Volume 14 Supplement 1

2nd Clinical Trials Methodology Conference: Methodology Matters

Open Access

The risks and benefits of RTPA in acute ischemic stroke for patients at high risk of intracranial haemorrhage and poor functional outcome: a secondary analysis of the IST-3 trial and systematic review of prediction models

  • Douglas Thompson1,
  • Gordon Murray1 and
  • William Whiteley1, 2
Trials201314(Suppl 1):O9


Published: 29 November 2013


Treating acute ischaemic stroke patients with iv-rtPA is of overall benefit. An associated increase in the risk of symptomatic intracranial haemorrhage (SICH) may cause considerable harm. We investigated (i) whether novel or existing prediction models could predict SICH or poor functional outcome in rtPA treated patients and (ii) whether ischemic stroke patients at either a high predicted risk of SICH or poor functional outcome experienced less benefit from rtPA.


We used the IST-3 trial data, an international, multicentre, open treatment randomised trial of rtPA versus control in 3035 acute ischemic stroke patients. We developed and internally evaluated a multivariate logistic regression model for SICH following rtPA including variables identified as important in a systematic review. We compared the discrimination (area under the receiver operating characteristic curve (AUROCC)) of our model with those existing in the medical literature. We calculated the absolute risk reduction of death or dependency with rtPA in patients at a low, medium or high predicted risk of SICH or poor functional outcome with each model.


Our model had similar discrimination for SICH (AUROCC 0.68 95% CI: 0.63-0.74) to nine previously developed models (HAT, SEDAN, SITS, GRASPS, SPAN-100, Stroke-TPI, DRAGON, THRIVE and a model with NIHSS and age). There was no evidence that patients at high predicted risk of SICH or poor functional outcome after stroke derived less benefit from rtPA.


We found no evidence to support a stratified approach in administering rtPA to acute ischaemic stroke patients at a high predicted risk of intracranial haemorrhage or poor functional outcome.

Authors’ Affiliations

Edinburgh MRC Hub for Trials Methodology Research, University of Edinburgh
Division of Clinical Neurosciences, University of Edinburgh, Western General Hospital


© Thompson et al; licensee BioMed Central Ltd. 2013

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.