Stroke Outcomes Research

About the program

Acute stroke research depends on data that are complete, consistent, and collected the same way at every participating hospital. This program applies data-management and biostatistical methods to multi-center stroke trials and registries at UTHealth Houston, where Dr. Hessabi has worked with the Biostatistics/Epidemiology/Research Design (BERD) Core since 2009. His roles on these studies have centered on the data: building and maintaining trial databases, monitoring and cleaning study data, and co-authoring the methods and results papers that come out of that work.

The earliest work came through the University of Texas Specialized Program in Acute Stroke (UT SPOTRIAS). For ARTSS, a pilot safety study of argatroban given with tPA, and for CLOTBUST-HF, a pilot study of an operator-independent transcranial ultrasound device, he developed Microsoft Access databases and performed data cleaning. He served as Data Monitor for ARTSS-2, a randomized, multi-center trial, and is a co-author of the 2015 Contemporary Clinical Trials paper describing its design and data coordination. He is also a co-author of a 2013 BMC Neurology report on quality assurance in the University of Texas Houston Stroke Registry, which examined how well abstracted registry data matched an expert-reviewed reference standard.

From 2019 to 2023 he was Data Manager for SELECT2, a randomized trial of endovascular thrombectomy in patients whose strokes had already produced a large ischemic core, and developed the trial's REDCap database. SELECT2 enrolled 352 patients at 31 centers in the United States, Canada, Europe, Australia, and New Zealand before it was stopped early for efficacy. Its primary results appeared in the New England Journal of Medicine in 2023, followed in 2024 by one-year outcomes in The Lancet and a prespecified analysis of transferred patients in JAMA Neurology. A related systematic review and meta-analysis in Annals of Clinical and Translational Neurology, on which he is a co-author, looks at stroke from the population side by asking why stroke occurs at younger ages in low- and middle-income countries.

The findings below summarize what each paper reported. They describe results in study populations and are not guidance for any individual treatment decision.

Key findings

1. 01

   In SELECT2, endovascular thrombectomy plus medical care led to better 90-day functional outcomes on the modified Rankin Scale than medical care alone in patients with large ischemic strokes (generalized odds ratio 1.51), and 20% of thrombectomy patients reached functional independence compared with 7% of the medical-care group. Mortality was similar between groups, and thrombectomy carried vascular complications such as arterial dissection and vessel perforation (New England Journal of Medicine, 2023).

2. 02

   At one year, the SELECT2 benefit persisted: thrombectomy still improved the distribution of modified Rankin Scale scores (generalized odds ratio 1.43). By then 45% of patients in the thrombectomy group had died compared with 52% in the medical-care group, a difference that was not statistically significant (The Lancet, 2024).

3. 03

   A prespecified SELECT2 analysis found that thrombectomy benefited both patients who arrived directly at a thrombectomy-capable center and patients transferred from another hospital, with no statistically significant difference in treatment effect between the two groups. The benefit was numerically larger in patients who presented directly, and longer transfer times and progression of ischemic change on imaging were associated with worse outcomes (JAMA Neurology, 2024).

4. 04

   A meta-analysis of 67 studies found that the mean age at stroke was lower in low- and middle-income countries than in high-income countries (63.1 versus 68.6 years). After adjustment for the Healthcare Access and Quality index, that difference was no longer statistically significant except among women, pointing to healthcare access and quality as a major contributor (Annals of Clinical and Translational Neurology, 2022).

5. 05

   In the University of Texas Houston Stroke Registry, the error rate in abstracted records fell from 4.8% in 2008 to 2.2% in 2011 as enhanced data quality assurance procedures were established, and agreement between registry data and the expert-reviewed reference standard was excellent for almost all variables checked (BMC Neurology, 2013).

6. 06

   In the CLOTBUST-HF pilot safety study of 20 patients, an operator-independent ultrasound device used alongside intravenous tPA was tolerated by all patients, none developed symptomatic intracerebral hemorrhage, and 40% of patients had complete recanalization of the blocked artery at two hours. The authors concluded that the approach warranted evaluation in a phase III efficacy trial (Stroke, 2013).

7. 07

   The ARTSS-2 methods paper reported that a covariate-adaptive randomization program kept the trial's three study arms balanced on study site, hemorrhage after thrombolysis (HAT) score, and distal internal carotid artery occlusion across the first 90 patients enrolled, and that the partly Bayesian safety stopping thresholds for both argatroban doses were not crossed (Contemporary Clinical Trials, 2015).

Collaborators

• Dr. Amrou Sarraj (UTHealth / University Hospitals Cleveland)
• Dr. Andrew Barreto (UTHealth)
• Dr. Mohammad H. Rahbar (UTHealth)
• The SELECT2 Investigators

Publications in this program

1. 2024·JAMA Neurology·22 citations

   Endovascular thrombectomy treatment effect in direct vs transferred patients with large ischemic strokes: a prespecified analysis of the SELECT2 trial

   Sarraj A, Hill MD, Hussain MS, Abraham MG, Ortega-Gutierrez S, Chen M, et al.

2. 2023·New England Journal of Medicine·520 citations

   Trial of endovascular thrombectomy for large ischemic strokes

   Sarraj A, Hassan AE, Abraham MG, Ortega-Gutierrez S, Kasner SE, Hussain MS, et al.

3. 2022·Annals of Clinical and Translational Neurology·82 citations

   Younger age of stroke in low-middle income countries is related to healthcare access and quality

   Rahbar MH, Medrano M, Diaz-Garelli F, Gonzalez Villaman C, Saroukhani S, Kim S, et al.