Introduction: Pancreas-kidney transplantation remains the gold standard treatment for patients with Type I Diabetes Mellitus and End-Stage Renal Failure as it is the only one that treats renal failure whilst also restoring normoglycaemia. The technique of choice for handling the exocrine component is enteric drainage, with published 5-year patient and graft survival rates being 85% and 75% respectively. Despite its long term success, it is a highly morbid procedure. An enteric leak (EL) of the duodenal stump is one of the more challenging complications specific to Pancreas Transplantation, with an incidence of 4-10%. Interestingly, recent research has demonstrated that the rates of ELs might be lower than previously thought. The objective of our study was to explore if this was true and to determine if there were additional factors that increased the risk of ELs.
Methods: All pancreas transplants at Westmead Hospital performed between January 2008 and August 2017 (n = 234) were analysed in order to identify ELs that occurred at any stage post-operatively. Risk factors for ELs including donor (age, gender, BMI, hypertension, diabetes, smoking status, vascular disease), patient (age, gender, BMI, hypertension, diabetes, smoking status, alcohol use, vascular disease, lowest blood pressure during admission) and transplantation procedure characteristics (vasopressor use, antibiotics administered, blood pressure, ischaemic time, anastomotic time, fluid boluses and graft types) were collected and analysed. Adjusting for possible confounders, a multivariate logistic regression model was used to assess the risk and predictors of enteric leaks using SPSS v. 22.0. 
Results and Discussion: Of the 234 recipients, 12 (5%) experienced an EL. Of these, 9 (75%) recipients had vascular disease, 6 (50%) recipients were ex-smokers, 1 (8%) recipient a current smoker and 3 (25%) recipients were obese with a BMI >30kg/m2. Six (50%) donors were overweight with a BMI >25kg/m2. All 12 patients were returned to theatre and converted to bladder drainage. At time of publication, no patients had experienced graft failure. The risk of EL increases by as much as 4.4 fold in recipients with vascular disease (OR: 4.4; 95% CI: 0.80-24.21; P=0.088). Although the association was weaker, other factors such as recipient BMI >24.2kg/m2 increased the risk of EL by as much as 1.8 fold (OR: 1.8; 95% CI: 0.4-9.3; P=0.46). The risk of EL increases by more than 3 fold in donors aged between 23 to 32 years compared to <22 years (OR: 3.1; 95% CI: 0.3-32.8; P=0.34). These odds increase by as much as 5 fold for donors ≥33 years (OR: 5.0; 95% CI: 0.5-52.5; P=0.180). The risk of EL increases 1.3 times in male recipients (OR: 1.3; 95% CI: 0.2-8.0; P=0.77) and when the donor was female (OR: 1.3; 95% CI: 0.3-6.9; P=0.75). Finally, the risk of EL increases by as much as two fold for a donor with a BMI >25.4kg/m2 (OR: 2.1; 95% CI: 0.3-15.3; P=0.463) and is 1.3 times higher for a donor with a BMI between 22.5-25.4kg/m2 (OR: 1.3; 95% CI: 0.2-10.9; P=0.80) compared to a donor with a BMI of <22.5kg/m2.
Conclusions: At Westmead in our patients over the past 10 years, the rate of enteric leaks is in agreement with the rates previously described in the literature. When controlling for all variables, there was weak evidence of an association between vascular disease and ELs. These findings identify possible risk factors for ELs, the need for further research in this area and careful donor selection.