Objective: To verify the feasibility of non-invasive ventilation (NIV) in immunocompromised children affected by ARDS.
Setting: University Hospital PICU.
Patients: Twenty-three consecutive immunocompromised children treated with NIV for ARDS.
Interventions: All consecutive patients received NIV through a face-mask or a helmet.
Results: No differences were found regarding admission data and severity scores between NIV responders and non-responders. Early and sustained improvement in PaO2/FiO2 ratio were observed in 82 and 74% of cases, respectively. 13 out of 23 patients (54.5%) avoided intubation and were discharged from the PICU; ten patients required intubation: two of them survived and eight patients died (two refractory hypoxemia, three septic shock, three multi-organ failure). PICU and intra-hospital mortality was significantly higher for NIV-nonresponders (P < 0.001). PICU stay was significantly shorter for NIV responders (P = 0.03). NIV responders had significantly lower heart and respiratory rate at the end of treatment (P < 0.001 and P = 0.048, respectively).
Conclusions: NIV administration is feasible and well tolerated in immunocompromised children with ARDS. A short NIV trial can be used to verify the usefulness of the technique. A randomized controlled trial is needed to confirm the efficacy of NIV in immunocompromised children requiring ventilatory support for ARDS.