Is Undernutrition Causally Associated with Deterioration of Outcomes in the Paediatric Intensive Care Unit (PICU): Systematic and Meta-analysis Review

Abstract

Abstract Background: Undernutrition (UN) refers to deficiencies or imbalances in a child’s intake and absorption of nutrients. Body mass index (BMI) and weight for height (W/H) provide sensitive markers for nutritional assessment as they reflect the child’s health status, growth and development. Objective: This systematic meta-analysis review examines the association between UN and paediatric intensive care unit (PICU) outcomes, such as mortality incidence, length of stay (LOS) and the need for and length of time on mechanical ventilation (MV). Methods: A search was conducted using the National Library of Medicine, SpringerLink and ScienceDirect. Observational studies on the association of low BMI or W/H with PICU outcomes were included. Pooled association estimates for the mortality rate and the need for MV were calculated using a random effects model. The fold change relative to normal nutrition (NN) was used to determine the LOS in the PICU and prolonged duration of MV. Results: The mortality incidence was 12% lower among the UN group compared to the NN group, but this was not statistically significant. The UN group had a 180% longer PICU stay than the NN group. An equal risk of MV was observed between the groups (RR: 1.08), but among MV patients in the UN group, the LOS was almost three-times longer than for the ventilated patients in the NN group. The study-specific results for mortality incidence receptor were moderately heterogeneous. I found that age, gender, study location and study quality could be confounders that impact this association. I concluded that children older than 3 years have a higher risk of mortality and the need for MV while in the PICU. In developing countries, UN patients are at a higher risk for increased LOS in the PICU and prolonged use of MV compared to NN patients. Conclusion: UN is associated with an increase in the risk of prolonged LOS in the PICU and prolong use the MV. However, children suffering from UN had a decreased incidence of death in the PICU compared to those with NN. Therefore, while the current evidence indicates a causal effect, publication bias and unmeasured confounders cannot be excluded.