Introduction: Reliable early stroke prognostication has important implications for goals of care planning, treatment guidance, and disposition planning. At present, prediction of patient long term functional outcomes remains challenging.
Hypothesis: Infarct volumes on early non-contrast CT may be a useful tool in predicting long term stroke functional outcomes.
Methods: Non-contrast CT images from 557 patients in the ALIAS2 trial were collected. Infarct and hemorrhage volumes at 24hrs post treatment were measured by 4 investigators blinded to patient clinical information. Abnormalities were outlined manually on each image slice and final volumes were calculated using Quantamo 1.0, a volumetric analysis tool. The location and volume of the acute abnormalities were correlated with modified Rankin Scales (mRS) scores at 90-day follow-up using Pearson correlation coefficients.
Results: There is a strong correlation between infarct volumes measured on CT at 24 hours with functional outcomes measured at 90 days r = 0.465 with every 18.6ml of brain infarct contributing to 1 point increase in mRS. Infarct volumes also correlated with number of hospital stays, patient death, and initial assessments using Alberta Stroke Program Early CT Score r= 0.211, 0.395 and -0.377 respectively. Acute hemorrhage volume correlated strongly with poor functional outcomes with 8.2ml of hemorrhage corresponding to 1 point increase in mRS r =0.468. A comparison of patients treated with placebo vs. high dose albumin showed no statistically significant difference in infarct volumes.
Conclusions: Measurement of brain infarction on CT at 24 hours post treatment can predict long-term functional outcomes and maybe a useful tool in guiding management.