Previous gastric banding increases morbidity and gastric leaks after laparoscopic sleeve gastrectomy for obesity

Laparoscopic sleeve gastrectomy (LSG) is performed in certain circumstances after failure of gastric banding. The goal of this study was to evaluate the impact of first-line gastric banding on the morbidity associated with secondary LSG for obesity.

The case records of 102 consecutive patients undergoing LSG were studied retrospectively. The technique of LSG was standardized. Two groups were compared: one with patients having undergone LSG after first-line gastric banding (n = 31) and the second, with patients having undergone first-line LSG (n = 71). Endpoints were overall morbidity and intra/postoperative complications including gastric leaks consecutive to staple line disruption as well as other septic or hemorrhagic complications. Multivariable analysis was performed to detect independent risk factors for morbidity.

Overall morbidity was significantly higher in patients having undergone LSG after first-line gastric banding compared with those undergoing first-line LSG (32.2% vs. 7%, P = 0.002). Gastric leaks secondary to staple line disruption also occurred statistically significantly more often in patients with first-line gastric banding (16.1% vs. 2.8%, P = 0.043). Waiting 6 months between gastric band removal and performing LSG did not prevent the increased morbidity compared with first-line LSG. Multivariable analysis revealed that among the factors analyzed (age, gender, comorbidity, body mass index, surgeon, first-line gastric banding), the only independent risk factor for staple line disruption was first-line gastric banding with an odds ratio = 6.6 (95% confidence interval = [1.2-36.3]).

Undergoing first-line gastric banding increases the risk of complications after secondary LSG. We recommend that patients who undergo LSG after a first-line gastric banding should be warned of the increased risks of morbidity or, alternatively, that LSG be performed preferentially as the initial procedure.