A systematic review of the definitions and prevalence of feeding intolerance in critically ill adults

Background & aims The term enteral feeding intolerance (FI) is frequently used in clinical practice and the literature, yet there is no standardised definition. FI is often quoted as a reason for failure to meet enteral nutrition (EN) targets but the lack of a consensus definition precludes accurate estimates of prevalence, predictors and clinical outcomes associated with FI. A systematic review was performed of studies in adult critical care patients to evaluate the definitions, relative risk, predictors and clinical outcomes of FI and to propose a uniform definition. Methods Database searches were completed in MEDLINE Ovid, Embase, CINAHL, PsycINFO, Google Scholar, NHS Evidence, Scopus and Web of Science. The search was performed in January and February 2021. Studies were included if they had an interventional, observational cohort or case–control study design and contained a definition of FI in critically ill adults. The following data were extracted from each included article: 1) study design; 2) study objective; 3) inclusion criteria; 4) population and setting; 5) sample size; 6) definition of FI; 7) prevalence of FI; 8) predictors of FI; 9) clinical outcome measures associated with FI. Studies were grouped based on the symptoms used to define FI with random effects meta-analysis. Results 89 unique studies containing a definition of FI were identified. Studies were categorised according to definition of FI into 3 groups: 1) Gastric residual volume (GRV) and/or gastrointestinal (GI) symptoms (n = 74); 2) Ability to achieve EN target (n = 5); 3) Composite definitions (n = 10). Meta-analysis showed a relative risk of FI of 0.55 [95% CI 0.45, 0.68] (p < 0.00001). The most frequently reported predictors of FI were use of vasoactive drugs, sedation or use of muscle relaxants, intra-abdominal pressure and APACHE II score. Conclusions FI is inconsistently defined in the literature but is reportedly common amongst critically ill adults. FI is most frequently defined by the presence of raised GRV and GI symptoms. However, studies show GRV to correlate poorly with delayed gastric emptying and this review demonstrated no correlation between GRV threshold and prevalence of FI. A standardised definition of FI is essential for future research and clinical practice. We propose a definition of FI including a failure to reach EN targets in addition to presence of GI symptoms. Protocol registration PROSPERO number CRD42020211879. Registered 29th September 2020.