Duration of Respiratory Support in Latin American NICUs: A Competing Risks Analysis by Altitude

Objective: We aimed to determine whether the duration of assisted ventilation is independently associated with receiving care at low- or high-altitude NICUs in Latin America among preterm infants born at ≤32 weeks of gestation.

Study Design: We conducted a multicenter observational cohort study using data from the EpicLatino registry (2015–2022), which contains prospectively collected clinical information from 32 NICUs in Latin America and the Caribbean. Assisted ventilation was defined as the use of any form of invasive mechanical ventilation, non-invasive positive pressure ventilation, or high-flow nasal cannula. Altitude was categorized as high (≥2,000 meters above sea level) or low (<2,000 meters). We used competing risks regression and mixed-effects models to evaluate the association between altitude and the duration of assisted ventilation, adjusting for clinical characteristics and center-level variation.

Results: We included 4,428 preterm infants; 81.4 % (n = 3,604) had assisted ventilation discontinue and 18.6 % (n = 824) died during hospitalization. The median duration of respiratory support was 8 days (4–23 IQR). In unadjusted analyses, infants in high-altitude centers (≥2,000 meters) had shorter durations of ventilation than those at lower altitudes (14 vs. 21 days; p < 0.001). In multivariable models adjusted for gestational age, necrotizing enterocolitis, admission period, and inborn status, altitude was not significantly associated with ventilation duration. A Cox model with shared frailty by center (HR = 1.17; 95 % CI: 0.87–1.57; p = 0.289), and a competing risks regression model with fixed effects for center (SHR = 1.09; 95 % CI: 0.86–1.39; p = 0.47).

Conclusions: Observed differences in the duration of assisted ventilation between high- and low- altitude centers were not independently associated with altitude after adjusting clinical characteristics and center. These findings suggest that institutional factors may explain much of the variation in ventilation practices. Standardized, context-adapted respiratory care protocols could help reduce practice variability across NICUs in the region.