|
Back to 2016 Annual Meeting
Empiric Post-Operative Metoclopramide Reduces Delayed Gastric Emptying After Pancreaticoduodenectomy
Brandon Chapman*, James McCullough, Douglas Overbey, Alessandro Paniccia, Cheryl Meguid, Ana Gleisner, Martin McCarter, Csaba Gajdos, Richard D. Schulick, Barish H. Edil
Department of Surgery, University of Colorado School of Medicine, Denver, CO
Introduction: Delayed gastric emptying (DGE) is one of the leading postoperative complications following pancreaticoduodenectomy. Treatment with erythromycin has been shown to reduce DGE; however, the use of metoclopramide has not been reported in the literature. The purpose of this study is to determine if empiric use of metoclopramide reduces DGE.
Methods: We analyzed data from a prospectively collected single institutional database identifying patients that underwent open pancreaticoduodenectomy from June 2012 to August 2015. Patient demographics and clinical information including preoperative, intraoperative, and postoperative outcomes were queried from the medical record. Empiric use of metoclopramide was defined as receiving scheduled doses within 24 hours of surgery. Predictors of ISGPS-defined DGE were determined on univariate analysis and variables with a p-value of <0.10 were utilized on forward stepwise multivariate logistic regression.
Results: We identified 128 consecutive patients undergoing open pancreaticoduodenectomy and 25 (19.5%) received empiric metoclopramide. Overall, DGE was diagnosed in 39 (30.5%) patients: 2 (5.1%) patients receiving empiric metoclopramide and 37 (95.0%) patients not receiving empiric metoclopramide. The median duration of metoclopramide use was 3 days (range 3-13). On univariate analysis, only empiric metoclopramide (p=0.007), bile leak (p=0.002), pancreatic fistula (p=0.002), and diabetes mellitus (DM) (p=0.007) were statistically associated with DGE (Table 1). On multivariate analysis, a history of DM (OR 2.43, 95% CI 1.06 to 5.60) and pancreatic fistula (OR 2.69, 95% CI 1.34 to 5.42) increased the odds of developing DGE and empiric metoclopramide was protective (OR 0.20, 95% CI 0.05 to 0.90). Additionally, length of stay was significantly longer in patients with DGE (14 days, range 9-41) versus those without DGE (8 days, range 5-35)(p<0.001).
Conclusions: DGE is a common complication after pancreaticoduodenectomy resulting in significant morbidity and prolonged hospital stay. Our findings suggest the routine use of empiric metoclopramide within 24 hours after surgery may reduce the odds of developing DGE by 80 percent. However, further study is needed to confirm these findings.
| | DGE (n=39)
| No DGE (n=89)
| p-value | Patient Demographics
| | | | Age (years)
| 69 (29-88)
| 65 (19-86)
| 0.331
| Male gender
| 22 (56.4)
| 50 (56.2)
| 0.981
| Diabetes Mellitus
| 14 (35.9)
| 13 (14.6)
| 0.007
| | Pathology | | | 0.266
| Pancreatic adenocarcinoma
| 16 (41.0)
| 42 (69.5)
| | Cholangiocarcinoma
| 6 (15.4)
| 10 (11.2)
| | Pancreatic NET
| 4 (10.3)
| 9 (10.1)
| | Ampullary adenocarcinoma
| 4 (10.3)
| 8 (9.0)
| | IPMN
| 4 (10.3)
| 6 (6.7)
| | Ampullary NET
| 0 (0)
| 3 (3.4)
| | Duodenal adenocarcinoma
| 2 (5.1)
| 1 (1.1)
| | Chronic pancreatitis
| 0 (0)
| 1 (1.1)
| | | Other | 3 (7.7)
| 9 (10.1)
| | Intra-Operative Outcomes
| | | | Classic Whipple
| 39 (100)
| 86 (96.6)
| 0.246
| Estimated blood loss (mL)
| 600 (150-7000)
| 500 (125-2000)
| 0.174
| Operative time (min)
| 373 (175-579)
| 373 (207-616)
| 0.425
| Post-Operative Outcomes
| | | | Empiric metoclopramide
| 2 (5.1)
| 23 (25.8)
| 0.007
| Bile leak
| 4 (10.3)
| 0 (0)
| 0.002
| Intra-abdominal abscess
| 4 (10.3)
| 5 (5.6)
| 0.345
| Wound infection
| 12 (30.8)
| 17 (19.1)
| 0.147
| Pancreatic fistula
| 16 (41.0)
| 14 (15.7)
| 0.002
| Length of stay (days)
| 14 (9-41)
| 8 (5-35)
| <0.001
|
Table 1. Patient demographics and post-operative outcomes on univariate analysis. Median (range), n (%)
Back to 2016 Annual Meeting
|
