Preoperative low-energy diets for patients with a body mass index >30 kg/m2 undergoing non-bariatric surgery: pilot feasibility randomized clinical trial and a systematic review and meta-analysis of efficacy data

Background: The current evidence for preoperative low-energy diets (LEDs) for patients with a BMI >30 kg/m2 before non-bariatric surgery rests on studies with a high risk of bias. An RCT is warranted to bridge this knowledge gap. Before a larger RCT, a pilot feasibility RCT was conducted to address potential hurdles for the larger trial. Methods: The pilot feasibility multicentre trial was conducted in Canada at four centres between 3 January 2024 and 23 October 2024. Patients were randomized (1: 1 randomly permuted online blocked allocation) to receive a 3-week LED protocol or standard care. All patients aged >18 years with a BMI >30 kg/m2 undergoing elective non-bariatric intra-abdominal or orthopaedic surgery were evaluated for enrolment. Main exclusion criteria were LED contraindications and surgery scheduled without at least 3 weeks notice. The primary outcome was descriptive including the following feasibility outcomes: recruitment rate, randomization percentage, intervention adherence, and follow-up completion. Clinical outcomes included anthropometric data. The primary outcome analysis was descriptive. Additionally, a random-effects meta-analysis was performed using data from the present study and data from previously published RCTs to evaluate 30-day postoperative morbidity. The last follow-up date was 14 January 2025. Results: Out of 373 patients approached, 288 were eligible. Of these, 175 consented to be randomized (60.8% (95% c.i. 54.9–66.4%)). Seventy four of these patients were still awaiting a surgical date when the trial closed, leaving 91 patients randomized (LED n = 45, control n = 46) in the present pilot trial. Out of these 91 randomized patients, 81 (89%, 95% c.i. 80.7–94.6%) had complete follow-up. LED adherence was 81.7% (95% c.i. 74.1–89.3%). Analysis of covariance suggested patients in the LED group lost more weight during the 3-week intervention period (MD 4.5kg, 95% c.i. −5.6 to −3.5). The pooled meta-analysis of 5 RCTs suggested a 19% relative risk reduction in postoperative morbidity favouring the intervention (RR 0.81, 95% c.i. 0.52–1.26, P = 0.351, I2 = 0%). Conclusion: The feasibility targets of this pilot RCT were not met for recruitment rate, randomization percentage, and complete follow-up. The pooled meta-analysis suggests that LEDs before non-bariatric surgery can effectively induce weight loss with reduced postoperative morbidity, which needs to be validated in a larger non-inferiority RCT with additional centres to meet the feasibility targets. Registration number: NCT03935451 (http://www.clinicaltrials.gov).