Recent Research

Outpatient vs. inpatient PE treatment, handheld echocardiography and length of stay, and more.


Outpatient PE treatment not inferior to inpatient management for low-risk patients

Outpatient care for pulmonary embolism (PE) can safely and effectively be used in place of inpatient care in patients with low risk of death, a study found.

Researchers conducted an open-label, randomized, non-inferiority trial at 19 emergency departments in Switzerland, France, Belgium and the U.S. between February 2007 and June 2010. The 344 subjects were at least 18 years of age with acute, symptomatic and objectively verified PE of severity index risk class I or II. They were randomized to initial outpatient treatment (discharged from the hospital ≤24 hours after randomization) or inpatient treatment with subcutaneous enoxaparin (≥5 days) followed by oral anticoagulation (≥90 days). The primary outcome was symptomatic, recurrent venous thromboembolism (VTE) within 90 days. Safety outcomes included major bleeding within 14 or 90 days and mortality within 90 days. Results appeared in the July 2 The Lancet.

In the primary analysis, one of 171 outpatients, a woman with cervical cancer who initially had bilateral segmental PE, developed recurrent VTE within 90 days, compared with none of the 168 inpatients (95% upper confidence limit [UCL], 2.7%; P=0.011). One patient in each treatment group died within 90 days (UCL, 2.1%; P=0.005), and two of 171 outpatients and no inpatients had major bleeding within 14 days (UCL, 3.6%; P=0.031). By 90 days, three outpatients but no inpatients had developed major bleeding (UCL, 4.5%; P=0.086). The major bleeds were intramuscular hematomas occurring on days 3 and 13. One additional outpatient developed major bleeding within 90 days (menometrorrhagia on day 50).

Outpatients had a mean of 3.4 fewer days of initial hospital stay, with similar hospital readmission rates, emergency department visits, and outpatient visits to doctors compared to the inpatient group. Outpatients had 2.6 more days of treatment with low-molecular-weight heparin than inpatients. Researchers wrote, “Although we showed non-inferiority for outpatient treatment with respect to major bleeding at 14 days, we did not achieve non-inferiority at 90 days because of an additional bleeding episode that occurred 50 days after randomization. However, given this time latency, it is unlikely that this bleeding event was related to randomization to outpatient treatment.”

Handheld echocardiography doesn't reduce length of stay vs. usual care

Length of stay on hospitalist service didn't differ for inpatients managed with handheld echocardiography versus usual care, a study found.

In a trial at a Chicago teaching hospital between July 2008 and March 2009, researchers randomly assigned 453 general medicine adult inpatients who were referred for standard echocardiography to usual care or care guided by handheld echocardiography. The hospitalists providing care were board-certified general internists who were not formally trained in echocardiography but had completed a 27-hour handheld echocardiography training program designed by the research team.

The handheld protocol entailed hospitalists recording the seven best-quality images up to six seconds in duration and immediately interpreting them via definitions that were meant to detect threshold levels of severity most pertinent to hospitalists—such as the potential presence of severe mitral regurgitation. Length of stay on the service of the referring hospitalist was the primary outcome, rather than total hospital length of stay, in order to reflect how handheld devices affected hospitalists' management decisions, including transfers to other services. Results were published in the AugustAmerican Journal of Medicine.

Patients cared for by handheld echocardiography had a 1.7% shorter length of stay (46.1 hours) on service than those with usual care (46.9 hours)—a difference that was not statistically significant. In post-hoc, subgroup analysis, patients referred for heart failure on whom handheld echocardiography was used had a shorter length of stay than usual-care patients with heart failure (P=0.0008). Among patients randomized to the handheld group, hospitalists performed echocardiography at a median of 16 hours after admission to their service and took a median of 10 minutes to do it. Among the 210 patients who underwent both handheld and usual echocardiography, hospitalists responded to handheld echocardiography by changing their management plans in 37% of patients. There was no significant difference between groups in the amount of time it took to receive echocardiography, or in 30-day readmissions or return visits to the ED.

The reduced length of stay with handheld echocardiography was neither statistically significant nor clinically meaningful, making the primary outcome a “true negative” result, the authors said. “Despite the favorable diagnostic performance of hand-carried echocardiography, hospitalists awaited standard echocardiography results before planning hospital discharge for most participants,” the authors noted—thus, not using the handheld version as a definitive test. This may be a result of study protocol limitations for using the handheld device, general limitations of the device with certain conditions, or hospitalists' concerns about liability, they noted. Evidence-based support for using handheld echocardiography with general inpatients is still lacking, they concluded. Future studies should ensure the handheld procedure is done immediately after admission, and should focus on groups the procedure is most likely to affect, like heart failure patients, they said.

Nesiritide not effective for acute heart failure

Nesiritide is not an effective treatment for acute decompensated heart failure, according to a recent controlled trial.

The study randomized more than 7,000 patients who were hospitalized with acute heart failure to receive either nesiritide or placebo for 24 to 168 hours in addition to standard care. End points were change in dyspnea and rehospitalization or death within 30 days. The results were published in the July 7 New England Journal of Medicine.

Patients taking nesiritide reported improved dyspnea at both six hours (44.5% vs. 42.1%, P=0.03) and 24 hours (68.2% vs. 66.1%, P=0.007), but these differences did not meet the study's prespecified minimum for statistical significance. There were no significant differences between the groups in rates of rehospitalization or death (9.4% on nesiritide vs. 10.1% on placebo, P=0.31), death at 30 days (3.6% on nesiritide vs. 4.0% on placebo) or worsening renal function (defined as a >25% decrease in estimated glomerular filtration rate, 31.4% on nesiritide vs. 29.5% on placebo, P=0.11). However, nesiritide did significantly increase rates of hypotension in patients who took it: 26.6% versus 15.3% (P<0.001).

Based on the results, study authors concluded that “nesiritide cannot be recommended for routine use in the broad population of patients with acute heart failure.” They noted that their findings are consistent with the results of the trial that led to FDA approval of nesiritide a decade ago. That study included about 500 patients and showed that nesiritide had a significant effect on dyspnea similar to that of nitroglycerin at three hours, but no significant effect at 24 hours.

The current study contradicted widespread assumptions about both the effectiveness of nesiritide in relieving dyspnea and its potential to harm patients by reducing survival and renal function, the authors noted. The results show the need for rigorously designed, adequately powered trials of new medications, the authors said. An accompanying editorial expanded on this conclusion, saying that “the FDA should be provided the full regulatory authority to require definitive trials and withdraw a drug if the sponsor does not conduct a far-reaching clinical end-point trial.”

Nonacute coronary syndrome inpatients with elevated troponin have higher mortality than ACS counterparts

Hospitalized patients with an elevated troponin level more often have a primary diagnosis that is not an acute coronary syndrome (ACS), and have poorer long-term survival and undergo fewer cardiac imaging studies than those with an ACS, according to a recent study.

Researchers identified 21,668 inpatients in the Veterans Affairs central databases in 2006 who had elevated troponin levels and were eventually discharged from the hospital. They divided patients into those diagnosed with nonacute or acute coronary syndromes, and compared use of cardiac imaging during hospitalization and within 90 days of discharge. Researchers also compared if patients were readmitted or died within one year after the initial hospitalization. Results were published in the July American Journal of Medicine.

Fifty-seven percent (n=12,400) of patients had a nonacute coronary syndrome condition, with the most common diagnostic descriptor being congestive heart failure and chronic coronary atherosclerosis. Their mortality rate was 22.8% one year after discharge, compared to 17.5% of patients with a diagnosis of ACS (odds ratio, 1.389; 95% CI, 1.298 to 1.487). Use of cardiac imaging was lower in patients with a nonacute coronary syndrome condition both during hospitalization and within 90 days of discharge (19.1% vs. 45% of ACS patients; OR, 0.564; 95% CI, 0.505 to 0.629). For patients with elevated troponin and nonacute coronary syndrome, the risk of death was approximately 2.5 times higher when no test was performed during the hospitalization (OR, 2.497; 95% CI, 2.134 to 2.921) and stayed higher when no test was performed within 90 days of discharge (OR, 2.068; 95% CI, 1.675 to 2.554). Specifically, the average time from hospitalization to death for these patients was 55.6 days when no cardiac imaging was done, 140.2 days with imaging during hospitalization, and 162.7 days with imaging within 90 days of discharge.

Patients with elevated troponin levels and nonacute coronary syndrome have poor outcomes after discharge, “in part, because of no proven therapies and a lack of strategy-based studies that can target high-risk subjects before hospital discharge,” the authors concluded. Study limitations include that outcomes were not determined in patients who didn't have troponins drawn, thus results can't be generalized to all hospitalized patients, they wrote. Those with biomarkers drawn were likely at higher risk, the authors noted. There was also some variability in troponin assays at some medical centers during the study time period, they added.

DNR decisions by surrogates common, take longer

More than half of do-not-resuscitate (DNR) orders were completed by a surrogate decision maker in a recent study at one urban, public hospital.

The retrospective cohort study was based on an electronic chart review of 668 patients who got a DNR while in the hospital. The decision was made by the patient in 191 cases (28.9%), a surrogate in 389 cases (58.2%) and as a shared decision by both parties in 88 cases (13.2%). Decisions made by surrogates were typically made later in a patient's hospital course. On hospital day three, 77.6% of patient decisions, 61.9% of surrogate decisions, and 58.0% of shared decisions had been made. Perhaps in part due to this delay, surrogate decisions were also made closer to a patient's death—a median of 1.0 days before death, compared to 3.5 days for patient decisions and 4.0 days for shared decisions. Surrogate decisions were more likely to be made for patients in the ICU than those in the regular wards (62.2% vs. 39.8%, P<0.001).

Based on the results, researchers suggested that surrogate decision making is the norm rather than the exception. They recommended that research and policies relating to DNRs take these results into account and incorporate best practices in surrogate communication, rather than assuming an autonomous patient. The finding that surrogate decisions were made later may relate to previous research showing that surrogate decision making is a process requiring multiple conversations. Prior research has also found that delayed DNR decisions are associated with higher hospital costs and longer lengths of stay, the researchers noted. Later decisions also could reduce the time available to provide end-of-life care.

Although the study's generalizability is limited by its single setting, other researchers may want to consider the findings and identify the causes of the delays in surrogate decision making. Interventions should be developed to optimize the decision making process for surrogates, the study authors concluded. The study was published in the July Journal of the American Geriatrics Society.

Multi-pronged intervention reduces readmissions, lowers costs

An intervention comprising inpatient coaching, home visits and post-discharge follow-up calls reduced 30-day readmission rates by 36% in a real-world setting, researchers reported.

Researchers conducted a quasi-experimental, prospective cohort study from January 2009 through June 2010 among a consecutive convenience sample of fee-for-service Medicare patients at six Rhode Island hospitals. Patients were admitted for cardiac or respiratory conditions or related symptoms, including shortness of breath, sudden weight gain, fever, cough and chest pain. The program expanded in January 2010 to include patients with any diagnosis. The intervention included a coach completing a hospital visit, a home visit, and two follow-up telephone calls. Coaches gave inpatients booklets to record their personal health records and their communication with outpatient clinicians. After discharge, the coaches tried to complete a home visit within three days, a first call within seven to 10 days, and a second call by 30 days. In their follow-up, coaches reinforced use of the personal health records, and emphasized how to recognize signs and symptoms of worsening health before emergencies happened. The second telephone call emphasized the importance of a follow-up visit with a physician and helped the patient locate other support services.

Of 1,888 approached to participate, more than half (1,042) consented, 257 completed the home visit, 238 completed the visit plus one phone call, and 191 received the home visit and both telephone calls. Before the Care Transition Intervention program began, the average readmission rate among the six hospitals was 21.1% (range, 18.1 to 23.1%). After the program, the 30-day readmission rate for those who didn't undergo any part of the intervention was 20%, compared to 12.8% for those who did (odds ratio [OR], 0.61; 95% CI, 0.42 to 0.88). Results appeared in the July 25 Archives of Internal Medicine.

The authors noted recruitment and retention challenges evident in their approximately 55% acceptance and 75% attrition rate among patients who agreed to a home visit. Still, the intervention significantly lowered readmissions in a real-world setting, they noted.

A second study in Archives outlined similar results from a prospective study of an advanced practice nurse-led transitional care program for patients with heart failure. Researchers looked at the effect of a three-month transitional care program in which an advanced practice nurse conducted a home visit within the first 72 hours of discharge, and at least eight home visits for each patient. The follow-up emphasized the nature of the heart failure, patient and caregiver goals, general health behavior and skills, and social support. Nurses were available by telephone seven days a week, and the nurses resumed hospital visits if a readmission occurred. The transitional care program reduced adjusted 30-day readmission rates by 48% compared to the period before the study, but had little effect on length of stay or total 60-day direct costs. The intervention reduced the hospital financial contribution by an average of $227 for each Medicare patient with heart failure.

An editorialist said the successes were tempered by low participation rates for both programs, and noted that the programs save money only if their implementation costs are covered by the savings. “Reimbursements are rarely provided for preventing negative outcomes,” he noted.

Apixaban associated with more bleeding, no change in ischemic events

A phase 3 trial of apixaban, a new factor Xa inhibitor, was halted after the drug was found to increase major bleeding in patients taking it after acute coronary syndrome.

The double-blind, randomized controlled trial included more than 7,000 patients that had recent acute coronary syndrome and at least two risk factors for recurrent ischemic events and were receiving standard antiplatelet therapy. The participants were randomized to receive either apixaban, 5 mg twice daily, or placebo, and the primary end point was cardiovascular death, myocardial infarction or ischemic stroke. However, after a median follow-up of 241 days, the trial was stopped due to differences between groups in the primary safety outcome—major bleeding. Such bleeding occurred in 1.3% (46 people) of those who received apixaban, compared to 0.5% (18 people) of those on placebo (hazard ratio, 2.59; P=0.001). There were also more intracranial and fatal bleeds in the active group.

No significant difference was found between the groups in rates of cardiovascular death, myocardial infarction or ischemic stroke. The combined outcome occurred in 7.5% of apixaban users compared to 7.9% of placebo users (P=0.51). The study authors noted that phase 2 trials of apixaban and another factor Xa inhibitor, rivaroxaban, had found increases in bleeding but also trends toward reductions in ischemic events. The early discontinuation of this trial—at a lower number of ischemic events than expected—leaves some uncertainty about whether a benefit to the drug could have been found in this study, they said.

This trial included high-risk patients, many with diabetes, heart failure or renal insufficiency, but no differences were seen among the subgroups in the study, such as those receiving aspirin plus clopidogrel versus aspirin alone or those who had or didn't have revascularization. Further investigation is required to determine if the results could also be different in other patient populations, the study authors suggested.

Still, the results of this trial, combined with those of other interventions, such as vitamin K antagonists, “raise doubt about whether meaningful incremental efficacy can be achieved with an acceptable risk of bleeding by combining a long-term oral anticoagulant with both aspirin and a P2Y12-receptor antagonist in patients with coronary disease,” the authors concluded. The study was funded by Bristol-Myers Squibb and Pfizer and appeared in the Aug. 25 New England Journal of Medicine.

Prior antibiotic use associated with higher mortality in gram-negative sepsis inpatients

Recent antibiotic use is associated with greater hospital mortality in patients with severe sepsis or septic shock attributed to gram-negative bacteremia, a study found.

In a retrospective cohort study, researchers identified 754 consecutive patients at a single academic hospital with gram-negative bacteremia complicated by severe sepsis or septic shock. Patients with gram-negative sepsis were identified by the presence of a blood culture positive only for gram-negative bacteria, combined with ICD-9 codes indicating acute organ dysfunction. Data between January 2002 and December 2007 were examined. Electronic inpatient and outpatient medical records for all patients in the multi-hospital and clinic healthcare system were reviewed to determine antibiotic exposure within the previous 90 days. The primary outcome was all-cause hospital mortality; secondary outcomes included antimicrobial susceptibility and proportion of patients receiving inappropriate initial antibiotic therapy. Results were published in the August Critical Care Medicine.

Seventy-nine percent of the patients studied were in the ICU, and the mean length of hospitalization was 10.2 ± 14.4 days at the time severe sepsis or septic shock occurred. The most common organisms isolated from blood cultures were Escherichia coli (30.8%), Klebsiella pneumonia (23.2%) and Pseudomonas aeruginosa (17.6%). Anti-pseudomonal agents were prescribed for most patients, as follows: cefepime (54.2%), ciprofloxacin (20.4%), aminoglycosides (18.2%), piperacillin-tazobactam (15.5%), and carbapenems (12.1%). Forty-one percent of patients had previous antimicrobial exposure in the past 90 days; of these, 50% had exposure to cefepime, 32.6% to ciprofloxacin, 28.7% to imipenem or meropenem, 19% to piperacillin-tazobactam, and 14.5% to aminoglycosides.

Patients with prior exposure were significantly more likely to have higher rates of inappropriate initial therapy (45.5% vs. 21.2%, P<0.001) and hospital mortality (51.3% vs. 34%, P<0.001) compared to those without prior antibiotic exposure. They had higher rates of resistance to cefepime (29% vs. 7%), piperacillin-tazobactam (31.9% vs. 11.5%), carbapenems (20% vs. 2.5%), ciprofloxacin (39.7% vs. 17.6%) and gentamicin (26.1% vs. 7.9%), with P<0.001 for all comparisons. Aside from recent antibiotic exposure, variables independently associated with hospital mortality included use of vasopressors, infection resulting from P. aeruginosa, inappropriate initial antibiotic therapy, increasing Acute Physiology and Chronic Health Evaluation II scores, and the number of acquired organ failures.

The reason for the association between mortality and previous antibiotic exposure is likely the greater degree of antimicrobial resistance in the causative pathogen(s) of patients who had received antibiotics in the previous 90 days, the authors wrote. Clinicians should look for and identify the presence of prior exposure to antibiotics when prescribing antibiotics to patients with severe sepsis or septic shock, they wrote, and institutions should formalize an approach to identify such prior exposure. In situations where recent antibiotic exposure is likely but details are unknown, combination empiric therapy directed against gram-negative bacteria is reasonable, they concluded.