![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Previous Article | Next Article 
Journal of Clinical Microbiology, October 2007, p. 3390-3392, Vol. 45, No. 10
0095-1137/07/$08.00+0 doi:10.1128/JCM.00505-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
A. van der Zee,10,
and
R. Jansen1*
Regional Public Health Laboratory Kennemerland, Boerhaavelaan 26, 2035 RC Haarlem, The Netherlands,1 Laboratory for Pathology and Medical Microbiology, PAMM Laboratories, Michelangelolaan 2, 5623 EJ Eindhoven, The Netherlands,2 Department of Medical Microbiology, Center of Infectious Diseases, Leiden University Medical Centre, P.O. Box 9600, 2300 RC Leiden, The Netherlands,3 Molecular Diagnostics, Jeroen Bosch Hospital, P.O. Box 90153, 5200 ME's-Hertogenbosch, The Netherlands,4 Department of Medical Microbiology, Radboud University Nijmegen Medical Center, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands,5 Public Health Laboratory Friesland, P.O. Box 21020, 8900 JA Leeuwarden, The Netherlands,6 Department of Medical Microbiology and Immunology, Meander Medical Center, Utrechtseweg 160, 3818 ES Amersfoort, The Netherlands,7 Laboratory for Infectious Diseases, van Ketwich Verschuurlaan 92, 9721 SW Groningen, The Netherlands,8 Department of Medical Microbiology, Academic Hospital Maastricht, P.O. Box 5800, 6202 AZ Maastricht, The Netherlands,9 Laboratory of Medical Microbiology, St. Elisabeth Hospital, P.O. Box 747, 5000 AS Tilburg, The Netherlands,10
Received 6 March 2007/ Returned for modification 4 May 2007/ Accepted 21 July 2007
 |
ABSTRACT |
|---|
 Top ABSTRACT TEXTREFERENCES |
|---|
|
TEXT |
|---|
|
TopABSTRACTTEXTREFERENCES |
|---|
A series of 12 sputum samples were prepared for this survey by the Regional Laboratory of Public Health in Haarlem, The Netherlands. By using sputum as a matrix, the samples of the survey closely resemble samples from patients, with its characteristic abundant microbial flora and mucoproteins.
Each sample consisted of 200 µl of sputum that was prepared from discarded patient material after it was screened for Legionella bacteria by culture and PCR. Samples 1 to 12 were spiked with Legionella bacteria according to the scheme shown in Table 1. The Legionella strains were a Legionella pneumophila serogroup 1 strain and a Legionella longbeachae strain. Legionella longbeachae is included in the panel because this species is the second most common causative agent of legionellosis and should be detected by the laboratories (9).
|
The performances of the laboratories are indicated in Table 2. For each correctly analyzed sample, i.e., the presence of a species and the correctly identified Legionella species, 2 points were obtained by the laboratory. If only the presence of an organism of the genus Legionella was reported, without the correct species name, 1 point was obtained. No points were obtained for a wrong diagnosis of a sample. Using Fisher's exact test, the P value between data was calculated. A P value of <0.05 was considered statistically significant.
|
The nine participating laboratories used five different protocols for DNA isolation. These methods encompass automated methods using magnetic affinity beads (the MagNA Pure and MagNA Pure Compact systems and the NucliSENS easyMAG system) and manual methods using affinity matrices (the QIAamp DNA mini kit and Puregene DNA purification kit) (Table 2). The five laboratories that used the MagNA Pure system had the lowest scores for samples 1 to 8, which were spiked with L. pneumophila only, due either to PCR inhibition or to a false-negative result for L. pneumophila. However, the correlation between a low score and the use of the MagNA Pure system might be coincidental, since in a controlled study, Wilson et al. reported that automated DNA extraction systems, including the MagNA Pure system, performed significantly better than manual extraction methods (13).
A blank sample (sample 9) that was added to the sputum sets was found negative by all participating laboratories (except laboratory A, which reported inhibition), indicating that Legionella contamination of laboratory reagents did not play an important role in this survey.
The PCR methods that were used by the laboratories differed considerably. The applied target genes were 16S rRNA, mip, and 5S rRNA genes (Table 2). Real-time PCR was done on LightCycler (Roche), iCycler (Bio-Rad), or ABI PRISM (Applied Biosystems) instruments, and the detection was done by TaqMan or fluorescence resonance energy transfer probes. Endpoint PCR products were detected by agarose gel electrophoresis or enzyme-linked immunoassay. Despite all these differences, neither the analytical sensitivity nor the proneness to inhibition of the PCRs is correlated to any of these parameters (Table 2).
As expected, a clear difference was found between the 16S rRNA and mip gene PCRs for their ability to detect L. longbeachae. The mip PCRs were designed to specifically detect L. pneumophila, while the 16S rRNA gene PCRs were designed to detect other Legionella species as well (3, 8-11, 14). This is clearly reflected in the scores for sample 10, which was spiked with L. longbeachae. None of the laboratories that used only a mip PCR (labs E, F, and G) detected Legionella in this sample, while the laboratories that used a 16S rRNA PCR (except lab C) correctly detected the Legionella species in this sample. The scores for samples 11 and 12, which contained both L. pneumophila and L. longbeachae, were comparable to samples 3 to 6, which were spiked with the same amounts of L. pneumophila.
The detection of Legionella species other than L. pneumophila is relevant for the diagnosis of legionellosis, since this disease can be caused by species other than L. pneumophila, such as L. longbeachae in immunocompetent patients and Legionella bozemanii in immunocompromised patients. The correct identification of the species is also relevant for epidemiological studies and for the identification of sources of infection. Patients infected with any of the Legionella species are treated the same. Therefore, the high specificity of the Legionella pneumophila mip gene-targeted PCRs is not advantageous for the diagnosis of legionellosis.
We conclude that the pretreatment of the sputum is important to prevent inhibition of the PCR. The DNA isolation method is of less importance; the low scores with the MagNA Pure and the MagNA Pure Compact systems might be coincidental. The two target genes, the 16S rRNA gene and the mip gene, perform equally well in detecting the lowest level of L. pneumophila in the samples. However, the 16S rRNA gene PCR is able to detect L. longbeachae, while the mip gene PCR does not detect this clinically relevant species. Therefore, in our view, the 16S rRNA gene PCR is preferred for the identification of patients with Legionnaires’ disease caused by either L. pneumophila or L. longbeachae.
|
FOOTNOTES |
|---|
Published ahead of print on 1 August 2007.
Present address: Laboratory of Pediatrics, Erasmus Medical Center Rotterdam, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands.
Present address: Laboratory for Molecular Diagnostics RLM/PAL, Albert Schweitzer Hospital, P.O. Box 899, 3300 AW Dordrecht, The Netherlands.
|
REFERENCES |
|---|
|
TopABSTRACTTEXTREFERENCES |
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Antimicrob. Agents Chemother. | Clin. Microbiol. Rev. |
|---|---|
| Clin. Vaccine Immunol. | ALL ASM JOURNALS |
|---|
