Abstract TP384: Recovery Trajectories in Acute Non-fatal Stroke
Introduction: Traditional approaches to the study of post stroke recovery assume that recovery follows a homogenous path, with deviations from the average explained in terms of socio-demographic and clinical differences. Latent class analysis (LCA) provides a powerful tool to investigate potential heterogeneity in post-stroke recovery pathways.
Aims: We aimed to 1) explore recovery trajectories in acute stroke over 3 months using LCA and 2) identify predictors of poor recovery.
Methods: Data from 202 patients admitted to Stroke Care Unit at the Royal Melbourne Hospital (Australia) between January and June 2011 were analyzed. The main outcome was disability (modified Rankin Score) pre-stroke, at discharge, and 3 months post stroke. Predictors of poor recovery were assessed using logistic regression analysis.
Results: Patients’ mean (SD) age was 68.6 (14.1), 58% were males and 82% had ischemic strokes. LCA identified 6 mutually exclusive recovery trajectories (see Figure 1), including: 1) No disability at onset, no impact from stroke (n=43); 2) Substantial disability, no impact from stroke (n=20); 3) Some disability, mild impact, full recovery (n=36); 4) No disability, mild impact, partial recovery (n=26); 5) No disability, high impact, partial recovery (n=50); 6) No disability, high impact, no recovery (n=27). Compared with those who made at least partial recovery (trajectories 1-5), those who made no recovery at 3 months were significantly younger (OR 0.97 95%CI 0.95,0.99) and more likely to be smokers (OR 2.48 95%CI 1.04,5.90). History of atrial fibrillation, ischemic heart disease, previous stroke, and stroke type were not associated with poor recovery (p≥0.103).
Conclusions: This study identifies stroke recovery pathways, providing a potentially useful tool for identifying individuals at risk of poor recovery and in need of targeted intervention.
- © 2012 by American Heart Association, Inc.