Abstract TP72: Predictors of IV tPA Use among Acute Ischemic Stroke patients presenting to a Regional Telestroke Network: The Partners Telestroke Center Experience

Abstract

Intro: Remote neurological consultation via telemedicine (TeleStroke) increases rates of IV tPA use and has gained acceptance through demonstrated reliability and similar outcomes to conventional tPA use. We sought to determine factors that predict tPA use among patients presenting to hospitals in our TeleStroke network. Methods: Using the Partners TeleStroke Network database, we analyzed all acute ischemic stroke (AIS) patients (n= 2,140/at 31 spoke hospitals) and the subset arriving ≤ 4.5 hr since last known well (LKW). All covariates associated with IV tPA use in univariate analysis at p-value <0.10 with less than 10% missing were included in multivariable logistic regression. Since NIHSS was only recorded for video consults, it was excluded from multivariable models. Results: Eighty percent (1720/2140) of all AIS patients that presented to spoke hospitals arrived ≤4.5 hr of LKW, and 38.1% (655/1720) of these received IV tPA. All patients who were not treated with tPA had a documented exclusion. Patients receiving tPA were more likely to be white, have higher NIHSS, arrive earlier from LKW time and trended toward younger, as compared to non-treated patients. They were also more likely to be evaluated by TeleStroke videoconferencing than telephone (Table 1). In multivariable analysis, decreasing LKW to arrival time [OR 1.01 per min, 95% CI 1.02-1.01] and use of videoconsultation [OR 3.64, 95% CI 2.65 - 5.00] were independently associated with rates of IV tPA use. Conclusion: Consistent with prior reports of conventional, front-door tPA use, age, stroke severity (NIHSS) and stroke duration (LKW to arrival time) also predict tPA use in TeleStroke environments. Use of videoconsultation is a novel independent predictor of tPA use, potentially due to increased confidence by providers in recommending treatment at a distance compared with the telephone alone. Replication of these findings in other independent telestroke network cohorts is recommended.