Illicit drug use and cerebral microbleeds in stroke and transient ischemic attack patients

Abstract

Background: Cerebral microbleeds (CMB) signal cerebral small vessel disease and are associated with ischemic stroke (IS) incidence, recurrence, and complications. While illicit drug use (IDU) is associated with cerebral small vessel disease, the association between CMB and IDU is understudied. We sought to delineate differences in vascular risk factors between IDU and CMB and determine the effect of this relationship on outcomes in IS/transient ischemic attack (TIA) patients.

Methods: We included 2001 consecutive IS and TIA patients (years 2009-2018) with a readable T2*gradient-echo MRI sequence. CMB rating followed standardized guidelines and CMB were grouped topographically into lobar, deep or infratentorial. IDU data (history and/or urine toxicology) was available for 1746 patients. The adverse composite outcome included pneumonia, urinary tract infection, deep venous thrombosis or death during hospitalization. Good functional outcome was defined as modified Rankin scale score < 3 and ambulatory on discharge. Univariate analysis was used to assess vascular risk factors and multivariable logistic regression was used to characterize the IDU/CMB relationship on outcomes.

Results: We observed IDU in 13.8 % (n=241), and CMB in 32.9% (n=575, 53.8% lobar, 27.3% deep and 18.8% infratentorial). Patients with IDU and at least one CMB were older (53.6±10.5 vs. 56.9±11.5, p=0.04), had a lower BMI (28.1±5.9 vs. 26.6±4.4, p=0.04), and were more likely to have had a previous IS/TIA (25.1% vs. 41.9%, p=0.01). IDU trended higher for those with severe CMB (10+) compared with those without CMB and 1-9 CMB (25% [n=9] vs 14.3% [n=1171] and 12.1% [n=65] respectively; p=0.07) without individual drug deviations from this pattern. Adverse and good functional outcomes were observed in 177 and 905 total patients, respectively. No significant interaction was observed between IDU and CMB with either adverse or functional composite outcomes.

Conclusion: IDU prevalence was high in our urban study population, and showed a borderline association with increasing CMB burden. Patients with CMB and IDU history were older and more likely to have had a previous IS/TIA. Further studies are required to clarify the clinical consequences related to the relationship between IDU and CMB.