Elizabeth Wantuch, MD, MS

Clinical question: Does fluid management guided by daily plasma natriuretic peptide-driven (BNP) levels in mechanically ventilated patients improve weaning outcomes compared with usual therapy dictated by clinical acumen?

Background: Ventilator weaning contributes at least 40% of the total duration of mechanical ventilation; strategies aimed at optimizing this process could provide substantial benefit. Previous studies have demonstrated that BNP levels prior to ventilator weaning independently predict weaning failure. No current objective practical guide to fluid management during ventilator weaning exists.

Study design: Randomized controlled trial.

Setting: Multiple international centers.

Synopsis: Three hundred four patients who met specific inclusion and exclusion criteria were randomized to either a BNP-driven or physician-guided strategy for fluid management during ventilator weaning. Patients with renal failure were excluded because of the influence of renal function on BNP levels.

All patients in both groups were ventilated with an automatic computer-driven weaning system to standardize the weaning process. In the BNP-driven group, diuretic use was higher, resulting in a more negative fluid balance and significantly shorter time to successful extubation (58.6 hours vs. 42.2 hours, P=0.03). The effect on weaning time was strongest in patients with left ventricular systolic dysfunction, whereas those with COPD seemed less likely to benefit. The two groups did not differ in baseline characteristics, length of stay, mortality, or development of adverse outcomes of renal failure, shock, or electrolyte disturbances.

Bottom line: Compared with physician-guided fluid management, a BNP-driven fluid management protocol decreased duration of ventilator weaning without significant differences in adverse events, mortality rate, or length of stay between the two groups.

Citation: Dessap AM, Roche-Campo F, Kouatchet A, et al. Natriuretic peptide-driven fluid management during ventilator weaning. Am J Respir Crit Care Med. 2012;186(12):1256-1263.

Visit our website for more physician reviews of recent HM-relevant literature.

Clinical question: Does fluid management guided by daily plasma natriuretic peptide-driven (BNP) levels in mechanically ventilated patients improve weaning outcomes compared with usual therapy dictated by clinical acumen?

Background: Ventilator weaning contributes at least 40% of the total duration of mechanical ventilation; strategies aimed at optimizing this process could provide substantial benefit. Previous studies have demonstrated that BNP levels prior to ventilator weaning independently predict weaning failure. No current objective practical guide to fluid management during ventilator weaning exists.

Study design: Randomized controlled trial.

Setting: Multiple international centers.

Synopsis: Three hundred four patients who met specific inclusion and exclusion criteria were randomized to either a BNP-driven or physician-guided strategy for fluid management during ventilator weaning. Patients with renal failure were excluded because of the influence of renal function on BNP levels.

All patients in both groups were ventilated with an automatic computer-driven weaning system to standardize the weaning process. In the BNP-driven group, diuretic use was higher, resulting in a more negative fluid balance and significantly shorter time to successful extubation (58.6 hours vs. 42.2 hours, P=0.03). The effect on weaning time was strongest in patients with left ventricular systolic dysfunction, whereas those with COPD seemed less likely to benefit. The two groups did not differ in baseline characteristics, length of stay, mortality, or development of adverse outcomes of renal failure, shock, or electrolyte disturbances.

Bottom line: Compared with physician-guided fluid management, a BNP-driven fluid management protocol decreased duration of ventilator weaning without significant differences in adverse events, mortality rate, or length of stay between the two groups.

Citation: Dessap AM, Roche-Campo F, Kouatchet A, et al. Natriuretic peptide-driven fluid management during ventilator weaning. Am J Respir Crit Care Med. 2012;186(12):1256-1263.

Visit our website for more physician reviews of recent HM-relevant literature.

Clinical question: Does fluid management guided by daily plasma natriuretic peptide-driven (BNP) levels in mechanically ventilated patients improve weaning outcomes compared with usual therapy dictated by clinical acumen?

Background: Ventilator weaning contributes at least 40% of the total duration of mechanical ventilation; strategies aimed at optimizing this process could provide substantial benefit. Previous studies have demonstrated that BNP levels prior to ventilator weaning independently predict weaning failure. No current objective practical guide to fluid management during ventilator weaning exists.

Study design: Randomized controlled trial.

Setting: Multiple international centers.

Synopsis: Three hundred four patients who met specific inclusion and exclusion criteria were randomized to either a BNP-driven or physician-guided strategy for fluid management during ventilator weaning. Patients with renal failure were excluded because of the influence of renal function on BNP levels.

All patients in both groups were ventilated with an automatic computer-driven weaning system to standardize the weaning process. In the BNP-driven group, diuretic use was higher, resulting in a more negative fluid balance and significantly shorter time to successful extubation (58.6 hours vs. 42.2 hours, P=0.03). The effect on weaning time was strongest in patients with left ventricular systolic dysfunction, whereas those with COPD seemed less likely to benefit. The two groups did not differ in baseline characteristics, length of stay, mortality, or development of adverse outcomes of renal failure, shock, or electrolyte disturbances.

Bottom line: Compared with physician-guided fluid management, a BNP-driven fluid management protocol decreased duration of ventilator weaning without significant differences in adverse events, mortality rate, or length of stay between the two groups.

Citation: Dessap AM, Roche-Campo F, Kouatchet A, et al. Natriuretic peptide-driven fluid management during ventilator weaning. Am J Respir Crit Care Med. 2012;186(12):1256-1263.

Visit our website for more physician reviews of recent HM-relevant literature.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. BNP-driven fluid management to improve ventilator weaning
2. Examining 30-day readmission patterns to reduce repeat hospitalizations
3. Impact of hospitalists’ workload on patient safety, care
4. Permanent atrial fibrillation is best controlled by diltiazem
5. Low-dose thrombolysis effective for pulmonary embolism
6. High mortality rate seen in surgical patients requiring CPR
7. ED visits common for acute-care patients post-discharge
8. Restrictive transfusion strategies effective for upper GI bleeding
9. Need for non-ICU acid suppression may be predictable
10. Recommended changes for adult immunizations

BNP-Driven Fluid Management Improves Ventilator Weaning

Clinical question: Does fluid management guided by daily plasma natriuretic peptide-driven (BNP) levels in mechanically ventilated patients improve weaning outcomes compared with usual therapy dictated by clinical acumen?

Background: Ventilator weaning contributes at least 40% of the total duration of mechanical ventilation; strategies aimed at optimizing this process could provide substantial benefit. Previous studies have demonstrated that BNP levels prior to ventilator weaning independently predict weaning failure. No current objective practical guide to fluid management during ventilator weaning exists.

Study design: Randomized controlled trial.

Setting: Multiple international centers.

Synopsis: In a multicenter randomized controlled trial, 304 patients who met specific inclusion and exclusion criteria were randomized to either a BNP-driven or physician-guided strategy for fluid management during ventilator weaning. Patients with renal failure were excluded because of the influence of renal function on BNP levels.

All patients in both groups were ventilated with an automatic computer-driven weaning system to standardize the weaning process. In the BNP-driven group, diuretic use was higher, resulting in a more negative fluid balance and significantly shorter time to successful extubation (58.6 hours vs. 42.2 hours, P=0.03). The effect on weaning time was strongest in patients with left ventricular systolic dysfunction, whereas those with COPD seemed less likely to benefit. The two groups did not differ in baseline characteristics, length of stay, mortality, or development of adverse outcomes of renal failure, shock, or electrolyte disturbances.

Bottom line: Compared with physician-guided fluid management, a BNP-driven fluid management protocol decreased duration of ventilator weaning without significant differences in adverse events, mortality rate, or length of stay between the two groups.

Citation: Dessap AM, Roche-Campo F, Kouatchet A, et al. Natriuretic peptide-driven fluid management during ventilator weaning. Am J Respir Crit Care Med. 2012;186(12):1256-1263.

30-Day Readmission Patterns for MI, Heart Failure, Pneumonia

Clinical question: Do patterns exist among patients readmitted within 30 days of discharge for acute myocardial infarction (AMI), heart failure, or pneumonia that could provide insight for improving strategies aimed at reducing readmission rates?

Background: Examining readmission timing, diagnoses, and patient demographics might provide information to better guide post-discharge programs aimed at reducing overall readmissions.

Study design: Retrospective review of Centers for Medicare & Medicaid Services (CMS) data.

Setting: Acute-care hospitals.

Synopsis: Using CMS hospitalization data for principal diagnoses of AMI, heart failure, or pneumonia from 2007 through 2009, the authors examined the percentage of 30-day readmissions occurring on each day after discharge; the most common readmission diagnoses; the median time to readmission for common readmission diagnoses; and the relationship between patient demographic characteristics, readmission diagnoses, and timing. They found total readmission rates of 24.8% for heart failure, 19.9% for AMI, and 18.3% for pneumonia. Approximately two-thirds of 30-day readmissions occurred within the first 15 days after discharge for each cohort. Neither readmission diagnoses nor timing varied by patient age, sex, or race.

Although the majority of readmissions do occur soon after discharge, it is important to note that about one-third of all readmissions occur 16 to 30 days after discharge. There also was a diverse spectrum of readmission diagnoses that were not associated with patient demographic characteristics. These findings suggest that current post-discharge strategies aimed at specific diseases or time periods might only address a fraction of the patients at risk for readmission.

Bottom line: Among Medicare patients hospitalized for heart failure, AMI, or pneumonia, 30-day readmissions were frequent throughout the entire period, and readmission diagnoses or timing did not vary by patient age, sex, or race.

Citation: Dharmarajan K, Hsich AF, Lin Z, et al. Diagnosis and timing of 30-day readmissions after hospitalization for heart failure, acute myocardial infarction, or pneumonia. JAMA. 2013;309(4):355-363.

Workload Might Impact Patient Safety and Quality of Care

Clinical question: Do hospitalists’ workloads affect patient quality of care and safety?

Background: Preventable medical errors contribute to a large number of patient deaths each year. It is unclear if a hospitalist’s clinical workload affects rates of medical errors or patient harm.

Study design: Cross-sectional cohort study.

Setting: Hospitalists enrolled in online physician community QuantiaMD.com.

Synopsis: There has been limited research evaluating the correlation between physician workload and patient safety. An online survey compared the responses of 506 out of 890 enrolled physicians on the impact of average patient census and several outcome measures of quality of care. Some 40% reported that their patient census exceeded their personal safe workload at least once a month. They also reported that less time for patient evaluations led to fewer discussions with patients and family members, more unnecessary medical work-ups, and lower patient satisfaction.

A limitation of this study is that this electronic survey had the potential for selection bias. It also only measured perceptions of safety and quality, and only used standard daytime shifts (excluding night, cross-cover, weekend, and holiday shifts), which might have been associated with significantly different conclusions.

Bottom line: Increase in workload has a negative perceived impact on patient safety and quality of care for attending hospitalists.

Citation: Michtalik HJ, Yeh HC, Pronovost P, et. al. Impact of attending physician workload on patient care: a survey of hospitalists. JAMA Intern Med. 2013;173(5):375-377.

Permanent Atrial Fibrillation Best Controlled by Diltiazem

Clinical question: Is there a difference between beta-blockers and calcium channel blockers for ventricular rate control and arrhythmia-related symptoms in patients with permanent atrial fibrillation?

Background: Rate control with beta-blockers or calcium channel blockers is recommended for the initial therapy of atrial fibrillation. However, studies comparing those drug classes or drugs within them are lacking.

Study design: Prospective, randomized, investigator-blind crossover study.

Setting: Majority of patients from an atrial fibrillation outpatient clinic at Baerum Hospital in Norway.

Synopsis: The RATe Control in Atrial Fibrillation (RATAF) study included 60 participants with permanent atrial fibrillation. The goal of the study was to compare the efficacy of diltiazem at 360 mg/day, verapamil at 240 mg/day, metoprolol at 100 mg/day, and carvedilol at 25mg/day on ventricular heart rate and related symptoms in atrial fibrillation. Patients had a mean age of 71, atrial fibrillation for more than three months, and mean heart rate of 96 beats/minute. Exclusion criteria included the presence of congestive heart failure or ischemic heart disease with the need for other medications that could compromise the study.

From this study, diltiazem was shown to have the greatest effect in lowering heart rate, and those patients taking this medication had decreased symptoms related to atrial fibrillation. Hospitalists should not rely solely on this study for their treatment choice in all atrial fibrillation patients, but in certain populations, they should consider diltiazem as their first-line drug.

Bottom line: Diltiazem was shown to have the greatest reduction in heart rate and symptoms related to permanent atrial fibrillation.

Citation: Ulimoen SR, Enger S, Carlson J, et al. Comparison of four single-drug regimens on ventricular rate and arrhythmia-related symptoms in patients with permanent atrial fibrillation. Am J Cardiol. 2013:111(2):225-230.

Low-Dose Thrombolysis Effective in Moderate Pulmonary Embolism

Clinical question: Can low-dose tissue plasminogen activator (tPA) help reduce pulmonary artery pressure in those with moderate pulmonary embolism (PE)?

Background: Studies have shown full-dose thrombolysis can effectively decrease pulmonary artery pressure in patients with massive PE. However, there are limited data regarding low-dose or “safe dose” thrombolytic therapy and its effect on pulmonary artery pressure.

Study design: Prospective, controlled, randomized study.

Setting: Single center.

Synopsis: The Moderate Pulmonary Embolism Treated with Thrombolysis (MOPETT) study enrolled patients with moderate PE, defined as signs and symptoms of PE plus computed tomographic pulmonary angiographic involvement of > 70% involvement of thrombus in ≥2 lobar or left/right main pulmonary arteries or high probability ventilation/perfusion scan (mismatch in ≥2 lobes). Patients in the thrombolysis group (n=61) were given low-dose tPA (100 mg tPA) and anticoagulation vs. the control group (n=60), which received only anticoagulation.

The study ran for 22 months, and the primary end points were pulmonary hypertension and recurrent PE. After analysis, low-dose thrombolysis was shown to significantly decrease pulmonary artery pressure and occurrence of recurrent PE compared to the control group.

This study demonstrates that, while the decision to use thrombolytics should always be made cautiously, hospitalists can consider low-dose thrombolysis in patients with moderate PE.

Bottom line: Low-dose thrombolysis, in addition to anticoagulation, in patients with moderate PE decreases pulmonary hypertension and recurrent PE.

Citation: Sharifi M, Bay C, Skrocki L, Rahimi F, Mehdipour M. Moderate pulmonary embolism treated with thrombolysis (from the “MOPETT” trial). Am J Cardiol. 2013;111(2):273-277.

High Mortality in Surgical Patients Requiring CPR

Clinical question: What are the incidence, characteristics, and 30-day-outcomes of CPR in surgical patients?

Background: Most studies of CPR are based on the medical population, and little is known about the utilization, risk factors, and outcomes of CPR in surgical patients.

Study design: Retrospective cohort study.

Setting: Two hundred fifty U.S. hospitals in the American College of Surgeons’ National Surgical Quality Improvement Program.

Synopsis: A total of 1.3 million surgical cases were studied in the data set. The overall incidence was 1 event per 203 cases. Most patients (77.6%) experienced a complication and did so on or before the day of CPR in three-fourths of cases. The incidence of CPR was the highest for cardiac surgery patients. Patients who received CPR had a mortality rate of 71.6%. Mortality rates of CPR patients increased with more comorbidities.

Additionally, older age and an American Society of Anesthesiologists (ASA) class of 5 was associated with higher mortality.

Limitations of this study included coding flaws in data collection, lack of capture of resuscitation-related injuries, and failure to account for changes in DNR orders.

Hospitalists should be mindful of risk factors contributing to CPR in surgical patients when performing perioperative evaluations.

Bottom line: Surgical patients who experience CPR have a high mortality rate, but many of these patients have pre-arrest complications that can be preventable.

Citation: Kazaure HS, Roman SA, Rosenthal RA, Sosa, JA. Cardiac arrest among surgical patients. JAMA Surg. 2013;148(1):14-21.

Emergency Department Visits are Frequent Post-Discharge

Clinical question: What role do ED visits contribute to the overall use of acute-care services within 30 days of hospital discharge?

Background: Hospital readmissions within 30 days of discharge are a marker of the quality of care and reflect the effectiveness of the discharge process. ED visits are also a marker of hospital-based acute care following discharge, but little is known about the role of the ED during the post-discharge period.

Study design: Prospective study.

Setting: Acute-care hospitals in California, Florida, and Nebraska.

Synopsis: Using the Healthcare Cost and Utilization Project state inpatient and ED databases, all discharges between July 1, 2008, and Sept. 31, 2009, were evaluated for residents aged 18 years or older from three hospitals in three states. After exclusions, 5 million index hospitalizations among 4 million unique patients were studied.

Approximately 40% of the more than 1 million post-discharge acute-care encounters involved a visit to the ED.

Limitations of this study include that the data was derived from only three states, and only hospital-based acute-care visits were measured (i.e. visits to physician offices were not included). As hospitalists, we are responsible for discharges and care transitions. Being sensitive to the common medical conditions resulting in post-discharge ED encounters might improve care transitions.

Bottom line: Hospital readmission rates underestimate ED use following discharge.

Citation: Vashi AA, Fox JP, Carr BG, et al. Use of hospital-based acute care among patients recently discharged from the hospital. JAMA. 2013;309(4):364-371.

Restrictive Transfusion Strategy Beneficial in Upper GI Bleeding

Clinical question: What is the hemoglobin threshold for transfusion of red cells in patients with acute upper GI bleeding?

Background: Controlled trials have shown that restrictive transfusion strategies (Hgb<7) are as effective as liberal transfusion strategies (Hgb<9) in critically ill patients. These studies have excluded patients with GI bleeding. In cases for which GI is not severe, the safest transfusion strategy is controversial.

Study design: Single-center study at Hospital de la Santa Creu i Sant Pau, Barcelona, Spain.

Synopsis: A total of 921 adult patients with acute upper GI bleeding were enrolled and assigned: 461 to a restrictive transfusion strategy (hemoglobin<7) and 460 to a liberal strategy (hemoglobin<9). Patients with massive exsanguinating bleeding, acute coronary syndrome, peripheral vasculopathy, stroke, transient ischemic attack, lower GI bleed, recent trauma or surgery, or low risk of rebleeding were excluded. The primary outcome measure was the rate of death from any cause within the first 45 days.

Secondary outcomes included the rate of further bleeding and the rate of in-hospital complications.

Statistically significant benefit in following a restrictive versus liberal strategy was demonstrated in all major outcomes: mortality (5% vs, 9%, P=0.02), rate of further bleeding (10% vs 18%, P=0.01), and rate of complications (40% vs. 48%, P=0.02).

The study is limited by its inability to be generalized to all patients with acute GI bleeding, as patients with massive exsanguinating bleeds and those with low risk of rebleeding were excluded.

Bottom line: Restrictive transfusion (Hgb<7) significantly improved outcomes for patients with acute upper GI bleeding.

Citation: Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013;368(1):11-21.

Need for Non-ICU Acid Suppression Might Be Predictable

Clinical question: What are the risk factors for nosocomial bleeding in non-critically-ill patients?

Background: Acid-suppressive medication has been shown to reduce the incidence of nosocomial GI bleed in the ICU, but current guidelines recommend against its use in non-critically-ill patients. However, a subgroup of these patients might possess a high enough risk for GI bleed that prophylaxis is warranted.

Study design: Cohort study.

Setting: Academic medical center in Boston.

Synopsis: A total of 75,723 admissions of adult patients hospitalized for three or more days were included. Exclusion criteria included primary discharge diagnosis of GI bleed; principal procedure code of cardiac catheterization; and bleeding episodes occurring while in the ICU or within 48 hours of transfer out of the ICU. The primary outcome was nosocomial GI bleed (>24 hours after admission) occurring outside the ICU.

Nosocomial GI bleeding occurred in 203 patients (0.27%). Independent risk factors for bleeding included age >60, male sex, liver disease, acute renal failure, sepsis, being on a medicine service, prophylactic anticoagulation, and coagulopathy. Based on the data, a scoring system was created that identified a high-risk group in whom the number needed to treat with acid-suppressive medication to prevent one bleed was 48.

The major limitations of this study are its observational nature and the need for validation of the proposed scoring system.

Bottom line: Risk for nosocomial GI bleeding appears predictable and supports the selective use of prophylactic acid suppression in non-critically-ill patients.

Citation: Herzig SJ, Rothberg MB, Feinbloom DB, et al. Risk factors for nosocomial gastrointestinal bleeding and use of acid-suppressive medication in non-critically ill patients. J Gen Intern Med. 2013; January 5. [Epub ahead of print].

Additions, Modifications, and Clarifications Regarding Adult Immunization

Clinical question: What are the changes to the recommended Adult Immunization Schedule for 2013?

Background: Despite the known public health benefits of immunization, adult vaccination rates remain low. The positive impact of strong provider recommendations regarding vaccines underscores the importance of provider awareness of vaccine schedules, precautions, and contraindications.

Study design: Annual Advisory Committee on Immunization Practices (ACIP) review.

Setting: Data from 2011 National Health Interview Survey.

Synopsis: Highlighted changes include: 1) a single dose of pneumococcal 13-valent conjugate (PCV13) vaccine is now recommended for all individuals over the age of 19 with qualifying conditions; 2) clarification regarding pneumococcal polysaccharide (PPSV23) vaccine illustrates that high-risk individuals will receive up to three doses (one or two doses prior to age 65, plus an additional dose after the age of 65); 3) one dose of tetanus, diphtheria, and acellular pertussis (Tdap) vaccine is now recommended for all adults, including individuals age >65; 4) pregnant woman are advised to receive Tdap between 27 and 36 weeks’ gestation, with each pregnancy to provide protection to their newborn in the first months of life; 5) quadrivalent formulations of the live attenuated influenza vaccine (LAIV) and most likely the inactivated influenza vaccine (IIV) will be available in the 2013-2014 influenza season to increase cross-reactive protection against influenza B; and 6) both injection and noninjection illicit drug users are recommended to receive hepatitis A vaccine.

Bottom line: Expanded recommendations for adult immunization provide more opportunities for the practicing hospitalist to improve vaccine capture.

Citation: Advisory Committee on Immunization Practices. Recommended adult immunization schedule: United States, 2013. Ann Int Med. 2013;158(3):191-199.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. BNP-driven fluid management to improve ventilator weaning
2. Examining 30-day readmission patterns to reduce repeat hospitalizations
3. Impact of hospitalists’ workload on patient safety, care
4. Permanent atrial fibrillation is best controlled by diltiazem
5. Low-dose thrombolysis effective for pulmonary embolism
6. High mortality rate seen in surgical patients requiring CPR
7. ED visits common for acute-care patients post-discharge
8. Restrictive transfusion strategies effective for upper GI bleeding
9. Need for non-ICU acid suppression may be predictable
10. Recommended changes for adult immunizations

BNP-Driven Fluid Management Improves Ventilator Weaning

Clinical question: Does fluid management guided by daily plasma natriuretic peptide-driven (BNP) levels in mechanically ventilated patients improve weaning outcomes compared with usual therapy dictated by clinical acumen?

Background: Ventilator weaning contributes at least 40% of the total duration of mechanical ventilation; strategies aimed at optimizing this process could provide substantial benefit. Previous studies have demonstrated that BNP levels prior to ventilator weaning independently predict weaning failure. No current objective practical guide to fluid management during ventilator weaning exists.

Study design: Randomized controlled trial.

Setting: Multiple international centers.

Synopsis: In a multicenter randomized controlled trial, 304 patients who met specific inclusion and exclusion criteria were randomized to either a BNP-driven or physician-guided strategy for fluid management during ventilator weaning. Patients with renal failure were excluded because of the influence of renal function on BNP levels.

All patients in both groups were ventilated with an automatic computer-driven weaning system to standardize the weaning process. In the BNP-driven group, diuretic use was higher, resulting in a more negative fluid balance and significantly shorter time to successful extubation (58.6 hours vs. 42.2 hours, P=0.03). The effect on weaning time was strongest in patients with left ventricular systolic dysfunction, whereas those with COPD seemed less likely to benefit. The two groups did not differ in baseline characteristics, length of stay, mortality, or development of adverse outcomes of renal failure, shock, or electrolyte disturbances.

Bottom line: Compared with physician-guided fluid management, a BNP-driven fluid management protocol decreased duration of ventilator weaning without significant differences in adverse events, mortality rate, or length of stay between the two groups.

Citation: Dessap AM, Roche-Campo F, Kouatchet A, et al. Natriuretic peptide-driven fluid management during ventilator weaning. Am J Respir Crit Care Med. 2012;186(12):1256-1263.

30-Day Readmission Patterns for MI, Heart Failure, Pneumonia

Clinical question: Do patterns exist among patients readmitted within 30 days of discharge for acute myocardial infarction (AMI), heart failure, or pneumonia that could provide insight for improving strategies aimed at reducing readmission rates?

Background: Examining readmission timing, diagnoses, and patient demographics might provide information to better guide post-discharge programs aimed at reducing overall readmissions.

Study design: Retrospective review of Centers for Medicare & Medicaid Services (CMS) data.

Setting: Acute-care hospitals.

Synopsis: Using CMS hospitalization data for principal diagnoses of AMI, heart failure, or pneumonia from 2007 through 2009, the authors examined the percentage of 30-day readmissions occurring on each day after discharge; the most common readmission diagnoses; the median time to readmission for common readmission diagnoses; and the relationship between patient demographic characteristics, readmission diagnoses, and timing. They found total readmission rates of 24.8% for heart failure, 19.9% for AMI, and 18.3% for pneumonia. Approximately two-thirds of 30-day readmissions occurred within the first 15 days after discharge for each cohort. Neither readmission diagnoses nor timing varied by patient age, sex, or race.

Although the majority of readmissions do occur soon after discharge, it is important to note that about one-third of all readmissions occur 16 to 30 days after discharge. There also was a diverse spectrum of readmission diagnoses that were not associated with patient demographic characteristics. These findings suggest that current post-discharge strategies aimed at specific diseases or time periods might only address a fraction of the patients at risk for readmission.

Bottom line: Among Medicare patients hospitalized for heart failure, AMI, or pneumonia, 30-day readmissions were frequent throughout the entire period, and readmission diagnoses or timing did not vary by patient age, sex, or race.

Citation: Dharmarajan K, Hsich AF, Lin Z, et al. Diagnosis and timing of 30-day readmissions after hospitalization for heart failure, acute myocardial infarction, or pneumonia. JAMA. 2013;309(4):355-363.

Workload Might Impact Patient Safety and Quality of Care

Clinical question: Do hospitalists’ workloads affect patient quality of care and safety?

Background: Preventable medical errors contribute to a large number of patient deaths each year. It is unclear if a hospitalist’s clinical workload affects rates of medical errors or patient harm.

Study design: Cross-sectional cohort study.

Setting: Hospitalists enrolled in online physician community QuantiaMD.com.

Synopsis: There has been limited research evaluating the correlation between physician workload and patient safety. An online survey compared the responses of 506 out of 890 enrolled physicians on the impact of average patient census and several outcome measures of quality of care. Some 40% reported that their patient census exceeded their personal safe workload at least once a month. They also reported that less time for patient evaluations led to fewer discussions with patients and family members, more unnecessary medical work-ups, and lower patient satisfaction.

A limitation of this study is that this electronic survey had the potential for selection bias. It also only measured perceptions of safety and quality, and only used standard daytime shifts (excluding night, cross-cover, weekend, and holiday shifts), which might have been associated with significantly different conclusions.

Bottom line: Increase in workload has a negative perceived impact on patient safety and quality of care for attending hospitalists.

Citation: Michtalik HJ, Yeh HC, Pronovost P, et. al. Impact of attending physician workload on patient care: a survey of hospitalists. JAMA Intern Med. 2013;173(5):375-377.

Permanent Atrial Fibrillation Best Controlled by Diltiazem

Clinical question: Is there a difference between beta-blockers and calcium channel blockers for ventricular rate control and arrhythmia-related symptoms in patients with permanent atrial fibrillation?

Background: Rate control with beta-blockers or calcium channel blockers is recommended for the initial therapy of atrial fibrillation. However, studies comparing those drug classes or drugs within them are lacking.

Study design: Prospective, randomized, investigator-blind crossover study.

Setting: Majority of patients from an atrial fibrillation outpatient clinic at Baerum Hospital in Norway.

Synopsis: The RATe Control in Atrial Fibrillation (RATAF) study included 60 participants with permanent atrial fibrillation. The goal of the study was to compare the efficacy of diltiazem at 360 mg/day, verapamil at 240 mg/day, metoprolol at 100 mg/day, and carvedilol at 25mg/day on ventricular heart rate and related symptoms in atrial fibrillation. Patients had a mean age of 71, atrial fibrillation for more than three months, and mean heart rate of 96 beats/minute. Exclusion criteria included the presence of congestive heart failure or ischemic heart disease with the need for other medications that could compromise the study.

From this study, diltiazem was shown to have the greatest effect in lowering heart rate, and those patients taking this medication had decreased symptoms related to atrial fibrillation. Hospitalists should not rely solely on this study for their treatment choice in all atrial fibrillation patients, but in certain populations, they should consider diltiazem as their first-line drug.

Bottom line: Diltiazem was shown to have the greatest reduction in heart rate and symptoms related to permanent atrial fibrillation.

Citation: Ulimoen SR, Enger S, Carlson J, et al. Comparison of four single-drug regimens on ventricular rate and arrhythmia-related symptoms in patients with permanent atrial fibrillation. Am J Cardiol. 2013:111(2):225-230.

Low-Dose Thrombolysis Effective in Moderate Pulmonary Embolism

Clinical question: Can low-dose tissue plasminogen activator (tPA) help reduce pulmonary artery pressure in those with moderate pulmonary embolism (PE)?

Background: Studies have shown full-dose thrombolysis can effectively decrease pulmonary artery pressure in patients with massive PE. However, there are limited data regarding low-dose or “safe dose” thrombolytic therapy and its effect on pulmonary artery pressure.

Study design: Prospective, controlled, randomized study.

Setting: Single center.

Synopsis: The Moderate Pulmonary Embolism Treated with Thrombolysis (MOPETT) study enrolled patients with moderate PE, defined as signs and symptoms of PE plus computed tomographic pulmonary angiographic involvement of > 70% involvement of thrombus in ≥2 lobar or left/right main pulmonary arteries or high probability ventilation/perfusion scan (mismatch in ≥2 lobes). Patients in the thrombolysis group (n=61) were given low-dose tPA (100 mg tPA) and anticoagulation vs. the control group (n=60), which received only anticoagulation.

The study ran for 22 months, and the primary end points were pulmonary hypertension and recurrent PE. After analysis, low-dose thrombolysis was shown to significantly decrease pulmonary artery pressure and occurrence of recurrent PE compared to the control group.

This study demonstrates that, while the decision to use thrombolytics should always be made cautiously, hospitalists can consider low-dose thrombolysis in patients with moderate PE.

Bottom line: Low-dose thrombolysis, in addition to anticoagulation, in patients with moderate PE decreases pulmonary hypertension and recurrent PE.

Citation: Sharifi M, Bay C, Skrocki L, Rahimi F, Mehdipour M. Moderate pulmonary embolism treated with thrombolysis (from the “MOPETT” trial). Am J Cardiol. 2013;111(2):273-277.

High Mortality in Surgical Patients Requiring CPR

Clinical question: What are the incidence, characteristics, and 30-day-outcomes of CPR in surgical patients?

Background: Most studies of CPR are based on the medical population, and little is known about the utilization, risk factors, and outcomes of CPR in surgical patients.

Study design: Retrospective cohort study.

Setting: Two hundred fifty U.S. hospitals in the American College of Surgeons’ National Surgical Quality Improvement Program.

Synopsis: A total of 1.3 million surgical cases were studied in the data set. The overall incidence was 1 event per 203 cases. Most patients (77.6%) experienced a complication and did so on or before the day of CPR in three-fourths of cases. The incidence of CPR was the highest for cardiac surgery patients. Patients who received CPR had a mortality rate of 71.6%. Mortality rates of CPR patients increased with more comorbidities.

Additionally, older age and an American Society of Anesthesiologists (ASA) class of 5 was associated with higher mortality.

Limitations of this study included coding flaws in data collection, lack of capture of resuscitation-related injuries, and failure to account for changes in DNR orders.

Hospitalists should be mindful of risk factors contributing to CPR in surgical patients when performing perioperative evaluations.

Bottom line: Surgical patients who experience CPR have a high mortality rate, but many of these patients have pre-arrest complications that can be preventable.

Citation: Kazaure HS, Roman SA, Rosenthal RA, Sosa, JA. Cardiac arrest among surgical patients. JAMA Surg. 2013;148(1):14-21.

Emergency Department Visits are Frequent Post-Discharge

Clinical question: What role do ED visits contribute to the overall use of acute-care services within 30 days of hospital discharge?

Background: Hospital readmissions within 30 days of discharge are a marker of the quality of care and reflect the effectiveness of the discharge process. ED visits are also a marker of hospital-based acute care following discharge, but little is known about the role of the ED during the post-discharge period.

Study design: Prospective study.

Setting: Acute-care hospitals in California, Florida, and Nebraska.

Synopsis: Using the Healthcare Cost and Utilization Project state inpatient and ED databases, all discharges between July 1, 2008, and Sept. 31, 2009, were evaluated for residents aged 18 years or older from three hospitals in three states. After exclusions, 5 million index hospitalizations among 4 million unique patients were studied.

Approximately 40% of the more than 1 million post-discharge acute-care encounters involved a visit to the ED.

Limitations of this study include that the data was derived from only three states, and only hospital-based acute-care visits were measured (i.e. visits to physician offices were not included). As hospitalists, we are responsible for discharges and care transitions. Being sensitive to the common medical conditions resulting in post-discharge ED encounters might improve care transitions.

Bottom line: Hospital readmission rates underestimate ED use following discharge.

Citation: Vashi AA, Fox JP, Carr BG, et al. Use of hospital-based acute care among patients recently discharged from the hospital. JAMA. 2013;309(4):364-371.

Restrictive Transfusion Strategy Beneficial in Upper GI Bleeding

Clinical question: What is the hemoglobin threshold for transfusion of red cells in patients with acute upper GI bleeding?

Background: Controlled trials have shown that restrictive transfusion strategies (Hgb<7) are as effective as liberal transfusion strategies (Hgb<9) in critically ill patients. These studies have excluded patients with GI bleeding. In cases for which GI is not severe, the safest transfusion strategy is controversial.

Study design: Single-center study at Hospital de la Santa Creu i Sant Pau, Barcelona, Spain.

Synopsis: A total of 921 adult patients with acute upper GI bleeding were enrolled and assigned: 461 to a restrictive transfusion strategy (hemoglobin<7) and 460 to a liberal strategy (hemoglobin<9). Patients with massive exsanguinating bleeding, acute coronary syndrome, peripheral vasculopathy, stroke, transient ischemic attack, lower GI bleed, recent trauma or surgery, or low risk of rebleeding were excluded. The primary outcome measure was the rate of death from any cause within the first 45 days.

Secondary outcomes included the rate of further bleeding and the rate of in-hospital complications.

Statistically significant benefit in following a restrictive versus liberal strategy was demonstrated in all major outcomes: mortality (5% vs, 9%, P=0.02), rate of further bleeding (10% vs 18%, P=0.01), and rate of complications (40% vs. 48%, P=0.02).

The study is limited by its inability to be generalized to all patients with acute GI bleeding, as patients with massive exsanguinating bleeds and those with low risk of rebleeding were excluded.

Bottom line: Restrictive transfusion (Hgb<7) significantly improved outcomes for patients with acute upper GI bleeding.

Citation: Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013;368(1):11-21.

Need for Non-ICU Acid Suppression Might Be Predictable

Clinical question: What are the risk factors for nosocomial bleeding in non-critically-ill patients?

Background: Acid-suppressive medication has been shown to reduce the incidence of nosocomial GI bleed in the ICU, but current guidelines recommend against its use in non-critically-ill patients. However, a subgroup of these patients might possess a high enough risk for GI bleed that prophylaxis is warranted.

Study design: Cohort study.

Setting: Academic medical center in Boston.

Synopsis: A total of 75,723 admissions of adult patients hospitalized for three or more days were included. Exclusion criteria included primary discharge diagnosis of GI bleed; principal procedure code of cardiac catheterization; and bleeding episodes occurring while in the ICU or within 48 hours of transfer out of the ICU. The primary outcome was nosocomial GI bleed (>24 hours after admission) occurring outside the ICU.

Nosocomial GI bleeding occurred in 203 patients (0.27%). Independent risk factors for bleeding included age >60, male sex, liver disease, acute renal failure, sepsis, being on a medicine service, prophylactic anticoagulation, and coagulopathy. Based on the data, a scoring system was created that identified a high-risk group in whom the number needed to treat with acid-suppressive medication to prevent one bleed was 48.

The major limitations of this study are its observational nature and the need for validation of the proposed scoring system.

Bottom line: Risk for nosocomial GI bleeding appears predictable and supports the selective use of prophylactic acid suppression in non-critically-ill patients.

Citation: Herzig SJ, Rothberg MB, Feinbloom DB, et al. Risk factors for nosocomial gastrointestinal bleeding and use of acid-suppressive medication in non-critically ill patients. J Gen Intern Med. 2013; January 5. [Epub ahead of print].

Additions, Modifications, and Clarifications Regarding Adult Immunization

Clinical question: What are the changes to the recommended Adult Immunization Schedule for 2013?

Background: Despite the known public health benefits of immunization, adult vaccination rates remain low. The positive impact of strong provider recommendations regarding vaccines underscores the importance of provider awareness of vaccine schedules, precautions, and contraindications.

Study design: Annual Advisory Committee on Immunization Practices (ACIP) review.

Setting: Data from 2011 National Health Interview Survey.

Synopsis: Highlighted changes include: 1) a single dose of pneumococcal 13-valent conjugate (PCV13) vaccine is now recommended for all individuals over the age of 19 with qualifying conditions; 2) clarification regarding pneumococcal polysaccharide (PPSV23) vaccine illustrates that high-risk individuals will receive up to three doses (one or two doses prior to age 65, plus an additional dose after the age of 65); 3) one dose of tetanus, diphtheria, and acellular pertussis (Tdap) vaccine is now recommended for all adults, including individuals age >65; 4) pregnant woman are advised to receive Tdap between 27 and 36 weeks’ gestation, with each pregnancy to provide protection to their newborn in the first months of life; 5) quadrivalent formulations of the live attenuated influenza vaccine (LAIV) and most likely the inactivated influenza vaccine (IIV) will be available in the 2013-2014 influenza season to increase cross-reactive protection against influenza B; and 6) both injection and noninjection illicit drug users are recommended to receive hepatitis A vaccine.

Bottom line: Expanded recommendations for adult immunization provide more opportunities for the practicing hospitalist to improve vaccine capture.

Citation: Advisory Committee on Immunization Practices. Recommended adult immunization schedule: United States, 2013. Ann Int Med. 2013;158(3):191-199.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. BNP-driven fluid management to improve ventilator weaning
2. Examining 30-day readmission patterns to reduce repeat hospitalizations
3. Impact of hospitalists’ workload on patient safety, care
4. Permanent atrial fibrillation is best controlled by diltiazem
5. Low-dose thrombolysis effective for pulmonary embolism
6. High mortality rate seen in surgical patients requiring CPR
7. ED visits common for acute-care patients post-discharge
8. Restrictive transfusion strategies effective for upper GI bleeding
9. Need for non-ICU acid suppression may be predictable
10. Recommended changes for adult immunizations

BNP-Driven Fluid Management Improves Ventilator Weaning

Clinical question: Does fluid management guided by daily plasma natriuretic peptide-driven (BNP) levels in mechanically ventilated patients improve weaning outcomes compared with usual therapy dictated by clinical acumen?

Background: Ventilator weaning contributes at least 40% of the total duration of mechanical ventilation; strategies aimed at optimizing this process could provide substantial benefit. Previous studies have demonstrated that BNP levels prior to ventilator weaning independently predict weaning failure. No current objective practical guide to fluid management during ventilator weaning exists.

Study design: Randomized controlled trial.

Setting: Multiple international centers.

Synopsis: In a multicenter randomized controlled trial, 304 patients who met specific inclusion and exclusion criteria were randomized to either a BNP-driven or physician-guided strategy for fluid management during ventilator weaning. Patients with renal failure were excluded because of the influence of renal function on BNP levels.

All patients in both groups were ventilated with an automatic computer-driven weaning system to standardize the weaning process. In the BNP-driven group, diuretic use was higher, resulting in a more negative fluid balance and significantly shorter time to successful extubation (58.6 hours vs. 42.2 hours, P=0.03). The effect on weaning time was strongest in patients with left ventricular systolic dysfunction, whereas those with COPD seemed less likely to benefit. The two groups did not differ in baseline characteristics, length of stay, mortality, or development of adverse outcomes of renal failure, shock, or electrolyte disturbances.

Bottom line: Compared with physician-guided fluid management, a BNP-driven fluid management protocol decreased duration of ventilator weaning without significant differences in adverse events, mortality rate, or length of stay between the two groups.

Citation: Dessap AM, Roche-Campo F, Kouatchet A, et al. Natriuretic peptide-driven fluid management during ventilator weaning. Am J Respir Crit Care Med. 2012;186(12):1256-1263.

30-Day Readmission Patterns for MI, Heart Failure, Pneumonia

Clinical question: Do patterns exist among patients readmitted within 30 days of discharge for acute myocardial infarction (AMI), heart failure, or pneumonia that could provide insight for improving strategies aimed at reducing readmission rates?

Background: Examining readmission timing, diagnoses, and patient demographics might provide information to better guide post-discharge programs aimed at reducing overall readmissions.

Study design: Retrospective review of Centers for Medicare & Medicaid Services (CMS) data.

Setting: Acute-care hospitals.

Synopsis: Using CMS hospitalization data for principal diagnoses of AMI, heart failure, or pneumonia from 2007 through 2009, the authors examined the percentage of 30-day readmissions occurring on each day after discharge; the most common readmission diagnoses; the median time to readmission for common readmission diagnoses; and the relationship between patient demographic characteristics, readmission diagnoses, and timing. They found total readmission rates of 24.8% for heart failure, 19.9% for AMI, and 18.3% for pneumonia. Approximately two-thirds of 30-day readmissions occurred within the first 15 days after discharge for each cohort. Neither readmission diagnoses nor timing varied by patient age, sex, or race.

Although the majority of readmissions do occur soon after discharge, it is important to note that about one-third of all readmissions occur 16 to 30 days after discharge. There also was a diverse spectrum of readmission diagnoses that were not associated with patient demographic characteristics. These findings suggest that current post-discharge strategies aimed at specific diseases or time periods might only address a fraction of the patients at risk for readmission.

Bottom line: Among Medicare patients hospitalized for heart failure, AMI, or pneumonia, 30-day readmissions were frequent throughout the entire period, and readmission diagnoses or timing did not vary by patient age, sex, or race.

Citation: Dharmarajan K, Hsich AF, Lin Z, et al. Diagnosis and timing of 30-day readmissions after hospitalization for heart failure, acute myocardial infarction, or pneumonia. JAMA. 2013;309(4):355-363.

Workload Might Impact Patient Safety and Quality of Care

Clinical question: Do hospitalists’ workloads affect patient quality of care and safety?

Background: Preventable medical errors contribute to a large number of patient deaths each year. It is unclear if a hospitalist’s clinical workload affects rates of medical errors or patient harm.

Study design: Cross-sectional cohort study.

Setting: Hospitalists enrolled in online physician community QuantiaMD.com.

Synopsis: There has been limited research evaluating the correlation between physician workload and patient safety. An online survey compared the responses of 506 out of 890 enrolled physicians on the impact of average patient census and several outcome measures of quality of care. Some 40% reported that their patient census exceeded their personal safe workload at least once a month. They also reported that less time for patient evaluations led to fewer discussions with patients and family members, more unnecessary medical work-ups, and lower patient satisfaction.

A limitation of this study is that this electronic survey had the potential for selection bias. It also only measured perceptions of safety and quality, and only used standard daytime shifts (excluding night, cross-cover, weekend, and holiday shifts), which might have been associated with significantly different conclusions.

Bottom line: Increase in workload has a negative perceived impact on patient safety and quality of care for attending hospitalists.

Citation: Michtalik HJ, Yeh HC, Pronovost P, et. al. Impact of attending physician workload on patient care: a survey of hospitalists. JAMA Intern Med. 2013;173(5):375-377.

Permanent Atrial Fibrillation Best Controlled by Diltiazem

Clinical question: Is there a difference between beta-blockers and calcium channel blockers for ventricular rate control and arrhythmia-related symptoms in patients with permanent atrial fibrillation?

Background: Rate control with beta-blockers or calcium channel blockers is recommended for the initial therapy of atrial fibrillation. However, studies comparing those drug classes or drugs within them are lacking.

Study design: Prospective, randomized, investigator-blind crossover study.

Setting: Majority of patients from an atrial fibrillation outpatient clinic at Baerum Hospital in Norway.

Synopsis: The RATe Control in Atrial Fibrillation (RATAF) study included 60 participants with permanent atrial fibrillation. The goal of the study was to compare the efficacy of diltiazem at 360 mg/day, verapamil at 240 mg/day, metoprolol at 100 mg/day, and carvedilol at 25mg/day on ventricular heart rate and related symptoms in atrial fibrillation. Patients had a mean age of 71, atrial fibrillation for more than three months, and mean heart rate of 96 beats/minute. Exclusion criteria included the presence of congestive heart failure or ischemic heart disease with the need for other medications that could compromise the study.

From this study, diltiazem was shown to have the greatest effect in lowering heart rate, and those patients taking this medication had decreased symptoms related to atrial fibrillation. Hospitalists should not rely solely on this study for their treatment choice in all atrial fibrillation patients, but in certain populations, they should consider diltiazem as their first-line drug.

Bottom line: Diltiazem was shown to have the greatest reduction in heart rate and symptoms related to permanent atrial fibrillation.

Citation: Ulimoen SR, Enger S, Carlson J, et al. Comparison of four single-drug regimens on ventricular rate and arrhythmia-related symptoms in patients with permanent atrial fibrillation. Am J Cardiol. 2013:111(2):225-230.

Low-Dose Thrombolysis Effective in Moderate Pulmonary Embolism

Clinical question: Can low-dose tissue plasminogen activator (tPA) help reduce pulmonary artery pressure in those with moderate pulmonary embolism (PE)?

Background: Studies have shown full-dose thrombolysis can effectively decrease pulmonary artery pressure in patients with massive PE. However, there are limited data regarding low-dose or “safe dose” thrombolytic therapy and its effect on pulmonary artery pressure.

Study design: Prospective, controlled, randomized study.

Setting: Single center.

Synopsis: The Moderate Pulmonary Embolism Treated with Thrombolysis (MOPETT) study enrolled patients with moderate PE, defined as signs and symptoms of PE plus computed tomographic pulmonary angiographic involvement of > 70% involvement of thrombus in ≥2 lobar or left/right main pulmonary arteries or high probability ventilation/perfusion scan (mismatch in ≥2 lobes). Patients in the thrombolysis group (n=61) were given low-dose tPA (100 mg tPA) and anticoagulation vs. the control group (n=60), which received only anticoagulation.

The study ran for 22 months, and the primary end points were pulmonary hypertension and recurrent PE. After analysis, low-dose thrombolysis was shown to significantly decrease pulmonary artery pressure and occurrence of recurrent PE compared to the control group.

This study demonstrates that, while the decision to use thrombolytics should always be made cautiously, hospitalists can consider low-dose thrombolysis in patients with moderate PE.

Bottom line: Low-dose thrombolysis, in addition to anticoagulation, in patients with moderate PE decreases pulmonary hypertension and recurrent PE.

Citation: Sharifi M, Bay C, Skrocki L, Rahimi F, Mehdipour M. Moderate pulmonary embolism treated with thrombolysis (from the “MOPETT” trial). Am J Cardiol. 2013;111(2):273-277.

High Mortality in Surgical Patients Requiring CPR

Clinical question: What are the incidence, characteristics, and 30-day-outcomes of CPR in surgical patients?

Background: Most studies of CPR are based on the medical population, and little is known about the utilization, risk factors, and outcomes of CPR in surgical patients.

Study design: Retrospective cohort study.

Setting: Two hundred fifty U.S. hospitals in the American College of Surgeons’ National Surgical Quality Improvement Program.

Synopsis: A total of 1.3 million surgical cases were studied in the data set. The overall incidence was 1 event per 203 cases. Most patients (77.6%) experienced a complication and did so on or before the day of CPR in three-fourths of cases. The incidence of CPR was the highest for cardiac surgery patients. Patients who received CPR had a mortality rate of 71.6%. Mortality rates of CPR patients increased with more comorbidities.

Additionally, older age and an American Society of Anesthesiologists (ASA) class of 5 was associated with higher mortality.

Limitations of this study included coding flaws in data collection, lack of capture of resuscitation-related injuries, and failure to account for changes in DNR orders.

Hospitalists should be mindful of risk factors contributing to CPR in surgical patients when performing perioperative evaluations.

Bottom line: Surgical patients who experience CPR have a high mortality rate, but many of these patients have pre-arrest complications that can be preventable.

Citation: Kazaure HS, Roman SA, Rosenthal RA, Sosa, JA. Cardiac arrest among surgical patients. JAMA Surg. 2013;148(1):14-21.

Emergency Department Visits are Frequent Post-Discharge

Clinical question: What role do ED visits contribute to the overall use of acute-care services within 30 days of hospital discharge?

Background: Hospital readmissions within 30 days of discharge are a marker of the quality of care and reflect the effectiveness of the discharge process. ED visits are also a marker of hospital-based acute care following discharge, but little is known about the role of the ED during the post-discharge period.

Study design: Prospective study.

Setting: Acute-care hospitals in California, Florida, and Nebraska.

Synopsis: Using the Healthcare Cost and Utilization Project state inpatient and ED databases, all discharges between July 1, 2008, and Sept. 31, 2009, were evaluated for residents aged 18 years or older from three hospitals in three states. After exclusions, 5 million index hospitalizations among 4 million unique patients were studied.

Approximately 40% of the more than 1 million post-discharge acute-care encounters involved a visit to the ED.

Limitations of this study include that the data was derived from only three states, and only hospital-based acute-care visits were measured (i.e. visits to physician offices were not included). As hospitalists, we are responsible for discharges and care transitions. Being sensitive to the common medical conditions resulting in post-discharge ED encounters might improve care transitions.

Bottom line: Hospital readmission rates underestimate ED use following discharge.

Citation: Vashi AA, Fox JP, Carr BG, et al. Use of hospital-based acute care among patients recently discharged from the hospital. JAMA. 2013;309(4):364-371.

Restrictive Transfusion Strategy Beneficial in Upper GI Bleeding

Clinical question: What is the hemoglobin threshold for transfusion of red cells in patients with acute upper GI bleeding?

Background: Controlled trials have shown that restrictive transfusion strategies (Hgb<7) are as effective as liberal transfusion strategies (Hgb<9) in critically ill patients. These studies have excluded patients with GI bleeding. In cases for which GI is not severe, the safest transfusion strategy is controversial.

Study design: Single-center study at Hospital de la Santa Creu i Sant Pau, Barcelona, Spain.

Synopsis: A total of 921 adult patients with acute upper GI bleeding were enrolled and assigned: 461 to a restrictive transfusion strategy (hemoglobin<7) and 460 to a liberal strategy (hemoglobin<9). Patients with massive exsanguinating bleeding, acute coronary syndrome, peripheral vasculopathy, stroke, transient ischemic attack, lower GI bleed, recent trauma or surgery, or low risk of rebleeding were excluded. The primary outcome measure was the rate of death from any cause within the first 45 days.

Secondary outcomes included the rate of further bleeding and the rate of in-hospital complications.

Statistically significant benefit in following a restrictive versus liberal strategy was demonstrated in all major outcomes: mortality (5% vs, 9%, P=0.02), rate of further bleeding (10% vs 18%, P=0.01), and rate of complications (40% vs. 48%, P=0.02).

The study is limited by its inability to be generalized to all patients with acute GI bleeding, as patients with massive exsanguinating bleeds and those with low risk of rebleeding were excluded.

Bottom line: Restrictive transfusion (Hgb<7) significantly improved outcomes for patients with acute upper GI bleeding.

Citation: Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013;368(1):11-21.

Need for Non-ICU Acid Suppression Might Be Predictable

Clinical question: What are the risk factors for nosocomial bleeding in non-critically-ill patients?

Background: Acid-suppressive medication has been shown to reduce the incidence of nosocomial GI bleed in the ICU, but current guidelines recommend against its use in non-critically-ill patients. However, a subgroup of these patients might possess a high enough risk for GI bleed that prophylaxis is warranted.

Study design: Cohort study.

Setting: Academic medical center in Boston.

Synopsis: A total of 75,723 admissions of adult patients hospitalized for three or more days were included. Exclusion criteria included primary discharge diagnosis of GI bleed; principal procedure code of cardiac catheterization; and bleeding episodes occurring while in the ICU or within 48 hours of transfer out of the ICU. The primary outcome was nosocomial GI bleed (>24 hours after admission) occurring outside the ICU.

Nosocomial GI bleeding occurred in 203 patients (0.27%). Independent risk factors for bleeding included age >60, male sex, liver disease, acute renal failure, sepsis, being on a medicine service, prophylactic anticoagulation, and coagulopathy. Based on the data, a scoring system was created that identified a high-risk group in whom the number needed to treat with acid-suppressive medication to prevent one bleed was 48.

The major limitations of this study are its observational nature and the need for validation of the proposed scoring system.

Bottom line: Risk for nosocomial GI bleeding appears predictable and supports the selective use of prophylactic acid suppression in non-critically-ill patients.

Citation: Herzig SJ, Rothberg MB, Feinbloom DB, et al. Risk factors for nosocomial gastrointestinal bleeding and use of acid-suppressive medication in non-critically ill patients. J Gen Intern Med. 2013; January 5. [Epub ahead of print].

Additions, Modifications, and Clarifications Regarding Adult Immunization

Clinical question: What are the changes to the recommended Adult Immunization Schedule for 2013?

Background: Despite the known public health benefits of immunization, adult vaccination rates remain low. The positive impact of strong provider recommendations regarding vaccines underscores the importance of provider awareness of vaccine schedules, precautions, and contraindications.

Study design: Annual Advisory Committee on Immunization Practices (ACIP) review.

Setting: Data from 2011 National Health Interview Survey.

Synopsis: Highlighted changes include: 1) a single dose of pneumococcal 13-valent conjugate (PCV13) vaccine is now recommended for all individuals over the age of 19 with qualifying conditions; 2) clarification regarding pneumococcal polysaccharide (PPSV23) vaccine illustrates that high-risk individuals will receive up to three doses (one or two doses prior to age 65, plus an additional dose after the age of 65); 3) one dose of tetanus, diphtheria, and acellular pertussis (Tdap) vaccine is now recommended for all adults, including individuals age >65; 4) pregnant woman are advised to receive Tdap between 27 and 36 weeks’ gestation, with each pregnancy to provide protection to their newborn in the first months of life; 5) quadrivalent formulations of the live attenuated influenza vaccine (LAIV) and most likely the inactivated influenza vaccine (IIV) will be available in the 2013-2014 influenza season to increase cross-reactive protection against influenza B; and 6) both injection and noninjection illicit drug users are recommended to receive hepatitis A vaccine.

Bottom line: Expanded recommendations for adult immunization provide more opportunities for the practicing hospitalist to improve vaccine capture.

Citation: Advisory Committee on Immunization Practices. Recommended adult immunization schedule: United States, 2013. Ann Int Med. 2013;158(3):191-199.

Clinical question: What is the in-hospital mortality risk associated with hospital-acquired Clostridium difficile infection after accounting for time to infection and baseline mortality risk at admission?

Background: Hospital-acquired C. diff infection (CDI) has been shown to be associated with a higher mortality rate and longer length of stay and cost. Previous studies have demonstrated an independent association of mortality with CDI, but have not incorporated time to infection and baseline mortality risk in the analyses.

Study design: Retrospective observational study.

Setting: Single-center, tertiary-care teaching hospital.

Synopsis: Patients who were hospitalized for more than three days were eligible. A baseline in-hospital mortality risk was estimated for each patient using an internally validated tool. A total of 136,877 admissions were identified. Mean baseline mortality risk was 1.8%. Overall rate of CDI was 1.02%.

Patients in the highest decile of baseline mortality risk had a higher rate of CDI than patients in the lowest decile (2.6% vs. 0.2%). Median time to diagnosis was 12 days. CDI was associated with an unadjusted fourfold higher risk of in-hospital death. When baseline mortality risk was included, the RR of death with CDI was 1.99 (95% CI 1.81-2.19).

Patients in the lowest decile of mortality risk had the highest risk of death (RR 45.70, 95% CI 11.35-183.98) compared with those in the highest decile (RR 1.29, 95% CI 1.11-1.50). Cox modeling estimated a threefold increase in death.

This study is limited by being single-site and the mortality risk model has not been validated externally. Results are also estimated from a small number of cases in the lower deciles.

Bottom line: CDI is associated with threefold higher in-hospital mortality. Patients with higher baseline mortality risk have a higher risk of CDI but have a lesser risk of dying compared with patients with lower baseline mortality risk. Hospitals should continue their efforts to reduce rates of CDI.

Citation: Oake N, Taljaard M, van Walraven C, Wilson K, Roth V, Forster AJ. The effect of hospital-acquired C. diff infection on in-hospital mortality. Arch Intern Med. 2010;170(20):1804-1810.

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

Clinical question: What is the in-hospital mortality risk associated with hospital-acquired Clostridium difficile infection after accounting for time to infection and baseline mortality risk at admission?

Background: Hospital-acquired C. diff infection (CDI) has been shown to be associated with a higher mortality rate and longer length of stay and cost. Previous studies have demonstrated an independent association of mortality with CDI, but have not incorporated time to infection and baseline mortality risk in the analyses.

Study design: Retrospective observational study.

Setting: Single-center, tertiary-care teaching hospital.

Synopsis: Patients who were hospitalized for more than three days were eligible. A baseline in-hospital mortality risk was estimated for each patient using an internally validated tool. A total of 136,877 admissions were identified. Mean baseline mortality risk was 1.8%. Overall rate of CDI was 1.02%.

Patients in the highest decile of baseline mortality risk had a higher rate of CDI than patients in the lowest decile (2.6% vs. 0.2%). Median time to diagnosis was 12 days. CDI was associated with an unadjusted fourfold higher risk of in-hospital death. When baseline mortality risk was included, the RR of death with CDI was 1.99 (95% CI 1.81-2.19).

Patients in the lowest decile of mortality risk had the highest risk of death (RR 45.70, 95% CI 11.35-183.98) compared with those in the highest decile (RR 1.29, 95% CI 1.11-1.50). Cox modeling estimated a threefold increase in death.

This study is limited by being single-site and the mortality risk model has not been validated externally. Results are also estimated from a small number of cases in the lower deciles.

Bottom line: CDI is associated with threefold higher in-hospital mortality. Patients with higher baseline mortality risk have a higher risk of CDI but have a lesser risk of dying compared with patients with lower baseline mortality risk. Hospitals should continue their efforts to reduce rates of CDI.

Citation: Oake N, Taljaard M, van Walraven C, Wilson K, Roth V, Forster AJ. The effect of hospital-acquired C. diff infection on in-hospital mortality. Arch Intern Med. 2010;170(20):1804-1810.

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

Clinical question: What is the in-hospital mortality risk associated with hospital-acquired Clostridium difficile infection after accounting for time to infection and baseline mortality risk at admission?

Background: Hospital-acquired C. diff infection (CDI) has been shown to be associated with a higher mortality rate and longer length of stay and cost. Previous studies have demonstrated an independent association of mortality with CDI, but have not incorporated time to infection and baseline mortality risk in the analyses.

Study design: Retrospective observational study.

Setting: Single-center, tertiary-care teaching hospital.

Synopsis: Patients who were hospitalized for more than three days were eligible. A baseline in-hospital mortality risk was estimated for each patient using an internally validated tool. A total of 136,877 admissions were identified. Mean baseline mortality risk was 1.8%. Overall rate of CDI was 1.02%.

Patients in the highest decile of baseline mortality risk had a higher rate of CDI than patients in the lowest decile (2.6% vs. 0.2%). Median time to diagnosis was 12 days. CDI was associated with an unadjusted fourfold higher risk of in-hospital death. When baseline mortality risk was included, the RR of death with CDI was 1.99 (95% CI 1.81-2.19).

Patients in the lowest decile of mortality risk had the highest risk of death (RR 45.70, 95% CI 11.35-183.98) compared with those in the highest decile (RR 1.29, 95% CI 1.11-1.50). Cox modeling estimated a threefold increase in death.

This study is limited by being single-site and the mortality risk model has not been validated externally. Results are also estimated from a small number of cases in the lower deciles.

Bottom line: CDI is associated with threefold higher in-hospital mortality. Patients with higher baseline mortality risk have a higher risk of CDI but have a lesser risk of dying compared with patients with lower baseline mortality risk. Hospitals should continue their efforts to reduce rates of CDI.

Citation: Oake N, Taljaard M, van Walraven C, Wilson K, Roth V, Forster AJ. The effect of hospital-acquired C. diff infection on in-hospital mortality. Arch Intern Med. 2010;170(20):1804-1810.

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

In This Edition

Literature at a Glance

A guide to this month’s studies

• Early ambulation and LOS in geriatric patients
• Patient-safety movement and hospital harm rates
• Lifestyle modification and weight loss
• Outcomes of transcatheter aortic-valve implantation
• Tool for predicting mortality in advanced dementia
• Residents’ opinion of new duty-hour regulations
• Renal ultrasound predictor for acute kidney injury
• Romiplostim use in immune thrombocytopenia

Increasing Ambulation within 48 Hours of Admission Decreases LOS by Two Days

Clinical question: Is there an association between an early increase in ambulation and length of stay (LOS) in geriatric patients admitted with an acute illness?

Background: Early ambulation leading to better recovery in such illnesses as pneumonia and myocardial infarction is well known, as is early ambulation after hip fracture surgery to prevent complications. However, no specific guidelines exist in regard to ambulation in older patients.

Study design: Prospective, nonblinded study.

Setting: Acute-care geriatric unit in an academic medical center.

Synopsis: A total of 162 patients 65 or older were studied. Data were collected during a four-month period in 2009. A Step Activity Monitor (SAM) was placed on admission. Patients were instructed to walk as usual. Investigators measured the number of steps taken per day and change in steps between the first and second day.

Patients averaged 662.1 steps per day, with a mean step change of 196.5 steps. The adjusted mean difference in LOS for patients who increased their total steps by 600 or more between the first and second day was 2.13 days (95% CI, 1.05-3.97). Patients who had low or negative changes in steps had longer LOS. The 32 patients who walked more than 600 steps were more likely to be men (P=0.02), independently ambulate (P<0.01), and have admitting orders of “ambulate with assist” (P=0.03).

One limitation of this study is that patients who walked more might have been less ill or very functional on admission.

Bottom line: Increasing ambulation early in a hospitalization (first two days) is associated with a decreased LOS in an elderly population.

Citation: Fisher SR, Kuo YF, Graham JE, Ottenbacher KJ, Ostir GV. Early ambulation and length of stay in older adults hospitalized for acute illness. Arch Intern Med. 2010;170(21):1942-1943.

Despite Efforts to Improve Patient Safety in Hospitals, No Reduction in Longitudinal Rates of Harm

Clinical question: As hospitals focus more on programs to improve patient safety, has the rate of harms decreased?

Background: Since the Institute of Medicine published a groundbreaking report (To Err is Human) a little more than a decade ago, policymakers, hospitals, and healthcare organizations have focused more on efforts to improve patient safety with the goal of reducing harms. It is not clear if these efforts have reduced harms.

Study design: Retrospective chart review.

Setting: Ten hospitals in North Carolina.

Synopsis: Ten charts per quarter were randomly selected from each hospital from January 2002 through December 2007. Internal and external reviewers used the IHI Global Trigger Tool for Measuring Adverse Events to identify rates of harm. Harms were classified into categories of severity and assessed for preventability.

Kappa scores were generally higher for internal reviewers, indicating higher reliability for internal reviewers. Internal reviewers identified 588 harms for 10,415 patient days (25.1 harms per 100 patient days), which occurred in 423 unique patients (18.1%). A majority (63.1%) of harms were considered preventable. Forty-one percent of harms were temporary and required intervention; 2.4% caused or contributed to a patient’s death.

There was no significant change over time in the rate of harms (regardless of reviewer type) even after adjusting for demographics.

This study is limited because it is based only in North Carolina hospitals. It was not powered to evaluate change in individual hospitals. There might have been unmeasurable improvements that were not accounted for by the trigger tool.

Bottom line: Despite a higher focus on patient safety, investigators did not find a decrease in the rate of harms. A majority of the harms were preventable. This study should not preclude efforts to continue to improve patient safety.

Citation: Landrigan CP, Parry GJ, Bones CB, Hackbarth AD, Goldmann DA, Sharek PJ. Temporal trends in rates of patient harm resulting from medical care. N Engl J Med. 2010;363(22):2124-2134.

Intensive Lifestyle Modification Improves Weight Loss in Severely Obese Individuals

Clinical question: Does the combination of diet modification and increased physical activity lead to weight loss and improve health risks in severely obese patients?

Background: Obesity is at epidemic proportions, but there are no evidence-based treatment guidelines for severe obesity.

Study design: Randomized, single-blind trial.

Setting: Community volunteers.

Synopsis: A total of 130 individuals with a body mass index (BMI) of ≥35 were randomized to receive lifestyle interventions consisting of diet and initial physical activity for 12 months, or diet for six months and delayed physical activity for the remainder of the year.

The initial-physical-activity group demonstrated greater weight loss at six months, but the overall weight loss did not differ between the two groups. At 12 months, the initial physical activity group lost 12.1 kg and the delayed-physical-activity group lost 9.87 kg. Both groups demonstrated significantly reduced blood pressure, reduced serum liver enzymes, and improved insulin resistance.

Candidates with a history of coronary artery disease, uncontrolled blood pressure, or diabetes were excluded. Participants were provided with prepackaged meal replacements for the first six months and received financial compensation for participation in the study.

This study is limited by the fact that a majority of the participants were female (85.1%). Providing meals to the participants also limits the application of this program to the general public.

Bottom line: The results of this study reflect the importance of diet and exercise on weight loss in obese individuals. However, adherence to the goals of the study required multiple individual and group meetings throughout the year, the provision of prepackaged meals, and some financial incentive for compliance.

Citation: Goodpaster GH, Delany JP, Otto AD, et al. Effects of diet and physical activity interventions on weight loss and cardiometabolic risk factors in severely obese adults: a randomized trial. JAMA. 2010;304 (16):1795-1802.

Transcatheter Aortic-Valve Implantation Is Superior to Standard Nonoperative Therapy for Symptomatic Aortic Stenosis

Clinical question: Is there a mortality benefit to transcatheter valve implantation over standard therapy in nonsurgical candidates with severe aortic stenosis (AS)?

Background: Untreated, symptomatic AS has a high rate of death, but a significant proportion of patients with severe aortic stenosis are poor surgical candidates. Available since 2002, transcatheter aortic-valve implantation (TAVI) is a promising, nonsurgical treatment option for severe AS. However, to date, TAVI has lacked rigorous clinical data.

Study design: Prospective, multicenter, randomized, active-treatment-controlled clinical trial.

Setting: Twenty-one centers, 17 of which were in the U.S.

Synopsis: A total of 358 patients with severe AS who were considered nonsurgical candidates were randomized to either TAVI or standard therapy. A majority (83.8%) of the patients in the standard group underwent balloon aortic valvuloplasty.

Researchers found a significant reduction (HR 0.55, 95% CI 0.40 to 0.74, P<0.001) in all-cause mortality at one year in those patients undergoing TAVI (30.7%) vs. standard therapy (50.7%). Additional benefits included lower rates of the composite endpoints of death from any cause or repeat hospitalization (42.5% vs. 71.6%, P<0.001) and NYHA Functional Class III or IV symptoms (25.2% vs. 58.0%, P<0.001) at one year. However, higher incidences of major strokes (5.0% vs. 1.6%, P=0.06) and major vascular complications (16.2% vs. 1.1%, P<0.001) were seen.

While the one-year mortality benefit of TAVI over standard nonoperative therapy was clearly demonstrated by this study, hospitalists should interpret these data cautiously with respect to their inpatient populations as exclusion criteria were extensive, including bicuspid or noncalcified aortic valve, LVEF less than 20%, and severe renal insufficiency. Additionally, the entity of standard therapy was poorly delineated.

Bottom line: TAVI should be considered in patients with severe aortic stenosis who are not suitable surgical candidates.

Citation: Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363(17):1597-1607.

ADEPT Score Better Predicts Six-Month Mortality in Nursing Home Residents with Advanced Dementia

Clinical question: Are current Medicare hospice eligibility guidelines accurate enough to predict six-month survival in nursing home residents with dementia when compared with the Advanced Dementia Prognostic Tool (ADEPT)?

Background: Incorrectly estimating the life expectancy in almost 5 million nursing home residents with dementia prevents enrollment to palliative care and hospice for those who would benefit most. Creating and validating a mortality risk score would allow increased services to these residents.

Study design: Prospective cohort study.

Setting: Twenty-one nursing homes in Boston.

Synopsis: A total of 606 nursing home residents with advanced dementia were recruited for this study. Each resident was assessed for Medicare hospice eligibility and assigned an ADEPT score. Mortality rate was determined six months later. These two assessment tools were compared regarding their ability to predict six-month mortality.

The mean ADEPT score was 10.1 (range of 1.0-32.5), with a higher score meaning worse prognosis. Sixty-five residents (10.7%) met Medicare hospice eligibility guidelines. A total of 111 residents (18.3%) died.

The ADEPT score was more sensitive (90% vs. 20%) but less specific (28.3% vs. 89%) than Medicare guidelines. The area under the receiver operating characteristic (AUROC) curve was 0.67 (95% CI, 0.62-0.72) for ADEPT and 0.55 (95% CI, 0.51-0.59) for Medicare.

ADEPT was slightly better than hospice guidelines in predicting six-month mortality.

This study was limited in that the resident data were collected at a single random time point and might not reflect reality, as with palliative care and hospice, there usually is a decline in status that stimulates the referrals.

Bottom line: The ADEPT score might better estimate the six-month mortality in nursing home residents with dementia, which can help expand the enrollment of palliative care and hospice for these residents.

Citation: Mitchell SL, Miller SC, Teno JM, Kiely DK, Davis RB, Shaffer ML. Prediction of 6-month survival of nursing home residents with advanced dementia using ADEPT vs hospice eligibility guidelines. JAMA. 2010;304(17):1929-1935.

Residents Concerned about How New ACGME Duty-Hour Restrictions Will Impact Patient Care and Education

Clinical question: How do residents believe the forthcoming revised ACGME Rules for Supervision and Duty Hours will impact their residency?

Background: On July 1, revised ACGME duty-hour rules go into effect, limiting PGY-1 residents to 16-hour duty periods and PGY-2 and above to 28 hours. The effect these recommendations will have on patient care and resident education is unknown.

Study design: Twenty-question electronic, anonymous survey.

Setting: Twenty-three medical centers in the U.S., including residents from all disciplines and years in training.

Synopsis: Twenty-two percent of residents responded to the survey (n=2,521). Overall, 48% of residents disagreed with this statement: “Overall the changes will have a positive effect on education,” while only 26% agreed. Approximately half of those surveyed agreed that the revisions would improve their quality of life, but the same percentage also believed the revisions would increase the length of their residencies.

Residents reacted negatively to the idea that the proposed changes would improve patient safety and quality of care delivered, promote education over service obligations, and prepare them to assume senior roles. In free-text comments, residents expressed concerns about an increased number of handoffs and decreased continuity of care.

While the sample size is large and diverse, results of this survey can be affected by voluntary response bias and, therefore, could be skewed toward more extreme responses (in this case, more negative responses). The wide distribution of the responses suggests this might not be the case.

Bottom line: Residents do not believe the new requirements—though they could improve their quality of life—will positively impact patient care and education.

Citation: Drolet BC, Spalluto LB, Fischer SA. Residents’ perspectives on ACGME regulation of supervision and duty hours—a national survey. N Engl J Med. 2010;363(23):e34(1)-e34.

Decision Rule Might Help Clinicians Decide When to Order Renal Ultrasound to Evaluate Hospitalized Patients with Acute Kidney Injury

Clinical question: Can a clinical prediction rule aid clinicians in deciding when to order a renal ultrasound (RUS) in hospitalized patients with acute kidney injury?

Background: RUS routinely is obtained in patients admitted with acute kidney injury (AKI) to rule out obstruction as a cause of AKI. It is not known if this test adds any additional information in the routine evaluation of AKI and if obtaining the test is cost-effective.

Study design: Cross-sectional study.

Setting: Yale-New Haven Hospital in Connecticut.

Synopsis: This study evaluated 997 inpatients with AKI who underwent RUS. Outcome events were RUS identification of hydronephrosis (HN) or hydronephrosis requiring intervention (HNRI). The patients were divided into two samples: 200 in derivation sample and 797 in validation sample. The derivation sample was used to identify specific factors associated with HN. Seven clinical variables were identified and were used to create three risk groups: low, medium, and high.

In the validation sample, 10.6% of patients had HN and 3.3% had HNRI. The negative predictive value for HN was 96.9%, sensitivity 91.8%, and negative likelihood ratio 0.27. The number needed to screen (NNS) low-risk patients for HN was 32 and 223 for HNRI. Based on their findings, if the patient was classified low-risk, clinicians might be able to delay or avoid ordering RUS.

The major limitation of this study was that it was based at a single institution. This study only evaluated RUS obtained in patients who were hospitalized and might not be applicable to outpatients.

Bottom line: RUS was not found to change clinical management in patients with AKI and classified as low-risk for HN. Limiting RUS to patients who are high-risk for obstruction will increase the chance of finding useful clinical information that can change management decisions and limit cost of unnecessary testing.

Citation: Licurse A, Kim MC, Dziura J, et al. Renal ultrasonography in the evaluation of acute kidney injury: developing a risk stratification framework. Arch Intern Med. 2010;170(21):1900-1907.

Romiplostim Has Higher Rate of Platelet Response and Fewer Adverse Events in Patients with Immune Thrombocytopenia

Clinical question: Does the use of romiplostim lead to increased platelet counts and lower rates of splenectomy and other adverse events when compared with standard therapy in patients with immune thrombocytopenia?

Background: Romiplostim is a thrombopoetin mimetic used to increase platelet counts in immune thrombocytopenia. Initial treatments for this disease involve glucocorticoids or intravenous immune globulin. Most patients require second-line medical or surgical therapies, including splenectomy.

Study design: Randomized, open-label controlled trial.

Setting: Eighty-five medical centers in North America, Europe, and Australia.

Synopsis: A total of 234 patients were randomized in a 2:1 ratio to receive either romiplostim or the medical standard of care. Co-primary endpoints were the incidence of treatment failure and the incidence of splenectomy; secondary endpoints included time to splenectomy, platelet count, platelet response, and quality of life. Treatment failure was defined as a platelet count of 20x109 per liter or lower for four weeks, or a major bleeding event.

At the end of 52 weeks, patients receiving romiplostim had higher platelet counts, fewer bleeding events, less need for splenectomy (9% vs. 36%), and a better quality of life.

The short-term use of romiplostim in this study was not associated with an increase in adverse events when compared with standard therapy. However, maintenance of the elevated platelet count, which results from romiplostim treatment, requires continuous use of the drug; the long-term effects are unknown.

Bottom line: In patients with immune thrombocytopenia, romiplostim leads to increased platelet counts, decreased bleeding events, and decreased need for splenectomy compared to standard of care. However, the cost of the medication, when compared with current therapies, could be prohibitive.

Citation: Kuter DJ, Rummel M, Boccia R, et al. Romiplostim or standard of care in patients with immune thrombocytopenia. N Engl J Med. 2010;363(20):1889-1899. TH

Pediatric HM Literature

Parental History Portends Persistent Chronic Abdominal Pain in Children

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.

Clinical question: What is the quality of the current evidence for potential prognostic factors of persistent chronic abdominal pain in children?

Background: Chronic abdominal pain (CAP) is a prevalent condition in childhood that might be associated with increased healthcare costs, including hospital admission. Retrospective studies have implicated psychosocial factors as being of prognostic relevance, but these are unable to offer greater insight into the relationship given the complex nature of this chronic illness.

Study design: Systematic review of literature.

Setting: MEDLINE, EMBASE, and PsycINFO.

Synopsis: The databases were searched through June 2008 for articles that focused on children 4 to 18 years of age; used criteria for CAP as defined by Apley and Naish, von Baeyer and Walker, or the Rome Committee; and demonstrated prospective determination of outcomes. Eight studies were included in the final review, and the levels of evidence were graded based on assessment of risk for bias.

Female sex and severity of baseline abdominal pain did not predict persistence of CAP, although there was conflicting evidence as to the role of psychological factors. There was moderate evidence that having a parent with functional gastrointestinal (GI) symptoms predicted persistence of CAP in children, and there was weak evidence that having parents who eschew psychological factors in favor of searching for an organic explanation predicted persistence of CAP.

Due to a limited number of studies or conflicting associations, conclusions regarding the following factors could not be drawn: age, educational level, duration of CAP, associated symptoms, socioeconomic status, and history of two or more surgical operations.

In addition, the univariate analysis used by the included studies might not be appropriate for such a multifactorially complex disease. Nevertheless, this study challenges the conventional wisdom that psychological factors predict persistence of pain and should remind clinicians to assess for parental functional GI disorders in this patient population.

Bottom line: Parental history of functional GI disorders predicts persistence of CAP in children.

Citation: Gieteling MJ, Bierma-Zeinstra SM, van Leeuwen Y, Passchier J, Berger MY. Prognostic factors for persistence of chronic abdominal pain in children. J Pediatr Gastroenterol Nutr. 2011;52(2):154-161.

In This Edition

Literature at a Glance

A guide to this month’s studies

• Early ambulation and LOS in geriatric patients
• Patient-safety movement and hospital harm rates
• Lifestyle modification and weight loss
• Outcomes of transcatheter aortic-valve implantation
• Tool for predicting mortality in advanced dementia
• Residents’ opinion of new duty-hour regulations
• Renal ultrasound predictor for acute kidney injury
• Romiplostim use in immune thrombocytopenia

Increasing Ambulation within 48 Hours of Admission Decreases LOS by Two Days

Clinical question: Is there an association between an early increase in ambulation and length of stay (LOS) in geriatric patients admitted with an acute illness?

Background: Early ambulation leading to better recovery in such illnesses as pneumonia and myocardial infarction is well known, as is early ambulation after hip fracture surgery to prevent complications. However, no specific guidelines exist in regard to ambulation in older patients.

Study design: Prospective, nonblinded study.

Setting: Acute-care geriatric unit in an academic medical center.

Synopsis: A total of 162 patients 65 or older were studied. Data were collected during a four-month period in 2009. A Step Activity Monitor (SAM) was placed on admission. Patients were instructed to walk as usual. Investigators measured the number of steps taken per day and change in steps between the first and second day.

Patients averaged 662.1 steps per day, with a mean step change of 196.5 steps. The adjusted mean difference in LOS for patients who increased their total steps by 600 or more between the first and second day was 2.13 days (95% CI, 1.05-3.97). Patients who had low or negative changes in steps had longer LOS. The 32 patients who walked more than 600 steps were more likely to be men (P=0.02), independently ambulate (P<0.01), and have admitting orders of “ambulate with assist” (P=0.03).

One limitation of this study is that patients who walked more might have been less ill or very functional on admission.

Bottom line: Increasing ambulation early in a hospitalization (first two days) is associated with a decreased LOS in an elderly population.

Citation: Fisher SR, Kuo YF, Graham JE, Ottenbacher KJ, Ostir GV. Early ambulation and length of stay in older adults hospitalized for acute illness. Arch Intern Med. 2010;170(21):1942-1943.

Despite Efforts to Improve Patient Safety in Hospitals, No Reduction in Longitudinal Rates of Harm

Clinical question: As hospitals focus more on programs to improve patient safety, has the rate of harms decreased?

Background: Since the Institute of Medicine published a groundbreaking report (To Err is Human) a little more than a decade ago, policymakers, hospitals, and healthcare organizations have focused more on efforts to improve patient safety with the goal of reducing harms. It is not clear if these efforts have reduced harms.

Study design: Retrospective chart review.

Setting: Ten hospitals in North Carolina.

Synopsis: Ten charts per quarter were randomly selected from each hospital from January 2002 through December 2007. Internal and external reviewers used the IHI Global Trigger Tool for Measuring Adverse Events to identify rates of harm. Harms were classified into categories of severity and assessed for preventability.

Kappa scores were generally higher for internal reviewers, indicating higher reliability for internal reviewers. Internal reviewers identified 588 harms for 10,415 patient days (25.1 harms per 100 patient days), which occurred in 423 unique patients (18.1%). A majority (63.1%) of harms were considered preventable. Forty-one percent of harms were temporary and required intervention; 2.4% caused or contributed to a patient’s death.

There was no significant change over time in the rate of harms (regardless of reviewer type) even after adjusting for demographics.

This study is limited because it is based only in North Carolina hospitals. It was not powered to evaluate change in individual hospitals. There might have been unmeasurable improvements that were not accounted for by the trigger tool.

Bottom line: Despite a higher focus on patient safety, investigators did not find a decrease in the rate of harms. A majority of the harms were preventable. This study should not preclude efforts to continue to improve patient safety.

Citation: Landrigan CP, Parry GJ, Bones CB, Hackbarth AD, Goldmann DA, Sharek PJ. Temporal trends in rates of patient harm resulting from medical care. N Engl J Med. 2010;363(22):2124-2134.

Intensive Lifestyle Modification Improves Weight Loss in Severely Obese Individuals

Clinical question: Does the combination of diet modification and increased physical activity lead to weight loss and improve health risks in severely obese patients?

Background: Obesity is at epidemic proportions, but there are no evidence-based treatment guidelines for severe obesity.

Study design: Randomized, single-blind trial.

Setting: Community volunteers.

Synopsis: A total of 130 individuals with a body mass index (BMI) of ≥35 were randomized to receive lifestyle interventions consisting of diet and initial physical activity for 12 months, or diet for six months and delayed physical activity for the remainder of the year.

The initial-physical-activity group demonstrated greater weight loss at six months, but the overall weight loss did not differ between the two groups. At 12 months, the initial physical activity group lost 12.1 kg and the delayed-physical-activity group lost 9.87 kg. Both groups demonstrated significantly reduced blood pressure, reduced serum liver enzymes, and improved insulin resistance.

Candidates with a history of coronary artery disease, uncontrolled blood pressure, or diabetes were excluded. Participants were provided with prepackaged meal replacements for the first six months and received financial compensation for participation in the study.

This study is limited by the fact that a majority of the participants were female (85.1%). Providing meals to the participants also limits the application of this program to the general public.

Bottom line: The results of this study reflect the importance of diet and exercise on weight loss in obese individuals. However, adherence to the goals of the study required multiple individual and group meetings throughout the year, the provision of prepackaged meals, and some financial incentive for compliance.

Citation: Goodpaster GH, Delany JP, Otto AD, et al. Effects of diet and physical activity interventions on weight loss and cardiometabolic risk factors in severely obese adults: a randomized trial. JAMA. 2010;304 (16):1795-1802.

Transcatheter Aortic-Valve Implantation Is Superior to Standard Nonoperative Therapy for Symptomatic Aortic Stenosis

Clinical question: Is there a mortality benefit to transcatheter valve implantation over standard therapy in nonsurgical candidates with severe aortic stenosis (AS)?

Background: Untreated, symptomatic AS has a high rate of death, but a significant proportion of patients with severe aortic stenosis are poor surgical candidates. Available since 2002, transcatheter aortic-valve implantation (TAVI) is a promising, nonsurgical treatment option for severe AS. However, to date, TAVI has lacked rigorous clinical data.

Study design: Prospective, multicenter, randomized, active-treatment-controlled clinical trial.

Setting: Twenty-one centers, 17 of which were in the U.S.

Synopsis: A total of 358 patients with severe AS who were considered nonsurgical candidates were randomized to either TAVI or standard therapy. A majority (83.8%) of the patients in the standard group underwent balloon aortic valvuloplasty.

Researchers found a significant reduction (HR 0.55, 95% CI 0.40 to 0.74, P<0.001) in all-cause mortality at one year in those patients undergoing TAVI (30.7%) vs. standard therapy (50.7%). Additional benefits included lower rates of the composite endpoints of death from any cause or repeat hospitalization (42.5% vs. 71.6%, P<0.001) and NYHA Functional Class III or IV symptoms (25.2% vs. 58.0%, P<0.001) at one year. However, higher incidences of major strokes (5.0% vs. 1.6%, P=0.06) and major vascular complications (16.2% vs. 1.1%, P<0.001) were seen.

While the one-year mortality benefit of TAVI over standard nonoperative therapy was clearly demonstrated by this study, hospitalists should interpret these data cautiously with respect to their inpatient populations as exclusion criteria were extensive, including bicuspid or noncalcified aortic valve, LVEF less than 20%, and severe renal insufficiency. Additionally, the entity of standard therapy was poorly delineated.

Bottom line: TAVI should be considered in patients with severe aortic stenosis who are not suitable surgical candidates.

Citation: Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363(17):1597-1607.

ADEPT Score Better Predicts Six-Month Mortality in Nursing Home Residents with Advanced Dementia

Clinical question: Are current Medicare hospice eligibility guidelines accurate enough to predict six-month survival in nursing home residents with dementia when compared with the Advanced Dementia Prognostic Tool (ADEPT)?

Background: Incorrectly estimating the life expectancy in almost 5 million nursing home residents with dementia prevents enrollment to palliative care and hospice for those who would benefit most. Creating and validating a mortality risk score would allow increased services to these residents.

Study design: Prospective cohort study.

Setting: Twenty-one nursing homes in Boston.

Synopsis: A total of 606 nursing home residents with advanced dementia were recruited for this study. Each resident was assessed for Medicare hospice eligibility and assigned an ADEPT score. Mortality rate was determined six months later. These two assessment tools were compared regarding their ability to predict six-month mortality.

The mean ADEPT score was 10.1 (range of 1.0-32.5), with a higher score meaning worse prognosis. Sixty-five residents (10.7%) met Medicare hospice eligibility guidelines. A total of 111 residents (18.3%) died.

The ADEPT score was more sensitive (90% vs. 20%) but less specific (28.3% vs. 89%) than Medicare guidelines. The area under the receiver operating characteristic (AUROC) curve was 0.67 (95% CI, 0.62-0.72) for ADEPT and 0.55 (95% CI, 0.51-0.59) for Medicare.

ADEPT was slightly better than hospice guidelines in predicting six-month mortality.

This study was limited in that the resident data were collected at a single random time point and might not reflect reality, as with palliative care and hospice, there usually is a decline in status that stimulates the referrals.

Bottom line: The ADEPT score might better estimate the six-month mortality in nursing home residents with dementia, which can help expand the enrollment of palliative care and hospice for these residents.

Citation: Mitchell SL, Miller SC, Teno JM, Kiely DK, Davis RB, Shaffer ML. Prediction of 6-month survival of nursing home residents with advanced dementia using ADEPT vs hospice eligibility guidelines. JAMA. 2010;304(17):1929-1935.

Residents Concerned about How New ACGME Duty-Hour Restrictions Will Impact Patient Care and Education

Clinical question: How do residents believe the forthcoming revised ACGME Rules for Supervision and Duty Hours will impact their residency?

Background: On July 1, revised ACGME duty-hour rules go into effect, limiting PGY-1 residents to 16-hour duty periods and PGY-2 and above to 28 hours. The effect these recommendations will have on patient care and resident education is unknown.

Study design: Twenty-question electronic, anonymous survey.

Setting: Twenty-three medical centers in the U.S., including residents from all disciplines and years in training.

Synopsis: Twenty-two percent of residents responded to the survey (n=2,521). Overall, 48% of residents disagreed with this statement: “Overall the changes will have a positive effect on education,” while only 26% agreed. Approximately half of those surveyed agreed that the revisions would improve their quality of life, but the same percentage also believed the revisions would increase the length of their residencies.

Residents reacted negatively to the idea that the proposed changes would improve patient safety and quality of care delivered, promote education over service obligations, and prepare them to assume senior roles. In free-text comments, residents expressed concerns about an increased number of handoffs and decreased continuity of care.

While the sample size is large and diverse, results of this survey can be affected by voluntary response bias and, therefore, could be skewed toward more extreme responses (in this case, more negative responses). The wide distribution of the responses suggests this might not be the case.

Bottom line: Residents do not believe the new requirements—though they could improve their quality of life—will positively impact patient care and education.

Citation: Drolet BC, Spalluto LB, Fischer SA. Residents’ perspectives on ACGME regulation of supervision and duty hours—a national survey. N Engl J Med. 2010;363(23):e34(1)-e34.

Decision Rule Might Help Clinicians Decide When to Order Renal Ultrasound to Evaluate Hospitalized Patients with Acute Kidney Injury

Clinical question: Can a clinical prediction rule aid clinicians in deciding when to order a renal ultrasound (RUS) in hospitalized patients with acute kidney injury?

Background: RUS routinely is obtained in patients admitted with acute kidney injury (AKI) to rule out obstruction as a cause of AKI. It is not known if this test adds any additional information in the routine evaluation of AKI and if obtaining the test is cost-effective.

Study design: Cross-sectional study.

Setting: Yale-New Haven Hospital in Connecticut.

Synopsis: This study evaluated 997 inpatients with AKI who underwent RUS. Outcome events were RUS identification of hydronephrosis (HN) or hydronephrosis requiring intervention (HNRI). The patients were divided into two samples: 200 in derivation sample and 797 in validation sample. The derivation sample was used to identify specific factors associated with HN. Seven clinical variables were identified and were used to create three risk groups: low, medium, and high.

In the validation sample, 10.6% of patients had HN and 3.3% had HNRI. The negative predictive value for HN was 96.9%, sensitivity 91.8%, and negative likelihood ratio 0.27. The number needed to screen (NNS) low-risk patients for HN was 32 and 223 for HNRI. Based on their findings, if the patient was classified low-risk, clinicians might be able to delay or avoid ordering RUS.

The major limitation of this study was that it was based at a single institution. This study only evaluated RUS obtained in patients who were hospitalized and might not be applicable to outpatients.

Bottom line: RUS was not found to change clinical management in patients with AKI and classified as low-risk for HN. Limiting RUS to patients who are high-risk for obstruction will increase the chance of finding useful clinical information that can change management decisions and limit cost of unnecessary testing.

Citation: Licurse A, Kim MC, Dziura J, et al. Renal ultrasonography in the evaluation of acute kidney injury: developing a risk stratification framework. Arch Intern Med. 2010;170(21):1900-1907.

Romiplostim Has Higher Rate of Platelet Response and Fewer Adverse Events in Patients with Immune Thrombocytopenia

Clinical question: Does the use of romiplostim lead to increased platelet counts and lower rates of splenectomy and other adverse events when compared with standard therapy in patients with immune thrombocytopenia?

Background: Romiplostim is a thrombopoetin mimetic used to increase platelet counts in immune thrombocytopenia. Initial treatments for this disease involve glucocorticoids or intravenous immune globulin. Most patients require second-line medical or surgical therapies, including splenectomy.

Study design: Randomized, open-label controlled trial.

Setting: Eighty-five medical centers in North America, Europe, and Australia.

Synopsis: A total of 234 patients were randomized in a 2:1 ratio to receive either romiplostim or the medical standard of care. Co-primary endpoints were the incidence of treatment failure and the incidence of splenectomy; secondary endpoints included time to splenectomy, platelet count, platelet response, and quality of life. Treatment failure was defined as a platelet count of 20x109 per liter or lower for four weeks, or a major bleeding event.

At the end of 52 weeks, patients receiving romiplostim had higher platelet counts, fewer bleeding events, less need for splenectomy (9% vs. 36%), and a better quality of life.

The short-term use of romiplostim in this study was not associated with an increase in adverse events when compared with standard therapy. However, maintenance of the elevated platelet count, which results from romiplostim treatment, requires continuous use of the drug; the long-term effects are unknown.

Bottom line: In patients with immune thrombocytopenia, romiplostim leads to increased platelet counts, decreased bleeding events, and decreased need for splenectomy compared to standard of care. However, the cost of the medication, when compared with current therapies, could be prohibitive.

Citation: Kuter DJ, Rummel M, Boccia R, et al. Romiplostim or standard of care in patients with immune thrombocytopenia. N Engl J Med. 2010;363(20):1889-1899. TH

Pediatric HM Literature

Parental History Portends Persistent Chronic Abdominal Pain in Children

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.

Clinical question: What is the quality of the current evidence for potential prognostic factors of persistent chronic abdominal pain in children?

Background: Chronic abdominal pain (CAP) is a prevalent condition in childhood that might be associated with increased healthcare costs, including hospital admission. Retrospective studies have implicated psychosocial factors as being of prognostic relevance, but these are unable to offer greater insight into the relationship given the complex nature of this chronic illness.

Study design: Systematic review of literature.

Setting: MEDLINE, EMBASE, and PsycINFO.

Synopsis: The databases were searched through June 2008 for articles that focused on children 4 to 18 years of age; used criteria for CAP as defined by Apley and Naish, von Baeyer and Walker, or the Rome Committee; and demonstrated prospective determination of outcomes. Eight studies were included in the final review, and the levels of evidence were graded based on assessment of risk for bias.

Female sex and severity of baseline abdominal pain did not predict persistence of CAP, although there was conflicting evidence as to the role of psychological factors. There was moderate evidence that having a parent with functional gastrointestinal (GI) symptoms predicted persistence of CAP in children, and there was weak evidence that having parents who eschew psychological factors in favor of searching for an organic explanation predicted persistence of CAP.

Due to a limited number of studies or conflicting associations, conclusions regarding the following factors could not be drawn: age, educational level, duration of CAP, associated symptoms, socioeconomic status, and history of two or more surgical operations.

In addition, the univariate analysis used by the included studies might not be appropriate for such a multifactorially complex disease. Nevertheless, this study challenges the conventional wisdom that psychological factors predict persistence of pain and should remind clinicians to assess for parental functional GI disorders in this patient population.

Bottom line: Parental history of functional GI disorders predicts persistence of CAP in children.

Citation: Gieteling MJ, Bierma-Zeinstra SM, van Leeuwen Y, Passchier J, Berger MY. Prognostic factors for persistence of chronic abdominal pain in children. J Pediatr Gastroenterol Nutr. 2011;52(2):154-161.

In This Edition

Literature at a Glance

A guide to this month’s studies

• Early ambulation and LOS in geriatric patients
• Patient-safety movement and hospital harm rates
• Lifestyle modification and weight loss
• Outcomes of transcatheter aortic-valve implantation
• Tool for predicting mortality in advanced dementia
• Residents’ opinion of new duty-hour regulations
• Renal ultrasound predictor for acute kidney injury
• Romiplostim use in immune thrombocytopenia

Increasing Ambulation within 48 Hours of Admission Decreases LOS by Two Days

Clinical question: Is there an association between an early increase in ambulation and length of stay (LOS) in geriatric patients admitted with an acute illness?

Background: Early ambulation leading to better recovery in such illnesses as pneumonia and myocardial infarction is well known, as is early ambulation after hip fracture surgery to prevent complications. However, no specific guidelines exist in regard to ambulation in older patients.

Study design: Prospective, nonblinded study.

Setting: Acute-care geriatric unit in an academic medical center.

Synopsis: A total of 162 patients 65 or older were studied. Data were collected during a four-month period in 2009. A Step Activity Monitor (SAM) was placed on admission. Patients were instructed to walk as usual. Investigators measured the number of steps taken per day and change in steps between the first and second day.

Patients averaged 662.1 steps per day, with a mean step change of 196.5 steps. The adjusted mean difference in LOS for patients who increased their total steps by 600 or more between the first and second day was 2.13 days (95% CI, 1.05-3.97). Patients who had low or negative changes in steps had longer LOS. The 32 patients who walked more than 600 steps were more likely to be men (P=0.02), independently ambulate (P<0.01), and have admitting orders of “ambulate with assist” (P=0.03).

One limitation of this study is that patients who walked more might have been less ill or very functional on admission.

Bottom line: Increasing ambulation early in a hospitalization (first two days) is associated with a decreased LOS in an elderly population.

Citation: Fisher SR, Kuo YF, Graham JE, Ottenbacher KJ, Ostir GV. Early ambulation and length of stay in older adults hospitalized for acute illness. Arch Intern Med. 2010;170(21):1942-1943.

Despite Efforts to Improve Patient Safety in Hospitals, No Reduction in Longitudinal Rates of Harm

Clinical question: As hospitals focus more on programs to improve patient safety, has the rate of harms decreased?

Background: Since the Institute of Medicine published a groundbreaking report (To Err is Human) a little more than a decade ago, policymakers, hospitals, and healthcare organizations have focused more on efforts to improve patient safety with the goal of reducing harms. It is not clear if these efforts have reduced harms.

Study design: Retrospective chart review.

Setting: Ten hospitals in North Carolina.

Synopsis: Ten charts per quarter were randomly selected from each hospital from January 2002 through December 2007. Internal and external reviewers used the IHI Global Trigger Tool for Measuring Adverse Events to identify rates of harm. Harms were classified into categories of severity and assessed for preventability.

Kappa scores were generally higher for internal reviewers, indicating higher reliability for internal reviewers. Internal reviewers identified 588 harms for 10,415 patient days (25.1 harms per 100 patient days), which occurred in 423 unique patients (18.1%). A majority (63.1%) of harms were considered preventable. Forty-one percent of harms were temporary and required intervention; 2.4% caused or contributed to a patient’s death.

There was no significant change over time in the rate of harms (regardless of reviewer type) even after adjusting for demographics.

This study is limited because it is based only in North Carolina hospitals. It was not powered to evaluate change in individual hospitals. There might have been unmeasurable improvements that were not accounted for by the trigger tool.

Bottom line: Despite a higher focus on patient safety, investigators did not find a decrease in the rate of harms. A majority of the harms were preventable. This study should not preclude efforts to continue to improve patient safety.

Citation: Landrigan CP, Parry GJ, Bones CB, Hackbarth AD, Goldmann DA, Sharek PJ. Temporal trends in rates of patient harm resulting from medical care. N Engl J Med. 2010;363(22):2124-2134.

Intensive Lifestyle Modification Improves Weight Loss in Severely Obese Individuals

Clinical question: Does the combination of diet modification and increased physical activity lead to weight loss and improve health risks in severely obese patients?

Background: Obesity is at epidemic proportions, but there are no evidence-based treatment guidelines for severe obesity.

Study design: Randomized, single-blind trial.

Setting: Community volunteers.

Synopsis: A total of 130 individuals with a body mass index (BMI) of ≥35 were randomized to receive lifestyle interventions consisting of diet and initial physical activity for 12 months, or diet for six months and delayed physical activity for the remainder of the year.

The initial-physical-activity group demonstrated greater weight loss at six months, but the overall weight loss did not differ between the two groups. At 12 months, the initial physical activity group lost 12.1 kg and the delayed-physical-activity group lost 9.87 kg. Both groups demonstrated significantly reduced blood pressure, reduced serum liver enzymes, and improved insulin resistance.

Candidates with a history of coronary artery disease, uncontrolled blood pressure, or diabetes were excluded. Participants were provided with prepackaged meal replacements for the first six months and received financial compensation for participation in the study.

This study is limited by the fact that a majority of the participants were female (85.1%). Providing meals to the participants also limits the application of this program to the general public.

Bottom line: The results of this study reflect the importance of diet and exercise on weight loss in obese individuals. However, adherence to the goals of the study required multiple individual and group meetings throughout the year, the provision of prepackaged meals, and some financial incentive for compliance.

Citation: Goodpaster GH, Delany JP, Otto AD, et al. Effects of diet and physical activity interventions on weight loss and cardiometabolic risk factors in severely obese adults: a randomized trial. JAMA. 2010;304 (16):1795-1802.

Transcatheter Aortic-Valve Implantation Is Superior to Standard Nonoperative Therapy for Symptomatic Aortic Stenosis

Clinical question: Is there a mortality benefit to transcatheter valve implantation over standard therapy in nonsurgical candidates with severe aortic stenosis (AS)?

Background: Untreated, symptomatic AS has a high rate of death, but a significant proportion of patients with severe aortic stenosis are poor surgical candidates. Available since 2002, transcatheter aortic-valve implantation (TAVI) is a promising, nonsurgical treatment option for severe AS. However, to date, TAVI has lacked rigorous clinical data.

Study design: Prospective, multicenter, randomized, active-treatment-controlled clinical trial.

Setting: Twenty-one centers, 17 of which were in the U.S.

Synopsis: A total of 358 patients with severe AS who were considered nonsurgical candidates were randomized to either TAVI or standard therapy. A majority (83.8%) of the patients in the standard group underwent balloon aortic valvuloplasty.

Researchers found a significant reduction (HR 0.55, 95% CI 0.40 to 0.74, P<0.001) in all-cause mortality at one year in those patients undergoing TAVI (30.7%) vs. standard therapy (50.7%). Additional benefits included lower rates of the composite endpoints of death from any cause or repeat hospitalization (42.5% vs. 71.6%, P<0.001) and NYHA Functional Class III or IV symptoms (25.2% vs. 58.0%, P<0.001) at one year. However, higher incidences of major strokes (5.0% vs. 1.6%, P=0.06) and major vascular complications (16.2% vs. 1.1%, P<0.001) were seen.

While the one-year mortality benefit of TAVI over standard nonoperative therapy was clearly demonstrated by this study, hospitalists should interpret these data cautiously with respect to their inpatient populations as exclusion criteria were extensive, including bicuspid or noncalcified aortic valve, LVEF less than 20%, and severe renal insufficiency. Additionally, the entity of standard therapy was poorly delineated.

Bottom line: TAVI should be considered in patients with severe aortic stenosis who are not suitable surgical candidates.

Citation: Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363(17):1597-1607.

ADEPT Score Better Predicts Six-Month Mortality in Nursing Home Residents with Advanced Dementia

Clinical question: Are current Medicare hospice eligibility guidelines accurate enough to predict six-month survival in nursing home residents with dementia when compared with the Advanced Dementia Prognostic Tool (ADEPT)?

Background: Incorrectly estimating the life expectancy in almost 5 million nursing home residents with dementia prevents enrollment to palliative care and hospice for those who would benefit most. Creating and validating a mortality risk score would allow increased services to these residents.

Study design: Prospective cohort study.

Setting: Twenty-one nursing homes in Boston.

Synopsis: A total of 606 nursing home residents with advanced dementia were recruited for this study. Each resident was assessed for Medicare hospice eligibility and assigned an ADEPT score. Mortality rate was determined six months later. These two assessment tools were compared regarding their ability to predict six-month mortality.

The mean ADEPT score was 10.1 (range of 1.0-32.5), with a higher score meaning worse prognosis. Sixty-five residents (10.7%) met Medicare hospice eligibility guidelines. A total of 111 residents (18.3%) died.

The ADEPT score was more sensitive (90% vs. 20%) but less specific (28.3% vs. 89%) than Medicare guidelines. The area under the receiver operating characteristic (AUROC) curve was 0.67 (95% CI, 0.62-0.72) for ADEPT and 0.55 (95% CI, 0.51-0.59) for Medicare.

ADEPT was slightly better than hospice guidelines in predicting six-month mortality.

This study was limited in that the resident data were collected at a single random time point and might not reflect reality, as with palliative care and hospice, there usually is a decline in status that stimulates the referrals.

Bottom line: The ADEPT score might better estimate the six-month mortality in nursing home residents with dementia, which can help expand the enrollment of palliative care and hospice for these residents.

Citation: Mitchell SL, Miller SC, Teno JM, Kiely DK, Davis RB, Shaffer ML. Prediction of 6-month survival of nursing home residents with advanced dementia using ADEPT vs hospice eligibility guidelines. JAMA. 2010;304(17):1929-1935.

Residents Concerned about How New ACGME Duty-Hour Restrictions Will Impact Patient Care and Education

Clinical question: How do residents believe the forthcoming revised ACGME Rules for Supervision and Duty Hours will impact their residency?

Background: On July 1, revised ACGME duty-hour rules go into effect, limiting PGY-1 residents to 16-hour duty periods and PGY-2 and above to 28 hours. The effect these recommendations will have on patient care and resident education is unknown.

Study design: Twenty-question electronic, anonymous survey.

Setting: Twenty-three medical centers in the U.S., including residents from all disciplines and years in training.

Synopsis: Twenty-two percent of residents responded to the survey (n=2,521). Overall, 48% of residents disagreed with this statement: “Overall the changes will have a positive effect on education,” while only 26% agreed. Approximately half of those surveyed agreed that the revisions would improve their quality of life, but the same percentage also believed the revisions would increase the length of their residencies.

Residents reacted negatively to the idea that the proposed changes would improve patient safety and quality of care delivered, promote education over service obligations, and prepare them to assume senior roles. In free-text comments, residents expressed concerns about an increased number of handoffs and decreased continuity of care.

While the sample size is large and diverse, results of this survey can be affected by voluntary response bias and, therefore, could be skewed toward more extreme responses (in this case, more negative responses). The wide distribution of the responses suggests this might not be the case.

Bottom line: Residents do not believe the new requirements—though they could improve their quality of life—will positively impact patient care and education.

Citation: Drolet BC, Spalluto LB, Fischer SA. Residents’ perspectives on ACGME regulation of supervision and duty hours—a national survey. N Engl J Med. 2010;363(23):e34(1)-e34.

Decision Rule Might Help Clinicians Decide When to Order Renal Ultrasound to Evaluate Hospitalized Patients with Acute Kidney Injury

Clinical question: Can a clinical prediction rule aid clinicians in deciding when to order a renal ultrasound (RUS) in hospitalized patients with acute kidney injury?

Background: RUS routinely is obtained in patients admitted with acute kidney injury (AKI) to rule out obstruction as a cause of AKI. It is not known if this test adds any additional information in the routine evaluation of AKI and if obtaining the test is cost-effective.

Study design: Cross-sectional study.

Setting: Yale-New Haven Hospital in Connecticut.

Synopsis: This study evaluated 997 inpatients with AKI who underwent RUS. Outcome events were RUS identification of hydronephrosis (HN) or hydronephrosis requiring intervention (HNRI). The patients were divided into two samples: 200 in derivation sample and 797 in validation sample. The derivation sample was used to identify specific factors associated with HN. Seven clinical variables were identified and were used to create three risk groups: low, medium, and high.

In the validation sample, 10.6% of patients had HN and 3.3% had HNRI. The negative predictive value for HN was 96.9%, sensitivity 91.8%, and negative likelihood ratio 0.27. The number needed to screen (NNS) low-risk patients for HN was 32 and 223 for HNRI. Based on their findings, if the patient was classified low-risk, clinicians might be able to delay or avoid ordering RUS.

The major limitation of this study was that it was based at a single institution. This study only evaluated RUS obtained in patients who were hospitalized and might not be applicable to outpatients.

Bottom line: RUS was not found to change clinical management in patients with AKI and classified as low-risk for HN. Limiting RUS to patients who are high-risk for obstruction will increase the chance of finding useful clinical information that can change management decisions and limit cost of unnecessary testing.

Citation: Licurse A, Kim MC, Dziura J, et al. Renal ultrasonography in the evaluation of acute kidney injury: developing a risk stratification framework. Arch Intern Med. 2010;170(21):1900-1907.

Romiplostim Has Higher Rate of Platelet Response and Fewer Adverse Events in Patients with Immune Thrombocytopenia

Clinical question: Does the use of romiplostim lead to increased platelet counts and lower rates of splenectomy and other adverse events when compared with standard therapy in patients with immune thrombocytopenia?

Background: Romiplostim is a thrombopoetin mimetic used to increase platelet counts in immune thrombocytopenia. Initial treatments for this disease involve glucocorticoids or intravenous immune globulin. Most patients require second-line medical or surgical therapies, including splenectomy.

Study design: Randomized, open-label controlled trial.

Setting: Eighty-five medical centers in North America, Europe, and Australia.

Synopsis: A total of 234 patients were randomized in a 2:1 ratio to receive either romiplostim or the medical standard of care. Co-primary endpoints were the incidence of treatment failure and the incidence of splenectomy; secondary endpoints included time to splenectomy, platelet count, platelet response, and quality of life. Treatment failure was defined as a platelet count of 20x109 per liter or lower for four weeks, or a major bleeding event.

At the end of 52 weeks, patients receiving romiplostim had higher platelet counts, fewer bleeding events, less need for splenectomy (9% vs. 36%), and a better quality of life.

The short-term use of romiplostim in this study was not associated with an increase in adverse events when compared with standard therapy. However, maintenance of the elevated platelet count, which results from romiplostim treatment, requires continuous use of the drug; the long-term effects are unknown.

Bottom line: In patients with immune thrombocytopenia, romiplostim leads to increased platelet counts, decreased bleeding events, and decreased need for splenectomy compared to standard of care. However, the cost of the medication, when compared with current therapies, could be prohibitive.

Citation: Kuter DJ, Rummel M, Boccia R, et al. Romiplostim or standard of care in patients with immune thrombocytopenia. N Engl J Med. 2010;363(20):1889-1899. TH

Pediatric HM Literature

Parental History Portends Persistent Chronic Abdominal Pain in Children

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.

Clinical question: What is the quality of the current evidence for potential prognostic factors of persistent chronic abdominal pain in children?

Background: Chronic abdominal pain (CAP) is a prevalent condition in childhood that might be associated with increased healthcare costs, including hospital admission. Retrospective studies have implicated psychosocial factors as being of prognostic relevance, but these are unable to offer greater insight into the relationship given the complex nature of this chronic illness.

Study design: Systematic review of literature.

Setting: MEDLINE, EMBASE, and PsycINFO.

Synopsis: The databases were searched through June 2008 for articles that focused on children 4 to 18 years of age; used criteria for CAP as defined by Apley and Naish, von Baeyer and Walker, or the Rome Committee; and demonstrated prospective determination of outcomes. Eight studies were included in the final review, and the levels of evidence were graded based on assessment of risk for bias.

Female sex and severity of baseline abdominal pain did not predict persistence of CAP, although there was conflicting evidence as to the role of psychological factors. There was moderate evidence that having a parent with functional gastrointestinal (GI) symptoms predicted persistence of CAP in children, and there was weak evidence that having parents who eschew psychological factors in favor of searching for an organic explanation predicted persistence of CAP.

Due to a limited number of studies or conflicting associations, conclusions regarding the following factors could not be drawn: age, educational level, duration of CAP, associated symptoms, socioeconomic status, and history of two or more surgical operations.

In addition, the univariate analysis used by the included studies might not be appropriate for such a multifactorially complex disease. Nevertheless, this study challenges the conventional wisdom that psychological factors predict persistence of pain and should remind clinicians to assess for parental functional GI disorders in this patient population.

Bottom line: Parental history of functional GI disorders predicts persistence of CAP in children.

Citation: Gieteling MJ, Bierma-Zeinstra SM, van Leeuwen Y, Passchier J, Berger MY. Prognostic factors for persistence of chronic abdominal pain in children. J Pediatr Gastroenterol Nutr. 2011;52(2):154-161.