Abstract Rationale The end-tidal alveolar dead space fraction (AVDSf) is a marker of mortality risk in pediatric acute respiratory distress syndrome (PARDS). The pathophysiology of alveolar dead space in PARDS is thought to be related to abnormal pulmonary perfusion and vascular dysfunction. We hypothesized that these vascular abnormalities may also extend to the central nervous system, and that AVDSf could serve as a marker for the development of delirium in children with PARDS. Methods This is a secondary analysis of children with PARDS enrolled in a single center randomized controlled trial of a lung protective ventilation strategy (REDVent, R01HL124666). Children who received ECMO were excluded. Maximum and median AVDSf (PaCO2 - PETCO2 (end-tidal CO2)/PaCO2) and oxygenation index (OI) in the first 24 hours of invasive mechanical ventilation (IMV) were calculated. The primary outcome was a psychiatric diagnosis of delirium within 21 days of IMV initiation. Baseline Pediatric Cerebral Performance Category scale (PCPC) scores and functional status were explored as potential effect modifiers. Results A total of 205 children who survived 21 days after IMV initiation were included. Of these, 27 (13%) were diagnosed with delirium, with a median time of 7 5,12 days from IMV initiation to diagnosis. Children with delirium had a longer duration of IMV, more days of paralysis, higher baseline PCPC score and were more likely to be immunosuppressed (all p 0.05). In multivariable modeling, neither median AVDSf nor OI in the first 24 hours of IMV was associated with delirium diagnosis. However, children with a PCPC score of 3 or higher, indicating at least moderate neurologic impairment, were less likely to be diagnosed with delirium compared to those with no neurologic impairment (OR 0.26, 0.08, 0.82 p = 0.021). Baseline PCPC and functional status did not modify the relationship between AVDSf and delirium. CONCLUSIONS In this sample, a psychiatric diagnosis of delirium was uncommon and not associated with initial markers of lung injury severity. Children with preexisting neurologic dysfunction appeared less likely to receive a delirium diagnosis, suggesting possible under-recognition in this group. Larger studies are needed to determine whether underdiagnosis of delirium influenced these findings. This abstract is funded by: NHLBI (National Heart, Lung, and Blood Institute)
Rinaldi et al. (Fri,) studied this question.