Back to 2011 Program

Introduction: Bariatric surgery has been shown to be beneficial in achieving weight loss and in decreasing long-term mortality. To date, however, evidence has suggested an increased mortality following bariatric surgery in patients ≥ 65 years. With adjustment of confounding variables, we hypothesized that the short-term operative outcome profile in those ≥ 65 years undergoing bariatric surgery would be comparable to that of younger persons. Methods: Patients with BMI ≥ 35 kg/m2 who underwent open and laparoscopic Roux-en-Y gastric bypass, open duodenal switch, laparoscopic adjustable gastric banding and vertical banded gastroplasty in the 2005 - 2009 American College of Surgeons National Surgical Quality Improvement Program were identified. Controlling for confounders and stratifying by open versus laparoscopic surgery, multivariate regression was used to predict the impact of age (≥ 65 years) on mortality, major events and prolonged length of stay (PLOS, > 90th percentile) at 30 days. Results: We identified 48,378 patients who underwent the above bariatric procedures between 2005 and 2009. Over the previous 5 years, the percentage of older patients undergoing bariatric surgery has increased from 1.92% in 2005 to 4.77% in 2009 (p < 0.001). There were only 72 deaths throughout the entire study period, 8 of which were in the ≥ 65 years cohort. The incidence of 30-day mortality in those 35-49, 50-64 and ≥ 65 years was 0.12%, 0.21% and 0.40%, respectively. Adjusting for confounders, multivariate regression analysis demonstrated advancing age to trend towards becoming a predictor of mortality but it was not significant (Table 1). Age ≥ 65 was a significant predictor of PLOS for both open and laparoscopic procedures. For those who underwent laparoscopic procedures, odds ratios were similar for PLOS for the 50-64 years and ≥ 65 years’ cohorts. While major adverse events were not predicted by age ≥ 65 for either open or laparoscopic procedures, BMI ≥ 55 kg/m2, severe ASA score, cardiac co-morbidities, albumin < 3 and creatinine > 1.5 were all predictors of major adverse events (not shown). Conclusions: This large, multi-hospital study demonstrates older age predicts short-term PLOS but not major events. The overall low death rates likely explain the observed trend toward significance in operative mortality in older patients. Thus surgeons should continue to promote caution when considering whether to operate on this patient population. Once corroborated, these results provide important information to patients, surgeons, hospitals, and payers prior to performing bariatric surgery in older persons with obesity.
Table 1. Predictors of 30-day Mortality and Prolonged Length of Stay

	30 Day Mortality OR (95% CI)		Prolonged Length of Stay OR (95% CI)
Factor	Open N=5017	Laparoscopic N=43361	Open N=4992	Laparoscopic N=43314
Age 35-49 yrs 50-64 yrs ≥ 65 yrs	Ref. 1.2 (0.4, 3.1) 3.3 (0.8, 14.3)	Ref. 1.2 (0.6, 2.4) 1.5 (0.5, 4.6)	Ref. 1.2 (1.0, 1.5) 2.1 (1.5, 2.9)	Ref. 1.2 (1.2, 1.3) 1.2 (1.1, 1.4)
BMI (kg/m2)45-49 50-54 55-59 ≥ 60	Ref. 1.8 (0.4, 8.1) 2.6 (0.6, 11.7) 5.3 (1.4, 20.1)	Ref. 2.4 (0.9, 6.0) 2.3 (0.8, 6.7) 4.0 (1.5, 11)	Ref. 1.2 (0.9-1.5) 1.4 (1.1, 1.8) 1.5 (1.2, 1.9)	Ref. 1.1 (1.0-1.2) 1.35 (1.2-1.5) 1.7 (1.6, 1.9)
C-index	0.78	0.73	0.65	0.63
Percentage of Events, % (N)	0.50 (25)	0.11 (47)	17.8 (888)	20.1 (8694)

Multivariate regression analysis, controlling for confounders, was used to formulate the above odds ratios. Confounders adjusted for include, but are not limited to, age, BMI, sex, race, ASA score, diabetes, pulmonary and cardiac comorbidities, albumin, creatinine, and admission year.