Association between one-session bilateral whole-lung lavage and periprocedural complications in patients with pulmonary alveolar proteinosis: a retrospective cohort study

BACKGROUND: Whole lung lavage (WLL) has been recognized as the most effective therapy of severe pulmonary alveolar proteinosis (PAP). Most centers perform the lavage of each lung in two sessions under general anesthesia at an interval of several days to weeks. Compared with two-session WLL, one-session bilateral sequential WLL only requires general anesthesia once. However, the safety of one-session WLL in PAP patients has not been assessed by large cohort studies. In this study, we aimed to investigate the association between the mode of WLL procedure (one-session or two-session) and the risk of periprocedural complications in PAP patients. METHODS: In this single-center retrospective cohort study, we included adult patients who were diagnosed as PAP and had undergone WLL procedures under general anesthesia from 2000 to 2022. Patients requiring extra-corporeal oxygenation during WLL were excluded. Since some patients received multiple WLL procedures, we considered each procedure in one-session or two-session group as a unique unit in our analysis. The primary outcome was the occurrence of any complications during hospitalization, including termination of WLL procedure due to fluid leakage or refractory hypoxemia, bronchospasm, delayed endotracheal extubation, cardiovascular event, pneumothorax, and fever. RESULTS: We included a total of 175 WLL procedures (118 patients), with 48 in the two-session group and 127 in the one-session group. Periprocedural complications occurred in 17 (35.4%) and 39 (30.7%) procedures in the two-session and the one-session groups, respectively. The risk of periprocedural complications did not differ significantly between groups, after adjusting the unbalanced confounders in a multivariable model (odds ratio 0.95, 95% confidence interval 0.34 to 2.69, P 0.929) or by inverse probability of treatment weighting (odds ratio 0.70, 95% confidence interval 0.30 to 1.54, P 0.379). Compared with the two-session WLL group, the one-session WLL group had a shorter postprocedural length of hospitalization and comparable decrease in alveolar-arterial oxygen tension gradient from baseline. CONCLUSIONS: One-session bilateral WLL was not associated with an increased risk of periprocedural complications compared with two-session WLL in PAP patients. Experienced physicians may consider performing one-session WLL in view of the comparable safety and efficacy and potential advantages of saving time.