Comparison of different detection methods for Mycoplasma pneumoniae infection in children with community-acquired pneumonia

Mycoplasma pneumoniae (M. pneumoniae, MP) is known as a common cause of community-acquired pneumonia (CAP) in children throughout the world. M. pneumoniae is the smallest prokaryotic microorganism, lacks a cell wall and has a high affinity for respiratory epithelial cells. A total of 40% or more cases of CAP in children are caused by M. pneumoniae, and approximately 18% of cases require hospitalization [1]. At present, it is believed that the pathogenic mechanism of M. pneumoniae is to cause direct injury through community-acquired respiratory distress syndrome (CARDS) toxin, which induces eosinophilia, increases cytokine production and induces the hyperreactivity of the airway, such as asthma, in animal models [2‐4]. In addition, the immunological response resulting from infection by M. pneumoniae causes pulmonary and extrapulmonary symptoms [5].

A worldwide increase in the prevalence of macrolide-resistant mycoplasma pneumoniae (MRMP) strains has been witnessed since 2000 [6]. In some regions of Asia, the resistance rates have been reported to be over 90% [7], whereas in North America and Europe, the rates have reached 25% [8, 9]. Compared to other patients with CAP, children with M. pneumoniae infection may be older and have prolonged symptoms of fever and cough [10]. The period of signs and symptoms will be shortened if accurate antimicrobial treatment is started early in the course of diseases. However, Mycoplasma pneumoniae pneumonia (MPP) in children cannot be diagnosed exactly in terms of clinical manifestations [11].

Mycoplasma pneumoniae pneumonia has the characteristics of insidious onset, mild pulmonary signs and low specificity of clinical manifestations and imaging. Therefore, effective and sensitive laboratory diagnostic methods are the main basis for the diagnosis of M. pneumoniae infection. At present, there are many clinical methods for detecting M. pneumoniae, such as culture, serology and molecular-based methods. However, as the gold standard, culture is limited in clinical utility because it is time-consuming and insensitive [12]. Both serology and molecular assays have their own advantages and disadvantages in clinical practice. In this paper, we aimed to evaluate the clinical application value of different methods to detect M. pneumoniae infection by comparing and analysing them to improve diagnostic efficiency.

M. pneumoniae is a major cause of infectious diseases worldwide. M. pneumoniae affects different tissues and organs, especially the respiratory tracts, of children in all age groups, and it has become the second leading pathogen after Streptococcus pneumoniae in children with community-acquired pneumonia [14]. The occurrence of CAP caused by M. pneumoniae varies with age. A previous study has reported that M. pneumoniae pneumonia is uncommon in children under 5 years of age but has a higher incidence among school-age children [1]. However, M. pneumoniae infections may occur in people from infancy through old age [15, 16]. From 2010 to 2012, 2638 children with pneumonia requiring hospitalization in three hospitals in the United States were enrolled in a study. Respiratory specimens were systematically collected and detected using real-time PCR, and M. pneumoniae was found to be more common in children ≥5 years (19% vs. 3%) [17]. In accordance with our data, the detection rates of M. pneumoniae by different methods in the three age groups were different, but the distributions were the same. School-aged children and adolescents were the most common ages affected, and there was rarely a significant infection before 3 years old, which was consistent with the epidemiological characteristics of M. pneumoniae.

At present, there are many methods for the laboratory diagnosis of M. pneumoniae infection, including culture, serological detection and molecular assays. Culture is often used for antimicrobial susceptibility testing or typing due to its high specificity. However, due to the complicated procedure, prolonged turnaround time and low sensitivity, culture is not recommended for routine testing [18, 19]. Serological tests are currently the most widely used detection methods of M. pneumoniae in clinical practice and are more sensitive than culture. Diagnostic sensitivity for serological tests of M. pneumoniae is determined by both the collection time of specimens and the performance characteristics of the methods. In our study, the positive rate of MP-IgM (PA) was 40%, which was higher than that of the others, which may have been attributed to the average time being 10 days before admission to our department. Although MP-IgM is the earliest antibody produced after M. pneumoniae infection, it still takes a certain period of time before it can be detected. In general, MP-IgM can be detected within approximately 1 week after infection, reaching a peak after 3 to 4 weeks, resulting in possible false negatives in the early stage [20]. It has been reported that the most accurate diagnosis of serology is obtained when paired sera collected at least 2 weeks apart are tested for both IgM and IgG at the same time, resulting in a 4-fold increase in titre [1]. However, in paediatrics, it is simply impossible to perform repeated blood sampling in a short time. MP-IgM (GICT), a rapid test kit based on immunochromatography, is a relatively new serological test. One study from Wei Li found that the specificity and sensitivity of MP-IgM (GICT) were 100 and 97.4%, respectively, compared with real-time PCR [21], which was not in line with our results. It might be that the dilution of the sample to only 100 μl caused false negatives. Due to the easy and rapid procedure (15 min), MP-IgM (GICT) is suitable for the identification of M. pneumoniae infections in paediatric outpatient departments. However, MP-IgM (PA) is still recommended for inpatients to reduce the false-negative rate.

With the development of molecular assays, nucleic acid amplification technology (NAAT) has gradually become an important method for the early rapid diagnosis of M. pneumoniae infection, which may help with early appropriate antibiotic therapy. Owing to the advantages of quick turnaround times, lower likelihood of contamination, higher sensitivity and higher specificity, as well as not being limited by time and immune function, NAAT is important in the early diagnosis of M. pneumoniae. In addition, NAAT has various detection formats, which can provide quantitative data, detect antimicrobial resistance genes and analyse the genetic relatedness of organisms [12]. However, there are a several limitations, such as contamination, which may result in false positives; difficulty in obtaining high-quality samples; sampling time point affecting results; and the possibility of PCR inhibitors leading to false negatives [22]. These limitations may have contributed to the lower positive rate of NAAT in this study.

We found that the rate of macrolide resistance accounted for 86.2% of cases. MRMP strains have been increasing in many countries since the first case was reported in Japan in 2000 [23] with the long-term widespread use of macrolides. Macrolides act on the ribosomal 50S subunit to inhibit protein synthesis. Mutations in the V region of the MP 23S rRNA domain can cause a decrease in affinity between the drug and the ribosome, leading to drug resistance [24]. Common mutations, at both position 2063 and position 2064, lead to high-level resistance, whereas positions 2067 and 2617 are associated with low-level resistance [25, 26]. In these 262 specimens, we found a mutation only at position 2063. MRMP in the Beijing population reached a high rate of more than 90% from 2008 to 2012 [27], which was similar to our results.

MP-DNA detection by RT-PCR technology amplifies gene fragments to diagnose pathogens, which has advantages in operation and sensitivity of detection [28]. We compared the diagnostic values of different methods for M. pneumoniae with RT-PCR as the standard. We found that MP-IgM (PA) had a high sensitivity of 74.0% and the highest Youden index and Kappa value, indicating that it was conducive to screening for MP infection. SAT is a recently developed method based on isothermal amplification of RNA [29, 30], which can be completed in approximately 3 h. In our study, the outcomes of SAT for M. pneumoniae, both MP-DNA and 7 RNAs, showed higher specificity but lower sensitivity. Specific RNA exists only in the proliferation stage of M. pneumoniae, indicating that increases in RNA levels may reflect bacterial multiplication [31]. Previous data have shown that the SAT positivity rates are significantly higher in untreated cases with MPP than in macrolide-treated MPP cases [32]. Therefore, one explanation for the “false-negative” SAT results is that long-term treatment might decrease the M. pneumoniae load. Our data showed that 78.3% of the patients received macrolide treatment before hospitalization. Another explanation might relate to the possible poor sampling procedural skills and the quality of the swab samples, leading some samples to be below the assay’s detection limit [33, 34]. In addition, multiplex PCR assays can be used to detect M. pneumoniae and other respiratory pathogens, but monoplex assays have higher sensitivity and specificity than multiplex assays [35], which was confirmed by our results.

Each detection method of M. pneumoniae has its own advantages and disadvantages. Although molecular assays are superior in detecting M. pneumoniae due to their rapid, sensitive and specific characteristics, they cannot take the place of serology [36]. A study in China analysed data from children hospitalized with MPP using IgM (PA) and RT-PCR [37]. The concordance was close to 90% for the two detection methods. However, 173 (7%) children with a positive PCR had a negative serological test, and there were only 72 (3%) IgM-positive children who were PCR negative. Different detection methods can be used not only to improve diagnostic specificity and sensitivity but also to reduce the false-negative rate and false-positive rate. No single test available can be reliable for the identification of M. pneumoniae infection, and a combination of various methods is the most reliable approach [38]. A positive PCR without serological evidence of infection may indicate that the specimen was collected too early in the course of M. pneumoniae infection for antibodies to develop. PCR results may become negative after a period of antibiotic treatment, while serological results would remain positive for a long time. Some reviews have concluded that no single test could reliably detect M. pneumoniae infection, but a combination of serological tests and PCR might be the most sensitive approach for early diagnosis in children [12, 39]. In our research, two different M. pneumoniae detection methods were combined and evaluated by RT-PCR, and the results showed that MP-IgM (PA) in combination with MP-RNA (SAT) can be used as a good screening method.

Our study had some limitations. First, this study was performed in a single centre and included no data on clinical features because there were no accurate ways to determine the specificity of these characteristics. Second, the lack of a gold standard diagnostic assay, such as microbiological culture and a 4-fold increase in IgM and IgG in paired serum, made it difficult to draw conclusions to confirm the most reliable and accurate method.

In conclusion, during the acute phases of M. pneumoniae pneumonia in children, the detection of MP-IgM (PA) together with MP-RNA (SAT) proved to be a reliable and precise diagnosis. Paediatricians need to select appropriate detection methods based on the age of children and the onset of the disease, and they need to consider various results to diagnose M. pneumoniae infection.