Rivaroxaban reduces CV events vs. placebo in STEMI patients
Rivaroxaban reduced cardiovascular (CV) events compared to placebo but also increased nonfatal major bleeding in patients with ST-segment elevation myocardial infarction (STEMI), a study found.
Researchers examined a subgroup of STEMI patients from the ATLAS ACS 2-TIMI 51 trial, which found rivaroxaban reduced CV events in patients with an acute coronary syndrome. The randomized, double-blind trial involved 15,528 patients with symptoms suggestive of ACS and a diagnosis of STEMI, non-ST-elevation MI or unstable angina. Patients were enrolled within one to seven days of admission and stabilized, then randomized to twice-daily rivaroxaban, 2.5 mg; rivaroxaban, 5 mg; or placebo. The subgroup analysis included 7,817 patients with STEMI, most of whom were taking aspirin (98.8%) and thienopyridine (96.8%).
Rivaroxaban reduced the primary endpoint of cardiovascular death, MI or stroke compared to placebo in an intention-to-treat (ITT) analysis (ITT: 8.4% vs. 10.6%; hazard ratio [HR], 0.81; P=0.02; modified ITT [mITT]: 8.3% vs. 9.7%; HR, 0.85; P=0.09). The reduction occurred within the first 30 days (ITT and mITT: 1.7% vs. 2.3%; P=0.04) and was apparent in analyses that included events with patients who were taking aspirin and thienopyridine therapies (ITT: 7.9% vs. 11.8%; P=0.01; mITT: 7.7% vs. 10.1%; P=0.06).
Both doses of rivaroxaban exhibited similar reductions in the primary endpoint compared to placebo, but 2.5 mg of rivaroxaban reduced cardiovascular death (HR, 0.06; P=0.01) and all-cause death (HR, 0.63; P=0.01) while 5 mg of rivaroxaban didn't. Compared to placebo, rivaroxaban increased major bleeding (2.2% vs. 0.6%; P<0.001) and intracranial hemorrhage (0.6% vs. 0.1%; P=0.02), but there wasn't a significant increase in fatal bleeding (0.2% vs. 0.1%; P=0.51). The lower dose of rivaroxaban led to significantly fewer major or minor bleeding events than the higher dose, as well as fewer medical attention bleeds and fatal bleeding events. Results were published online March 7 by the Journal of the American College of Cardiology.
Previous studies have shown a cardiovascular benefit to adding 2.5 mg of rivaroxaban daily to aspirin and clopidogrel, and these studies, along with the current one, “underscore the efficacy of greater degrees of antithrombotic therapy beyond aspirin and standard dose clopidogrel following a STEMI,” the researchers wrote. They cautioned that the current study excluded individuals with an elevated risk of bleeding, so results don't apply to those patients. Overall, the lower rivaroxaban dose exhibited a better safety profile than the 5-mg dose, “thus, the addition of rivaroxaban 2.5 mg twice daily may offer an effective strategy to reduce thrombotic events in patients following a STEMI,” the researchers concluded.
Atrial fibrillation associated with cognitive impairment, dementia regardless of stroke history
Atrial fibrillation (AF) was associated with a higher risk for cognitive impairment and dementia, with or without a history of clinical stroke, a meta-analysis found.
Twenty-one studies were included in the meta-analysis, seven that looked at the association of AF with cognitive impairment or dementia after stroke and 14 that examined the association between AF and cognitive impairment or dementia in a broader population, including patients with or without a history of stroke.
The review appeared in the March 5 Annals of Internal Medicine.
In a combined analysis, AF was significantly associated with cognitive impairment (relative risk [RR], 1.40). There was significant heterogeneity among studies, mainly from variability among prospective studies and possibly due to variances in outcome measures. So researchers incorporated a random-effects model and did several sensitivity analyses and found that pooled estimates were virtually the same for prospective and cross-sectional studies. Restricting the analysis to studies of dementia, which is more reliably diagnosed than cognitive impairment, eliminated the significant heterogeneity without changing the pooled estimate substantially (RR, 1.38).
Limiting the analysis to the eight studies that defined cognitive impairment as a mini-mental state exam (MMSE) score of 24 or less or cognitive decline as a reduction in MMSE score of 3 points or more did not appreciably change the results (RR, 1.38). Investigating subtypes of dementia did not reveal any significant association between AF and Alzheimer's disease (RR, 1.22); however, the association was significant for vascular dementia (RR, 1.72). Limiting the analysis to participants without a history of stroke and studies that adjusted for stroke in multivariate analyses did not appreciably affect the primary results (RR, 1.34), nor did restricting the analysis to studies that specifically excluded patients with a history of stroke (RR, 1.37).
Seven studies reported an association between AF and cognitive impairment or dementia after stroke (RR, 2.70). Although prospective and cross-sectional studies showed overlapping risk estimates, the association was stronger in prospective studies (RR, 3.01).
The researchers wrote, “On the basis of this systematic review and meta-analysis of all available data, future research should carefully distinguish between types of dementia, and investigators should consider cognitive function as a new outcome to be assessed in interventional studies for the treatment of AF.”
Similar glycemic control found with basal bolus and basal plus regimens
For hospitalized patients with type 2 diabetes, a regimen of daily glargine supplemented with corrective doses of glulisine controlled blood glucose as well as a standard basal bolus regimen, a recent study found.
The multicenter trial included 375 medical or surgical patients with type 2 diabetes usually treated with diet, oral antidiabetic agents or insulin at a dose of 0.4 unit/kg/day or less. During hospitalization, they were randomized to a basal bolus regimen (glargine once daily and glulisine before meals), a basal plus regimen (glargine once daily and corrective doses of glulisine given by sliding scale), or sliding-scale regular insulin.
Noting that clinicians have been reluctant to follow recommendations to switch from sliding scale to basal bolus in part because of hypoglycemia risks, the authors hypothesized that a single daily dose of basal insulin might provide similar glucose control and lower hypoglycemia rates than a basal bolus regimen. The results were published by Diabetes Care on Feb. 22.
The study found that the basal bolus and basal plus regimens improved mean daily blood glucose after the first day of therapy by about the same amount. Both basal groups also had significantly lower mean daily blood glucose than the sliding-scale group. There were also significantly fewer patients with more than two consecutive glucose measurements above 240 mg/dL: 0% of the basal bolus group, 2% of the basal plus group and 19% of the sliding-scale group. Hypoglycemia (blood glucose under 70 mg/dL) occurred in 16% of basal bolus patients, 13% of basal plus patients and 3% of sliding-scale patients. The groups had similar rates of severe hypoglycemia.
Basal plus and basal bolus resulted in similar glycemic control, superior to that found with sliding scale, the study authors concluded. Based on these results, basal plus is an effective alternative to basal bolus for patients similar to those in this trial. The study was limited by its exclusion of ICU patients and those who had hepatic disease, severe hyperglycemia or a usual insulin dose of more than 0.4 unit/kg/day. Higher insulin doses or a standard basal bolus regimen might be best for those patients, the authors noted.
Guideline- and nonguideline-based treatment common in febrile neutropenia
Receipt of guideline-based treatment is common for inpatients with febrile neutropenia, but so is receipt of nonguideline-based treatment, a recent study found.
Researchers used a large administrative database to study the treatment of 25,231 patients with solid tumors who were hospitalized for febrile neutropenia (FN). Patients were treated between Jan. 1, 2000, and March 31, 2010, at more than 600 U.S. hospitals. Researchers examined treatment within 48 hours of admission to get a sense of initial decision making. They looked at use of guideline-based antibiotics as well as nonguideline-based treatments, vancomycin and granulocyte colony-stimulating factors (GCSF). They stratified patients into low- and high-risk groups and examined the effect of treatment on nonroutine hospital discharge and death. Results were published online March 4 by JAMA Internal Medicine.
Seventy-nine percent of patients received guideline-based antibiotics, while 37% received vancomycin and 63% received GCSF. Patients treated at hospitals with high volumes of FN patients, or by physicians who treat large volumes of FN patients, were more likely to receive guideline-based antibiotics (odds ratios [ORs], 1.56 and 1.19, respectively; P<0.05). The same was true of patients managed by hospitalists (OR, 1.49; P<0.05), who were also more likely to use vancomycin. In general, the use of vancomycin increased from 17% in 2000 to 55% in 2010; use of GCSF decreased from 73% to 55% in that time period. Use of guideline-based antibiotics decreased the likelihood of discharge to a nursing facility (OR, 0.77) and death (OR, 0.63) for low-risk patients but didn't make a significant difference for high-risk patients.
The fact that prompt treatment with guideline-based antibiotics reduced in-hospital death for lower-risk patients “suggests that antibiotic choice can be used as a quality metric for FN,” the authors wrote. An invited commenter, however, questioned whether guideline adherence should be used as a surrogate for quality, noting that guideline recommendations often aren't based on level 1 evidence and that it's important to individualize treatment decisions. Also, some guidelines suggest GCSFs can be considered in the care of FN patients, so it's not clear if GCSF use in this study represented high- or low-quality care, the commenter added. “Guideline adherence is perhaps a sibling of quality care,” not a twin, he wrote.
The commenter also noted that the finding of an association between hospital and physician experience with FN and treatment choices lined up with previous research, and taken together “suggests that education and clinical decision support have the potential to increase guideline adherence.” Electronic health record systems could be modified to prompt physicians about guideline-adherent care, he added, and to require them to explain their reasons for use of vancomycin and GCSFs.
Dabigatran noninferior to warfarin for preventing VTE recurrence
In patients who had a previous venous thromboembolism, dabigatran prevented recurrence about as well as warfarin and caused fewer bleeding events, according to two recent manufacturer-sponsored trials of extended treatment.
The first double-blind trial involved almost 3,000 patients with venous thromboembolism (VTE) who were thought to be at high risk of recurrence. They completed three months of initial therapy and then were randomized to warfarin or twice-daily dabigatran at a dose of 150 mg. The length of study treatment ranged from six to 36 months. Recurrent VTE occurred in 1.8% of dabigatran patients compared to 1.3% of warfarin patients, a finding that met the study's pre-specified threshold for noninferiority. The dabigatran patients had fewer major bleeds and fewer major or clinically relevant bleeds, but only the latter outcome showed a statistically significant difference.
In the second double-blind trial, about 1,300 similar patients were randomized to the same dabigatran treatment or to placebo. The dabigatran group had many fewer VTE recurrences (0.4% vs. 5.6%; hazard ratio, 0.08; P<0.001) than the placebo group, but more major (0.3% vs. 0) and major or clinically relevant (5.3% vs. 1.8%) bleeds. The results of both trials were published together in the Feb. 21 New England Journal of Medicine.
The study authors concluded that dabigatran was effective in extended treatment of VTE and carried a lower risk of major or clinically relevant bleeding than warfarin, but a higher one than placebo. They noted that the efficacy of dabigatran compared to placebo was similar to that shown by rivaroxaban and warfarin in other trials. However, the study was not able to resolve concerns about an association between dabigatran and acute coronary syndromes (ACS). In the first trial, ACS occurred in 13 (0.9%) dabigatran patients and 3 (0.2%) warfarin patients, and it occurred in one patient in each of the groups in the placebo trial.
Based on results like these, targeted anticoagulants are an appealing alternative to warfarin, according to an accompanying editorial, which noted that dabigatran has not yet been FDA-approved for extended treatment of VTE. However, the editorialist cautioned, there are still questions and concerns about these new drugs, including the current lack of an antidote and the challenge of selecting appropriate patients for treatment.
Meta-analyses results don't support hydroxyethyl starch in critically ill
Hydroxyethyl starch for fluid resuscitation fared poorly in two recent meta-analyses involving critically ill patients.
In the first study, reviewers identified 38 randomized controlled trials that compared hydroxyethyl starch (HES) to crystalloids, albumin or gelatins in 10,880 critically ill patients receiving acute volume resuscitation. The risk ratio for death among those who received HES was 1.07—a figure which included results from seven trials performed by a researcher whose later work was retracted due to scientific misconduct. When those seven trials (and their 590 patients) were excluded, HES was found to be significantly associated with death (risk ratio [RR], 1.09) among the remaining 10,290 patients, as well as with increased renal failure among 8,725 patients (RR, 1.27) and with increased use of renal replacement therapy among 9,258 patients (RR, 1.32).
In addition to emphasizing “the potentially important and adverse effects of scientific misconduct,” this study demonstrates the importance of revising and revisiting recommendations and guidelines, editorialists wrote. The study doesn't settle the controversy about colloids vs. crystalloids because the HES comparators were too heterogeneous—a fault of the existing literature pool, not the meta-analysis itself, they wrote. High- quality trials that compare HES 130/0.4 starch vs. crystalloids need to be done; in the meantime, the results of this analysis suggest the harms of HES probably outweigh the benefits, and HES shouldn't be used in critically ill patients who need acute volume resuscitation, they wrote. The meta-analysis and editorial appeared Feb. 20 in JAMA.
In the second study, reviewers found nine studies of 3,456 patients with sepsis who had been randomized to HES 130/0.38-0.45 vs. crystalloid, or HES 130/0.38-0.45 vs. human albumin. While HES didn't seem to affect the relative risk of death for relevant trials (RR, 1.04 in eight trials), the relative risk was 1.11 in the predefined analysis of trials with low risk of bias (four trials). Renal replacement was used more in the HES group (RR, 1.36, five trials) and acute kidney injury was more common (RR, 1.18, four trials). Also, more patients in the HES group were transfused with red blood cells (RR, 1.29, three trials), and more in this group had serious adverse events (RR, 1.30, four trials). The authors concluded that it “seems unlikely that hydroxyethyl starch 130/0.38-0.45 provides overall clinical benefit for patients with sepsis.” The study was published Feb. 15 by BMJ.
For inpatient VTE prevention, rivaroxaban noninferior to enoxaparin
Rivaroxaban is noninferior to enoxaparin for standard-duration venous thromboembolism (VTE) prophylaxis in acutely ill patients, a recent study found.
Researchers randomly assigned 8,101 patients from 52 countries to receive either 40 mg of subcutaneous enoxaparin per day for 10 ±4 days and oral placebo for 35 ±4 days, or subcutaneous placebo for 10 ±4 days and 10 mg of oral rivaroxaban per day for 35 ±4 days. All patients were at least 40 years old and had been hospitalized for an acute medical illness; median hospitalization time was 11 days. More than 30% of patients had two or more acute medical conditions. The primary outcomes were the composite of asymptomatic proximal or symptomatic VTE up to day 10, and up to day 35. The main safety outcome was the composite of major and non-major but clinically relevant bleeding. Results were published in the Feb. 7 New England Journal of Medicine.
The primary efficacy outcome occurred in 2.7% of both rivaroxaban and enoxaparin patients at day 10. At day 35, however, rivaroxaban was associated with lower risk: 4.4% (n=131) of patients versus 5.7% (n=175) with enoxaparin (relative risk, 0.77; P=0.02). Bleeding risk, however, was higher with rivaroxaban, occurring in 2.8% (n=111) of that group and 1.2% (n=49) of the enoxaparin group at day 10 (P<0.001). At day 35, bleeding rates also differed, at 4.1% (n=164) for the rivaroxaban group versus 1.7% (n=67) for the enoxaparin group (P<0.001). Although extended-duration rivaroxaban was associated with a decrease in deaths related to VTE and an increase in deaths related to bleeding, the all-cause death rate was no different between the two drugs.
The study was limited by its inclusion of asymptomatic proximal deep venous thrombosis—as detected on ultrasonography—as part of the main outcome, the authors noted. Ultrasonography isn't routinely performed on medical patients, so the performance of the test on day 10 may have influenced the natural history of the disease by resulting in treatment of asymptomatic disease. This, in turn, could explain a risk reduction at day 35 that was lower than expected, they wrote.
More at-home deaths in Medicare patients, but also more end-of-life ICU stays
More Medicare patients died at home in 2009 than in 2000, but there was also an increase in intensive care use during the last 30 days of life, a recent study found.
The retrospective cohort study included a random 20% sample of Medicare fee-for-service beneficiaries (more than 800,000 patients total) who died in 2000, 2005 or 2009. Researchers compared these patients' sites of death, places of care in the last 30 days, hospice use and health care transitions at the end of life. Results appeared in the Feb. 6 Journal of the American Medical Association.
Between 2000 and 2009, the percentage of patients dying in acute care hospitals dropped from 32.6% to 24.6%. The percentage of patients in hospice at the time of death also changed significantly, from 21.6% to 42.2%. However, the study also found an increase in the use of the intensive care unit (ICU) in the final 30 days, from 24.3% of deaths in 2000 to 29.2% in 2009. The researchers noted an increase in health care transitions at the end of life too, especially in the last three days (10.3% of patients in 2000 versus 14.2% in 2009). Specifically focusing on the patients who used hospice in 2009, they found that 28.4% of them had been in for three days or less and, of those, 40.3% had come from the ICU.
Previous reports have also shown that more elderly patients are dying at home, but this study shows that these patients are not necessarily receiving less aggressive care, the authors concluded. The finding that hospice admissions were short and followed ICU stays suggests that the increasing use of hospice may not reduce resource utilization. Although the study was not able to collect data on patient preferences, the authors speculated that the observed patterns of care were not the result of patient choice and could be improved.
An accompanying editorial also called for greater attention to patient preferences and goals, as well as provision of active treatments such as intravenous fluids or antibiotics outside the hospital whenever possible. Set criteria for ICU admissions, based on likely benefit and life expectancy, could also be helpful in reducing inappropriate and costly care, the editorialist said.
Endovascular therapy not superior to IV tPA for acute ischemic stroke
For patients with acute ischemic stroke, endovascular therapy isn't superior to standard treatment with intravenous tissue plasminogen activator (IV tPA), a study found.
Researchers in a multicenter study randomly assigned 362 acute ischemic stroke patients to IV tPA or to endovascular therapy (intra-arterial thrombolysis with recombinant tPA, mechanical clot disruption or retrieval, or a combination of these). All patients were randomized within 4.5 hours of stroke onset and received treatment as soon as possible after randomization. (IV tPA patients got therapy within 4.5 hours of symptom onset, and endovascular patients got it within six hours.) The main outcome was survival free of disability at 90 days, defined by a modified Rankin score of 0 or 1 (scale out of 6, with 0 indicating no symptoms and 6 indicating death). Results were adjusted for age, sex, stroke severity and presence of atrial fibrillation and were published online Feb. 6 by The New England Journal of Medicine.
The median time from stroke onset to the beginning of treatment was 2.75 hours for IV tPA and 3.75 hours for endovascular therapy. At three months, 30.4% of patients (n=55) in the endovascular therapy group and 34.8% in the IV tPA group (n=63) had survived without disability (adjusted odds ratio, 0.71; P=0.16). Six percent of patients in each group experienced fatal or nonfatal symptomatic intracranial hemorrhage within seven days. There were no significant differences between groups in the rates of other serious adverse events or in the case fatality rate. A subgroup analysis suggested that the absence of superiority of endovascular treatment didn't depend on the type of center, stroke subtype or time to endovascular treatment.
The authors had hypothesized that the downside of endovascular treatment, in terms of time spent, would be eclipsed by its faster and more effective revascularization compared to IV tPA, they wrote. However, the results didn't show endovascular therapy achieved superior outcomes and “do not provide support for the use of the more invasive and expensive endovascular therapy over intravenous treatment,” the authors concluded.