Innovative Computerized Support System Shortens Ventilation Duration in Children with Lung Failure

Researchers at Children's Hospital Los Angeles have conducted the first clinical trial demonstrating that a computerized decision support (CDS) tool can effectively reduce the duration of mechanical ventilation needed by children suffering from acute lung failure. This study, published in NEJM Evidence, shows promising results for leveraging AI-driven tools to enhance respiratory care in pediatric patients.

The trial involved 248 children diagnosed with pediatric acute respiratory distress syndrome (PARDS). Participants were randomized to receive either standard ventilation managed solely by physicians or an advanced ventilation strategy guided by a lung and diaphragm-protective CDS tool. The results were significant: children managed with the CDS experienced a median weaning time of approximately 0.09 days, compared to 1.04 days in the traditional care group.

Further benefits included better preservation of respiratory muscle strength during early spontaneous breathing trials and improved functional status at ICU discharge among the CDS group. Notably, children leaving the ICU after CDS-supported ventilation showed fewer signs of debilitation, suggesting a reduction in ventilator-associated complications.

The CDS system functions by recommending frequent adjustments to ventilator settings based on ongoing assessment of each child's breathing patterns. Dr. Robinder Khemani, the study's lead author, emphasized the significance: "This is the first trial in both adults and children to test a computerized ventilation approach aimed at safeguarding lung and diaphragm function. The ability to make individualized, reproducible decisions in real-time can greatly improve patient outcomes."

Despite the positive findings, both groups demonstrated similar respiratory strength at hospital discharge, indicating that the early benefits may primarily influence immediate recovery and ICU experience.

This innovative approach addresses the complexity of mechanical ventilation management, where balancing risks and individual needs is often challenging. Incorporating computer-driven decision support offers a reproducible method to optimize ventilation strategies, potentially transforming pediatric respiratory care.

For more details, see the full study: Robinder G. Khemani et al, 'Randomized Trial of Lung and Diaphragm Protective Ventilation in Children,' NEJM Evidence, 2025.