Investigating the prevalence, predictors and outcomes associated with enteral feeding intolerance in critically ill adults: a retrospective cohort study
Investigating the prevalence, predictors and outcomes associated with enteral feeding intolerance in critically ill adults: a retrospective cohort study
Background & aims: enteral feeding intolerance (EFI) is a common cause for failure to meet nutritional targets in critically ill patients. However, EFI is poorly defined, and prevalence varies according to the definition used. Predictors and outcomes associated with EFI have also been difficult to quantify due to the lack of a standardised definition. A recent systematic review proposed a definition of EFI as a failure to meet 80% of nutritional targets accompanied by one or more of the following gastrointestinal symptoms: vomiting/regurgitation, abdominal distention or diarrhoea. The aim of this study was to determine the prevalence of EFI in critically ill adults according to the definition proposed and: i) to examine associated predictors and ii) the impact on patients’ outcomes including intensive care unit (ICU) length of stay and mortality.
Methods: a single centre retrospective cohort study collected data on 400 patients admitted to a tertiary critical care unit in the UK. Data were collected on demographic variables, use of mechanical ventilation, medication, nutrition (calorie/protein targets; calories/protein received), nursed prone, gastrointestinal symptoms, intensive care length of stay, and 60-day mortality. Non-parametric tests and multivariable logistic regression were used to examine predictors of EFI and associations with ICU length of stay and 60-day mortality.
Results: 47.7% (n=191/400) of patients experienced at least one day of EFI. The median duration of intolerance was 1 day (range 1-21 days). Statistically significant predictors of EFI were fentanyl use (p<0.001), dexmedetomidine use (p=0.045) and hepato-pancreato-biliary disease (p=0.011). ICU length of stay was statistically significantly increased in patients with EFI compared to those without (p<0.001).
Conclusions: EFI is common in critically ill patients; however, for the majority, this intolerance is of short duration. Early identification and treatment strategies may reduce ICU length of stay, but further research is needed to investigate this.
Critical illness, Feeding intolerance, Intensive care, Prevalence
Jenkins, Bethan
ed7a7ed9-8725-424e-86df-efc1f81d5aba
Hossain, M. Jamal
3b4f5a47-c0a3-407b-88c0-ec936e70faf3
Beja, Maria Margarida
b4037fc5-4df9-402d-82c9-6ec4bca57af1
Calder, Philip C.
1797e54f-378e-4dcb-80a4-3e30018f07a6
Marino, Luise V.
ae989c9e-0696-45fd-bfb9-a8a72c5c2623
16 April 2026
Jenkins, Bethan
ed7a7ed9-8725-424e-86df-efc1f81d5aba
Hossain, M. Jamal
3b4f5a47-c0a3-407b-88c0-ec936e70faf3
Beja, Maria Margarida
b4037fc5-4df9-402d-82c9-6ec4bca57af1
Calder, Philip C.
1797e54f-378e-4dcb-80a4-3e30018f07a6
Marino, Luise V.
ae989c9e-0696-45fd-bfb9-a8a72c5c2623
Jenkins, Bethan, Hossain, M. Jamal, Beja, Maria Margarida, Calder, Philip C. and Marino, Luise V.
(2026)
Investigating the prevalence, predictors and outcomes associated with enteral feeding intolerance in critically ill adults: a retrospective cohort study.
Clinical Nutrition ESPEN, 73, [103269].
(doi:10.1016/j.clnesp.2026.103269).
Abstract
Background & aims: enteral feeding intolerance (EFI) is a common cause for failure to meet nutritional targets in critically ill patients. However, EFI is poorly defined, and prevalence varies according to the definition used. Predictors and outcomes associated with EFI have also been difficult to quantify due to the lack of a standardised definition. A recent systematic review proposed a definition of EFI as a failure to meet 80% of nutritional targets accompanied by one or more of the following gastrointestinal symptoms: vomiting/regurgitation, abdominal distention or diarrhoea. The aim of this study was to determine the prevalence of EFI in critically ill adults according to the definition proposed and: i) to examine associated predictors and ii) the impact on patients’ outcomes including intensive care unit (ICU) length of stay and mortality.
Methods: a single centre retrospective cohort study collected data on 400 patients admitted to a tertiary critical care unit in the UK. Data were collected on demographic variables, use of mechanical ventilation, medication, nutrition (calorie/protein targets; calories/protein received), nursed prone, gastrointestinal symptoms, intensive care length of stay, and 60-day mortality. Non-parametric tests and multivariable logistic regression were used to examine predictors of EFI and associations with ICU length of stay and 60-day mortality.
Results: 47.7% (n=191/400) of patients experienced at least one day of EFI. The median duration of intolerance was 1 day (range 1-21 days). Statistically significant predictors of EFI were fentanyl use (p<0.001), dexmedetomidine use (p=0.045) and hepato-pancreato-biliary disease (p=0.011). ICU length of stay was statistically significantly increased in patients with EFI compared to those without (p<0.001).
Conclusions: EFI is common in critically ill patients; however, for the majority, this intolerance is of short duration. Early identification and treatment strategies may reduce ICU length of stay, but further research is needed to investigate this.
Text
Bethan Jernkins_Clin Nutr ESPEN_Accepted version
- Accepted Manuscript
Restricted to Repository staff only until 1 April 2027.
Request a copy
More information
Accepted/In Press date: 28 March 2026
e-pub ahead of print date: 1 April 2026
Published date: 16 April 2026
Keywords:
Critical illness, Feeding intolerance, Intensive care, Prevalence
Identifiers
Local EPrints ID: 511411
URI: http://eprints.soton.ac.uk/id/eprint/511411
ISSN: 2405-4577
PURE UUID: cf8fcdb3-c4eb-490b-87ca-6112273c3a82
Catalogue record
Date deposited: 14 May 2026 16:30
Last modified: 15 May 2026 02:07
Export record
Altmetrics
Contributors
Author:
Bethan Jenkins
Author:
M. Jamal Hossain
Author:
Maria Margarida Beja
Author:
Luise V. Marino
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics