Better Poststroke Outcome Follows Antidepressant Use

Antidepressant medications can effectively treat post-stroke depression (PSD), potentially leading to improvement in outcomes and decreases in overall health care services utilization by patients, a pair of new studies indicates.

After an acute stroke, the first of the two studies found, health care utilization was significantly higher among patients with PSD than in patients who had strokes but did not have PSD. The study, funded by the Department of Veterans Affairs (VA), appeared in the November 2006, Stroke.

“PSD is common among stroke survivors, and it is associated with worse functional outcomes and increased poststroke mortality,” Huanguang Jia, Ph.D., of the VA Medical Center in Gainesville, Fla., and colleagues pointed out. “Limited information is available about its impact on health care use.”

In a national retrospective, observational study, Jia and his colleagues evaluated the impact of PSD on health care use by 5,825 veterans with acute stroke, identified from VA inpatient databases. The team used VA and Medicare inpatient and outpatient data and VA pharmacy information to determine the subjects' PSD status and health care utilization.

Patients were determined to have PSD if they had an inpatient or outpatient diagnosis of depression or received one of the antidepressants on the VA formulary system within 12 months after the index stroke. Jia and his colleagues found that 41 percent of the veterans with acute stroke in their sample had PSD.

Health care use was determined by the number of hospital stays, outpatient visits, and the cumulative length of inpatient stays.

Compared with subjects who had suffered strokes but did not have PSD, those with the disorder had more average number of inpatient stays (2.4 versus 1.8) and outpatient visits (30.7 versus 20). Patients with PSD also had longer length of inpatient stays (25 versus 17 days). Each of these differences remained statistically significant even after adjustment for patients' demographic and clinical factors.

“More specifically, we estimated that patients with PSD had inpatient stays 1.2 times, outpatient visits 1.3 times, and length of stays 1.4 times that of the patients without PSD, respectively, within the first 12 months [following] the index stroke,” Jia explained.

He and his colleagues concluded that early identification and treatment of PSD may help “reduce overall cost of care, prevent premature deaths, and improve functional recovery and quality of life.”

Treatment Is Safe and Effective

In the second study, Yan Chen, M.P.H., Ph.D., a postdoctoral student in the School of Pharmacy at the University of Cincinnati, and colleagues performed a metaanalysis of randomized, placebo-controlled trials of antidepressants in patients with PSD. The team reviewed studies published since 1984.

Outcome measures used in the studies varied and included response rate, depression rating scale scores, recovery from neurologic impairments, and improvement in activities of daily living (ADLs) after stroke.

A total of 1,320 patients were identified as being treated for PSD in 16 randomized, controlled trials. The pooled response rates in the active groups were 65.18 percent (234 responders/359 patients) and 44.37 percent for those in placebo groups (138 responders/311 patients). This difference was statistically significant.

From baseline to endpoint, patients in the active group had significantly greater improvement in depressive symptoms than patients in the placebo group. In addition, Chen and his colleagues found that longer duration of treatment correlated with the degree of improvement in depressive symptoms.

Three PSD clinical trials reported treatment effects on neurological impairment using the internationally recognized Scandinavian Neurological Stroke Scale (SNSS). While patients in the active-treatment groups did show improvement compared with patients in the placebo groups, the differences were not statistically significant. Three other studies, using other standardized assessments of neurological impairment such as the Chinese Stroke Scale, did find statistically significant improvement in treatment groups compared with placebo groups.

Antidepressant Lead to Improvement

Treatment effects of antidepressants on ADLs were reported in five studies. On average, the active-treatment groups showed statistically significant improvement, compared with placebo groups, as measured by the Barthel Index, a standardized measure of functional impairment. No significant difference was seen, however, between the two groups on a different neurological-impairment assessment tool, the Functional Independence Measure.

While Chen and his colleagues found no statistically significant improvements in SNSS scores, and the outcomes related to ADLs were inconsistent across different scale measurements, “the small number of studies available for the evaluation of these outcomes [limited] the [study's] power” to detect significant improvements.

“An important finding observed in this study,” Chen and his coauthors concluded, “is the significant relationship between the degree of improvement in depressive symptoms and duration of treatment. The benefits of antidepressants in reducing depressive symptoms became significant after [three] or [four] weeks of treatment, and increased with continued treatment.”

Effective treatment of PSD, Chen noted, “may contribute to the recovery of stroke-induced deficits. Studies in which antidepressants were evaluated for the prevention of PSD also suggest that antidepressants may help or accelerate the rehabilitation of patients with stroke. Neuronal recovery in stroke may be mediated through the effects of selective serotonin-reuptake inhibitors on brain-derived neurotrophic factor and neurogenesis.”

“The Impact of Poststroke Depression on Healthcare Use by Veterans With Acute Stroke,” is posted at<http://stroke.ahajournals.org/cgi/content/abstract/37/11/2796>;“ Treatment Effects of Antidepressants in Patients With Post-Stroke Depression: A Meta-Analysis,” is posted at<www.theannals.com/cgi/content/abstract/40/12/2115>.▪