Introduction: Children with dependence on invasive mechanical ventilation (IMV) are at risk for prolonged hospitalizations. While children with technology dependence utilize a significant proportion of PICU resources, there is limited information on the characteristics of their admissions. Acute infections are common reasons for admission in pediatrics. A better understanding of ventilator trajectories for children with suspected infection could optimize resource use and support timely dispositions. This study aimed to (1) characterize ventilator setting trajectories during PICU admission for suspected infection among children with pre-existing tracheostomies and IMV dependence, and (2) assess variability in discharge ventilator settings by clinical and demographic factors. Methods: We conducted a retrospective cohort study using an electronic health record (EHR) dataset from 13 PICUs. Admissions from 1/2012-12/2017 were included for patients with suspected or confirmed infection on admission. Patients with baseline IMV dependence were identified using ICD codes. Ventilator settings were extracted from EHR flowsheets; missing values were imputed using last observation carried forward. Positive end-expiratory pressure (PEEP) was used as a surrogate marker for ventilator titration. Associations between clinical and demographic characteristics and ventilator settings were evaluated using linear and logistic regression models. Results: 1657 encounters met inclusion criteria. Median PEEP at both admission and discharge was 6 (IQR 5, 8). Among survivors, 806/1570 (51.3%) had no change in PEEP during hospitalization, 180/1570 (11.5%) were discharged on decreased PEEP, and 152/1570 (9.7%) on higher PEEP. Male sex (adjusted odds ratio aOR 1.28, 95% CI: 1.04–1.58) and hospital site were associated with unchanged PEEP during admission, while age and season were not (age aOR 1.01, 95% CI: 0.97–1.06). Younger age was independently associated with higher discharge PEEP (p < 0.001), and significant inter-site variation was observed. Conclusions: Most encounters for children with IMV dependence involved minimal ventilator titration. The observed inter-site variability and minimal ventilator adjustment during admission highlight opportunities to standardize admission and discharge criteria in this population.
Pilarz et al. (Sun,) studied this question.