Genotyping and refractory risk factors of mycoplasma pneumoniae pneumonia in Suzhou, China

Abstract

Objective

To investigate the genotyping and drug resistance of Mycoplasma pneumoniae (MP) epidemic strains in children and analyze the risk factors for refractory Mycoplasma pneumoniae pneumonia (RMPP).

Method

Nasopharyngeal aspirates (NPA) were collected from hospitalized children with MP infection from September to October 2023. Tracheoscopy was performed when necessary, and Bronchoalveolar Lavage Fluid (BALF) was collected. Polymerase Chain Reaction (PCR) capillary electrophoresis were used to detect respiratory pathogens, and a nested PCR based on the MP P1 gene monitored MP subtypes. The MP-23 S rRNA V region was amplified and sequenced. Clinical data and laboratory results were analyzed for RMPP risk factors.

Result

From 261 children diagnosed with Mycoplasma pneumoniae pneumonia (MPP), cough (100%) and fever (93.1%) were the most common symptoms. Moist rales (65.5%) were the prevalent pulmonary signs. Bronchoscopy was required in 44.4% of cases. Drug resistance genes were detected in 258 cases (98.9%). Genotyping showed 192 cases (73.56%) as type I and 69 cases (26.4%) as type II. Type I required more bronchoscope interventions (54.2%) compared to type II (17.4%, P < 0.001). Among the children, 92 (35.2%) were classified as RMPP. The RMPP group had longer hospitalization (9.14 ± 4.38 vs. 6.7 ± 1.81 days), higher fever ratio (100% vs. 89.9%), and longer febrile duration (8.72 ± 2.52 vs. 4.56 ± 2.50 days) (all P < 0.05). Higher rates of shortness of breath (7.6% vs. 1.2%) and decreased breath sounds (30.4% vs. 16.6%) were noted in the RMPP group (P < 0.05). Additionally, higher proportions of elevated CRP, ALT, AST, LDH, and D-D dimer were found in the RMPP group, along with a greater need for bronchoscopy (70.7% vs. 43.2%). Multivariate logistic regression identified total febrile duration, mucus plug formation, elevated AST, LDH, and D-D dimer as RMPP risk factors. Receiver operator characteristic (ROC) curve indicated that total febrile duration, LDH, and D-D dimer could serve as predictive markers for RMPP.

Conclusion

Type I predominated among MP strains in Suzhou, with a high prevalence of drug resistance. Type I infections were associated with a higher likelihood of requiring bronchoscopy. Prolonged fever, mucus plug formation, elevated AST, LDH, and D-D dimer were independent risk factors for RMPP, with total febrile duration, LDH, and D-D dimer serving as predictive markers.

Clinical trial number

Not applicable.

Background

Since its initial description in the 1940s and subsequent recognition as a highly evolved pathogenic bacterium, Mycoplasma pneumoniae (MP) has become a global cause of primary atypical pneumonia [1]. Today, MP is one of the most prevalent pathogens responsible for respiratory infections in children, particularly community-acquired pneumonia (CAP). Research suggests that MP may account for approximately 4 to 8% of cases of community-acquired bacterial pneumonia (CABP) during endemic periods. However, during epidemics, this organism can contribute to as much as 20 to 40% of CABP in the general population, reaching as high as 70% in enclosed settings [2, 3]. The incidence of Mycoplasma pneumoniae pneumonia (MPP) in children varies from 10 to 30% of community-acquired pneumonia cases [4, 5]. The incidence of refractory Mycoplasma Pneumoniae pneumonia (RMPP) is increasing year by year, and its mechanism is a focus of research. In MP infections, adhesion and colonization of host cells are mainly accomplished by adhesins, with the P1 gene playing a pivotal role. By encoding the P1 protein, the P1 gene facilitates MP adhesion and colonization on host cells. Additionally, the P1 protein serves as a crucial immunogen, eliciting a robust immune response from the host. Mutation of the P1 gene results in the absence of the encoded P1 protein, rendering MP non-virulent and non-toxic [6].

The aim of this study is to investigate the genotyping of MP strains and the expression of resistance genes in the Suzhou area. Additionally, we will explore the risk factors for children with MP infection progressing to RMPP to identify suitable predictive indicators. The goal is to enable early detection of children who may develop RMPP, allowing for timely intervention to shorten the duration of illness and reduce the occurrence of later complications.

Methods

Patients and specimen collection

This study is a prospective cross-sectional study. During the study period (September to October 2023), all participants’ parents or guardians provided written informed consent. Nasopharyngeal aspirates (NPA) were collected from the patients within 24 h, and patients were enrolled if their MP PCR test was positive. The NPA samples were tested for MP resistance genes and genotyping. Clinical data and laboratory results from admission were collected, including blood routine, liver and kidney function tests, coagulation routine, and pretransfusion examination. Chest radiography performed within 24 h prior to admission or within the subsequent 24 h was also collected. Some children requiring it underwent bronchoscopy and airway lavage.

Group: children with MPP who have received appropriate treatment with macrolide antibiotics for 7 days or more continue to experience fever, worsening clinical signs, and progressive lung imaging findings, or develop extrapulmonary complications [7] were classified into the RMPP group, while the remaining children were classified into the MPP group.

Nasopharyngeal aspirates

Nasopharyngeal aspirates were obtained from each patient using a sterile plastic catheter, gently inserted into the lower pharynx through the nasal cavity. These samples were used to detect common microorganisms.

Electronic bronchoscopy

Some children required bronchoscopy and lavage therapy, particularly those with significant atelectasis and pulmonary mass lesions that showed poor absorption after treatment. Before the procedure, parents were informed of the potential surgical risks, and their informed consent was obtained. Pediatric patients fasted from both solids and liquids for at least 6–8 h prior to the procedure. Intramuscular atropine sulfate, at a dose of 0.01–0.02 mg/kg, was administered as premedication. The procedure was conducted under the supervision of experienced anesthesiologists. An electronic bronchoscope (Olympus CV260, Tokyo, Japan) was inserted into each lobe. Foreign body forceps or brushes were used to clear any plastic sputum deposits, ensuring an unobstructed airway. The affected area was rinsed three times with 1 ml/kg of prewarmed sterile 0.9% saline solution, which was then collected using a sterile sputum-collecting pipe (Falcon 50 ml, Becton-Dickinson, Rutherford, NJ, USA). The collected bronchoalveolar lavage fluid (BALF) sample was used for etiological examination.

Microbiological analysis

Nasopharyngeal aspirates and BALF samples were tested for 11 types of viruses, along with MP and Chlamydia pneumoniae (CP). Bacteria were assessed by inoculating NPA and BALF samples onto blood plates and examining them after 18–20 h of incubation. Bacterial growth exceeding 10³ colony-forming units/ml was considered significant. Mycoplasma pneumoniae, CP, and viruses such as RespiratorySyncytial virus, Adenovirus, Influenza virus (A, B, H1N1, H3N2), Parainfluenza virus, Boca virus, Human rhinovirus, Human coronavirus, and Human metapneumovirus were detected by PCR using a 13 Respiratory Pathogen Multiplex Detection Kit (PCR Capillary Electrophoresis Fragment Analysis) (Hailshi Gene Technology Co., Ltd, Ningbo, China), following the manufacturer’s instructions.

Mycoplasma pneumoniae resistance gene detection

All samples underwent amplification using MP-23SrRNA V region PCR for gene detection. DNA extraction was performed with a DNA extraction kit (BioPerfectus, China). The concentration of the extracted DNA samples ranged from 80 to 200 ng/µL, with purity A values between 1.6 and 1.9, meeting the conditions for PCR amplification. Primers were designed based on sequences from GenBank and relevant literature, with synthesis by Sangon Biotech Co., Ltd (Shanghai, China). PCR amplification was carried out using specified reaction systems and conditions. The 10 µL PCR product was subjected to 1.5% agarose gel electrophoresis and stained with EB. Gel imaging analysis was conducted to interpret the electrophoresis results. The amplified product was sent to Sangon Biotech Co., Ltd (Shanghai, China). for sequencing and compared with the standard strain M129.

Mycoplasma pneumoniae genotyping by nested multiplex PCR

DNA was extracted from clinical samples and reference MP strains. MP strains M129 and FH were selected to represent strain types I and II, respectively. Genomic DNA extraction from these strains was performed using the DNA Purification Kit (B518267, Sangon Biotech Co., Ltd., Shanghai, China) following the manufacturer’s instructions. Primers targeting the two variable regions of RepMP4 and RepMP2/3 within the P1 gene were designed based on a previously described method [8] and synthesized by Sangon Biotech Co., Ltd. (Table 1). Nested PCR was then performed using the following primer pairs: ADH2-forward (F)/ADH2-reverse (R), ADH3F/ADH3R, ADH3/MP2/3-R1, and MP2/3-R2/MP2/3-F2 (Table 1). The PCR reaction mixture included 10X Buffer Concentrate (Shinegene, Shanghai, China; 4 µl), 25 mM MgCl₂ (3 µl), 10 mM dNTPs (Shinegene; 3 µl), 10 µM of forward and reverse primers (1 µl each), extracted DNA as a template (5 µl), and AmpliTaq (0.3 µl, Takara Biotechnology, Co., Ltd., Shanghai, China). The reaction volume was adjusted to 25 µl with DNase-free bi-distilled H₂O. PCR amplification commenced with an initial denaturation at 95℃ for 5 min, followed by 30 cycles at 95℃ for 30 s, 50℃ for 30 s, 72℃ for 2.5 min, and a final extension at 72℃ for 5 min. Amplification was conducted using a Bio-Rad real-time PCR amplifier (Bio-Rad Laboratories, Inc.).

Table 1 Polymerase chain reaction primer sequences of Mycoplasma pneumonia P1 genotypes

Three P1 type I and three P1 type II strains were randomly selected from nested multiplex PCR products. The amplified products were sent to Sangon Biotech Co., Ltd. for sequencing of the P1 gene. The sequenced genes were then compared with the P1 genes of reference strains M129 and FH, representing strain types I and II, respectively, archived in GenBank (https://www.ncbi.nlm.nih.gov/nuccore/) (Access Nos. M18639 and FJ215693). Comparisons were conducted using the Basic Local Alignment Search Tool (BLAST) database (https://blast.ncbi.nlm.nih.gov/Blast.cgi).

Statistical analysis

Measurement data were presented as the mean ± standard deviation (SD). Categorical data were expressed as percentages. Group comparisons were conducted using the t-test or Fisher’s exact test, as appropriate. Risk factors were analyzed using logistic regression, and ROC curves were generated. A P value < 0.05 was considered statistically significant.

Results

General information and MPP clinical manifestations

In this study, NPA or BALF were collected from 261 children admitted to Children’s Hospital of Soochow University and diagnosed with MPP. Among them, there were 134 males and 127 females, with ages ranging from 3 months 14 days to 15 years, and an average age of 5.99 ± 3.41 years.

Fever and cough were the most common clinical manifestations of MPP. All children presented with cough, while 244 cases (93.5%) had fever. Among these, 14 cases (5.4%) exhibited low-grade fever, 98 cases (37.5%) had moderate fever, and 132 cases (50.6%) had high fever. The average duration of fever was 6.03 ± 3.20 days, with 122 cases (46.7%) experiencing fever for 7 days or more. Other clinical manifestations included wheezing in 25 cases (9.5%) and shortness of breath in 9 cases (3.4%). The main pulmonary signs observed were moist rales in 171 cases (65.5%), wheezing sound in 42 cases (16.1%), and diminished breath sounds in 56 cases (21.5%).

Pulmonary imaging revealed pleural effusion in 33 cases (12.6%), atelectasis in 18 cases (6.9%), bronchiectasis in 8 patients (3.1%), and pulmonary cavities in 4 patients (1.5%).

Mycoplasma pneumoniae drug resistance

Genetic testing for drug resistance was conducted in all cases, revealing a total of 258 instances of drug resistance genes, with a detection rate of 98.9%. Only 3 cases tested negative for drug resistance genes. The clinical presentations of the 3 children with no detected drug resistance genes were relatively mild; their fever did not exceed 7 days, and bronchoscopy was unnecessary. Additionally, chest imaging did not reveal atelectasis, and AST and ALT levels were within normal ranges.

Mycoplasma pneumoniae classification and clinical manifestations

Mycoplasma pneumoniae genotyping was conducted in all cases, revealing 192 cases (73.6%) with type I and 69 cases (26.4%) with type II. All patients with type I exhibited drug resistance, whereas all 3 patients without drug resistance were classified as type II. There were no significant differences in age, sex, length of stay, or the proportion of RMPP between the two groups.

There were no significant differences in clinical manifestations such as fever, cough, wheezing, and shortness of breath. However, a slightly higher proportion of children with type II had a fever lasting more than 10 days, accounting for 18.8%, compared to 11.5% in type I cases. No significant differences were observed in pulmonary signs and imaging findings between the two groups.

In terms of laboratory examination, platelet elevation was more pronounced in children with type II, with a median platelet count of 336 × 10⁹/L, significantly higher than that of type I (300 × 10⁹/L). Although the proportion of CRP > 40 mg/dl and LDH elevation was slightly higher in type II children compared to type I children, the difference was not statistically significant.

Regarding treatment, the proportion of children requiring oxygen therapy was similar between the two groups. However, the proportion of children requiring bronchoscopic intervention was higher in type I cases. Specifically, bronchoscopic intervention was necessary in 54.2% of children in the type I group compared to only 17.4% in the type II group. Specific data are provided in Table 2.

Table 2 Clinical characteristics and laboratory examination of different types of MP

Clinical features of RMPP

According to the RMPP definition, the 261 children enrolled in this study were categorized into the RMPP group or the MPP group, comprising 92 (35.2%) and 169 (64.8%) cases, respectively. The average duration of illness and fever before admission was significantly higher in the RMPP group compared to the MPP group (8.94 ± 2.83 vs. 7.75 ± 3.39, 7.60 ± 2.95 vs. 4.25 ± 2.29, P < 0.05). However, there were no significant differences in the proportion of macrolide antibiotic use and the duration of use before admission between the two groups.

In the RMPP group, the length of hospital stay, fever incidence, total febrile duration, and occurrence of thermal spikes were significantly higher compared to the MPP group, with statistically significant differences observed. Additionally, the proportion of shortness of breath and diminished breath sounds was significantly higher in the RMPP group. Chest imaging revealed a higher prevalence of atelectasis in the RMPP group. Furthermore, CRP, ALT, AST, LDH, and D-D dimer levels were significantly elevated in the RMPP group compared to the MPP group. The proportion of mucus plugs observed during bronchoscopy was also significantly higher in the RMPP group. Although the proportion of mixed infections was higher in the RMPP group than in the MPP group, the difference was not statistically significant in univariate analysis. Moreover, the RMPP group exhibited a higher percentage of patients requiring bronchoscopy and specialized treatment. Detailed data are provided in Table 3.

Table 3 Comparison of clinical features and laboratory tests of RMPP and MPP

Risk factors and prediction of RMPP

Using multivariate logistic regression analysis to assess the relationship between indicators that showed significant differences in univariate analysis (P < 0.05) and the occurrence of RMPP, specifically including: duration of illness before admission, duration of fever before admission, fever, high fever, shortness of breath, total febrile duration, low breath sound, pleural effusion, CRP > 40 mg/dl, AST > 44U/L, ALT > 30U/L, LDH > 420U/L, D-Dimer > 550ug/L, and mucus plug formation. Ultimately, the total febrile duration, AST > 44U/L, LDH > 420U/L, D-Dimer > 550ug/L, and mucus plug formation were identified as independent risk factors for RMPP. The remaining indicators were not included in the model. The risk of developing RMPP in children with AST greater than 44 U/L is 7.422 times higher than in those with lower AST levels. For those with LDH greater than 420 U/L, the risk increases to 5.916 times compared to children with lower LDH levels. Additionally, children with D-Dimer levels exceeding 550 ug/L face a risk that is 2.188 times higher than those with lower levels. Furthermore, the likelihood of progressing to RMPP in children with mucus plugs is 6.815 times greater than in those without. Detailed data are presented in Table 4.

Table 4 Multivariate logistic regression of risk factors for RMPP

Receiver operating characteristic curve analysis was performed on the aforementioned measurement data, revealing that total febrile duration, LDH, and D-D dimer exhibited better predictive performance, with AUC values of 0.884, 0.648, and 0.689, respectively. The optimal cutoff point for total febrile duration was determined to be 6.5 days, with a sensitivity and specificity of 90.2% and 76.9%, respectively. For LDH, the optimal cutoff point was 372.8 U/L, resulting in a sensitivity of 38.6% and specificity of 88.0%. Similarly, the optimal cutoff point for D-D dimer was 720 ug/L, with a sensitivity and specificity of 48.8% and 83.5%, respectively. Detailed data are presented in Table 5; Fig. 1.

Table 5 ROC curve of RMPP

Fig. 1

ROC curve of RMPP

Discussion

Mycoplasma Pneumonia is currently highly prevalent in China, yet the mechanisms driving fluctuations in its incidence remain unclear. It has been suggested that shifts in the proportion of strains with specific P1 types or concurrent increases in the incidence of several strains may lead to epidemics or changes in immunity. Furthermore, it is believed that the genotype of MP may be undergoing changes, resulting in diverse genetic material in each epidemic. A recent study reported the detection of polyclonal strains within a single epidemic [9]. In 2008, Kenri et al. [10] established a nested multiplex PCR typing method based on the P1 gene sequence, which effectively detects MP strains of type I, II, and Variant 2 (including V2a and V2b). The P1 typing of MP may vary between different years. Cousin observed that MP protein typing could alternate across different years. Pereyre noted a steady increase in the proportion of type 2 since 1996 [10, 11]. In Japan, type switching occurs approximately every 8–10 years, with a transition period of 2–3 years from one type to another [8]. A decade ago, our team studied the typing of MP in the Suzhou area. We collected respiratory secretion specimens from 313 children with acute respiratory tract MP infections and found that 303 cases (97.12%) were type I, 8 cases (2.56%) were type II, and 1 was V2 variant [12]. In this study, among 261 MP samples, 192 cases (73.56%) were type I, and 69 cases (26.4%) were type II. The incidence of type II has increased in recent years in the Suzhou area, but type I still predominates.

The effect of different types of MP on clinical manifestations remains inconclusive. Nilsson found no significant difference in disease severity between subtypes [13]. Zhang observed that all type 2 children exhibited severe symptoms and complications with liver function damage, although the insufficient number of cases precluded statistical significance [14]. Our previous study indicated that children with type 1 are more susceptible to extra-pulmonary complications and have a higher incidence of RMPP. In this study, no significant difference in clinical manifestations was observed among the different types, although the proportion of type 1 children requiring tracheoscopy was higher.

In recent years, with the widespread use of antibiotics, MP has developed a certain level of tolerance to them. Macrolide antibiotics target the V region of the 23 S rRNA of MP ribosomes, forming covalent bonds that inhibit peptide bond formation. Previous studies have outlined several mechanisms of MP resistance to macrolide antibiotics: (1) Gene mutations in the V region of 23 S rRNA that prevent covalent bonding with macrolides, (2) Efflux mechanisms, and (3) Methylating passivating enzymes [15]. Our team conducted drug resistance gene testing on MP samples collected from 2013 to 2014, revealing a detection rate of 92.5%. In this current study, we detected only 3 cases of children without drug resistance genes, indicating a drug resistance rate as high as 98.9%, higher than that observed a decade ago.

Research has shown that MP with drug-resistant genes tends to cause longer fever durations and pre-admission illness courses compared to those without drug resistance. Additionally, these children require steroid treatment more frequently and for longer periods and are more prone to mixed infections [16]. In this study, although the clinical manifestations of the three children in the non-drug-resistant group appeared milder, the small sample size prevents definitive conclusions about clinical significance.

Furthermore, despite the high prevalence of drug-resistant genes in this region, most children responded effectively to macrolide antibiotics in clinical treatment. Notably, 166 children with positive drug-resistant genes did not develop RMPP. This may be partly related to the body’s complex immune system, as there is a discrepancy between in vitro resistance and in vivo efficacy. Additionally, MP has a self-limiting nature, and the clinical use of corticosteroids may inhibit the immune response induced by MP in the body.

Common clinical manifestations of MP infection include fever and a persistent dry cough, with severe cases potentially presenting with shortness of breath. In this study, all children exhibited coughing, and over 90% had a fever. Chest imaging findings can vary, but severe cases often show extensive inflammation, atelectasis, pleural effusion, and may even develop necrotizing pneumonia or obstructive bronchiolitis [17]. Although PCR is currently the fastest and most sensitive testing method, it can yield false-negative and false-positive results [18]. Additionally, in outpatient settings, it is often not feasible to wait for definitive pathogen results. Therefore, this study provides a basis for the early clinical diagnosis and empirical treatment of MPP.

The proportion of severe cases and RMPP has gradually increased in recent years, drawing significant interest from scholars due to its serious clinical manifestations and complications. Researchers have focused on identifying risk factors associated with RMPP to find relevant clinical indicators for early warning and intervention, ultimately reducing the occurrence of later complications. Studies have indicated that several factors are significantly correlated with the occurrence of RMPP, including a fever duration of ≥ 10 days, pleural effusion, extrapulmonary complications, lung consolidation involving more than two-thirds of the lung, and CRP levels > 40 mg/L [19]. Other studies have shown that IL-6, IL-10, LDH, PT, and D-dimer can be used as predictive indicators for RMPP [20]. In our study, we found that prolonged febrile duration, mucus plug formation, and elevated AST, LDH, and D-D dimer levels were all risk factors for RMPP. Further investigation revealed that total febrile duration, LDH, and D-D dimer levels had better predictive value for RMPP. Febrile duration > 6.5 days, LDH levels > 372.8 U/L, and D-D dimer levels > 720 ug/L were identified as optimal prediction thresholds.

Conclusion

In the Suzhou area, type I predominated among MP cases, with a high proportion exhibiting drug resistance genes. Notably, more children with type I required tracheoscopic intervention. Additionally, prolonged fever, mucus plug formation, and elevated levels of AST, LDH, and D-D dimer were identified as independent risk factors for RMPP. Total febrile duration, LDH, and D-D dimer levels emerged as effective predictors of RMPP.