Cannabis Use and Outcomes in Patients With Aneurysmal Subarachnoid Hemorrhage
Background and Purpose—The incidence of cannabis use in patients with aneurysmal subarachnoid hemorrhage (aSAH) and its impact on morbidity, mortality, and outcomes are unknown. Our objective was to evaluate the relationship between cannabis use and outcomes in patients with aSAH.
Methods—Records of consecutive patients admitted with aSAH between 2010 and 2015 were reviewed. Clinical features and outcomes of aSAH patients with negative urine drug screen and cannabinoids-positive (CB+) were compared. Regression analyses were used to assess for associations.
Results—The study group consisted of 108 patients; 25.9% with CB+. Delayed cerebral ischemia was diagnosed in 50% of CB+ and 23.8% of urine drug screen negative patients (P=0.01). CB+ was independently associated with development of delayed cerebral ischemia (odds ratio, 2.68; 95% confidence interval, 1.03–6.99; P=0.01). A significantly higher number of CB+ than urine drug screen negative patients had poor outcome (35.7% versus 13.8%; P=0.01). In univariate analysis, CB+ was associated with the composite end point of hospital mortality/severe disability (odds ratio, 2.93; 95% confidence interval, 1.07–8.01; P=0.04). However, after adjusting for other predictors, this effect was no longer significant.
Conclusions—We offer preliminary data that CB+ is independently associated with delayed cerebral ischemia and possibly poor outcome in patients with aSAH. Our findings add to the growing evidence on the association of cannabis with cerebrovascular risk.
Cannabis is the most widely used illicit drug in the world.1 During the past few years, several states in the United States and many European countries have legalized cannabis for medicinal or recreational consumption. There are reports, however, that this drug may not be as safe as commonly perceived. Mounting evidence suggests that cannabis consumption may place users—particularly younger individuals—at increased risk for cerebrovascular events.2,3 The incidence of cannabis use in patients with aneurysmal subarachnoid hemorrhage (SAH) and its impact on outcomes are unknown. In this study, we sought to investigate these questions.
Study Design and Subjects
This study was approved by the Institutional Review Board. Using an institutional repository, we reviewed the data on consecutive patients aged ≥18 years admitted with the diagnosis of aneurysmal SAH from January 2010 through October 2015. SAH was defined as sudden headache with or without coma or focal deficit, and presence of subarachnoid blood, confirmed by computed tomographic scan. We excluded SAH cases associated with trauma, arteriovenous malformations, vasculitis/vasculopathy, mycotic aneurysms, infectious/neoplastic lesions, and angiogram-negative SAH. The following admission data were collected: age, sex, urine drug screen (UDS), Fisher and Hunt/Hess grades, location and the largest aneurysm diameter, and presenting mean arterial pressure. Other clinical data registered included trough serum sodium during hospitalization, trough mean arterial pressure during the first 3 admission days, and occurrence of delayed cerebral ischemia (DCI). The criteria for DCI were new focal neurological signs or deterioration of the level of consciousness ≥48 hours of presentation with no evidence of worsening hemorrhage, plus radiological evidence of cerebral infarction; with or without angiographic evidence of vasospasm. UDS was performed using routine laboratory-based panel for detection of common substances of abuse, with sensitivity of 50 ng/mL threshold for cannabinoids detection. Primary end points were development of DCI, in-hospital mortality, and modified Rankin Scale score on outpatient follow-up. Patients who had inadequate follow-up information documented were excluded from outcome analysis, but not from determining DCI association. All patients at our institution were treated in accordance with the American Heart Association guidelines.4
Continuous variables were reported as means±SD when distribution was normal, and medians with interquartile range when not normally distributed. Categorical variables were analyzed using χ2 and Fisher exact tests, when appropriate. Non-normally distributed variables were analyzed using the Mann–Whitney U test. To avoid the confounding effects of other drugs, patients with UDS positive for other illicit substances, including combination of cannabinoids and other drugs, were excluded. We compared baseline features of cannabinoids-positive (CB+) patients to those with negative UDS (UDS−).
Prespecified dichotomization of the modified Rankin Scale score was used, with poor outcome defined as a composite of in-hospital death and severe disability (modified Rankin Scale scores, 4 and 5) on follow-up. The potential impact of each admission characteristic on the development of DCI and poor outcome was analyzed in univariate and multivariate regression models.
During the study period, 213 patients were admitted with the diagnosis of aneurysmal SAH. Presentation UDS was performed in 135; mean age 42.8±15.4 and men comprising 41.4%. There was no difference in age between patients with and without UDS checked (Table I in the online-only Data Supplement). Posterior circulation aneurysms comprised 24%, and 3% had multiple aneurysms. UDS was negative in 59.3%, CB+ in 25.9%, and in the rest, positive for other toxins (including combination of cannabinoids and other drugs). Baseline characteristics between CB+ and UDS− patients were no different, except that CB+ patients presented with significantly higher Hunt/Hess grades (P=0.02; Table II in the online-only Data Supplement).
Fifty percent of CB+ and 23.8% of UDS− patients suffered DCI (P=0.01). There was no difference in the mean number of days from admission to diagnosis of DCI. In univariate analysis, Fisher grade, Hunt/Hess, and CB+ were significantly associated with occurrence of DCI. Multivariate analysis using these predictors showed an independent association of CB+ with DCI (Table 1).
Although in-hospital mortality rate was higher in CB+ than UDS− patients, this did not reach statistical significance (14.3% versus 3.8%; P=0.052). Eighty-nine patients who remained alive had follow-up data, with average follow-up latency from discharge of 58.7±25 days. A significantly larger number of CB+ patients met the end point of poor outcome, compared with UDS− patients (35.7% versus 13.8%; P=0.01). Univariate analysis showed significant correlation of Hunt/Hess, DCI, and CB+ with poor outcome (Table 2). However, the association of CB+ was no longer significant when these 3 characteristics were analyzed via multiple regression.
This study showed that recent cannabis use, evident by CB+ UDS on admission, is associated with increased risk of DCI and possibly poor outcome in patients with aneurysmal SAH. To our knowledge, this is the first study to investigate and ascertain this relationship.
We found that CB+ patients had more severe symptoms (Hunt/Hess) than UDS−. We also determined higher rate of poor outcome in CB+ patients, compared with individuals with UDS−. However, CB+ status did not independently predict poor outcome or death. The adverse impact of CB+ on outcomes found in univariate analysis could be through its association with DCI. This was illustrated by DCI—not CB+—remaining an independent predictor for poor outcome in multivariate analysis (Table 2).
The higher prevalence of DCI associated with CB+ in our patients may be explained by the evidence indicating increased risk of ischemic stroke in cannabis users, particularly young adults.5 Δ9-tetrahydrocannabinol, the main psychoactive component of cannabis, increases oxidative stress and induces cerebral mitochondrial dysfunction.5 Cannabis may additionally cause reversible cerebral vasoconstriction syndrome.6 In a series of young patients with stroke, multifocal intracranial stenosis associated with cannabis use was reported in 21%, with reversibility of the stenotic foci after cannabis cessation.7 Studies using transcranial Doppler have demonstrated significantly elevated pulsatility indices, suggesting cerebrovascular resistance, and increased systolic velocities in marijuana users, compared with controls.8 In heavy users, these increases persisted after a month of monitored abstinence.8 Elevated cerebrovascular resistance may increase susceptibility to DCI.9
Several limitations to our study should be noted. Although our institution strictly adheres to standard guidelines, variability in management by different providers and guideline evolution of over time may have caused a lack of uniformity in patient care. Overall, however, we did not feel that these factors significantly affected our results. Retrospective analysis limited our ability to detect and incorporate other potentially relevant clinical features, including cardiopulmonary comorbidities, infections, recurrent aneurysmal rupture, hydrocephalus, and seizures. Four patients developed hydrocephalus, 3 suffered recurrent aneurysm rupture, and 6 had seizures. None of these patients died or were left with severe disability. Omission of patients with UDS positive for other drugs reduced our cohort size, but this was done to maintain purity of the control group. The decisions to obtain UDS were not uniform and may have been influenced by unmeasured factors not included in the multivariable model.
We also could not differentiate new cannabis use from residual drug excretion, considering that cannabinoids may persist in the urine for several days/weeks. Therefore, establishing true temporal relationship was difficult. The same is true with regard to extricating chronic use from single-episode consumption. Finally, the possibility of false-positive/negative UDS could not be completely eliminated.
In conclusion, these preliminary findings append the growing evidence that cannabis use is associated with cerebrovascular risk. A larger body of data are needed to expand on these findings and determine if cannabis use is indeed associated with poor outcome.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.116.013099/-/DC1.
- Received February 10, 2016.
- Revision received March 2, 2016.
- Accepted March 3, 2016.
- © 2016 American Heart Association, Inc.
- 1.↵Prevalence of Drug Use among the General Population. World Drug Report 2012. United Nations Office on Drugs and Crime. http://www.unodc.org/unodc/en/data-and-analysis/WDR-2012.html. Accessed January 3, 2016.
- Hackam DG
- Connolly ES Jr,
- Rabinstein AA,
- Carhuapoma JR,
- Derdeyn CP,
- Dion J,
- Higashida RT,
- et al
- Wolff V,
- Schlagowski AI,
- Rouyer O,
- Charles AL,
- Singh F,
- Auger C,
- et al
- Ducros A,
- Boukobza M,
- Porcher R,
- Sarov M,
- Valade D,
- Bousser MG
- Wolff V,
- Lauer V,
- Rouyer O,
- Sellal F,
- Meyer N,
- Raul JS,
- et al