Prevalence, management, health-care burden, and 90-day outcomes of prolonged mechanical ventilation in the paediatric intensive care unit (LongVentKids): an international, prospective, cross-sectional cohort study.

Abstract

Background: The number of children requiring prolonged mechanical ventilation (PMV) has increased with the advancement of medical care. We aimed to estimate the prevalence of PMV worldwide, document demographic and clinical characteristics of children requiring PMV in paediatric intensive care units (PICUs), and to understand variation in clinical practice and health-care burden.

Methods: This international, multicentre, cross-sectional cohort study screened participating PICUs in 28 countries for children aged >37 postgestational weeks to 17 years who had been receiving mechanical ventilation (MV; invasive or non-invasive) for at least 14 consecutive days. Screening days took place every 90 days for 3 years. Patients were eligible for inclusion in the analysis if they had been receiving MV (invasive or non-invasive) for at least 14 consecutive days by their first day of screening. Eligible patients were followed up on the subsequent screening day 90 days later or at time of hospital discharge, whichever came first. Outcome data were recorded in a validated web-based case report file. The primary outcome was the prevalence of PMV. Secondary outcomes were mortality, duration of MV, tracheostomy, and number of complications. All outcomes were assessed at 90 days post-screening. The study was registered with ClinicalTrials.gov, NCT04112459.

Findings: Between Sept 4, 2019 and Dec 7, 2022, 14 595 children were screened on four separate screening days in 158 PICUs, and 2773 patients had been receiving MV for at least 14 days and were included in the analysis. The point prevalence of PMV was 25·8% (IQR 24·1-28·5). Median age was 0·4 years (IQR 0·2-5·3) and median weight was 8·1 kg (IQR 4·7-19·1). 625 (24·0%) of 2610 patients had a history of prematurity (<37 weeks gestational age at birth). 90-day outcome data were collected for 2430 patients. 441 (18·2%) of 2430 patients had died within 90 days. 649 (29·8%) of 2176 patients who initiated ventilation support upon hospital admission had a tracheostomy placed after the first 14 days of MV. The median time to tracheostomy placement after MV initiation was 26 days (IQR 18-52). 462 (21·2%) of 2176 patients had at least one failed extubation between MV initiation and their first screening date. 556 (25·6%) of 2174 patients who started MV upon hospital admission required MV for 21 days or less, whereas 1618 (74·4%) patients required MV for 22 days or more; 90-day mortality did not differ between these groups (18·2% vs 20·30%, p=0·288). Complications were recorded for 810 (38·4%) 2109 patients who initiated MV upon hospital admission; of these 539 (67%) had ventilator-associated pneumonia, and 212 (39%) of 539 patients had multiple episodes of ventilator-associated pneumonia.

Interpretation: Timing of tracheostomy was variable, and duration of MV was longer than previously reported. The large variability in patients requiring MV and the associated health-care burden and outcomes across PICUs suggest that further investigation of the factors influencing the care of children with MV is warranted.

Funding: Réseau en Santé Respiratoire du Québec (Respiratory Research Network of Quebec), Fonds de la recherche en santé du Québec, Women and Children Health Research Institute-Clinical/Community Research Integration and Support Program, Réseau mère-enfant de la francophonie.

Translations: For the French and Spanish translations of the abstract see Supplementary Materials section.