Background: The guidelines for the requirement of Legionella urinary antigen tests on admission for patients hospitalized with community-acquired pneumonia differ in Japan, the United States, and Europe. We aimed to evaluate the association between the timing of Legionella urinary antigen testing and inhospital mortality in patients with atypical pneumonia.
Methods: We identified 654,708 patients with atypical pneumonia from July 2010 to March 2021 using the Japanese national inpatient database. The patients were divided into groups that underwent Legionella urinary antigen tests on the day of admission (test group, n = 229,649) and those that underwent testing after the day of admission or were untested (control group, n = 425,059). A propensity score-stabilized inverse probability of treatment weighting analysis was performed to compare inhospital mortality, length of hospital stay, and total hospitalization costs between the two groups. Odds ratios (ORs) or differences and their 95% confidence intervals (CIs) were calculated using generalized linear models.
Results: The tested group had a significantly lower 30-day inhospital mortality than that of the control group (7.7% vs. 9.0%; OR: 0.83 [95% CIs, 0.81-0.86]). The tested group also had a significantly shorter length of stay (difference, -2.3 [-2.6 to - 2.0] days and total hospitalization costs (-396 [-508 to - 285] US dollars) than that of the control group.
Conclusions: Legionella urinary antigen testing upon admission is associated with better outcomes in patients with atypical pneumonia. Legionella urinary antigen testing performed on the day of admission is recommended for hospitalized patients with atypical pneumonia.
Keywords: Inhospital mortality; Legionella pneumonia; length of stay; urinary antigen tests.