Lap band outcomes from 19,221 patients across centers and over a decade within the state of New York

Introduction: We sought to determine the rate of revision and explant of the laparoscopic adjustable gastric banding (LAGB) over a ten-year period in the state of New York.

Methods: Following IRB approval, the SPARCS administrative database was used to identify LAGB placement from 2004 to 2010. We tracked patients who underwent band placement with subsequent removal/revision, followed by conversion to either Roux-en-Y gastric bypass (RYBG) or sleeve gastrectomy (SG) between 2004 and 2013. McNemar test and Chi-square test were used to compare complications between primary procedure and subsequent revision and to compare complication rates and mortality rates, respectively. Log-rank test was used to assess patient characteristics and comorbidities. p < 0.05 was considered significant.

Results: During a 7-year period, there were 19,221 records of LAGB placements and 6567 records of revisions or removal. We were able to follow up 3158 (16.43 %) who subsequently underwent a band removal or revision over the course of this period. An additional 3606 patients had no records in the state of New York following the procedure, thus making the rate of revision 20.22 %. Initial revision procedures were coded as band removal in 32.77 % (n = 1035), band revision in 30.53 % (n = 964), band removal and replacement in 19.09 % (n = 603), removal and conversion to SG in 5.64 % (n = 178), or removal and conversion to RYGB in 11.97 % (n = 378). From the 3158 patients, 2515 (79.64 %) required only one revision. Six hundred and forty-three patients underwent two or more revisions. Thirty-one out of 3158 (0.0098 %) patients had complications at their initial operation, but 919 (29.1 %) had complications during revision (p < 0.0001).

Conclusions: Over a 7-year period, at least 20.22 % of LAGB required removal or revision. Based on all case numbers, total revision rate may be as high as 34.2 %. Although the band is believed to be a reversible procedure, revisional procedures are significantly more morbid than the initial procedure.