Post-stroke acute heart failure in patients with large vessel occlusion undergoing endovascular treatment: A pooled analysis of individual patient data from multicenter studies with mediation analysis.

内容摘要

Cardiac complications rank among the leading contributors to poor outcomes in ischemic stroke patients along with the neurological impairment, while the risk stratification and prognostic significance of post-stroke acute heart failure (PSHF) remain poorly characterized. This study aimed to investigate the incidence, predictors, and impacts of PSHF in patients with large vessel occlusion stroke (LVO) undergoing endovascular treatment (EVT). Given that cardioembolic stroke inherently involves underlying cardiac pathology, a secondary aim was to test whether the effect of stroke severity on PSHF was modified by cardioembolic etiology and whether PSHF mediated the effect of stroke severity on functional outcome in these patients. In a pooled analysis of individual patient data from four multicenter prospective studies conducted in China between January 2014 and June 2023, we included 3,415 patients with LVO who underwent EVT. The primary outcome was very poor functional outcome, defined as 90-day modified Rankin Scale (mRS) 5-6. Multivariable regression models, interaction testing, and mediation analysis were used, with adjustment for clinically relevant covariates including demographic characteristics, vascular risk factors, baseline stroke severity, imaging characteristics, and treatment-related factors. PSHF developed in 278 patients (8.14%), with its incidence reaching peak at 1 day after stroke onset. PSHF was significantly associated with a higher rate of very poor outcome (62.23% versus 31.08%, adjusted odds ratio (aOR) 3.09, 95% confidence interval (CI) [2.25, 4.24]). A significant interaction was observed between cardioembolism and the baseline National Institutes of Health Stroke Scale (NIHSS) score (p for interaction = 0.016). Moderate-to-severe stroke significantly increased the risk of PSHF in patients with cardioembolic stroke (aOR 1.91, 95% CI [1.28, 2.87]), but not in those with non-cardioembolic stroke (aOR 0.97, 95% CI [0.81, 1.82]). Mediation analysis showed that PSHF mediated 7.70% (95% CI [2.40, 12.40]) of the effect of moderate‑to‑severe stroke on very poor outcome among cardioembolic patients. The main methodological limitations were the pooled design using studies with different protocols and the potential for residual unmeasured confounding. PSHF was significantly associated with very poor outcome in LVO patients undergoing EVT. Moderate-to-severe cardioembolic LVO substantially elevated the risk of PSHF, with PSHF partially mediating the adverse prognostic impact of stroke severity. Early risk assessment and monitoring for PSHF may optimize management in this high-risk population.