Clinical outcomes in pediatric intestinal failure: a meta-analysis and meta-regression

Background: Intestinal failure (IF) is associated with significant morbidity and mortality, yet specific parameters that determine medium- and long-term outcomes remain ill defined. Objective: The aim of this study was to determine the long-term outcomes in childhood IF and identify patient characteristics associated with clinical endpoints. Design: MEDLINE and EMBASE were searched for cohorts of >10 pediatric-onset IF patients with >12 mo follow-up. Random-effects meta-analysis and meta-regression weighted by follow-up duration were used to calculate clinical outcome rates and patient factors associated with outcomes. Primary outcome was mortality rate; secondary outcomes included neurodevelopmental status, transplantation, IF-associated liver disease (IFALD), enteral autonomy, and sepsis. Results: In total, 175 cohorts (9318 patients and 34,549 y follow-up) were included in the meta-analysis. Overall mortality was 5.2% per y (95% CI: 4.3, 6.0) and was associated with sepsis and IFALD on meta-regression. Mortality rate improved with time from 5.9% per y pre-2000 to 4.5% per y post-2005. Sepsis rate was also predictive of IFALD and liver failure. Enteral autonomy was associated with small bowel length but not presence of ileo-cecal valve. There was a relative lack of data on neurodevelopmental outcomes. Conclusions: Sepsis is the primary modifiable factor associated with mortality and liver failure, whereas enteral autonomy correlates with small-bowel length. No clear parameters have been identified that accurately predict neurodevelopmental outcomes, and hence further research is needed. Together, our findings are helpful for parental counseling and resource planning, and support targeting reduction in sepsis.