Abstract WP311: Estimating Supported Outcomes in Intracerebral Hemorrhage Patients with Withdrawal of Life Support through Propensity Score Matching
Background: Observational reports have suggested that a self-fulfilling prognostic pessimism may lead to withdrawal of life support (WOLS) in more severe intracerebral hemorrhage patients (ICHs) who might otherwise have acceptable clinical outcomes. Our objective was to estimate outcomes in a cohort of ICHs who had WOLS, if instead they had life support continued.
Methods: Multivariable modeling was used to create a propensity score (PS) for WOLS in a Boston single center cohort of ICHs with follow up modified Rankin Scale (mRS) or known death at 6 months. Individual ICHs with WOLS were then matched to individual ICHs without WOLS and baseline variables and outcomes compared.
Results: There were 576 ICHs with all variables available for analysis; mean age was 72 years, 44% women, 83% white and WOLS in 45%. Factors used to create the PS for WOLS included age, GCS, ICH and IVH volume, diabetes, ICH location, SBP, sex, race, history of HTN, DM, CAD, atrial fibrillation, warfarin, intubation, and EVD. WOLS/non-WOLS matches were possible for 81 pairs (N=162 total), and the groups were well matched on all PS variables; key values included mean age (73 years), GCS (9.5), ICH volume (37cc) and % intubation (57) and EVD use (23); more patients had dementia in the WOLS group than non-WOLS (24 vs. 11%, p=0.03). Follow up mRS in the WOLS group was 3 in 1.2%, 5 in 2.5% and 6 (death) in 96%; in the matched non-WOLS group, % in mRS scores 0-6 were 4.9, 8.6, 3.7, 24.7, 25.9, 16.0 and 16.0 respectively (p<0.0001 for difference in distribution). Using mRS <= 3 as an acceptable outcome, % in WOLS/non-WOLS were 1.2%/42% (difference p<0.0001).
Conclusions: The proportion of ICHs with WOLS that, without WOLS, might have had an acceptable clinical outcome was estimated to be 42%, suggesting that for every 2-3 less ICHs with WOLS, one more patient might have an acceptable clinical outcome. This proportion might vary depending on an individual patient’s interpretation of what constitutes an acceptable outcome.
- © 2012 by American Heart Association, Inc.