Diagnostic Utility and Clinical Significance of Naso- and Oropharyngeal Samples Used in a PCR Assay To Diagnose Mycoplasma pneumoniae Infection in Children with Community-Acquired Pneumonia

Abstract

PCR assays of naso- and oropharyngeal samples among hospitalized children appear equally effective for the diagnosis of serologically confirmed community-acquired mycoplasmal pneumonia. However, the combination of results from both sites yields optimal sensitivity (57%), specificity (98%), and positive (92%) and negative (82%) predictive values when compared with Mycoplasma pneumoniae enzyme-linked immunosorbent assay.

Controversy exists over the role of PCR assays performed on upper respiratory tract (URT) samples for the diagnosis of mycoplasmal pneumonia. It has been suggested that the most sensitive and prompt method of diagnosis in children is a combination of URT PCR and acute-phase serum immunoglobulin M (IgM) enzyme immunoassay (4, 11). However, concerns exist about the potential inability to distinguish between URT carriage of M. pneumoniae versus active disease in children with PCR-positive URT specimens in the setting of pneumonia (6). Furthermore, the optimal site and method of sampling the URT for the detection of M. pneumoniae by PCR are not established in children.

Patients.

We report on 21 children with serologically proven M. pneumoniae infection and 42 approximately matched M. pneumoniae seronegative controls selected from 154 consecutively enrolled children hospitalized with clinically and radiologically confirmed lower respiratory infections, as described previously (8, 9). The Institutional Review Board of the University of Texas Southwestern Medical Center at Dallas approved the protocol.

Laboratory detection of M. pneumoniae.

M. pneumoniae enzyme-linked immunosorbent assays (ELISA) were performed by the Diagnostic Mycoplasma Laboratory at The University of Alabama at Birmingham. The assay was considered positive if the IgM was ≥1:10 or if there was ≥4-fold rise in IgG titer (1). Twenty-one of 154 children (13.6%) had serologically proven acute mycoplasmal pneumonia. Forty-two approximately matched M. pneumoniae seronegative control subjects were selected from the same study population.

Both nasopharyngeal (Np) and oropharyngeal (Op) swabs were obtained within 24 h of admission by using dacron-tipped swabs (Fisher Scientific). Swab tips were placed in 200-μl PCR buffer and stored at −70°C. Upon thawing, 25 μl of protease solution was added, and samples were lysed in Buffer AL (QIAamp blood kit; QIAGEN GmbH) and sonicated at 37°C. Bacterial DNA was extracted by centrifuge columns (QIAGEN GmbH). Single-step DNA amplification was performed with a programmable thermal cycler (GeneAmp PCR system 9600; Perkin-Elmer Cetus, Norwalk, Conn.), based on a method modified from that of Williamson et al. (7, 12 ). The 100-μl reaction mixture consisted of 40 μl of DNA eluent and 60 μl of HotStarTaq DNA polymerase master mix (QIAGEN GmbH). The published primer sequences targeted the highly conserved region of the 16S rDNA gene of M. pneumoniae and amplified a unique 290-bp region (12). A 10-μl volume of amplified product was separated by electrophoresis in polyacrylamide gel and visualized under UV light after ethidium bromide staining. We confirmed the identification of M. pneumoniae by means of real-time PCR assay targeting the P1 adhesion protein gene (6). M. pneumoniae DNA extract isolated from M. pneumoniae stock (ATCC 29342) and sterile distilled water were amplified with each assay as positive and negative controls, respectively. We demonstrated that the lower limit of detection of the PCR assay was <10 CFU/ml.

Statistics.

Independent Student's t tests were used for normally distributed continuous data, and the Mann-Whitney U test was used for nonparametric data. The Yates corrected chi-square test or Fisher's exact test was used for categorical data where appropriate (two-tailed). A P value of less than 0.05 was defined as significant.

Twelve of the 21 (57%) children diagnosed with M. pneumoniae by serum ELISA had positive M. pneumoniae PCR assays from either Np or Op swabs; 9 out of the 12 were Np swabs and 9 out of the 12 were Op swabs (six children had positive Np and Op swabs). One of the 42 control children with pneumonia had positive PCR assays for M. pneumoniae from both Np and Op swabs. Because convalescent-phase serum from this patient was not available, we could not rule out acute mycoplasmal infection with certainty. The detection of M. pneumoniae by PCR was significantly greater for children with positive M. pneumoniae ELISA than for children with negative ELISA (odds ratio, 54.7; range, 5.9 to 1,279.3). Compared with M. pneumoniae ELISA as the diagnostic standard, PCR performed on Np and Op samples demonstrated a sensitivity of 57.1%, a specificity of 97.6%, a positive predictive value of 92.3%, and a negative predictive value of 82.0%. Table 1 presents characteristics of and comparisons among the children with positive M. pneumoniae serology by URT PCR status.

TABLE 1.

Comparison of 21 hospitalized children with pneumonia and positive M. pneumoniae ELISA by URT PCR status (Op- or Np sample)

Characteristica	URT mycoplasmal PCR resultb		P value
	Negative	Positive
No. of subjects	9	12
Age (mo)	79 (62)	55 (54)	0.20
Male:female	3:6	7:5	0.39
Duration of prior symptoms (days)	6.2 (4.0)	5.4 (5.0)	0.69
Antibiotic therapy within the 2 weeks preceding admission (%)	33	25	1.0
Macrolide or azalide treatment within the 2 weeks preceding admission (%)c	33	0	0.06
Tmax before admission was ≥39°C	4/6	4/11	0.34
Wheezing present at time of admission (%)	44	33	0.67
Duration of oxygen therapy (days)	0.78 (0)	1.7 (1.25)	0.045
Duration of hospitalization (days)	5.4 (4.0)	4.3 (4.0)	0.41
CXR showed unequivocal consolidation and/or pleural effusion (%)	78	50	0.37
CXR showed interstitial infiltrate (%)	0	0
CXR showed hyperinflated lungs (%)	0	0
WBC (106/ml)	18.7 (18.5)e	13.6 (11.8)d	0.24
Band forms (%)	8.4 (10)f	6.3 (3.0)d	0.69
Procalcitonin (ng/ml)	10.1 (0.3)	1.9 (0.6)	0.57

^aCXR, chest roentgenogram; WBC, white blood cell count; T_max, maximum temperature.

^bValues are means (unless otherwise indicated, values are numbers of patients). Values in parentheses are medians.

^cAgents active against mycoplasma included erythromycin, azithromycin, and clarithromycin.

^dResults were available from 12 patients.

^eResults were available from six patients.

^fResults were available from five patients.

Although URT PCR has been advocated for the diagnosis of M. pneumoniae pneumonia in children, the optimal site and method of sampling is not clear. We found that PCR assays of Np or Op dacron-tipped swabs appear equally effective for the detection of M. pneumoniae. However, combining test results from both sites provides the greatest diagnostic yield. In adults with symptoms of upper and/or lower respiratory tract infection, Gnarpe et al. found no significant difference between Np and Op swabs for the detection of M. pneumoniae by PCR (5). They also observed that sampling both sites gave the greatest yield. Reznikov et al. compared Np aspirates to Op swabs in children and found no significant difference in the detection of M. pneumoniae by PCR (10). They did note that Np aspirate specimens were significantly more likely to be rejected than Op swabs because of PCR inhibitors or lack of respiratory epithelial material. Based on the available data, no anatomic site or method is clearly superior for obtaining URT samples. Combining test results from more than one site does appear to improve diagnostic accuracy.

In this investigation of children hospitalized with acute community-acquired pneumonia, PCR of Np and Op swabs had a high positive predictive value (92.3%) and specificity (97.6%) for the diagnosis of M. pneumoniae compared with acute and convalescent ELISA. Dorigo-Zetsma et al. tested 74 control children without signs of respiratory infection or recent antibiotic treatment and found no samples to be PCR positive for M. pneumoniae on Op sampling, while 8% of children with signs of respiratory infection had URT specimens that were PCR positive (3). The combination of these findings indicates a low false-positive rate when URT samples from children are tested for M. pneumoniae DNA. A positive URT M. pneumoniae PCR for a child with community-acquired pneumonia is suggestive of M. pneumoniae pneumonia rather than respiratory carriage.

URT PCR status was examined in relation to clinical manifestations of M. pneumoniae pneumonia in children diagnosed by acute and convalescent ELISA. The duration of oxygen therapy was found to be significantly longer in PCR-positive specimens than in PCR-negative specimens. Therefore, a positive URT PCR assay may signify a greater severity of pneumonia. In support of our findings, Dorigo-Zetsma et al., utilizing semiquantitative M. pneumoniae PCR on Op swabs, documented a significant relationship between M. pneumoniae DNA load and severity of disease in hospitalized, nonhospitalized, and asymptomatic contacts (2).