Definition, prevalence, and outcome of feeding intolerance in intensive care: a systematic review and meta-analysis

Clinicians and researchers frequently use the phrase 'feeding intolerance' (FI) as a descriptive term in enterally fed critically ill patients. We aimed to: (1) determine what is the most accepted definition of FI; (2) estimate the prevalence of FI; and (3) evaluate whether FI is associated with important outcomes. Systematic searches of peer-reviewed publications using PubMed, MEDLINE, and Web of Science were performed with studies reporting FI extracted. We identified 72 studies defining FI. In 33 studies, the definition was based on large gastric residual volumes (GRVs) together with other gastrointestinal symptoms, while 30 studies relied solely on large GRVs, six studies used inadequate delivery of enteral nutrition (EN) as a threshold, and three studies gastrointestinal symptoms without reference to GRV. The median volume used to define a 'large' GRV was 250 ml (ranges from 75 to 500 ml). The pooled proportion (n = 31 studies) of FI was 38.3% (95% CI 30.7-46.2). Five studies reported outcomes, all of them observed adverse outcome in FI patients. In three studies, respectively, FI was associated with increased mortality and ICU length-of-stay. In summary, FI is inconsistently defined but appears to occur frequently. There are preliminary data indicating that FI is associated with adverse outcomes. A standard definition of FI is required to determine the accuracy of these preliminary data.