Matthew Edwardson, MD

Whiteley WN, Thompson D, Murray G, Cohen G, Lindley RI, Wardlaw J, and Sandercock P. Targeting Recombinant Tissue-Type Plasminogen Activator in Acute Ischemic Stroke Based on Risk of Intracranial Hemorrhage or Poor Functional Outcome: An Analysis of the Third International Stroke Trial. Stroke. 2014


Given the extensive data now available from multiple clinical trials of IV-TPA, one might expect to find clinical factors predisposing patients to harm versus benefit that could help guide treatment decisions. Whiteley and colleagues explored this question by analyzing data from the IST-3 trial. They were able to predict which subjects were most likely to experience symptomatic ICH (sICH) or poor functional outcome. Surprisingly, the patients at highest risk for sICH or poor functional outcome also received the most benefit from IV-TPA.



The authors used logistic regression to develop a new model to predict sICH and poor functional outcome in subjects from the IST-3 trial and compared this new model to several other prediction models validated on prior datasets. Nearly all of the models showed at least moderate ability to predict sICH and poor functional outcome. A very simplified model that included only age and NIHSS was equally effective at prediction. The models were stratified into mild, moderate, and severe risk of sICH or poor functional outcome. Paradoxically, the authors found that IV-TPA was most beneficial for those subjects who fell into the severe categories.

This study reinforces lessons learned from IST-3 and provides insights into our lack of a precise predictor for those who will develop sICH. IST-3 made the case for eliminating advanced age and high NIHSS as contraindications to IV-TPA. The current study puts these findings into perspective. The very factors that predict a higher rate of sICH and poor functional outcome also predict less long term disability in response to treatment with IV-TPA. Of course those with large vessel occlusion are more likely to benefit from IV-TPA and more likely to have sICH, which may explain these counterintuitive results. The current study also highlights the need to identify a much more specific prediction marker for sICH. Using current sICH markers to decide when to withhold TPA does more harm than good. While awaiting a better biomarker for sICH, we should forge ahead and not exclude patients from IV-TPA simply because of advanced age or high NIHSS.