Recent Research

Post-stroke depression risk, side effects from statins, and more.

Inpatient data can predict post-stroke depression risk

A clinical prediction model could help predict stroke patients' risk of developing depression using data collected within a week after stroke, according to a recent study.

The study included 410 consecutive stroke patients in the Netherlands who were able to communicate adequately in the week after stroke. They were followed for six to eight weeks after stroke for the development of major depressive disorder. Multivariable logistic regression models were fitted and a bootstrap-backward selection process was used to reduce the prediction model. Results were published in the September Stroke.

The final model included a medical history of depression or other psychiatric disorders (increased odds ratio [OR] of depression, 7.22 [95% CI, 3.63 to 14.35]), hypertension (OR, 0.49 [95% CI, 0.26 to 0.92]) or angina pectoris (OR, 2.82 [95% CI 1.33 to 6.21]), and the patient's Bethel item dressing score. Compared to patients who dressed completely independently as the reference, patients who needed help but could do about half of their dressing unaided had reduced risk of depression (OR, 0.26 [95% CI, 0.08 to 0.82]) and those who were completely dependent had increased risk (OR, 1.57 [95% CI, 0.80 to 3.09]).

The resulting model had acceptable discrimination, the researchers concluded, with an area under the receiver-operating characteristic curve of 0.78 (95% CI, 0.72 to 0.85) and P value of the U-statistic of 0.96. Patients in the lowest risk category had a 2% rate of depression, compared to 82% in the highest risk group. One limitation of the model is that 69% of the patients assessed for inclusion in the study were too ill to participate, so the findings cannot be generalized to the many patients unable to communicate in the week after stroke.

Despite this limitation, and the need for validation in other populations, this model (which researchers named the Post-stroke Depression Prediction Scale [DePreS]) could be useful to clinicians in daily care to estimate their patients' risk for depression. If use of the model were followed by depression treatment and follow-up, it could improve the care of stroke patients, the study authors wrote.

Guideline-based therapy linked to acute MI survival in oldest old

Increased use of guideline-based therapy appeared to improve survival after acute myocardial infarction (MI) in patients 85 years of age and older, a recent study has found.

Researchers looked at patients at least 85 years of age and older who were hospitalized for acute MI in six biennial periods from 1997 to 2007 in 11 medical centers in Massachusetts as part of the Worcester Heart Attack Study. Trends in 90-day survival after discharge and use of guideline-based medications for MI during the hospital stay and at discharge were examined. Guideline-based medications were defined as aspirin, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, beta-blockers and lipid-lowering agents. The authors used sequential multivariable Cox regression models to assess the relationship among medication use, year, and survival rates 90 days after discharge. Results were published in the September American Journal of Medicine.

A total of 1,137 patients were included in the study sample, 527 hospitalized between 1997 and 2001 and 610 hospitalized between 2003 and 2007. Patients ranged in age from 85 to 105 years; average age in the first group was 89.1 years, and average age in the second group was 88.7 years. The researchers found that 90-day survival rates were higher for those hospitalized between 2003 and 2007 compared with 1997 to 2001 (69.1% vs. 59.8%; P<0.05). The average number of guideline-based medications prescribed also increased from 1.8 in 1997 to 2.9 in 2007 (P<0.001).

The authors also looked at trends in use of procedural intervention and thrombolytics and found that a larger percentage of patients had percutaneous coronary intervention during hospitalization in 2007 compared with 1997 (14.6% vs. <1%; P<0.001), in contrast to rates of coronary artery bypass grafting, which remained below 1% throughout the study period. Thrombolytic use decreased significantly from 1997 to 2007 (10.3% vs. 0%; P<0.001). The unadjusted hazard ratio for mortality at 90 days after discharge in 2003-2007 versus 1997-2001 was 0.73, but the relationship did not remain significant after adjustment for patient characteristics and guideline-based medication use (hazard ratio, 1.26).

The authors noted that their study involved only Massachusetts, that they couldn't determine whether patients were eligible for each of the included types of therapy, and that they had no data available on medication adherence after discharge, among other limitations. However, they concluded that 90-day survival after hospitalization for MI improved in their population-based sample of adults 85 years of age and older from 1997 to 2007. “These encouraging trends seem to have been predominantly explained by increases in the use of guideline-based cardiac medications,” the authors wrote. They called for further studies to determine the effects of treatment on long-term survival, as well as the clinical and public health effects of medication adherence.

Colistin associated with higher rate of AKI than polymyxin B

Critically ill patients who were given intravenous colistin had a greater incidence of acute kidney injury (AKI) than those given polymyxin B, a study found.

In a retrospective cohort study, researchers examined data on 67 adult patients who received intravenous polymyxin B (PB) between January 2008 and June 2009 and 106 patients who received colistin from January 2009 through June 2010, both for at least 72 hours. Patients with normal renal function had PB dosed at 15,000 to 25,000 units/kg per day as a continuous infusion over 24 hours. Colistin was dosed at 5 mg/kg of ideal body weight per day or actual body weight or as a fixed dose of 150 mg every 12 hours. RIFLE (risk, injury, failure, loss and end-stage kidney disease) criteria were used to define nephrotoxicity and to evaluate the severity of acute renal failure. Both agents were mostly used to treat multidrug-resistant Pseudomonas aeruginosa and Acinetobacter baumannii infections, and most patients were in the ICU.

The mean treatment duration was 11.7 days for colistin and 12.5 days for PB (P=0.66). Nephrotoxicity developed in 60.4% of colistin patients and 41.8% of PB patients (P=0.02). Among those who developed it, the difference in time to peak serum creatinine wasn't significantly different between groups. Most patients who developed acute renal failure fell into the reversible kidney injury category. Only one patient progressed to end-stage renal disease (from the colistin group). Results were published online July 9 in Clinical Infectious Diseases.

The study is limited by its retrospective nature and small sample size and by the fact that the drugs were studied in two different time periods, so potential variations in practice could have affected results. Also, dose adjustments or drug discontinuations due to acute renal failure weren't captured. Overall, the study adds more strength to previous results showing a temporal relationship between using these agents and development of nephrotoxicity, the authors wrote, and patients who take them should be closely monitored.

Majority of TTEs appropriate but may not change management

Most transthoracic echocardiograms (TTEs), even those done for appropriate clinical indications, have little effect on patient management, according to a recent study.

Researchers in Texas retrospectively reviewed medical records of TTEs at one medical center and classified them according to appropriate use criteria developed in 2011 and assessed clinical impact. The two cardiologists who judged appropriate use criteria were blinded to clinical impact, and the two cardiologists who assessed clinical impact were blinded to appropriate use criteria. When the cardiologists disagreed and could not reach consensus on either clinical impact or appropriate use, a third cardiologist, also blinded, adjudicated. A TTE could be classified as leading to an active change in care, continuation of current care, or no change in care. The study's main outcome measures were prevalence of appropriate, inappropriate and uncertain TTEs based on appropriate use criteria and the prevalence of the three clinical impact categories. The results were published online July 22 by JAMA Internal Medicine.

All TTEs ordered at University of Texas Southwestern Medical Center from April 1 through April 30, 2011, were reviewed and 535 were included in the analysis. Most of the study patients were women (58.7%); 55.3% were white, 21.1% were African American and 8.2% were Hispanic. The mean age was 58 years. Fifty-seven percent of TTEs were ordered for inpatients, and general internists (38.5%) and cardiologists (31.2%) were the specialists most likely to order the tests.

The researchers found that 31.8% of TTEs led to an active change in care, 46.9% led to continuation of current care and 21.3% led to no change in care. Based on the 2011 appropriate use criteria, 91.8% of TTEs were appropriate, 4.3% were inappropriate, and 3.9% were of uncertain appropriateness. A similar proportion of appropriate and inappropriate TTEs led to active change in care (32.2% vs. 21.7%, respectively; P=0.29). Outpatient TTEs were slightly less likely than inpatient ones to be judged appropriate (86.5% vs. 95.7%, respectively; P<0.001).

The study was limited by its reliance on electronic medical records, which may have provided incomplete information and led to misclassification of clinical impact, the authors noted. They also pointed out that they were unable to assess patient satisfaction and that their results may not be generalizable to other practice settings, among other limitations. They concluded, however, that while almost all TTEs in their study were considered appropriate by 2011 criteria, fewer than one-third led to an active change in care. In addition, almost half led to continuation of current care and approximately 21% resulted in no change in care.

“The discrepancy between appropriateness and clinical impact is striking and suggests that the [appropriate use criteria] as currently implemented are unlikely to facilitate optimal use of TTE,” they wrote. They called for further research examining the necessity of TTE in medical care.

The author of one of two invited commentaries said that while the existing appropriate use criteria are “meticulous,” the evidence on how and when to best use echocardiography is slim. The current study, he said, “demonstrates that the concepts of appropriateness and usefulness may diverge considerably. Transthoracic echocardiograms cost more than 71 billion per year to Medicare alone, and many TTE procedures performed by the book may still not lead to improved outcomes.” He called for additional randomized trials of diagnostic testing focusing on specific clinical scenarios. “Such trials may cost more than collecting observational data, but for common tests and indications, their results may be definitive, eventually saving far more money,” he wrote.

The second invited commentary noted the limitations of a retrospective review of electronic medical records and said that the results should be confirmed prospectively before major conclusions are drawn. “Certainly, the [appropriate use criteria] are not without remaining flaws and ideally should result in a categorization scheme that can be demonstrated to have a consistent, but necessarily invariable, effect on medical decision making,” the authors wrote. “This retrospective study points the way for further prospective studies looking at the impact of echocardiography and how it affects physician decision making.”

Care transitions program may help reduce 30-day readmissions

A recent study suggests a program that helps older patients transition from hospital to home or other care facilities may reduce readmissions, though editorialists disagreed on the relevance of the results.

Researchers studied outcomes from the acute care units of 11 hospitals that implemented Project BOOST (Better Outcomes by Optimizing Safe Transitions), a Society of Hospital Medicine initiative to promote best practices for hospital discharge care transitions. The main outcome was 30-day readmissions, although researchers also looked at length of stay in BOOST units compared to control units. Data were collected from December 2008 through June 2010. Compared to the 19 BOOST hospitals in the initial cohort that didn't participate, those that volunteered to participate in the study tended to be academic facilities, as well as larger and in urban locales.

The average rate of 30-day readmissions was 14.7% before BOOST compared to 12.7% 12 months later (P=0.010)—an absolute reduction of 2% and a relative reduction of 13.6% (P=0.054). Readmission rates for matched control units that didn't implement BOOST were 14% in the pre-intervention period and 14.1% in the post-intervention period. Length of stay in BOOST and control units didn't differ significantly. Results were published in the August Journal of Hospital Medicine.

The results are “encouraging,” but their external validity may be limited to facilities with the resources to report to a national database and to capture analytics at an individual unit level, the researchers wrote. They also noted that while the study indicates modest gains in reducing readmissions are possible through a concerted effort at changing hospital processes, the Centers for Medicare and Medicaid Services' goal of a 20% reduction in readmissions may be difficult to achieve.

An editorial praised the flexibility of BOOST as a program but noted that the study results are limited by the large number of facilities that dropped out of the program or failed to submit data, which may have led to results biased toward success. Also, the overall results appear to be driven almost entirely by success at a single site, it said. “In the end, it seems unlikely that this iteration of the BOOST program produced broad reductions in readmission rates,” the editorialists wrote.

Side effects rare with statins

Side effects associated with statin therapy are not common, and simvastatin and pravastatin may be safer and more tolerable than other drugs in the class, reported a meta-analysis that totaled nearly a quarter-million people.

Researchers systematically reviewed 55 two-armed placebo-controlled trials and 80 two- or multiarmed active-comparator trials that enrolled 246,955 individuals with and without cardiovascular disease to evaluate different statins. Individual statins showed no significant differences from controls for the following side effects:

  • Myalgia: 43,531 participants, statins vs. controls, odds ratio (OR), 1.07 (95% CI, 0.89 to 1.29; I2, 22.1%). Simvastatin had lower odds than atorvastatin (OR, 0.56; 95% CI, 0.42 to 0.75; I2, 0.0%).
  • Creatine kinase elevation: 101,324 participants, statins vs. controls, OR, 1.13 (95% CI, 0.85 to 1.51; I2, 20.4%). Pitavastatin resulted in significantly more elevations than control (OR, 3.63; 95% credible interval, 1.10 to 14.10).
  • Discontinuations because of adverse events: 76,462 participants, statins compared to controls, OR, 0.95 (95% CI, 0.83 to 1.08; I2, 21.9%). Simvastatin was significantly more tolerable than atorvastatin (OR, 0.61; 95% CI, 0.42 to 0.89; I2, 71.9%) or rosuvastatin (OR, 0.49; 95% CI, 0.27 to 0.88; I2, 0.0%).

Statins as a class resulted in significantly higher odds of diabetes (OR, 1.09; 95% CI, 1.02 to 1.16) and transaminase elevations (OR, 1.51; 95% CI, 1.24 to 1.84) compared with control. However, there was no evidence of increased risk of cancer among 100,523 participants (OR, 0.96; 95% credible interval, 0.91 to 1.02; I2, 0.0%), nor was there evidence of potential head-to-head differences between statins for these outcomes. Results were published in the July Circulation: Cardiovascular Quality and Outcomes.

The researchers noted that when statins were compared, there were numerous statistically detectable differences favoring simvastatin and pravastatin. Dose-level comparisons showed there were higher odds of discontinuations with higher doses of atorvastatin and rosuvastatin, while higher doses of atorvastatin, fluvastatin, lovastatin and simvastatin were associated with higher odds of transaminase elevations. Simvastatin at its highest doses was associated with creatine kinase elevations (OR, 4.14; 95% credible interval, 1.08 to 16.24).

The researchers wrote, “At the population level, mortality and cardiovascular benefits of statin therapy greatly overweigh its potential harms, even taking into account the recent finding that statin use is associated with a modest increase in diabetes mellitus incidence. At the individual level, however, there may be a risk of exposing a large group of individuals to the (primarily minor) harms of statin therapy for the benefit of a smaller number of individuals. This brings into sharp focus the importance of correctly identifying the set of individuals who stand to benefit from statin therapy.”

Drug combo during CPR, stress-dose hydrocortisone in postresuscitation shock associated with favorable outcome

Vasopressin-epinephrine and methylprednisolone during CPR and stress-dose hydrocortisone in postresuscitation shock were associated with better neurologically favorable survival to hospital discharge compared with epinephrine plus a saline placebo in patients in cardiac arrest who required vasopressors, a recent study has found.

Researchers in Greece performed a randomized, double-blind, placebo-controlled parallel-group trial from Sept. 1, 2008, to Oct. 1, 2010, in three tertiary care centers. Patients were assigned to receive vasopressin, 20 IU/CPR cycle, plus epinephrine, 1 mg/CPR cycle or saline placebo plus epinephrine (1 mg/CPR cycle) for the first five CPR cycles after randomization, plus additional epinephrine if needed. A CPR cycle lasted approximately three minutes. In the first CPR cycle after randomization, patients in the vasopressin-epinephrine group received methylprednisolone, 40 mg, and controls received saline placebo. Patients in postresuscitation shock received stress-dose hydrocortisone, 300 mg/d, for a maximum of seven days with gradual tapering.

The study's primary outcomes were return of spontaneous circulation for at least 20 minutes and survival to hospital discharge with a Cerebral Performance Category (CPC) score of 1 or 2. The CPC score includes five categories: good cerebral performance, moderate cerebral disability, severe cerebral disability, coma or vegetative state, and death. A score of 1 indicates consciousness and ability to work and live normally; a score of 2 indicates consciousness and ability to conduct independent daily living activities but the presence of such disorders and hemiplegia, seizures and cognitive changes.

The study results appeared in the July 17 Journal of the American Medical Association.

Two hundred sixty-eight consecutive patients with cardiac arrest were included in the study, 130 assigned to the vasopressin-epinephrine group and 138 assigned to the control group. The mean patient age was approximately 63 years, and most patients (68%) were men. Follow-up was complete in all resuscitated patients.

Patients in the vasopressin-epinephrine group were more likely to have return of spontaneous circulation for at least 20 minutes and were more likely to survive to hospital discharge with a CPC score of 1 or 2 (83.9% vs. 65.9% and 13.9% vs. 5.1%, respectively; P=0.005 and P=0.02). Patients in the vasopressin-epinephrine group who experienced postresuscitation shock were also more likely than patients with postresuscitation shock in the control group to have neurologically favorable survival to discharge (21.1% vs. 8.2%; P=0.02), as well as better hemodynamics, central venous oxygen saturation and organ dysfunction. Adverse event rates were similar between vasopressin-epinephrine patients and controls.

The authors noted that their study did not determine prevasopressor CPR hemodynamics, physiologic variables after return of spontaneous circulation, stress hormone levels at baseline, or myocardial function after cardiac arrest, among other limitations. However, based on their findings, they concluded that compared with epinephrine and saline placebo, vasopressin-epinephrine and methylprednisolone during CPR and stress-dose hydrocortisone in postresuscitation shock improved neurologically favorable survival to hospital discharge among patients with in-hospital cardiac arrest requiring vasopressors.

Intermittent pneumatic compression helps reduce VTE risk

Intermittent pneumatic compression (IPC) to the lower limbs can reduce venous thromboembolism (VTE) risk in hospitalized patients and is especially effective when combined with pharmacologic thromboprophylaxis, a recent meta-analysis found.

Researchers searched MEDLINE, EMBASE, and Cochrane databases for randomized controlled trials that compared IPC with pharmacologic thromboprophylaxis, thromboembolic deterrent stockings (TEDS), no prophylaxis, and a combination of IPC and pharmacologic thromboprophylaxis. They excluded trials that used IPC for surgery only or for less than 24 hours after surgery or that compared different types of IPC without a placebo group. The main outcomes were deep venous thrombosis (DVT) and pulmonary embolism (PE).

Results were published in the Aug. 27 Circulation.

In total, 16,164 hospitalized patients from 70 trials in 15 countries met inclusion criteria and were included in the meta-analysis. The overall quality of the trials was “modest,” the researchers said. They found that IPC was more effective than no IPC prophylaxis in reducing DVT (7.3% vs. 16.7%; absolute risk reduction [ARR], 9.4%; 95% CI, 7.9% to 10.9%; P<0.01) and PE (1.2% vs. 2.8%; ARR, 1.6%; 95% CI, 0.9% to 2.3%; P<0.01). IPC was also more effective than TEDS in reducing DVT and seemed to be as effective as pharmacologic thromboprophylaxis but with a reduced risk of bleeding (relative risk [RR], 0.41; 95% CI, 0.25 to 0.65, P<0.01). Adding pharmacologic thromboprophylaxis to IPC further reduced the risk of DVT (RR, 0.54; 95% CI, 0.32 to 0.91; P=0.02) compared to IPC alone.

The researchers noted that underuse of thromboprophylaxis is common in many facilities and that IPC is an “attractive option” because it can be used in almost all inpatients, including those with active bleeding or at higher risk of bleeding.

“IPC should be used as early as possible for hospitalized patients who have contraindications to pharmacologic thromboprophylaxis, and pharmacologic thromboprophylaxis should be added to the IPC instead of replacing it when the risk of bleeding subsides for patients who are at high-risk of venous thromboembolism,” they concluded, in line with the American College of Chest Physicians' clinical practice guidelines on preventing thrombosis.