I. A. Van Wieren1, J. Thumma1, O. Varban1, J. Dimick1 1University Of Michigan,Department Of Surgery, The Center For Healthcare Outcomes & Policy,Ann Arbor, MI, USA
Introduction: Sleeve gastrectomy has emerged as the most common bariatric procedure. However, there is emerging data that this procedure can result in lifestyle limiting gastroesophageal reflux. It is unclear whether these symptoms are severe enough to offset the benefits of the procedure in terms of weight loss and other positive outcomes. Using a validated disease-specific instrument, we evaluated the extent to which reflux symptoms after sleeve gastrectomy affected patients’ satisfaction with the surgery.
Methods: We studied 6,633 patients who underwent laparoscopic sleeve gastrectomy (2013 to 2017) from Michigan bariatric surgical collaborative. We used the GERD-HRQL score which is 10 questions each ranging from 0 for no symptoms to 5 for severe symptoms. To assess the impact of sleeve gastrectomy we calculated change in this score before versus after the procedure. We divided the delta GERD score into quintiles: the bottom quintile represents worsening of GERD symptoms from baseline to 1-year and the top quintile represents improvement in symptoms. We then looked at the relationship between delta GERD score and patient satisfaction at 1-year. We used univariate and multivariate generalized linear mixed models to assess the variation in satisfaction explained by change in GERD score/delta GERD, percent excess body weight loss (%EBWL) at 1-year and other patient outcomes (serious complications, readmission and reoperations). We controlled for patient factors (age, gender, race and comorbidities) and year of surgery.
Results: The average change in GERD score was 1.63 (range: -48 to 48). However, the change in GERD score varied across quintiles with -9.0 point (range: -48 to -3) worsening in the bottom quintile verses a 13.9 point (range: 7 to 48) improvement in the top quintile. Overall, 77.7% of patients were satisfied, but the proportion of patients satisfied was highly dependent on whether there reflux symptoms improved or worsened. For example, in the bottom quintile only 48.9% were satisfied compared to 78.1% in the top quintile. In a multivariate model, changes in GERD score explained 10.5% of the variation in 1-year satisfaction. In fact, change in GERD score predicted the most variation in 1-year patient satisfaction, especially among whose symptoms worsened the most. For patients in the worst quintile, reflux symptoms explained 30.6% of variation compared to 2.2% with little change or improvement in reflux (quintiles 2-5). In univariate analyses, %EBWL explained only 2% of variation in satisfaction and <1% was explained by 30-day patient outcomes (serious complications, readmissions or reoperations).
Conclusion: In this state wide study of sleeve gastrectomy in Michigan, we demonstrated that reflux symptoms are the most important determinant of 1-year satisfaction after sleeve gastrectomy particularly among patients whose symptoms worsened the most.