Frank Wharton, MD

Background: Hospital readmission has received national attention as an indicator of poor healthcare quality and unnecessary costs. While some studies suggest that some readmissions are preventable, the exact number is unknown. Understanding preventability and the views of front-line clinicians might help hospitalists balance multifactorial compromise between throughput and length of stay.

Study design: Retrospective cohort study.

Setting: Two community tertiary-care hospitals and two suburban hospitals staffed by a single hospitalist group in Portland, Ore.

Synopsis: A group of 17 hospitalists analyzed the inpatient and outpatient charts of 300 consecutive patients readmitted within 21 days of discharge using a structured data collection tool to record patient characteristics, process measures, and perceived preventability of the readmission. Patients were either discharged by internal-medicine hospitalists or had an internal-medicine consultation during their initial stay.

Characteristics of readmitted patients and initial hospital stays were similar to those previously reported in other studies. More than 60% of readmissions were deemed preventable (15%) or possibly preventable (46%). As prevention strategies for these readmissions, hospitalists most frequently suggested interventions under their control, such as longer initial hospital stay (23%). Other potential interventions focused on system approaches, including outpatient appointments, case management, palliative care, and home health.

In most cases (96%), the reviewer was not the discharging hospitalist. Only six hospitalists performed 83% of the reviews, and inter-reviewer reliability was not assessed. Other limitations included confounding biases, such as timing of review, source hospital reviewed, reviewer optimism, and previous primary-care experience of reviewer.

Bottom line: Prevention of hospital readmissions will require a balance between increased length of stay and system-based team approaches beyond the direct control of hospitalists.

Citation: Koekkoek D, Bayley KB, Brown A, Rustvold DL. Hospitalists assess the causes of early hospital readmissions. J Hosp Med. 2011;6:383-388.

Patients with Complex, Severe Illnesses and Persistent Physiologic Abnormalities Have Higher Risk of Intensive-Care-Unit Readmission

Clinical question: Which patient characteristics increase the risk for intensive-care-unit (ICU) readmission?

Background: Patients are often discharged from the ICU based on subjective criteria, workload, and bed demand, making ICU readmission a tempting quality indicator. Studies have examined institutional characteristics leading to ICU readmission, but few have shown how patient case mix affects longer lengths of ICU and hospital stays and higher in-hospital mortality.

Study design: Retrospective observational cohort study using a large, multi-institutional U.S. database.

Setting: Computerized data collection and analysis system from 97 intensive- and cardiac-care units at 35 hospitals in the United States.

Synopsis: Patient-level characteristics and outcomes of 229,375 initial ICU admissions from 2001 to 2009 were compared for patients with and without ICU readmission using the Acute Physiology and Chronic Health Evaluation (APACHE) IV database (Cerner Corporation, Kansas City, Mo.). Multivariable logistic regression analysis identified potential characteristics increasing the risk for ICU readmission. Case-mix adjusted outcomes for mortality and length of stay were calculated.

A total of 13,980 (6.1%) patients were readmitted to the ICUs. Characteristics associated with ICU readmission were similar to those previously reported. In addition to illness severity at initial admission, readmission was associated with complications, poor response to therapy, or persistent physiologic abnormalities at ICU discharge. Patients who were readmitted to the ICU had a risk-adjusted, fourfold greater probability of in-hospital mortality and a 2.5-fold increase in length of hospital stay.

Data were obtained solely from hospitals with an APACHE system, and included only one hospital from the Northeast. Not all ICUs from a single-study hospital were included. DNR orders were not available in the database for analysis.

Bottom line: Persistent physiologic derangements at ICU discharge are as influential on ICU readmission and poor outcomes as other known risk factors, potentially warranting slightly longer ICU stays for these specific patients.

Citation: Kramer AA, Higgins TL, Zimmerman JE. Intensive care unit readmissions in U.S. hospitals: patient characteristics, risk factors, and outcomes. Crit Care Med. 2012;40:3-10.

Clopidogrel Might Not Worsen Bleeding Complications During Surgery

Clinical question: Does clopidogrel cause bleeding complications during vascular surgery?

Background: There is no consensus and few large trials to decide if preoperative clopidogrel use causes bleeding complications in vascular surgeries.

Study design: Prospective.

Setting: New England academic and community centers.

Synopsis: The authors enrolled 10,406 patients from 15 academic and community centers in New England. These patients underwent carotid endarterectomy, lower-extremity bypass, and abdominal aortic aneurysm repair. Ruptured aortic aneurysms were excluded. The use of aspirin, clopidogrel, both, or no anti-platelet therapy within 48 hours before surgery was recorded. The outcomes measured were postoperative bleeding requiring reoperation and the need for packed red blood cell transfusions.

Clopidogrel therapy (n=229) compared with no anti-platelet therapy (n=2,010) did well when measured by reoperation (clopidogrel 0.9%, none 1.5%, P=0.74), incidence of transfusion (clopidogrel 0%, none 18%, P=0.1), and volume of transfusion (clopidogrel 0 units, none 0.7 units, P=0.1). However, the significance, especially for reoperation, is not impressive.

The power of the study could be improved with more clopidogrel users. Another limitation is that the aspirin and clopidogrel platelet effect lasts longer than the 48-hour cutoff. As there were more bleeding complications in the no-anti-platelet group, there is a concern for bias in how the patients were selected and treated.

Bottom line: Clopidogrel might be safe to continue in vascular surgeries, but larger and more valid studies are needed.

Citation: Stone DH, Goodney PP, Schanzer A, et al. Clopidogrel is not associated with major bleeding complications during peripheral arterial surgery. J Vasc Surg. 2011;54:779-784.

Tissue-Defined TIAs Have Better Prognostic Benefits Than Time-Defined TIAs

Citation: Is risk stratification for recurrent stroke better when brain imaging is performed after a transient ischemic attack (TIA)?

Background: The presence of a TIA is a known risk factor for recurrent stroke. There is debate in changing the definition of TIA from time-based to tissue-based. However, it is not known if this will improve management.

Study design: Multicenter observation cohort.

Setting: Twelve independent international research centers.

Synopsis: The current definition of a TIA is a neurological deficit that resolves within 24 hours. To determine the urgency of an admission and work-up, the ABCD2 score (age, blood pressure, clinical symptoms, duration, and diabetes) is often used as a prognostic tool for recurrent stroke. The authors enrolled 4,574 patients with the traditional diagnosis of TIA. In addition to calculating their ABCD2 score, they were then classified as tissue-positive (infarction) or tissue-negative per MRI or CT. At both seven and 90 days after TIA, both imaging modalities when combined with ABCD2 were able to identify risk for recurrent stroke better than if they were used alone.

Limitations included the images being interpreted by their individual healthcare centers and that the stroke centers had lower recurrent rates of stroke compared with community centers. If only a CT can be used, it should be noted that the sensitivity is lower when compared with an MRI in this study.

Bottom line: This categorization of tissue- or non-tissue-positive TIAs improved the prognostic information provided by the ABCD2 score and might improve management decisions.

Citation: Giles MF, Albers GW, Amarenco P, et al. Early stroke risk and ABCD2 score performance in tissue- vs time-defined TIA: a multicenter study. Neurology. 2011;77:1222-1228.

ETT Comparable to MPI as Initial Diagnostic Test for Women Suspected of CAD

Clinical question: Is the exercise treadmill test (ETT) equivalent to myocardial perfusion imaging (MPI) as an initial diagnostic test for low-risk women capable of exercising with suspected coronary artery disease (CAD)?

Background: To date, the evidence base for diagnostic testing in women with suspected CAD has been limited, with no randomized trials to guide appropriate clinical decision-making in the choice of noninvasive testing strategies. The aim of this trial was to provide comparative-effectiveness data for women capable of performing exercise testing with regard to whether the addition of imaging to standard ETT provided incremental clinical benefit over index ETT alone.

Study design: Prospective randomized.

Settings: Forty-three cardiology practices across the U.S.

Synopsis: A total of 824 women 40 years of age or older with intermediate pretest CAD likelihood were randomized to ETT or MPI. All the women included in the study were symptomatic with suspected CAD, had an interpretable ECG, and were >5 on DASI (Duke Activity Status Index). A total of 17 primary end points were confirmed, including three nonfatal myocardial infarctions, one heart failure hospitalization, 12 acute coronary syndrome hospitalizations, and only one sudden cardiac death reported in 772 women.

At two years, major adverse cardiovascular events (MACE)-free survival was identical (98%) for women randomized to the ETT or exercise MPI arm (P=0.59). The observed two-year MACE rate was 1.7% for ETT and 2.3% for exercise MPI. For secondary end points, the overall rate of hospitalization for chest pain was 3%. By randomization, the two-year rate of hospitalization for chest pain symptoms was 3% for women in the ETT arm and 4% for those in the exercise MPI arm (P=0.39). An additional six women died from non-cardiac causes (ETT arm, 0.5%; exercise MPI arm, 1%; P=0.39).

Bottom line: ETT is of comparable diagnostic efficacy to MPI as an initial diagnostic test for women suspected of CAD, capable of exercising, based on two-year outcome of cardiac death, nonfatal myocardial infarction, or hospital admission for acute coronary syndrome or heart failure.

Citation: Shaw LJ, Mieres JH, Hendel RH, et al. Comparative effectiveness of exercise electrocardiography with or without myocardial perfusion single photon emission computed tomography in women with suspected coronary artery disease: results from the What Is the Optimal Method for Ischemia Evaluation in Women (WOMEN) trial. Circulation. 2011;124:1239-1249.

Controlled Forced Diuresis Is More Effective in High-Risk Patients for CI-AKI Prevention

Clinical question: Is a system of hydration that creates high urine output while maintaining fluid balance better than routine hydration in high-risk patients for preventing contrast-induced acute kidney injury (CI-AKI)?

Background: Hydration with sodium bicarbonate solution and administration of N-acetylcysteine (NAC) help prevent CI-AKI in low- to medium-risk patients. A recent study (PRINCE) suggests that increasing the urine flow rate (≥150 mL/hr) reduces the toxic effect of contrast, but this regimen requires forced diuresis with high doses of furosemide that may result in further renal damage due to negative fluid balance. The RenalGuard system (PLC Medical Systems Inc., Franklin, Mass.) can simultaneously achieve high urine output and maintain fluid balance.

Study design: Randomized, investigator-driven, open-label study comparing two different hydration strategies.

Setting: Multicenter trial in Italy involving patients with chronic kidney disease scheduled for angiography who met the inclusion and exclusion criteria for high risk of developing CI-AKI.

Synopsis: The study included 292 patients randomized to receive sodium bicarbonate solution (control group) versus hydration with saline controlled by the RenalGuard system with furosemide. Both groups received NAC, though at different doses and routes of administration. CI-AKI occurred in 20.5% (30 of 146) in the control group versus 11% (16 of 146) in the RenalGuard group.

Bottom line: Controlled forced diuresis to achieve high urine flow rate is more effective than conventional hydration with sodium bicarbonate solution in high-risk patients for preventing CI-AKI. Larger studies are needed to confirm and define the role of the RenalGuard system.

Citation: Briguori C, Visconti G, Focaccio A, et al. Renal insufficiency after contrast media administration trial II (REMEDIAL II). Circulation. 2011;124:1260-1269.

Severe Asymptomatic Carotid Artery Stenosis Does Not Increase Stroke or Mortality Risks after CAB Surgery

Clinical question: What is the risk for stroke in patients with asymptomatic carotid artery stenosis (CAS) who are undergoing coronary artery bypass grafting (CABG)?

Background: Stroke occurs as a complication of CABG in approximately 2% of patients. The vast majority occur within the first 24 hours of surgery and are associated with a high mortality rate.

Study design: Retrospective cohort.

Setting: A single institution in Washington, D.C.

Synopsis: Data were collected on 878 consecutive patients who had undergone carotid ultrasound before CABG over a six-year period. Patients with severe CAS of >75% (n=117) were compared with those with <75% stenosis (n=761) to assess the rates of in-hospital stroke and mortality.

Patients with severe CAS had similar rates of stroke compared with those without severe CAS (3.4% vs. 3.6%). Additionally, there was no difference in the incidence of in-hospital complications or mortality between the two groups.

Bottom line: The cause of stroke after CABG is complex and multifactorial, but severe CAS alone appears to not be the biggest risk factor for stroke in patients undergoing CABG.

Citation: Mahmoudi M, Hill PC, Xue Z, et al. Patients with severe asymptomatic carotid artery stenosis do not have a higher risk of stroke and mortality after coronary artery bypass surgery. Stroke. 2011;42:2801-2805.

Cardiac Arrest Survivors Have Long-Term Memory Deficits

Clinical question: Are there any lasting cognitive deficits in patients surviving out-of-hospital cardiac arrest (OHCA) from ventricular fibrillation (VF)?

Background: Although advances have been made in the rapid intervention of OHCAs, there are no population-based, age-adjusted studies of the cognitive and neurologic outcomes of long-term survivors when the “chain of survival” is used (rapid emergency services, bystander CPR, early defibrillation, and advanced care).

Study design: Prospective population-based, age-adjusted study.

Setting: Single hospital in Olmsted County, Minn.

Synopsis: Of 332 OHCA patients, 47 survivors were enrolled for neurologic and neuropsychological testing at least six months after a near-death experience (median time since arrest, 7.8 years). Neurologic examination did not reveal any focal deficits related to the event, but long-term survivors had lower scores on measures of long-term memory and learning efficiency (P=0.001). Nearly all survivors were functionally independent at the time of testing. Interestingly, there was no correlation between prolonged call-to-shock time and cognitive ability, suggesting that there is a possibility of positive neurologic outcomes with a call-to-shock time as late as 10 minutes in OHCA. Older age was also not identified as a negative prognostic factor.

A limitation of the study was the small sample size, though larger populations of survivors are difficult to find. The precise meaning of “cognitive impairment” was also controversial and should be better defined for any future studies.

Bottom line: Long-term survivors of OHCA from VF have long-term memory impairment compared with the normal population at the same age and education level.

Citation: Mateen FJ, Josephs KA, Trenerry MR, et al. Long-term cognitive outcomes following out-of-hospital cardiac arrest. Neurology. 2011;77:1438-1445.

Clinical Shorts

Risk of esophageal adenocarcinoma due to Barrett’s esophagus in the absence of dysplasia is minor

Danish population cohort study shows annual risk for esophageal adenocarcinoma for patients with Barrett’s esophagus, with or without low-grade dysplasia, to be four to five times lower than previously reported.

Citation: Hvid-Jensen F, Pedersen L, Drewes AM, Sørensen HT, Funch-Jensen P. Incidence of adenocarcinoma among patients with Barrett’s esophagus. N Engl J Med. 2011;365:1375-1383.

Pneumonia after stroke increases 30-day and ONE-year mortality

In this cohort, the rate of stroke-associated pneumonia was 7.1% with increased 30-day and one-year mortality. Organized stroke care did not affect this rate but reduced 30-day mortality.

Citation: Finlayson O, Kapral M, Hall R, et al. Risk factors, inpatient care, and outcomes of pneumonia after ischemia stroke. Neurology. 2011;77:1338-1345.

Therapeutic hypothermia for cardiac arrest does not affect time to awakening

A retrospective review at a large academic institution showed that cardiac arrest survivors enrolled in therapeutic hypothermia protocols awoke within three days of the arrest, just like nonhypothermia patients.

Citation: Fugate JE, Wijdicks EFM, White R, Rabinstein AA. Does therapeutic hypothermia affect time to awakening in cardiac arrest survivors? Neurology. 2011;77:1346-1350.

Variability in changes in forced expiratory volume in 1 second over time in COPD

The rate of change in FEV(1) among patients with COPD is highly variable, with increased rates of decline among current smokers, patients with bronchodilator reversibility, and patients with emphysema.

Citation: Vestbo J, Edwards LD, Scanlon PD, et al. Changes in forced expiratory volume in 1 second over time in COPD. N Engl J Med. 2011;365:1184-1192.

Baclofen reduces benzodiazepine requirements in alcohol withdrawal syndrome

A small prospective, randomized, double-blind, placebo-controlled trial did demonstrate decreased requirements for lorazepam, but no improvement in symptoms, by using scheduled low doses of baclofen during alcohol withdrawal.

Citation: Lyon JE, Khan RA, Gessert CE, Larson PL, Renier CM. Treating alcohol withdrawal with oral baclofen: a randomized, double-blind, placebo-controlled trial. J Hosp Med. 2011;6:474-479.

ESRD patients have a higher long-term risk of peptic ulcer rebleeding

This prospective cohort study showed that the incidence of rebleeding is highest in the first year in end-stage renal disease patients when compared with controls. Consideration should be given to long-term use of proton pump inhibitors or other gastroprotective agents in this at-risk population.

Citation: Wu C, Wu M, Kuo K, et al. Long-term peptic ulcer rebleeding risk estimation in patients undergoing haemodialysis: a 10-year nationwide cohort study. Gut. 2011;60:1038-1042.

Issue

The Hospitalist - 2012(01)

Publications

The Hospitalist

Read more about In the Literature: Physician Reviews of HM-Related Research

Sections

Author(s)

Steven Deitelzweig, MD, SFHM

Indranill Basu Ray, MD

William Carter, MD

David Lee, MD, MBA

Renee Meadows, MD, FACP, SFHM Srinivas Vuppala, MD

Frank Wharton, MD

Author(s)

Steven Deitelzweig, MD, SFHM

Indranill Basu Ray, MD

William Carter, MD

David Lee, MD, MBA

Renee Meadows, MD, FACP, SFHM Srinivas Vuppala, MD

Frank Wharton, MD

In This Edition

Literature At A Glance

A guide to this month’s studies

Hospitalist views on readmission prevention

Characteristics of hospital ICU readmission

Effect of clopidogrel on bleeding outcomes in vascular surgery

Time-versus tissue-based diagnosis of TIA

ETT versus ETT with imaging for the diagnosis of CAD in women

Effect of high urine output with adequate hydration on contrast-induced nephropathy

Stroke rate in CABG patients with severe carotid artery stenosis

Effect of cardiac arrest on long-term cognition