Age demonstrates limited predictive utility for functional outcomes after thrombectomy in patients aged ≥70 years...

The clinical factor impact on outcomes after endovascular thrombectomy (EVT) in patients aged ≥70 years remains incompletely understood. We aimed to identify predictors of good outcomes after EVT in patients aged ≥70 years, defined as a modified Rankin Scale (mRS) score of 0-2.

This retrospective, single-center cohort study included 94 patients aged ≥70 years with acute ischemic stroke who underwent EVT. Participants were stratified into septuagenarians (n = 44) and octo/nonagenarians (n = 50). We evaluated post-EVT modified thrombolysis in cerebral infarction reperfusion grade, symptomatic intracerebral hemorrhage, and mRS score at 3 months follow-up as outcomes. Both multivariable (LR) and Bayesian logistic regression (BLR) and sensitivity analyses were conducted to derive adjusted odds ratio (aOR) and assess the probabilistic associations between clinical variables and outcomes.

At presentation, the median ischemic core was higher in octo/nonagenarians compared to septuagenarians (20 mL vs. 4 mL, p = 0.0464); median Alberta Stroke Program Early CT Score was lower (7 vs. 8, p = 0.0112). Higher Fazekas grades of leukoaraiosis were more frequent in octo/nonagenarians (p = 0.0297) than in septuagenarians. Good mRS outcomes were achieved by 27.3% of septuagenarians vs. 8.0% of octo/nonagenarians (p = 0.0274). In the multivariable LR, age was not an independent predictor of poor outcomes (aOR 2.40; 95% CI, 0.65-10.08; p = 0.1991). BLR identified higher National Institutes of Health Stroke Scale scores [odds ratio (OR) 0.90; 95% credible interval, 0.81-0.98] associated with poorer outcomes, whereas prior intravenous thrombolysis (OR 6.59; 1.16-23.09) predicted better outcomes. BLR did not show probabilistic certainty of age as a predictor of functional outcomes. Including infarct core in the model did not impact sensitivity analysis results.

Age was not independently associated with functional outcomes. Age-related differences in outcomes may be mediated by initial stroke characteristics rather than age.

ClinicalTrials.gov, identifier NCT06953427.