ACUTE PANCREATITIS IN PATIENTS WITH A HISTORY OF BARIATRIC SURGERY: IS IT LESS SEVERE?
Paul T. Kroner*1, C. Roberto Simons-Linares2, Alex M. Kesler1, Peter Abader1, Mohammad Afsh1, Juan E. Corral1, Prabhleen Chahal2, Massimo Raimondo1
1Mayo Clinic Florida, Jacksonville, FL; 2Gastroenterology and Hepatology, Cleveland Clinic, Cleveland, OH
Introduction
Bariatric surgery (BS) leads to weight loss and causes alterations in gastrointestinal and pancreatic peptides that assist in treating diabetes mellitus (DM). This raises question on acute pancreatitis (AP) clinical outcomes in this population. Older studies showed improved clinical outcomes of AP in post-BS patients. However, these studies were performed at the time when Roux-en-Y gastric bypass was the BS procedure of choice, while sleeve gastrectomy predominates today. Our aim was to assess mortality, morbidity, and resource utilization of AP in patients with BS in the past 5 years.
Methods
Observational retrospective cohort study using the NIS 2012-2016, the largest public inpatient database in the US. Patients with a principal diagnostic ICD9-10CM code for AP were included. None were excluded. Stratification for the coexistence of history of BS was performed. The primary outcome was determining inpatient mortality of AP in patients with history of BS, stratified for biliary AP etiology. Secondary outcomes were determining morbidity, resource utilization, length of hospital stay (LOS), total hospital charges and costs, which were adjusted for inflation using the Consumer Price Index. Propensity score matching was used to create a matching population for gender, age and Charlson Comorbidity Index. Multivariate analyses were used to adjust for income in patient zip code, hospital region, location, size and teaching status.
Results
920,615 patients with AP were identified, of which 15,345 had undergone BS. After propensity matching 8,220 patients with BS had AP. The mean age was 53 years and 85% were female. On multivariate analysis, the mortality odds for AP was 0.42 (p<0.01), biliary AP 0.29 (<0.04) in the history of BS group compared to patients with no BS history. For secondary outcomes, adjusted odds of acute kidney insufficiency (AKI), shock, ICU, multi-organ failure, ERCP, costs, charges, and LOS were all lower for patients with AP who had history of BS compared to patients who did not (Table 1). Patients with biliary AP, showed even lower odds of AKI, ICU, multi-organ failure, costs, charges, and LOS, but higher odds of cholecystectomy (Table 2).
Conclusion
Patients with acute pancreatitis with history of bariatric surgery have lower odds of mortality, morbidity and resource utilization compared to patients with no history of BS. One can speculate this may be due to post-surgical alterations in pancreatic and gastrointestinal functions including hormonal and anatomical changes. Interestingly, patients with biliary AP and BS seemed to have even lower odds of morbidity and mortality than patients with non-biliary AP, suggesting an added benefit with milder disease course. Studies are needed to determine the exact mechanism which contributes to improved outcomes, as well as stratification for the different types of BS procedures.
Table 1
| Variable | aOR | 95% CI | p-value |
| Overall AP Mortality | 0.42 | 0.28-0.63 | <0.01 |
| AKI | 0.47 | 0.41-0.53 | <0.01 |
| Shock | 0.78 | 0.92-1.06 | 0.03 |
| ICU | 0.41 | 0.32-0.52 | <0.01 |
| TPN | 1.08 | 0.83-1.19 | 0.53 |
| CT Abdomen | 0.93 | 0.69-1.26 | 0.65 |
| Ultrasound | 0.98 | 0.70-1.38 | 0.92 |
| Multi-organ Failure | 0.46 | 0.40-0.52 | <0.01 |
| All Cholecystectomies | 1.53 | 1.39-1.68 | <0.01 |
| Open cholecystectomy | 1.28 | 0.91-1.78 | 0.15 |
| Lap cholecystectomy | 1.53 | 1.39-1.69 | <0.01 |
| ERCP | 0.69 | 0.59-0.80 | <0.01 |
| aMean | 95% CI | p-value | |
| Additional Costs | -$2,805 | -$3,401 - -$2,208 | <0.01 |
| Additional Charges | -$10,850 | -$13,318 - -$8,382 | <0.01 |
| Additional LOS | -1.00 | -1.2 - -0.82 | <0.01 |
Adjusted odds ratios and additional means for the evaluated factors in patients with AP and history of BS compared to patients with no history of BS.
Table 2
| Variable | aOR | 95% CI | p-value |
| Biliary AP Mortality | 0.29 | 0.09-1.39 | <0.04 |
| AKI | 0.39 | 0.29-0.53 | <0.01 |
| Shock | 0.65 | 0.38-1.10 | 0.11 |
| ICU | 0.26 | 0.14-0.51 | <0.01 |
| TPN | 0.71 | 0.38-1.33 | 0.29 |
| CT Abdomen | 1.22 | 0.68-2.16 | 0.50 |
| Ultrasound | 0.83 | 0.47-1.45 | 0.51 |
| Multi-organ Failure | 0.36 | 0.26-0.48 | <0.01 |
| All Cholecystectomies | 1.57 | 1.34-1.83 | <0.01 |
| Open Cholecystectomy | 1.32 | 0.91-1.91 | 0.14 |
| Lap Cholecystectomy | 1.49 | 1.28-1.74 | <0.01 |
| ERCP | 0.56 | 0.46-0.68 | <0.01 |
| aMean | 95% CI | p-value | |
| Additional Costs | -$2,450 | -$3,223 - -$1,677 | <0.01 |
| Additional Charges | -$9,027 | -$12,813 - -$5,240 | <0.01 |
| Additional LOS | -1.3 | -1.54 - -1.01 | <0.01 |
Adjusted odds ratios and additional means for the evaluated factors in patients with Biliary AP and history of BS compared to patients with no history of BS.
Back to 2019 Abstracts
