Computed tomography-derived low muscle status and clinical outcomes after isolated coronary artery bypass grafting: A systematic review and meta-analysis with implications for perioperative nutrition risk stratification

Background and aims Computed tomography (CT)-derived measures of low muscle quantity and quality may indicate reduced physiologic reserve, but their prognostic value after isolated coronary artery bypass grafting (CABG) is uncertain. We conducted a PRISMA-guided systematic review and conservative meta-analysis of adjusted estimates to examine associations between preoperative CT-derived low muscle status and outcomes after isolated CABG. Methods We searched PubMed, Embase, Web of Science, Cochrane CENTRAL, and ClinicalTrials.gov from inception to December 23, 2025. Cohort studies of adults undergoing isolated CABG were eligible if they assessed preoperative low muscle status using CT (dual-energy X-ray absorptiometry eligible) and reported mortality and/or postoperative complications. Overlapping cohorts were excluded. Adjusted hazard ratios (HRs) were pooled using random-effects meta-analysis (restricted maximum likelihood) with the Hartung–Knapp–Sidik–Jonkman adjustment. Results Four independent cohorts ( n = 3931) met inclusion criteria and used CT-based definitions spanning lumbar and thoracic muscle quantity/quality measures. Three cohorts ( n = 3452) contributed adjusted estimates for long-term mortality. Across studies, the direction of association was unfavorable, but quantitative synthesis was exploratory and highly uncertain (pooled HR 2.71, 95% confidence interval 0.38–19.40; I 2 = 92.3%; approximate 95% prediction interval 0.47–15.6). Secondary outcomes were too heterogeneous for pooling and were summarized narratively. Conclusions Preoperative CT-derived low muscle status may serve as a pragmatic vulnerability flag in isolated CABG when CT is already available, but current pooled evidence for long-term mortality remains exploratory, highly heterogeneous, and statistically imprecise. Standardized measurement, more comparable adjustment strategies, and harmonized outcome reporting are needed before broader clinical translation.