Back to Annual Meeting Posters

Introduction: Bariatric gastric bypass surgery techniques have evolved over the past decade in an effort to minimize complications. One serious immediate complication after Roux-en-Y gastric bypass (RYGB) is postoperative bleeding. We examined the impact of postoperative bleeding on patient outcomes after RYGB and studied the trend over the past decade.
Methods: We used the Nationwide Inpatient Sample (NIS) to calculate outcomes every 5 years from 2000 to 2010. The NIS is the largest nationally representative publically available inpatient database in the United States. Patients were included if they had an ICD-9 CM code indicating an open or laparoscopic RYGB. Exclusion criteria were age <18, previous weight loss surgery, history of a GI malignancy, inflammatory bowel disease, infectious colitis, and non-elective admission. Significant bleeding was defined as ICD-9 CM code for packed red blood cell transfusion postoperatively. Bleeding-related endoscopy rate was defined as the difference in percent endoscopy between patients with and without postoperative bleeding. Additional length of stay and additional charge were defined as the difference in median length of stay and charge, respectively, between the patients with and without postoperative bleeding. Charge was adjusted for inflation using the consumer price index, and is presented in 2010 US dollars. Comorbidities were identified using the Charlson comorbidity index.
Results: The incidence of RYGB increased markedly from 2000 to 2005, and then stablized from 2005 to 2010. The proportion of laparoscopic RYGB has increased from 0% in 2000 to 90% in 2010. Although the mean age has remained consistent, age distribution has broadened over time, and the comorbidity burden has grown. Over time, the proportion of surgeries done at teaching or urban hospitals decreased. Postoperative bleeding rate increased by 33% between 2000 and 2010, as did the bleeding-attributable risk of shock. However, bleeding-attributable mortality decreased over time, as did the rate of endoscopy and reoperation for bleeding. Additional length of stay secondary to postoperative bleeding has decreased over time; however, total length of stay and total financial burden related to postoperative bleeding have increased steadily and in 2010 they were 3681 person-days and $37 million, respectively (Table 1).
Conclusion: The number of patients undergoing RYGB has substantially increased from 2000 to 2005, and has since stabilized. The postoperative bleeding rate has increased substantially since 2000, but bleeding-attributable mortality, rate of endoscopy and rate of reoperation for bleeding has decreased. The total length of stay and financial burden of postoperative bleeding in patients undergoing RYGB continue to increase.

	2000	2005	2010
Number of RYGB surgery
Any RYGB	23,697	88,571	71,199
Open RYGB	100%	27%	10%
Laparoscopic RYGB	0%	73%	90%
Median age (IQR)	41 (33-48)	42 (34-51)	44 (35-53)
Female	84%	82%	79%
Charlson comorbidity index	0: 67% 1: 25% 2: 5%	0: 57% 1: 34% 2: 7%	0: 50% 1: 37% 2: 10%
Teaching hospital	79%	55%	58%
Urban hospital	98%	97%	92%
Postoperative bleeding
Any RYGB	1.8% (426 patients)	2.2% (1956 patients)	2.4% (1681 patients)
Open RYGB	1.8% (426 patients)	3.1% (751 patients)	5.6% (412 patients)
Laparoscopic RYGB	-	1.9% (1209 patients)	2.0% (1269 patients)
Bleeding-attributable risk of death	4.1%	1.64%	1.9%
Bleeding-attributable risk of shock	0.0%	1.8%	2.0%
Reoperation for bleeding	26.6%	14.8%	6.8%
Endoscopy for bleeding	5.0%	8.9%	3.5%
Bleeding-attributable LOS (median, days): Bleeding-attributable LOS (patient-days):	3.1 1325	2.1 4010	2.2 3681
Bleeding-attributable hospitalization charge (Indexed to 2010 USD)	Median: $34,722 Total: $14,791,572	Median:$17,191 Total: $33,625,596	Median: $22,104 Total: $ 37,156,824