Association of revisions or conversions after sleeve gastrectomy with annual bariatric center procedural volume in the state of New York.

<bold>Introduction: </bold>Although bariatric center procedural volume has been associated with early perioperative safety, data on the effect of such volume and long-term outcomes after sleeve gastrectomy (SG) are limited. This study aims to examine the relationship between annual bariatric center SG volume and the incidence of revisions or conversions (RC) after SG.<bold>Methods: </bold>The New York Statewide Planning and Research Cooperative System database was used to identify all patients who underwent SG between 2006 and 2012. Subsequent RC events were captured up to 2016. Bariatric centers having annual SG volume less than 45, between 45 and 65, and greater than 65 were defined as low (LV), medium (MV), and high volume (HV), respectively. Multivariable Cox proportional hazard regression analysis was performed to compare the risk of having RC among centers with different yearly sleeve volumes.<bold>Results: </bold>We identified 8389 patients who underwent SG. The overall estimated cumulative incidence of RC was 0.5% (95% CI 0.3-0.6%) at 1 year, 6.2% (95% CI 5.4-7.0%) at 5 years, and 15.3% (95% CI 12.6-18.0%) at 8 years after SG. The estimated cumulative incidence of RC for LV, MV, and HV at 8 years after SG was 16.7% (95% CI 11.1-22.3%), 15.5% (95% CI 11.2-19.8%), and 13.7% (95% CI 9.4-17.9%), respectively. HV centers have lower risk of RC compared to LV (hazard ratio 0.65; 95% CI 0.48-0.88) and MV (hazard ratio 0.75; 95% CI 0.57-0.98). LV and MV centers have comparable risk of RC (hazard ratio 1.15; 95% CI 0.87-1.51). Patients having the initial SG performed in LV were the least likely to have RC in the same institution (46.1% of LV, 13.2% of MV and 22.3% of HV; p < 0.0001).<bold>Conclusion: </bold>Patients undergoing SG at LV centers experience the highest risk of subsequent RC. This effect persists after adjusting for patient-level factors. These data underline the relationship between volume threshold and long-term effect. [ABSTRACT FROM AUTHOR]