Abstract WP346: HIAT2 Predicts Poor Functional Outcome, Palliative Care Involvement, and In-hospital Mortality in tPA Treated and Untreated Ischemic Stroke Patients

Abstract

Introduction: The HIAT2 (Houston Intra Arterial Therapy-2) score has been used to predict poor functional outcome in acute ischemic stroke (AIS) patients undergoing endovascular therapy (ET). Little is known about HIAT2’s ability to predict other outcomes in non-ET samples. Aim: To test the ability of HIAT2 to predict poor functional outcome (mRS 4-6), in-hospital mortality, and inpatient palliative care (PC) consult in (1) ET, (2) IV tPA, and (3) untreated patients. Methods: A retrospective review of consecutive AIS patients presenting to our comprehensive stroke center (CSC) from March 2014 to April 2015. Demographic and clinical data were collected. HIAT2 was calculated as follows: age (≤59=0, 60-79=2, ≥80=4), glucose (<150=0, ≥150=1), NIHSS (≤10=0, 11-20=1, ≥21=2), ASPECTS Score (8-10=0, ≤7=3). We used AUC to measure the ability of the HIAT2 score to predict our three outcomes of interest. Results: Among the 776 AIS patients admitted to our CSC, 6.6% received acute ET, 14.6% received IV tPA, and 79.8% received neither. For ET patients HIAT2 had an AUC of 0.592 for mRS 4-6, AUC 0.569 for PC, and AUC 0.656 for death. For tPA patients HIAT2 had an AUC of 0.686 for mRS 4-6, AUC 0.798 for PC, and AUC 0.825 for death. For untreated patients HIAT2 had an AUC of 0.629 for mRS 4-6, AUC 0.649 for PC, and AUC 0.641 for death. In the tPA treated sample, a HIAT2 score ≥4 had a sensitivity and specificity of 0.436 and 0.819 in predicting mRS 4-6, sensitivity and specificity of 0.667 and 0.782 for PC, and sensitivity and specificity of 0.733 and 0.806 for death. Discussion: Our results suggest that the variables used to create the HIAT2 score are useful in predicting poor outcomes in untreated, tPA treated, and ET patients. Despite its ability to predict poor outcome in these samples, treatment should not be withheld from patients that otherwise qualify.