Repeat sleeve gastrectomy: optimization of outcomes by modifying the indications and technique

Background: Few series are available concerning repeat sleeve gastrectomy (re-SG), and series have reported contradictory results concerning morbidity rates, with limited data concerning weight loss.

Objective: Evaluate the short- and medium-term outcomes of re-SG.

Setting: University hospital, France, public practice.

Methods: Between June 2007 and March 2016, all patients undergoing re-SG (n = 46 patients) were included. Re-SG was proposed for patients with insufficient excess weight loss (EWL) (≤50%) or renewed weight gain with excessively high residual gastric volume (>250 mL and/or large gastric pouch). The primary efficacy endpoint was the overall complication rate of re-SG. The secondary efficacy endpoints were operative data, evaluation of weight loss, and correction of co-morbidities, risk factors for gastric leak (GL), by comparing 2 periods (period 1, January 2004-December 2013: blue/green or purple staplers without reinforcement; period 2, after December 2013: black staplers with reinforcement) and comparison of weight loss according to the indication for re-SG.

Results: The re-SG group consisted of 46 patients (35 women, mean age: 47.5 yr). The mean body mass index (BMI) before SG was 47.2 kg/m² (35-63.6). The mean time interval between SG and re-SG was 73 months (11-106). The BMI before re-SG was 41.2 kg/m² (29-54.7). Indications for surgery were insufficient weight loss in 25 patients (54.3%) and weight regain in 21 patients (45.7%). A large gastric pouch was visible in 4 patients (8.6%). The mean operating time was 97.6 minutes (45-220). One death (2.1%) and 7 complications (15.2%) were observed. The mean length of hospital stay was 3.6 days (1-30). At last follow-up, mean BMI was 32.1 kg/m² (20.3-41.3) and mean EWL was 62.3% (18-127.2). When analyzing risk factors for GL, residual gastric volume between 250 and 350 mL was associated with a higher GL rate compared with a volume ≥350 mL, and re-SG performed during period 1 was associated with a higher GL rate than re-SG performed during period 2 (17.4% versus 0%; P = .13). Re-SG performed for weight regain was associated with a significantly higher additional weight loss compared with re-SG performed for insufficient weight loss (mean additional EWL of 45.9%; P = .06).

Conclusion: Re-SG is feasible, but it requires adaptation of the surgical procedure to decrease complications. Results on weight loss are acceptable, but the best indications for re-SG were a gastric volume>350 mL and in the case of weight regain with the exception of technical failure of the primary SG.