Angela Horton, MD

Clinical question: What is the relationship between well-being and demographic factors, educational debt, and medical knowledge in internal-medicine residents?

Background: Physician distress during training is common and can negatively impact patient care. There has never been a study of internal-medicine residents nationally that examined the patterns of distress across demographic factors or the association of these factors with medical knowledge.

Study design: Cross-sectional study.

Setting: U.S. internal-medicine residency programs.

Synopsis: Of the 21,208 U.S. internal-medicine residents who completed the 2008 in-training examination, 77.3% had both survey and demographic data available for analysis. Nearly 15% of these 16,394 residents rated quality of life “as bad as it can be” or “somewhat bad,” and 32.9% felt somewhat or very dissatisfied with work-life balance.

Overall burnout, high levels of weekly emotional exhaustion, and weekly depersonalization were reported by 51.5%, 45.8%, and 28.9% of residents, respectively. Symptoms of emotional exhaustion decreased as training increased, while depersonalization increased after the first postgraduate year. Residents reporting quality of life “as bad as it can be,” emotional exhaustion, or debt greater than $200,000 had mean exam scores 2.7, 4.2, and 5 points, respectively, lower than others surveyed.

Although unlikely given the study design, nonresponse bias could affect these results. Not all demographic variables or domains of well-being were studied, and self-reported educational debt could have been misclassified. Nonetheless, findings suggest that distress remains among residents despite the changes made to duty-hour regulations in 2003.

Bottom line: Suboptimal quality of life and burnout were common among internal-medicine residents nationally; symptoms of burnout were associated with higher debt and lower exam scores.

Citation: West CP, Shanafelt TD, Kolars JC. Quality of life, burnout, educational debt, and medical knowledge among internal medicine residents. JAMA. 2011;306:952-960.

Visit our website for more physician reviews of HM-related research.

Clinical question: What is the relationship between well-being and demographic factors, educational debt, and medical knowledge in internal-medicine residents?

Background: Physician distress during training is common and can negatively impact patient care. There has never been a study of internal-medicine residents nationally that examined the patterns of distress across demographic factors or the association of these factors with medical knowledge.

Study design: Cross-sectional study.

Setting: U.S. internal-medicine residency programs.

Synopsis: Of the 21,208 U.S. internal-medicine residents who completed the 2008 in-training examination, 77.3% had both survey and demographic data available for analysis. Nearly 15% of these 16,394 residents rated quality of life “as bad as it can be” or “somewhat bad,” and 32.9% felt somewhat or very dissatisfied with work-life balance.

Overall burnout, high levels of weekly emotional exhaustion, and weekly depersonalization were reported by 51.5%, 45.8%, and 28.9% of residents, respectively. Symptoms of emotional exhaustion decreased as training increased, while depersonalization increased after the first postgraduate year. Residents reporting quality of life “as bad as it can be,” emotional exhaustion, or debt greater than $200,000 had mean exam scores 2.7, 4.2, and 5 points, respectively, lower than others surveyed.

Although unlikely given the study design, nonresponse bias could affect these results. Not all demographic variables or domains of well-being were studied, and self-reported educational debt could have been misclassified. Nonetheless, findings suggest that distress remains among residents despite the changes made to duty-hour regulations in 2003.

Bottom line: Suboptimal quality of life and burnout were common among internal-medicine residents nationally; symptoms of burnout were associated with higher debt and lower exam scores.

Citation: West CP, Shanafelt TD, Kolars JC. Quality of life, burnout, educational debt, and medical knowledge among internal medicine residents. JAMA. 2011;306:952-960.

Visit our website for more physician reviews of HM-related research.

Clinical question: What is the relationship between well-being and demographic factors, educational debt, and medical knowledge in internal-medicine residents?

Background: Physician distress during training is common and can negatively impact patient care. There has never been a study of internal-medicine residents nationally that examined the patterns of distress across demographic factors or the association of these factors with medical knowledge.

Study design: Cross-sectional study.

Setting: U.S. internal-medicine residency programs.

Synopsis: Of the 21,208 U.S. internal-medicine residents who completed the 2008 in-training examination, 77.3% had both survey and demographic data available for analysis. Nearly 15% of these 16,394 residents rated quality of life “as bad as it can be” or “somewhat bad,” and 32.9% felt somewhat or very dissatisfied with work-life balance.

Overall burnout, high levels of weekly emotional exhaustion, and weekly depersonalization were reported by 51.5%, 45.8%, and 28.9% of residents, respectively. Symptoms of emotional exhaustion decreased as training increased, while depersonalization increased after the first postgraduate year. Residents reporting quality of life “as bad as it can be,” emotional exhaustion, or debt greater than $200,000 had mean exam scores 2.7, 4.2, and 5 points, respectively, lower than others surveyed.

Although unlikely given the study design, nonresponse bias could affect these results. Not all demographic variables or domains of well-being were studied, and self-reported educational debt could have been misclassified. Nonetheless, findings suggest that distress remains among residents despite the changes made to duty-hour regulations in 2003.

Bottom line: Suboptimal quality of life and burnout were common among internal-medicine residents nationally; symptoms of burnout were associated with higher debt and lower exam scores.

Citation: West CP, Shanafelt TD, Kolars JC. Quality of life, burnout, educational debt, and medical knowledge among internal medicine residents. JAMA. 2011;306:952-960.

Visit our website for more physician reviews of HM-related research.

Clinical question: To what extent does diagnostic phlebotomy contribute to hospital-acquired anemia (HAA) during acute myocardial infarction (AMI)?

Background: During AMI, hospital-acquired HAA is associated with higher mortality and poorer health status. Moderate to severe HAA (nadir hemoglobin level <11 g/dL) has been shown to be prognostically important. The contribution of diagnostic phlebotomy blood loss on HAA is unknown and is a potentially modifiable factor.

Study design: Retrospective observational cohort study.

Setting: Fifty-seven U.S. hospitals.

Synopsis: Using Cerner Corp.'s Health Facts database, information was collected on 17,676 patients with AMI. Moderate to severe HAA developed in 3,551 (20%) patients who were not anemic upon admission. The diagnostic blood loss was estimated by assuming minimal blood volume per adult tube required to perform the lab work obtained. The mean phlebotomy volume was higher in patients with HAA compared with patients without HAA (173.8 mL vs. 83.5 mL; P<0.001). There was significant variation of diagnostic blood loss between hospitals. The risk of HAA increased by 18% (RR 1.18; 95% CI, 1.13-1.22) for every 50 mL of diagnostic blood loss.

Patients with HAA were noted to have greater disease severity and comorbidities. No causal inference can be made given the observational nature of the study. Randomized trials are needed to evaluate if strategies to reduce diagnostic blood loss can reduce HAA and improve clinical outcomes for patients with AMI.

Bottom line: Diagnostic blood loss is associated with development of hospital-acquired anemia in patients with acute myocardial infarction.

Citation: Salisbury AC, Reid KJ, Alexander KP, et al. Diagnostic blood loss from phlebotomy and hospital-acquired anemia during acute myocardial infarction. Arch Intern Med. 2011;171:1646-1653.

For more physician reviews of HM-relevant research, visit our website.

Clinical question: To what extent does diagnostic phlebotomy contribute to hospital-acquired anemia (HAA) during acute myocardial infarction (AMI)?

Background: During AMI, hospital-acquired HAA is associated with higher mortality and poorer health status. Moderate to severe HAA (nadir hemoglobin level <11 g/dL) has been shown to be prognostically important. The contribution of diagnostic phlebotomy blood loss on HAA is unknown and is a potentially modifiable factor.

Study design: Retrospective observational cohort study.

Setting: Fifty-seven U.S. hospitals.

Synopsis: Using Cerner Corp.'s Health Facts database, information was collected on 17,676 patients with AMI. Moderate to severe HAA developed in 3,551 (20%) patients who were not anemic upon admission. The diagnostic blood loss was estimated by assuming minimal blood volume per adult tube required to perform the lab work obtained. The mean phlebotomy volume was higher in patients with HAA compared with patients without HAA (173.8 mL vs. 83.5 mL; P<0.001). There was significant variation of diagnostic blood loss between hospitals. The risk of HAA increased by 18% (RR 1.18; 95% CI, 1.13-1.22) for every 50 mL of diagnostic blood loss.

Patients with HAA were noted to have greater disease severity and comorbidities. No causal inference can be made given the observational nature of the study. Randomized trials are needed to evaluate if strategies to reduce diagnostic blood loss can reduce HAA and improve clinical outcomes for patients with AMI.

Bottom line: Diagnostic blood loss is associated with development of hospital-acquired anemia in patients with acute myocardial infarction.

Citation: Salisbury AC, Reid KJ, Alexander KP, et al. Diagnostic blood loss from phlebotomy and hospital-acquired anemia during acute myocardial infarction. Arch Intern Med. 2011;171:1646-1653.

For more physician reviews of HM-relevant research, visit our website.

Clinical question: To what extent does diagnostic phlebotomy contribute to hospital-acquired anemia (HAA) during acute myocardial infarction (AMI)?

Background: During AMI, hospital-acquired HAA is associated with higher mortality and poorer health status. Moderate to severe HAA (nadir hemoglobin level <11 g/dL) has been shown to be prognostically important. The contribution of diagnostic phlebotomy blood loss on HAA is unknown and is a potentially modifiable factor.

Study design: Retrospective observational cohort study.

Setting: Fifty-seven U.S. hospitals.

Synopsis: Using Cerner Corp.'s Health Facts database, information was collected on 17,676 patients with AMI. Moderate to severe HAA developed in 3,551 (20%) patients who were not anemic upon admission. The diagnostic blood loss was estimated by assuming minimal blood volume per adult tube required to perform the lab work obtained. The mean phlebotomy volume was higher in patients with HAA compared with patients without HAA (173.8 mL vs. 83.5 mL; P<0.001). There was significant variation of diagnostic blood loss between hospitals. The risk of HAA increased by 18% (RR 1.18; 95% CI, 1.13-1.22) for every 50 mL of diagnostic blood loss.

Patients with HAA were noted to have greater disease severity and comorbidities. No causal inference can be made given the observational nature of the study. Randomized trials are needed to evaluate if strategies to reduce diagnostic blood loss can reduce HAA and improve clinical outcomes for patients with AMI.

Bottom line: Diagnostic blood loss is associated with development of hospital-acquired anemia in patients with acute myocardial infarction.

Citation: Salisbury AC, Reid KJ, Alexander KP, et al. Diagnostic blood loss from phlebotomy and hospital-acquired anemia during acute myocardial infarction. Arch Intern Med. 2011;171:1646-1653.

For more physician reviews of HM-relevant research, visit our website.

Clinical question: What is the association between time to clinical stability (TCS) and post-discharge death or readmission in patients hospitalized with community-acquired pneumonia (CAP)?

Background: In patients with CAP, inflammatory response during hospitalization might be associated with adverse outcomes after discharge. Studies have not evaluated if time to clinical stability, a reflection of inflammatory response, can be used to identify patients at high risk of adverse outcomes after discharge.

Study design: Retrospective cohort study.

Setting: Veterans Hospital, Louisville, Ky.

Synopsis: Of 464 hospitalized patients with CAP, those with TCS >3 days had a higher rate of readmission or death within 30 days after discharge compared with those who had a TCS =3 days (26% versus 15%; OR 1.98; 95% CI, 1.19-3.3; P=0.008). Longer TCS during hospitalization was associated with a significantly increased risk of adverse outcomes (adjusted OR 1.06, 1.54, 2.40, 10.53 if TCS was reached at days 2, 3, 4, 5 versus Day 1, respectively). The authors proposed that patients with delays in reaching clinical stability should receive a special discharge management approach to decrease the risk of morbidity and mortality after discharge; this may include close observation, home visits, and a follow-up clinic appointment within 10 days.

As a retrospective cohort study, unaccounted-for confounders might exist between TCS and adverse outcomes. The small sample size precluded development of a fully predictive model. Additionally, the population studied was elderly men in a single hospital, which might limit generalizability.

Bottom line: Hospitalized patients with community-acquired pneumonia whose time to clinical stability was greater than three days had a higher risk of readmission or death within 30 days after discharge.

Citation: Aliberti S, Peyrani P, Filardo G, et al. Association between time to clinical stability and outcomes after discharge in hospitalized patients with community-acquired pneumonia. Chest. 2011;140:482-488.

For more physician reviews of HM-relevant research, visit our website.

Clinical question: What is the association between time to clinical stability (TCS) and post-discharge death or readmission in patients hospitalized with community-acquired pneumonia (CAP)?

Background: In patients with CAP, inflammatory response during hospitalization might be associated with adverse outcomes after discharge. Studies have not evaluated if time to clinical stability, a reflection of inflammatory response, can be used to identify patients at high risk of adverse outcomes after discharge.

Study design: Retrospective cohort study.

Setting: Veterans Hospital, Louisville, Ky.

Synopsis: Of 464 hospitalized patients with CAP, those with TCS >3 days had a higher rate of readmission or death within 30 days after discharge compared with those who had a TCS =3 days (26% versus 15%; OR 1.98; 95% CI, 1.19-3.3; P=0.008). Longer TCS during hospitalization was associated with a significantly increased risk of adverse outcomes (adjusted OR 1.06, 1.54, 2.40, 10.53 if TCS was reached at days 2, 3, 4, 5 versus Day 1, respectively). The authors proposed that patients with delays in reaching clinical stability should receive a special discharge management approach to decrease the risk of morbidity and mortality after discharge; this may include close observation, home visits, and a follow-up clinic appointment within 10 days.

As a retrospective cohort study, unaccounted-for confounders might exist between TCS and adverse outcomes. The small sample size precluded development of a fully predictive model. Additionally, the population studied was elderly men in a single hospital, which might limit generalizability.

Bottom line: Hospitalized patients with community-acquired pneumonia whose time to clinical stability was greater than three days had a higher risk of readmission or death within 30 days after discharge.

Citation: Aliberti S, Peyrani P, Filardo G, et al. Association between time to clinical stability and outcomes after discharge in hospitalized patients with community-acquired pneumonia. Chest. 2011;140:482-488.

For more physician reviews of HM-relevant research, visit our website.

Clinical question: What is the association between time to clinical stability (TCS) and post-discharge death or readmission in patients hospitalized with community-acquired pneumonia (CAP)?

Background: In patients with CAP, inflammatory response during hospitalization might be associated with adverse outcomes after discharge. Studies have not evaluated if time to clinical stability, a reflection of inflammatory response, can be used to identify patients at high risk of adverse outcomes after discharge.

Study design: Retrospective cohort study.

Setting: Veterans Hospital, Louisville, Ky.

Synopsis: Of 464 hospitalized patients with CAP, those with TCS >3 days had a higher rate of readmission or death within 30 days after discharge compared with those who had a TCS =3 days (26% versus 15%; OR 1.98; 95% CI, 1.19-3.3; P=0.008). Longer TCS during hospitalization was associated with a significantly increased risk of adverse outcomes (adjusted OR 1.06, 1.54, 2.40, 10.53 if TCS was reached at days 2, 3, 4, 5 versus Day 1, respectively). The authors proposed that patients with delays in reaching clinical stability should receive a special discharge management approach to decrease the risk of morbidity and mortality after discharge; this may include close observation, home visits, and a follow-up clinic appointment within 10 days.

As a retrospective cohort study, unaccounted-for confounders might exist between TCS and adverse outcomes. The small sample size precluded development of a fully predictive model. Additionally, the population studied was elderly men in a single hospital, which might limit generalizability.

Bottom line: Hospitalized patients with community-acquired pneumonia whose time to clinical stability was greater than three days had a higher risk of readmission or death within 30 days after discharge.

Citation: Aliberti S, Peyrani P, Filardo G, et al. Association between time to clinical stability and outcomes after discharge in hospitalized patients with community-acquired pneumonia. Chest. 2011;140:482-488.

For more physician reviews of HM-relevant research, visit our website.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Procalcitonin to guide antimicrobial use in ICU patients
2. Platelet reactivity and event rates after PCI
3. Conservative treatment of necrotizing pancreatitis
4. Use of MRI to diagnose Takotsubo cardiomyopathy
5. Rates of inadvertent medication discontinuation after hospitalization
6. Role of history and physical exam in diagnosing medical illness
7. Red-cell distribution width and rates of mortality

Procalcitonin-Guided Therapy Decreases Antimicrobial Duration in ICU Patients

Clinical question: Can utilization of serum procalcitonin (PCT) levels safely reduce antimicrobial exposure in ICU patients?

Background: Serum PCT levels are elevated in bacterial infections and sepsis and have been used in some settings to guide antimicrobial therapy. Randomized controlled trials have demonstrated reduction of antibiotic use with PCT measurement. This systematic review assessed the safety and effectiveness of PCT measurements in reducing antimicrobial exposure in ICU patients.

Study design: Systematic review.

Setting: Adult medical and surgical ICUs.

Synopsis: A search of MEDLINE and EMBASE yielded 1,018 publications related to PCT, critically ill patients, and antimicrobial therapy. Six randomized controlled trials involving 1,476 patients were reviewed. The duration of antimicrobial use was significantly decreased in all five studies that evaluated treatment duration. The remaining study only assessed the impact of PCT on initiation of antimicrobial therapy and did not demonstrate decreased antimicrobial exposure. Compared with the control group, patients randomized to PCT-guided therapy had relative reductions in duration of first antibiotic course by 21%, down to 38%, and decreases in days of antimicrobial therapy per 1,000 ICU patient days by 20%, down to 23%. PCT intervention also was associated with a 23% to 37% increase in days alive without antimicrobial therapy during the first 28 days. The length of ICU stay was significantly decreased in two studies but was not significantly different in the other studies. There were no significant differences in rates of mortality, infection relapse, or days free of mechanical ventilation.

Bottom line: PCT guidance reduced antimicrobial exposure of ICU patients without increasing rates of mortality or infection relapse.

Citation: Agarwal R, Schwartz DN. Procalcitonin to guide duration of antimicrobial therapy in intensive care units: a systematic review. Clin Infect Dis. 2011;53:379-387.

High Residual Platelet Reactivity Increases Risk of Cardiovascular Events among Patients with Acute Coronary Syndromes Undergoing PCI

Clinical question: Is high residual platelet reactivity (HRPR) in patients receiving clopidogrel associated with increased risk of ischemic events after percutaneous coronary intervention (PCI)?

Background: Studies have demonstrated an increased risk of cardiovascular events associated with HRPR in patients receiving clopidogrel while undergoing PCI. However, these studies have used a variety of platelet function tests, and thresholds for positive tests have not been established. In addition, these studies have enrolled heterogeneous populations with short-term follow-up and few have included patients with acute coronary syndromes (ACSs).

Study design: Prospective cohort study.

Setting: Cardiology service of a referral center in Italy.

Synopsis: This study included 1,789 consecutive patients with ACS undergoing PCI. Patients were given 325 mg aspirin and a 600-mg loading dose of clopidogrel on admission, followed by a daily maintenance dose of aspirin 325 mg and clopidogrel 75 mg for at least six months. Platelet reactivity was measured using adenosine diphosphate (ADP) testing, and those with HRPR (≥70% platelet aggregation) were given increased dosing of clopidogrel or switched to ticlopidine using ADP test guidance. At two-year follow-up, patients with HRPR experienced an increased composite endpoint of cardiac death, myocardial infarction, urgent coronary revascularization, and stroke with a combined event rate of 14.6% in patients with HRPR and 8.7% in patients with low residual platelet reactivity (95% CI, 1.6%-11.1%; P=0.003). Stent thrombosis was also higher in HRPR patients (absolute risk increase 3.2%; 95% CI, 0.4%-6.7%; P=0.01).

This study is nonrandomized, and residual unmeasured confounding cannot be excluded. In addition, use of non-antiplatelet drugs and adherence to recommended drugs might have influenced outcomes.

Bottom line: HRPR is associated with increased risk of ischemic events in patients with ACS receiving antiplatelet agents after PCI.

Citation: Parodi G, Marcucci R, Valenti R, et al. High residual platelet reactivity after clopidogrel loading and long-term cardiovascular events among patients with acute coronary syndromes undergoing PCI. JAMA. 2011;306:1215-1223.

Conservative Treatment of Necrotizing Pancreatitis Is Associated with Improved Outcomes

Clinical question: What are the outcomes of conservative and interventional management of necrotizing pancreatitis?

Background: Open necrosectomy was historically the treatment of choice for necrotizing pancreatitis, but, currently, pancreatic necrosis is only managed invasively when complicated by infection. Other changes in management over time have included a shift in the timing of intervention and the use of minimally invasive techniques. Existing studies do not reflect these changes in practice patterns and have been limited by small sample sizes or the exclusion of important subgroups of patients.

Study design: Prospective cohort study.

Setting: Twenty-one Dutch hospitals.

Synopsis: This study included 639 patients with acute necrotizing pancreatitis confirmed by imaging. Overall mortality was 15%. Conservative treatment was performed in 62% of the patients with a mortality of 7%; however, patients with organ failure (pulmonary, circulatory, and/or renal) who received conservative therapy had a mortality rate of 37%. Intervention (percutaneous drainage, video-assisted retroperitoneal debridement, endoscopic transluminal necrosectomy, laparotomy) in patients with suspected or confirmed infected pancreatic necrosis was performed on 38% of the patients with associated mortality of 27%. Interventions performed within the first 14 days of hospitalization resulted in a 56% mortality rate, whereas interventions performed after Day 29 resulted in a 15% mortality rate (P<0.001). Patients with organ failure experienced significantly greater mortality compared with patients with no organ failure (35% vs. 2%; P<0.001). Primary percutaneous drainage was associated with fewer complications than was primary necrosectomy (42% vs. 64%; P=0.003).

This study was nonrandomized, and final decisions regarding management were left to the treating physician. Notably, while there was no significant difference in APACHE II scores between the conservative and intervention groups, intervention patients had more severe pancreatic disease and scored higher on other measures of disease severity.

Bottom line: Patients with necrotizing pancreatitis can frequently be managed conservatively, though the presence of organ failure and parenchymal necrosis confer poorer prognosis. When intervention is indicated, postponing intervention and utilizing minimally invasive techniques decrease morbidity and mortality.

Citation: Van Santvoort HC, Bakker OJ, Bollen TL, et al. A conservative and minimally invasive approach to necrotizing pancreatitis improves outcome. Gastroenterology. 2011;141:1254-1263.

Cardiac MRI Complements Clinical Findings in Diagnosis of Stress Cardiomyopathy

Clinical question: What are the clinical features and cardiovascular MRI findings in patients with stress (Takotsubo) cardiomyopathy?

Background: Stress cardiomyopathy (SC) is characterized by acute and profound reversible left ventricular dysfunction that is thought to result from increased sympathetic activity related to emotional and/or physical stress. Prior studies evaluating the clinical features of SC were limited by small sample sizes and single-center enrollment, and cardiac MRI use in SC has not been well studied.

Study design: Prospective cohort study.

Setting: Seven North American and European tertiary-care centers.

Synopsis: This study enrolled 256 patients who met diagnostic criteria for SC according to Mayo criteria. Postmenopausal women were most commonly affected; only 11% of participants were men and 8% were women younger than 50 years old. An identifiable stressor was found in 71% of the patients. Clinical presentation was notable for symptoms of acute coronary syndrome (ACS) in 88% of patients, abnormal electrocardiogram in 87%, and elevated troponin T in 90%. Coronary angiography was normal in three-fourths of patients, and no patients had features of acute plaque rupture. Typical apical ballooning was seen on left ventriculography in 82% of patients.

Cardiac MRI findings included severe left ventricular dysfunction in a noncoronary distribution, myocardial edema in areas of regional wall abnormalities, absence of high signal areas in late gadolinium enhancement (e.g., absence of necrosis/fibrosis), and increased early gadolinium uptake (i.e. early inflammation). Repeat cardiac MRI four weeks after initial diagnosis showed near or complete resolution of imaging findings.

Bottom line: Stress cardiomyopathy typically presents like ACS, usually affects postmenopausal women, is often preceded by a stressful event, and is characterized by cardiac MRI findings of regional wall motion abnormalities, reversible myocardial injury, and the absence of fibrosis. Cardiac MRI may be valuable in diagnosing SC in patients who present without the classic clinical features.

Citation: Eitel I, von Knobelsdorff-Brenkenhoff F, Bernhardt P, et al. Clinical characteristics and cardiovascular magnetic resonance findings in stress (Takotsubo) cardiomyopathy. JAMA. 2011;306:277-286.

Increased Risk of Potentially Inadvertent Medication Discontinuation Following Acute-Care Hospitalization

Clinical question: Are medications for chronic diseases inadvertently discontinued after acute-care hospitalization, and is this risk increased in patients who had an ICU stay?

Background: Transitions of care are associated with medical errors. Two such transitions are a shift from the ICU to floor setting and from the inpatient to outpatient setting. Medications for chronic diseases might be held during hospitalization for a variety of reasons, and medication errors may occur if these drugs are not restarted when the acute problem resolves or the patient is discharged from the hospital.

Study design: Population-based cohort study.

Setting: Ontario, Canada.

Synopsis: Using four separate databases, administrative records were reviewed for 396,380 patients aged >65 years who were continuous users of at least one of five evidence-based medication groups for common chronic diseases. Medications included statins, antiplatelet/anticoagulant agents, levothyroxine, respiratory inhalers, and gastric acid suppressants. The primary outcome was potentially unintentional medication discontinuation (measured by failure to renew the prescription at 90 days) for hospitalized versus nonhospitalized patients. All medication groups had statistically significant adjusted odds ratios ranging from 1.18 (95% CI, 1.14-1.23) for discontinuation of levothyroxine to 1.86 (95% CI, 1.77-1.97) for discontinuation of antiplatelet/anticoagulant medications. Treatment in an ICU further increased this risk compared with nonhospitalized patients, and increased the risk for medication discontinuation in four of the five medication groups when compared with patients hospitalized without ICU treatment.

Important study limitations include the lack of appropriate clinical details to classify medication discontinuation as unintentional and the inability of administrative data to prove causality. This study highlights the importance of medication reconciliation and calls attention to inadvertent medication discontinuation during care transitions (see “Reconciliation Act,”).

Bottom line: Patients discharged from the hospital, particularly after ICU treatment, have a higher risk of discontinuation of long-term medications for chronic medical problems when compared with nonhospitalized patients.

Citation: Bell CM, Brener SS, Gunraj N, et al. Association of ICU or hospital admission with unintentional discontinuation of medications for chronic diseases. JAMA. 2011;306:840-847.

Clinical Exam Remains Valuable in the Diagnosis of Patients Admitted to a Medicine Service

Clinical question: Is a clinical exam useful in the diagnosis of newly admitted patients to a general medicine service?

Background: The clinical exam, which comprises the history and physical examination, has long been described as essential to the diagnosis of illness. However, the literature supporting this claim is limited to the ambulatory setting. There has not been evaluation of the clinical exam as a diagnostic tool in the hospital setting, where more advanced testing is readily available.

Study design: Retrospective chart review.

Setting: Urban academic medical center.

Synopsis: The study included 442 adult patients consecutively admitted from the emergency department to the general medicine service who were separately assessed by one senior resident (SR) and one experienced hospital physician (HP) not involved with the case. The SR and HP each made an initial diagnosis and documented the most helpful component(s) in arriving at that diagnosis. Outcomes included comparison of the SR and HP’s admission diagnosis with the discharge diagnosis, and the diagnostic value of the various components of the clinical exam and initial studies.

Compared with the discharge diagnosis, the SR’s initial diagnosis was correct in 80.1% of cases, while the HP was correct in 84.4%. The patient’s history was the most important element in the initial assessment, independently influencing approximately 20% of the correct diagnoses for both physicians. Approximately 60% of correct diagnoses were established using the history and/or physical, and more than 90% were made using a combination of history, physical exam, and/or basic tests (admission labs, electrocardiogram, chest X-ray).

The generalizability of these results is limited by the retrospective, single-center study design, involvement of only one resident physician, and the lack of information regarding number of experienced clinicians and types of diagnoses.

Bottom line: Among patients admitted to a general medicine service, the most powerful tool in obtaining an accurate diagnosis was the combination of a patient’s history and a physical exam.

Citation: Paley L, Zornitzki T, Cohen J, et al. Utility of clinical examination in the diagnosis of emergency department patients admitted to the department of medicine of an academic hospital. Arch Intern Med. 2011;171:1394-1396.

RDW Predicts All-Cause Mortality and Bloodstream Infection in ICU Patients

Clinical question: Among patients admitted to the ICU, is red blood cell distribution width (RDW) a reliable indicator of mortality?

Background: The RDW is an inexpensive test that is commonly included in routine laboratory studies. It has been associated with multiple disease processes and found to be a strong predictor of mortality in the general adult population. However, there has been limited study of the association between RDW and outcomes in critically ill patients.

Study design: Observational cohort study.

Setting: Urban tertiary-care academic medical center.

Synopsis: Data from 51,413 adult patients who received critical care between 1997 and 2007 were obtained from a computerized registry and evaluated for the primary outcome of 30-day mortality after critical-care initiation. Secondary outcomes included 90-day, 365-day, and in-hospital mortality, as well as bloodstream infection. Logistic regression examined both primary and secondary outcomes in association with pre-established RDW quintiles. After multivariable adjustment, RDW was found to be associated with mortality at 30, 90, and 365 days, in addition to in-hospital mortality. The highest RDW quintile (RDW >15.8%) had an adjusted OR of 2.61 (95% CI, 2.37-2.86; P<0.001) for the primary outcome, with similar risk for secondary outcomes of mortality. A subgroup of 18,525 patients with blood culture data was analyzed and an adjusted OR of 1.44 was found in the highest RDW quintile for the secondary outcome of bloodstream infection.

Bottom line: Red blood cell distribution width is a strong independent predictor of all-cause mortality and bloodstream infection in patients receiving intensive care.

Citation: Bazick HS, Chang D, Mahadevappa K, Gibbons FK, Christopher KB. Red cell distribution width and all-cause mortality in critically ill patients. Crit Care Med. 2011;39:1913-1921.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Procalcitonin to guide antimicrobial use in ICU patients
2. Platelet reactivity and event rates after PCI
3. Conservative treatment of necrotizing pancreatitis
4. Use of MRI to diagnose Takotsubo cardiomyopathy
5. Rates of inadvertent medication discontinuation after hospitalization
6. Role of history and physical exam in diagnosing medical illness
7. Red-cell distribution width and rates of mortality

Procalcitonin-Guided Therapy Decreases Antimicrobial Duration in ICU Patients

Clinical question: Can utilization of serum procalcitonin (PCT) levels safely reduce antimicrobial exposure in ICU patients?

Background: Serum PCT levels are elevated in bacterial infections and sepsis and have been used in some settings to guide antimicrobial therapy. Randomized controlled trials have demonstrated reduction of antibiotic use with PCT measurement. This systematic review assessed the safety and effectiveness of PCT measurements in reducing antimicrobial exposure in ICU patients.

Study design: Systematic review.

Setting: Adult medical and surgical ICUs.

Synopsis: A search of MEDLINE and EMBASE yielded 1,018 publications related to PCT, critically ill patients, and antimicrobial therapy. Six randomized controlled trials involving 1,476 patients were reviewed. The duration of antimicrobial use was significantly decreased in all five studies that evaluated treatment duration. The remaining study only assessed the impact of PCT on initiation of antimicrobial therapy and did not demonstrate decreased antimicrobial exposure. Compared with the control group, patients randomized to PCT-guided therapy had relative reductions in duration of first antibiotic course by 21%, down to 38%, and decreases in days of antimicrobial therapy per 1,000 ICU patient days by 20%, down to 23%. PCT intervention also was associated with a 23% to 37% increase in days alive without antimicrobial therapy during the first 28 days. The length of ICU stay was significantly decreased in two studies but was not significantly different in the other studies. There were no significant differences in rates of mortality, infection relapse, or days free of mechanical ventilation.

Bottom line: PCT guidance reduced antimicrobial exposure of ICU patients without increasing rates of mortality or infection relapse.

Citation: Agarwal R, Schwartz DN. Procalcitonin to guide duration of antimicrobial therapy in intensive care units: a systematic review. Clin Infect Dis. 2011;53:379-387.

High Residual Platelet Reactivity Increases Risk of Cardiovascular Events among Patients with Acute Coronary Syndromes Undergoing PCI

Clinical question: Is high residual platelet reactivity (HRPR) in patients receiving clopidogrel associated with increased risk of ischemic events after percutaneous coronary intervention (PCI)?

Background: Studies have demonstrated an increased risk of cardiovascular events associated with HRPR in patients receiving clopidogrel while undergoing PCI. However, these studies have used a variety of platelet function tests, and thresholds for positive tests have not been established. In addition, these studies have enrolled heterogeneous populations with short-term follow-up and few have included patients with acute coronary syndromes (ACSs).

Study design: Prospective cohort study.

Setting: Cardiology service of a referral center in Italy.

Synopsis: This study included 1,789 consecutive patients with ACS undergoing PCI. Patients were given 325 mg aspirin and a 600-mg loading dose of clopidogrel on admission, followed by a daily maintenance dose of aspirin 325 mg and clopidogrel 75 mg for at least six months. Platelet reactivity was measured using adenosine diphosphate (ADP) testing, and those with HRPR (≥70% platelet aggregation) were given increased dosing of clopidogrel or switched to ticlopidine using ADP test guidance. At two-year follow-up, patients with HRPR experienced an increased composite endpoint of cardiac death, myocardial infarction, urgent coronary revascularization, and stroke with a combined event rate of 14.6% in patients with HRPR and 8.7% in patients with low residual platelet reactivity (95% CI, 1.6%-11.1%; P=0.003). Stent thrombosis was also higher in HRPR patients (absolute risk increase 3.2%; 95% CI, 0.4%-6.7%; P=0.01).

This study is nonrandomized, and residual unmeasured confounding cannot be excluded. In addition, use of non-antiplatelet drugs and adherence to recommended drugs might have influenced outcomes.

Bottom line: HRPR is associated with increased risk of ischemic events in patients with ACS receiving antiplatelet agents after PCI.

Citation: Parodi G, Marcucci R, Valenti R, et al. High residual platelet reactivity after clopidogrel loading and long-term cardiovascular events among patients with acute coronary syndromes undergoing PCI. JAMA. 2011;306:1215-1223.

Conservative Treatment of Necrotizing Pancreatitis Is Associated with Improved Outcomes

Clinical question: What are the outcomes of conservative and interventional management of necrotizing pancreatitis?

Background: Open necrosectomy was historically the treatment of choice for necrotizing pancreatitis, but, currently, pancreatic necrosis is only managed invasively when complicated by infection. Other changes in management over time have included a shift in the timing of intervention and the use of minimally invasive techniques. Existing studies do not reflect these changes in practice patterns and have been limited by small sample sizes or the exclusion of important subgroups of patients.

Study design: Prospective cohort study.

Setting: Twenty-one Dutch hospitals.

Synopsis: This study included 639 patients with acute necrotizing pancreatitis confirmed by imaging. Overall mortality was 15%. Conservative treatment was performed in 62% of the patients with a mortality of 7%; however, patients with organ failure (pulmonary, circulatory, and/or renal) who received conservative therapy had a mortality rate of 37%. Intervention (percutaneous drainage, video-assisted retroperitoneal debridement, endoscopic transluminal necrosectomy, laparotomy) in patients with suspected or confirmed infected pancreatic necrosis was performed on 38% of the patients with associated mortality of 27%. Interventions performed within the first 14 days of hospitalization resulted in a 56% mortality rate, whereas interventions performed after Day 29 resulted in a 15% mortality rate (P<0.001). Patients with organ failure experienced significantly greater mortality compared with patients with no organ failure (35% vs. 2%; P<0.001). Primary percutaneous drainage was associated with fewer complications than was primary necrosectomy (42% vs. 64%; P=0.003).

This study was nonrandomized, and final decisions regarding management were left to the treating physician. Notably, while there was no significant difference in APACHE II scores between the conservative and intervention groups, intervention patients had more severe pancreatic disease and scored higher on other measures of disease severity.

Bottom line: Patients with necrotizing pancreatitis can frequently be managed conservatively, though the presence of organ failure and parenchymal necrosis confer poorer prognosis. When intervention is indicated, postponing intervention and utilizing minimally invasive techniques decrease morbidity and mortality.

Citation: Van Santvoort HC, Bakker OJ, Bollen TL, et al. A conservative and minimally invasive approach to necrotizing pancreatitis improves outcome. Gastroenterology. 2011;141:1254-1263.

Cardiac MRI Complements Clinical Findings in Diagnosis of Stress Cardiomyopathy

Clinical question: What are the clinical features and cardiovascular MRI findings in patients with stress (Takotsubo) cardiomyopathy?

Background: Stress cardiomyopathy (SC) is characterized by acute and profound reversible left ventricular dysfunction that is thought to result from increased sympathetic activity related to emotional and/or physical stress. Prior studies evaluating the clinical features of SC were limited by small sample sizes and single-center enrollment, and cardiac MRI use in SC has not been well studied.

Study design: Prospective cohort study.

Setting: Seven North American and European tertiary-care centers.

Synopsis: This study enrolled 256 patients who met diagnostic criteria for SC according to Mayo criteria. Postmenopausal women were most commonly affected; only 11% of participants were men and 8% were women younger than 50 years old. An identifiable stressor was found in 71% of the patients. Clinical presentation was notable for symptoms of acute coronary syndrome (ACS) in 88% of patients, abnormal electrocardiogram in 87%, and elevated troponin T in 90%. Coronary angiography was normal in three-fourths of patients, and no patients had features of acute plaque rupture. Typical apical ballooning was seen on left ventriculography in 82% of patients.

Cardiac MRI findings included severe left ventricular dysfunction in a noncoronary distribution, myocardial edema in areas of regional wall abnormalities, absence of high signal areas in late gadolinium enhancement (e.g., absence of necrosis/fibrosis), and increased early gadolinium uptake (i.e. early inflammation). Repeat cardiac MRI four weeks after initial diagnosis showed near or complete resolution of imaging findings.

Bottom line: Stress cardiomyopathy typically presents like ACS, usually affects postmenopausal women, is often preceded by a stressful event, and is characterized by cardiac MRI findings of regional wall motion abnormalities, reversible myocardial injury, and the absence of fibrosis. Cardiac MRI may be valuable in diagnosing SC in patients who present without the classic clinical features.

Citation: Eitel I, von Knobelsdorff-Brenkenhoff F, Bernhardt P, et al. Clinical characteristics and cardiovascular magnetic resonance findings in stress (Takotsubo) cardiomyopathy. JAMA. 2011;306:277-286.

Increased Risk of Potentially Inadvertent Medication Discontinuation Following Acute-Care Hospitalization

Clinical question: Are medications for chronic diseases inadvertently discontinued after acute-care hospitalization, and is this risk increased in patients who had an ICU stay?

Background: Transitions of care are associated with medical errors. Two such transitions are a shift from the ICU to floor setting and from the inpatient to outpatient setting. Medications for chronic diseases might be held during hospitalization for a variety of reasons, and medication errors may occur if these drugs are not restarted when the acute problem resolves or the patient is discharged from the hospital.

Study design: Population-based cohort study.

Setting: Ontario, Canada.

Synopsis: Using four separate databases, administrative records were reviewed for 396,380 patients aged >65 years who were continuous users of at least one of five evidence-based medication groups for common chronic diseases. Medications included statins, antiplatelet/anticoagulant agents, levothyroxine, respiratory inhalers, and gastric acid suppressants. The primary outcome was potentially unintentional medication discontinuation (measured by failure to renew the prescription at 90 days) for hospitalized versus nonhospitalized patients. All medication groups had statistically significant adjusted odds ratios ranging from 1.18 (95% CI, 1.14-1.23) for discontinuation of levothyroxine to 1.86 (95% CI, 1.77-1.97) for discontinuation of antiplatelet/anticoagulant medications. Treatment in an ICU further increased this risk compared with nonhospitalized patients, and increased the risk for medication discontinuation in four of the five medication groups when compared with patients hospitalized without ICU treatment.

Important study limitations include the lack of appropriate clinical details to classify medication discontinuation as unintentional and the inability of administrative data to prove causality. This study highlights the importance of medication reconciliation and calls attention to inadvertent medication discontinuation during care transitions (see “Reconciliation Act,”).

Bottom line: Patients discharged from the hospital, particularly after ICU treatment, have a higher risk of discontinuation of long-term medications for chronic medical problems when compared with nonhospitalized patients.

Citation: Bell CM, Brener SS, Gunraj N, et al. Association of ICU or hospital admission with unintentional discontinuation of medications for chronic diseases. JAMA. 2011;306:840-847.

Clinical Exam Remains Valuable in the Diagnosis of Patients Admitted to a Medicine Service

Clinical question: Is a clinical exam useful in the diagnosis of newly admitted patients to a general medicine service?

Background: The clinical exam, which comprises the history and physical examination, has long been described as essential to the diagnosis of illness. However, the literature supporting this claim is limited to the ambulatory setting. There has not been evaluation of the clinical exam as a diagnostic tool in the hospital setting, where more advanced testing is readily available.

Study design: Retrospective chart review.

Setting: Urban academic medical center.

Synopsis: The study included 442 adult patients consecutively admitted from the emergency department to the general medicine service who were separately assessed by one senior resident (SR) and one experienced hospital physician (HP) not involved with the case. The SR and HP each made an initial diagnosis and documented the most helpful component(s) in arriving at that diagnosis. Outcomes included comparison of the SR and HP’s admission diagnosis with the discharge diagnosis, and the diagnostic value of the various components of the clinical exam and initial studies.

Compared with the discharge diagnosis, the SR’s initial diagnosis was correct in 80.1% of cases, while the HP was correct in 84.4%. The patient’s history was the most important element in the initial assessment, independently influencing approximately 20% of the correct diagnoses for both physicians. Approximately 60% of correct diagnoses were established using the history and/or physical, and more than 90% were made using a combination of history, physical exam, and/or basic tests (admission labs, electrocardiogram, chest X-ray).

The generalizability of these results is limited by the retrospective, single-center study design, involvement of only one resident physician, and the lack of information regarding number of experienced clinicians and types of diagnoses.

Bottom line: Among patients admitted to a general medicine service, the most powerful tool in obtaining an accurate diagnosis was the combination of a patient’s history and a physical exam.

Citation: Paley L, Zornitzki T, Cohen J, et al. Utility of clinical examination in the diagnosis of emergency department patients admitted to the department of medicine of an academic hospital. Arch Intern Med. 2011;171:1394-1396.

RDW Predicts All-Cause Mortality and Bloodstream Infection in ICU Patients

Clinical question: Among patients admitted to the ICU, is red blood cell distribution width (RDW) a reliable indicator of mortality?

Background: The RDW is an inexpensive test that is commonly included in routine laboratory studies. It has been associated with multiple disease processes and found to be a strong predictor of mortality in the general adult population. However, there has been limited study of the association between RDW and outcomes in critically ill patients.

Study design: Observational cohort study.

Setting: Urban tertiary-care academic medical center.

Synopsis: Data from 51,413 adult patients who received critical care between 1997 and 2007 were obtained from a computerized registry and evaluated for the primary outcome of 30-day mortality after critical-care initiation. Secondary outcomes included 90-day, 365-day, and in-hospital mortality, as well as bloodstream infection. Logistic regression examined both primary and secondary outcomes in association with pre-established RDW quintiles. After multivariable adjustment, RDW was found to be associated with mortality at 30, 90, and 365 days, in addition to in-hospital mortality. The highest RDW quintile (RDW >15.8%) had an adjusted OR of 2.61 (95% CI, 2.37-2.86; P<0.001) for the primary outcome, with similar risk for secondary outcomes of mortality. A subgroup of 18,525 patients with blood culture data was analyzed and an adjusted OR of 1.44 was found in the highest RDW quintile for the secondary outcome of bloodstream infection.

Bottom line: Red blood cell distribution width is a strong independent predictor of all-cause mortality and bloodstream infection in patients receiving intensive care.

Citation: Bazick HS, Chang D, Mahadevappa K, Gibbons FK, Christopher KB. Red cell distribution width and all-cause mortality in critically ill patients. Crit Care Med. 2011;39:1913-1921.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Procalcitonin to guide antimicrobial use in ICU patients
2. Platelet reactivity and event rates after PCI
3. Conservative treatment of necrotizing pancreatitis
4. Use of MRI to diagnose Takotsubo cardiomyopathy
5. Rates of inadvertent medication discontinuation after hospitalization
6. Role of history and physical exam in diagnosing medical illness
7. Red-cell distribution width and rates of mortality

Procalcitonin-Guided Therapy Decreases Antimicrobial Duration in ICU Patients

Clinical question: Can utilization of serum procalcitonin (PCT) levels safely reduce antimicrobial exposure in ICU patients?

Background: Serum PCT levels are elevated in bacterial infections and sepsis and have been used in some settings to guide antimicrobial therapy. Randomized controlled trials have demonstrated reduction of antibiotic use with PCT measurement. This systematic review assessed the safety and effectiveness of PCT measurements in reducing antimicrobial exposure in ICU patients.

Study design: Systematic review.

Setting: Adult medical and surgical ICUs.

Synopsis: A search of MEDLINE and EMBASE yielded 1,018 publications related to PCT, critically ill patients, and antimicrobial therapy. Six randomized controlled trials involving 1,476 patients were reviewed. The duration of antimicrobial use was significantly decreased in all five studies that evaluated treatment duration. The remaining study only assessed the impact of PCT on initiation of antimicrobial therapy and did not demonstrate decreased antimicrobial exposure. Compared with the control group, patients randomized to PCT-guided therapy had relative reductions in duration of first antibiotic course by 21%, down to 38%, and decreases in days of antimicrobial therapy per 1,000 ICU patient days by 20%, down to 23%. PCT intervention also was associated with a 23% to 37% increase in days alive without antimicrobial therapy during the first 28 days. The length of ICU stay was significantly decreased in two studies but was not significantly different in the other studies. There were no significant differences in rates of mortality, infection relapse, or days free of mechanical ventilation.

Bottom line: PCT guidance reduced antimicrobial exposure of ICU patients without increasing rates of mortality or infection relapse.

Citation: Agarwal R, Schwartz DN. Procalcitonin to guide duration of antimicrobial therapy in intensive care units: a systematic review. Clin Infect Dis. 2011;53:379-387.

High Residual Platelet Reactivity Increases Risk of Cardiovascular Events among Patients with Acute Coronary Syndromes Undergoing PCI

Clinical question: Is high residual platelet reactivity (HRPR) in patients receiving clopidogrel associated with increased risk of ischemic events after percutaneous coronary intervention (PCI)?

Background: Studies have demonstrated an increased risk of cardiovascular events associated with HRPR in patients receiving clopidogrel while undergoing PCI. However, these studies have used a variety of platelet function tests, and thresholds for positive tests have not been established. In addition, these studies have enrolled heterogeneous populations with short-term follow-up and few have included patients with acute coronary syndromes (ACSs).

Study design: Prospective cohort study.

Setting: Cardiology service of a referral center in Italy.

Synopsis: This study included 1,789 consecutive patients with ACS undergoing PCI. Patients were given 325 mg aspirin and a 600-mg loading dose of clopidogrel on admission, followed by a daily maintenance dose of aspirin 325 mg and clopidogrel 75 mg for at least six months. Platelet reactivity was measured using adenosine diphosphate (ADP) testing, and those with HRPR (≥70% platelet aggregation) were given increased dosing of clopidogrel or switched to ticlopidine using ADP test guidance. At two-year follow-up, patients with HRPR experienced an increased composite endpoint of cardiac death, myocardial infarction, urgent coronary revascularization, and stroke with a combined event rate of 14.6% in patients with HRPR and 8.7% in patients with low residual platelet reactivity (95% CI, 1.6%-11.1%; P=0.003). Stent thrombosis was also higher in HRPR patients (absolute risk increase 3.2%; 95% CI, 0.4%-6.7%; P=0.01).

This study is nonrandomized, and residual unmeasured confounding cannot be excluded. In addition, use of non-antiplatelet drugs and adherence to recommended drugs might have influenced outcomes.

Bottom line: HRPR is associated with increased risk of ischemic events in patients with ACS receiving antiplatelet agents after PCI.

Citation: Parodi G, Marcucci R, Valenti R, et al. High residual platelet reactivity after clopidogrel loading and long-term cardiovascular events among patients with acute coronary syndromes undergoing PCI. JAMA. 2011;306:1215-1223.

Conservative Treatment of Necrotizing Pancreatitis Is Associated with Improved Outcomes

Clinical question: What are the outcomes of conservative and interventional management of necrotizing pancreatitis?

Background: Open necrosectomy was historically the treatment of choice for necrotizing pancreatitis, but, currently, pancreatic necrosis is only managed invasively when complicated by infection. Other changes in management over time have included a shift in the timing of intervention and the use of minimally invasive techniques. Existing studies do not reflect these changes in practice patterns and have been limited by small sample sizes or the exclusion of important subgroups of patients.

Study design: Prospective cohort study.

Setting: Twenty-one Dutch hospitals.

Synopsis: This study included 639 patients with acute necrotizing pancreatitis confirmed by imaging. Overall mortality was 15%. Conservative treatment was performed in 62% of the patients with a mortality of 7%; however, patients with organ failure (pulmonary, circulatory, and/or renal) who received conservative therapy had a mortality rate of 37%. Intervention (percutaneous drainage, video-assisted retroperitoneal debridement, endoscopic transluminal necrosectomy, laparotomy) in patients with suspected or confirmed infected pancreatic necrosis was performed on 38% of the patients with associated mortality of 27%. Interventions performed within the first 14 days of hospitalization resulted in a 56% mortality rate, whereas interventions performed after Day 29 resulted in a 15% mortality rate (P<0.001). Patients with organ failure experienced significantly greater mortality compared with patients with no organ failure (35% vs. 2%; P<0.001). Primary percutaneous drainage was associated with fewer complications than was primary necrosectomy (42% vs. 64%; P=0.003).

This study was nonrandomized, and final decisions regarding management were left to the treating physician. Notably, while there was no significant difference in APACHE II scores between the conservative and intervention groups, intervention patients had more severe pancreatic disease and scored higher on other measures of disease severity.

Bottom line: Patients with necrotizing pancreatitis can frequently be managed conservatively, though the presence of organ failure and parenchymal necrosis confer poorer prognosis. When intervention is indicated, postponing intervention and utilizing minimally invasive techniques decrease morbidity and mortality.

Citation: Van Santvoort HC, Bakker OJ, Bollen TL, et al. A conservative and minimally invasive approach to necrotizing pancreatitis improves outcome. Gastroenterology. 2011;141:1254-1263.

Cardiac MRI Complements Clinical Findings in Diagnosis of Stress Cardiomyopathy

Clinical question: What are the clinical features and cardiovascular MRI findings in patients with stress (Takotsubo) cardiomyopathy?

Background: Stress cardiomyopathy (SC) is characterized by acute and profound reversible left ventricular dysfunction that is thought to result from increased sympathetic activity related to emotional and/or physical stress. Prior studies evaluating the clinical features of SC were limited by small sample sizes and single-center enrollment, and cardiac MRI use in SC has not been well studied.

Study design: Prospective cohort study.

Setting: Seven North American and European tertiary-care centers.

Synopsis: This study enrolled 256 patients who met diagnostic criteria for SC according to Mayo criteria. Postmenopausal women were most commonly affected; only 11% of participants were men and 8% were women younger than 50 years old. An identifiable stressor was found in 71% of the patients. Clinical presentation was notable for symptoms of acute coronary syndrome (ACS) in 88% of patients, abnormal electrocardiogram in 87%, and elevated troponin T in 90%. Coronary angiography was normal in three-fourths of patients, and no patients had features of acute plaque rupture. Typical apical ballooning was seen on left ventriculography in 82% of patients.

Cardiac MRI findings included severe left ventricular dysfunction in a noncoronary distribution, myocardial edema in areas of regional wall abnormalities, absence of high signal areas in late gadolinium enhancement (e.g., absence of necrosis/fibrosis), and increased early gadolinium uptake (i.e. early inflammation). Repeat cardiac MRI four weeks after initial diagnosis showed near or complete resolution of imaging findings.

Bottom line: Stress cardiomyopathy typically presents like ACS, usually affects postmenopausal women, is often preceded by a stressful event, and is characterized by cardiac MRI findings of regional wall motion abnormalities, reversible myocardial injury, and the absence of fibrosis. Cardiac MRI may be valuable in diagnosing SC in patients who present without the classic clinical features.

Citation: Eitel I, von Knobelsdorff-Brenkenhoff F, Bernhardt P, et al. Clinical characteristics and cardiovascular magnetic resonance findings in stress (Takotsubo) cardiomyopathy. JAMA. 2011;306:277-286.

Increased Risk of Potentially Inadvertent Medication Discontinuation Following Acute-Care Hospitalization

Clinical question: Are medications for chronic diseases inadvertently discontinued after acute-care hospitalization, and is this risk increased in patients who had an ICU stay?

Background: Transitions of care are associated with medical errors. Two such transitions are a shift from the ICU to floor setting and from the inpatient to outpatient setting. Medications for chronic diseases might be held during hospitalization for a variety of reasons, and medication errors may occur if these drugs are not restarted when the acute problem resolves or the patient is discharged from the hospital.

Study design: Population-based cohort study.

Setting: Ontario, Canada.

Synopsis: Using four separate databases, administrative records were reviewed for 396,380 patients aged >65 years who were continuous users of at least one of five evidence-based medication groups for common chronic diseases. Medications included statins, antiplatelet/anticoagulant agents, levothyroxine, respiratory inhalers, and gastric acid suppressants. The primary outcome was potentially unintentional medication discontinuation (measured by failure to renew the prescription at 90 days) for hospitalized versus nonhospitalized patients. All medication groups had statistically significant adjusted odds ratios ranging from 1.18 (95% CI, 1.14-1.23) for discontinuation of levothyroxine to 1.86 (95% CI, 1.77-1.97) for discontinuation of antiplatelet/anticoagulant medications. Treatment in an ICU further increased this risk compared with nonhospitalized patients, and increased the risk for medication discontinuation in four of the five medication groups when compared with patients hospitalized without ICU treatment.

Important study limitations include the lack of appropriate clinical details to classify medication discontinuation as unintentional and the inability of administrative data to prove causality. This study highlights the importance of medication reconciliation and calls attention to inadvertent medication discontinuation during care transitions (see “Reconciliation Act,”).

Bottom line: Patients discharged from the hospital, particularly after ICU treatment, have a higher risk of discontinuation of long-term medications for chronic medical problems when compared with nonhospitalized patients.

Citation: Bell CM, Brener SS, Gunraj N, et al. Association of ICU or hospital admission with unintentional discontinuation of medications for chronic diseases. JAMA. 2011;306:840-847.

Clinical Exam Remains Valuable in the Diagnosis of Patients Admitted to a Medicine Service

Clinical question: Is a clinical exam useful in the diagnosis of newly admitted patients to a general medicine service?

Background: The clinical exam, which comprises the history and physical examination, has long been described as essential to the diagnosis of illness. However, the literature supporting this claim is limited to the ambulatory setting. There has not been evaluation of the clinical exam as a diagnostic tool in the hospital setting, where more advanced testing is readily available.

Study design: Retrospective chart review.

Setting: Urban academic medical center.

Synopsis: The study included 442 adult patients consecutively admitted from the emergency department to the general medicine service who were separately assessed by one senior resident (SR) and one experienced hospital physician (HP) not involved with the case. The SR and HP each made an initial diagnosis and documented the most helpful component(s) in arriving at that diagnosis. Outcomes included comparison of the SR and HP’s admission diagnosis with the discharge diagnosis, and the diagnostic value of the various components of the clinical exam and initial studies.

Compared with the discharge diagnosis, the SR’s initial diagnosis was correct in 80.1% of cases, while the HP was correct in 84.4%. The patient’s history was the most important element in the initial assessment, independently influencing approximately 20% of the correct diagnoses for both physicians. Approximately 60% of correct diagnoses were established using the history and/or physical, and more than 90% were made using a combination of history, physical exam, and/or basic tests (admission labs, electrocardiogram, chest X-ray).

The generalizability of these results is limited by the retrospective, single-center study design, involvement of only one resident physician, and the lack of information regarding number of experienced clinicians and types of diagnoses.

Bottom line: Among patients admitted to a general medicine service, the most powerful tool in obtaining an accurate diagnosis was the combination of a patient’s history and a physical exam.

Citation: Paley L, Zornitzki T, Cohen J, et al. Utility of clinical examination in the diagnosis of emergency department patients admitted to the department of medicine of an academic hospital. Arch Intern Med. 2011;171:1394-1396.

RDW Predicts All-Cause Mortality and Bloodstream Infection in ICU Patients

Clinical question: Among patients admitted to the ICU, is red blood cell distribution width (RDW) a reliable indicator of mortality?

Background: The RDW is an inexpensive test that is commonly included in routine laboratory studies. It has been associated with multiple disease processes and found to be a strong predictor of mortality in the general adult population. However, there has been limited study of the association between RDW and outcomes in critically ill patients.

Study design: Observational cohort study.

Setting: Urban tertiary-care academic medical center.

Synopsis: Data from 51,413 adult patients who received critical care between 1997 and 2007 were obtained from a computerized registry and evaluated for the primary outcome of 30-day mortality after critical-care initiation. Secondary outcomes included 90-day, 365-day, and in-hospital mortality, as well as bloodstream infection. Logistic regression examined both primary and secondary outcomes in association with pre-established RDW quintiles. After multivariable adjustment, RDW was found to be associated with mortality at 30, 90, and 365 days, in addition to in-hospital mortality. The highest RDW quintile (RDW >15.8%) had an adjusted OR of 2.61 (95% CI, 2.37-2.86; P<0.001) for the primary outcome, with similar risk for secondary outcomes of mortality. A subgroup of 18,525 patients with blood culture data was analyzed and an adjusted OR of 1.44 was found in the highest RDW quintile for the secondary outcome of bloodstream infection.

Bottom line: Red blood cell distribution width is a strong independent predictor of all-cause mortality and bloodstream infection in patients receiving intensive care.

Citation: Bazick HS, Chang D, Mahadevappa K, Gibbons FK, Christopher KB. Red cell distribution width and all-cause mortality in critically ill patients. Crit Care Med. 2011;39:1913-1921.