diagnostic yield and high cost.

Epilepsy is the most common neurologic cause of loss of consciousness but is estimated to account for less than 5% of patients with syncope.⁵ A thorough and thoughtful neurologic history and examination is often enough to distinguish between syncope, convulsive syncope, epileptic convulsions, and pseudosyncope.

In syncope, the loss of consciousness usually occurs 30 seconds to several minutes after standing. It presents with or without a prodrome (warmth, palpitations, and diaphoresis) and can be relieved with supine positioning. True loss of consciousness usually lasts less than a minute and is accompanied by loss of postural tone, with little or no fatigue in the recovery period.⁶

Conversely, in convulsive syncope, the prodrome can include pallor and diaphoresis. Loss of consciousness lasts about 30 seconds but is accompanied by fixed gaze, upward eye deviation, nuchal rigidity, tonic spasms, myoclonic jerks, tonic-clonic convulsions, and oral automatisms.⁶

Pseudosyncope is characterized by a prodrome of lightheadedness, shortness of breath, chest pain, and tingling sensations, followed by episodes of apparent loss of consciousness that last longer than several minutes and occur multiple times a day. During these episodes, patients purposefully try to avoid trauma when they lose consciousness, and almost always keep their eyes closed, in contrast to syncopal episodes, when the eyes are open and glassy.⁷

ROLE OF ELECTROENCEPHALOGRAPHY

If the diagnosis remains unclear after the history and neurologic examination, EEG is recommended (class IIa, ie, reasonable, can be useful) during tilt-table testing, as it can help differentiate syncope, pseudosyncope, and epilepsy.¹

In an epileptic convulsion, EEG shows epileptiform discharges, whereas in syncope, it shows diffuse brainwave slowing with delta waves and a flatline pattern. In pseudosyncope and psychogenic nonepileptic seizures, EEG shows normal activity.⁸

Routine EEG is not recommended if there are no specific neurologic signs of epilepsy or if the history and neurologic examination indicate syncope or pseudosyncope.¹

Structural brain disease does not typically present with transient global cerebral hypoperfusion resulting in syncope, so magnetic resonance imaging and computed tomography have a low diagnostic yield. Studies have revealed that for the 11% to 58% of patients who undergo neuroimaging, it establishes a diagnosis in only 0.2% to 1%.⁹ For this reason and in view of their high cost, these imaging tests should not be routinely ordered in the evaluation of syncope.^4,10 Similarly, carotid artery imaging should not be routinely ordered if there is no focal neurologic finding suggesting unilateral ischemia.¹⁰

CASE CONTINUED

In our 40-year-old patient, the history suggests dehydration, as she recently started taking a diuretic. Thus, laboratory testing is reasonable.

Loss of bladder control is often interpreted as a red flag for neurologic disease, but syncope can often present with urinary incontinence. Urinary incontinence may also occur in epileptic seizure and in nonepileptic events such as syncope. A pooled analysis by Brigo et al¹¹ determined that urinary incontinence had no value in distinguishing between epilepsy and syncope. Therefore, this physical finding should not incline the clinician to one diagnosis or the other.


Given our patient’s presentation, findings on physical examination, and absence of focal neurologic deficits, she should not undergo neuroimaging for syncope evaluation. The more likely cause of her syncope is orthostatic intolerance (orthostatic hypotension or vasovagal syncope) in the setting of intravascular volume depletion, likely secondary to diuretic use. Obtaining orthostatic vital signs is mandatory, and this confirms the diagnosis.

A 40-year-old woman with a history of hypertension, who was recently started on a diuretic, presents to the emergency department after a witnessed syncopal event. She reports a prodrome of lightheadedness, nausea, and darkening of her vision that occurred a few seconds after standing, followed by loss of consciousness. She had a complete, spontaneous recovery after 10 seconds, but upon arousal she noticed she had lost bladder control.

Her blood pressure is 120/80 mm Hg supine, 110/70 mm Hg sitting, and 90/60 mm Hg standing. She has no focal neurologic deficits. The cardiac examination is normal, without murmurs, and electrocardiography shows sinus tachycardia (heart rate 110 bpm) without other abnormalities. Results of laboratory testing are unremarkable.

Should you order neuroimaging to evaluate for syncope?

DEFINITIONS, CLASSIFICATIONS

Syncope is an abrupt loss of consciousness due to transient global cerebral hypoperfusion, with a concomitant loss of postural tone and rapid, spontaneous recovery.¹ Recovery from syncope is characterized by immediate restoration of orientation and normal behavior, although the period after recovery may be accompanied by fatigue.²

The European Society of Cardiology² has classified syncope into 3 main categories: reflex (neurally mediated) syncope, syncope due to orthostatic hypotension, and cardiac syncope. Determining the cause is critical, as this determines the prognosis.

KEYS TO THE EVALUATION

According to the 2017 American College of Cardiology/American Heart Association (ACC/AHA) and the 2009 European Society of Cardiology guidelines, the evaluation of syncope should include a thorough history, taken from the patient and witnesses, and a complete physical examination. This can identify the cause of syncope in up to 50% of cases and differentiate between cardiac and noncardiac causes. Features that point to cardiac syncope include age older than 60, male sex, known heart disease, brief prodrome, syncope during exertion or when supine, first syncopal event, family history of sudden cardiac death, and abnormal physical examination.¹

Features that suggest noncardiac syncope are young age; syncope only when standing; recurrent syncope; a prodrome of nausea, vomiting, and a warm sensation; and triggers such as dehydration, pain, distressful stimulus, cough, laugh micturition, defecation, and swallowing.¹

Electrocardiography should follow the history and physical examination. When done at presentation, electrocardiography is diagnostic in only about 5% of cases. However, given the importance of the diagnosis, it remains an essential part of the initial evaluation of syncope.³

If a clear cause of syncope is identified at this point, no further workup is needed, and the cause of syncope should be addressed.¹ If the cause is still unclear, the ACC/AHA guidelines recommend further evaluation based on the clinical presentation and risk stratification.

Routine use of additional testing is costly; tests should be ordered on the basis of their potential diagnostic and prognostic value. Additional evaluation should follow a stepwise approach and can include targeted blood work, autonomic nerve evaluation, tilt-table testing, transthoracic echocardiography, stress testing, electrocardiographic monitoring, and electrophysiologic testing.¹

Syncope is rarely a manifestation of neurologic disease, yet 11% to 58% of patients with a first episode of uncomplicated syncope undergo extensive neuroimaging with magnetic resonance imaging, computed tomography, electroencephalography (EEG), and carotid ultrasonography.⁴ Evidence suggests that routine neurologic testing is of limited value given its low diagnostic yield and high cost.

Epilepsy is the most common neurologic cause of loss of consciousness but is estimated to account for less than 5% of patients with syncope.⁵ A thorough and thoughtful neurologic history and examination is often enough to distinguish between syncope, convulsive syncope, epileptic convulsions, and pseudosyncope.

In syncope, the loss of consciousness usually occurs 30 seconds to several minutes after standing. It presents with or without a prodrome (warmth, palpitations, and diaphoresis) and can be relieved with supine positioning. True loss of consciousness usually lasts less than a minute and is accompanied by loss of postural tone, with little or no fatigue in the recovery period.⁶

Conversely, in convulsive syncope, the prodrome can include pallor and diaphoresis. Loss of consciousness lasts about 30 seconds but is accompanied by fixed gaze, upward eye deviation, nuchal rigidity, tonic spasms, myoclonic jerks, tonic-clonic convulsions, and oral automatisms.⁶

Pseudosyncope is characterized by a prodrome of lightheadedness, shortness of breath, chest pain, and tingling sensations, followed by episodes of apparent loss of consciousness that last longer than several minutes and occur multiple times a day. During these episodes, patients purposefully try to avoid trauma when they lose consciousness, and almost always keep their eyes closed, in contrast to syncopal episodes, when the eyes are open and glassy.⁷

ROLE OF ELECTROENCEPHALOGRAPHY

If the diagnosis remains unclear after the history and neurologic examination, EEG is recommended (class IIa, ie, reasonable, can be useful) during tilt-table testing, as it can help differentiate syncope, pseudosyncope, and epilepsy.¹

In an epileptic convulsion, EEG shows epileptiform discharges, whereas in syncope, it shows diffuse brainwave slowing with delta waves and a flatline pattern. In pseudosyncope and psychogenic nonepileptic seizures, EEG shows normal activity.⁸

Routine EEG is not recommended if there are no specific neurologic signs of epilepsy or if the history and neurologic examination indicate syncope or pseudosyncope.¹

Structural brain disease does not typically present with transient global cerebral hypoperfusion resulting in syncope, so magnetic resonance imaging and computed tomography have a low diagnostic yield. Studies have revealed that for the 11% to 58% of patients who undergo neuroimaging, it establishes a diagnosis in only 0.2% to 1%.⁹ For this reason and in view of their high cost, these imaging tests should not be routinely ordered in the evaluation of syncope.^4,10 Similarly, carotid artery imaging should not be routinely ordered if there is no focal neurologic finding suggesting unilateral ischemia.¹⁰

CASE CONTINUED

In our 40-year-old patient, the history suggests dehydration, as she recently started taking a diuretic. Thus, laboratory testing is reasonable.

Loss of bladder control is often interpreted as a red flag for neurologic disease, but syncope can often present with urinary incontinence. Urinary incontinence may also occur in epileptic seizure and in nonepileptic events such as syncope. A pooled analysis by Brigo et al¹¹ determined that urinary incontinence had no value in distinguishing between epilepsy and syncope. Therefore, this physical finding should not incline the clinician to one diagnosis or the other.


Given our patient’s presentation, findings on physical examination, and absence of focal neurologic deficits, she should not undergo neuroimaging for syncope evaluation. The more likely cause of her syncope is orthostatic intolerance (orthostatic hypotension or vasovagal syncope) in the setting of intravascular volume depletion, likely secondary to diuretic use. Obtaining orthostatic vital signs is mandatory, and this confirms the diagnosis.

Reports of cancer date back thousands of years to Egyptian texts. Its existence baffled scientists until the 1950s, when Watson, Crick, and Franklin discovered the structure of DNA, laying the groundwork for identifying the genetic pathways leading to cancer. Currently, cancer is a leading global cause of death and the second leading cause of death in the United States.^1,2

In an effort to curtail cancer and its related morbidity and mortality, population-based screening programs have been implemented with tests that identify precancerous lesions and, preferably, early-stage rather than late-stage cancer.

Screening for cancer can lead to early diagnosis and prevent death from cancer, but the topic continues to provoke controversy.

VALUE OF SCREENING QUESTIONED

In a commentary in the March 2019 Cleveland Clinic Journal of Medicine, Kim et al³ argued that cancer screening is not very effective and that we need to find the balance between the potential benefit and harm.

Using data from the US Preventive Services Task Force (USPSTF) and various studies, the authors showed that although screening can prevent some deaths from breast, colon, prostate, and lung cancer, at least 3 times as many people who are screened still die of those diseases. Given that screening does not eliminate all cancer deaths, has not been definitely shown to decrease the all-cause mortality rate, and has the potential to harm through false-positive results, overdiagnosis, and overtreatment, the authors questioned the utility of screening and encouraged us to discuss the benefits and harms with our patients.

In view of the apparently meager benefit, the USPSTF has relaxed its recommendations for screening for breast and prostate cancer in average-risk populations in recent years, a move that has evoked strong reactions from some clinicians. Proponents of screening argue that preventing late-stage cancers can save money, as the direct and indirect costs of morbidity associated with late-stage cancers are substantial, and that patients prefer screening when a test is available. Current models of screening efficacy do not take these factors into account.⁴

Kim et al, in defending the USPSTF’s position, suggested that the motivation for aggressive testing may be a belief that no harm is greater than the benefit of saving a life. They illustrated this through a Swiftian “modest proposal,” ie, universal prophylactic organectomy to prevent cancer. This hypothetical extreme measure would nearly eliminate the risk of cancer in the removed organs and prevent overdiagnosis and overtreatment of malignancies, but at substantial harm and cost.

In response to this proposal, we would like to point out the alternative extreme: stop all cancer screening programs. The pendulum would swing from what was previously considered a benefit—cancer prevention—to a harm, ie, cancer.

Observational studies, systematic reviews, meta-analyses, and modeling studies show that screening for cervical, colorectal, breast, and prostate cancer decreases disease-specific mortality.^5–11

For example, in lung cancer, the National Lung Screening Trial demonstrated reductions in disease-specific and overall mortality in patients at high risk who underwent low-dose screening computed tomography.¹²

In breast cancer, a systematic review demonstrated decreased disease-specific mortality for women ages 50 through 79 who underwent screening mammography.¹³

In cervical cancer, lower rates of cancer-related death and invasive cancer have also been shown with screening.¹⁴

In colorectal cancer, great strides have been made in reducing both the incidence of and mortality from this disease over the past 30 years through fecal occult blood testing. Early detection shifts the 5-year survival rate—14% for late-stage cancer—to over 90%.¹⁵ Colorectal cancer screening has also been shown to be cost-effective, with savings in excess of $30,000 per life-year gained from screening.¹⁶

Moreover, recent data from the Prostate, Lung, Colorectal, and Ovarian Cancer (PLCO) screening trial¹⁷ demonstrated a 2-fold higher overall non-cancer-related mortality rate in participants who did not adhere to screening compared with those who were fully adherent to all sex-specific PLCO screening tests when adjusted for age, sex, and ethnicity. Although a possible explanation is that people who adhere to screening recommendations are also likely to have a healthier lifestyle overall, the association persisted (although it was slightly attenuated) even after adjusting for medical risk and behavioral factors.

ON THIS WE CAN AGREE

Like Kim et al, we also believe an informed discussion of screening should occur with each patient—and challenge Kim et al to design an efficient and practical approach to allow providers to do so in a busy office visit aimed to address and manage other competing diseases.

In addition, medical science needs to improve. Methods to increase the efficacy of screening and decrease risks should be explored; these include improving test and operator performance, reducing nonadherence to screening, investigating novel biomarkers or precursors of cancer and pathways that escape current detection, and devising better risk-stratification tools.

Bodies such as the USPSTF should use models that account for factors not considered previously but important when informing patients of potential benefits and harm. Examples include varying sensitivities and specificities at different rounds of testing and accounting for the variability in risk or efficacy affected by race, ethnicity, sex, and patient preferences.

We practice in the era of evidence-based medicine. Guidelines and recommendations are based on the available evidence. As more studies are published, disease mechanisms are better understood, and the effects of previous recommendations are evaluated, cancer screening programs will be further refined or replaced. The balance between benefit and harm will be further delineated.

Kim et al knocked on the door of personalized medicine, where individual screening will be based on individual risk. Until that door is opened, screening should be personalized through the risk-benefit discussions we have with our patients. Ultimately, the choice to undergo screening is the patient’s.

Reports of cancer date back thousands of years to Egyptian texts. Its existence baffled scientists until the 1950s, when Watson, Crick, and Franklin discovered the structure of DNA, laying the groundwork for identifying the genetic pathways leading to cancer. Currently, cancer is a leading global cause of death and the second leading cause of death in the United States.^1,2

In an effort to curtail cancer and its related morbidity and mortality, population-based screening programs have been implemented with tests that identify precancerous lesions and, preferably, early-stage rather than late-stage cancer.

Screening for cancer can lead to early diagnosis and prevent death from cancer, but the topic continues to provoke controversy.

VALUE OF SCREENING QUESTIONED

In a commentary in the March 2019 Cleveland Clinic Journal of Medicine, Kim et al³ argued that cancer screening is not very effective and that we need to find the balance between the potential benefit and harm.

Using data from the US Preventive Services Task Force (USPSTF) and various studies, the authors showed that although screening can prevent some deaths from breast, colon, prostate, and lung cancer, at least 3 times as many people who are screened still die of those diseases. Given that screening does not eliminate all cancer deaths, has not been definitely shown to decrease the all-cause mortality rate, and has the potential to harm through false-positive results, overdiagnosis, and overtreatment, the authors questioned the utility of screening and encouraged us to discuss the benefits and harms with our patients.

In view of the apparently meager benefit, the USPSTF has relaxed its recommendations for screening for breast and prostate cancer in average-risk populations in recent years, a move that has evoked strong reactions from some clinicians. Proponents of screening argue that preventing late-stage cancers can save money, as the direct and indirect costs of morbidity associated with late-stage cancers are substantial, and that patients prefer screening when a test is available. Current models of screening efficacy do not take these factors into account.⁴

Kim et al, in defending the USPSTF’s position, suggested that the motivation for aggressive testing may be a belief that no harm is greater than the benefit of saving a life. They illustrated this through a Swiftian “modest proposal,” ie, universal prophylactic organectomy to prevent cancer. This hypothetical extreme measure would nearly eliminate the risk of cancer in the removed organs and prevent overdiagnosis and overtreatment of malignancies, but at substantial harm and cost.

In response to this proposal, we would like to point out the alternative extreme: stop all cancer screening programs. The pendulum would swing from what was previously considered a benefit—cancer prevention—to a harm, ie, cancer.

Observational studies, systematic reviews, meta-analyses, and modeling studies show that screening for cervical, colorectal, breast, and prostate cancer decreases disease-specific mortality.^5–11

For example, in lung cancer, the National Lung Screening Trial demonstrated reductions in disease-specific and overall mortality in patients at high risk who underwent low-dose screening computed tomography.¹²

In breast cancer, a systematic review demonstrated decreased disease-specific mortality for women ages 50 through 79 who underwent screening mammography.¹³

In cervical cancer, lower rates of cancer-related death and invasive cancer have also been shown with screening.¹⁴

In colorectal cancer, great strides have been made in reducing both the incidence of and mortality from this disease over the past 30 years through fecal occult blood testing. Early detection shifts the 5-year survival rate—14% for late-stage cancer—to over 90%.¹⁵ Colorectal cancer screening has also been shown to be cost-effective, with savings in excess of $30,000 per life-year gained from screening.¹⁶

Moreover, recent data from the Prostate, Lung, Colorectal, and Ovarian Cancer (PLCO) screening trial¹⁷ demonstrated a 2-fold higher overall non-cancer-related mortality rate in participants who did not adhere to screening compared with those who were fully adherent to all sex-specific PLCO screening tests when adjusted for age, sex, and ethnicity. Although a possible explanation is that people who adhere to screening recommendations are also likely to have a healthier lifestyle overall, the association persisted (although it was slightly attenuated) even after adjusting for medical risk and behavioral factors.

ON THIS WE CAN AGREE

Like Kim et al, we also believe an informed discussion of screening should occur with each patient—and challenge Kim et al to design an efficient and practical approach to allow providers to do so in a busy office visit aimed to address and manage other competing diseases.

In addition, medical science needs to improve. Methods to increase the efficacy of screening and decrease risks should be explored; these include improving test and operator performance, reducing nonadherence to screening, investigating novel biomarkers or precursors of cancer and pathways that escape current detection, and devising better risk-stratification tools.

Bodies such as the USPSTF should use models that account for factors not considered previously but important when informing patients of potential benefits and harm. Examples include varying sensitivities and specificities at different rounds of testing and accounting for the variability in risk or efficacy affected by race, ethnicity, sex, and patient preferences.

We practice in the era of evidence-based medicine. Guidelines and recommendations are based on the available evidence. As more studies are published, disease mechanisms are better understood, and the effects of previous recommendations are evaluated, cancer screening programs will be further refined or replaced. The balance between benefit and harm will be further delineated.

Kim et al knocked on the door of personalized medicine, where individual screening will be based on individual risk. Until that door is opened, screening should be personalized through the risk-benefit discussions we have with our patients. Ultimately, the choice to undergo screening is the patient’s.

Reports of cancer date back thousands of years to Egyptian texts. Its existence baffled scientists until the 1950s, when Watson, Crick, and Franklin discovered the structure of DNA, laying the groundwork for identifying the genetic pathways leading to cancer. Currently, cancer is a leading global cause of death and the second leading cause of death in the United States.^1,2

In an effort to curtail cancer and its related morbidity and mortality, population-based screening programs have been implemented with tests that identify precancerous lesions and, preferably, early-stage rather than late-stage cancer.

Screening for cancer can lead to early diagnosis and prevent death from cancer, but the topic continues to provoke controversy.

VALUE OF SCREENING QUESTIONED

In a commentary in the March 2019 Cleveland Clinic Journal of Medicine, Kim et al³ argued that cancer screening is not very effective and that we need to find the balance between the potential benefit and harm.

Using data from the US Preventive Services Task Force (USPSTF) and various studies, the authors showed that although screening can prevent some deaths from breast, colon, prostate, and lung cancer, at least 3 times as many people who are screened still die of those diseases. Given that screening does not eliminate all cancer deaths, has not been definitely shown to decrease the all-cause mortality rate, and has the potential to harm through false-positive results, overdiagnosis, and overtreatment, the authors questioned the utility of screening and encouraged us to discuss the benefits and harms with our patients.

In view of the apparently meager benefit, the USPSTF has relaxed its recommendations for screening for breast and prostate cancer in average-risk populations in recent years, a move that has evoked strong reactions from some clinicians. Proponents of screening argue that preventing late-stage cancers can save money, as the direct and indirect costs of morbidity associated with late-stage cancers are substantial, and that patients prefer screening when a test is available. Current models of screening efficacy do not take these factors into account.⁴

Kim et al, in defending the USPSTF’s position, suggested that the motivation for aggressive testing may be a belief that no harm is greater than the benefit of saving a life. They illustrated this through a Swiftian “modest proposal,” ie, universal prophylactic organectomy to prevent cancer. This hypothetical extreme measure would nearly eliminate the risk of cancer in the removed organs and prevent overdiagnosis and overtreatment of malignancies, but at substantial harm and cost.

In response to this proposal, we would like to point out the alternative extreme: stop all cancer screening programs. The pendulum would swing from what was previously considered a benefit—cancer prevention—to a harm, ie, cancer.

Observational studies, systematic reviews, meta-analyses, and modeling studies show that screening for cervical, colorectal, breast, and prostate cancer decreases disease-specific mortality.^5–11

For example, in lung cancer, the National Lung Screening Trial demonstrated reductions in disease-specific and overall mortality in patients at high risk who underwent low-dose screening computed tomography.¹²

In breast cancer, a systematic review demonstrated decreased disease-specific mortality for women ages 50 through 79 who underwent screening mammography.¹³

In cervical cancer, lower rates of cancer-related death and invasive cancer have also been shown with screening.¹⁴

In colorectal cancer, great strides have been made in reducing both the incidence of and mortality from this disease over the past 30 years through fecal occult blood testing. Early detection shifts the 5-year survival rate—14% for late-stage cancer—to over 90%.¹⁵ Colorectal cancer screening has also been shown to be cost-effective, with savings in excess of $30,000 per life-year gained from screening.¹⁶

Moreover, recent data from the Prostate, Lung, Colorectal, and Ovarian Cancer (PLCO) screening trial¹⁷ demonstrated a 2-fold higher overall non-cancer-related mortality rate in participants who did not adhere to screening compared with those who were fully adherent to all sex-specific PLCO screening tests when adjusted for age, sex, and ethnicity. Although a possible explanation is that people who adhere to screening recommendations are also likely to have a healthier lifestyle overall, the association persisted (although it was slightly attenuated) even after adjusting for medical risk and behavioral factors.

ON THIS WE CAN AGREE

Like Kim et al, we also believe an informed discussion of screening should occur with each patient—and challenge Kim et al to design an efficient and practical approach to allow providers to do so in a busy office visit aimed to address and manage other competing diseases.

In addition, medical science needs to improve. Methods to increase the efficacy of screening and decrease risks should be explored; these include improving test and operator performance, reducing nonadherence to screening, investigating novel biomarkers or precursors of cancer and pathways that escape current detection, and devising better risk-stratification tools.

Bodies such as the USPSTF should use models that account for factors not considered previously but important when informing patients of potential benefits and harm. Examples include varying sensitivities and specificities at different rounds of testing and accounting for the variability in risk or efficacy affected by race, ethnicity, sex, and patient preferences.

We practice in the era of evidence-based medicine. Guidelines and recommendations are based on the available evidence. As more studies are published, disease mechanisms are better understood, and the effects of previous recommendations are evaluated, cancer screening programs will be further refined or replaced. The balance between benefit and harm will be further delineated.

Kim et al knocked on the door of personalized medicine, where individual screening will be based on individual risk. Until that door is opened, screening should be personalized through the risk-benefit discussions we have with our patients. Ultimately, the choice to undergo screening is the patient’s.

“Twin berries on one stem, grievous damage has been done to both in regarding the Humanities and Science in any other light than complemental.”
—Sir William Osler¹

The year 2019 marks the 100th anniversary of Sir William Osler’s last public speech. Still reeling from the death of his only son in World War I, he had been asked to give the presidential inaugural address of the Classical Association at Oxford. It was the first time a physician had received the honor, and Osler took the assignment very seriously. He chose to speak about “The old humanities and the new science,” and to call for a reunification of the two fields. “Humanists have not enough Science” he warned, “and Science sadly lacks the Humanities…this unhappy divorce…should never have taken place.”¹  Later, he said that it was the speech to which he had given the greatest thought and preparation. It was in fact Osler’s personal legacy: 2 months later he turned 70, and 7 months later he was dead.

Revisiting the address today, what can Osler teach the high-tech physician of today, when doctors have become “providers” and patients “consumers”? Is Osler’s message still relevant to our craft, or has he simply become an icon of professional nostalgia with little value for our times?

THE NEED FOR THE HUMANITIES IN MEDICINE

Medicine has certainly grown both powerful and successful. Yet it is also confronting hurdles that would have been unimaginable in Osler’s time. Physicians are now the professionals with the highest suicide rate,² a burnout rate as high as 70%,^3,4 rampant depression,⁵ dwindling empathy,⁶ a predominantly negative perception by the public,^7,8 and a disturbing propensity to quit.⁹ These, of course, may just be symptoms of an increasingly meaningless environment wherein doctors have become small cogs in a medical-industrial complex they can’t control or even understand. Still, is it possible that something more personal may have been lost in the way we now select and educate physicians? Could this, in turn, make us less resilient?

In this regard, Osler’s last public speech serves as an enduring reminder of the need for the humanities in medicine. Osler not only believed it, but throughout his life never missed a chance to express in words, writings, and deeds that the humanities are indeed “the hormones” of the profession. In 1919 he warned against the risk of separating our humanistic tradition from the sciences, and urged us “to infect [anyone] with the spirit of the Humanities,” since to him that was “the greatest single gift in education.”¹

Unfortunately, the humanities are slippery, not easily quantifiable, hard to define, and seemingly incompatible with an evidence-based approach. Quite understandably, today’s data-obsessed medicine views them with suspicion. But besides reminding us that in medicine not all that counts can be counted, and not all that can be counted counts, the humanities are in fact a fundamental component of the physician’s skill set.

In a multicenter survey of 5 medical schools,¹⁰ there was indeed a correlation between students’ exposure to the humanities and many of the personal qualities whose absence we lament in today’s medicine: empathy, tolerance for ambiguity, emotional intelligence, and prevention of burnout. Most significant was a strong correlation with wisdom, as measured by the 21-item Brief Wisdom Screening Scale.¹¹ That all these traits may correlate with humanities exposure is intuitive, since the humanities not only teach tolerance and compassion, but also capture the collective experience of those who came before us. Hence, they teach us wisdom. Wisdom is not an ACGME competency, but it’s undoubtedly a prerequisite for the art of healing.¹² In fact, wisdom may very well be the fundamental trait that characterizes a well-rounded physician, since it encompasses empathy, resilience, comfort with ambiguity, and the capacity to learn from the past. Not surprisingly, wisdom in the world was Osler’s closing wish in 1919.

The humanities can also nurture the very personal qualities we desire in physicians. For example, observing drama fosters empathy,¹³ as does taking an elective in medical humanities.¹⁴ Drawing enhances the reading of faces,¹⁵ and observing art improves the art of clinical observation.¹⁶ Reading good literature prompts better detection of emotions,¹⁷ and reflective writing improves students’ well-being.¹⁸ Playing a musical instrument reduces burnout.¹⁹ And an undergraduate major in the humanities correlates with greater tolerance for ambiguity,²⁰ a highly desirable trait in physicians, since it means openness to new ideas and the capacity to better cope with difficult situations.²¹

In fact, some of the qualities fostered by the humanities even translate into better patient care. For instance, tolerance for ambiguity correlates with more positive attitudes towards patients who have frustrating complaints,²² with lower use of resources,²³ and with a career choice in direct patient care.²⁴ Hence, it has been suggested that it should be a prerequisite for medical school admission.²⁵ Physicians’ empathy is also beneficial, since it correlates with a lower rate of complications and better outcomes in the care of diabetic patients.²⁶ This should not come as a surprise. As Hippocrates put it 2,500 years ago, “some patients, though conscious that their condition is perilous, recover their health simply through their contentment with the goodness of the physician.”²⁷

Lastly, studying the humanities may provide crucial antibodies against the pain and suffering that are unavoidable staples of the human condition. To paraphrase Osler, the humanities might vaccinate us against the difficulties of our profession. Hippocrates himself had suggested that “it is well to superintend the sick to make them well, to care for the healthy to keep them well, but also to care for one’s self…”²⁷ That is why many institutions now require medical students to take humanities courses.²⁸

Yet this effort may amount to a rearguard action that arrives too late and provides too little. The humanities should probably be taught before medical school.²⁹ After all, if it’s possible to make a scientist out of a humanist (Osler was living proof), the experience of the past decades seems to suggest that it’s considerably harder to make a humanist out of a scientist—hence the need to revisit undergraduate curricula and admission criteria to medical school, so that students can receive an adequate foundation in both arenas. Ironically, students express positive attitudes toward a liberal education and think it would actually help them as physicians.³⁰ Yet they also understand that the selection process remains tilted towards the sciences.^30–32

For Osler, scientific evidence was important but not a substitute for a humanistic approach. As he reminded students, “The practice of medicine is an art based on science,”³³ whose main goals are to prevent disease, relieve suffering, and heal the sick. To do so, one ought to care more “for the individual patient than for the special features of the disease.”³⁴ But he warned them, “It is much harder to acquire the art than the science.”³⁵ In fact, “The practice of medicine is a calling in which your heart will be exercised equally with your head.”³³ Hence the need to “cultivate equally well hearts and heads.”³⁴ Almost foreseeing our infatuation with guidelines, he also warned against turning medicine into assembly-line work. There are “two great types of practitioners—the routinist and the rationalist,” he said in 1900, and “into the clutches of the demon routine the majority of us ultimately come.”³⁶

Like most great people, Osler was a man of lights, shadows, and contradictions, probably not quite the saint we wish to believe. Yet he provides insights that are as valid today as they were for his own times, and possibly even more so. His 1919 speech is a paean to the humanities, but also a potential eulogy. As a Victorian physician, Osler was a blend of the new science and the old humanities. He knew that “the old art cannot possibly be replaced by, but must be absorbed in, the new science.”³⁵ Yet he could also see the upcoming split between the two cultures, and he tried to warn us. He could in fact foresee the end of an entire way of life. As he said in his address, “there must be a very different civilization or there will be no civilization at all.”¹

The crisis we face in medicine today may indeed be a symptom of a much larger cultural shift. As Osler himself put it, “The philosophies of one age have become the absurdities of the next, and the foolishness of yesterday has become the wisdom of tomorrow.”³³ Like Osler, we live in times of transition that require us to act. If in 1910 Flexner gave us science,³⁷ Osler in 1919 reminded us that medicine also needs the humanities. We ought to heed his message and reconcile the two fields. The alternative is a future full of tricorders and burned-out technicians, but sorely lacking in healers.

“Twin berries on one stem, grievous damage has been done to both in regarding the Humanities and Science in any other light than complemental.”
—Sir William Osler¹

The year 2019 marks the 100th anniversary of Sir William Osler’s last public speech. Still reeling from the death of his only son in World War I, he had been asked to give the presidential inaugural address of the Classical Association at Oxford. It was the first time a physician had received the honor, and Osler took the assignment very seriously. He chose to speak about “The old humanities and the new science,” and to call for a reunification of the two fields. “Humanists have not enough Science” he warned, “and Science sadly lacks the Humanities…this unhappy divorce…should never have taken place.”¹  Later, he said that it was the speech to which he had given the greatest thought and preparation. It was in fact Osler’s personal legacy: 2 months later he turned 70, and 7 months later he was dead.

Revisiting the address today, what can Osler teach the high-tech physician of today, when doctors have become “providers” and patients “consumers”? Is Osler’s message still relevant to our craft, or has he simply become an icon of professional nostalgia with little value for our times?

THE NEED FOR THE HUMANITIES IN MEDICINE

Medicine has certainly grown both powerful and successful. Yet it is also confronting hurdles that would have been unimaginable in Osler’s time. Physicians are now the professionals with the highest suicide rate,² a burnout rate as high as 70%,^3,4 rampant depression,⁵ dwindling empathy,⁶ a predominantly negative perception by the public,^7,8 and a disturbing propensity to quit.⁹ These, of course, may just be symptoms of an increasingly meaningless environment wherein doctors have become small cogs in a medical-industrial complex they can’t control or even understand. Still, is it possible that something more personal may have been lost in the way we now select and educate physicians? Could this, in turn, make us less resilient?

In this regard, Osler’s last public speech serves as an enduring reminder of the need for the humanities in medicine. Osler not only believed it, but throughout his life never missed a chance to express in words, writings, and deeds that the humanities are indeed “the hormones” of the profession. In 1919 he warned against the risk of separating our humanistic tradition from the sciences, and urged us “to infect [anyone] with the spirit of the Humanities,” since to him that was “the greatest single gift in education.”¹

Unfortunately, the humanities are slippery, not easily quantifiable, hard to define, and seemingly incompatible with an evidence-based approach. Quite understandably, today’s data-obsessed medicine views them with suspicion. But besides reminding us that in medicine not all that counts can be counted, and not all that can be counted counts, the humanities are in fact a fundamental component of the physician’s skill set.

In a multicenter survey of 5 medical schools,¹⁰ there was indeed a correlation between students’ exposure to the humanities and many of the personal qualities whose absence we lament in today’s medicine: empathy, tolerance for ambiguity, emotional intelligence, and prevention of burnout. Most significant was a strong correlation with wisdom, as measured by the 21-item Brief Wisdom Screening Scale.¹¹ That all these traits may correlate with humanities exposure is intuitive, since the humanities not only teach tolerance and compassion, but also capture the collective experience of those who came before us. Hence, they teach us wisdom. Wisdom is not an ACGME competency, but it’s undoubtedly a prerequisite for the art of healing.¹² In fact, wisdom may very well be the fundamental trait that characterizes a well-rounded physician, since it encompasses empathy, resilience, comfort with ambiguity, and the capacity to learn from the past. Not surprisingly, wisdom in the world was Osler’s closing wish in 1919.

The humanities can also nurture the very personal qualities we desire in physicians. For example, observing drama fosters empathy,¹³ as does taking an elective in medical humanities.¹⁴ Drawing enhances the reading of faces,¹⁵ and observing art improves the art of clinical observation.¹⁶ Reading good literature prompts better detection of emotions,¹⁷ and reflective writing improves students’ well-being.¹⁸ Playing a musical instrument reduces burnout.¹⁹ And an undergraduate major in the humanities correlates with greater tolerance for ambiguity,²⁰ a highly desirable trait in physicians, since it means openness to new ideas and the capacity to better cope with difficult situations.²¹

In fact, some of the qualities fostered by the humanities even translate into better patient care. For instance, tolerance for ambiguity correlates with more positive attitudes towards patients who have frustrating complaints,²² with lower use of resources,²³ and with a career choice in direct patient care.²⁴ Hence, it has been suggested that it should be a prerequisite for medical school admission.²⁵ Physicians’ empathy is also beneficial, since it correlates with a lower rate of complications and better outcomes in the care of diabetic patients.²⁶ This should not come as a surprise. As Hippocrates put it 2,500 years ago, “some patients, though conscious that their condition is perilous, recover their health simply through their contentment with the goodness of the physician.”²⁷

Lastly, studying the humanities may provide crucial antibodies against the pain and suffering that are unavoidable staples of the human condition. To paraphrase Osler, the humanities might vaccinate us against the difficulties of our profession. Hippocrates himself had suggested that “it is well to superintend the sick to make them well, to care for the healthy to keep them well, but also to care for one’s self…”²⁷ That is why many institutions now require medical students to take humanities courses.²⁸

Yet this effort may amount to a rearguard action that arrives too late and provides too little. The humanities should probably be taught before medical school.²⁹ After all, if it’s possible to make a scientist out of a humanist (Osler was living proof), the experience of the past decades seems to suggest that it’s considerably harder to make a humanist out of a scientist—hence the need to revisit undergraduate curricula and admission criteria to medical school, so that students can receive an adequate foundation in both arenas. Ironically, students express positive attitudes toward a liberal education and think it would actually help them as physicians.³⁰ Yet they also understand that the selection process remains tilted towards the sciences.^30–32

For Osler, scientific evidence was important but not a substitute for a humanistic approach. As he reminded students, “The practice of medicine is an art based on science,”³³ whose main goals are to prevent disease, relieve suffering, and heal the sick. To do so, one ought to care more “for the individual patient than for the special features of the disease.”³⁴ But he warned them, “It is much harder to acquire the art than the science.”³⁵ In fact, “The practice of medicine is a calling in which your heart will be exercised equally with your head.”³³ Hence the need to “cultivate equally well hearts and heads.”³⁴ Almost foreseeing our infatuation with guidelines, he also warned against turning medicine into assembly-line work. There are “two great types of practitioners—the routinist and the rationalist,” he said in 1900, and “into the clutches of the demon routine the majority of us ultimately come.”³⁶

Like most great people, Osler was a man of lights, shadows, and contradictions, probably not quite the saint we wish to believe. Yet he provides insights that are as valid today as they were for his own times, and possibly even more so. His 1919 speech is a paean to the humanities, but also a potential eulogy. As a Victorian physician, Osler was a blend of the new science and the old humanities. He knew that “the old art cannot possibly be replaced by, but must be absorbed in, the new science.”³⁵ Yet he could also see the upcoming split between the two cultures, and he tried to warn us. He could in fact foresee the end of an entire way of life. As he said in his address, “there must be a very different civilization or there will be no civilization at all.”¹

The crisis we face in medicine today may indeed be a symptom of a much larger cultural shift. As Osler himself put it, “The philosophies of one age have become the absurdities of the next, and the foolishness of yesterday has become the wisdom of tomorrow.”³³ Like Osler, we live in times of transition that require us to act. If in 1910 Flexner gave us science,³⁷ Osler in 1919 reminded us that medicine also needs the humanities. We ought to heed his message and reconcile the two fields. The alternative is a future full of tricorders and burned-out technicians, but sorely lacking in healers.

“Twin berries on one stem, grievous damage has been done to both in regarding the Humanities and Science in any other light than complemental.”
—Sir William Osler¹

The year 2019 marks the 100th anniversary of Sir William Osler’s last public speech. Still reeling from the death of his only son in World War I, he had been asked to give the presidential inaugural address of the Classical Association at Oxford. It was the first time a physician had received the honor, and Osler took the assignment very seriously. He chose to speak about “The old humanities and the new science,” and to call for a reunification of the two fields. “Humanists have not enough Science” he warned, “and Science sadly lacks the Humanities…this unhappy divorce…should never have taken place.”¹  Later, he said that it was the speech to which he had given the greatest thought and preparation. It was in fact Osler’s personal legacy: 2 months later he turned 70, and 7 months later he was dead.

Revisiting the address today, what can Osler teach the high-tech physician of today, when doctors have become “providers” and patients “consumers”? Is Osler’s message still relevant to our craft, or has he simply become an icon of professional nostalgia with little value for our times?

THE NEED FOR THE HUMANITIES IN MEDICINE

Medicine has certainly grown both powerful and successful. Yet it is also confronting hurdles that would have been unimaginable in Osler’s time. Physicians are now the professionals with the highest suicide rate,² a burnout rate as high as 70%,^3,4 rampant depression,⁵ dwindling empathy,⁶ a predominantly negative perception by the public,^7,8 and a disturbing propensity to quit.⁹ These, of course, may just be symptoms of an increasingly meaningless environment wherein doctors have become small cogs in a medical-industrial complex they can’t control or even understand. Still, is it possible that something more personal may have been lost in the way we now select and educate physicians? Could this, in turn, make us less resilient?

In this regard, Osler’s last public speech serves as an enduring reminder of the need for the humanities in medicine. Osler not only believed it, but throughout his life never missed a chance to express in words, writings, and deeds that the humanities are indeed “the hormones” of the profession. In 1919 he warned against the risk of separating our humanistic tradition from the sciences, and urged us “to infect [anyone] with the spirit of the Humanities,” since to him that was “the greatest single gift in education.”¹

Unfortunately, the humanities are slippery, not easily quantifiable, hard to define, and seemingly incompatible with an evidence-based approach. Quite understandably, today’s data-obsessed medicine views them with suspicion. But besides reminding us that in medicine not all that counts can be counted, and not all that can be counted counts, the humanities are in fact a fundamental component of the physician’s skill set.

In a multicenter survey of 5 medical schools,¹⁰ there was indeed a correlation between students’ exposure to the humanities and many of the personal qualities whose absence we lament in today’s medicine: empathy, tolerance for ambiguity, emotional intelligence, and prevention of burnout. Most significant was a strong correlation with wisdom, as measured by the 21-item Brief Wisdom Screening Scale.¹¹ That all these traits may correlate with humanities exposure is intuitive, since the humanities not only teach tolerance and compassion, but also capture the collective experience of those who came before us. Hence, they teach us wisdom. Wisdom is not an ACGME competency, but it’s undoubtedly a prerequisite for the art of healing.¹² In fact, wisdom may very well be the fundamental trait that characterizes a well-rounded physician, since it encompasses empathy, resilience, comfort with ambiguity, and the capacity to learn from the past. Not surprisingly, wisdom in the world was Osler’s closing wish in 1919.

The humanities can also nurture the very personal qualities we desire in physicians. For example, observing drama fosters empathy,¹³ as does taking an elective in medical humanities.¹⁴ Drawing enhances the reading of faces,¹⁵ and observing art improves the art of clinical observation.¹⁶ Reading good literature prompts better detection of emotions,¹⁷ and reflective writing improves students’ well-being.¹⁸ Playing a musical instrument reduces burnout.¹⁹ And an undergraduate major in the humanities correlates with greater tolerance for ambiguity,²⁰ a highly desirable trait in physicians, since it means openness to new ideas and the capacity to better cope with difficult situations.²¹

In fact, some of the qualities fostered by the humanities even translate into better patient care. For instance, tolerance for ambiguity correlates with more positive attitudes towards patients who have frustrating complaints,²² with lower use of resources,²³ and with a career choice in direct patient care.²⁴ Hence, it has been suggested that it should be a prerequisite for medical school admission.²⁵ Physicians’ empathy is also beneficial, since it correlates with a lower rate of complications and better outcomes in the care of diabetic patients.²⁶ This should not come as a surprise. As Hippocrates put it 2,500 years ago, “some patients, though conscious that their condition is perilous, recover their health simply through their contentment with the goodness of the physician.”²⁷

Lastly, studying the humanities may provide crucial antibodies against the pain and suffering that are unavoidable staples of the human condition. To paraphrase Osler, the humanities might vaccinate us against the difficulties of our profession. Hippocrates himself had suggested that “it is well to superintend the sick to make them well, to care for the healthy to keep them well, but also to care for one’s self…”²⁷ That is why many institutions now require medical students to take humanities courses.²⁸

Yet this effort may amount to a rearguard action that arrives too late and provides too little. The humanities should probably be taught before medical school.²⁹ After all, if it’s possible to make a scientist out of a humanist (Osler was living proof), the experience of the past decades seems to suggest that it’s considerably harder to make a humanist out of a scientist—hence the need to revisit undergraduate curricula and admission criteria to medical school, so that students can receive an adequate foundation in both arenas. Ironically, students express positive attitudes toward a liberal education and think it would actually help them as physicians.³⁰ Yet they also understand that the selection process remains tilted towards the sciences.^30–32

For Osler, scientific evidence was important but not a substitute for a humanistic approach. As he reminded students, “The practice of medicine is an art based on science,”³³ whose main goals are to prevent disease, relieve suffering, and heal the sick. To do so, one ought to care more “for the individual patient than for the special features of the disease.”³⁴ But he warned them, “It is much harder to acquire the art than the science.”³⁵ In fact, “The practice of medicine is a calling in which your heart will be exercised equally with your head.”³³ Hence the need to “cultivate equally well hearts and heads.”³⁴ Almost foreseeing our infatuation with guidelines, he also warned against turning medicine into assembly-line work. There are “two great types of practitioners—the routinist and the rationalist,” he said in 1900, and “into the clutches of the demon routine the majority of us ultimately come.”³⁶

Like most great people, Osler was a man of lights, shadows, and contradictions, probably not quite the saint we wish to believe. Yet he provides insights that are as valid today as they were for his own times, and possibly even more so. His 1919 speech is a paean to the humanities, but also a potential eulogy. As a Victorian physician, Osler was a blend of the new science and the old humanities. He knew that “the old art cannot possibly be replaced by, but must be absorbed in, the new science.”³⁵ Yet he could also see the upcoming split between the two cultures, and he tried to warn us. He could in fact foresee the end of an entire way of life. As he said in his address, “there must be a very different civilization or there will be no civilization at all.”¹

The crisis we face in medicine today may indeed be a symptom of a much larger cultural shift. As Osler himself put it, “The philosophies of one age have become the absurdities of the next, and the foolishness of yesterday has become the wisdom of tomorrow.”³³ Like Osler, we live in times of transition that require us to act. If in 1910 Flexner gave us science,³⁷ Osler in 1919 reminded us that medicine also needs the humanities. We ought to heed his message and reconcile the two fields. The alternative is a future full of tricorders and burned-out technicians, but sorely lacking in healers.

To the Editor: Regarding the article about a man with rapidly progressive pleural effusion by Zoumot et al in the January 2019 issue,¹ there was some inconsistency between the teaching points and the actions taken.

Question 1 asked what was the most likely cause of the patient’s pleuritic chest pain. Pulmonary embolism was an unlikely diagnosis, given the patient’s presentation and his normal D-dimer level, which the text acknowledges, but then proceeds to state that computed tomographic angiography of the chest was done anyway.

After pleural effusion was diagnosed, question 2 asked what was the best management strategy for the patient at that time. The best management strategy was to give oral antibiotics with close follow-up because the patient was at low risk of a poor outcome, but he was advised to be admitted for intravenous antibiotics anyway.

I’m not quite sure of the point of the didactic exercise when actions are not consistent with the analytic rationale for testing and treatment.

To the Editor: Regarding the article about a man with rapidly progressive pleural effusion by Zoumot et al in the January 2019 issue,¹ there was some inconsistency between the teaching points and the actions taken.

Question 1 asked what was the most likely cause of the patient’s pleuritic chest pain. Pulmonary embolism was an unlikely diagnosis, given the patient’s presentation and his normal D-dimer level, which the text acknowledges, but then proceeds to state that computed tomographic angiography of the chest was done anyway.

After pleural effusion was diagnosed, question 2 asked what was the best management strategy for the patient at that time. The best management strategy was to give oral antibiotics with close follow-up because the patient was at low risk of a poor outcome, but he was advised to be admitted for intravenous antibiotics anyway.

I’m not quite sure of the point of the didactic exercise when actions are not consistent with the analytic rationale for testing and treatment.

To the Editor: Regarding the article about a man with rapidly progressive pleural effusion by Zoumot et al in the January 2019 issue,¹ there was some inconsistency between the teaching points and the actions taken.

Question 1 asked what was the most likely cause of the patient’s pleuritic chest pain. Pulmonary embolism was an unlikely diagnosis, given the patient’s presentation and his normal D-dimer level, which the text acknowledges, but then proceeds to state that computed tomographic angiography of the chest was done anyway.

After pleural effusion was diagnosed, question 2 asked what was the best management strategy for the patient at that time. The best management strategy was to give oral antibiotics with close follow-up because the patient was at low risk of a poor outcome, but he was advised to be admitted for intravenous antibiotics anyway.

I’m not quite sure of the point of the didactic exercise when actions are not consistent with the analytic rationale for testing and treatment.

In Reply: We thank Dr. Davidson for his comments. Indeed, the teaching points may appear inconsistent with the actual patient journey in this case. In the real world, physicians from different teams and specialties are involved in the care of a patient, and medical practice may not strictly adhere to guidelines.

In question 1, the emergency department physician decided to proceed with computed tomographic pulmonary angiography to rule out pulmonary embolism. Based on best practice guidelines, pulmonary angiography was not indicated, as the clinical pretest probability of pulmonary embolism was low, supported by the patient’s negative D-dimer test. When we wrote the article, as we already had the scan, we used it to support the learning points in terms of findings on computed tomography at the early stage of a developing empyema, and also to support that the scan was in fact not indicated (not the other way around).

As for question 2, specific data-driven guidelines do not exist on how best to manage patients with bronchopneumonia with an early evolving parapneumonic effusion. In the text that follows question 2, we stated that management as an inpatient or outpatient would have been reasonable. Although we considered the patient at low risk for a poor outcome, we offered inpatient admission at the time for better control of his severe pleuritic pain (this could have been made clearer in the text), as well as close monitoring of his evolving parapneumonic effusion, and we do not believe that this contradicts the teaching points of this case.

In Reply: We thank Dr. Davidson for his comments. Indeed, the teaching points may appear inconsistent with the actual patient journey in this case. In the real world, physicians from different teams and specialties are involved in the care of a patient, and medical practice may not strictly adhere to guidelines.

In question 1, the emergency department physician decided to proceed with computed tomographic pulmonary angiography to rule out pulmonary embolism. Based on best practice guidelines, pulmonary angiography was not indicated, as the clinical pretest probability of pulmonary embolism was low, supported by the patient’s negative D-dimer test. When we wrote the article, as we already had the scan, we used it to support the learning points in terms of findings on computed tomography at the early stage of a developing empyema, and also to support that the scan was in fact not indicated (not the other way around).

As for question 2, specific data-driven guidelines do not exist on how best to manage patients with bronchopneumonia with an early evolving parapneumonic effusion. In the text that follows question 2, we stated that management as an inpatient or outpatient would have been reasonable. Although we considered the patient at low risk for a poor outcome, we offered inpatient admission at the time for better control of his severe pleuritic pain (this could have been made clearer in the text), as well as close monitoring of his evolving parapneumonic effusion, and we do not believe that this contradicts the teaching points of this case.

In Reply: We thank Dr. Davidson for his comments. Indeed, the teaching points may appear inconsistent with the actual patient journey in this case. In the real world, physicians from different teams and specialties are involved in the care of a patient, and medical practice may not strictly adhere to guidelines.

In question 1, the emergency department physician decided to proceed with computed tomographic pulmonary angiography to rule out pulmonary embolism. Based on best practice guidelines, pulmonary angiography was not indicated, as the clinical pretest probability of pulmonary embolism was low, supported by the patient’s negative D-dimer test. When we wrote the article, as we already had the scan, we used it to support the learning points in terms of findings on computed tomography at the early stage of a developing empyema, and also to support that the scan was in fact not indicated (not the other way around).

As for question 2, specific data-driven guidelines do not exist on how best to manage patients with bronchopneumonia with an early evolving parapneumonic effusion. In the text that follows question 2, we stated that management as an inpatient or outpatient would have been reasonable. Although we considered the patient at low risk for a poor outcome, we offered inpatient admission at the time for better control of his severe pleuritic pain (this could have been made clearer in the text), as well as close monitoring of his evolving parapneumonic effusion, and we do not believe that this contradicts the teaching points of this case.

To the Editor: I enjoyed reading “Should metformin be used in every patient with type 2 diabetes” by Makin and Lansang in the January 2019 issue.¹

I just wanted to point out that metformin is a frequent cause of low serum vitamin B₁₂ levels, and serum vitamin B₁₂ levels should be monitored intermittently in patients using metformin.

To the Editor: I enjoyed reading “Should metformin be used in every patient with type 2 diabetes” by Makin and Lansang in the January 2019 issue.¹

I just wanted to point out that metformin is a frequent cause of low serum vitamin B₁₂ levels, and serum vitamin B₁₂ levels should be monitored intermittently in patients using metformin.

To the Editor: I enjoyed reading “Should metformin be used in every patient with type 2 diabetes” by Makin and Lansang in the January 2019 issue.¹

I just wanted to point out that metformin is a frequent cause of low serum vitamin B₁₂ levels, and serum vitamin B₁₂ levels should be monitored intermittently in patients using metformin.