Factors Predicting In-Hospital Mortality in Bariatric Surgery: An Analysis from the National Inpatient Sample Database

Muhammad Asad Khan, MD, John N. Afthinos, MD, FACS, Karen E. Gibbs, MD, FACS

Staten Island University Hospital

Objective: Bariatric surgery is being performed in increasing numbers each year. Centers of excellence standards were set up to improve the quality and monitor post-operative morbidity and mortality. Bariatric surgery has matured into a field which maintains high standards for safety and quality. We sought to evaluate the in-hospital outcomes from a large, prospectively collected database to determine predictors of in-hospital mortality to aid in pre-operative assessment of these challenging patients.

Methods: The National Inpatient Sample database was queried for primary bariatric operations performed from 2005 – 2009. Revisional surgery and biliopancreatic diversion-duodenal switch procedures were excluded. Patient comorbid conditions, insurance status, ethnicity, age and gender were evaluated. In-hospital morbidity and mortality was tabulated. A multivariate logistic regression was performed to select factors which contributed to increased mortality.

Results: The weighted national estimate of bariatric procedures performed was 548,106. Laparoscopic Roux-en-Y gastric bypass was the most commonly performed procedure (60.7%). The overall in-hospital mortality was 0.1% (Table 1). Statistically significant predictors of in-hospital mortality included age > 50, male gender, open procedure, COPD, obstructive sleep apnea, peripheral vascular disease and congestive heart failure (Table 2).

Conclusions: Operative mortality in bariatric surgery remains very low, rivaling that of more commonly accepted procedures (i.e. laparoscopic cholecystectomy). Despite the many comorbidities associated with the morbidly obese patient, excellent outcomes can be achieved when these patients are managed appropriately. These outcomes speak to the level of maturity and dedication to quality patient care in the field of bariatric surgery. Even with the recent introduction of the sleeve gastrectomy, outcomes remain excellent.

Table1. Bariatric procedure type and corresponding mortality rate

Procedure type	ICD-9 Code	N (%)	Mortality
Open GBP	44.31, 44.39	67340 (12.3%)	232 (0.3%)
Lap GBP	44.38	332566 (60.7%)	278 (0.1%)
Lap Gastric band	44.95	114272 (20.8%)	36 (0.0%)
Lap Gastroplasty	44.68	12627 (2.3%)	5 (0.0%)
Sleeve Gastrectomy	43.89	21301 (3.9%)	25 (0..1%)

Table 2. Multivariate logistic regression analysis for predictors of mortality

	Relative risk (95% CI)	P-value
Age >50	3.2 (2.5-4.0)	<.001
Male gender	2.5 (1.5-4.0)	<.001
Hypertension	04 (0.3-0.5)	<.001
Sleep apnea	2.4 (1.7-3.3)	<.001
Chronic pulmonary disease	2.3 (1.6-11.2)	<.001
Diabetes	0.6v(0.5-1.8)	<.001
Congestive heart failure	5.4 (2.6-11.2)	<.001
Smoking	1.4 (0.8-2.2)	NS
Open procedure	4.1 (3.3-5.0)	<.001

---

Session: Podium Presentation

Program Number: S121