Letters To The Editor

To the Editor: I read with great interest the article by Thomas et al, “Interpreting SPRINT: How low should you go?”¹

Hypertension is the most prevalent modifiable risk factor, affecting almost one in every three people in the United States.² Moreover, only half of people with hypertension have their blood pressure under control to the current standard of lower than 140/90 mm Hg.² The Systolic Blood Pressure Intervention Trial (SPRINT) tested a lower goal systolic pressure, ie, less than 120 mm Hg, and found it more beneficial than the standard goal of less than 140 mm Hg.³

A drawback of SPRINT that Thomas et al did not address in their interpretation of the trial is that the two study groups were not homogeneous in terms of the antihypertensive drugs used. Antihypertensive drugs do not only lower blood pressure—some of them have additional pleiotropic effects, making their use more advantageous in special situations. For example, renin-angiotensin-aldosterone system (RAAS) blockers—ie, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, and mineralocorticoid receptor antagonists—are disease-modfying drugs in heart failure, as are certain beta-blockers.⁴ The cardiovascular benefit seen in the intensive-treatment group in SPRINT compared with the standard-therapy group was primarily due to a reduction in heart failure (a 38% relative risk reduction, P = .0002),³ for which RAAS blockers and beta-adrenergic blocking drugs have been shown consistently to be beneficial. But the intensive- and standard-therapy groups were not homogeneous in terms of the use of RAAS blockers and beta-blockers.

So, was the cardiovascular benefit attained in the intensive-treatment group in SPRINT due to the benefit of lower blood pressure or to the drugs used?

To the Editor: I read with great interest the article by Thomas et al, “Interpreting SPRINT: How low should you go?”¹

Hypertension is the most prevalent modifiable risk factor, affecting almost one in every three people in the United States.² Moreover, only half of people with hypertension have their blood pressure under control to the current standard of lower than 140/90 mm Hg.² The Systolic Blood Pressure Intervention Trial (SPRINT) tested a lower goal systolic pressure, ie, less than 120 mm Hg, and found it more beneficial than the standard goal of less than 140 mm Hg.³

A drawback of SPRINT that Thomas et al did not address in their interpretation of the trial is that the two study groups were not homogeneous in terms of the antihypertensive drugs used. Antihypertensive drugs do not only lower blood pressure—some of them have additional pleiotropic effects, making their use more advantageous in special situations. For example, renin-angiotensin-aldosterone system (RAAS) blockers—ie, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, and mineralocorticoid receptor antagonists—are disease-modfying drugs in heart failure, as are certain beta-blockers.⁴ The cardiovascular benefit seen in the intensive-treatment group in SPRINT compared with the standard-therapy group was primarily due to a reduction in heart failure (a 38% relative risk reduction, P = .0002),³ for which RAAS blockers and beta-adrenergic blocking drugs have been shown consistently to be beneficial. But the intensive- and standard-therapy groups were not homogeneous in terms of the use of RAAS blockers and beta-blockers.

So, was the cardiovascular benefit attained in the intensive-treatment group in SPRINT due to the benefit of lower blood pressure or to the drugs used?

To the Editor: I read with great interest the article by Thomas et al, “Interpreting SPRINT: How low should you go?”¹

Hypertension is the most prevalent modifiable risk factor, affecting almost one in every three people in the United States.² Moreover, only half of people with hypertension have their blood pressure under control to the current standard of lower than 140/90 mm Hg.² The Systolic Blood Pressure Intervention Trial (SPRINT) tested a lower goal systolic pressure, ie, less than 120 mm Hg, and found it more beneficial than the standard goal of less than 140 mm Hg.³

A drawback of SPRINT that Thomas et al did not address in their interpretation of the trial is that the two study groups were not homogeneous in terms of the antihypertensive drugs used. Antihypertensive drugs do not only lower blood pressure—some of them have additional pleiotropic effects, making their use more advantageous in special situations. For example, renin-angiotensin-aldosterone system (RAAS) blockers—ie, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, and mineralocorticoid receptor antagonists—are disease-modfying drugs in heart failure, as are certain beta-blockers.⁴ The cardiovascular benefit seen in the intensive-treatment group in SPRINT compared with the standard-therapy group was primarily due to a reduction in heart failure (a 38% relative risk reduction, P = .0002),³ for which RAAS blockers and beta-adrenergic blocking drugs have been shown consistently to be beneficial. But the intensive- and standard-therapy groups were not homogeneous in terms of the use of RAAS blockers and beta-blockers.

So, was the cardiovascular benefit attained in the intensive-treatment group in SPRINT due to the benefit of lower blood pressure or to the drugs used?

To the Editor: In their review,¹ Thomas et al noted that the benefits of intensive blood pressure lowering seen in the SPRINT study² were not observed in the Action to Control Cardiovascular Risk in Diabetes-Blood pressure (ACCORD BP) trial³ or in the Secondary Prevention of Small Subcortical Strokes (SPS3) trial.⁴ In addition to the reasons discussed in their review, the discrepancy may be due to the surprisingly low rate of statin use in the patients enrolled in SPRINT. Even though 61% of the patients in SPRINT had a 10-year Framingham risk score greater than 15%, only 44% of the patients were on statin therapy. In comparison, rates of statin use in ACCORD BP and SPS3 were 65% and 83%, respectively.

A possible interaction between statin use and intensive blood pressure lowering is consistent with previous data on angiotensin-converting enzyme (ACE) inhibitor use in high-risk populations.

The Heart Outcomes Prevention Evaluation (HOPE) trial,⁵ in which only 29% of patients received lipid-lowering therapy, found that ACE inhibitor use was associated with a significant reduction in a composite cardiovascular outcome, whereas the Prevention of Events With Angiotensin-Converting Enzyme Inhibitor Therapy (PEACE) trial,⁶ in which 70% of patients were on lipid-lowering therapy, did not show a benefit for ACE inhibitor therapy. In addition, there are many drug interactions between statins and calcium channel blockers, potentially limiting options for simultaneous aggressive treatment of lipid levels and blood pressure.

In summary, aggressive use of statins may confer sufficient cardiovascular protection when aggressive antihypertensive therapy provides little or no incremental benefit. Hopefully, further analyses of these trials will shed light on this important question.

To the Editor: In their review,¹ Thomas et al noted that the benefits of intensive blood pressure lowering seen in the SPRINT study² were not observed in the Action to Control Cardiovascular Risk in Diabetes-Blood pressure (ACCORD BP) trial³ or in the Secondary Prevention of Small Subcortical Strokes (SPS3) trial.⁴ In addition to the reasons discussed in their review, the discrepancy may be due to the surprisingly low rate of statin use in the patients enrolled in SPRINT. Even though 61% of the patients in SPRINT had a 10-year Framingham risk score greater than 15%, only 44% of the patients were on statin therapy. In comparison, rates of statin use in ACCORD BP and SPS3 were 65% and 83%, respectively.

A possible interaction between statin use and intensive blood pressure lowering is consistent with previous data on angiotensin-converting enzyme (ACE) inhibitor use in high-risk populations.

The Heart Outcomes Prevention Evaluation (HOPE) trial,⁵ in which only 29% of patients received lipid-lowering therapy, found that ACE inhibitor use was associated with a significant reduction in a composite cardiovascular outcome, whereas the Prevention of Events With Angiotensin-Converting Enzyme Inhibitor Therapy (PEACE) trial,⁶ in which 70% of patients were on lipid-lowering therapy, did not show a benefit for ACE inhibitor therapy. In addition, there are many drug interactions between statins and calcium channel blockers, potentially limiting options for simultaneous aggressive treatment of lipid levels and blood pressure.

In summary, aggressive use of statins may confer sufficient cardiovascular protection when aggressive antihypertensive therapy provides little or no incremental benefit. Hopefully, further analyses of these trials will shed light on this important question.

To the Editor: In their review,¹ Thomas et al noted that the benefits of intensive blood pressure lowering seen in the SPRINT study² were not observed in the Action to Control Cardiovascular Risk in Diabetes-Blood pressure (ACCORD BP) trial³ or in the Secondary Prevention of Small Subcortical Strokes (SPS3) trial.⁴ In addition to the reasons discussed in their review, the discrepancy may be due to the surprisingly low rate of statin use in the patients enrolled in SPRINT. Even though 61% of the patients in SPRINT had a 10-year Framingham risk score greater than 15%, only 44% of the patients were on statin therapy. In comparison, rates of statin use in ACCORD BP and SPS3 were 65% and 83%, respectively.

A possible interaction between statin use and intensive blood pressure lowering is consistent with previous data on angiotensin-converting enzyme (ACE) inhibitor use in high-risk populations.

The Heart Outcomes Prevention Evaluation (HOPE) trial,⁵ in which only 29% of patients received lipid-lowering therapy, found that ACE inhibitor use was associated with a significant reduction in a composite cardiovascular outcome, whereas the Prevention of Events With Angiotensin-Converting Enzyme Inhibitor Therapy (PEACE) trial,⁶ in which 70% of patients were on lipid-lowering therapy, did not show a benefit for ACE inhibitor therapy. In addition, there are many drug interactions between statins and calcium channel blockers, potentially limiting options for simultaneous aggressive treatment of lipid levels and blood pressure.

In summary, aggressive use of statins may confer sufficient cardiovascular protection when aggressive antihypertensive therapy provides little or no incremental benefit. Hopefully, further analyses of these trials will shed light on this important question.

In Reply: We thank the readers for their important and insightful comments and questions.

Dr. Yilmaz raises the point that there was no mandate in the SPRINT trial to preferentially use any specific class of antihypertensive medications in either group. However, there was greater use of all drug classes (including diuretics and renin-angiotensin-aldosterone blockers) in the intensive-treatment group.¹ (This information was included as a supplementary appendix in the main paper, and as Table 1 in our review.) Could this have contributed to the primary cardiovascular outcome benefit seen in the intensive-therapy group, largely driven by a decreased incidence of heart failure, or could it even have masked the symptoms of heart failure rather than preventing it^2,3? While this is plausible, since the SPRINT trial was designed as a “treat to target” study and not as an antihypertensive medication efficacy study, it is difficult to conclusively answer the question of potential pleiotropic effects of antihypertensive medications influencing the trial results. The authors did not comment on this in the main paper, and we agree that further analysis would be helpful in exploring this important question.

Dr. Edwards raises the question whether antihypertensive therapy confers additional cardiovascular benefit over aggressive use of statins. Statin use in the SPRINT cohort (both intensive and standard groups) was low at baseline, despite this being a population at high cardiovascular risk.¹ It is unclear whether treatment practices pertaining to lipid management could have changed during the course of the trial in participants within the SPRINT cohort, particularly after the new lipid guidelines were published. The recently published HOPE-3 trial indicated cardiovascular benefit with statins used as a primary prevention strategy in older persons with intermediate cardiovascular risk.^4,5 Notably, outcomes with combination therapy in this trial using a statin plus antihypertensive therapy were not significantly better than with statin alone, except in the subgroup of participants who were in the upper third of systolic blood pressure levels, where combination appeared to benefit more. This study, of course, was done in a population with lower cardiovascular risk than in SPRINT, and the antihypertensive medications used (candesartan and hydrochlorothiazide) were not at maximal doses. There is also a question of whether use of chlorthalidone in HOPE-3 may have been more effective.

We agree with Dr. Edwards that this is an important question that merits further exploration, especially in the broader context of treatment based on cardiovascular risk.

In Reply: We thank the readers for their important and insightful comments and questions.

Dr. Yilmaz raises the point that there was no mandate in the SPRINT trial to preferentially use any specific class of antihypertensive medications in either group. However, there was greater use of all drug classes (including diuretics and renin-angiotensin-aldosterone blockers) in the intensive-treatment group.¹ (This information was included as a supplementary appendix in the main paper, and as Table 1 in our review.) Could this have contributed to the primary cardiovascular outcome benefit seen in the intensive-therapy group, largely driven by a decreased incidence of heart failure, or could it even have masked the symptoms of heart failure rather than preventing it^2,3? While this is plausible, since the SPRINT trial was designed as a “treat to target” study and not as an antihypertensive medication efficacy study, it is difficult to conclusively answer the question of potential pleiotropic effects of antihypertensive medications influencing the trial results. The authors did not comment on this in the main paper, and we agree that further analysis would be helpful in exploring this important question.

Dr. Edwards raises the question whether antihypertensive therapy confers additional cardiovascular benefit over aggressive use of statins. Statin use in the SPRINT cohort (both intensive and standard groups) was low at baseline, despite this being a population at high cardiovascular risk.¹ It is unclear whether treatment practices pertaining to lipid management could have changed during the course of the trial in participants within the SPRINT cohort, particularly after the new lipid guidelines were published. The recently published HOPE-3 trial indicated cardiovascular benefit with statins used as a primary prevention strategy in older persons with intermediate cardiovascular risk.^4,5 Notably, outcomes with combination therapy in this trial using a statin plus antihypertensive therapy were not significantly better than with statin alone, except in the subgroup of participants who were in the upper third of systolic blood pressure levels, where combination appeared to benefit more. This study, of course, was done in a population with lower cardiovascular risk than in SPRINT, and the antihypertensive medications used (candesartan and hydrochlorothiazide) were not at maximal doses. There is also a question of whether use of chlorthalidone in HOPE-3 may have been more effective.

We agree with Dr. Edwards that this is an important question that merits further exploration, especially in the broader context of treatment based on cardiovascular risk.

In Reply: We thank the readers for their important and insightful comments and questions.

Dr. Yilmaz raises the point that there was no mandate in the SPRINT trial to preferentially use any specific class of antihypertensive medications in either group. However, there was greater use of all drug classes (including diuretics and renin-angiotensin-aldosterone blockers) in the intensive-treatment group.¹ (This information was included as a supplementary appendix in the main paper, and as Table 1 in our review.) Could this have contributed to the primary cardiovascular outcome benefit seen in the intensive-therapy group, largely driven by a decreased incidence of heart failure, or could it even have masked the symptoms of heart failure rather than preventing it^2,3? While this is plausible, since the SPRINT trial was designed as a “treat to target” study and not as an antihypertensive medication efficacy study, it is difficult to conclusively answer the question of potential pleiotropic effects of antihypertensive medications influencing the trial results. The authors did not comment on this in the main paper, and we agree that further analysis would be helpful in exploring this important question.

Dr. Edwards raises the question whether antihypertensive therapy confers additional cardiovascular benefit over aggressive use of statins. Statin use in the SPRINT cohort (both intensive and standard groups) was low at baseline, despite this being a population at high cardiovascular risk.¹ It is unclear whether treatment practices pertaining to lipid management could have changed during the course of the trial in participants within the SPRINT cohort, particularly after the new lipid guidelines were published. The recently published HOPE-3 trial indicated cardiovascular benefit with statins used as a primary prevention strategy in older persons with intermediate cardiovascular risk.^4,5 Notably, outcomes with combination therapy in this trial using a statin plus antihypertensive therapy were not significantly better than with statin alone, except in the subgroup of participants who were in the upper third of systolic blood pressure levels, where combination appeared to benefit more. This study, of course, was done in a population with lower cardiovascular risk than in SPRINT, and the antihypertensive medications used (candesartan and hydrochlorothiazide) were not at maximal doses. There is also a question of whether use of chlorthalidone in HOPE-3 may have been more effective.

We agree with Dr. Edwards that this is an important question that merits further exploration, especially in the broader context of treatment based on cardiovascular risk.

To the Editor: We read the IM Board Review article by Hanouneh et al in the February issue of the Journal with great interest.¹ The authors described an interesting case of a young woman presenting with what initially seemed to be jaundice of acute onset, with rapid progression to acute encephalopathy and worsening liver failure. The patient was eventually diagnosed with fulminant Wilson disease and, thankfully, underwent successful liver transplant. We thank the authors for their in-depth review of the common causes of acute liver failure, the general approach to management, and the tailored treatment of Wilson disease in such settings.

However, we believe that several aspects merit further attention. First, on initial presentation and investigation, it would have been important to consider cholestatic hepatobiliary pathologic processes (eg, choledocholithiasis, cholangitis, primary biliary cirrhosis, primary sclerosing cholangitis), given the characteristic liver panel results.

Second, the authors rightly pointed out that hemolytic anemia is common in patients with acute liver failure secondary to Wilson disease. However, it is important to keep in mind that additional testing should include Coombs testing (typically negative in Wilson disease) and examination of the peripheral smear to exclude other etiologies, since such conditions as thrombotic thrombocytopenic purpura may present with multiorgan failure as well.²

Third, the authors report that Kayser-Fleischer rings are pathognomonic for Wilson disease. However, many reports in peer-reviewed medical journals suggest that this may not be the case and the overall clinical picture should be
considered.³

Fourth, while the authors focus their attention on liver transplant, several other treatments deserve mentioning. We agree that liver transplant is considered the only lifesaving treatment. But in certain situations, molecular absorbent recirculation systems and hemodialysis may provide temporary support while awaiting transportation to a liver transplant center or actual liver transplant.⁴

To the Editor: We read the IM Board Review article by Hanouneh et al in the February issue of the Journal with great interest.¹ The authors described an interesting case of a young woman presenting with what initially seemed to be jaundice of acute onset, with rapid progression to acute encephalopathy and worsening liver failure. The patient was eventually diagnosed with fulminant Wilson disease and, thankfully, underwent successful liver transplant. We thank the authors for their in-depth review of the common causes of acute liver failure, the general approach to management, and the tailored treatment of Wilson disease in such settings.

However, we believe that several aspects merit further attention. First, on initial presentation and investigation, it would have been important to consider cholestatic hepatobiliary pathologic processes (eg, choledocholithiasis, cholangitis, primary biliary cirrhosis, primary sclerosing cholangitis), given the characteristic liver panel results.

Second, the authors rightly pointed out that hemolytic anemia is common in patients with acute liver failure secondary to Wilson disease. However, it is important to keep in mind that additional testing should include Coombs testing (typically negative in Wilson disease) and examination of the peripheral smear to exclude other etiologies, since such conditions as thrombotic thrombocytopenic purpura may present with multiorgan failure as well.²

Third, the authors report that Kayser-Fleischer rings are pathognomonic for Wilson disease. However, many reports in peer-reviewed medical journals suggest that this may not be the case and the overall clinical picture should be
considered.³

Fourth, while the authors focus their attention on liver transplant, several other treatments deserve mentioning. We agree that liver transplant is considered the only lifesaving treatment. But in certain situations, molecular absorbent recirculation systems and hemodialysis may provide temporary support while awaiting transportation to a liver transplant center or actual liver transplant.⁴

To the Editor: We read the IM Board Review article by Hanouneh et al in the February issue of the Journal with great interest.¹ The authors described an interesting case of a young woman presenting with what initially seemed to be jaundice of acute onset, with rapid progression to acute encephalopathy and worsening liver failure. The patient was eventually diagnosed with fulminant Wilson disease and, thankfully, underwent successful liver transplant. We thank the authors for their in-depth review of the common causes of acute liver failure, the general approach to management, and the tailored treatment of Wilson disease in such settings.

However, we believe that several aspects merit further attention. First, on initial presentation and investigation, it would have been important to consider cholestatic hepatobiliary pathologic processes (eg, choledocholithiasis, cholangitis, primary biliary cirrhosis, primary sclerosing cholangitis), given the characteristic liver panel results.

Second, the authors rightly pointed out that hemolytic anemia is common in patients with acute liver failure secondary to Wilson disease. However, it is important to keep in mind that additional testing should include Coombs testing (typically negative in Wilson disease) and examination of the peripheral smear to exclude other etiologies, since such conditions as thrombotic thrombocytopenic purpura may present with multiorgan failure as well.²

Third, the authors report that Kayser-Fleischer rings are pathognomonic for Wilson disease. However, many reports in peer-reviewed medical journals suggest that this may not be the case and the overall clinical picture should be
considered.³

Fourth, while the authors focus their attention on liver transplant, several other treatments deserve mentioning. We agree that liver transplant is considered the only lifesaving treatment. But in certain situations, molecular absorbent recirculation systems and hemodialysis may provide temporary support while awaiting transportation to a liver transplant center or actual liver transplant.⁴

In Reply: We thank Dr. Mirrakhimov and colleagues for bringing important questions to our attention.

In terms of the differential diagnosis of cholestatic liver injury, we agree that pathologic processes such choledocholithiasis, cholangitis, primary biliary cirrhosis, and primary sclerosing cholangitis should be generally considered. However, in the case we described, the patient had no abdominal pain or fever, which makes choledocholithiasis or cholangitis very unlikely. Primary biliary cirrhosis and primary sclerosing cholangitis can cause chronic liver disease but should not be considered in the differential diagnosis of acute liver injury (acute hepatitis), such as in the case we described.

We agree that the hemolytic anemia typically seen in patients with Wilson disease is Coombs-negative, and that Coombs testing and a peripheral smear should be performed. Both were negative in our patient.

We also agree with Dr. Mirrakhimov and colleagues that Kayser-Fleischer rings are not necessarily specific for Wilson disease and can be seen in patients with other forms of cholestatic liver disease such as primary biliary cirrhosis. However, Kayser-Fleischer rings are pathognomonic for acute liver failure from Wilson disease. In other words, when Kayser-Fleischer rings are seen in a patient with acute liver failure, the diagnosis is Wilson disease until proven otherwise.

We discussed on page 112 of our article other treatments such as plasmapheresis as adjunctive therapy to bridge patients with acute liver failure secondary to Wilson disease to transplant. However, liver transplant is still the only definitive and potentially curative treatment.

In Reply: We thank Dr. Mirrakhimov and colleagues for bringing important questions to our attention.

In terms of the differential diagnosis of cholestatic liver injury, we agree that pathologic processes such choledocholithiasis, cholangitis, primary biliary cirrhosis, and primary sclerosing cholangitis should be generally considered. However, in the case we described, the patient had no abdominal pain or fever, which makes choledocholithiasis or cholangitis very unlikely. Primary biliary cirrhosis and primary sclerosing cholangitis can cause chronic liver disease but should not be considered in the differential diagnosis of acute liver injury (acute hepatitis), such as in the case we described.

We agree that the hemolytic anemia typically seen in patients with Wilson disease is Coombs-negative, and that Coombs testing and a peripheral smear should be performed. Both were negative in our patient.

We also agree with Dr. Mirrakhimov and colleagues that Kayser-Fleischer rings are not necessarily specific for Wilson disease and can be seen in patients with other forms of cholestatic liver disease such as primary biliary cirrhosis. However, Kayser-Fleischer rings are pathognomonic for acute liver failure from Wilson disease. In other words, when Kayser-Fleischer rings are seen in a patient with acute liver failure, the diagnosis is Wilson disease until proven otherwise.

We discussed on page 112 of our article other treatments such as plasmapheresis as adjunctive therapy to bridge patients with acute liver failure secondary to Wilson disease to transplant. However, liver transplant is still the only definitive and potentially curative treatment.

In Reply: We thank Dr. Mirrakhimov and colleagues for bringing important questions to our attention.

In terms of the differential diagnosis of cholestatic liver injury, we agree that pathologic processes such choledocholithiasis, cholangitis, primary biliary cirrhosis, and primary sclerosing cholangitis should be generally considered. However, in the case we described, the patient had no abdominal pain or fever, which makes choledocholithiasis or cholangitis very unlikely. Primary biliary cirrhosis and primary sclerosing cholangitis can cause chronic liver disease but should not be considered in the differential diagnosis of acute liver injury (acute hepatitis), such as in the case we described.

We agree that the hemolytic anemia typically seen in patients with Wilson disease is Coombs-negative, and that Coombs testing and a peripheral smear should be performed. Both were negative in our patient.

We also agree with Dr. Mirrakhimov and colleagues that Kayser-Fleischer rings are not necessarily specific for Wilson disease and can be seen in patients with other forms of cholestatic liver disease such as primary biliary cirrhosis. However, Kayser-Fleischer rings are pathognomonic for acute liver failure from Wilson disease. In other words, when Kayser-Fleischer rings are seen in a patient with acute liver failure, the diagnosis is Wilson disease until proven otherwise.

We discussed on page 112 of our article other treatments such as plasmapheresis as adjunctive therapy to bridge patients with acute liver failure secondary to Wilson disease to transplant. However, liver transplant is still the only definitive and potentially curative treatment.

To the Editor: I appreciated the article on cognitive biases and diagnostic error by Mull et al in the November 2015 issue.¹ They presented an excellent description of the pitfalls of diagnosis as reflected in a case of a patient misdiagnosed with heart failure who ultimately died of pulmonary tuberculosis complicated by pulmonary embolism (the latter possibly from using the wrong form of heparin). To the points they raised,  I would like to add a few of my own about diagnosis in general and heart failure in particular.

First, any initial diagnosis not confirmed objectively within the first 24 hours should be questioned, and other possibilities should be investigated. I have found this to be essential for every day’s stay in the hospital and for every outpatient visit. The authors mention checklists as part of the solution to the problem of misdiagnosis, and I would suggest that confirmation of initial diagnoses be built into these checklists.

In the case of a presumptive diagnosis of an acute exacerbation of heart failure treated empirically with diuretics, the diagnosis should be confirmed by the next day’s response to the diuretics, ie, increased urine output, a lower respiratory rate, and a fall in the pro-B-type natriuretic peptide level. Moreover, a change in the radiographic appearance should be seen, and respiratory and pulmonary function should improve after the first 24 hours on oxygen supplementation plus diuretics. Daily patient weights are also critical in determining response to a diuretic, and are rarely done accurately. I order weights and review them daily for patients like this.

Second, it is good to look at things yourself, including the patient, medication lists, laboratory values, and radiographic films. The attending physician should look at the radiographs together with a senior radiologist. Seeing no improvement or change on the second hospital day, or seeing signs incompatible with heart failure, one could order computed tomography of the chest and begin to entertain pulmonary diagnoses.

Even vital signs can be questionable. For example, in the case presented here, with a temperature of 99°F, a heart rate of 105, and a pulse oxygenation saturation of 89%, a respiratory rate of 24 seems unbelievably low. In my experience, the respiratory rate is recorded erroneously most of the time unless it is recorded electronically or checked at the bedside by the physician using a timepiece with a sweep second-hand.

Additionally, I have found that ordering several days’ laboratory tests (eg, complete blood cell counts, chemistry panels) in advance, in many cases, risks missing important findings and wastes time, energy, and the patient’s blood. I have learned to evaluate each patient daily and to order the most pertinent laboratory tests. With electronic medical records, I can check laboratory results as soon as they are available.  

To the Editor: I appreciated the article on cognitive biases and diagnostic error by Mull et al in the November 2015 issue.¹ They presented an excellent description of the pitfalls of diagnosis as reflected in a case of a patient misdiagnosed with heart failure who ultimately died of pulmonary tuberculosis complicated by pulmonary embolism (the latter possibly from using the wrong form of heparin). To the points they raised,  I would like to add a few of my own about diagnosis in general and heart failure in particular.

First, any initial diagnosis not confirmed objectively within the first 24 hours should be questioned, and other possibilities should be investigated. I have found this to be essential for every day’s stay in the hospital and for every outpatient visit. The authors mention checklists as part of the solution to the problem of misdiagnosis, and I would suggest that confirmation of initial diagnoses be built into these checklists.

In the case of a presumptive diagnosis of an acute exacerbation of heart failure treated empirically with diuretics, the diagnosis should be confirmed by the next day’s response to the diuretics, ie, increased urine output, a lower respiratory rate, and a fall in the pro-B-type natriuretic peptide level. Moreover, a change in the radiographic appearance should be seen, and respiratory and pulmonary function should improve after the first 24 hours on oxygen supplementation plus diuretics. Daily patient weights are also critical in determining response to a diuretic, and are rarely done accurately. I order weights and review them daily for patients like this.

Second, it is good to look at things yourself, including the patient, medication lists, laboratory values, and radiographic films. The attending physician should look at the radiographs together with a senior radiologist. Seeing no improvement or change on the second hospital day, or seeing signs incompatible with heart failure, one could order computed tomography of the chest and begin to entertain pulmonary diagnoses.

Even vital signs can be questionable. For example, in the case presented here, with a temperature of 99°F, a heart rate of 105, and a pulse oxygenation saturation of 89%, a respiratory rate of 24 seems unbelievably low. In my experience, the respiratory rate is recorded erroneously most of the time unless it is recorded electronically or checked at the bedside by the physician using a timepiece with a sweep second-hand.

Additionally, I have found that ordering several days’ laboratory tests (eg, complete blood cell counts, chemistry panels) in advance, in many cases, risks missing important findings and wastes time, energy, and the patient’s blood. I have learned to evaluate each patient daily and to order the most pertinent laboratory tests. With electronic medical records, I can check laboratory results as soon as they are available.  

To the Editor: I appreciated the article on cognitive biases and diagnostic error by Mull et al in the November 2015 issue.¹ They presented an excellent description of the pitfalls of diagnosis as reflected in a case of a patient misdiagnosed with heart failure who ultimately died of pulmonary tuberculosis complicated by pulmonary embolism (the latter possibly from using the wrong form of heparin). To the points they raised,  I would like to add a few of my own about diagnosis in general and heart failure in particular.

First, any initial diagnosis not confirmed objectively within the first 24 hours should be questioned, and other possibilities should be investigated. I have found this to be essential for every day’s stay in the hospital and for every outpatient visit. The authors mention checklists as part of the solution to the problem of misdiagnosis, and I would suggest that confirmation of initial diagnoses be built into these checklists.

In the case of a presumptive diagnosis of an acute exacerbation of heart failure treated empirically with diuretics, the diagnosis should be confirmed by the next day’s response to the diuretics, ie, increased urine output, a lower respiratory rate, and a fall in the pro-B-type natriuretic peptide level. Moreover, a change in the radiographic appearance should be seen, and respiratory and pulmonary function should improve after the first 24 hours on oxygen supplementation plus diuretics. Daily patient weights are also critical in determining response to a diuretic, and are rarely done accurately. I order weights and review them daily for patients like this.

Second, it is good to look at things yourself, including the patient, medication lists, laboratory values, and radiographic films. The attending physician should look at the radiographs together with a senior radiologist. Seeing no improvement or change on the second hospital day, or seeing signs incompatible with heart failure, one could order computed tomography of the chest and begin to entertain pulmonary diagnoses.

Even vital signs can be questionable. For example, in the case presented here, with a temperature of 99°F, a heart rate of 105, and a pulse oxygenation saturation of 89%, a respiratory rate of 24 seems unbelievably low. In my experience, the respiratory rate is recorded erroneously most of the time unless it is recorded electronically or checked at the bedside by the physician using a timepiece with a sweep second-hand.

Additionally, I have found that ordering several days’ laboratory tests (eg, complete blood cell counts, chemistry panels) in advance, in many cases, risks missing important findings and wastes time, energy, and the patient’s blood. I have learned to evaluate each patient daily and to order the most pertinent laboratory tests. With electronic medical records, I can check laboratory results as soon as they are available.  

In Reply: We thank Dr. Field for his insights and personal observations related to diagnosis and biases that contribute to diagnostic errors.

Dr. Field’s comment about the importance of revisiting one’s initial working diagnosis is consistent with our proposed diagnostic time out. A diagnostic time out can incorporate a short checklist and aid in debiasing clinicians when findings do not fit the case presentation, such as lack of response to diuretic therapy. Being mindful of slowing down and not necessarily rushing to judgment is another important component.¹ Of note, the residents in our case did revisit their initial working diagnosis, as suggested by Dr. Field. Questions from learners have great potential to serve as debiasing instruments and should always be encouraged. Those who do not work with students can do the same by speaking with nurses or other members of the healthcare team, who offer observations that busy physicians might miss.

Our case highlights the problem that we lack objective criteria to diagnose symptomatic heart failure. While B-type natriuretic factor (BNP) has a strong negative predictive value, serial BNP measurements have not been established to be helpful in the management of heart failure.² Although certain findings on chest radiography have strong positive and negative likelihood associations, the role of serial chest radiographs is less clear.³ Thus, heart failure remains a clinical diagnosis in current practice.

As Dr. Field points out, the accuracy and performance characteristics of diagnostic testing, such as the respiratory rate, need to be considered in conjunction with debiasing strategies to achieve higher diagnostic accuracy. Multiple factors can contribute to low-performing or misinterpreted diagnostic tests, and inaccurate vital signs have been shown to be similarly prone to potential error.⁴

Finally, we wholeheartedly agree with Dr. Field’s comment on unnecessary testing.  High-value care is appropriate care. Using Bayesian reasoning to guide testing, monitoring the treatment course appropriately, and eliminating waste is highly likely to improve both value and diagnostic accuracy. Automated, ritual ordering of daily tests can indicate that thinking has been shut off, leaving clinicians susceptible to premature closure of the diagnostic process as well as the potential for “incidentalomas” to distract them from the right diagnosis, all the while leading to low-value care such as wasteful spending, patient dissatisfaction, and hospital-acquired anemia.⁵ We believe that deciding on a daily basis what the next day’s tests will be can be another powerful debiasing habit, one with benefits beyond diagnosis.

In Reply: We thank Dr. Field for his insights and personal observations related to diagnosis and biases that contribute to diagnostic errors.

Dr. Field’s comment about the importance of revisiting one’s initial working diagnosis is consistent with our proposed diagnostic time out. A diagnostic time out can incorporate a short checklist and aid in debiasing clinicians when findings do not fit the case presentation, such as lack of response to diuretic therapy. Being mindful of slowing down and not necessarily rushing to judgment is another important component.¹ Of note, the residents in our case did revisit their initial working diagnosis, as suggested by Dr. Field. Questions from learners have great potential to serve as debiasing instruments and should always be encouraged. Those who do not work with students can do the same by speaking with nurses or other members of the healthcare team, who offer observations that busy physicians might miss.

Our case highlights the problem that we lack objective criteria to diagnose symptomatic heart failure. While B-type natriuretic factor (BNP) has a strong negative predictive value, serial BNP measurements have not been established to be helpful in the management of heart failure.² Although certain findings on chest radiography have strong positive and negative likelihood associations, the role of serial chest radiographs is less clear.³ Thus, heart failure remains a clinical diagnosis in current practice.

As Dr. Field points out, the accuracy and performance characteristics of diagnostic testing, such as the respiratory rate, need to be considered in conjunction with debiasing strategies to achieve higher diagnostic accuracy. Multiple factors can contribute to low-performing or misinterpreted diagnostic tests, and inaccurate vital signs have been shown to be similarly prone to potential error.⁴

Finally, we wholeheartedly agree with Dr. Field’s comment on unnecessary testing.  High-value care is appropriate care. Using Bayesian reasoning to guide testing, monitoring the treatment course appropriately, and eliminating waste is highly likely to improve both value and diagnostic accuracy. Automated, ritual ordering of daily tests can indicate that thinking has been shut off, leaving clinicians susceptible to premature closure of the diagnostic process as well as the potential for “incidentalomas” to distract them from the right diagnosis, all the while leading to low-value care such as wasteful spending, patient dissatisfaction, and hospital-acquired anemia.⁵ We believe that deciding on a daily basis what the next day’s tests will be can be another powerful debiasing habit, one with benefits beyond diagnosis.

In Reply: We thank Dr. Field for his insights and personal observations related to diagnosis and biases that contribute to diagnostic errors.

Dr. Field’s comment about the importance of revisiting one’s initial working diagnosis is consistent with our proposed diagnostic time out. A diagnostic time out can incorporate a short checklist and aid in debiasing clinicians when findings do not fit the case presentation, such as lack of response to diuretic therapy. Being mindful of slowing down and not necessarily rushing to judgment is another important component.¹ Of note, the residents in our case did revisit their initial working diagnosis, as suggested by Dr. Field. Questions from learners have great potential to serve as debiasing instruments and should always be encouraged. Those who do not work with students can do the same by speaking with nurses or other members of the healthcare team, who offer observations that busy physicians might miss.

Our case highlights the problem that we lack objective criteria to diagnose symptomatic heart failure. While B-type natriuretic factor (BNP) has a strong negative predictive value, serial BNP measurements have not been established to be helpful in the management of heart failure.² Although certain findings on chest radiography have strong positive and negative likelihood associations, the role of serial chest radiographs is less clear.³ Thus, heart failure remains a clinical diagnosis in current practice.

As Dr. Field points out, the accuracy and performance characteristics of diagnostic testing, such as the respiratory rate, need to be considered in conjunction with debiasing strategies to achieve higher diagnostic accuracy. Multiple factors can contribute to low-performing or misinterpreted diagnostic tests, and inaccurate vital signs have been shown to be similarly prone to potential error.⁴

Finally, we wholeheartedly agree with Dr. Field’s comment on unnecessary testing.  High-value care is appropriate care. Using Bayesian reasoning to guide testing, monitoring the treatment course appropriately, and eliminating waste is highly likely to improve both value and diagnostic accuracy. Automated, ritual ordering of daily tests can indicate that thinking has been shut off, leaving clinicians susceptible to premature closure of the diagnostic process as well as the potential for “incidentalomas” to distract them from the right diagnosis, all the while leading to low-value care such as wasteful spending, patient dissatisfaction, and hospital-acquired anemia.⁵ We believe that deciding on a daily basis what the next day’s tests will be can be another powerful debiasing habit, one with benefits beyond diagnosis.

To the Editor: I read with great interest the commentary by Lau and Naugler¹ regarding how much allergen-specific immunoglobulin E (IgE) testing is too much. The authors made a number of important conclusions that directly contradict the international consensus statement on IgE antibody test performance published by the Clinical Laboratory Standards Institute (CLSI) in 2009 (2nd edition)² and updated in 2016 (3rd edition) in the I/LA-20 guidance document.³

The most important conclusion of the CLSI I/LA-20 panel was to reaffirm the golden rule of diagnostic allergy testing, which states that allergen-specific IgE antibody detected by either skin testing or serology methods is simply a marker for sensitization and thus only one of many risk factors for allergic disease. IgE positivity is not synonymous with the presence of allergic disease without a positive clinical history.⁴ Clinicians, since the time that IgE was discovered as the reagin in 1967, have tried to use the presence of IgE antibody as detected either by skin testing or serology as the definitive indicator of allergic disease. This is simply inappropriate. Both skin testing and serology are diagnostic tests that indicate sensitization (the presence of IgE antibody) and not disease. The clinician using a positive clinical history of allergic symptoms, objectively collected, must make the link between sensitization (IgE antibody positivity) and allergic disease. 

Lau and Naugler make this same mistake and conclude from their Figure 1 data that “serum antigen-specific IgE testing is not a reliable diagnostic tool.” They use the Wians criterion⁵ of the summed diagnostic sensitivity and specificity of 170 to indicate if a test is clinically useful. They determined the sums of the diagnostic sensitivity and specificity for 89 allergen specificities, most of which they report as below 170. Among the specificities they cover are select aeroallergens, food allergens, venoms, and drugs. Importantly, they use a positive threshold of 0.35 kU/L for only some of their specificities, and they consider a sum of the diagnostic sensitivity and specificity equal to or greater than 170 as clinically relevant.

While Wians’ analysis may have been appropriate for laboratory tests like glucose and even prostate-specific antigen that associate closely with defining a disease state, this criterion is inappropriate for IgE antibody tests that do not directly identify allergic disease. There is peer-reviewed literature on nonreactors based on their clinical history with a validated positive IgE skin test, IgE antibody serology, or food challenge tests.^6,7 Thus, the data in their Figure 1 have no value in defining the performance of IgE antibody tests of sensitization.

Moreover, their report is vague on the actual IgE antibody assay method that was used. This information is important because we know that different IgE assay methods measure different populations of IgE antibody.^2,3 Also, the report does not define whether the participants who provided sera for testing actually had physician-defined allergic disease based on an objective clinical history.

The act of determining optimal cutoff values to maximize the “diagnostic” sensitivity and specificity is appropriate for many laboratory tests, but for allergen-specific IgE antibody analyses, it should be considered inappropriate. These are tests of sensitization, not disease. The IgE antibody result should be reported down to the regulatory-cleared and manufacturer-defined analytical sensitivity, which for the principal IgE antibody autoanalyzers used worldwide is 0.1 kU/L.⁸  These concerns essentially invalidate the conclusions of their report. Unfortunately, they leave the reader with misleading negative impressions about the utility of IgE antibody analyses that are extensively validated methods.

Finally, contrary to the assertions of the authors, current commentaries on the topic of relative diagnostic performance of skin testing and autoanalyzer-based IgE serology tests support the conclusion that, especially for aeroallergens, both the in vivo skin test and the current autoanalyzer-based in vitro serology tests provide overlapping, indistinguishable, and thus comparable diagnostic sensitivity and specificity results.^9,10 Unfortunately, the authors refer to the 2008 Bernstein practice parameter that is out of date in relation to autoanalyzer technology, which has advanced by 2016.

Thus, contrary to the assertions of Lau and Naugler, IgE antibody serology has a clear, well-defined, and positive role in defining sensitization as a key part of the diagnostic workup of a patient who is suspected of having allergic disease. As with any laboratory test, IgE antibody measurements need to be judiciously ordered and used by the clinician only when there is a strong pretest likelihood, based on the patient’s clinical history, of allergic disease.

To the Editor: I read with great interest the commentary by Lau and Naugler¹ regarding how much allergen-specific immunoglobulin E (IgE) testing is too much. The authors made a number of important conclusions that directly contradict the international consensus statement on IgE antibody test performance published by the Clinical Laboratory Standards Institute (CLSI) in 2009 (2nd edition)² and updated in 2016 (3rd edition) in the I/LA-20 guidance document.³

The most important conclusion of the CLSI I/LA-20 panel was to reaffirm the golden rule of diagnostic allergy testing, which states that allergen-specific IgE antibody detected by either skin testing or serology methods is simply a marker for sensitization and thus only one of many risk factors for allergic disease. IgE positivity is not synonymous with the presence of allergic disease without a positive clinical history.⁴ Clinicians, since the time that IgE was discovered as the reagin in 1967, have tried to use the presence of IgE antibody as detected either by skin testing or serology as the definitive indicator of allergic disease. This is simply inappropriate. Both skin testing and serology are diagnostic tests that indicate sensitization (the presence of IgE antibody) and not disease. The clinician using a positive clinical history of allergic symptoms, objectively collected, must make the link between sensitization (IgE antibody positivity) and allergic disease. 

Lau and Naugler make this same mistake and conclude from their Figure 1 data that “serum antigen-specific IgE testing is not a reliable diagnostic tool.” They use the Wians criterion⁵ of the summed diagnostic sensitivity and specificity of 170 to indicate if a test is clinically useful. They determined the sums of the diagnostic sensitivity and specificity for 89 allergen specificities, most of which they report as below 170. Among the specificities they cover are select aeroallergens, food allergens, venoms, and drugs. Importantly, they use a positive threshold of 0.35 kU/L for only some of their specificities, and they consider a sum of the diagnostic sensitivity and specificity equal to or greater than 170 as clinically relevant.

While Wians’ analysis may have been appropriate for laboratory tests like glucose and even prostate-specific antigen that associate closely with defining a disease state, this criterion is inappropriate for IgE antibody tests that do not directly identify allergic disease. There is peer-reviewed literature on nonreactors based on their clinical history with a validated positive IgE skin test, IgE antibody serology, or food challenge tests.^6,7 Thus, the data in their Figure 1 have no value in defining the performance of IgE antibody tests of sensitization.

Moreover, their report is vague on the actual IgE antibody assay method that was used. This information is important because we know that different IgE assay methods measure different populations of IgE antibody.^2,3 Also, the report does not define whether the participants who provided sera for testing actually had physician-defined allergic disease based on an objective clinical history.

The act of determining optimal cutoff values to maximize the “diagnostic” sensitivity and specificity is appropriate for many laboratory tests, but for allergen-specific IgE antibody analyses, it should be considered inappropriate. These are tests of sensitization, not disease. The IgE antibody result should be reported down to the regulatory-cleared and manufacturer-defined analytical sensitivity, which for the principal IgE antibody autoanalyzers used worldwide is 0.1 kU/L.⁸  These concerns essentially invalidate the conclusions of their report. Unfortunately, they leave the reader with misleading negative impressions about the utility of IgE antibody analyses that are extensively validated methods.

Finally, contrary to the assertions of the authors, current commentaries on the topic of relative diagnostic performance of skin testing and autoanalyzer-based IgE serology tests support the conclusion that, especially for aeroallergens, both the in vivo skin test and the current autoanalyzer-based in vitro serology tests provide overlapping, indistinguishable, and thus comparable diagnostic sensitivity and specificity results.^9,10 Unfortunately, the authors refer to the 2008 Bernstein practice parameter that is out of date in relation to autoanalyzer technology, which has advanced by 2016.

Thus, contrary to the assertions of Lau and Naugler, IgE antibody serology has a clear, well-defined, and positive role in defining sensitization as a key part of the diagnostic workup of a patient who is suspected of having allergic disease. As with any laboratory test, IgE antibody measurements need to be judiciously ordered and used by the clinician only when there is a strong pretest likelihood, based on the patient’s clinical history, of allergic disease.

To the Editor: I read with great interest the commentary by Lau and Naugler¹ regarding how much allergen-specific immunoglobulin E (IgE) testing is too much. The authors made a number of important conclusions that directly contradict the international consensus statement on IgE antibody test performance published by the Clinical Laboratory Standards Institute (CLSI) in 2009 (2nd edition)² and updated in 2016 (3rd edition) in the I/LA-20 guidance document.³

The most important conclusion of the CLSI I/LA-20 panel was to reaffirm the golden rule of diagnostic allergy testing, which states that allergen-specific IgE antibody detected by either skin testing or serology methods is simply a marker for sensitization and thus only one of many risk factors for allergic disease. IgE positivity is not synonymous with the presence of allergic disease without a positive clinical history.⁴ Clinicians, since the time that IgE was discovered as the reagin in 1967, have tried to use the presence of IgE antibody as detected either by skin testing or serology as the definitive indicator of allergic disease. This is simply inappropriate. Both skin testing and serology are diagnostic tests that indicate sensitization (the presence of IgE antibody) and not disease. The clinician using a positive clinical history of allergic symptoms, objectively collected, must make the link between sensitization (IgE antibody positivity) and allergic disease. 

Lau and Naugler make this same mistake and conclude from their Figure 1 data that “serum antigen-specific IgE testing is not a reliable diagnostic tool.” They use the Wians criterion⁵ of the summed diagnostic sensitivity and specificity of 170 to indicate if a test is clinically useful. They determined the sums of the diagnostic sensitivity and specificity for 89 allergen specificities, most of which they report as below 170. Among the specificities they cover are select aeroallergens, food allergens, venoms, and drugs. Importantly, they use a positive threshold of 0.35 kU/L for only some of their specificities, and they consider a sum of the diagnostic sensitivity and specificity equal to or greater than 170 as clinically relevant.

While Wians’ analysis may have been appropriate for laboratory tests like glucose and even prostate-specific antigen that associate closely with defining a disease state, this criterion is inappropriate for IgE antibody tests that do not directly identify allergic disease. There is peer-reviewed literature on nonreactors based on their clinical history with a validated positive IgE skin test, IgE antibody serology, or food challenge tests.^6,7 Thus, the data in their Figure 1 have no value in defining the performance of IgE antibody tests of sensitization.

Moreover, their report is vague on the actual IgE antibody assay method that was used. This information is important because we know that different IgE assay methods measure different populations of IgE antibody.^2,3 Also, the report does not define whether the participants who provided sera for testing actually had physician-defined allergic disease based on an objective clinical history.

The act of determining optimal cutoff values to maximize the “diagnostic” sensitivity and specificity is appropriate for many laboratory tests, but for allergen-specific IgE antibody analyses, it should be considered inappropriate. These are tests of sensitization, not disease. The IgE antibody result should be reported down to the regulatory-cleared and manufacturer-defined analytical sensitivity, which for the principal IgE antibody autoanalyzers used worldwide is 0.1 kU/L.⁸  These concerns essentially invalidate the conclusions of their report. Unfortunately, they leave the reader with misleading negative impressions about the utility of IgE antibody analyses that are extensively validated methods.

Finally, contrary to the assertions of the authors, current commentaries on the topic of relative diagnostic performance of skin testing and autoanalyzer-based IgE serology tests support the conclusion that, especially for aeroallergens, both the in vivo skin test and the current autoanalyzer-based in vitro serology tests provide overlapping, indistinguishable, and thus comparable diagnostic sensitivity and specificity results.^9,10 Unfortunately, the authors refer to the 2008 Bernstein practice parameter that is out of date in relation to autoanalyzer technology, which has advanced by 2016.

Thus, contrary to the assertions of Lau and Naugler, IgE antibody serology has a clear, well-defined, and positive role in defining sensitization as a key part of the diagnostic workup of a patient who is suspected of having allergic disease. As with any laboratory test, IgE antibody measurements need to be judiciously ordered and used by the clinician only when there is a strong pretest likelihood, based on the patient’s clinical history, of allergic disease.