Can we improve the management of inoperable malignant bowel obstruction? Results of a feasibility study of elemental diet as an alternative to parenteral nutrition in patients with advanced gynaecological cancer

Abstract

Purpose Nutrition support in inoperable bowel obstruction (IBO) remains challenging. Parenteral nutrition (PN) is recommended if the prognosis is > 2 months. An elemental diet (ED) is licensed for strictures in Crohn’s disease but has not been used in malignant bowel obstruction. The aim of this study was to evaluate the use of ED in patients with IBO and provide a proof of concept of ED as an acceptable feeding option. Methods This was a mixed-methods single-arm feasibility study. The primary endpoint was to provide a ‘proof of concept’ of ED as an acceptable feeding option for patients with IBO. Secondary endpoints included taste acceptability, incidences of vomiting and pain, the proportion of women who tolerated ED, the number of cartons drunk, quality of life (QOL) and the number of women treated with chemotherapy. Patients (> 18 years) with CT-confirmed IBO who could tolerate 500 ml of liquid in 24 h remained on the trial for 2 weeks. Results A total of 29 patients were recruited; of those, 19 contributed to the analysis for the primary endpoint; 13 (68.4%) participants tolerated the ED; 26 patients contributed to MSAS and EORTC QLQ questionnaires at baseline to allow for the assessment of symptoms. At the start of the study, 18 (69%) of patients experienced vomiting, reducing to 4 (25%) by the end of day 15 of the study; 24 (92%) of patients reported pain at consent, reducing to 12 (75%) by the end of day 15. QOL scores improved from 36.2 (95% CI 27.7–44.7) at baseline to 53.1 (95% CI 40.3–66) at the end of day 15; 16 (84%) participants commenced chemotherapy within the first week of starting ED. The number of cartons across all participants showed a median of 1.3 cartons per day (range 0.8 to 2.5). Conclusion ED is well tolerated by patients with IBO caused by gynaecological malignancies and may have a positive effect on symptom burden and QOL.