Long-term non-invasive ventilation in children with central nervous system disorders: A systematic review and meta-analysis

BACKGROUND: Long-term non-invasive ventilation (LT-NIV) is commonly used to treat sleep-related breathing disorders (SRBD) in children. Children with central nervous system (CNS) disorders experience a high rate of SRBDs. However, the outcomes of LT-NIV use for children with CNS disorders remains unclear. METHODS: This systematic review is a sub-study of a scoping review on LT-NIV use in children. The scoping review search strategy identified studies of children using LT-NIV from January 1990 to March 2024. These results were searched for studies of children with CNS disorders. To identify studies for meta-analysis, studies were grouped as: 1) studies of children with CNS disorders as part of broader a broader group of children using LT-NIV; 2) studies exclusively of children with CNS disorders using LT-NIV; and 3) studies of children with congenital central hypoventilation syndrome using LT-NIV. The Non-Randomized Studies of Interventions tool was used to assessed risk of bias. RESULTS: A total of 55 studies met inclusion criteria and included 2,015 children with CNS disorders using LT-NIV. Nineteen studies reported outcomes specific to children with CNS disorders. Meta-analysis of four studies showed no difference in mortality between children with and without CNS disorders(1.23, 95% CI: 0.40-3.79). Hospitalization rates across four studies suggested a higher rate of hospitalization in children with CNS disorders compared to children without CNS disorders. Meta-analysis of three studies showed reductions in the apnoea-hypopnoea index following LT-NIV use; the response to LT-NIV, however, varied across individual studies. CONCLUSION: LT-NIV use may benefit some children with CNS disorders, particularly through improvements in sleep-related breathing disorders. However, data remain limited, and uncertainty persists regarding the impact on mortality, hospitalization, and other important outcomes.