Invalidity of JRS atypical pneumonia prediction score in Omicron variant of COVID-19 pneumonia.

BACKGROUND

We evaluated whether the Japanese Respiratory Society (JRS) atypical pneumonia prediction score can be adapted for the diagnosis of COVID-19 pneumonia due to Omicron BA.1, BA.2, and BA.5 subvariants.

METHODS

We enrolled a total of 547 patients with community-acquired COVID-19 pneumonia. Of the COVID-19 pneumonia patients, 198 cases were the Omicron BA.1 subvariant, 127 cases were the Omicron BA.2 subvariant, and 222 cases were the Omicron BA.5 subvariant. Patients with extremely severe pneumonia were excluded and finally 524 patients were analyzed.

RESULTS

Rates of conformity for the six predictors were identical among the three Omicron groups, and high rates of conformity were observed in the following predictors: adventitious sounds; etiological agent; and a peripheral WBC count. The sensitivities of the diagnosis of atypical pneumonia in patients with COVID-19 pneumonia based on four or more predictors were 49.0% in the BA.1 subvariant group, 58.1% in the BA.2 subvariant group, and 51.0% in the BA.5 subvariant group. The diagnostic sensitivities for Omicron BA.1, BA.2, and BA.5 subvariant groups were 96.6%, 100%, and 96.4% for non-elderly (aged <60 years) patients and 28.4%, 29.7%, and 34.2% for elderly (aged ≥60 years) patients, respectively.

CONCLUSIONS

In Omicron variant of COVID-19, the JRS atypical pneumonia prediction score is a useful tool for distinguishing between COVID-19 pneumonia and bacterial pneumonia only in patients aged <60 years.