By Alexander E. Merkler, MD

Assistant Professor of Neurology, Weill Cornell Medical College

Dr. Merkler reports no financial relationships relevant to this field of study.

SYNOPSIS: In a Phase II, randomized, multicenter prospective trial, intravenous glyburide failed to improve outcomes in patients with large hemispheric infarction, although there was a reduction in neuroimaging and biomarkers of cerebral edema.

SOURCE: Sheth KN, Elm JJ, Molyneaux BJ, et al. Safety and efficacy of intravenous glyburide on brain swelling after large hemispheric infarction (GAMES-RP): A randomized, double-blind, placebo-controlled phase 2 trial. Lancet Neurol 2016;15:1160-1169.

Large hemispheric infarction (LHI) accounts for up to 10% of all ischemic strokes and confers a high degree of morbidity and mortality.1 Malignant cerebral edema often develops after LHI, which leads to brain herniation and death. Without surgery, mortality is more than 70%.1 Decompressive hemicraniectomy has been shown to significantly reduce mortality, although approximately 40% of survivors are left severely disabled with a modified Rankin Scale (mRS) of 4 (unable to walk without assistance and unable to attend to own bodily needs without assistance).2 Therefore, further strategies are necessary to reduce morbidity and improve neurological outcomes in patients with LHI.

Preclinical studies have shown that glyburide, a sulfonylurea receptor 1 (SUR1) inhibitor reduces cerebral edema, improves survival, and is associated with better neurological function. Thus, the GAMES-RP was created as a Phase II clinical trial to assess the safety and efficacy of intravenous glyburide at reducing brain swelling in patients with LHI.

GAMES-RP was a double-blind, prospective, randomized trial performed across 18 hospitals in the United States. Patients were randomized to receive either intravenous glyburide or placebo within 10 hours of LHI stroke onset. LHI was defined by a diffusion weighted image (DWI) lesion volume of 82-300 cm3. Patients were allowed to have received intravenous thrombolysis with tPA but not endovascular thrombectomy. Intravenous glyburide was given as a bolus and continued for 72 hours. The primary outcome was the proportion of patients with an mRS of 0-4 at 90 days without undergoing decompressive hemicraniectomy.

Of 77 patients, 41 were randomized to receive intravenous glyburide and 44 to receive placebo. Overall, there was no difference in the primary outcome, mRS 0-4 without decompressive hemicraniectomy, in the glyburide and placebo groups (41% vs. 39%, P = 0.77). In addition, there were no differences in secondary outcomes — rates of decompressive craniectomy or death by 14 days, change in ipsilateral swelling at 72-96 hours, or change in lesional swelling at 72-96 hours — between the groups. However, there was a strong, although nonsignificant, trend toward improved functional outcome (P = 0.12) in patients who received glyburide. Furthermore, patients who received glyburide had significantly decreased midline shift at 72-96 hours compared to patients who received placebo (4.6 mm vs. 8.5 mm, P = 0.0006) and decreased concentrations of metalloproteinase 9 (MMP-9), a biomarker of brain edema (P = 0.006). Finally, although nonsignificant, there was a trend toward reduced mortality in patients receiving intravenous glyburide (17% vs. 36%, P = 0.06) at 90 days.

To account for variations in practice of using decompressive hemicraniectomy across the different centers, the authors performed a post-hoc analysis of participants without decompressive hemicraniectomy and found an effect of intravenous glyburide at reducing hemispheric swelling (median 49 cm3 for glyburide group vs. 77 cm3 in the placebo group, P = 0.04) but not for lesional swelling (median 41 cm3 vs. 75 cm3, P = 0.15). Finally, the rate of serious adverse events in the intravenous glyburide group and placebo group was similar.


This study failed to show that intravenous glyburide led to a decreased need for decompressive hemicraniectomy in patients with LHI. However, intravenous glyburide did reduce edema as measured by decreased midline shift and decreased concentrations of MMP-9. In addition, adverse effects were no more common in patients receiving intravenous glyburide than in patients receiving placebo, and in a shift analysis, there was a strong trend toward improved functional outcomes in patients receiving intravenous glyburide.

Perhaps decompressive hemicraniectomy was not the best outcome to measure; although there are published guidelines recommending when to perform hemicraniectomy for LHI, the decision was ultimately left to individual centers, which often have varying institutional policies on when to perform surgery. For example, in this study, 90% of all the surgeries that were performed occurred in only eight of the 19 centers, which together enrolled only half of the participants.

Overall, although intravenous glyburide did not reduce the need for decompressive hemicraniectomy, there was a strong trend toward improved functional outcomes and reduced mortality, as well as a reduction in neuroimaging and biomarkers of edema. Further study is necessary given the high morbidity and mortality associated with LHI.


  1. Huttner HB, Schwab S. Malignant middle cerebral artery infarction: Clinical characteristics, treatment strategies, and future perspectives. Lancet Neurol 2009;8:949-958.
  2. Vahedi K, Hofmeijer J, Juettler E, et al. Early decompressive surgery in malignant infarction of the middle cerebral artery: A pooled analysis of three randomised controlled trials. Lancet Neurol 2007;6:215-222.