Characterising oral feeding intolerance in patients with acute pancreatitis

Abstract

Nutrition is an important component of the management of acute pancreatitis (AP), particularly given the role of oral feeding intolerance (OFI) in prolonging hospitalisation, increasing treatment costs, and reducing patient quality of life. Oral feeding intolerance is a common complication of AP, characterised by relapse of symptoms following the reintroduction of oral diet. The pooled global incidence of OFI is currently unknown and, despite preliminary investigations of potential predictors by several studies, its pathogenesis is poorly understood. The first aim of this thesis was to determine the global incidence of OFI through systematic review of the literature and meta-analysis of all reported incidences of OFI. Secondly, this thesis aimed to evaluate the current evidence regarding the predictors of OFI, by extracting data from relevant literature, and meta-analysing the results where possible. The third aim was to investigate whether gastrointestinal motility, can be used as predictor of OFI, by determining the relationship between OFI and the gastroparesis cardinal symptom index (GCSI), and developing a predictive nomogram for use in clinical settings. The results of this thesis showed that OFI is a common complication that affects approximately 1 in 6 patients with AP. Additionally, meta-analyses of potential predictors that were identified in the literature demonstrated that serum lipase prior to refeeding, pleural effusion, and (peri)pancreatic collections are all significant predictors of OFI. Finally, GI motility, measured by the GCSI, was found to be significantly associated with the development of OFI. The GCSI nausea/vomiting sub-score obtained on day 2 of admission was found to have the greatest predictive value of all GCSI scores investigated, and was used to develop a nomogram to predict OFI risk. Future research should focus on two main areas; firstly, superiority trials are needed to determine the optimal mode and timing of refeeding, in order to better inform nutritional management guidelines. Secondly, future studies should build upon the existing research regarding predictors of OFI, particularly serum lipase prior to refeeding and the GCSI nausea/vomiting sub-score on day 2 of admission, which were both identified as promising predictive measures in this thesis. External validation of the predictive nomogram developed and presented in this thesis is also necessary, if it is to be a clinically useful predictive tool.