Recent Research

PE hospitalizations, vena cava filter placements both on the rise

More Medicare patients are being hospitalized with pulmonary embolism (PE) and more are receiving inferior vena cava filters, according to an analysis of recent trends.

The study included all 556,658 Medicare beneficiaries who had a principal discharge diagnosis of PE between 1999 and 2010. Of these patients, 94,427 had a filter placed. Over the studied time period, the annual number of PE hospitalizations increased (from 31,746 in 1999 to 54,392 in 2010), as did the number of hospitalizations with filter placement (from 5,003 in 1999 to 8,928 in 2010). Results appeared in the March 8 Journal of the American College of Cardiology.

Comparing the 71% increase in PE hospitalizations with the 78% increase in filters revealed a statistically insignificant increase in the rate of filter placement per 1,000 PE hospitalizations (from 157.6 to 164.1; P=0.11), the authors found. These results were similar across demographic subgroups, although rates of filter placement were higher among black patients and patients ages 85 years and over. The study also found wide geographic variation, with the highest rate of filter use in the South Atlantic region and the lowest in the Mountain region. The researchers also looked at mortality in the studied PE patients. After adjustment, the whole patient population's 30-day mortality declined from 12.7% to 9.0% and 1-year mortality dropped from 26.3% to 22.4% (P<0.001 for both time periods). The declines were seen both among patients who did and did not receive filters.

There are a number of potential explanations for the results, study authors said. Mortality might have declined because filters were effective in preventing fatal PEs or because improvements in filter-placement technique and equipment have reduced procedure-related mortality. However, mortality also decreased among patients who didn't have filters. Another explanation is that more sensitive diagnostic tools increased diagnosis of PE, particularly in less sick patients, leading the entire cohort to have lower mortality. Filters may have been used more frequently in healthier patients, reducing mortality in the filter-receiving group. Or these factors may have combined to produce the observed results, the authors said.

“Collectively, our results suggest that more permissive use of technology has occurred over time in the setting of persistent controversy for net benefit,” they concluded. They called for future research to identify which patient subgroups would benefit most from filter placement. An accompanying editorial agreed, noting that the vast majority of filters are placed in stable PE patients, when those who are unstable may be most likely to benefit.

Preadmission glucocorticoids associated with 30-day poststroke mortality

Patients taking glucocorticoids before hospital admission for stroke may be at higher mortality risk, according to a recent study.

Researchers in Denmark performed a population-based cohort study using medical registry data to examine whether an association existed between preadmission glucocorticoid use and 30-day mortality in patients with a first inpatient diagnosis of stroke. Glucocorticoid use was categorized as current use (if the last prescription was filled ≤90 days before admission), former use, and nonuse, and current use was additionally classified as new or long-term. The study's primary outcome was 30-day mortality rate ratio (RR). The results appeared in the March Stroke.

Overall, 100,042 patients had a first-time stroke between 2004 and 2012, and of these, 83,735 had ischemic stroke, 11,779 had intracerebral hemorrhage (ICH), and 4,528 had subarachnoid hemorrhage (SAH). Median age was 74 years for those with ischemic stroke, 72 years for those with ICH, and 58 years for those with SAH. Glucocorticoids were currently used by 4,645 patients (4.6%), while 1,535 (1.5%) were former users and 93,682 (93.8%) were categorized as nonusers.

The researchers found that absolute mortality risk for ischemic stroke, ICH, and SAH was higher among current glucocorticoid users compared with nonusers (19.5% vs. 10.2%, 46.5% vs. 34.4%, and 35.0% vs. 23.2%, respectively). The adjusted 30-day mortality RR was also higher in current users than in nonusers for ischemic stroke (RR, 1.58; 95% CI, 1.46 to 1.71), with a larger effect seen among new users (RR, 1.80; 95% CI, 1.62 to 1.99). The effect on adjusted 30-day mortality RR was more modest among new users versus nonusers for ICH (RR, 1.26; 95% CI, 1.09 to 1.45) and SAH (RR, 1.40; 95% CI, 1.01 to 1.93). No notable association was seen between mortality and former glucocorticoid use.

The researchers noted that their results could have been affected by residual and unmeasured cofounding and variations in disease severity. In addition, they said, their study was observational and detailed clinical data were not available. However, they concluded that their results indicate an association between current use of systemic glucocorticoids and higher short-term mortality risk after first hospitalization for ischemic stroke, SAH, and ICH.

Despite the study's limitations, they wrote, “We consider a causal relation possible because the association followed a dose gradient and persisted after adjustment for [Charlson Comorbidity Index] and diseases for which glucocorticoids are prescribed. Hence, our observations merit clinical attention when prescribing glucocorticoids to patients at risk of stroke.”

Troponin associated with readmission in heart failure patients

In hospitalized heart failure patients, elevated troponin levels are associated with increased length of stay and short- and long-term mortality, a recent meta-analysis found.

The systematic review and meta-analysis included 26 studies of patients hospitalized with acute decompensated heart failure. Detectable or elevated cardiac troponin was associated with longer length of stay, with an odds ratio (OR) of 1.05 (95% CI, 1.01 to 1.10). These patients also had an increased risk of inpatient mortality (OR, 2.57; 95% CI, 2.27 to 2.91). A composite outcome of mortality and major adverse events during hospitalization was also associated with elevated troponin (OR, 1.33; 95% CI, 1.03 to 1.71).

The risk continued postdischarge, with patients who had detectable or elevated troponin during hospitalization having more than double the risk of mortality and readmission in the next 6 months (mortality OR, 2.11; 95% CI, 1.43 to 3.12; combined mortality and readmission OR, 2.81; 95% CI, 1.60 to 4.92). The risk continued in the year after discharge (mortality OR, 2.21; 95% CI, 1.46 to 3.35; combined OR, 2.30; 95% CI, 1.78 to 2.99) and more than 1 year after discharge (mortality OR, 3.69; 95% CI, 2.64 to 5.18; combined OR, 3.49; 95% CI, 2.08 to 5.84).

This review, which the authors called the most comprehensive to date on the topic, confirmed previous observations of associations between elevated troponin and negative outcomes in heart failure patients. The findings are biologically plausible, they noted, theorizing that the acute myocardial injury indicated by elevated troponin promotes ventricular remodeling and heart failure progression.

The results show that cardiac troponin could provide important prognostic information for the care of hospitalized heart failure patients, the authors said. They proposed using it in conjunction with other biomarkers recently shown to be prognostic indicators for heart failure, including brain natriuretic peptide/N-terminal pro-brain natriuretic peptide, soluble ST2, and cystatin C. The analysis was published by the Journal of Hospital Medicine on Feb. 18.

Pioglitazone may reduce risk of second stroke in nondiabetics

Pioglitazone appears to prevent cardiovascular events in patients without diabetes who have insulin resistance and cerebrovascular disease, according to a recent study.

The international, double-blind Insulin Resistance Intervention after Stroke (IRIS) trial randomized 1,939 patients to pioglitazone and 1,937 patients to placebo for a median follow-up period of 4.8 years. Patients in both groups were 63.5 years old on average. Results were published online on Feb. 17 by the New England Journal of Medicine.

Participants were at least 40 years old, were insulin-resistant, and had had a qualifying ischemic stroke or transient ischemic attack during the 6 months before randomization. Patients with diabetes were excluded from the trial.

The primary outcome of stroke or myocardial infarction (MI) occurred in 175 patients (9.0%) in the pioglitazone group and in 228 patients (11.8%) in the placebo group (P=0.007). This finding did not change after researchers adjusted for covariates. Among secondary outcomes, the rate of progression to diabetes was significantly lower in the pioglitazone group than in the placebo group (3.8% vs. 7.7%, P<0.001), and the drug had no significant effect on cognition compared to placebo.

Compared to patients taking placebo, patients in the pioglitazone group had more weight gain (mean 2.6 kg gain vs. mean 0.5 kg loss, P<0.001) and higher rates of edema (35.6% vs. 24.9%, P<0.001), shortness of breath (17.6% vs. 15.1%, P=0.03), and serious bone fractures (5.1% vs. 3.2%, P=0.003). The total incidence of death, hospitalization, and cancer did not differ significantly between groups.

The results suggest that giving pioglitazone to 100 patients similar to those in the trial for about 5 years could prevent 3 patients from experiencing stroke or MI. But during the same period, the treatment would be expected to lead to bone fractures requiring surgery or hospitalization in 2 patients.

“These findings may tempt clinicians to rush to prescribe pioglitazone, but many caveats remain,” said an accompanying editorial. Enrolled patients met strict criteria, including exclusion for heart failure, and had little neurologic impairment, so the response to pioglitazone may be different in patients with differing characteristics, the editorialist noted. “However, pioglitazone represents a potentially important therapy for the secondary prevention of vascular events in appropriately selected patients with cerebrovascular disease.”

COPD exacerbations with higher eosinophils associated with shorter length of stay

A higher eosinophil count is associated with shorter length of stay for patients with moderate COPD exacerbations, a recent study found.

The new data came from a subgroup analysis of a large randomized controlled trial. It included 243 COPD patients with a mean age of 71 years. Their median absolute eosinophil count was 100 cells/µL, and the inpatient mortality rate was 3%. Results were published in the February CHEST.

A quarter of the studied patients had a peripheral blood eosinophil count over 200 cells/µL or 2% of the total leukocyte count at admission. Those patients, all of whom were treated with oral corticosteroids, had significantly shorter mean length of stay than those below the eosinophil cutoff: 5.0 days (range, 1 to 19 days) versus 6.5 days (range, 1 to 33 days). This finding was independent of treatment prior to admission. Readmission rates were similar between those with high and low eosinophils. Eosinophilic exacerbations were usually not associated with elevated C-reactive protein.

Patients with eosinophilic exacerbations are clinically indistinguishable from other COPD exacerbation patients, the study authors noted. However, this study's results suggest that they may respond rapidly to corticosteroid treatment. This information could reduce treatment failure rates in COPD exacerbations and help identify patients who would most benefit from treatment with corticosteroids or antibiotics, the authors said.

They cautioned that the study was limited by being a subgroup analysis of a larger trial that was not powered to investigate recovery and outcomes from eosinophilic exacerbations. Therefore, the results should be confirmed in a prospective clinical trial that controls for corticosteroid prescription and predefined discharge criteria, the authors said.