Cognitive Outcomes in Randomized Controlled Trials of Coronary Artery Bypass Graft Surgery From 2005 to 2025: A Systematic Review.

Overview

abstract

BACKGROUND: Cognitive decline after coronary artery bypass graft (CABG) is common and affects morbidity, mortality, and quality of life. We systematically reviewed randomized CABG trial control arms to characterize cognitive assessments, testing frequency, attrition, and ability to detect perioperative change. METHODS: We searched MEDLINE, Embase, Cochrane Library, and PsycINFO for randomized controlled trials of CABG surgery that included at least one arm of patients solely undergoing CABG and that reported at least one objective cognitive assessment, from January 2005 to February 2025. Trials with mixed cardiac surgery or only subjective measures were excluded. We summarized task frequency, cognitive domains, and attrition. For tasks assessed preoperatively and postoperatively in ≥3 trials, we reported control group means and SDs. Risk of bias was assessed using the Cochrane Risk of Bias tool among 6 bias domains. This study was supported by NIH-R01NS123639. RESULTS: Of 3494 screened studies, 2284 were CABG trials, and only 71 (3.1%) reported cognitive evaluation. These involved 15 925 patients (79% men; mean age, 64.2 years; median follow-up, 90 days) and used 145 unique cognitive tasks, with the Trail Making Test Part B (40 of 71; 56.3%) and Part A (38 of 71; 53.5%) being the most frequently administered. Among 7 tasks with sufficient data, none detected preoperative to postoperative changes. Attrition rates averaged 18.9%, with a broad range of 0 to 62%. CONCLUSIONS: Cognitive assessment is uncommon in CABG trials, and commonly used tests rarely detect change. Heterogeneity precluded meta-analysis, and high attrition raises concerns about selection and survivorship bias. To evaluate cognitive impact after CABG, trials need standardized, sensitive assessment strategies resilient to attrition and feasible for broad deployment.