Introduction:

The advantage of nucleic acid amplification techniques (NAAT) is their extreme sensitivity and specificity when compared with traditional techniques. Multiplex formats might solve the practical shortcoming of detecting only the infectious agent that is searched for. Real-time assays enable a one-tube assay that is suitable for high-throughput applications, reducing the assay time and limiting potential contamination between samples. A comparison between different NAAT is needed.

Objectives:

To compare the sensitivity and specificity of real-time (RT) mono- and multiplex nucleic acid sequence-based amplification (NASBA), and mono-PCR for the detection of M. pneumoniae, C. pneumoniae and Legionella spp. in respiratory specimens from hospitalised and outpatients with community-acquired pneumonia (CAP).

Methods:

A total of 244 respiratory specimens were collected from 142 patients with CAP (116 throat swabs, 112 sputa and 16 other respiratory specimens). NASBA was done by using the NucliSens Basic Kit (bioMérieux), PCR was done as described earlier. PCR detects L. pneumophila whereas NASBA was designed to detect Legionella spp. All samples with discordant results were reanalysed. Definition of the expanded gold standard used to calculate the sensitivities of the tests applied: true positive if positive by at least two NAAT.

Results:

The sensitivities of the different techniques, compared with an expanded gold standard, were 77.8, 100, and 100% for detection of M. pneumoniae; and 50, 100, and 50 for detection of L. pneumophila by PCR, RT mono-NASBA, and RT multiplex NASBA respectively. If positive by any method, the sensitivities were 63.2, 92.1, and 71.1 for detection of M. pneumoniae; 57.1, 71.4 and 28.6 for detection of L. pneumophila by PCR, RT mono-NASBA, and RT multiplex NASBA, respectively. Two samples were found to be C. pneumoniae positive only by RT mono NASBA which is an insufficient number to calculate the sensitivity of the tests applied.

Conclusions:

Mono RT NASBA is more sensitive than mono PCR. When the former is compared with RT multiplex NASBA, the mono NASBA is the most sensitive test. This confirms results presented earlier by using spiked clinical specimens. The RT multiplex NASBA could become a fast, and user-friendly diagnostic tool for the detection of these organisms in respiratory specimens although it seems to be less sensitive. Further comparison on larger numbers of clinical specimens is necessary.