Glucose-insulin-potassium therapy in patients with ST-segment elevation myocardial infarction

Context The clinical benefit of glucose-insulin-potassium (GIK) infusion in patients with ST-segment elevation myocardial infarction (STEMI) is unclear. While some smaller trials suggest benefit, in the CREATE-ECLA trial, GIK infusion had no effect on 30-day mortality in 20 201 patients. Objectives To determine the association between GIK infusion therapy and 30-day and 6-month outcomes in patients with STEMI. Design, Setting, and Participants Primary analysis of the OASIS-6 GIK randomized controlled trial of 2748 patients with acute STEMI; prespecified analyses of the combined trial data from the OASIS-6 GIK and CREATE-ECLA GIK trial populations of 22 943 patients with acute STEMI; subgroup analysis on the timing of initiation of GIK infusion therapy and outcomes; and post hoc analyses exploring whether GIK infusion may cause early harm by increasing glucose and potassium levels and net fluid gain. Intervention High-dose GIK solution consisting of 25% glucose, 50 U/L of regular insulin, and 80 mEq/L of potassium infused at 1.5 mL/kg per hour for 24 hours. Main Outcome Measures Mortality rates at 30 days and 6 months in the OASIS-6 GIK trial and rates of death, heart failure, and the composite of death or heart failure at 3 and 30 days in the combined OASIS-6 GIK and CREATE-ECLA GIK trial populations. Results At 6 months, 148 (10.8%) GIK infusion patients and 143 (10.4%) control patients died in the OASIS-6 trial ( hazard ratio [HR], 1.04; 95% CI, 0.83-1.31; P=.72); 153 (11.1%) GIK patients and 185 (13.5%) control patients had heart failure ( HR, 0.83; 95% CI, 0.67-1.02; P=.08); and 240 (17.5%) GIK patients and 264 (19.2%) control patients had a composite of death or heart failure ( HR, 0.91; 95% CI, 0.76-1.08; P=.27). In the prespecified analyses of the combined trial data, there were 712 deaths (6.2%) in the GIK group and 632 deaths (5.5%) in the control group at 3 days ( HR, 1.13; 95% CI, 1.02-1.26; P=.03). This difference disappeared by 30 days, with 1108 deaths (9.7%) in the GIK group and 1068 (9.3%) in the control group ( HR, 1.04; 95% CI, 0.96-1.13; P=.33). GIK therapy increased levels of glucose, potassium, and net fluid gain postinfusion, all 3 of which predicted death after adjusting for multiple confounders. Adjusting for glucose, potassium, and net fluid gain eliminated the apparent increase in mortality at 3 days observed with GIK infusion, suggesting a direct association with these factors. Administration of GIK infusion within 4 hours of symptom onset yielded no benefit compared with later initiation. Conclusions Infusion of GIK provided no benefit and may cause early harm following STEMI. Avoidance of infusion-related hyperglycemia, hyperkalemia, and net fluid gain may be advisable in future studies of metabolic modulation in patients with STEMI. Trial Registration clinicaltrials.gov Identifier: NCT00064428.