Evaluation of Legionella diagnostic prediction score in patients with SARS-CoV-2 Omicron pneumonia

Legionella pneumonia is an important cause of community-acquired pneumonia (CAP). The Japanese Respiratory Society (JRS) pneumonia guideline 2024 proposed use of the Legionella Diagnostic Prediction Score for the management of CAP in adults. The committee for the JRS pneumonia guideline is required to verify the validity of the Legionella Diagnostic Prediction Score for the next revision. In addition, it is necessary to determine appropriate cutoff scores by examining all pneumonia cases. In the present study, we validated the usefulness of the Legionella Diagnostic Prediction Score using SARS-CoV-2 Omicron CAP. We analyzed 116 patients with L. pneumophila CAP and 947 patients with SARS-CoV-2 Omicron CAP. Among Omicron cases, the median Legionella Diagnostic Prediction Score was identical among BA.1, BA.2, BA.5, XBB lineage, BA.2.86 and JN.1 subvariants. The median Legionella Diagnostic Prediction Score was significantly higher in the L. pneumophila CAP group than the SARS-CoV-2 Omicron CAP group (4 vs 1, p < 0.0001). When targeting all 947 patients with Omicron subvariants, the diagnostic sensitivity and specificity for the presumptive diagnosis of L. pneumophila CAP were 90.5 % and 90.8 %, respectively, when a total score >= 3 points was set as the cutoff level. When the cutoff score was >= 4 points, the diagnostic sensitivity and specificity for presumptive diagnosis of L. pneumophila CAP were 76.7 % and 99.6 %, respectively. Our results demonstrated that the Legionella Diagnostic Prediction Score had good diagnostic ability during the SARS-CoV-2 Omicron variant epidemic period. To set optimal indicators and cutoff values for the Legionella Diagnostic Prediction Score, the policy of the committee for the JRS pneumonia guideline is to continue testing for all pneumonia types.