Two Quality Control Exercises Involving Nucleic Acid Amplification Methods for Detection of Mycoplasma pneumoniae and Chlamydophila pneumoni
http://www.100md.com
微生物临床杂志 2006年第3期
Laboratory of Medical Microbiology, Department of Medicine, Universitaire Instelling Antwerpen, Universiteitsplein 1 S3, B-2610 Wilrijk Belgium1
ABSTRACT
The quality performance of laboratories for the detection of Mycoplasma pneumoniae and Chlamydophila pneumoniae by two quality control (QC) exercises with a 2-year interval was investigated. For the 2002 QC exercise, specimens were spiked with M. pneumoniae at concentrations of 5,000, 500, 50, and 0 color-changing units (CCU)/100 μl. The limit of detectability was 50 CCU/100 μl. Therefore, this concentration was omitted from the 2004 panel and was excluded from the analysis. In 2002, 2 out of 12 participants obtained 100% correct results, 2 out of 12 produced false-positive results, and 10 out of 12 had between 0 out of 9 and 8 out of 9 correct positive results. In 2004, correct results were obtained in 15 out of 18 tests, and no false-positive results were reported. In 2002, specimens were spiked with C. pneumoniae at concentrations of 490, 49, 4.9, and 0 inclusion-forming units/100 μl (IFU/100 μl). In the 2004 panel, samples spiked with a lower dilution of 0.49 IFU/100 μl were added to the panel. For the C. pneumoniae QC, correct results were produced in 12 out of 16 and 13 out of 18 tests in 2002 and in 2004, respectively. Both multiplex PCR and nucleic acid sequence-based amplification (NASBA) formats scored a smaller number of samples positive than the monoplex reactions.
INTRODUCTION
A multitude of nucleic acid amplification tests (NAATs) have been described for the detection of Mycoplasma pneumoniae and Chlamydophila pneumoniae in respiratory specimens (6, 17). In addition to in-house PCR tests, commercial kits are becoming available, such as the LCx for C. pneumoniae (Abbott Laboratories) (8) and the NucliSens Basic kit (bioMerieux), for which the primers and the target-specific biotinylated capture probe are to be synthesized for each target by the user (15).
Although all tests aim to be rapid, sensitive, specific, and easy to perform, results of NAATs may be unreliable because of cross-contamination, inappropriate treatment of the clinical samples leading to loss of target nucleic acid, or the presence of inhibitors (2, 5, 12, 14, 26, 27).
To date, only one study compared the results of different NAATs for the detection of C. pneumoniae in respiratory specimens in different centers (8), and four studies compared the results of amplification methods performed in different centers for the detection of C. pneumoniae in atheroma specimens (2, 3, 11, 21). To our knowledge, no such studies have been published for M. pneumoniae.
The aim of this study was to assess the quality performance of laboratories for the detection of M. pneumoniae and C. pneumoniae by two quality control (QC) exercises with a 2-year interval.
MATERIALS AND METHODS
Participating laboratories. The participating laboratories are as follows, in alphabetical order: Academisch Ziekenhuis Vrije Universiteit Brussel, Brussels, Belgium; Algemeen Ziekenhuis Sint Augustinus, Wilrijk, Belgium; Algemeen Ziekenhuis Sint Jan, Brugges, Belgium; bioMerieux, Boxtel, The Netherlands; Centre Hospitalier Regional de la Citadelle, Liege, Belgium; Cliniques Universitaires Universite Catholique de Louvain de Mont-Godinne, Yvoir, Belgium; Groupement des Centres Hospitaliers de Jolimont-Lobbes et de Tubize-Nivelles, La Louviere, Belgium; Institut Jules Bordet, Brussels, Belgium; Institution de Pathologie et de Genetique-Loverval, Loverval, Belgium; Leids Universitair Medisch Centrum, Leiden, The Netherlands; Onze Lieve Vrouwe Ziekenhuis, Aalst, Belgium; Ospedale Maggiore di Milano, Milan, Italy; Public Health Laboratory Friesland, Leeuwarden, The Netherlands; Universite Libre de Bruxelles-Erasme, Brussels, Belgium; University Medical Center-Brugmann, Brussels, Belgium; University of Antwerp, Wilrijk, Belgium; Universitair Ziekenhuis Katholieke Universiteit Leuven, Leuven, Belgium; University Hospital Antwerp, Edegem, Belgium; Virga Jesse Ziekenhuis, Hasselt, Belgium; and Ziekenhuizen Noord Antwerpen, Antwerp, Belgium.
Preparation of proficiency panels. A stock suspension of M. pneumoniae strain ATCC 29085 was quantitated by incubation of 10-fold dilutions in triplicate in SP4 medium at 37°C. The cultures were monitored for 2 months, and the titer was expressed as color-changing units (CCU) per milliliter; 1 CCU corresponds to 10 to 100 cells (4).
A stock suspension of C. pneumoniae strain ATCC VR-1355 was quantified by incubation of five replicates of 10-fold dilutions on confluent layers of Hep-2 cells. The vials were centrifuged at 3,500 rpm at 25°C for 60 min and incubated at 37°C. After 1 h, the medium was replaced by fresh culture medium containing 1 mg/liter cycloheximide. After 3 days, cells were fixed with 96% ethanol and stained by the fluorescent antibody technique with specific mouse monoclonal antibodies and fluorescein-labeled rabbit anti-mouse immunoglobulin (both from Dako A/S, Glostrup, Denmark). The titer was expressed as inclusion-forming units (IFU) per milliliter.
Bronchoalveolar lavage (BAL) specimens were collected from patients in the University Hospital of Antwerp and stored at –20°C. BAL pools were negative for M. pneumoniae and C. pneumoniae as tested by nucleic acid sequence-based amplification (NASBA) (16, 21) and by in-house-developed PCR (12, 27).
The 2002 proficiency panel consisted of parts A and B: part A was prepared in sterile physiologic saline, and part B was prepared in BAL. The negative samples were prepared in a laminar flow cabinet in a separate room prior to the preparation of the positive samples. Materials and pipettes had never been used for M. pneumoniae- or C. pneumoniae-related work previously.
M. pneumoniae parts A and B each contained four negative samples and three samples with 500 CCU/100 μl of solution (see Table 5); part A also contained three samples with 50 CCU/100 μl, and part B contained three samples with 5,000 CCU/100 μl. For C. pneumoniae (see Table 7), parts A and B contained four negative samples each and three samples with 49 IFU/100 μl; furthermore, part A contained three samples with 4.9 IFU/100 μl, and part B contained three samples with 490 IFU/100 μl.
For the 2004 proficiency tests, all samples were prepared in BAL. Each panel contained four negative samples, prepared as described above. Three different BAL pools were used: pool 1 was fluid, pool 2 was more viscous, and pool 3 was blood stained. Three samples of the M. pneumoniae panel (see Table 6) were spiked with 5,000 CCU/100 μl, and three samples were spiked with 500 CCU/100 μl. The C. pneumoniae proficiency panel (see Table 8) included one sample with 490 IFU/100 μl, one sample with 49 IFU/100 μl, one sample with 4.9 IFU/100 μl, and three samples with 0.49 IFU/100 μl BAL.
The proficiency panels were coded and sent refrigerated on the same day or were stored at –80°C until they were shipped frozen to the participants. The samples were to be tested as routine specimens, and the results were returned within 4 weeks together with a questionnaire collecting information on the procedures applied.
Quality control assurance. Fifteen serial dilutions of M. pneumoniae were tested by NASBA and PCR (12, 16) after one and two freeze-thaw cycles, mimicking the conditions of the preparation of the panels and on arrival at the participating laboratory. Similar tests were performed for the detection of nine serial dilutions of C. pneumoniae by NASBA and PCR (18, 27). Furthermore, each BAL pool was spiked in triplicate with a serial dilution of M. pneumoniae and tested by both NAATs. Quality control assurance of the panels was performed by NASBA and the in-house-developed PCR.
Methods used in the participating centers for the extraction of M. pneumoniae and C. pneumoniae nucleic acids. Each laboratory followed its standard procedures for M. pneumoniae and C. pneumoniae detection. Thus, the procedures varied from laboratory to laboratory.
The following commercially available extraction kits were used (Tables 1 to 4): Amplicor HCV specimen preparation kit (Roche, Mannheim, Germany), Nucleospin tissue (Clontech, Mount View, CA), NucliSens Basic kit (bioMerieux, Boxtel, The Netherlands), Puregene genomic DNA for gram-positive bacteria (Gentrasystems, Minneapolis, MN), QIAamp DNA minikit (QIAGEN, Hilden, Germany), QIAamp DNA blood kit (QIAGEN), QIAamp stool kit (QIAGEN), QIAamp DNA tissue kit (QIAGEN), Roche Amplicor Sputum Preparation kit (Roche), and the Roche High Pure PCR Template Preparation kit (Roche).
Three participants used in-house nucleic acid extraction methods for the M. pneumoniae or C. pneumoniae 2002 proficiency panels: participant number 5, 10 min of boiling; participant number 10, 5 cycles of 1 min of freeze boiling; and participant number 14, a proteinase K pretreatment followed by 10 min of boiling.
Methods used in the participating centers for the amplification and detection of M. pneumoniae and C. pneumoniae nucleic acids. Participant 1. Conventional nucleic acid sequence-based amplification (NASBA) for the detection of M. pneumoniae RNA in 2002 was done as described earlier (14, 15). Monoplex real-time NASBA for the detection of M. pneumoniae RNA was done as described by Loens et al. (16). Conventional and monoplex real-time NASBA for the detection of C. pneumoniae RNA was done as described by Loens et al. (18). Multiplex (MX)-real-time NASBA for the detection of RNA from both organisms was done as described previously (19).
Participant 2. M. pneumoniae DNA was detected as described by Hardegger et al. (10). C. pneumoniae DNA was amplified by an in-house-developed real-time PCR based on the PstI fragment-updated sequence (7) using primers CPNFW (5'-TGGAGATAAAATGGCTGGACG-3') and CPNREV (5'-TATGGCATATCCGCTTCGG-3') and detection probe (5'-6-carboxyfluorescein [FAM]-CACGGAAATAAAGGTGTTGTTTCCAAAATCG-6-carboxytetramethylrhodamine [TAMRA]-3') and the same amplification conditions as those for M. pneumoniae amplification.
Participant 3. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers MYPN3 (5'-AGGCTTCAAGTGGACAAAGTGAC-3') and MYPN4 (5'-GATTGTYCCTGCTGGYCCAT-3') and detection probe MYPNF (5'-FAM-ACCACACCAAGTTCACGAGCGCTACG-TAMRA-3') with the following amplification conditions: 1 cycle of 2 min at 50°C followed by 1 cycle of 10 min at 95°C and 45 cycles of 15 s at 95°C and 60 s at 60°C.
C. pneumoniae DNA was also amplified by an in-house-developed real-time PCR using primers CHPN1 (5'-GAGATGGAGCAAATCCTAAAAGCTA-3') and CHPN2 (5'-AAATAGGTTGAGTCAACGACTTAAGGT-3') and detection probe CHPN-F (5'-FAM-TCAGCCATAACGCCGTGAATACGTTCTC-TAMRA-3') with the same amplification conditions used for M. pneumoniae.
Participant 4. In 2002, M. pneumoniae DNA was amplified by using the Minerva kit (Minerva Biolabs, Berlin, Germany) according to the instructions of the manufacturer. In 2004, an in-house-developed assay was applied using primers Mpneu02-FW (5'-GCCGGCAGTGGCAGTC-3') and Mpneu02-RV (5'-AGCCGCTTCGGTTCGG-3') and detection probe Mpneu02-MGB (5'-FAM-AACCACGTATGATCCC-nonfluorescent quencher [NFQ]-3') with the following amplification conditions: 1 cycle of 2 min at 50°C, 1 cycle of 10 min at 95°C, and 45 cycles of 15 s at 95°C and 60 s at 60°C.
For the detection of C. pneumoniae, primers CHLpnFW (5'-TGGCTAGGCCATTGAGAGTGA-3') and CHLpnRV (5'-GTTATGGATGGAGGGACTACTTTTG-3') and detection probe CHLpnMGB (5'-FAM-CTCAGCGCTTGCC-NFQ-3') were used. Amplification conditions were 1 cycle of 1 min at 95°C followed by 45 cycles of 15 s at 95°C and 60 s at 60°C.
Participant 5. In 2002, M. pneumoniae DNA was amplified by an in-house-developed PCR using primers OJPU1 (5'-GCCACCCTCGGGGGCAGTCAG-3') and OJPU3 (5'-GAGTCGGGATTCCCCGCGGAGG-3') (12). The following amplification conditions were used: 1 cycle of 3 min at 90°C, followed by 39 cycles of 1 min at 90°C, 2 min at 67°C, and 2 min at 72°C, and 1 final cycle of 10 min at 72°C.
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using slightly modified major outer membrane protein (MOMP) VD2 primers of Tondella et al. (26), FPCPN (5'-CTCGTTGGTTTATTCGGAGTT-3') and RPCPN (5'-CCAAGAGAAAGAGGTGTCTGTG-3'), and the slightly modified MOMP VD2 detection probe described by Tondella et al. (26), CPN (5'-FAM-AAGGTACTACTGTAAATGCAAATGAACTACCAAACGTTTCTTTAAGTAACGGAG-TAMRA-3').
In 2004, M. pneumoniae DNA was amplified by an own in-house-developed real-time PCR using primers M823F (5'-AGCGAACCGAGAGTGGTCAA-3') and M973R (5'-GATTGGCCAGATCCAGATGTG-3') and the detection probe (5'-FAM-CTCCAGGGCGCTGAGGCCACT-TAMRA-3'). Amplification conditions were 1 cycle of 2 min at 50°C and 1 cycle of 10 min at 95°C, followed by 50 cycles of 15 s at 95°C and 60 s at 60°C each.
C. pneumoniae DNA was amplified as described by Welti et al. (32).
Participant 6. In 2002 and 2004, M. pneumoniae DNA was amplified as described previously by Ieven et al. (12) and Ursi et al. (30), respectively. For C. pneumoniae DNA amplification, PCR was applied in 2002 and 2004 as described by Ursi et al. (29) and Hoymans et al. (11), respectively, using for the latter the MOMP VD4 primers and one detection probe described by Tondella et al. (26). Amplification conditions were the same as those for M. pneumoniae.
Participant 7. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using reverse primer 5'-CCAGGGCACATAATCCAACAC-3' and forward primer 5'-AAGGAACAAACTGATCCCACTTCT-3' and detection probe 5'-FAM-TCTCCACCGGGTTCAACCTTGTGG-NFQ-3', with the following amplification conditions 1 cycle of 2 min at 50°C, 1 cycle of 10 min at 95°C, and 40 cycles of 15 s at 95°C and 60 s at 60°C.
Participant 8. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers MPF (5'-CCAACCTCCATGTAGCTGATAGCT-3') and MPR (5'-TATCGCCAGGTAAAAACTCCTTCT-3') and the detection probe (5'-FAM-ATCCTTGTTGTAAGGCTTGTAATCG-TAMRA-3'). Amplification conditions were 1 cycle of 2 min at 50°C and 1 cycle of 10 min at 95°C, followed by 45 cycles of 15 s at 95°C and 60 s at 60°C.
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers CPF (5'-AAGGGCTATAAAGGCGTTGCT-3') and CPR (5'-TGGTCGCAGACTTTGTTCCA-3') and detection probe (5'-FAM-TCCCCTTGCCAACAGACGCTGG-TAMRA-3'). The amplification conditions used were the same as those for M. pneumoniae.
Participant 9. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers FPMPN (5'-TCTGGCGTGGATCTCTCCC-3') and RPMPN (5'-GACACTCGTGCTTGGTAACTGC-3'). Detection was done by using SYBR green. Amplification conditions were 1 cycle at 95°C for 10 min followed by 50 cycles each of 10 s at 95°C and 20 s at 65°C. In 2004, the same assay conditions were used, but real-time detection was done using probe MPN (5'-FAM-GAAGGAATGATAAGGCTTCAAGTGGACAAAGTG-TAMRA-3').
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using the slightly modified MOMP VD2 primers of Tondella et al. (26), FPCPN (5'-CTCGTTGGTTTATTCGGAGTT-3') and RPCPN (5'-CCAAGAGAAAGAGGTGTCTGTG-3'). Detection was done by using SYBR green. Amplification conditions were the same as those for M. pneumoniae. In 2004, the slightly modified MOMP VD2 detection probe described by Tondella et al. (26), CPN (5'-FAM-AAGGTACTACTGTAAATGCAAATGAAC-TAMRA-3'), was used.
Participant 10. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers MNP1FW (5'-CCAACCAAACAACAACGTTCA-3') and MNP1REV (5'-CCTTGACTGGAGGCCGTTAA-3') and detection probe MPN (5'-FAM-TCAATCCGAATAACGGTGACTTCTTACCACTG-TAMRA-3') in 2004. Amplification conditions were 1 cycle at 50°C for 2 min followed by 1 cycle at 95°C for 10 min and 40 cycles each 15 s at 95°C and 60 s at 60°C.
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers CPF1 (5'-GGACCTTACCTGGACTTGACATGT-3') and CPR1 (5'-CCATGCAGCACCTGTGTATCTG-3') and detection probe (5'-FAM-TGACCACTGTAGAAATACAGCTTTCCGCAAGG-TAMRA-3'). The amplification conditions used were the same as those for M. pneumoniae.
Participant 11. The M. pneumoniae and C. pneumoniae monoplex real-time PCR assays were done as described previously by Templeton et al. (24, 25). For the real-time multiplex PCR, primers, probes, and amplification conditions similar to those in the monoplex assays were used.
Participant 12. M. pneumoniae and C. pneumoniae DNA was detected as described by Abele-Horn et al. and Tong and Sillis (1, 27), respectively.
Participant 13. C. pneumoniae DNA was amplified by an in-house real-time PCR developed by participant number 14 using primers chlapneu 171F and chlapneu 250R and detection probe chlapneu 200T. Amplification conditions were 1 cycle at 50°C for 2 min and 1 cycle at 95°C for 10 min, followed by 45 cycles each of 15 s at 95°C and 60 s at 60°C.
Participant 14. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers MP-T1 (5'-ACACCAAGTT CGCGAGTGCTA-3') and MP-T2 (5'-CCGTCCTGCGTGGTTAAACTAT-3') and detection probe MP-S (5'-FAM-ATCCCGACTCGTTAAAGCAGGATAAGATT-TAMRA-3'). Amplification conditions were 1 cycle at 50°C for 2 min and 1 cycle at 95°C for 10 min, followed by 45 cycles each of 15 s at 95°C and 60 s at 60°C.
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers chlapneu 171F (5'-CGGCTAGAAATCAATTATAAGACTGAAG-3') and chlapneu 250R (5'-TGGCGAATGACACCATGATC-3') and detection probe chlapneu 200T (5'-FAM-AAATCTGCATCTCCCTCACGAATATGCTCA-TAMRA-3'). The amplification conditions used were the same as those for M. pneumoniae.
Participant 15. Conventional and monoplex real-time NASBA for the detection of C. pneumoniae RNA was done as described previously (23). MX-real-time NASBA was done as described previously (19).
Participant 16. M. pneumoniae and C. pneumoniae DNA was detected as described by Hardegger et al. and Welti et al. (10, 32), respectively.
Participant 17. M. pneumoniae and C. pneumoniae MOMP VD4 DNA was detected as described by Hardegger et al. and Tondella et al. (10, 26), respectively.
Participant 18. M. pneumoniae and C. pneumoniae MOMP VD2 DNA was detected by monoplex PCR using primers and probes described by Grondahl et al. and Tondella et al. (9, 26), respectively. Amplification conditions were 1 cycle of 10 min at 94°C followed by 40 cycles of 30 s at 94°C, 50°C, and 72°C, with a final cycle of 5 min at 72°C. The same primers, probes, and amplification conditions were used in the multiplex PCR.
Participant 19. The commercially available Pneumoplex assay (Prodesse, Waukesha, Wis.) was used for the detection of both M. pneumoniae and C. pneumoniae according to the instructions of the manufacturer.
Participant 20. M. pneumoniae DNA was amplified for participant 20 as described by Ursi et al. (30).
Internal controls. Most participants used a generic internal control such as the ones described by Hoymans et al., Ieven et al., and Tong and Sillis (11, 12, 27) or phocine herpes virus (25), except for participant 1, who measured for the presence of U1A mRNA in 2002, participant 2, who spiked another aliquot of the same sample with the respective organism, and the commercially available assays, which had their own internal controls.
Statistical analysis. The Fisher exact test or the chi-squared test was applied for the calculation of the significance between results during the preparation of the QC panels and for the performance of the different assays.
RESULTS
Preparation of the QC panels. To monitor the stability of the samples in the panels, aliquots kept at –80°C were examined by PCR and NASBA after one and two freezing-thawing cycles. All samples tested positive in 9 to 15 different runs for C. pneumoniae and M. pneumoniae, respectively, except for the samples with the lowest concentrations of M. pneumoniae, which produced positive results in 2 out of 15 PCR runs and 6 out of 15 NASBA runs (P = 0.10) in 2002.
When spiked BAL pools were tested by NASBA, an input of 50 CCU/100 μl, 500 CCU/100 μl, and 5000 CCU/100 μl of M. pneumoniae yielded 9 out of 9 positive results each time. When spiked BAL pools were tested by NASBA, an input of 4.9 IFU/100 μl, 49 IFU/100, and 490 IFU/100 μl of C. pneumoniae yielded 9 out of 9 positive results each time.
Performance of the laboratories. Some laboratories provided two amplification results obtained by two different procedures, such as mono- and multiplex reactions. Primers and probes used, as well as amplification conditions applied, were highly diverse.
Tables 1 to 4 show the different extraction, amplification, and detection procedures applied, as well as the target molecules used.
The 2002 M. pneumoniae QC. For M. pneumoniae, 13 datasets were obtained (Table 5). All procedures were in-house developed, except one participant, no. 4, who used the commercial kit from Minerva Biolabs GmbH (Berlin, Germany).
Eleven out of 13 assays applied one PCR and two NASBA; 9 were based on real-time detection, 3 used agarose gel electrophoresis, and 1 used electrochemiluminescent detection. The target most often used was the P1 gene in 9 out of 13 (69.2%) assays.
One data set, from participant 10, was excluded from the global analysis because of the uniformly negative results. All samples containing 5,000 CCU/100 μl were found positive. The samples containing 500 CCU/100 μl were scored positive in 20 out of 36 (55.5%) of the cases when suspended in saline and in 21 out of 36 (58.3%) of the cases when suspended in BAL fluid (P = 0.8). The lowest concentrations tested, 50 CCU/100 μl, scored positive in 10 out of 36 (27.8%) of the cases, a score which is not different from that obtained during the assessment of the panels before their distribution (P = 0.3).
Participant 2 did not include any negative controls and scored 3 out of 8 (37.5%) false-positive results. False-positive results were reported by a second laboratory (no. 5), and inhibition of the reaction in three instances was reported by two laboratories. In 7 out of 13 tests (53.9%), an internal control was used to monitor inhibition of the reaction in each sample. Eight (66.7%) participants reported the use of dUTP-uracyl-N-glycosylase (dUTP-UDG) to avoid false-positive results due to carryover, but nevertheless participants 2 and 5 using dUTP-UDG produced false-positive results.
The 2004 M. pneumoniae QC. The 2004 M. pneumoniae QC resulted in 18 datasets; 11 sets were obtained from 10 laboratories that participated in the 2002 QC (Table 6).
All participants used a commercially available nucleic acid extraction kit. Agarose gel electrophoresis was no longer used by any of the participants. In 16 out of 18 assays PCR was used, NASBA was used twice. Real-time detection was applied in 15 out of 18 tests, and an enzyme immunoassay (EIA) was applied in 3 out of 18 tests. The target most often used was the P1 gene (66.7%).
The samples containing 5,000 CCU/100 μl scored positive in 51 out of 54 (94.1%) of the cases, and those containing 500 CCU/100 μl were scored positive in 49 out of 54 (90.2%) of the cases. Four participants applied a multiplex assay. In three instances, participants 1, 11, and 18, an in-house-developed multiplex assay was performed in parallel with a monoplex real-time reaction; in one of these the multiplex reaction was less sensitive. The commercially available multiplex assay, the Pneumoplex assay from Prodesse (Waukesha, Wis.) (used by participant 19), was also less sensitive.
An internal control to monitor inhibition of the reaction in each sample was used in 15 out of 18 (83.3%) of the tests. Nine (60.0%) participants reported the use of dUTP-UDG to avoid false-positive results; none were recorded. One participant did not include negative controls.
The 2002 C. pneumoniae QC. The C. pneumoniae panels resulted in 16 datasets, of which 10 were delivered within the requested period of 4 weeks (Table 7). Twelve out of 16 amplification assays applied a PCR, 4 used a NASBA, 13 datasets used real-time detection, 2 participants used agarose gel electrophoresis, and 1 used electrochemiluminescent detection. Targets most often used were the 16S rRNA (NASBA), a cloned PstI fragment, the MOMP gene, or a 16S rRNA gene fragment in four (25%), four (25%), three (18.8%), and three (18.8%) of the assays, respectively. The other targets were only used once and were the PmP4 gene and a cytadhesin gene.
No false-positive results were recorded among the 124 samples. Samples containing 490 IFU/100 μl scored positive in 44 out of 45 (97.8%) instances, samples containing 49 IFU/100 μl suspended in saline scored positive in 46 out of 48 (95.8%) cases, and samples suspended in BAL fluid scored positive in 39 out of 45 (86.7%) cases (P = 0.1). The lowest concentration tested, 4.9 IFU/100 μl, scored positive in 32 out of 48 (66.7%) of the cases.
The multiplex NASBA, participant 15b, targeting the M. pneumoniae, C. pneumoniae, and Legionella pneumophila 16S rRNA, was significantly less sensitive than the monoplex NASBA of participant 15; both were in-house developed in the same laboratory.
Six out of 16 tests (37.5%) used an internal control to monitor inhibition of the reaction in each sample. Nine (64.3%) participants reported the use of dUTP-UDG.
The 2004 C. pneumoniae QC. The C. pneumoniae panels resulted in 18 datasets (Table 8). Thirteen sets were obtained from 12 laboratories that participated in the 2002 QC.
Sixteen assays applied a PCR, and 2 applied a NASBA; 15 datasets used real-time detection, and in 3 assays an EIA was used. Targets most often used were the MOMP gene, the cloned PstI fragment, and the 16S rRNA gene in 6 out of 18 (33.3%), 5 out of 18 (27.8%), and in 3 out of 18 (16.7%) assays, respectively.
Among 72 C. pneumoniae-negative samples, 4 were reported positive by one participant who did not use dUTP-UDG. Samples containing 490 IFU/100 μl were scored positive for 17 out of 18 (94.4%) of the cases, samples containing 49 IFU/100 μl scored positive in 14 out of 18 (83.3%) cases, and those containing 4.9 IFU/100 μl scored positive in 15 out of 18 (83.3%) of the cases. The lowest concentration tested, 0.49 IFU/100 μl, scored positive in 18 out of 54 (33.3%) of the cases.
Three participants submitted results of a multiplex NAAT: a real-time in-house developed NASBA (participant 1), an in-house-developed PCR (participant 18), and the commercially available Pneumoplex from Prodesse. The multiplex PCR was less sensitive than the monoplex counterpart. Laboratory 19, applying the Pneumoplex, did not detect a single C. pneumoniae-positive sample.
In 15 out of 18 tests (83.3%), an internal control was used to monitor inhibition of the reaction in each sample. Nine (56.3%) participants reported the use of dUTP-UDG to avoid false-positive results. One participant did not include negative controls.
Samples prepared with BAL pool 1 (clear), pool 2 (more viscous), and pool 3 (blood stained) spiked with 0.49 IFU/100 μl scored positive in 4 out of 17, 8 out of 17, and 4 out of 17 cases, respectively. The results of participant 8 were not included in this calculation due to the high number of false-positive results.
Generally, participants targeting the MOMP gene (participants 6, 8, 9, 17, and 18) found fewer samples to be C. pneumoniae positive than participants targeting the 16S rRNA, 16S rRNA gene, or PstI fragment. Participants 13 and 14 applied the same in-house real-time assay and obtained identical results.
DISCUSSION
QC for the NAATs for M. pneumoniae and C. pneumoniae are particularly important, because in the clinical laboratory there is no practical alternative for the detection of both agents. The two QCs reported here illustrate the technical evolution between 2002 and 2004: disappearance of gel electrophoresis for the detection of the amplicons, a substantial increase in monitoring of inhibition of the amplification reaction, and an increased use of positive and negative samples and of multiplex amplification reactions. The use of dUTP-UDG remained constant. Inclusion of positive and negative controls did not differ between the two QCs.
Gradually, laboratories performing NAATs for the detection of both M. pneumoniae and C. pneumoniae use the same amplification conditions for both assays. The advantage is that in the same run different organisms can be targeted, decreasing the total turnaround time.
In 2002 two participants obtained false-positive results in the M. pneumoniae panel (one of these performing routinely up to 50 tests on a monthly basis). In the 2004 M. pneumoniae panel, both participants reported all samples correctly. No false-positive C. pneumoniae results were recorded in 2002, and 11 out of 15 participants produced correct results. In 2004, one participant produced false-positive results for all four C. pneumoniae-negative samples. Other PCR quality assessment studies have also recorded false-positive results (8, 20, 22, 31, 33). Perhaps this problem will not be eliminated until sample processing can be automated.
In this quality control study, the major problem was the occurrence of false negatives, especially when testing the M. pneumoniae 2002 proficiency panel. The results obtained by the participants in 2002 for the samples containing 50 CCU/100 μl of M. pneumoniae are in line with those of the reference laboratory (P = 0.13). This low concentration was not included in 2004. There were no differences between the suspensions prepared in saline (20 positive results) and those prepared in BAL (21 positive results). Participant 10, which failed to detect any M. pneumoniae-positive samples, did not participate in the M. pneumoniae 2004 panel and also failed to detect small numbers of C. pneumoniae in 2002 and 2004.
The lowest concentration of M. pneumoniae is clearly at the limit of detectability, therefore only the 5,000 and 500 CCU/100 μl samples were taken into consideration for the global analysis. Thus, 2 out of 12 participants produced correct results, 2 out of 12 produced false-positive results, and 9 participants produced between 0 of 9 and 8 of 9 correct results. There was considerable improvement in 2004, when 15 out of 18 participants produced 100% M. pneumoniae correct results without any false positives.
Suspensions of both M. pneumoniae and C. pneumoniae in saline and in BAL were included in the 2002 panels to reveal possible inhibitors: two samples in saline were found in one laboratory to inhibit the reaction, and one sample in BAL was found in a second laboratory. Both laboratories used an extraction kit from Roche. No inhibitors were recorded in 2004, although one of the BAL pools was blood stained. Both laboratories reporting inhibition in the 2002 panel used the same extraction procedure for the 2004 M. pneumoniae panel. Participant 8 did not change the amplification and detection procedures. For the 2004 panel, participant 4 replaced the Minerva Biolabs kit with an in-house-developed real-time PCR.
Two out of three participants (18b and 19) obtaining false-negative results in the 2004 M. pneumoniae QC applied a multiplex NAAT, one of them being a commercial kit (Pneumoplex; Prodesse). Two other participants using an in-house-developed multiplex NAAT obtained correct results.
Four participants failed to detect C. pneumoniae in the sample containing 49 IFU/100 μl in the 2004 QC, two of them applied a multiplex NAAT, with one being a commercial kit (Pneumoplex; Prodesse).
In the 2004 QC tests, multiplex NAAT was applied. PCR was used by participants 11, 18, and 19 (Table 6 and Table 8), and NASBA was used by participants 1 and 15. The multiplex PCRs scored a smaller number of samples positive than most of the monoplex tests. One multiplex PCR was the commercially available Pneumoplex (Prodesse). Although the limit of detection of this assay was reported to be 5 CCU/ml for M. pneumoniae and 0.01 50% tissue culture infective dose/ml for C. pneumoniae and 10 copies of recombinant DNA for each organism (13), the test did not perform well in this evaluation. Participant 19 tested a second panel of each organism with similar results. The assay was performed correctly. The manufacturer was contacted and is aware of the sensitivity problems of the Pneumoplex assay. They intend to improve the sensitivity of the test.
Multiplex assays are somewhat less sensitive than monoplex assays, but until the number of organisms present in clinical specimens of diseased individuals is known, it is impossible to state whether the degree of sensitivity attained is acceptable.
From previous investigations, we have learned that the sensitivity of molecular diagnostics using respiratory samples may be compromised by the presence of inhibitory factors in the samples (2, 5, 8, 14, 28, 29). Therefore, three different BAL pools were used in the 2004 QC panels. There was no significant difference in the positivity rates between the three different BAL pools used to prepare the suspensions, although the suspension in the blood-stained BAL had the lowest number of positive results. Unexpectedly, the viscous suspension scored a higher number of positive results than the clear BAL suspension.
There was considerable variation between the different PCR protocols applied by the participants. Evaluation of the various PCR protocols showed no apparent association between their performance and the particular variables of the PCR method used, except for the C. pneumoniae MOMP amplification protocols. Participants 6, 17, and 18 used the protocol described by Tondella et al. (26) and obtained lower positivity scores than participants using a different protocol and targeting a different gene for the detection of C. pneumoniae. Participant 9 had a slightly better positivity score using slightly modified MOMP VD2 primers and detection probe. However, the results of such a comparison must be interpreted with caution due to the relatively small number of samples, the small number of participating laboratories, and the high diversity of the methods used.
The discrepant results from two successive QC exercises observed among different laboratories, some of which lacked experience in NAAT, illustrate the importance of training of personnel and the use of negative and positive controls in the preparative and amplification phases. This is illustrated by the three laboratories that reported false-positive results; one of them did not use negative controls in the preparation, amplification, and detection procedures.
The different performance characteristics of the amplification-based assays used may explain discrepant findings from published studies that used NAATs to determine M. pneumoniae and C. pneumoniae prevalence in patient populations. Sensitivity and specificity issues should be addressed before publishing clinical and epidemiological studies of M. pneumoniae and C. pneumoniae infections based on the detection of bacterial DNA and RNA in clinical specimens by NAAT.
This study also underlines the need for reference reagents and standard operating procedures to enable experienced technicians to perform quality control assessment of nucleic acid amplification methods and thus perform reliable diagnostic molecular amplification techniques on a routine basis.
ACKNOWLEDGMENTS
The organization of the QC exercises for the Belgian Centers for Molecular Diagnostics was supported by the public health service, RIZIV-INAMI.
REFERENCES
Abele-Horn, M., U. Busch, H. Nitzschiko, et al. 1998. Molecular approaches to diagnosis of pulmonary diseases due to Mycoplasma pneumoniae. J. Clin. Microbiol. 36:548-551.
Apfalter, P., F. Blasi, J. Boman, C. A. Gaydos, M. Kundi, M. Maass, A. Makristathis, A. Meijer, R. Nadrchal, K. Persson, M. L. Rotter, C. Y. W. Tong, G. Stanek, and A. M. Hirschl. 2001. Multicenter comparison trial of DNA extraction methods and PCR assays for detection of Chlamydia pneumoniae in endarterectomy specimens. J. Clin. Microbiol. 39:519-524.
Apfalter, P., O. Assadian, F. Blasi, J. Boman, C. A. Gaydos, M. Kundy, A. Makristathis, M. Nehr, M. L. Rotter, and A. M. Hirschl. 2002. Reliability of nested PCR for detection of Chlamydia pneumoniae DNA in atheromas: results from a multicenter study applying standardized protocols. J. Clin. Microbiol. 40:4428-4434.
Bernet, C., M. Garret, B. de Barbeyrac, C. BeBear, and J. Bonnet. 1989. Detection of Mycoplasma pneumoniae by using the polymerase chain reaction. J. Clin. Microbiol. 27:2492-2496.
Birkebaek, N. H., J. S. Jensen, T. Seefeldt, J. Degn, B. Huniche, P. L. Andersen, and L. Ostergaard. 2000. Chlamydia pneumoniae infection in adults with chronic cough compared with healthy blood donors. Eur. Respir. J. 16:108-111.
Boman, J., C. A. Gaydos, and T. C. Quinn. 1999. Molecular diagnosis of Chlamydia pneumoniae infection. J. Clin. Microbiol. 37:3791-3799.
Campbell, L. A., M. Perez Melgosa, D. J. Hamilton, C. C. Kuo, and J. T. Grayston. 1992. Detection of Chlamydia pneumoniae by polymerase chain reaction. J. Clin. Microbiol. 30:434-439.
Chernesky, M., M. Smieja, J. Schachter, J. Summersgill, L. Schindler, N. Solomon, K. Campbell, L. A. Campbell, A. Cappuccio, C. Gaydos, S. Chong, J. Moncada, J. Phillips, D. Jang, B. J. Wood, A. Perich, M. Hammerschlag, M. Cerney, and J. Mahony. 2002. Comparison of an industry derived LCx Chlamydia pneumoniae PCR research kit to in-house assays performed in five laboratories. J. Clin. Microbiol. 40:2357-2362.
Grndahl, B., W. Puppe, A. Hoppe, I. Kühne, J. A. I. Weigl, and H.-J. Schmitt. 1999. Rapid identification of nine microorganisms causing acute respiratory tract infections by single tube multiplex reverse transcription PCR: feasibility study. J. Clin. Microbiol. 37:1-7.
Hardegger, D., D. Nadal, W. Bossart, M. Altwegg, and F. Dutly. 2000. Rapid detection of M. pneumoniae in clinical samples by real-time PCR. J. Microbiol. Methods 41:45-51.
Hoymans, V. Y., J. M. Bosmans, D. Ursi, W. Martinet, F. Wuyts, E. Van Marck, M. Altwegg, and C. J. Vrints. 2004. Immunohistostaining assays for detection of Chlamydia pneumoniae in atherosclerotic arteries indicate cross-reactions with nonchlamydial plaque constituents. J. Clin. Microbiol. 42:3219-3224.
Ieven, M., D. Ursi, H. Van Bever, W. Quint, H. G. M. Niesters, and H. Goossens. 1996. Detection of Mycoplasma pneumoniae by two polymerase chain reactions and role of M. pneumoniae in acute respiratory tract infections in pediatric patients. J. Infect. Dis. 173:1445-14452.
Khana, M., J. Fan, K. Pehler-Harrington, C. Waters, P. Douglass, J. Stallock, S. Kehl, and K. J. Henrickson. 2005. The Pneumoplex assay, a multiplex PCR-enzyme hybridization assay that allows simultaneous detection of five organisms, Mycoplasma pneumoniae, Chlamydia (Chlamydophila) pneumoniae, Legionella pneumophila, Legionella micdadei, and Bordetella pertussis, and its real-time counterpart. J. Clin. Microbiol. 43:565-571.
Loens, K., D. Ursi, M. Ieven, P. van Aarle, P. Sillekens, P. Oudshoorn, and H. Goossens. 2002. Detection of Mycoplasma pneumoniae in spiked clinical samples by nucleic acid sequence-based amplification. J. Clin. Microbiol. 40:1339-1345.
Loens, K., M. Ieven, D. Ursi, H. Foolen, P. Sillekens, and H. Goossens. 2003. Application of NucliSens basic kit for the detection of Mycoplasma pneumoniae in respiratory specimens. J. Microbiol. Methods 54:127-130.
Loens, K., M. Ieven, D. Ursi, T. Beck, M. Overdijk, P. Sillekens, and H. Goossens. 2003. Detection of Mycoplasma pneumoniae by real-time nucleic acid sequence-based amplification. J. Clin. Microbiol. 41:4448-4450.
Loens, K., D. Ursi, H. Goossens, and M. Ieven. 2003. Minireview: molecular diagnosis of Mycoplasma pneumoniae infections. J. Clin. Microbiol. 41:4915-4923.
Loens, K., T. Beck, H. Goossens, D. Ursi, M. Overdijk, P. Sillekens, and M. Ieven. 2006. Development of conventional and real-time NASBA for the detection of Chlamydophila pneumoniae in respiratory specimens. J. Clin. Microbiol., submitted for publication.
Loens, K., T. Beck, H. Wouters, D. Ursi, M. Overdijk, P. Sillekens, H. Goossens, and M. Ieven. 2004. Comparison of real-time mono and multiplex NASBA for the detection of M. pneumoniae, C. pneumoniae and L. pneumophila in respiratory specimens. 14th European Congress of Clinical Microbiology and Infectious Diseases, European Society for Clinical Microbiology and Infectious Diseases, 01/05-04/05/2004, Prague, Tsjech Republic. Clin. Microbiol. Infect. 10(Suppl. 3):7.
Muir, P., A. Ras, P. E. Klapper, G. M. Cleator, K. Korn, C. Aepinus, A. Fomsgaard, P. Palmer, A. Samuelsson, A. Tenorio, B. Weissbrich, and A. M. Van Loon. 1999. Multicenter quality assessment of PCR methods for detection of enteroviruses. J. Clin. Microbiol. 37:1409-1414.
Ramirez, J. A., and the Chlamydia pneumoniae/Atherosclerosis Study Group. 1996. Isolation of Chlamydia pneumoniae from the coronary artery of a patient with coronary atherosclerosis. Ann. Intern. Med. 125:979-982.
Schirm, J., A. M. Van Loon, E. Valentine-Thon, P. E. Klapper, J. Reid, and G. M. Cleator. 2002. External quality assessment program for qualitative and quantitative detection of hepatitis C virus RNA in diagnostic virology. J. Clin. Microbiol. 40:2973-2980.
Sillekens, P., M. Overdijk, I. Roosmalen, K. Loens, M. Ieven, P. Haima, and M. Jacobs. 2002. Molecular diagnosis of Chlamydia pneumoniae infections by using NASBA with real time detection. 12th European Congress of Clinical Microbiology and Infectious Diseases, European Society for Clinical Microbiology and Infectious Diseases, 24/04-27/04/2002, Milan, Italy. Clin. Microbiol. Infect. Suppl. P730.
Templeton, K. E., S. A. Scheltinga, A. W. Graffelman, P. H. van den Broek, and E. Claas. 2003. Comparison and evaluation of real-time PCR, real-time nucleic acid sequence-based amplification, conventional PCR, and serology for diagnosis of Mycoplasma pneumoniae. J. Clin. Microbiol. 41:4366-4371.
Templeton, K. E., S. A. Scheltingha, W. C. J. F. M. Van den Eeden, A. W. Graffelman, P. H. van den Broek, and E. Claas. 2005. Improved diagnosis of etiology of community-acquired pneumonia using real-time PCR. Clin. Infect. Dis. 41:345-351.
Tondella, M. L. L., D. F. Talkington, B. P. Holloway, S. F. Dowell, K. Cowley, M. Soriano-Gabarro, M. S. Elkind, and B. S. Fields. 2002. Development and evaluation of real-time PCR-based fluorescence assays for detection of Chlamydia pneumoniae. J. Clin. Microbiol. 40:575-583.
Tong, C. Y. W., and M. Sillis. 1993. Detection of Chlamydia pneumoniae and Chlamydia psittaci in sputum samples by PCR. J. Clin. Pathol. 46:313-317.
Ursi, J. P., D. Ursi, M. Ieven, and S. R. Pattyn. 1992. Utility of an internal control for the polymerase chain reaction. APMIS 100:635-639.
Ursi, D., M. Ieven, H. P. Van Bever, and H. Goossens. 1998. Construction of an internal control for the detection of Chlamydia pneumoniae by PCR. Mol. Cell. Probes 12:235-238.
Ursi, D., K. Dirven, K. Loens, M. Ieven, and H. Goossens. 2003. Detection of Mycoplasma pneumoniae in respiratory samples by real-time PCR using an inhibition control. J. Microbiol. Methods 55:149-153.
Valentine-Thon, E., A. M. Van Loon, J. Schirm, J. Reid, P. E. Klapper, and G. M. Cleator. 2001. European proficiency testing program for molecular detection and quantitation of hepatitis B virus DNA. J. Clin. Microbiol. 39:4407-4412.
Welti, M., K. Jaton, M. Altwegg, R. Sahli, A. Wenger, and J. Bille. 2003. Development of a multiplex real-time quantitative PCR assay to detect Chlamydia pneumoniae, Legionella pneumophila, and Mycoplasma pneumoniae in respiratory tract secretions. Diagn. Microbiol. Infect. Dis. 45:85-95.
Zaaijer, H. L., H. T. M. Cuypers, H. W. Reesink, I. N. Winkel, G. Gerken, and P. N. Lelie. 1993. Reliability of polymerase chain reaction for detection of hepatitis C virus. Lancet 341:722-724.(K. Loens, T. Beck, D. Urs)
ABSTRACT
The quality performance of laboratories for the detection of Mycoplasma pneumoniae and Chlamydophila pneumoniae by two quality control (QC) exercises with a 2-year interval was investigated. For the 2002 QC exercise, specimens were spiked with M. pneumoniae at concentrations of 5,000, 500, 50, and 0 color-changing units (CCU)/100 μl. The limit of detectability was 50 CCU/100 μl. Therefore, this concentration was omitted from the 2004 panel and was excluded from the analysis. In 2002, 2 out of 12 participants obtained 100% correct results, 2 out of 12 produced false-positive results, and 10 out of 12 had between 0 out of 9 and 8 out of 9 correct positive results. In 2004, correct results were obtained in 15 out of 18 tests, and no false-positive results were reported. In 2002, specimens were spiked with C. pneumoniae at concentrations of 490, 49, 4.9, and 0 inclusion-forming units/100 μl (IFU/100 μl). In the 2004 panel, samples spiked with a lower dilution of 0.49 IFU/100 μl were added to the panel. For the C. pneumoniae QC, correct results were produced in 12 out of 16 and 13 out of 18 tests in 2002 and in 2004, respectively. Both multiplex PCR and nucleic acid sequence-based amplification (NASBA) formats scored a smaller number of samples positive than the monoplex reactions.
INTRODUCTION
A multitude of nucleic acid amplification tests (NAATs) have been described for the detection of Mycoplasma pneumoniae and Chlamydophila pneumoniae in respiratory specimens (6, 17). In addition to in-house PCR tests, commercial kits are becoming available, such as the LCx for C. pneumoniae (Abbott Laboratories) (8) and the NucliSens Basic kit (bioMerieux), for which the primers and the target-specific biotinylated capture probe are to be synthesized for each target by the user (15).
Although all tests aim to be rapid, sensitive, specific, and easy to perform, results of NAATs may be unreliable because of cross-contamination, inappropriate treatment of the clinical samples leading to loss of target nucleic acid, or the presence of inhibitors (2, 5, 12, 14, 26, 27).
To date, only one study compared the results of different NAATs for the detection of C. pneumoniae in respiratory specimens in different centers (8), and four studies compared the results of amplification methods performed in different centers for the detection of C. pneumoniae in atheroma specimens (2, 3, 11, 21). To our knowledge, no such studies have been published for M. pneumoniae.
The aim of this study was to assess the quality performance of laboratories for the detection of M. pneumoniae and C. pneumoniae by two quality control (QC) exercises with a 2-year interval.
MATERIALS AND METHODS
Participating laboratories. The participating laboratories are as follows, in alphabetical order: Academisch Ziekenhuis Vrije Universiteit Brussel, Brussels, Belgium; Algemeen Ziekenhuis Sint Augustinus, Wilrijk, Belgium; Algemeen Ziekenhuis Sint Jan, Brugges, Belgium; bioMerieux, Boxtel, The Netherlands; Centre Hospitalier Regional de la Citadelle, Liege, Belgium; Cliniques Universitaires Universite Catholique de Louvain de Mont-Godinne, Yvoir, Belgium; Groupement des Centres Hospitaliers de Jolimont-Lobbes et de Tubize-Nivelles, La Louviere, Belgium; Institut Jules Bordet, Brussels, Belgium; Institution de Pathologie et de Genetique-Loverval, Loverval, Belgium; Leids Universitair Medisch Centrum, Leiden, The Netherlands; Onze Lieve Vrouwe Ziekenhuis, Aalst, Belgium; Ospedale Maggiore di Milano, Milan, Italy; Public Health Laboratory Friesland, Leeuwarden, The Netherlands; Universite Libre de Bruxelles-Erasme, Brussels, Belgium; University Medical Center-Brugmann, Brussels, Belgium; University of Antwerp, Wilrijk, Belgium; Universitair Ziekenhuis Katholieke Universiteit Leuven, Leuven, Belgium; University Hospital Antwerp, Edegem, Belgium; Virga Jesse Ziekenhuis, Hasselt, Belgium; and Ziekenhuizen Noord Antwerpen, Antwerp, Belgium.
Preparation of proficiency panels. A stock suspension of M. pneumoniae strain ATCC 29085 was quantitated by incubation of 10-fold dilutions in triplicate in SP4 medium at 37°C. The cultures were monitored for 2 months, and the titer was expressed as color-changing units (CCU) per milliliter; 1 CCU corresponds to 10 to 100 cells (4).
A stock suspension of C. pneumoniae strain ATCC VR-1355 was quantified by incubation of five replicates of 10-fold dilutions on confluent layers of Hep-2 cells. The vials were centrifuged at 3,500 rpm at 25°C for 60 min and incubated at 37°C. After 1 h, the medium was replaced by fresh culture medium containing 1 mg/liter cycloheximide. After 3 days, cells were fixed with 96% ethanol and stained by the fluorescent antibody technique with specific mouse monoclonal antibodies and fluorescein-labeled rabbit anti-mouse immunoglobulin (both from Dako A/S, Glostrup, Denmark). The titer was expressed as inclusion-forming units (IFU) per milliliter.
Bronchoalveolar lavage (BAL) specimens were collected from patients in the University Hospital of Antwerp and stored at –20°C. BAL pools were negative for M. pneumoniae and C. pneumoniae as tested by nucleic acid sequence-based amplification (NASBA) (16, 21) and by in-house-developed PCR (12, 27).
The 2002 proficiency panel consisted of parts A and B: part A was prepared in sterile physiologic saline, and part B was prepared in BAL. The negative samples were prepared in a laminar flow cabinet in a separate room prior to the preparation of the positive samples. Materials and pipettes had never been used for M. pneumoniae- or C. pneumoniae-related work previously.
M. pneumoniae parts A and B each contained four negative samples and three samples with 500 CCU/100 μl of solution (see Table 5); part A also contained three samples with 50 CCU/100 μl, and part B contained three samples with 5,000 CCU/100 μl. For C. pneumoniae (see Table 7), parts A and B contained four negative samples each and three samples with 49 IFU/100 μl; furthermore, part A contained three samples with 4.9 IFU/100 μl, and part B contained three samples with 490 IFU/100 μl.
For the 2004 proficiency tests, all samples were prepared in BAL. Each panel contained four negative samples, prepared as described above. Three different BAL pools were used: pool 1 was fluid, pool 2 was more viscous, and pool 3 was blood stained. Three samples of the M. pneumoniae panel (see Table 6) were spiked with 5,000 CCU/100 μl, and three samples were spiked with 500 CCU/100 μl. The C. pneumoniae proficiency panel (see Table 8) included one sample with 490 IFU/100 μl, one sample with 49 IFU/100 μl, one sample with 4.9 IFU/100 μl, and three samples with 0.49 IFU/100 μl BAL.
The proficiency panels were coded and sent refrigerated on the same day or were stored at –80°C until they were shipped frozen to the participants. The samples were to be tested as routine specimens, and the results were returned within 4 weeks together with a questionnaire collecting information on the procedures applied.
Quality control assurance. Fifteen serial dilutions of M. pneumoniae were tested by NASBA and PCR (12, 16) after one and two freeze-thaw cycles, mimicking the conditions of the preparation of the panels and on arrival at the participating laboratory. Similar tests were performed for the detection of nine serial dilutions of C. pneumoniae by NASBA and PCR (18, 27). Furthermore, each BAL pool was spiked in triplicate with a serial dilution of M. pneumoniae and tested by both NAATs. Quality control assurance of the panels was performed by NASBA and the in-house-developed PCR.
Methods used in the participating centers for the extraction of M. pneumoniae and C. pneumoniae nucleic acids. Each laboratory followed its standard procedures for M. pneumoniae and C. pneumoniae detection. Thus, the procedures varied from laboratory to laboratory.
The following commercially available extraction kits were used (Tables 1 to 4): Amplicor HCV specimen preparation kit (Roche, Mannheim, Germany), Nucleospin tissue (Clontech, Mount View, CA), NucliSens Basic kit (bioMerieux, Boxtel, The Netherlands), Puregene genomic DNA for gram-positive bacteria (Gentrasystems, Minneapolis, MN), QIAamp DNA minikit (QIAGEN, Hilden, Germany), QIAamp DNA blood kit (QIAGEN), QIAamp stool kit (QIAGEN), QIAamp DNA tissue kit (QIAGEN), Roche Amplicor Sputum Preparation kit (Roche), and the Roche High Pure PCR Template Preparation kit (Roche).
Three participants used in-house nucleic acid extraction methods for the M. pneumoniae or C. pneumoniae 2002 proficiency panels: participant number 5, 10 min of boiling; participant number 10, 5 cycles of 1 min of freeze boiling; and participant number 14, a proteinase K pretreatment followed by 10 min of boiling.
Methods used in the participating centers for the amplification and detection of M. pneumoniae and C. pneumoniae nucleic acids. Participant 1. Conventional nucleic acid sequence-based amplification (NASBA) for the detection of M. pneumoniae RNA in 2002 was done as described earlier (14, 15). Monoplex real-time NASBA for the detection of M. pneumoniae RNA was done as described by Loens et al. (16). Conventional and monoplex real-time NASBA for the detection of C. pneumoniae RNA was done as described by Loens et al. (18). Multiplex (MX)-real-time NASBA for the detection of RNA from both organisms was done as described previously (19).
Participant 2. M. pneumoniae DNA was detected as described by Hardegger et al. (10). C. pneumoniae DNA was amplified by an in-house-developed real-time PCR based on the PstI fragment-updated sequence (7) using primers CPNFW (5'-TGGAGATAAAATGGCTGGACG-3') and CPNREV (5'-TATGGCATATCCGCTTCGG-3') and detection probe (5'-6-carboxyfluorescein [FAM]-CACGGAAATAAAGGTGTTGTTTCCAAAATCG-6-carboxytetramethylrhodamine [TAMRA]-3') and the same amplification conditions as those for M. pneumoniae amplification.
Participant 3. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers MYPN3 (5'-AGGCTTCAAGTGGACAAAGTGAC-3') and MYPN4 (5'-GATTGTYCCTGCTGGYCCAT-3') and detection probe MYPNF (5'-FAM-ACCACACCAAGTTCACGAGCGCTACG-TAMRA-3') with the following amplification conditions: 1 cycle of 2 min at 50°C followed by 1 cycle of 10 min at 95°C and 45 cycles of 15 s at 95°C and 60 s at 60°C.
C. pneumoniae DNA was also amplified by an in-house-developed real-time PCR using primers CHPN1 (5'-GAGATGGAGCAAATCCTAAAAGCTA-3') and CHPN2 (5'-AAATAGGTTGAGTCAACGACTTAAGGT-3') and detection probe CHPN-F (5'-FAM-TCAGCCATAACGCCGTGAATACGTTCTC-TAMRA-3') with the same amplification conditions used for M. pneumoniae.
Participant 4. In 2002, M. pneumoniae DNA was amplified by using the Minerva kit (Minerva Biolabs, Berlin, Germany) according to the instructions of the manufacturer. In 2004, an in-house-developed assay was applied using primers Mpneu02-FW (5'-GCCGGCAGTGGCAGTC-3') and Mpneu02-RV (5'-AGCCGCTTCGGTTCGG-3') and detection probe Mpneu02-MGB (5'-FAM-AACCACGTATGATCCC-nonfluorescent quencher [NFQ]-3') with the following amplification conditions: 1 cycle of 2 min at 50°C, 1 cycle of 10 min at 95°C, and 45 cycles of 15 s at 95°C and 60 s at 60°C.
For the detection of C. pneumoniae, primers CHLpnFW (5'-TGGCTAGGCCATTGAGAGTGA-3') and CHLpnRV (5'-GTTATGGATGGAGGGACTACTTTTG-3') and detection probe CHLpnMGB (5'-FAM-CTCAGCGCTTGCC-NFQ-3') were used. Amplification conditions were 1 cycle of 1 min at 95°C followed by 45 cycles of 15 s at 95°C and 60 s at 60°C.
Participant 5. In 2002, M. pneumoniae DNA was amplified by an in-house-developed PCR using primers OJPU1 (5'-GCCACCCTCGGGGGCAGTCAG-3') and OJPU3 (5'-GAGTCGGGATTCCCCGCGGAGG-3') (12). The following amplification conditions were used: 1 cycle of 3 min at 90°C, followed by 39 cycles of 1 min at 90°C, 2 min at 67°C, and 2 min at 72°C, and 1 final cycle of 10 min at 72°C.
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using slightly modified major outer membrane protein (MOMP) VD2 primers of Tondella et al. (26), FPCPN (5'-CTCGTTGGTTTATTCGGAGTT-3') and RPCPN (5'-CCAAGAGAAAGAGGTGTCTGTG-3'), and the slightly modified MOMP VD2 detection probe described by Tondella et al. (26), CPN (5'-FAM-AAGGTACTACTGTAAATGCAAATGAACTACCAAACGTTTCTTTAAGTAACGGAG-TAMRA-3').
In 2004, M. pneumoniae DNA was amplified by an own in-house-developed real-time PCR using primers M823F (5'-AGCGAACCGAGAGTGGTCAA-3') and M973R (5'-GATTGGCCAGATCCAGATGTG-3') and the detection probe (5'-FAM-CTCCAGGGCGCTGAGGCCACT-TAMRA-3'). Amplification conditions were 1 cycle of 2 min at 50°C and 1 cycle of 10 min at 95°C, followed by 50 cycles of 15 s at 95°C and 60 s at 60°C each.
C. pneumoniae DNA was amplified as described by Welti et al. (32).
Participant 6. In 2002 and 2004, M. pneumoniae DNA was amplified as described previously by Ieven et al. (12) and Ursi et al. (30), respectively. For C. pneumoniae DNA amplification, PCR was applied in 2002 and 2004 as described by Ursi et al. (29) and Hoymans et al. (11), respectively, using for the latter the MOMP VD4 primers and one detection probe described by Tondella et al. (26). Amplification conditions were the same as those for M. pneumoniae.
Participant 7. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using reverse primer 5'-CCAGGGCACATAATCCAACAC-3' and forward primer 5'-AAGGAACAAACTGATCCCACTTCT-3' and detection probe 5'-FAM-TCTCCACCGGGTTCAACCTTGTGG-NFQ-3', with the following amplification conditions 1 cycle of 2 min at 50°C, 1 cycle of 10 min at 95°C, and 40 cycles of 15 s at 95°C and 60 s at 60°C.
Participant 8. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers MPF (5'-CCAACCTCCATGTAGCTGATAGCT-3') and MPR (5'-TATCGCCAGGTAAAAACTCCTTCT-3') and the detection probe (5'-FAM-ATCCTTGTTGTAAGGCTTGTAATCG-TAMRA-3'). Amplification conditions were 1 cycle of 2 min at 50°C and 1 cycle of 10 min at 95°C, followed by 45 cycles of 15 s at 95°C and 60 s at 60°C.
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers CPF (5'-AAGGGCTATAAAGGCGTTGCT-3') and CPR (5'-TGGTCGCAGACTTTGTTCCA-3') and detection probe (5'-FAM-TCCCCTTGCCAACAGACGCTGG-TAMRA-3'). The amplification conditions used were the same as those for M. pneumoniae.
Participant 9. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers FPMPN (5'-TCTGGCGTGGATCTCTCCC-3') and RPMPN (5'-GACACTCGTGCTTGGTAACTGC-3'). Detection was done by using SYBR green. Amplification conditions were 1 cycle at 95°C for 10 min followed by 50 cycles each of 10 s at 95°C and 20 s at 65°C. In 2004, the same assay conditions were used, but real-time detection was done using probe MPN (5'-FAM-GAAGGAATGATAAGGCTTCAAGTGGACAAAGTG-TAMRA-3').
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using the slightly modified MOMP VD2 primers of Tondella et al. (26), FPCPN (5'-CTCGTTGGTTTATTCGGAGTT-3') and RPCPN (5'-CCAAGAGAAAGAGGTGTCTGTG-3'). Detection was done by using SYBR green. Amplification conditions were the same as those for M. pneumoniae. In 2004, the slightly modified MOMP VD2 detection probe described by Tondella et al. (26), CPN (5'-FAM-AAGGTACTACTGTAAATGCAAATGAAC-TAMRA-3'), was used.
Participant 10. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers MNP1FW (5'-CCAACCAAACAACAACGTTCA-3') and MNP1REV (5'-CCTTGACTGGAGGCCGTTAA-3') and detection probe MPN (5'-FAM-TCAATCCGAATAACGGTGACTTCTTACCACTG-TAMRA-3') in 2004. Amplification conditions were 1 cycle at 50°C for 2 min followed by 1 cycle at 95°C for 10 min and 40 cycles each 15 s at 95°C and 60 s at 60°C.
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers CPF1 (5'-GGACCTTACCTGGACTTGACATGT-3') and CPR1 (5'-CCATGCAGCACCTGTGTATCTG-3') and detection probe (5'-FAM-TGACCACTGTAGAAATACAGCTTTCCGCAAGG-TAMRA-3'). The amplification conditions used were the same as those for M. pneumoniae.
Participant 11. The M. pneumoniae and C. pneumoniae monoplex real-time PCR assays were done as described previously by Templeton et al. (24, 25). For the real-time multiplex PCR, primers, probes, and amplification conditions similar to those in the monoplex assays were used.
Participant 12. M. pneumoniae and C. pneumoniae DNA was detected as described by Abele-Horn et al. and Tong and Sillis (1, 27), respectively.
Participant 13. C. pneumoniae DNA was amplified by an in-house real-time PCR developed by participant number 14 using primers chlapneu 171F and chlapneu 250R and detection probe chlapneu 200T. Amplification conditions were 1 cycle at 50°C for 2 min and 1 cycle at 95°C for 10 min, followed by 45 cycles each of 15 s at 95°C and 60 s at 60°C.
Participant 14. M. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers MP-T1 (5'-ACACCAAGTT CGCGAGTGCTA-3') and MP-T2 (5'-CCGTCCTGCGTGGTTAAACTAT-3') and detection probe MP-S (5'-FAM-ATCCCGACTCGTTAAAGCAGGATAAGATT-TAMRA-3'). Amplification conditions were 1 cycle at 50°C for 2 min and 1 cycle at 95°C for 10 min, followed by 45 cycles each of 15 s at 95°C and 60 s at 60°C.
C. pneumoniae DNA was amplified by an in-house-developed real-time PCR using primers chlapneu 171F (5'-CGGCTAGAAATCAATTATAAGACTGAAG-3') and chlapneu 250R (5'-TGGCGAATGACACCATGATC-3') and detection probe chlapneu 200T (5'-FAM-AAATCTGCATCTCCCTCACGAATATGCTCA-TAMRA-3'). The amplification conditions used were the same as those for M. pneumoniae.
Participant 15. Conventional and monoplex real-time NASBA for the detection of C. pneumoniae RNA was done as described previously (23). MX-real-time NASBA was done as described previously (19).
Participant 16. M. pneumoniae and C. pneumoniae DNA was detected as described by Hardegger et al. and Welti et al. (10, 32), respectively.
Participant 17. M. pneumoniae and C. pneumoniae MOMP VD4 DNA was detected as described by Hardegger et al. and Tondella et al. (10, 26), respectively.
Participant 18. M. pneumoniae and C. pneumoniae MOMP VD2 DNA was detected by monoplex PCR using primers and probes described by Grondahl et al. and Tondella et al. (9, 26), respectively. Amplification conditions were 1 cycle of 10 min at 94°C followed by 40 cycles of 30 s at 94°C, 50°C, and 72°C, with a final cycle of 5 min at 72°C. The same primers, probes, and amplification conditions were used in the multiplex PCR.
Participant 19. The commercially available Pneumoplex assay (Prodesse, Waukesha, Wis.) was used for the detection of both M. pneumoniae and C. pneumoniae according to the instructions of the manufacturer.
Participant 20. M. pneumoniae DNA was amplified for participant 20 as described by Ursi et al. (30).
Internal controls. Most participants used a generic internal control such as the ones described by Hoymans et al., Ieven et al., and Tong and Sillis (11, 12, 27) or phocine herpes virus (25), except for participant 1, who measured for the presence of U1A mRNA in 2002, participant 2, who spiked another aliquot of the same sample with the respective organism, and the commercially available assays, which had their own internal controls.
Statistical analysis. The Fisher exact test or the chi-squared test was applied for the calculation of the significance between results during the preparation of the QC panels and for the performance of the different assays.
RESULTS
Preparation of the QC panels. To monitor the stability of the samples in the panels, aliquots kept at –80°C were examined by PCR and NASBA after one and two freezing-thawing cycles. All samples tested positive in 9 to 15 different runs for C. pneumoniae and M. pneumoniae, respectively, except for the samples with the lowest concentrations of M. pneumoniae, which produced positive results in 2 out of 15 PCR runs and 6 out of 15 NASBA runs (P = 0.10) in 2002.
When spiked BAL pools were tested by NASBA, an input of 50 CCU/100 μl, 500 CCU/100 μl, and 5000 CCU/100 μl of M. pneumoniae yielded 9 out of 9 positive results each time. When spiked BAL pools were tested by NASBA, an input of 4.9 IFU/100 μl, 49 IFU/100, and 490 IFU/100 μl of C. pneumoniae yielded 9 out of 9 positive results each time.
Performance of the laboratories. Some laboratories provided two amplification results obtained by two different procedures, such as mono- and multiplex reactions. Primers and probes used, as well as amplification conditions applied, were highly diverse.
Tables 1 to 4 show the different extraction, amplification, and detection procedures applied, as well as the target molecules used.
The 2002 M. pneumoniae QC. For M. pneumoniae, 13 datasets were obtained (Table 5). All procedures were in-house developed, except one participant, no. 4, who used the commercial kit from Minerva Biolabs GmbH (Berlin, Germany).
Eleven out of 13 assays applied one PCR and two NASBA; 9 were based on real-time detection, 3 used agarose gel electrophoresis, and 1 used electrochemiluminescent detection. The target most often used was the P1 gene in 9 out of 13 (69.2%) assays.
One data set, from participant 10, was excluded from the global analysis because of the uniformly negative results. All samples containing 5,000 CCU/100 μl were found positive. The samples containing 500 CCU/100 μl were scored positive in 20 out of 36 (55.5%) of the cases when suspended in saline and in 21 out of 36 (58.3%) of the cases when suspended in BAL fluid (P = 0.8). The lowest concentrations tested, 50 CCU/100 μl, scored positive in 10 out of 36 (27.8%) of the cases, a score which is not different from that obtained during the assessment of the panels before their distribution (P = 0.3).
Participant 2 did not include any negative controls and scored 3 out of 8 (37.5%) false-positive results. False-positive results were reported by a second laboratory (no. 5), and inhibition of the reaction in three instances was reported by two laboratories. In 7 out of 13 tests (53.9%), an internal control was used to monitor inhibition of the reaction in each sample. Eight (66.7%) participants reported the use of dUTP-uracyl-N-glycosylase (dUTP-UDG) to avoid false-positive results due to carryover, but nevertheless participants 2 and 5 using dUTP-UDG produced false-positive results.
The 2004 M. pneumoniae QC. The 2004 M. pneumoniae QC resulted in 18 datasets; 11 sets were obtained from 10 laboratories that participated in the 2002 QC (Table 6).
All participants used a commercially available nucleic acid extraction kit. Agarose gel electrophoresis was no longer used by any of the participants. In 16 out of 18 assays PCR was used, NASBA was used twice. Real-time detection was applied in 15 out of 18 tests, and an enzyme immunoassay (EIA) was applied in 3 out of 18 tests. The target most often used was the P1 gene (66.7%).
The samples containing 5,000 CCU/100 μl scored positive in 51 out of 54 (94.1%) of the cases, and those containing 500 CCU/100 μl were scored positive in 49 out of 54 (90.2%) of the cases. Four participants applied a multiplex assay. In three instances, participants 1, 11, and 18, an in-house-developed multiplex assay was performed in parallel with a monoplex real-time reaction; in one of these the multiplex reaction was less sensitive. The commercially available multiplex assay, the Pneumoplex assay from Prodesse (Waukesha, Wis.) (used by participant 19), was also less sensitive.
An internal control to monitor inhibition of the reaction in each sample was used in 15 out of 18 (83.3%) of the tests. Nine (60.0%) participants reported the use of dUTP-UDG to avoid false-positive results; none were recorded. One participant did not include negative controls.
The 2002 C. pneumoniae QC. The C. pneumoniae panels resulted in 16 datasets, of which 10 were delivered within the requested period of 4 weeks (Table 7). Twelve out of 16 amplification assays applied a PCR, 4 used a NASBA, 13 datasets used real-time detection, 2 participants used agarose gel electrophoresis, and 1 used electrochemiluminescent detection. Targets most often used were the 16S rRNA (NASBA), a cloned PstI fragment, the MOMP gene, or a 16S rRNA gene fragment in four (25%), four (25%), three (18.8%), and three (18.8%) of the assays, respectively. The other targets were only used once and were the PmP4 gene and a cytadhesin gene.
No false-positive results were recorded among the 124 samples. Samples containing 490 IFU/100 μl scored positive in 44 out of 45 (97.8%) instances, samples containing 49 IFU/100 μl suspended in saline scored positive in 46 out of 48 (95.8%) cases, and samples suspended in BAL fluid scored positive in 39 out of 45 (86.7%) cases (P = 0.1). The lowest concentration tested, 4.9 IFU/100 μl, scored positive in 32 out of 48 (66.7%) of the cases.
The multiplex NASBA, participant 15b, targeting the M. pneumoniae, C. pneumoniae, and Legionella pneumophila 16S rRNA, was significantly less sensitive than the monoplex NASBA of participant 15; both were in-house developed in the same laboratory.
Six out of 16 tests (37.5%) used an internal control to monitor inhibition of the reaction in each sample. Nine (64.3%) participants reported the use of dUTP-UDG.
The 2004 C. pneumoniae QC. The C. pneumoniae panels resulted in 18 datasets (Table 8). Thirteen sets were obtained from 12 laboratories that participated in the 2002 QC.
Sixteen assays applied a PCR, and 2 applied a NASBA; 15 datasets used real-time detection, and in 3 assays an EIA was used. Targets most often used were the MOMP gene, the cloned PstI fragment, and the 16S rRNA gene in 6 out of 18 (33.3%), 5 out of 18 (27.8%), and in 3 out of 18 (16.7%) assays, respectively.
Among 72 C. pneumoniae-negative samples, 4 were reported positive by one participant who did not use dUTP-UDG. Samples containing 490 IFU/100 μl were scored positive for 17 out of 18 (94.4%) of the cases, samples containing 49 IFU/100 μl scored positive in 14 out of 18 (83.3%) cases, and those containing 4.9 IFU/100 μl scored positive in 15 out of 18 (83.3%) of the cases. The lowest concentration tested, 0.49 IFU/100 μl, scored positive in 18 out of 54 (33.3%) of the cases.
Three participants submitted results of a multiplex NAAT: a real-time in-house developed NASBA (participant 1), an in-house-developed PCR (participant 18), and the commercially available Pneumoplex from Prodesse. The multiplex PCR was less sensitive than the monoplex counterpart. Laboratory 19, applying the Pneumoplex, did not detect a single C. pneumoniae-positive sample.
In 15 out of 18 tests (83.3%), an internal control was used to monitor inhibition of the reaction in each sample. Nine (56.3%) participants reported the use of dUTP-UDG to avoid false-positive results. One participant did not include negative controls.
Samples prepared with BAL pool 1 (clear), pool 2 (more viscous), and pool 3 (blood stained) spiked with 0.49 IFU/100 μl scored positive in 4 out of 17, 8 out of 17, and 4 out of 17 cases, respectively. The results of participant 8 were not included in this calculation due to the high number of false-positive results.
Generally, participants targeting the MOMP gene (participants 6, 8, 9, 17, and 18) found fewer samples to be C. pneumoniae positive than participants targeting the 16S rRNA, 16S rRNA gene, or PstI fragment. Participants 13 and 14 applied the same in-house real-time assay and obtained identical results.
DISCUSSION
QC for the NAATs for M. pneumoniae and C. pneumoniae are particularly important, because in the clinical laboratory there is no practical alternative for the detection of both agents. The two QCs reported here illustrate the technical evolution between 2002 and 2004: disappearance of gel electrophoresis for the detection of the amplicons, a substantial increase in monitoring of inhibition of the amplification reaction, and an increased use of positive and negative samples and of multiplex amplification reactions. The use of dUTP-UDG remained constant. Inclusion of positive and negative controls did not differ between the two QCs.
Gradually, laboratories performing NAATs for the detection of both M. pneumoniae and C. pneumoniae use the same amplification conditions for both assays. The advantage is that in the same run different organisms can be targeted, decreasing the total turnaround time.
In 2002 two participants obtained false-positive results in the M. pneumoniae panel (one of these performing routinely up to 50 tests on a monthly basis). In the 2004 M. pneumoniae panel, both participants reported all samples correctly. No false-positive C. pneumoniae results were recorded in 2002, and 11 out of 15 participants produced correct results. In 2004, one participant produced false-positive results for all four C. pneumoniae-negative samples. Other PCR quality assessment studies have also recorded false-positive results (8, 20, 22, 31, 33). Perhaps this problem will not be eliminated until sample processing can be automated.
In this quality control study, the major problem was the occurrence of false negatives, especially when testing the M. pneumoniae 2002 proficiency panel. The results obtained by the participants in 2002 for the samples containing 50 CCU/100 μl of M. pneumoniae are in line with those of the reference laboratory (P = 0.13). This low concentration was not included in 2004. There were no differences between the suspensions prepared in saline (20 positive results) and those prepared in BAL (21 positive results). Participant 10, which failed to detect any M. pneumoniae-positive samples, did not participate in the M. pneumoniae 2004 panel and also failed to detect small numbers of C. pneumoniae in 2002 and 2004.
The lowest concentration of M. pneumoniae is clearly at the limit of detectability, therefore only the 5,000 and 500 CCU/100 μl samples were taken into consideration for the global analysis. Thus, 2 out of 12 participants produced correct results, 2 out of 12 produced false-positive results, and 9 participants produced between 0 of 9 and 8 of 9 correct results. There was considerable improvement in 2004, when 15 out of 18 participants produced 100% M. pneumoniae correct results without any false positives.
Suspensions of both M. pneumoniae and C. pneumoniae in saline and in BAL were included in the 2002 panels to reveal possible inhibitors: two samples in saline were found in one laboratory to inhibit the reaction, and one sample in BAL was found in a second laboratory. Both laboratories used an extraction kit from Roche. No inhibitors were recorded in 2004, although one of the BAL pools was blood stained. Both laboratories reporting inhibition in the 2002 panel used the same extraction procedure for the 2004 M. pneumoniae panel. Participant 8 did not change the amplification and detection procedures. For the 2004 panel, participant 4 replaced the Minerva Biolabs kit with an in-house-developed real-time PCR.
Two out of three participants (18b and 19) obtaining false-negative results in the 2004 M. pneumoniae QC applied a multiplex NAAT, one of them being a commercial kit (Pneumoplex; Prodesse). Two other participants using an in-house-developed multiplex NAAT obtained correct results.
Four participants failed to detect C. pneumoniae in the sample containing 49 IFU/100 μl in the 2004 QC, two of them applied a multiplex NAAT, with one being a commercial kit (Pneumoplex; Prodesse).
In the 2004 QC tests, multiplex NAAT was applied. PCR was used by participants 11, 18, and 19 (Table 6 and Table 8), and NASBA was used by participants 1 and 15. The multiplex PCRs scored a smaller number of samples positive than most of the monoplex tests. One multiplex PCR was the commercially available Pneumoplex (Prodesse). Although the limit of detection of this assay was reported to be 5 CCU/ml for M. pneumoniae and 0.01 50% tissue culture infective dose/ml for C. pneumoniae and 10 copies of recombinant DNA for each organism (13), the test did not perform well in this evaluation. Participant 19 tested a second panel of each organism with similar results. The assay was performed correctly. The manufacturer was contacted and is aware of the sensitivity problems of the Pneumoplex assay. They intend to improve the sensitivity of the test.
Multiplex assays are somewhat less sensitive than monoplex assays, but until the number of organisms present in clinical specimens of diseased individuals is known, it is impossible to state whether the degree of sensitivity attained is acceptable.
From previous investigations, we have learned that the sensitivity of molecular diagnostics using respiratory samples may be compromised by the presence of inhibitory factors in the samples (2, 5, 8, 14, 28, 29). Therefore, three different BAL pools were used in the 2004 QC panels. There was no significant difference in the positivity rates between the three different BAL pools used to prepare the suspensions, although the suspension in the blood-stained BAL had the lowest number of positive results. Unexpectedly, the viscous suspension scored a higher number of positive results than the clear BAL suspension.
There was considerable variation between the different PCR protocols applied by the participants. Evaluation of the various PCR protocols showed no apparent association between their performance and the particular variables of the PCR method used, except for the C. pneumoniae MOMP amplification protocols. Participants 6, 17, and 18 used the protocol described by Tondella et al. (26) and obtained lower positivity scores than participants using a different protocol and targeting a different gene for the detection of C. pneumoniae. Participant 9 had a slightly better positivity score using slightly modified MOMP VD2 primers and detection probe. However, the results of such a comparison must be interpreted with caution due to the relatively small number of samples, the small number of participating laboratories, and the high diversity of the methods used.
The discrepant results from two successive QC exercises observed among different laboratories, some of which lacked experience in NAAT, illustrate the importance of training of personnel and the use of negative and positive controls in the preparative and amplification phases. This is illustrated by the three laboratories that reported false-positive results; one of them did not use negative controls in the preparation, amplification, and detection procedures.
The different performance characteristics of the amplification-based assays used may explain discrepant findings from published studies that used NAATs to determine M. pneumoniae and C. pneumoniae prevalence in patient populations. Sensitivity and specificity issues should be addressed before publishing clinical and epidemiological studies of M. pneumoniae and C. pneumoniae infections based on the detection of bacterial DNA and RNA in clinical specimens by NAAT.
This study also underlines the need for reference reagents and standard operating procedures to enable experienced technicians to perform quality control assessment of nucleic acid amplification methods and thus perform reliable diagnostic molecular amplification techniques on a routine basis.
ACKNOWLEDGMENTS
The organization of the QC exercises for the Belgian Centers for Molecular Diagnostics was supported by the public health service, RIZIV-INAMI.
REFERENCES
Abele-Horn, M., U. Busch, H. Nitzschiko, et al. 1998. Molecular approaches to diagnosis of pulmonary diseases due to Mycoplasma pneumoniae. J. Clin. Microbiol. 36:548-551.
Apfalter, P., F. Blasi, J. Boman, C. A. Gaydos, M. Kundi, M. Maass, A. Makristathis, A. Meijer, R. Nadrchal, K. Persson, M. L. Rotter, C. Y. W. Tong, G. Stanek, and A. M. Hirschl. 2001. Multicenter comparison trial of DNA extraction methods and PCR assays for detection of Chlamydia pneumoniae in endarterectomy specimens. J. Clin. Microbiol. 39:519-524.
Apfalter, P., O. Assadian, F. Blasi, J. Boman, C. A. Gaydos, M. Kundy, A. Makristathis, M. Nehr, M. L. Rotter, and A. M. Hirschl. 2002. Reliability of nested PCR for detection of Chlamydia pneumoniae DNA in atheromas: results from a multicenter study applying standardized protocols. J. Clin. Microbiol. 40:4428-4434.
Bernet, C., M. Garret, B. de Barbeyrac, C. BeBear, and J. Bonnet. 1989. Detection of Mycoplasma pneumoniae by using the polymerase chain reaction. J. Clin. Microbiol. 27:2492-2496.
Birkebaek, N. H., J. S. Jensen, T. Seefeldt, J. Degn, B. Huniche, P. L. Andersen, and L. Ostergaard. 2000. Chlamydia pneumoniae infection in adults with chronic cough compared with healthy blood donors. Eur. Respir. J. 16:108-111.
Boman, J., C. A. Gaydos, and T. C. Quinn. 1999. Molecular diagnosis of Chlamydia pneumoniae infection. J. Clin. Microbiol. 37:3791-3799.
Campbell, L. A., M. Perez Melgosa, D. J. Hamilton, C. C. Kuo, and J. T. Grayston. 1992. Detection of Chlamydia pneumoniae by polymerase chain reaction. J. Clin. Microbiol. 30:434-439.
Chernesky, M., M. Smieja, J. Schachter, J. Summersgill, L. Schindler, N. Solomon, K. Campbell, L. A. Campbell, A. Cappuccio, C. Gaydos, S. Chong, J. Moncada, J. Phillips, D. Jang, B. J. Wood, A. Perich, M. Hammerschlag, M. Cerney, and J. Mahony. 2002. Comparison of an industry derived LCx Chlamydia pneumoniae PCR research kit to in-house assays performed in five laboratories. J. Clin. Microbiol. 40:2357-2362.
Grndahl, B., W. Puppe, A. Hoppe, I. Kühne, J. A. I. Weigl, and H.-J. Schmitt. 1999. Rapid identification of nine microorganisms causing acute respiratory tract infections by single tube multiplex reverse transcription PCR: feasibility study. J. Clin. Microbiol. 37:1-7.
Hardegger, D., D. Nadal, W. Bossart, M. Altwegg, and F. Dutly. 2000. Rapid detection of M. pneumoniae in clinical samples by real-time PCR. J. Microbiol. Methods 41:45-51.
Hoymans, V. Y., J. M. Bosmans, D. Ursi, W. Martinet, F. Wuyts, E. Van Marck, M. Altwegg, and C. J. Vrints. 2004. Immunohistostaining assays for detection of Chlamydia pneumoniae in atherosclerotic arteries indicate cross-reactions with nonchlamydial plaque constituents. J. Clin. Microbiol. 42:3219-3224.
Ieven, M., D. Ursi, H. Van Bever, W. Quint, H. G. M. Niesters, and H. Goossens. 1996. Detection of Mycoplasma pneumoniae by two polymerase chain reactions and role of M. pneumoniae in acute respiratory tract infections in pediatric patients. J. Infect. Dis. 173:1445-14452.
Khana, M., J. Fan, K. Pehler-Harrington, C. Waters, P. Douglass, J. Stallock, S. Kehl, and K. J. Henrickson. 2005. The Pneumoplex assay, a multiplex PCR-enzyme hybridization assay that allows simultaneous detection of five organisms, Mycoplasma pneumoniae, Chlamydia (Chlamydophila) pneumoniae, Legionella pneumophila, Legionella micdadei, and Bordetella pertussis, and its real-time counterpart. J. Clin. Microbiol. 43:565-571.
Loens, K., D. Ursi, M. Ieven, P. van Aarle, P. Sillekens, P. Oudshoorn, and H. Goossens. 2002. Detection of Mycoplasma pneumoniae in spiked clinical samples by nucleic acid sequence-based amplification. J. Clin. Microbiol. 40:1339-1345.
Loens, K., M. Ieven, D. Ursi, H. Foolen, P. Sillekens, and H. Goossens. 2003. Application of NucliSens basic kit for the detection of Mycoplasma pneumoniae in respiratory specimens. J. Microbiol. Methods 54:127-130.
Loens, K., M. Ieven, D. Ursi, T. Beck, M. Overdijk, P. Sillekens, and H. Goossens. 2003. Detection of Mycoplasma pneumoniae by real-time nucleic acid sequence-based amplification. J. Clin. Microbiol. 41:4448-4450.
Loens, K., D. Ursi, H. Goossens, and M. Ieven. 2003. Minireview: molecular diagnosis of Mycoplasma pneumoniae infections. J. Clin. Microbiol. 41:4915-4923.
Loens, K., T. Beck, H. Goossens, D. Ursi, M. Overdijk, P. Sillekens, and M. Ieven. 2006. Development of conventional and real-time NASBA for the detection of Chlamydophila pneumoniae in respiratory specimens. J. Clin. Microbiol., submitted for publication.
Loens, K., T. Beck, H. Wouters, D. Ursi, M. Overdijk, P. Sillekens, H. Goossens, and M. Ieven. 2004. Comparison of real-time mono and multiplex NASBA for the detection of M. pneumoniae, C. pneumoniae and L. pneumophila in respiratory specimens. 14th European Congress of Clinical Microbiology and Infectious Diseases, European Society for Clinical Microbiology and Infectious Diseases, 01/05-04/05/2004, Prague, Tsjech Republic. Clin. Microbiol. Infect. 10(Suppl. 3):7.
Muir, P., A. Ras, P. E. Klapper, G. M. Cleator, K. Korn, C. Aepinus, A. Fomsgaard, P. Palmer, A. Samuelsson, A. Tenorio, B. Weissbrich, and A. M. Van Loon. 1999. Multicenter quality assessment of PCR methods for detection of enteroviruses. J. Clin. Microbiol. 37:1409-1414.
Ramirez, J. A., and the Chlamydia pneumoniae/Atherosclerosis Study Group. 1996. Isolation of Chlamydia pneumoniae from the coronary artery of a patient with coronary atherosclerosis. Ann. Intern. Med. 125:979-982.
Schirm, J., A. M. Van Loon, E. Valentine-Thon, P. E. Klapper, J. Reid, and G. M. Cleator. 2002. External quality assessment program for qualitative and quantitative detection of hepatitis C virus RNA in diagnostic virology. J. Clin. Microbiol. 40:2973-2980.
Sillekens, P., M. Overdijk, I. Roosmalen, K. Loens, M. Ieven, P. Haima, and M. Jacobs. 2002. Molecular diagnosis of Chlamydia pneumoniae infections by using NASBA with real time detection. 12th European Congress of Clinical Microbiology and Infectious Diseases, European Society for Clinical Microbiology and Infectious Diseases, 24/04-27/04/2002, Milan, Italy. Clin. Microbiol. Infect. Suppl. P730.
Templeton, K. E., S. A. Scheltinga, A. W. Graffelman, P. H. van den Broek, and E. Claas. 2003. Comparison and evaluation of real-time PCR, real-time nucleic acid sequence-based amplification, conventional PCR, and serology for diagnosis of Mycoplasma pneumoniae. J. Clin. Microbiol. 41:4366-4371.
Templeton, K. E., S. A. Scheltingha, W. C. J. F. M. Van den Eeden, A. W. Graffelman, P. H. van den Broek, and E. Claas. 2005. Improved diagnosis of etiology of community-acquired pneumonia using real-time PCR. Clin. Infect. Dis. 41:345-351.
Tondella, M. L. L., D. F. Talkington, B. P. Holloway, S. F. Dowell, K. Cowley, M. Soriano-Gabarro, M. S. Elkind, and B. S. Fields. 2002. Development and evaluation of real-time PCR-based fluorescence assays for detection of Chlamydia pneumoniae. J. Clin. Microbiol. 40:575-583.
Tong, C. Y. W., and M. Sillis. 1993. Detection of Chlamydia pneumoniae and Chlamydia psittaci in sputum samples by PCR. J. Clin. Pathol. 46:313-317.
Ursi, J. P., D. Ursi, M. Ieven, and S. R. Pattyn. 1992. Utility of an internal control for the polymerase chain reaction. APMIS 100:635-639.
Ursi, D., M. Ieven, H. P. Van Bever, and H. Goossens. 1998. Construction of an internal control for the detection of Chlamydia pneumoniae by PCR. Mol. Cell. Probes 12:235-238.
Ursi, D., K. Dirven, K. Loens, M. Ieven, and H. Goossens. 2003. Detection of Mycoplasma pneumoniae in respiratory samples by real-time PCR using an inhibition control. J. Microbiol. Methods 55:149-153.
Valentine-Thon, E., A. M. Van Loon, J. Schirm, J. Reid, P. E. Klapper, and G. M. Cleator. 2001. European proficiency testing program for molecular detection and quantitation of hepatitis B virus DNA. J. Clin. Microbiol. 39:4407-4412.
Welti, M., K. Jaton, M. Altwegg, R. Sahli, A. Wenger, and J. Bille. 2003. Development of a multiplex real-time quantitative PCR assay to detect Chlamydia pneumoniae, Legionella pneumophila, and Mycoplasma pneumoniae in respiratory tract secretions. Diagn. Microbiol. Infect. Dis. 45:85-95.
Zaaijer, H. L., H. T. M. Cuypers, H. W. Reesink, I. N. Winkel, G. Gerken, and P. N. Lelie. 1993. Reliability of polymerase chain reaction for detection of hepatitis C virus. Lancet 341:722-724.(K. Loens, T. Beck, D. Urs)