Strategies to Promote Resiliency: A Randomized Embedded Multifactorial Adaptative Platform (REMAP) Clinical Trial to Study Interventions to Improve Recovery After Surgery in High-Risk Patients

Objective: In a randomized, embedded, multifactorial, adaptive platform (REMAP) trial, we hypothesized that perioperative metformin would improve postoperative time alive and out of the hospital, defined by 90-day hospital-free days (HFD-90), among nondiabetic aged adults. Background: As our population ages, patients are increasingly frail requiring an emphasis on treatments to counteract their diminished resilience, especially following the stress of surgery. Growing literature supports metformin as an antiaging and anti-inflammatory therapy with beneficial effects extending into the perioperative period. Methods: At-risk adults (≥60 years) scheduled for elective surgical interventions were randomized to placebo (N = √3) or metformin (N = 1:1:1; 500 mg:1000 mg:1500 mg) for short (7-28 days), intermediate (29-90 days), or long (>90 days) preoperative durations. An adaptive sample size of (N = 1000-2500) would identify at least a 15% improvement in HFD-90 for >1 metformin doses. Using intention-to-treat analysis, Bayesian ordinal logistic regression compared HFD-90 and frequentist logistic regression compared 90-day reoperation and readmission. Results: Before trial closure, we randomized 302 (N = 106 placebo, N = 196 metformin [N = 64, 500 mg; N = 66, 1000 mg; and N = 66, 1500 mg]) patients without differences in baseline demographics (age 68 ± 6 years, 45% females, and 92% White race) or interventions (spine [29%], general [38%], colorectal [13%], and other [20%]). The odds of HFD-90 did not significantly differ between all doses and duration of metformin or placebo. There were no differences in the odds of reintervention (OR = 1.1 [95% CI, 0.6-2.0]) or readmission (OR = 1.5 [95% CI, 0.7-2.8]). Conclusions: Pretreatment with metformin did not improve postoperative outcomes in this REMAP trial, although trial enrollment was markedly limited by the COVID-19 pandemic and is underpowered.