Value of pneumococcal PCR in respiratory samples for exclusion of pneumococcal pneumonia.

Abstract

Streptococcus pneumoniae is the main aetiological agent in bacterial pneumonia. Therefore pneumococcal PCR is often included in respiratory multiplex PCR panels, both commercial and in-house. But respiratory PCR results for S. pneumoniae are difficult to interpret due to frequent non-pathogenic colonization on the mucosal surface of the upper airways with pneumococci or to cross-reaction of the PCR target in non-pneumococcal streptococci. In this study we investigated the value of lytA gene pneumococcal PCR in patients presenting with pneumonia. To assess the utility of lytA gene detection for S. pneumoniae in a respiratory multiplex quantitative PCR (qPCR) panel for patients presenting with pneumonia. A retrospective study was conducted for lytA gene results as target for S. pneumoniae in hospitalized patients who were diagnosed with pneumonia and for which a respiratory multiplex PCR panel was performed. Patients were classified as 'probable', 'possible' or 'unlikely' of having a pneumococcal pneumonia. A sensitivity of 71.4% and specificity of 89.6% were found, corresponding to a negative predictive value and positive predictive value of 97.6% and 34.2%, respectively, when considering 'probable' versus 'possible/unlikely'. In the PCR-positive cases we found a statistically significant difference in semi-quantitative Ct values between the 'probable' and the 'possible/unlikely' groups. We conclude that a negative qPCR for the lytA gene in a respiratory sample is highly predictive of a negative S. pneumoniae culture and is possibly sufficient to exclude S. pneumoniae as a causative agent. Respiratory pneumococcal PCR has a high negative predictive value for pneumococcal disease but the positive predictive value is low.