AbstractBackground and Purpose—
We performed a meta-analysis to assess whether leukoaraiosis on brain computed tomographic scans of acute ischemic stroke patients treated with intravenous thrombolysis is associated with an increased risk of symptomatic intracerebral hemorrhage (sICH) or poor functional outcome at 3 to 6 months after stroke, or both.Methods—
We searched PubMed and pooled relevant data in meta-analyses using random effects models. Using odds ratios (OR), we quantified the strength of association between the presence and severity of leukoaraiosis and post-thrombolysis sICH or 3- to 6-month modified Rankin Score >2.Results—
Eleven eligible studies (n=7194) were pooled in meta-analysis. The risk of sICH was higher in patients with leukoaraiosis (OR, 1.55; 95% confidence interval [CI], 1.17–2.06; P=0.002) and severe leukoaraiosis (OR, 2.53; 95% CI, 1.92–3.34; P<0.0001) compared with patients without leukoaraiosis. Leukoaraiosis was an independent predictor of sICH in 6 included studies (n=4976; adjusted OR, 1.75; 95% CI, 1.35–2.27; P<0.0001). OR for leukoaraiosis and poor 3- to 6-month outcome was 2.02 (95% CI, 1.54–2.65; P<0.0001), with significant statistical heterogeneity (I2, 75.7%; P=0.002). In adjusted analyses, leukoaraiosis was an independent predictor of poor outcome (n=3688; adjusted OR, 1.61; 95% CI, 1.44–1.79; P<0.0001). In post hoc analyses, including only leukoaraiosis patients in randomized controlled trials (IST-3 [third International Stroke Trial], NINDS [National Institute of Neurological Disorders and Stroke], ECASS-1–2 [European Cooperative Acute Stroke Study]; n=2234), tissue-type plasminogen activator versus control was associated with higher sICH risk (OR, 5.50; 95% CI, 2.49–12.13), but lower poor outcome risk (OR, 0.75; 95% CI, 0.60–0.95).Conclusions—
Leukoaraiosis might increase post-intravenous thrombolysis sICH risk and poor outcome poststroke. Despite increased sICH risk, intravenous tissue-type plasminogen activator treatment has net clinical benefit in patients with leukoaraiosis. Given the risk of bias/confounding, these results should be considered hypothesis-generating and do not justify withholding intravenous thrombolysis.