Can frailty predict complications and mortality in patients undergoing total knee arthroplasty? A systematic review and meta-analysis.
Objective To evaluate the value of frailty in predicting postoperative complications and mortality in patients undergoing total knee arthroplasty (TKA). Methodology PubMed, Scopus, EMBASE and CENTRAL databases were screened up to 20th November 2025 for observational and randomized studies evaluating postoperative outcomes in frail TKA patients. Data on postoperative complications, readmission and reoperation rates, length of hospital stay and mortality were extracted. A random-effects model was used to calculate pooled risk ratios (RRs) and odds ratios (ORs) with 95% confidence intervals (CIs). Results Sixteen studies with over 14 million patients were included. Frail patients had a significantly higher risk of 30-day mortality with RR 4.40 (95% CI: 2.18-8.91; p < 0.00001); overall postoperative complications with RR 1.85 (95% CI: 1.35-2.52; p= 0.0001); 30-day readmission with RR: 2.24 (95% CI: 1.74-2.87; p < 0.00001) and 30-day reoperation with RR: 1.81 (95% CI: 1.48-2.22; p < 0.0001). Frailty was also linked to an increased risk of surgical site infections with RR 1.56 ( 95% CI: 0.86-2.83) and prolonged hospital stay with mean difference MD 0.86 days (95% CI: 0.55-1.16; p<0.00001). Individual complications, such as myocardial infarction, renal complications and pulmonary embolism, were consistently more frequent in frail patients. Heterogeneity was moderate to high in some outcomes, attributed to differences in frailty assessment tools and study designs. Conclusion Within the limitations, frailty is a strong, independent predictor of increased postoperative complications, mortality and prolonged hospital stays in patients undergoing TKA. Incorporating frailty assessments into preoperative evaluations can enhance risk stratification and guide interventions to improve perioperative care in this vulnerable population.Prospero Registration Number: CRD420251035972.