Selective Serotonin Reuptake Inhibitors and Risk of Hemorrhagic Stroke
Background and Purpose— Selective serotonin reuptake inhibitors (SSRI) are widely prescribed. Several reports have observed an increased bleeding risk associated with SSRI use, which is hypothesized to be secondary to their antiplatelet effect.
Methods— We tested the hypothesis that SSRIs increase the risk for or potentiate the risk of hemorrhagic stroke associated with antiplatelets and anticoagulants.
Results— In multivariate analysis, we found no increased risk associated with SSRI use for intracerebral hemorrhage (odds ratio=1.1, 95% CI: 0.7 to 1.8; P=0.63) or subarachnoid hemorrhage (odds ratio=0.6, 95% CI: 0.4 to 1.0; P=0.054). In addition, potentiation of risk with warfarin or antiplatelets was not observed.
Conclusions— Further studies with larger populations would be needed to exclude a small increase in intracranial hemorrhage risk with SSRI use.
Approximately 99 000 hemorrhagic strokes occur each year in the United States, accounting for 10% to 15% of all strokes.1 Known risk factors for hemorrhagic stroke include increasing age, prior ischemic stroke, hypertension, smoking, and treatment with anticoagulant and antiplatelet medications. Between 1988 and 2002, 67.6 million new patients were started on the selective serotonin reuptake inhibitors (SSRI) fluoxetine, paroxetine, and sertraline.2 Thus, a potential effect of SSRIs on bleeding risk is of major significance.
Serotonin is released by platelets at the site of vascular injury to promote clot formation. When the endothelial layer is damaged, platelets come in contact with subendothelial collagen, which leads to release of intracellular serotonin granules that promote clot formation. Platelets contain a serotonin reuptake transporter that is similar to in the brain. It has been demonstrated that SSRIs inhibit this transporter in platelets.3,4 One study demonstrated an 83% decrease in platelet serotonin concentrations after 2 weeks of therapeutic paroxetine administration.4 We tested the hypothesis that SSRIs increase the risk for hemorrhagic stroke and potentiate the risk associated with antiplatelet and anticoagulant drugs.
As part of our Genetic and Environmental Risk Factors for Hemorrhagic Stroke study (NS36695), cases of intracerebral (ICH) and subarachnoid hemorrhage (SAH) were identified in the Greater Cincinnati region from May 1997 to August 2001 and from July 2002 to October 20055 through retrospective screening of all area emergency rooms and discharge ICD-9 codes and prospective screening of neurosurgery admissions at the busiest hospitals. Patients were approached for enrollment in a genetic sampling and interview arm. Subjects who agreed were matched by age, race, and gender to population-based controls by random digit dialing. All medications taken 2 weeks before index date were recorded. The index date for cases was the date of onset; for controls it was 1 to 2 weeks before the interview, to avoid recall bias. SSRIs used for the current analysis included citalopram, escitalopram, fluoxetine, paroxetine, and sertraline. Classification of antiplatelet drugs included aspirin, clopidogrel, and dipyridamole. Because of the small number of subjects on both nonaspirin antiplatelet drugs and SSRIs (5 patients), the analysis was limited to aspirin alone.
The data were managed and analyzed using SAS version 9.1 (SAS Institute). Univariate and multivariable analysis for the association of independent variables with stroke was achieved using a conditional logistic modeling approach (PROC LOGISTIC). The multivariable analysis used a backward elimination procedure for risk factors with significance levels of P<0.10. Interaction terms of SSRI with and without aspirin or warfarin use were included into secondary models.
During the study period, 2441 patients with ICH and 894 patients with SAH were identified, of which 500 ICH (20%) and 416 SAH (46.5%) cases were enrolled in the case-control study. Comparison of interviewed and noninterviewed patients found that interviewed patients were younger for both SAH (50.9 versus 57.7 years) and ICH (65 versus 70.2 years). After adjustment for age, there was no significant difference in the prevalence of hypertension, diabetes, heavy alcohol use, race, gender, or SSRI use between interviewed and noninterviewed patients with ICH or SAH. There was a trend toward higher rates of current smoking among interviewed SAH patients compared with noninterviewed patients (50.4 versus 43.5%; P=0.054).
Of the 916 hemorrhagic stroke patients enrolled, 71 (7.8%) were on an SSRI at the time of stroke, and of 1776 demographically matched controls, 158 (8.9%) were on an SSRI. The Table presents the univariate and multivariate analysis of SSRI risk for all hemorrhages. After controlling for multiple risk factors, SSRI use was not independently associated with increased risk for hemorrhagic stroke (odds ratio [OR]=0.8, 95% CI: 0.5 to 1.2; P=0.25).
Of ICH and SAH subtypes, 44 and 27, respectively, were using an SSRI. Multivariate analyses revealed no increased risk associated with SSRI for ICH (OR=1.1, 95% CI: 0.7 to 1.8; P=0.63) or SAH (OR=0.6, 95% CI: 0.4 to 1.0; P=0.054). In interaction analysis, the use of SSRIs and warfarin did not confer a significantly greater risk of hemorrhagic stroke (OR=4.7, 95% CI: 1.2 to 18.4) than warfarin alone (OR=3.0, 95% CI: 1.8 to 5.0). The use of SSRIs and aspirin did not confer a significantly greater risk (OR=0.9, 95% CI: 0.5 to 1.5) than aspirin alone (OR=1.1, 95% CI: 0.8 to 1.3). After stratification by subtype, SSRI potentiation of aspirin was not observed for ICH (OR=1.1 95% CI: 0.5 to 2.5) or SAH (OR=0.4, 95% CI: 0.1 to 1.7).
We did not find an increased risk of hemorrhagic stroke or its subtypes with SSRI use, nor did we find evidence that SSRIs potentiate the bleeding risk conferred by warfarin or aspirin. These results are important given the widespread use of SSRIs, the prior reports of systemic bleeding complications, and the specific use of SSRI for poststroke depression.
A growing number of case reports attribute bleeding events to SSRI use. Sites of bleeding include genitourinary, respiratory, and gastrointestinal systems.6–8 Two population-based cohort studies found SSRI users were more likely to be hospitalized for gastrointestinal bleeding than nonusers.9,10 Both studies found that the bleeding risk of aspirin was potentiated by SSRI treatment.9,10
Our results are in accordance with 2 prior reports which assessed intracranial bleeding risk associated with SSRIs.11,12 A nested case-control study in a cohort of antidepressant users found no association between SSRI use and intracranial hemorrhage.10 Although no increased risk of hemorrhage was observed, the study identified only 65 cases of hemorrhage of which 7 were exposed to SSRIs. Another nested case-control study based on information from a patient registry concluded that SSRI exposure was not a risk factor for ICH.12 The study included 21 ICH cases exposed to SSRI. Hypertension, diabetes, atrial fibrillation, and ischemic heart disease were adjusted for using drug prescriptions as proxy measures.
The present study includes a larger number of SSRI-exposed cases of hemorrhagic stroke (n=71) and controls and a more comprehensive list of relevant risk factors. Our results strengthen the argument that SSRI use does not lead to an increased risk of hemorrhagic stroke. After an ischemic stroke, many patients experience depression and are frequently placed on an SSRI along with antiplatelet or anticoagulant drugs, and our findings provide reassurance that no major increase in hemorrhage risk is expected.
A limitation is that our study had 81% power to detect an OR of 1.6 or greater and a smaller risk may not have been detected. Although our interviewed cases were similar to noninterviewed cases, they were slightly younger. Thus, our study may miss an increased risk of hemorrhagic stroke with SSRI use among an older population.
Further study is needed to determine whether SSRI use increases the risk of hemorrhage among patients with a prior ischemic stroke or has a small effect on risk.
- Received April 17, 2007.
- Accepted April 27, 2007.
Kissela B, Schneider A, Kleindorfer D, Khoury J, Miller R, Alwell K, Woo D, Szaflarski J, Gebel J, Moomaw C, Pancioli A, Jauch E, Shukla R, Broderick J. Stroke in a biracial population: the excess burden of stroke among blacks. Stroke. 2004; 35: 426–431.
Aldred G. An analysis of use of Prozac, Paxil, and Zoloft in USA 1988–2002. In: Alliance for Research Protection; 2004. Available at: http://www.ahrp.org/perlfect/search/search_form.html.
Javors MA, Houston JP, Tekell JL, Brannan SK, Frazer A. Reduction of platelet serotonin content in depressed patients treated with either paroxetine or desipramine. Int J Neuropsychopharmacol. 2000; 3: 229–235.
Woo D, Sauerbeck LR, Kissela BM, Khoury JC, Szaflarski JP, Gebel J, Shukla R, Pancioli AM, Jauch EC, Menon AG, Deka R, Carrozzella JA, Moomaw CJ, Fontaine RN, Broderick JP. Genetic and environmental risk factors for intracerebral hemorrhage: preliminary results of a population-based study. Stroke. 2002; 33: 1190–1196; discussion 1196.
de Abajo FJ, Rodriguez LA, Montero D. Association between selective serotonin reuptake inhibitors and upper gastrointestinal bleeding: population based case-control study. Bmj. 1999; 319: 1106–1109.
Bak S, Tsiropoulos I, Kjaersgaard JO, Andersen M, Mellerup E, Hallas J, Garcia Rodriguez LA, Christensen K, Gaist D. Selective serotonin reuptake inhibitors and the risk of stroke: a population-based case-control study. Stroke. 2002; 33: 1465–1473.