Outcome of high-flow nasal cannula for acute hypoxemic respiratory failure due to pneumocystis pneumonia

Abstract

Background: Despite increasing use of high-flow nasal cannula (HFNC) in acute hypoxemic respiratory failure (AHRF), data regarding the selection of HFNC as the initial ventilation in AHRF with Pneumocystis pneumonia (PCP) are limited.

Methods: Eighty-two consecutive HIV-negative PCP patients with AHRF were divided into three groups: HFNC (n = 23); HFNC followed by mechanical ventilation (HFNC-MV) (n = 29); and mechanical ventilation (MV) (n = 30). Clinical outcomes and physiologic effects were compared between the groups by retrospective chart review.

Results: Compared with the HFNC and the MV groups, the HFNC-MV group tended to have a higher incidence of barotrauma, and had significantly lower HFNC/MV- and ICU-free days. The 60-day mortality was 52% in the HFNC-MV group, 13% in the HFNC group, and 30% in the MV group (P = 0.01). Multivariate analysis indicated that HFNC-MV was independently associated with mortality (OR, 9.68; 95% CI, 1.79–52.29; P = 0.008). Repeated measures ANOVA revealed that there was a significant improvement in PaO₂/FiO₂ (P = 0.02) and A-aO2 (alveolar-arterial gradient) (P = 0.02) in the HFNC group in comparison with the other groups during the first 96 hours of ventilation. In the HFNC-MV group, PaO₂ increased in survivors and decreased in non-survivors within 3 hours of HFNC initiation. The cut-off value of predicting mortality was <53% increase of baseline PaO₂ at 3 hour with the sensitivity of 87% and the specificity of 57%.

Conclusions: HFNC can be the initial choice for ventilation of AHRF with PCP. Due to high mortality with inappropriate HFNC use, however, switching to MV should be considered in case without early improvement in oxygenation.