Development and validation of a predictive model for feeding intolerance in intensive care unit patients with sepsis

Abstract

Background: Feeding intolerance in patients with sepsis is associated with a lower enteral nutrition (EN) intake and worse clinical outcomes. The aim of this study was to develop and validate a predictive model for enteral feeding intolerance in the intensive care unit patients with sepsis. Methods: In this dual-center, retrospective, case-control study, a total of 195 intensive care unit patients with sepsis were enrolled from June 2018 to June 2020. Data of 124 patients for 27 clinical indicators from one hospital were used to train the model, and data from 71 patients from another hospital were used to assess the external predictive performance. The predictive models included logistic regression, naive Bayesian, random forest, gradient boosting tree, and deep learning (multilayer artificial neural network) models. Results: Eighty-six (44.1%) patients were diagnosed with enteral feeding intolerance. The deep learning model achieved the best performance, with areas under the receiver operating characteristic curve of 0.82 (95% confidence interval = 0.74–0.90) and 0.79 (95% confidence interval = 0.68–0.89) in the training and external sets, respectively. The deep learning model showed good calibration; based on the decision curve analysis, the model's clinical benefit was considered useful. Lower respiratory tract infection was the most important contributing factor, followed by peptide EN and shock. Conclusions: The new prediction model based on deep learning can effectively predict enteral feeding intolerance in intensive care unit patients with sepsis. Simple clinical information such as infection site, nutrient type, and septic shock can be useful in stratifying a septic patient's risk of EN intolerance.

Gut Microbiota and Enteral Nutrition Tolerance in Non-Abdominal Infection Septic ICU Patients: An Observational Study

“…Heyland et al [5] reported in an international survey that age, region (Asia), burns, and a high APACHE II score may be associated with EN intolerance. Hu et al [12] reported 15 factors that may affect EN tolerance, such as pneumonia, shock, infection site, and EN formula. However, this study design was not suitable for studying the relationship between gut microbiota and EN tolerance, as the population had great heterogeneity, which may lead to difficulties in the analysis of the data.…”

Clinical practice guidelines for nutritional assessment and monitoring of adult ICU patients in China

“…The studies included 10 prospective studies [ [114] , [115] , [116] , [117] , [118] , [119] , [120] , [121] , [122] , [123] ] and 16 retrospective studies. [ [124] , [125] , [126] , [127] , [128] , [129] , [130] , [131] , [132] , [133] , [134] , [135] , [136] , [137] , [138] , [139] ] The overall incidence rate of feeding intolerance in critically ill patients was found to be 0.40 (95% CI: 0.34 to 0.46), indicating a substantial association with adverse patient-centered outcomes. Particularly noteworthy is the OR for all-cause ICU mortality risk, which was 1.99 (95% CI: 1.69 to 2.35).…”

The future of artificial intelligence in clinical nutrition

“…It refers to the use of data from electronic health records to predict the occurrence of certain events or the best time to give treatment. Hu et al [18] developed and validated a predictive model for enteral feeding intolerance (EFI) in ICU patients with sepsis. In this dual-center, retrospective, case-control study, a total of 195 ICU patients with sepsis, who stayed at an ICU for at least 7 days and received enteral nutrition, were enrolled.…”