|
Back to 2016 Annual Meeting
The Impact of Hepaticojejunostomy Anastomotic Leaks After Pancreatoduodenectomy: An Uncommon Complication but a Devastating Source of Morbidity and Mortality
Andrea Jester*, Catherine Chung, david becerra, E M. Kilbane, Michael G. House, Nicholas J. Zyromski, C. Max Schmidt, Attila Nakeeb, Eugene P. Ceppa
Surgery, Indiana University School of Medicine, Indianapolis, IN
Purpose: Pancreaticojejunostomy (PJ) leak is a common source of major morbidity after pancreatoduodenectomy (PD). Isolated hepaticojejunostomy leaks (HJ) are less frequent after PD and not well studied. The purpose of this study was to determine if isolated HJ leaks or the combination of PJ + HJ leaks result in worse postoperative outcomes when compared to no leak or isolated PJ leaks.
Methods: Consecutive cases of PD (n=924) were reviewed at a single high-volume institution over an 8-year period (November 2006-October 2014). Ninety-day outcomes were monitored through a prospectively maintained database. A retrospective analysis was performed to determine if there were statistically significant differences in patients with no leak, isolated PJ leak, isolated HJ leak, or combined PJ + HJ leaks.
Results: PJ leaks were identified in 268 out of 924 patients (29%); 169 grade A (63%), 64 grade B (23%), and 32 grade C (12%). Isolated HJ leaks were identified in 23 out of 924 PD patients (2.5%), and PJ + HJ leaks were identified in 32 patients (3.5%). Demographics including age, gender and ASA were similar between the 4 groups. Those with HJ leaks or PJ + HJ leaks had a significantly increased risk of 90-day all cause morbidity when compared to those with isolated PJ leaks (56% and 59% vs. 34 %, respectively, p = 0.04). Those with no leak experienced significantly less morbidity (23 %, p = 0.01). Isolated HJ and PJ + HJ leaks were associated with a significantly increased risk of unplanned re-intubation, prolonged ventilator times, septic shock, and renal failure. The median length of stay was longer for isolated HJ and PJ + HJ leaks when compared to isolated PJ leaks and no leak (18 and 21 days vs. 11 and 8 days, p = 0.01). The need for re-operation was significantly increased in the isolated HJ and PJ + HJ groups when compared to the isolated PJ leak and no leak groups (22 and 21.7 % vs. 7.8 and 4.5 %, respectively, p = 0.001). Re-admission rates were similar amongst the PJ, HJ and PJ + HJ leak groups (25, 28 and 31 %, respectively, p = 0.8), but were higher than the no leak group (14 %, p = 0.02). Overall 30-day mortality was 3.6 %. Isolated HJ leaks and PJ + HJ leaks were associated with higher 30-day mortality rates compared to those with isolated PJ leaks or no leak (13 and 12.5 % vs. 4.5 and 2.7 %, p = 0.04 and 0.02, respectively).
Conclusions: Leaks associated with the hepaticojejunostomy anastomosis after pancreatoduodenectomy result in significant risk for postoperative morbidity and mortality.
Morbidity and mortality after pancreatoduodenectomy: No leak vs. PJ leak vs. HJ leak vs. PJ + HJ leak
| Group | # Pts | Overall 90-day Major Morbidity | 30-day Mortality | Re-operation | 30-day Hospital Re-admission | Septic Shock | LOS (days) | | No Leak | 602 | 23.3 % | 2.7 % | 4.5 % | 14.7 % | 2.7 % | 8 | | PJ Leak | 268 | 34 % * | 4.5 % | 7.8 % | 25 % * | 9.7 % | 11 | | HJ Leak | 23 | 56 % * # | 13 % * # | 22 % * # | 26 % * | 18 % * # | 18 * # | | PJ + HJ Leak | 32 | 59 % * # | 12.5 % * # | 22 % * # | 31 % * | 34 % * # | 21 * # |
* denotes p < 0.05 no leak; # denotes p < 0.05 PJ leak
Back to 2016 Annual Meeting
|
