Comparison and Evaluation of Real-Time PCR, Real-Time Nucleic Acid Sequence-Based Amplification, Conventional PCR, and Serology for Diagnosis of Mycoplasma pneumoniae

Abstract

Mycoplasma pneumoniae is a common cause of community-acquired pneumonia and lower-respiratory-tract infections. Diagnosis has traditionally been obtained by serological diagnosis, but increasingly, molecular techniques have been applied. However, the number of studies actually comparing these assays is limited. The development of a novel duplex real-time PCR assay for detection of M. pneumoniae in the presence of an internal control real-time PCR is described. In addition, real-time nucleic acid sequence-based amplification (NASBA) on an iCycler apparatus is evaluated. These assays were compared to serology and a conventional PCR assay for 106 clinical samples from patients with lower-respiratory-tract infection. Of the 106 samples, 12 (11.3%) were positive by all the molecular methods whereas serology with acute sample and convalescent samples detected 6 (5.6%) and 9 (8.5%), respectively. Clinical symptoms of the patients with Mycoplasma-positive results were compared to those of the other patients with lower-respiratory-tract infections, and it was found that the results for mean lower age numbers as well as the presence of chills, increased erythrocyte sedimentation rate, and raised C-reactive protein levels showed significant differences. Molecular methods are superior for diagnosis of M. pneumoniae, providing more timely diagnosis. In addition, using real-time methods involves less hands-on time and affords the ability to monitor the reaction in the same tube.

Mycoplasma pneumoniae is reported to cause 6 to 20% of community-acquired pneumonia (CAP) and lower-respiratory-tract infections (LRTI) in older children and adults (3, 11). The incidence of M. pneumoniae in adults with CAP and LRTI ranges from 1 to 30%, depending on the population studied and the diagnostic test used (19, 22, 27). M. pneumoniae is difficult to grow in cultures; therefore, clinical diagnosis relies mainly on serology and, in recent years, on molecular techniques (9). It is important to establish M. pneumoniae as the pathogen by laboratory diagnosis, as the clinical presentation is not significantly different from that seen with other pathogens causing CAP (29). Since the organism is not sensitive to β-lactam antibiotics, which are often used for empirical treatment of LRTI, a rapid diagnostic method is required for the prescription of effective antibiotics (7).

Serological methods lack sufficient sensitivity in the acute phase of the disease. An accurate diagnosis with convalescent-phase samples is often made many days after the onset of disease (28). Sensitivity and specificity values are between 55 and 100%, depending on the serological method used and the patient population tested (1, 4, 8, 10, 15, 18, 25). PCR has been shown to offer the potential of increased sensitivity and rapidity compared to other diagnostic tests. For the diagnosis of M. pneumoniae infections, therefore, nucleic acid amplification techniques have been introduced in many diagnostic laboratories as a valuable test (9). Two targets are primarily used for M. pneumoniae amplification assays. For isothermal nucleic acid sequence-based amplification (NASBA), 16S rRNA has obviously been used as a target, and for PCR, the 16S rRNA genes or the P1 adhesion gene have been described. PCR assays targeting the P1 gene have been reported to be more sensitive than those targeted at the 16S rRNA (16). Sensitivities range from 65 to 90% and specificities range from 90 to 100%, depending on the PCR format and the reference system used (2, 5, 6, 7, 13, 14, 21, 23, 26, 30). Confirmation of PCR results by Southern blot hybridization, seminested application, or enzyme immunoassay detection is important for reliable amplification of PCR in the diagnostic laboratory. Real-time PCR methods using molecular-beacon detection allow single-tube PCR amplification and detection with no need for post-PCR analysis.

In various studies, PCR has been compared to serology (6, 7, 13, 14, 23, 30) and NASBA has been compared to conventional PCR (20) for the diagnosis of M. pneumonia infection. No studies have compared the different diagnostic values of real-time PCR, real-time NASBA, conventional PCR, and serology in an adult population.

In this study, a real-time PCR assay for M. pneumoniae was designed using molecular beacons as probes. In a single tube, an internal control real-time PCR was duplexed with M. pneumoniae real-time PCR for monitoring DNA extraction in the clinical samples and to determine the presence of inhibitors. In addition, real-time NASBA was developed for M. pneumoniae for use with an iCycler real-time detection system. The diagnostic significance has been evaluated on samples collected in a 32-month prospective study among adults with LRTI. Each sample was analyzed by real-time PCR, real-time NASBA using a NucliSens Basic kit, and conventional PCR. Acute- and convalescent-phase sera were also analyzed by a complement fixation test (CFT) and a Serodia particle precipitation assay (PPA) to enable a comparison between molecular methods and serology for the diagnosis of M. pneumoniae infection.

MATERIALS AND METHODS

Bacterial strains.

Data for the bacterial strains used to test the specificity of the real-time PCR are presented in Table 1. A suspension of colonies of all bacterial isolates was made in 0.9% NaCl prior to nucleic acid extraction. M. pneumoniae strain P1 1428 (ATCC 29085) was quantitated at the University of Antwerp (Antwerp, Belgium). The titer was expressed in color-changing units (CCU) per milliliter, with one CCU corresponding to 10 to 100 cells (20). This strain was introduced in a proficiency panel for quality control of an M. pneumoniae PCR used in the first national external quality assessment for Belgian laboratories (see report [First External Quality Assessment in Belgian Laboratories performing Molecular Microbiology] [http://www.uia.ac.be/cmd/tests/tests.html]).

TABLE 1.

Bacterial species and strains

Species	Strain or type	Sourcea
Mycoplasma pneumoniae type 1	ATCC 29085 (PI 1428)	ATCC
Mycoplasma pneumoniae type 2	ATCC 15492 (MAC)	ATCC
Mycoplasma fermentans	NC10117	NCTC
Mycoplasma hominis	NC10111	NCTC
Mycoplasma genitalium	ATCC 33530 (G-37)	ATCC
Mycoplasma orale	NC10112	NCTC
Mycoplasma buccale	NC10136	NCTC
Mycoplasma salivarium	NC10113	NCTC
Ureaplasma urealyticum	Clinical isolate	UZA
Legionella pneumophila	ATCC 33152	ATCC
Chlamydia pneumoniae	TW-183
Moraxella catarrhalis	ATCC 25238	ATCC
Haemophilus influenzae	ATCC 43065	ATCC
Streptococcus pneumoniae	ATCC 49150	ATCC
Streptococcus pyogenes	ATCC 12344	ATCC
Enterococcus faecalis	ATCC 12984	ATCC
Staphylococcus aureus	ATCC 12600	ATCC
Klebsiella pneumoniae	ATCC 13883	ATCC
Escherichia coli	ATCC 11775	ATCC
Neisseria meningitidis	ATCC 13090	ATCC
Pseudomonas aeruginosa	ATCC 10145	ATCC
Enterobacter aerogenes	ATCC 13048	ATCC

^aATCC, American Type Culture Collection; NCTC, National Collection of Type Cultures (Central Public Health Laboratory, London, England); UZA, Universitair Zeikenhuis, Antwerp, Belgium.

Patients and samples.

From November 1998 through June 2001, 145 adults with signs of LRTI who consulted a general practitioner in the Leiden, The Netherlands, area were seen. Provided informed consent was given, patients attending the general practitioner practice as well as those seen at home were included in the study. Patients were seen a median of 7 days (range, 1 to 28 days) after the presentation of symptoms. Entry criteria were (i) age over 18 years, (ii) the presence of an pulmonary auscultation abnormality(ies), and (iii) at least two of the three following criteria: fever (i.e., temperature of >38°C) or fever in the previous 48 h, dyspnea or cough (productive or nonproductive), and tachypnea, malaise, or confusion. Patients with terminal illness or other conditions that would preclude completion were excluded. Clinical data, including a chest radiograph, were collected on each patient.

A cotton-tipped-swab sample was taken by the investigator either at the clinic or at home from each patient at the first visit, processed in the laboratory on the day of collection, and stored at −70°C prior to nucleic acid isolation for the molecular assays. Blood samples were collected from each patient at the first visit, and one more blood sample was taken after 10 to 14 days. The sera were stored at −20°C and used for serological testing. Sputum was collected if produced by the patient.

Serology for M. pneumoniae.

Using a Serodia-MycoII kit (Fujirebio, Tokyo, Japan), a microparticle agglutination test for detection of M. pneumoniae antibodies was performed. The assay detects antibodies by using a gelatin particle agglutination assay, with undefined membrane components of M. pneumoniae Mac as an antigen. The assay was performed according to the manufacturer's instructions, and an immunoglobulin M antibody titer of ≥1:320 was regarded as a positive result.

Paired sera were analyzed by CFT using a commercially available antigen (Virion, Ruschlikon, Switzerland). Briefly, the assay was performed by making serial twofold dilutions of the paired sera. The sera were reacted with the Mycoplasma antigen and guinea pig whole complement (BioWhittaker, Walkersville, Md.). The antibody titers were obtained by assessing the end point of hemolysis of the red blood cells. A titer of ≥128 or a fourfold rise in titer was regarded as a positive result.

Nucleic acid isolation.

Nucleic acids from M. pneumoniae-positive material, clinical specimens, and other bacterial isolates were extracted with a QiaAmp DNA kit (Qiagen, Hilden, Germany). All samples were extracted according to the manufacturer's instructions, resulting in 200 μl of purified nucleic acids, which was stored at −20°C. Negative controls were included in each run. For the control, sterile distilled water was added instead of a specimen.

Conventional PCR.

PCR amplification was performed using primers described by Ieven et al. (16). Briefly, 10 μl of isolated DNA was amplified with P1-specific gene primers for 40 cycles. Cycling conditions were 3 min at 94°C followed by 40 cycles of 30 s at 94°C, 30 s at 65°C, and 45 s at 72°C followed by a 10-min hold at 72°C. The product was detected by enzymatic reaction with a probe specific to the P1 product; the probe was labeled with digoxigenin. Inhibition was determined by spiking samples with an M. pneumoniae control in a separate amplification.

Primers and probes for Mycoplasma real-time PCR.

Using criteria required for the design of molecular-beacon assays, primer and molecular-beacon sequences were selected from a P1 cytadhesin gene sequence of M. pneumoniae (accession no. AF286371). The PCR primers were designed using the primer 3 program (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi) to ensure the absence of secondary structures. The molecular beacon was designed using the Mfold Zuker program (http://www.bioinfo.rpi.edu/applications/mfold/). Additional criteria for a good molecular beacon included a melting temperature of 8°C over the melting temperature of the primers and a relatively short amplicon (<150 bp). The stem sequence was selected to have a melting temperature compatible with that of the molecular beacon. The beacon formed a stable structure (with no secondary structures) at 50 to 55°C (the proposed annealing temperature). A BLAST search was performed to check the specificity of the DNA sequences of the primers and probe. The fluorescent reporter on the 5′ end of the probe was 6-carboxy-fluorescein (FAM), and the quencher on the 3′ end was Dabcyl. Biolegio (Malden, The Netherlands) prepared the molecular beacons and primers. Selected primers and probes are shown in Table 2.

TABLE 2.

Primers and probes for PCR

Assay format	Target	Primer or probe sequencea	Primer or probe	Amplicon size
Real-time PCR	Mpn P1	ATTCCCGAACAAAATAATGA	Upstream primer	151
	Mpn P1	GTTTGACAAAGTCCGTGAAG	Downstream primer	151
	MpnP1	FAM-GCTGCCCCAAAGCCACCCTGATCACCCGGCAGC-Dabcyl	Molecular beacon
	PhHVgB	ATGCATTTAAAACCCTCAAA	Upstream primer	140
	PhHVgB	GCATCAACTTCTTCGACAAT	Downstream primer	140
	PhHVgB	Cy-5-GTCGCCCCTGGTTTTTATCGTACGGGAACAGGCGAC-BHQ2	Molecular beacon
NASBAb	Mpn16S	FAM-GCGCACCCTAATACATGCAAGTCGATCGAAAGTGCGC-Dabcyl	Molecular beacon

^aUnderlined nucleotide designations indicate the stem structure of the beacon.

^bPrimer data for NASBA have been published by Loens et al. (20).

Primers and probes for internal-control real-time PCR.

A real-time PCR assay for phocine herpes virus (PhHV) was used to monitor inhibition of the real-time PCR. The initial assay described by Niesters (24) used TaqMan probes; therefore, the assay was redesigned for the use of molecular beacons. Primer and probe sequences (Table 2) were selected from PhHV sequences (accession no. U92270). The PhHV assay design was performed under conditions that mimicked those of the Mycoplasma real-time PCR assay to facilitate the multiplexing of the two assays.

Real-time PCR.

Real-time PCR was performed in 50 μl of a reaction mixture consisting of 25 μl of platinum Supermix (Invitrogen), 3.5 mM MgCl₂, 0.4 μM concentrations of each Mycoplasma primer, 0.2 μM concentrations of each PhHV primer, a 0.34 μM concentration of the Mycoplasma molecular beacon, a 0.2 μM concentration of the PhHV molecular beacon, and 10 μl of the template. The PCR thermal profile consisted of an initial incubation of 2 min at 50°C and 2 min at 95°C followed by 50 cycles of 30 s at 95°C, 30 s at 50°C, and 30 s at 72°C. Amplification, detection, and data analysis were performed with an iCycler IQ real-time detection system (Bio-Rad, Veenendaal, The Netherlands). Each sample was spiked with 10³ copies of PhHV that were coextracted with the sample, and the assay was performed as a duplex PCR.

Inter- and intraassay variability.

DNA was extracted from M. pneumoniae ATCC 15492 (Mycobacterium avium complex) and stored in AE buffer (50 mM Na acetate [pH 5.3], 10 mM EDTA [pH 8.0]; Qiagen, Hilden, Germany). The DNA was diluted to a concentration equivalent to 50 CCU/100 μl and stored in small aliquots at −20°C. To determine inter- and intra-assay variation, an aliquot was thawed and run in quintuplicate in five consecutive runs of the multiplex real-time PCR assay.

NASBA.

NASBA is an isothermal amplification of RNA. An RNA polymerase binds a promoter site that had been attached to the target RNA by specific primers. The assay was performed as described by Loens et al. (20) but with the detection and analysis being performed on an iCycler IQ real-time detection system (Bio-Rad) using a molecular beacon as the probe. The molecular beacon and primers were designed in accordance with the characteristics of the 16S rRNA gene (16S rDNA) of M. pneumoniae. The fluorescent reporter in the beacon was FAM, and the beacon was quenched by Dabcyl. Primers were prepared by Eurogentec (Seraing, Belgium), and the molecular beacons were prepared by Biolegio.

The NASBA was performed using a NucliSens basic kit (BioMerieux, Boxtel, The Netherlands). Double-stranded DNA (including a T7 promoter site) was made by adding 5 μl of the template RNA to a 10-μl reaction mixture containing primers, RNase H, and reverse transcriptase. This mixture was heated for 2 min at 65°C and subsequently cooled to 41°C for 2 min. Thereafter, 5 μl of enzyme mixture containing T7 polymerase was added at 41°C. Final concentrations were 100 mM KCl and 0.2 μM for the molecular beacon and each of the primers. After the addition of the enzymes, the reaction volume was mixed well and placed on the iCycler for 90 min at 41°C, with readings performed every 45 s. All detection and data analysis were performed on an iCycler IQ real-time detection system.

Analysis of results.

All data and laboratory results were coded and entered into a database. A result was considered a true positive for M. pneumoniae when the conventional PCR described by Ieven et al. gave a positive result (16). The sensitivity and specificity were calculated for each assay against this reference. The chi-square test was used for analysis of the differences in the frequencies of detection between pairs of groups. The Student t test was used to compare the means from patients with and without M. pneumoniae. Differences with a P value of <0.05 were considered to be statistically significant.

RESULTS

Real-time PCR specificity and sensitivity.

The real-time PCR assay specifically amplified DNA from both M. pneumoniae type 1 and type 2 but not from any of the other Mycoplasma species or other respiratory bacteria listed in Table 1. Sensitivity was determined by analyzing dilutions of DNA extracted from an M. pneumoniae-positive ATCC strain. From tests of serial 10-fold dilutions of DNA in Tris buffer, the lowest level of detection was determined to be 5 CCU/100 μl. The results of the Belgian proficiency panel showed that of the 20 samples tested, 19 had the correct result with the real-time PCR assay. Only one low-positive sample did not give a correct result. The reproducibility of the assay was determined by testing samples containing 50 CCU/100 μl in quintuplicate in the multiplex assay to determine interassay and intra-assay variability. As determined from the threshold cycle (Ct) values obtained from five consecutive runs, the interrun variation was 0.7 for this standard sample. The mean of the intra-assay variability was 0.28 (range, 0.1 to 0.7).

Evaluation of PCR inhibition.

Application of PCR amplification for microbiological diagnosis of clinical specimens requires sufficient controls. Besides specific PCR controls, the DNA isolation process and potential inhibition should be monitored as well. The real-time PCR assay for M. pneumoniae was duplexed with a reaction to amplify a PhHV spike that was used as an internal control.

A dilution series of PhHV coamplified with the M. pneumoniae targets did not significantly affect the efficiency and sensitivity of the reaction. A fixed amount of PhHV dilution was added to the lysis buffer prior to nucleic acid extraction. The amount of this PhHV spike was chosen to give a positive result at a Ct value of 33 to 34. Table 3 shows that coamplification of PhHV resulted in M. pneumoniae amplification Ct values similar to those of amplification without spikes.

TABLE 3.

Ct values for M. pneumoniae dilution series after duplex real-time PCR amplification with and without the PhHV spike

CCU of M. pneumoniae/ml	Ct value
	Without spike	With spike
10,000	27.9	28.2
1,000	32.1	31.6
100	35.1	35.3
10	39.4	40.3

Clinical evaluation of M. pneumoniae real-time PCR.

During the 32-month period, 145 adult patients were seen and diagnosed as having a LRTI; X-ray results showed that 30 had an infiltrate. Of these adults, a complete set of samples was available for 106, enabling the performance of the serological assays and the three molecular assays. Out of the samples from 106 adult patients, those of 12 (11.3%) were positive by the real-time PCR. The conventional PCR and the NASBA, which targeted a different part of the genome (Table 4), confirmed these positive results. The molecular methods showed 100% agreement, and the sensitivity and specificity for the real-time PCR were 100% (Table 5). No inhibition was detected in the samples tested, as amplification of the PhHV spike was detected at a Ct value of 33 to 34.

TABLE 4.

Laboratory findings with positive results for each assay

Patient no.	Titer for:				Result of:
	Serodia sample 1	Serodia sample 2	CFT sample 1	CFT sample 2	PCR	Real-time PCR	NASBA
1	320	1,240	<8	64	+	+	+
2	80	>2,560	32	>128	+	+	+
3	80	80	8	128	+	+	+
4	320	>2,560	32	>128	+	+	+
5	320	640	32	>128	+	+	+
6	80	80	<8	64	+	+	+
7	80	80	<8	<8	+	+	+
8	320	640	<8	<8	+	+	+
9	>2,560	>2,560	>128	>128	+	+	+
10	80	>2,560	8	>128	+	+	+
11	80	80	8	128	+	+	+
12	320	640	<8	<8	+	+	+

^aFor Serodia assays, a titer of ≥320 represented a positive result; for CFT, a titer of >128 or fourfold rise in titer represented a positive result.

TABLE 5.

Sensitivity and specificity for techniques compared to conventional PCR^a

Test	% Sensitivity	% Specificity	% Positivity (no. of samples [n = 106])
Real-time PCR	100	100	11.3 (12)
NASBA	100	100	11.3 (12)
Serodia (acute-phase sample)	50	100	5.6 (6)
Serodia (convalescent-phase sample)	66	100	7.5 (8)
CFT	75	100	8.5 (9)

^aSee reference 16 for details on conventional PCR.

Evaluation of serological testing.

Serological testing showed that 9 (8.5%) and 8 (7.5%) out of the 106 patients had a positive CFT result and a positive Serodia PPA result, respectively (Table 4). By testing the convalescent-phase sample instead of the acute-phase sample in the Serodia assay, two more positive results were obtained. Six samples were positive by both the Serodia PPA and the CFT. Three samples were positive in the CFT and negative in the Serodia assay, and two samples were positive by Serodia and negative by CFT. These five serologically discrepant samples were all positive by the molecular methods. There was one sample that was positive in the three molecular assays and negative by both of the serological methods.

Clinical data.

Although the number of positive samples was limited, the clinical data was evaluated as well. Diagnosis for M. pneumoniae was made on the basis of the PCR results. Comparison of clinical data of patients diagnosed with M. pneumoniae (n = 12) and those who were M. pneumoniae negative (n = 94) revealed that a lower mean age, the presence of chills, and the detection of an erythrocyte sedimentation rate (ESR) and C-reactive protein (>50 mg/liter) were indicative of possible Mycoplasma infection (Table 6). Rhinitis was significantly more abundant in the Mycoplasma-negative patients (P < 0.001). It was also found that none of the patients with Mycoplasma infection had pulmonary disease, painful cervical lymph nodes, or vomiting. No significant difference was seen between the two groups regarding sputum production, fever, infiltrate on a chest X ray, and patient sex. In addition, no dual infections were seen (conventional microbiological results not shown).

TABLE 6.

Clinical data for Mycoplasma-positive and -negative groups

Characteristic	No. of patients with:		P	Odds ratio (confidence interval [crude])
	Positive Mycoplasma PCR results (n = 12)	Negative Mycoplasma PCR results (n = 94)
Sex (female)	4 (33%)	52 (55%)	0.15	0.4 (0.1-1.4)
Mean age in yr (SD)	43 (9)	51 (16)	0.011
Pulmonary disease	0 (0%)	19 (20%)	0.12
Fever	12 (100%)	77 (82%)	0.21
No sputum	5 (42%)	16 (17%)	0.058	3.5 (1.0-12.4)
Rhinitis	1 (8%)	58 (62%)	<0.001	0.1 (0.1-1.4)
Chills	4 (33%)	52 (55%)	0.15	0.4 (0.0-0.5)
Vomiting	0 (0%)	18 (19%)	0.21
Painful lymph nodes	0 (0%)	9 (10%)	0.59
Infiltrate on chest X ray (n = 100)	4 (33%)	16 (18%)	0.25	2.3 (0.6-8.4)
CRPa ≥ 50 mg/liter (n = 101)	10 (83%)	44 (49%)	0.03	5.1 (1.1-24.7)
ESRb (n = 102)	12 (100%)	52 (58%)	0.003

^aCRP, C-reactive protein.

^bThe ESR normal levels were adjusted for age and sex as follows: for females of age 18 to 51 years, the normal level was 0 to 25 mm/h; for females of age 51 to 66 years, the normal level was 0 to 30; for males of age 18 to 51 years, the normal level was 0 to 15; for males of age 51 to 66 years, the normal level was 0 to 20; for males and females of more than 66 years of age, the normal level was 0 to 40.

DISCUSSION

The standard laboratory method for the diagnosis of M. pneumoniae as an etiological agent for LRTI and CAP has been culture or serology (9). PCR has been shown to be a better diagnostic test than conventional techniques (6, 7, 13, 14, 23, 30), and Loens et al. showed recently that NASBA is a good alternative as well (20).

In the present study, an internally controlled real-time PCR assay that targets the P1 adhesion gene for the diagnosis of M. pneumoniae infections was designed. Real-time PCR-based fluorescence assays have advantages over conventional PCR (14, 24). The fluorescent probes provide additional specificity for the PCR without the requirement of post-PCR processing. This obviously reduces the potential risk of product carryover. More importantly, real-time PCR leads to a significant reduction in the time to results and, therefore, better patient management.

The real-time PCR assay for M. pneumoniae, which was shown to be specific and to have good analytical sensitivity, was clinically evaluated using a group of patients with LRTI. In a group of 106 patients, 12 (11%) positive results were detected, which is similar to the rate of M. pneumoniae infection in adult populations found in other studies (19). For patients with CAP, rates of 15 to 30% (22) or as low as 1 or 3% (27) have been reported. The M. pneumoniae real-time PCR-positive results were confirmed by a conventional PCR (16). This PCR targets the same gene; therefore, another confirmation was sought in the NASBA. The NASBA reaction is targeted to the 16s rRNA, and the same positives were again detected, resulting in 100% agreement of the results obtained by the molecular methods. An additional feature of this NASBA is that the amplified products are detected using a fluorescent probe in real time and that this isothermal NASBA amplification and detection can be performed on an iCycler IQ real-time detection system. The NASBA reaction detects not only M. pneumoniae types 1 and 2 but also M. genitalium, which has the same 16S rRNA sequence. Although some reports show that M. genitalium is found in the respiratory tract (5, 17), no evidence of its presence was detected in patients in our study group, as all NASBA-positive results were real-time PCR positive as well.

Diagnosis of M. pneumoniae by molecular methods appears superior to that by serology. Using the acute-phase sera, the Serodia PPA missed half (6 of 12) of the active infections of M. pneumoniae. Adding the convalescent-phase sample increased the sensitivity to 66%. The cutoff for a positive result in the PPA assay was set at a titer of ≥320. Nonspecific results can be obtained with a titer of ≥160. A few samples with an acute-phase titer of 160 showed no increase in titer in the convalescent-phase serum. A titer of ≥320 is obviously a better indication of M. pneumoniae infection, especially since in a diagnostic laboratory the convalescent-phase serum is not always obtained.

One sample was negative by both serological methods but positive using molecular methods. In this case, the serum samples had been collected only 8 days apart, which could have affected the sensitivity of the second sample. Although in some studies, a larger number of positive results have been obtained by serology than by PCR methods (7, 30), in the present study all the serologically positive results were also positive by PCR and NASBA.

The results show that real-time PCR using throat swabs is a very good method for the diagnosis of M. pneumoniae and that the samples obtained were obviously of high quality. PhHV was used as the internal control and was added prior to extraction, which enabled the whole process to be evaluated. To implement this assay in routine diagnosis, a DNA isolation and inhibition control reaction was designed and was included in a duplex reaction using FAM and Cy5 as fluorophores. The amplification efficiency and sensitivity of the M. pneumoniae and PhHV real-time duplex PCR were comparable to those of the individual assays. No inhibition of the PCR was seen in DNA extracted from the throat swabs used in our study. However, sputa and nasopharyngeal aspirates have been described as exhibiting greater sensitivity (16, 26), and these specimens are more likely to be inhibitory. Therefore, to reliably detect microorganisms in clinical specimens by molecular methods and to avoid false-negative results, the use of internal controls is indispensable.

Foy et al. reported several clinical parameters associated with M. pneumoniae infection (12), whereas others have found that the clinical presentation was difficult to discriminate from those of other infections (7, 28). Although in this study only 106 patients were analyzed for the presence of M. pneumoniae, the clinical factors were assessed. Significant differences between the patients diagnosed with M. pneumoniae included results for mean lower age, the presence of chills, increased ESR, and raised C-reactive protein levels. Obviously, conclusions are difficult with the small number of positive cases seen in this study. However, as the presence of chest infiltrate, sputum, and fever was not found to be predictive for M. pneumoniae as distinct from other pathogens causing respiratory tract infections, diagnosis by laboratory confirmation is required.

In conclusion, real-time PCR and real-time NASBA are sensitive, specific, and rapid methods for M. pneumoniae diagnosis in adults. Real-time methods are clearly superior to conventional diagnostic assays (including conventional PCR) and provide quantitative data for the pathogen present. Real-time PCR with the PhHV multiplexed in the same tube provides a reliable tool for implementation in the diagnostic laboratory and will improve patient management.