Results of Surgical and Medical Rehabilitation for Adult Patients With Type III Intestinal Failure in a Comprehensive Unit Today: Building a New Model to Predict Parenteral Nutrition Independency

Background Short-bowel syndrome remains the primary cause of intestinal failure (IF) in adult patients. We aim to report the long-term results of medical and surgical rehabilitation in a cohort of patients with type III IF (III-IF) and develop a formula to predict parenteral nutrition (PN) independency. Methods We used a retrospective analysis of a prospective database for III-IF patients undergoing autologous gastrointestinal reconstruction surgery (AGIRS) from March 2006 to August 2018. Analyzed variables included demographic data, postsurgical intestinal length (PSIL), postsurgical anatomy, teduglutide (TED) treatment, and PN volume reduction. Univariate analysis, Cox regression, logistic regression forward stepwise models, and receiver operating characteristic (ROC) curve were done using SPSS v20. Results AGIRS was performed in 88 patients. The most frequent anatomy at first visit was type 1. Prevailing anatomy after surgery was type 3. Eight patients started TED; 6 achieved freedom from PN. At a mean follow-up time of 1606.1 +/- 1190.25 days, freedom from PN survival was achieved in 83%. Variables identified at the logistic regression analysis led to a novel formula to predict intestinal rehabilitation, including PSIL, presence of ileocecal valve, and use of TED as part of postsurgical treatment. Conclusions AGIRS in this group of patients enabled intestinal length increase and also intestinal anatomy conversion into a more favorable type for intestinal rehabilitation. TED treatment was useful to discontinue PN in patients with classical negative anatomical predictors. The novel predicting formula has an ROC area under the curve = 0.82. Further studies are necessary to validate this formula.