|Home | About | Journals | Submit | Contact Us | Français|
The safety of thrombolytic therapy in patients with cocaine-associated acute ischemic stroke (CIS) is unknown.
We conducted a retrospective review of CIS patients who presented to our stroke center. Thrombolytic treatment was compared between cocaine-positive (n=29) and cocaine-negative (n=75) patients. We also compared CIS patients treated with t-PA versus those who did not receive t-PA (n=58). Safety outcomes were determined by the incidence of symptomatic intracerebral hemorrhage, in-hospital mortality, and mRS at hospital discharge.
There were no complications in t-PA treated CIS patients. Cocaine-positive and cocaine-negative treated patients had similar stroke severity and safety outcomes. CIS patients treated with t-PA had more severe strokes on baseline NIHSS but similar safety outcomes compared with non-treated CIS patients.
Thrombolytic therapy for CIS appears to be safe in this small study. Further research is needed to more definitively assess safety and efficacy of t-PA for CIS.
Cocaine causes acute cerebrovascular disorders such as ischemic stroke and intracranial hemorrhage1, 2. The potential mechanisms of acute ischemic stroke (AIS) associated with cocaine (CIS) include arrhythmias, cardiomyopathies, septic emboli from endocarditis in IV drug abusers, promotion of a pro-thrombotic state3, post-myocardial infarction related thromboembolism, vasospasm, and vasculitis.4 Several of these mechanisms may pose an increased risk for intracerebral hemorrhage (ICH) after IV t-PA. Cocaine causes acute hypertension which may also pose an increased risk of ICH after IV t-PA. Indeed, ICH has been reported to occur in patients with cocaine associated acute coronary syndromes treated with thrombolytics.5 Given these concerns, physicians are recommended to exercise caution in considering the use of IV-tPA for CIS6 in the absence of any published data. We therefore retrospectively analyzed a consecutive series of prospectively collected patients with CIS treated with t-PA, and compared their safety and outcomes to toxicology proven cocaine-negative patients who received t-PA within three hours of symptom onset. We also compared the safety and outcome of CIS patients treated with t-PA versus those who did not receive t-PA.
A retrospective chart review was performed of prospectively collected data of patients admitted to our stroke service during 2004–2007 who were identified as having recent cocaine exposure confirmed by a positive urine toxicology screen for the cocaine metabolite, benzolecgonine. Baseline demographics, admission vital signs, NIHSS scores, neuroimaging, creatinine kinase, troponin, and glucose were obtained. We also obtained IV t-PA use, hemorrhagic transformation, symptomatic ICH (sICH), discharge disposition, and modified Rankin scale scores (mRS) on discharge. Symptomatic ICH was defined as a parenchymal hematoma (PH-1 or PH-2) with an increase in the NIHSS of ≥4 points. Asymptomatic ICH was defined as any other hemorrhagic infarction. Favorable outcome was defined as discharge mRS≤2. Favorable disposition was defined as inpatient rehabilitation or home. We compared CIS patients who received IV t-PA within 3 hours of symptom onset versus patients who received IV t-PA during the same time period, but had no detectable cocaine metabolites on urine drug screen. Our center has no formal policy on selection of patients for toxicology screening. In addition, we compared CIS patients treated with IV t-PA with CIS patients who did not receive t-PA. For statistical analysis, categorical variables were compared using χ2 or Fisher exact tests where appropriate. Continuous variables were compared using 2-sample t tests or Mann–Whitney U tests where appropriate.
We identified 87 patients who had CIS (median age, 48; 71% were male and 69% were African-American). The median NIHSS at presentation was 7 (0–40). Favorable outcome was seen in 46 patients (57%).
IV t-PA was administered in 29 CIS patients (33.3%) and in 75 patients with a negative urine cocaine study (Table 1). Cocaine users were younger and more frequently male. Cocaine users had higher initial diastolic blood pressure and lower glucose levels. Cocaine users had higher CPK levels but there were no statistical differences in CK-MB or troponin. Patients in both groups had similar baseline NIHSS. There were no significant differences in safety or outcome (HT, sICH, mRS at discharge, discharge disposition or mortality) regardless of whether the patients used cocaine.
CIS patients treated with IV t-PA had significantly higher median NIHSS scores compared with those patients who did not receive t-PA (Table 2). Despite the increased stroke severity, the rates of sICH, mortality, favorable disposition, and favorable mRS at the time of discharge, were similar between t-PA and non-tPA treated groups. Finally, we also report stroke etiologies between cocaine positive and cocaine negative patients. We found a difference in the distribution of TOAST classifications among AIS patients based on cocaine exposure (Fisher's exact = 0.016).
Cocaine is a well-known precipitant of ischemic stroke1, 4 and there is concern cocaine may pose an enhanced risk for ICH after t-PA5. While the safety of t-PA in the setting of CIS is unknown, the short time window to administer t-PA within three hours of AIS onset may make determining the presence of cocaine unfeasible. We report the largest series of patients with CIS to our knowledge and treated with t-PA and find that there were no events of sICH after t-PA. Compared with cocaine-negative patients who received t-PA, mortality was lower in the cocaine-positive t-PA treated patients; however, this difference was not statistically significant. Both cocaine-positive and cocaine-negative treated patients had similar outcomes at hospital discharge. Diastolic blood pressures were higher in the cocaine-positive group compared to the cocaine-negative group. However, all patients treated with t-PA met blood pressure AHA guidelines for thrombolysis and did not experience acute spikes in blood pressure during or after t-PA infusion. Moreover, despite having more severe deficits, patients with CIS who received t-PA did not have significant differences in sICH or mortality compared with those CIS patients who did not receive t-PA. Interestingly, it has been suggested that long-term cognitive impairment in cocaine users is the result of diffuse cerebral small vessel disease7. Such patients might therefore be at increased risk for intracerebral hemorrhage; while we did find an increased incidence of small vessel disease in cocaine positive patients, we did not see an increased risk of sICH in this population.
This study has several limitations including its retrospective approach and small sample number. Although we report a large series of patients, this retrospective study is not powered to detect significant differences in sICH. Accordingly, the results must be interpreted with caution given that the outcomes of t-PA treated patients may be different if a substantially higher number of CIS patients were studied. There is a selection bias in designing a “cocaine-negative” patient group as not all patients on our stroke service undergo a toxicology screen, and older patients with vascular risk factors can have positive toxicology studies8. Finally, the standard urine toxicology screen for the cocaine metabolite used at our hospital tests positive for three days after cocaine use. Therefore, this study does not rule out the possibility that acutely intoxicated patients may have a worse outcome after t-PA administration.
In conclusion, the results from this study suggest that t-PA may be safe to administer in CIS and that early outcomes after CIS do not appear to be adversely affected by t-PA. Often, in an emergency department setting, physicians do not know if a patient with AIS was using cocaine because the toxicology screen may not return in time to consider IV t-PA. We provide preliminary evidence that IV t-PA could be used in this patient population, but further studies are needed to more definitively determine the safety and efficacy of IV t-PA for CIS patients.
Acknowledgements and Funding
This work was supported by AHA Award 0475008N (SIS), NIH training grant T32NS04712, and P50 NS044227.