Mycoplasma pneumoniae accounts for 4%–20% of community-acquired pneumonia (CAP) in pediatric patients worldwide.1,2 This frequency varies widely, because—among other things—diagnostic tests for this bacterium are not sensitive and specific in most hospitals and health institutions. This means that in many countries including Colombia, there is no epidemiologic surveillance program to monitor infections caused by this bacterium. Additionally, it is common for primary care sites to lack reliable rapid tests to confirm this pathogen microbiologically. Many infections and even outbreaks are falsely attributed to this bacterium, go unnoticed or are underdiagnosed, which prevents our understanding of the real impact of this bacterium on public health.3
According to the Centers for Disease Control and Prevention, diagnosis of M. pneumoniae is performed using culture, serology and molecular tests.4 However, bacteria require enriched growth media that are only available in reference laboratories.56 In the absence of culture, paired serology (IgM and IgG antibodies) against M. pneumoniae has become the definitive diagnosis.6–8 The problem with this technique is that it requires between 4 and 8 weeks to take the sample in the convalescent phase, which restricts its utility in diagnosis and clinical decision-making.9 Due to this limitation, it has been decided to measure IgM titers in the acute phase; however, because these antibodies can remain high for months or possibly years, they may not appear in very young children or during re-infection,6 and the positive predictive value of this test can be as low as 15%,10 the results of the test are unreliable.
The use of molecular tests based on PCR enable detection of M. pneumoniae DNA in studied samples. The potential advantages of PCR are that it does not require the microorganism to be viable in the sample to test positive, results can possibly be obtained in hours, and it has high sensitivity.11,12 However, these tests are rarely offered in primary care services given the infrastructure required to conduct them.13
However, the results obtained in some studies leave doubt about the true utility of this diagnostic test for M. pneumoniae, especially because its sensitivity in sputum samples varies widely (9%–100%) and it has poor concordance with paired serology (kappa <0.4), which is considered definitive diagnosis of M. pneumoniae infection.14,15 Additionally, the high analytic sensitivity of PCR, which in some cases can detect <1002 CFU/mL16 or 50 fg of target DNA,17 does not reliably differentiate infection from colonization.6 Thus, both acute phase IgM and molecular and conventional techniques often have false positive results even in healthy individuals.
The objective of this study was to compare the performance of serology in the acute phase (IgM) and paired phase (IgG and IgM) with that of commercial PCR for the diagnosis of M. pneumoniae in children with and without CAP.
MATERIALS AND METHODS
Study Design: Cross-sectional Study
Participants and Sample
The inclusion and exclusion criteria of the population were published elsewhere.18 Briefly, children between 1 month and 17 years of age with CAP who had <2 weeks of symptoms were included. Those children with immunosuppression, neurologic or psychiatric disorders, structural lung diseases or who had received >72 hours of antibiotics were excluded.
Additionally, a control group (non-CAP group) was included that consisted of a convenience sample of 61 children who required ambulatory surgery without evidence of respiratory infection or antimicrobial therapy in the last 15 days.
In children in the CAP group, induced sputum (IS) samples and blood samples were taken in the acute and convalescent phases (from 4 to 8 weeks after the acute phase). In children in the non-CAP group, a tracheal aspirate was obtained when they were intubated before start of surgery, and a blood sample was taken during their elective procedure. The respiratory samples were transported and kept in cold chain (2°C–6°C) for a maximum time of 5 hours and stored at −80°C, while the sera was stored at −20°C.
The detection of IgM and IgG antibodies against M. pneumoniae in serum was performed with the quantitative immunoenzymatic assay for Mycoplasma pneumoniae IgG/IgM® ELISA (Vircell SL, Granada, Spain) according to the recommendations of the manufacturer.
In several hospitals where children with CAP were treated, there was a rapid serologic test that qualitatively detected IgM type antibodies (Immunocard® Mycoplasma, Bioscense, Inc) in serum for their diagnosis. When available, these data were collected and analyzed independently.
DNA was extracted from 500 μL of sputum previously treated with 2% NALC/NaOH or from tracheal aspirate, for which the QIAmp DNA commercial kit (Qiagen, Valencia, CA). In each round, a negative extraction control was included to rule out cross-contamination. DNA was stored at −80°C until performing the molecular tests.
Detection of M. pneumoniae by PCR
The Seeplex ACE Pneumobacter Detection commercial kit (Seegene, Songpagu, Korea) was used, which detects 5 other respiratory pathogens in addition to M. pneumoniae: Legionella pneumophila, Streptococcus pneumoniae, Haemophilus influenzae, Bordetella pertussis and Chlamydophila pneumoniae. All the recommendations of the manufacturer were followed.
Diagnosis of M. pneumoniae
A paired serology or positive PCR amplification resulted in diagnosis of pneumonia by M. pneumoniae.4 Serologic positivity was defined as a 4-fold increase in the IgG or IgM antibody titer of the convalescent phase serum with respect to the acute phase, as long as these values exceeded the positivity threshold described by the commercial kit (>11). In the case of PCR, the result was considered positive if a band of the same size as the positive control was observed, which was ~583 bp in the agarose gel and corresponded to M. pneumoniae.19
Additionally, the positive or negative result of the IgM serology in the acute phase was considered—including both the qualitative test (rapid test) used in some of the participating institutions (when the treating physician considered it necessary) and the immunoassay ELISA used by us—following the specifications of the merchant.
The qualitative variables were expressed using relative and absolute frequencies. Normal tests were applied to the quantitative tests, and the data are expressed as the median and interquartile range (IQR). The proportions of positive results of each of the diagnostic techniques used were estimated both in children with CAP and without CAP. The percentage of diagnostic agreement between PCR and paired serology was also evaluated. To determine the concordance between these tests, Cohen’s kappa coefficient was used. The concordance was considered low when the kappa score was <0.6, moderate to good when it was between 0.6 and 0.8, and excellent when it was >0.8.20 The statistical analyses and the Cohen’s kappa coefficient were performed with the statistical programs SPSS® v.21.0 and Epidat® v.3.1, respectively.
Characteristics of Children and Frequency of M. pneumoniae
In the group with CAP, 524 IS samples were obtained, as well as 497 blood samples in the acute phase and 491 in the convalescent phase. There were 13.9% of these children (73/525) with positive tests (either PCR or paired serology, or both) for M. pneumoniae, and 3.3% in the group without CAP. In the non-CAP group, we collected blood samples in acute phase in 60 of the 61 children. One child in the CAP group and none of the children without CAP had received macrolides in the 3 months before the time of enrollment. The clinical data of the children in the CAP group were previously described.18
Serologic Tests and PCR Findings
Among the 491 children with CAP who had serology available in the acute and convalescent phases, 6.7%33 exhibited quadrupled antibody titers against M. pneumoniae, 4.5% exhibited quadrupled antibody titers against IgG, 1.2% exhibited quadrupled antibody titers against IgM and 1.0% exhibited quadrupled antibody titers against both (Fig. 1).
The PCR studied in IS was positive in 10.3% (54/524) of children in the CAP group, 14 of whom also had quadrupled IgG and/or IgM titers. The remaining 40 were positive only by PCR (Table 1). In children without CAP, PCR was positive in 3.3% (2/60) (Fig. 2).
We found no significant differences between the symptoms appearance and sampling of the patients who had positive results in the multiplex PCR (5.4 days ± 4.1), the serology (2.6 days ± 2.5) or were positive for both tests (4.1 days ± 2.1). The results were similar when we analyzed the difference in days for those children with acute positive IgM (5.4 days ± 4.0) and IgG (5.4 days± 4.8).
There was a difference between the age of CAP [median 35 months (IQR: 20.5–52)] and non-CAP group [median 54 months (IQR: 29–73.5)] (P = 0.003). However, there were no statistical differences between the positivity of the PCR [CAP group: median 36 months (IQR: 21–53) vs. non-CAP group: median 46 months (IQR: 8–83), P: 0.748] and acute phase serology [CAP group: median 39 months (IQR: 21–63) vs. non-CAP group: median 53 months (IQR: 32–75), P: 0.196] by age between both groups.
Agreement and Concordance Between the Diagnostic Tests
The highest agreement between the techniques was 91% between paired serology (IgM or IgG) and PCR (Table 1). Although this agreement was high, the concordance calculated by the kappa index, which considers the discrepancies, was unacceptable at 0.3596 (95% CI: 0.1267–0.5024).
Additionally, PCR was compared with the different types of IgM analysis performed in the acute phase—including both ours (quantitative, Vircell) and that of hospital institutions (qualitative, Immunocard Mycoplasma, Bioscense, Inc). The agreement was 71% between quantitative IgM in the acute phase and PCR and 70% between the qualitative test and PCR; the concordance by kappa index was 0.2252 (95% CI: 0.1481–0.3023) and 0.1681 (95% CI: −0.0347 to 0.3710), respectively (Table 2).
In our group of 525 children with CAP and 61 controls, detection of M. pneumoniae DNA by PCR in IS had the highest diagnostic yield (10.3%), while the paired serology (by IgG or IgM) was positive in 6.7% of those positive cases. Taking both tests into account, M. pneumoniae explained 13.9% of these cases in our children. However, our data show that there is no diagnostic concordance between the tests. Given that the IgM test in the acute phase, which is available in many of our hospital institutions (Immunocard), was positive in one-third of the patients with CAP and 40% of children without CAP, the existing difficulties for the diagnosis of this microorganism become evident.
Our results, which were obtained from IS samples from children with CAP, add to the findings generated by other authors in different situations. Spuesens et al,21 in the Netherlands, reported that 16.2% of pharyngeal swab samples from children between 3 months and 16 years of age with upper respiratory infection had positive real-time PCR results for M. pneumoniae. Similarly, Wagner et al22 in Switzerland documented 39.7% (146) positive cases for this bacterium when evaluating the diagnostic performance of a multiplex RT-PCR to detect atypical bacteria in 368 different respiratory samples (sputum, BAL, respiratory swabs and tracheal and nasopharyngeal secretions) of patients with symptoms of atypical pneumonia. Both authors conclude that current diagnostic tests do not allow differentiation between asymptomatic carriers and symptomatic M. pneumoniae infections.
In this sense, Meyer Sauteur et al9 reported the presence of M. pneumoniae in the respiratory tract of 21%–56% of asymptomatic children without causing any type of disease in their host. Likewise, Spuesens et al (2013) found that M. pneumoniae DNA in asymptomatic children can be detected up to 3 months after the initial sample21 and that the DNA load of this bacteria in children with and without respiratory infection is similar (P: 0.11). Therefore, studies are needed to determine the real role of M. pneumoniae in the host lung and the possible triggers to induce disease processes in the host.
Although paired serology has been considered definitive diagnosis of M. pneumoniae infection, its sensitivity is higher than that of the molecular technique14,23 and its specificity is affected in these cases because it can become positive after any acute infection by M. pneumoniae, including upper respiratory ones that are often subclinical. Given that it is not useful in clinical practice due to the need for a convalescent sample when the possible outcomes of the patient have already occurred, the diagnosis in most hospital institutions rests only in the detection of IgM in the acute phase. In our environment, this is often a qualitative test (Inmunocard) of low performance compared with paired serology (31.1%).24,25 This suggests that the measurement of IgM antibodies at a single moment of infection is not a reliable indicator for the detection of M. pneumoniae in lower respiratory infections. The same occurs with IgM in the acute phase of the ELISA IgG/IgM (Vircell SL, Granada, Spain) used by us; only some of the initially positive children for IgM exhibit quadrupled antibody titers later in the convalescent phase.
As with the detection of M. pneumoniae by PCR in respiratory samples of asymptomatic children, it is also clear that a positive IgM result can be detected in healthy children. Nir-Paz et al reported that at least 20% of children, adolescents, and young adults without respiratory symptoms who participated in an evaluation study of 8 serologic tests for M. pneumoniae had a positive IgM result for any of the techniques evaluated.7 Similar to our findings, Spuessens et al found no significant differences in the frequency of IgM antibodies between the groups of symptomatic and asymptomatic children (12.6% vs. 9.2%, P = 0.23).21
It is estimated that IgM antibody levels peak between the third and sixth weeks and then gradually decrease.26,27 It is clear that IgM-type antibodies generally increase in the first infection by M. pneumoniae, but their increase is minimal in patients who are re-infected with the bacteria; thus, the utility of this test in older children and adults should be reconsidered.28 It is expected that these children have had previous exposures to the bacteria, given the epidemiologic cyclic dynamics of these infections.
There was low concordance observed between the diagnostic tests in our study. Reviewing the literature, there are few studies that compare the performance of PCR and serology in large pediatric samples.6 In general, a lower concordance was observed between children 0 and 3 years of age (0.486) compared with children who are >6 years (0.783).29 A Chinese study that included 3146 children hospitalized for CAP, in which sputum PCR was considered a reference test, found that the proportion of false-positive paired serologies [serology (+)/multiplex-PCR (−)] taken in an interval of only 7 days between both samples was 60% (78/130) in children <3 years, 39% (51/132) in children between 3 and 6 years and 31% (50/160) in children >6 years. Additionally, the proportion of false-negative serologies [serology (−)/multiplex-PCR (+)] was 21% (14/66), 4% (3/84) and 2% (2/112) among children 0–3, 3–6 and >6 years of age, respectively.30 Chang et al also reported a large discrepancy between the results obtained using IgM and RT-PCR; only 12.6% (23/182) of the patients showed positive results for both tests (kappa index of 0.445). RT-PCR did not detect M. pneumoniae DNA in 47.7% of the plasma samples that were positive for IgM.30 Other authors present similar results.15,31,32 The most likely explanation for these findings is that acute phase IgM antibodies appear positive with negative PCRs because of the duration of IgM antibody circulation after an infection by M. pneumoniae and the differences in the sensitivity of conventional PCR versus multiple PCR (higher in the former). On the other hand, positive PCR without quadrupled titers may be due to the presence of bacterial DNA,21 colonization by M. pneumoniae or lack of response to elevate plasma levels of IgM in some patients.6,9,33
Some authors have proposed alternatives such as the amplification of M. pneumoniae ribosomal RNA. The researchers found that simultaneous amplification and testing of M. pneumoniae performed on pharyngeal swabs or sputum obtained from children between 1 month and 10 years of age with pneumonia was more sensitive than real-time PCR, with a detection limit of 101 CFU/mL, and showed good agreement with IgM in the acute phase (Kappa index 0.797).34
Given the above, it is necessary to develop new tests or improve existing tests for the diagnosis of M. pneumoniae.
Our study has some limitations. Due to a budget deficit, the number of children in our control group was smaller than that in the group with pneumonia, which could affect the comparisons between the 2 groups. Although lower respiratory samples were analyzed in both groups, including IS in children with CAP and tracheal aspirate in children without CAP, it was not possible to perform a second measurement of IgM and IgG comparable to the convalescent sample in children without CAP for ethical reasons.
Finally, and although some authors question the exclusion of children with respiratory symptoms as controls in etiologic studies of pneumonia,35 we believe that the ideal controls to evaluate the diagnostic tests are children without respiratory symptoms. This is because it decreases the probability of detecting false-positive CAP by pathogens of the upper respiratory tract, and the actual frequency of colonization in “healthy” children is better reflected.
Our study allows us to conclude that among the alternatives evaluated, PCR in IS has the best diagnostic performance; the sensitivity of IgM in the acute phase is low, and it is equally positive in children with CAP and healthy children and thus useless for the diagnosis of M. pneumoniae in these cases; and the concordance between diagnostic tests is very poor. Therefore, it is necessary to optimize these tests or develop new ones to optimize the diagnosis of CAP caused by M. pneumoniae.
We thank all the participating institutions (Department of Health and Social Protection of Antioquia, Metrosalud ESE, Hospital Foundation San Vicente de Paul, Pablo Tobón Uribe Hospital, CES Clinic, San Rafael de Itagüí ESE Hospital, Marco Fidel Suárez Hospital, Manuel Uribe Ángel ESE Hospital, General Hospital of Medellín “Luz Castro de Gutierrez,” Soma Clinic, Santa Ana Children’s Clinic, Health Services Provider Institution of the University of Antioquia, Las Américas Clinic and Sagrado Corazón Clinic) for their support of the study, and we thank all the children and their families. The resources of the research group were used to collect, analyze and interpret the results.
1. Waites KB, Balish MF, Atkinson TP. New insights into the pathogenesis and detection of Mycoplasma pneumoniae
infections. Future Microbiol. 2008;3:635–648.
2. Waites KB, Xiao L, Liu Y, et al. Mycoplasma pneumoniae
from the respiratory tract and beyond. Clin Microbiol Rev. 2017;30:747–809.
3. Diaz MH, Winchell JM. The evolution of advanced molecular diagnostics for the detection and characterization of Mycoplasma pneumoniae
. Front Microbiol. 2016;7:232.
4. Pneumonia | Mycoplasma pneumoniae
| Diagnostic Methods for Labs | CDC [Internet]. 2019 [cited 2019 Mar 2]. Available from: https://www.cdc.gov/pneumonia/atypical/mycoplasma/hcp/diagnostic-methods.html
. Accessed January 28, 2020.
5. Pérez I, Gómez M, González Rico S. El diagnóstico convencional de Mycoplasma pneumoniae
como agente causal de Neumonías Adquiridas en la Comunidad (NAC). Rev Soc Venez Microbiol. 2007;27:73–78.
6. Loens K, Ieven M. Mycoplasma pneumoniae
: current knowledge on nucleic acid amplification techniques and serological diagnostics. Front Microbiol. 2016;7:448.
7. Nir-Paz R, Michael-Gayego A, Ron M, et al. Evaluation of eight commercial tests for Mycoplasma pneumoniae
antibodies in the absence of acute infection. Clin Microbiol Infect. 2006;12:685–688.
8. Marchello C, Dale AP, Thai TN, et al. Prevalence of atypical pathogens in patients with cough and community-acquired pneumonia: a meta-analysis. Ann Fam Med. 2016;14:552–566.
9. Meyer Sauteur PM, Unger WW, Nadal D, et al. Infection with and carriage of Mycoplasma pneumoniae
. Front Microbiol. 2016;7:329.
10. Mejía S, Rueda Z, López L, et al. Diagnóstico serológico de la infección por Mycoplasma pneumoniae
en niños hospitalizados por neumonía adquirida en la comunidad, 2011. 2012 Poster presented at: VIII Encuentro Nacional de Investigación en Enfermedades Infecciosas; Armenia, Colombia.Agosto;
11. Blanco S, Fuenzalida L, Bas A, et al. Comparison of 2 molecular assays and a serologic test in diagnosing Mycoplasma pneumoniae
infection in paediatrics patients. Diagn Microbiol Infect Dis. 2011;71:463–466.
12. Loens K, Van Heirstraeten L, Malhotra-Kumar S, et al. Optimal sampling sites and methods for detection of pathogens possibly causing community-acquired lower respiratory tract infections. J Clin Microbiol. 2009;47:21–31.
13. SLIPE. Consenso de la Sociedad Latinoamericana de Infectología Pediátrica (SLIPE) sobre Neumonía Adquirida en la Comunidad (NAC). Rev Soc Bol Ped. 2010;XXIV:1–24.
14. Kumar S, Garg IB, Sethi GR. Mycoplasma pneumoniae
in community-acquired lower respiratory tract infections. Indian J Pediatr. 2018;85:415–419.
15. Thurman KA, Walter ND, Schwartz SB, et al. Comparison of laboratory diagnostic procedures for detection of Mycoplasma pneumoniae
in community outbreaks. Clin Infect Dis. 2009;48:1244–1249.
16. Ratliff AE, Duffy LB, Waites KB. Comparison of the illumigene Mycoplasma
DNA amplification assay and culture for detection of Mycoplasma pneumoniae
. J Clin Microbiol. 2014;52:1060–1063.
17. Thurman KA, Warner AK, Cowart KC, et al. Detection of Mycoplasma pneumoniae
, chlamydia pneumoniae, and legionella spp. in clinical specimens using a single-tube multiplex real-time PCR
assay. Diagn Microbiol Infect Dis. 2011;70:1–9.
18. Copete AR, Aguilar YA, Rueda ZV, et al. Genotyping and macrolide resistance of Mycoplasma pneumoniae
identified in children
with community-acquired pneumonia in Medellín, Colombia. Int J Infect Dis. 2018;66:113–120.
19. Park J, Kim J, Rheem I, et al. Evaluation of seeplexTM pneumobacter multiplex PCR
Kit for the detection of respiratory bacterial pathogens in pediatric patients. Ann Lab Med. 2009;29:307–313.
20. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–174.
21. Spuesens EB, Fraaij PL, Visser EG, et al. Carriage of Mycoplasma pneumoniae
in the upper respiratory tract of symptomatic and asymptomatic children
: an observational study. PLoS Med. 2013;10:e1001444.
22. Wagner K, Springer B, Imkamp F, et al. Detection of respiratory bacterial pathogens causing atypical pneumonia by multiplex Lightmix® RT-PCR
. Int J Med Microbiol. 2018;308:317–323.
23. Zhang L, Zong ZY, Liu YB, et al. PCR
for diagnosing Mycoplasma pneumoniae
infection: a systematic review & meta-analysis. Indian J Med Res. 2011;134:270–280.
24. Ozaki T, Nishimura N, Ahn J, et al. Utility of a rapid diagnosis
kit for Mycoplasma pneumoniae
pneumonia in children
, and the antimicrobial susceptibility of the isolates. J Infect Chemother. 2007;13:204–207.
25. Ueda R, Okamura J, Inoue Y, et al. Rapid detection of Mycoplasma pneumoniae
IgM antibodies using immunoCard mycoplasma kit compared with complement fixation (CF) tests and clinical application. Eur Respir J. 2012;40(suppl 56):P2466.
26. Lee W-J, Huang E-Y, Tsai C-M, et al. Role of serum Mycoplasma pneumoniae
IgA, IgM, and IgG in the diagnosis
of Mycoplasma pneumoniae
-related pneumonia in school-age children
and adolescents. Clin Vaccine Immunol CVI. 2017;24: pii: e00471–16.
27. Matas Andreu L, Molinos Abós S, Fernández Rivas G, et al. Diagnóstico serológico de las infecciones por Mycoplasma pneumoniae
. Enfermedades Infecc Microbiol Clínica. 2006;24:19–23.
28. Daxboeck F, Krause R, Wenisch C. Laboratory diagnosis
of Mycoplasma pneumoniae
infection. Clin Microbiol Infect. 2003;9:263–273.
29. Wang L, Feng Z, Zhao M, et al. A comparison study between GeXP-based multiplex-PCR
assay for Mycoplasma pneumoniae
detection in children
with community acquired pneumonia. BMC Infect Dis. 2017;17:518.
30. Chang HY, Chang LY, Shao PL, et al. Comparison of real-time polymerase chain reaction and serological tests for the confirmation of Mycoplasma pneumoniae
infection in children
with clinical diagnosis
of atypical pneumonia. J Microbiol Immunol Infect. 2014;47:137–144.
31. Beersma MF, Dirven K, van Dam AP, et al. Evaluation of 12 commercial tests and the complement fixation test for Mycoplasma pneumoniae
-specific immunoglobulin G (IgG) and IgM antibodies, with PCR
used as the “gold standard”. J Clin Microbiol. 2005;43:2277–2285.
32. Templeton KE, Scheltinga SA, Graffelman AW, et al. Comparison and evaluation of real-time PCR
, real-time nucleic acid sequence-based amplification, conventional PCR
, and serology
of Mycoplasma pneumoniae
. J Clin Microbiol. 2003;41:4366–4371.
33. Sillis M. The limitations of IgM assays in the serological diagnosis
of Mycoplasma pneumoniae
infections. J Med Microbiol. 1990;33:253–258.
34. Li W, Fang Y, Shen H, et al. Evaluation of a real-time method of simultaneous amplification and testing in diagnosis
of Mycoplasma pneumoniae
infection in children
with pneumonia. PLoS ONE. 2017;12e0177842.
35. Higdon MM, Hammitt LL, Deloria Knoll M, et al.; PERCH Study Group. Should controls with respiratory symptoms be excluded from case-control studies of pneumonia etiology? reflections from the PERCH Study. Clin Infect Dis. 2017;64(suppl_3):S205–S212.