Multicenter Cohort of 768 High-risk Patients With Symptomatic Gallstones, Adding Cholecystectomy Did Not Increase Length of Stay or Perioperative Morbidity, Supporting it as a Safe Single-stage Option

Overview

Despite numerous single-center reports and meta-analyses, few studies have directly compared outcomes of LSG with versus without concomitant cholecystectomy exclusively in a high-risk cohort as defined above. To the best of our knowledge, only one prior study (Dincer and Dogan, 2019) specifically examined this high-risk population, and it included only 27 patients in the concomitant group. This multicenter retrospective comparative study - conducted in three research institutes in Egypt and Saudi Arabia - aims to evaluate the safety of concomitant cholecystectomy during LSG in high-risk patients with severe obesity and symptomatic cholelithiasis, using a control group of high-risk patients who underwent LSG alone. Our study provides the largest cohort to date specifically addressing this question in a high-risk population.

Data was collected from the archived medical records of the surgical departments of all the three research centers (in Zagazig city, Egypt, and the Najran University Hospital, Najran city, Saudi Arabia), and were retrospectively analyzed for all patients with inclusion criteria and the laparoscopic sleeve gastrectomy was done, with or without concomitant laparoscopic cholecystectomy during the period from January 2015 to January 2025 (the study period). Patients were divided into: * Group A (LSG alone): High-risk patients with symptomatic stones who underwent LSG without cholecystectomy. * Group B (LSG with concomitant cholecystectomy): High-risk patients with symptomatic stones who underwent LSG and simultaneous laparoscopic cholecystectomy. The decision to perform concomitant cholecystectomy was clinical, based on surgeon preference, patient consent, and intraoperative findings. Group A patients were those in whom the surgeon opted to defer cholecystectomy due to concerns about operative time, complexity, patient refusal, or significant comorbidities that might increase the risk of a combined procedure. This selection bias was addressed using propensity-score matching. A total of 2,512 patients underwent LSG during the study period across the three centres \[1,452 at Zagazig University Hospitals, 567 at Al-Ahrar Teaching Hospital, and 493 at Najran University Hospital\]. Among them, 805 patients (32.1%) met the high-risk criteria (≥ 3 comorbidities, age ≥ 55 years, or ASA ≥ III) and had symptomatic cholelithiasis. Of these, 518 (64.3%) underwent LSG alone (Group A), and 287 (35.7%) underwent concomitant cholecystectomy (Group B). After excluding 37 patients lost to 30-day follow-up (23 from Group A, 14 from Group B), the final analysis included 495 patients in Group A and 273 patients in Group B. See the flow chart for the allocation process throughout the study Postoperative Care and Outcome Measures All patients followed an enhanced recovery after surgery (ERAS) protocol: early mobilization, multimodal analgesia, and gradual dietary progression. Discharge criteria included tolerance of oral intake, adequate pain control, and afebrile status. The primary outcome was overall 30 day postoperative complication rate, while secondary outcomes were: * Operative time (skin incision to skin closure) * Length of hospital stay (LOS) * Specific complications: bleeding (requiring transfusion or reoperation), surgical site infection (superficial/deep), staple line leakage, bile leak, trocar site eventration (previously termed "port site hernia") * 30 day readmission and reoperation rates * 30 day mortality * Clavien Dindo grade of complications Data Collection Data were extracted from electronic medical records, operative logs, and outpatient charts using a standardised form. Variables included demographics, BMI, comorbidities, ASA class, preoperative ultrasound findings, operative details, intraoperative events, and 30 day outcomes. Loss to follow up was defined as no documented contact or clinic visit within 30 days post discharge. Statistical analysis Continuous variables were expressed as mean ± standard deviation (SD) or median with interquartile range (IQR) and compared using Student's t-test or Mann-Whitney U test. Categorical variables were expressed as frequencies and percentages and compared using chi-square or Fisher's exact test (with Fisher's exact test used for comparisons with expected cell counts \<5). A p-value \<0.05 was considered significant. To account for multiple comparisons, the Bonferroni correction was applied; a corrected p-value \<0.003 was considered significant for the 15 comparisons. Post-hoc power analysis was performed: based on the observed overall complication rates (5.5% vs 5.9%) and sample sizes (n=495 and 273), the study had approximately 85% power to detect a 5% absolute difference in overall complications (α=0.05, two-tailed). For rare events (e.g., staple line leakage), the study was underpowered. To address selection bias, propensity-score matching was performed using the following covariates: age, sex, BMI, number of comorbidities, ASA class, gallbladder wall thickness, and history of acute cholecystitis. Patients were matched 1:1 using a caliper of 0.2 of the standard deviation of the logit of the propensity score. Standardised mean differences (SMD) were calculated before and after matching; SMD \<0.1 indicated good balance. Multivariable logistic regression was performed to identify independent predictors of overall complications, with concomitant cholecystectomy as the primary variable of interest, adjusting for age, sex, BMI, number of comorbidities, ASA class, gallbladder wall thickness, history of acute cholecystitis, and centre. Adjusted odds ratios (OR) with 95% confidence intervals (CI) were reported. Sensitivity analyses included: (1) centre-stratified analysis using random-effects meta-analysis; (2) temporal trend analysis by year of surgery; (3) subgroup analyses by each component of the high-risk definition (age ≥55, ≥3 comorbidities, ASA ≥III); (4) exclusion of patients with acute cholecystitis; and (5) stratification by procedural sequence (cholecystectomy-first vs. sleeve-first) in Group B. All analyses were performed using SPSS version 22.0 (IBM Corp., Armonk, NY, USA) and R version 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria).