Abstract
Abstract only
Background:
Tools to predict outcome after endovascular reperfusion therapy (ERT) for acute ischemic stroke (AIS) have previously included only pre-treatment variables. We sought to derive and validate an outcome prediction score based on readily available pre-treatment and treatment factors.
Methods:
The derivation cohort consisted of 516 patients with anterior circulation AIS from 9 centers from September 2009-July 2011. The validation cohort consisted of 110 patients with anterior circulation AIS from the Penumbra Pivotal Trial. Multivariable logistic regression identified predictors of good outcome, defined as a modified Rankin Score (mRS) of < 2, in the derivation sample; model beta coefficients were used to assign point scores. Discrimination was tested using C-statistics. We then validated the score in the Penumbra cohort and performed calibration (predicted versus observed good outcome) in both cohorts.
Results:
Good outcome at 3 months was noted in 189 (36.8%) patients in the derivation cohort. The independent predictors of good outcome were
A
ge (2 pts: <60; 1 pt: 60-79; 0 pts: >79),
N
IHSS score (4 pts: 0-10; 2 pts: 11-20; 0 pts: > 20),
L
ocation of clot (2 pts: M2; 1 pt: M1; 0 pts: ICA),
R
ecanalization (5 pts: TICI 2 or 3), and
S
ymptomatic hemorrhage (2 pts: none, HT1-2, or PH1; 0 pts: PH2). The outcome (SNARL) score demonstrated good discrimination in the derivation cohort (C-statistic 0.78, 95% CI 0.72-0.78) and validation cohort (C-statistic 0.74, 95% CI 0.64-0.84). There was excellent calibration in each cohort (Figure).
Conclusions:
The SNARL score is a validated tool to determine the probability of functional recovery among AIS treated with endovascular reperfusion strategies. Unlike previous scores that did not include treatment factors such as successful recanalization or hemorrhagic complications, our score can be applied to patients after treatment and may provide guidance to physicians, patients, and families about expected functional outcome.