Abstract
Background
This study aimed to evaluate whether Mycoplasma pneumoniae resistance gene detection can independently guide antibiotic therapy for Mycoplasma pneumoniae pneumonia in children and to identify key predictors for antibiotic adjustment.
Methods
We conducted a retrospective cohort study of children with Mycoplasma pneumoniae pneumonia who underwent resistance gene testing. Participants were recruited from two centers (Affiliated Hospital of North Sichuan Medical College and People's Hospital of Nanbu County) between January 2023 and October 2025. Resistance gene testing results served as the basis for assigning patients to either the positive or negative group. The expression of Mycoplasma pneumoniae resistance genes was analyzed. The positive group was further divided into a “switch group” (switched to doxycycline) and a “maintenance group” (continued azithromycin). We performed a comparative analysis of demographic, clinical, and laboratory parameters across the groups. Multivariate logistic regression identified factors associated with switching antibiotics, and ROC curves assessed predictive performance.
Results
A total of 150 children were enrolled, with 13 in the macrolide resistance gene-negative group and 137 in the positive group, resulting in a macrolide resistance gene positivity rate of 91.3%. Of these patients, 65 patients were in the switch group and 72 patients were in the maintenance group. Significant differences were observed in baseline characteristics (age), inflammatory markers (CRP, PCT, lymphocyte count), disease severity (severe MPP, pleural effusion, bronchoscopy), and clinical management (peak fever, hospital stay, antibiotic duration) (P < 0.05). Logistic regression analysis identified increased age, elevated CRP, and prolonged hospital stay as independent predictors for switching to second-line antibiotics in macrolide-resistant Mycoplasma pneumoniae.
Conclusion
In the Nanchong area, macrolide resistance mediated by the 23S rRNA A2063G mutation predominates. Resistance gene detection alone indicates only potential resistance but cannot independently guide antibiotic switching. Among gene-positive children, elevated CRP and older age are early predictors of switching need, while prolonged hospital stay serves as a retrospective confirmatory marker. Clinical decisions should integrate these biomarkers rather than relying solely on genetic results.
Keywords: age, azithromycin, c-Reactive protein, doxycycline, hospital stay, macrolide-resistant mycoplasma pneumoniae, predictive factor, resistance gene
Introduction
The incidence of Mycoplasma pneumoniae (M. pneumoniae) infection exhibits marked geographical heterogeneity in attack rates, and epidemics follow a cyclic pattern occurring roughly every 3–5 years (1, 2). Since 2023, there has been a marked global increase in the incidence of M. pneumoniae infections. In China, it has emerged as a predominant pathogen responsible for community-acquired pneumonia (CAP) in children, significantly impacting both the epidemiological landscape of pediatric respiratory infections and related clinical management strategies (3–5).
Macrolides are the first-line treatment for Mycoplasma pneumoniae pneumonia (MPP), with azithromycin being the preferred choice in China (6). However, the widespread use of macrolides has led to increasing resistance, altering the drug susceptibility profile (7). The primary resistance mechanism of macrolide-resistant Mycoplasma pneumoniae (MRMP) involves mutations at different sites of the 23S rRNA gene (8). These mutations confer varying degrees of resistance; for instance, the A2063G mutation often results in resistance to both 14-membered (e.g., erythromycin) and 15-membered (e.g., azithromycin) macrolides (9, 10). MRMP resistance rates differ significantly by region, with higher rates in East Asia than in Europe and America, and exceed 90.0% in China (11–17).
Infections with MRMP significantly increase the risk of severe Mycoplasma pneumoniae pneumonia (SMPP), leading to prolonged fever, hospital stays, and antibiotic use, and complicating treatment decisions and increasing costs (18, 19). Doxycycline has been suggested as a first-line alternative for confirmed MRMP cases (16). However, its use in children is limited by the risk of permanent tooth discoloration and enamel hypoplasia and other adverse effects (20, 21).
Thus, early identification of macrolide resistance is crucial for guiding clinical decisions and optimizing treatment strategies (22). In this context, Targeted Next-Generation Sequencing (t-NGS) can analyze the genetic information of pathogens to identify resistance and virulence genes (23, 24). The contributions of resistance genes from different strains to the phenotypic resistance vary. Existing studies have not yet fully elucidated the causal relationship between the detection of resistance genes and clinical resistance, and the detection of resistance genes does not necessarily predict treatment failure with macrolides (25, 26). The epidemiological characteristics of MRMP in children in Nanchong are not well defined. This study aimed to bridge this evidence gap and assess the necessity of switching to second-line drugs and the independent predictors for using new-generation tetracyclines based on detected resistance genes.
Methods
Study design and population
A retrospective analysis was conducted on children diagnosed with MPP and who underwent resistance gene testing from January 2023 to October 2025 at the Affiliated Hospital of North Sichuan Medical College in Nanchong and the People's Hospital of Nanbu County. A positive result for resistance gene testing was defined as the presence of a mutation in the mycoplasma resistance gene. Clinical and testing data of the children were collected, with the following inclusion criteria: (1) completion of t-NGS testing during hospitalization, (2) age between 29 days and 14 years, and (3) all children were diagnosed with MPP based on the Guidelines for the Diagnosis and Treatment of MPP in Children (2023 Edition) (6). Exclusion criteria included: (1) pre-existing chronic conditions-autoimmune disorders, congenital cardiac anomalies, or bronchopulmonary maldevelopment, and (2) incomplete clinical data.
Detection of drug resistance and selection of antibiotics
All analyzed samples were clinical residual specimens processed by a collaborating laboratory using standardized protocols for high-throughput targeted sequencing, with detection and bioinformatics methods based on prior studies (27). This was a retrospective cohort study using electronic medical records. Antibiotic therapy decisions were made by attending physicians based on clinical judgment and local treatment guidelines, not by study protocol. Treatment data were extracted from routine clinical documentation, reflecting real-world practice patterns.
Clinical data collection
Clinical information was retrieved from the electronic hospital charts of admitted children. The data included gender, age, duration of fever, fever peak, type and duration of antibiotic use, length of hospital stay, presence of lung consolidation, pleural effusion, atelectasis, initial white blood cell count (WBC), C-reactive protein (CRP), absolute neutrophil count, absolute lymphocyte count, procalcitonin (PCT), albumin (ALB), alanine aminotransferase (ALT), total bile acids (TBA), total bilirubin (TBIL), lactate dehydrogenase (LDH), prognosis, and results of t-NGS.
Statistical analysis
Data were analyzed using SPSS version 27.0.1.0(IBM Corp., Armonk, NY, USA). Non-normally distributed continuous variables are presented as medians and interquartile ranges (IQR), and the Mann–Whitney U-test was used. Categorical variables were analyzed using the χ2-test or Fisher's exact test. Multivariable logistic regression was employed to identify risk factors for switching antibiotics. A combined predictive indicator was developed using multivariable logistic regression. Model discrimination was assessed by receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC). A P-value < 0.05 was considered statistically significant.
Ethics approval and consent to participate
This study was approved by the Medical Ethics Committees of the Affiliated Hospital of North Sichuan Medical College in Nanchong and the People's Hospital of Nanbu County in 2025 (2025ER74-1, NYLS20250102). Informed consent was waived by the ethics committees of both participating hospitals, as the study involved only anonymized, pre-existing clinical data without additional patient risk.
Results
Detection of MPP drug resistance genes
This study included 150 children with MPP who underwent t-NGS testing during hospitalization at two centers between January 2023 and October 2025. Of these, resistance gene testing identified 137 (91.3%) children with the A2063G mutation and 13 (8.7%) without macrolide resistance mutations (Figure 1). This 23S rRNA alteration confers macrolide resistance. Antibiotic selection (azithromycin vs. doxycycline) did not differ significantly between resistance-positive and -negative groups (Table 1). Of the 137 children with macrolide resistance genes, 72 (52.6%) continued azithromycin and 65 (47.4%) were switched to doxycycline (Table 1). These 137 children constituted the primary study population (Table 2); their median age was 5.58 years (IQR 3.96–7.50 years), 56.2% were female, and 54.7% had severe MPP. Clinical outcomes were satisfactory in 100.0% (72/72) of the maintenance group and 96.9% (63/65) of the switch group, with no significant difference between groups (P = 0.223) (Table 2).
Figure 1.
Analysis of Mycoplasma resistance gene detection of 150 children who underwent targeted next-generation sequencing (t-NGS), 13 (8.7%) were negative for macrolide resistance mutations and 137 (91.3%) were positive. All positive cases harbored the A2063G mutation in the 23S rRNA gene. The right panel shows the distribution within the positive group.
Table 1.
Comparison of antibiotic therapy strategies by macrolide resistance gene Status in 150 children with Mycoplasma pneumoniae pneumonia.
| Variables | Total (n = 150) | Negative group (n = 13) | Positive group (n = 137) | Statistic | P |
|---|---|---|---|---|---|
| Antibiotic, n (%) | χ2 = 0.385 | 0.535 | |||
| Azithromycin | 80 (53.3%) | 8 (61.5%) | 72 (52.6%) | ||
| Doxycycline | 70 (46.7%) | 5 (38.5%) | 65 (47.4%) |
χ2, Chi-square test.
Table 2.
Clinical characteristics of 137 children with macrolide-resistant Mycoplasma pneumoniae pneumonia by antibiotic treatment strategy.
| Variables | Total (N = 137) | Maintenance group (n = 72) | Switch group (n = 65) | Statistic | p |
|---|---|---|---|---|---|
| Age, M (Q1, Q3) | 5.58 (3.96,7.50) | 4.50 (2.31, 7.13) | 6.58 (4.79, 8.71) | z:4.005 | <0.001 |
| CRP, M (Q1, Q3) | 14.60 (3.85,28.29) | 5.13 (1.62, 12.33) | 25.29 (17.58, 34.91) | z:6.589 | <0.001 |
| WBC, M (Q1, Q3) | 8.28 (6.61,10.71) | 8.58 (6.84, 10.77) | 8.24 (6.54, 10.45) | z:0.662 | 0.508 |
| Neutrophil. count, M (Q1, Q3) | 5.03 (3.85,6.89) | 4.75 (3.40, 6.91) | 5.42 (4.35, 6.85) | z:1.414 | 0.157 |
| Lymphocyte. count, M (Q1, Q3) | 2.25 (1.54,3.43) | 2.58 (1.79, 4.32) | 2.00 (1.31, 2.89) | z:2.849 | 0.004 |
| ALB, M (Q1, Q3) | 45.50 (41.90,48.20) | 45.25 (42.05, 48.10) | 46.10 (41.65, 48.45) | z:0.580 | 0.562 |
| ALT, M (Q1, Q3) | 16.00 (11.00,22.10) | 16.80 (12.00, 22.82) | 16.00 (10.50, 20.00) | z:1.111 | 0.266 |
| TBA, M (Q1, Q3) | 4.50 (3.00,7.60) | 4.80 (3.10, 7.90) | 4.20 (2.90, 6.70) | z:1.149 | 0.251 |
| TBIL, M (Q1, Q3) | 5.70 (4.55,7.50) | 5.70 (4.62, 7.50) | 5.70 (4.40, 7.50) | z:0.231 | 0.818 |
| PCT, M (Q1, Q3) | 0.07 (0.05,0.15) | 0.07 (0.05, 0.12) | 0.11 (0.05, 0.21) | z:2.364 | 0.018 |
| LDH, M (Q1, Q3) | 288.00 (248.00,359.00) | 291.00 (252.00, 341.50) | 285.00 (238.50, 372.50) | z:0.123 | 0.902 |
| Fever top, M (Q1, Q3) | 39.30 (38.50,40.00) | 39.00 (37.97, 39.60) | 39.50 (39.00, 40.00) | z:2.927 | 0.003 |
| Duration of fever, M (Q1, Q3) | 5.00 (1.00,6.00) | 4.00 (0.56, 6.00) | 5.00 (2.00, 7.00) | z:1.836 | 0.066 |
| Hospital stay, M (Q1, Q3) | 9.00 (8.00,11.50) | 9.00 (7.00, 11.00) | 10.00 (9.00, 12.00) | z:3.224 | 0.001 |
| Antibiotic duration, M (Q1, Q3) | 9.00 (7.00,11.00) | 9.00 (6.25, 10.00) | 10.00 (8.00, 12.00) | z:2.375 | 0.018 |
| Sex, n (%) | χ2:0.724 | 0.395 | |||
| 1 | 60 (43.8%) | 34 (47.2%) | 26 (40.0%) | ||
| 2 | 77 (56.2%) | 38 (52.8%) | 39 (60.0%) | ||
| Mixed infection, n (%) | χ2:0.592 | 0.442 | |||
| 0 | 42 (30.7%) | 20 (27.8%) | 22 (33.8%) | ||
| 1 | 95 (69.3%) | 52 (72.2%) | 43 (66.2%) | ||
| SMPP, n (%) | χ2:4.864 | 0.027 | |||
| 0 | 62 (45.3%) | 39 (54.2%) | 23 (35.4%) | ||
| 1 | 75 (54.7%) | 33 (45.8%) | 42 (64.6%) | ||
| Prognosis, n (%) | – | 0.223 | |||
| 0 | 135 (98.5%) | 72 (100.0%) | 63 (96.9%) | ||
| 1 | 2 (1.5%) | 0 (0.0%) | 2 (3.1%) | ||
| Pleural effusion, n (%) | χ2:7.131 | 0.008 | |||
| 0 | 117 (85.4%) | 67 (93.1%) | 50 (76.9%) | ||
| 1 | 20 (14.6%) | 5 (6.9%) | 15 (23.1%) | ||
| Pulmonary atelectasis, n (%) | χ2:0.247 | 0.619 | |||
| 0 | 127 (92.7%) | 68 (94.4%) | 59 (90.8%) | ||
| 1 | 10 (7.3%) | 4 (5.6%) | 6 (9.2%) | ||
| Lung consolidation, n (%) | χ2:1.917 | 0.166 | |||
| 0 | 78 (56.9%) | 45 (62.5%) | 33 (50.8%) | ||
| 1 | 59 (43.1%) | 27 (37.5%) | 32 (49.2%) | ||
| Bronchoscopy, n (%) | χ2:5.227 | 0.022 | |||
| 0 | 56 (40.9%) | 36 (50.0%) | 20 (30.8%) | ||
| 1 | 81 (59.1%) | 36 (50.0%) | 45 (69.2%) |
Z, Mann–Whitney test; χ2, Chi-square test; -, Fisher exact; M, Median; Q1 1st Quartile; Q3, 3st Quartile; WBC, white blood cell count; CRP, C-reactive protein; PCT, procalcitonin; LDH, lactate dehydrogenase; ALB, albumin; ALT, alanine aminotransferase; TBA, total bile acids; TBIL, total bilirubin; SMPP, severe Mycoplasma pneumoniae pneumonia. Sex: 1 = male, 2 = female; SMPP: 0 = general MPP, 1 = severe MPP; Prognosis: 0 = satisfactory, 1 = unsatisfactory; Mixed infection, Pleural effusion, Pulmonary atelectasis, Lung consolidation, Bronchoscopy: 0 = no, 1 = yes.
Clinical characteristics and antibiotic adjustment strategies in children with macrolide-resistant MPP
Among the 137 children with macrolide resistance genes, 72 continued azithromycin therapy (maintenance group), while 65 were switched from azithromycin to doxycycline (switch group). Compared with the maintenance group, children in the switch group were older and had longer hospital stays, as well as higher fever peaks, CRP, and PCT levels. The switch group also had higher proportions of prolonged hospital stay, antibiotic use duration, severe MPP, pleural effusion, and bronchoscopy requirement, with all differences being statistically significant (all P < 0.05). After rigorous statistical evaluation, no evidence of significant inter-group divergence was detected for any of the following variables: patient gender, duration of fever, complete blood count (WBC, neutrophil count), liver function (ALB, ALT, TBA, TBIL), LDH, or the incidence of atelectasis, lung consolidation, mixed infections, and ultimate clinical prognosis (all P > 0.05) (Table 2).
Risk factors for switching from azithromycin to doxycycline
To identify independent predictors of antibiotic switching, we first performed univariate analysis (Table 2). Variables with P < 0.05 and/or clinical relevance were subsequently entered into a multivariate logistic regression model. After adjusting for potential confounders, older age (OR 1.228, 95% CI 1.049–1.438, P = 0.011), elevated CRP (OR 1.061, 95% CI 1.024–1.099, P = 0.001), and prolonged hospital stay (OR 1.310, 95% CI 1.043–1.646, P = 0.020) remained independently associated with an increased risk of antibiotic switching (Figure 2).
Figure 2.
Multivariate logistic regression analysis of factors associated with antibiotic switching in children with macrolide-resistant Mycoplasma pneumoniae. Squares represent odds ratios and horizontal lines represent 95% confidence intervals. The dashed vertical line indicates the null value (OR = 1.0). SMPP, severe Mycoplasma pneumoniae pneumonia.
Predictors of second-line therapy escalation in MRMP: age, hospital stay and CRP
A receiver operating characteristic (ROC) curve analysis was performed to assess the predictive performance, either individually or in combination, of age, hospital stay, and CRP levels, for the need to switch to second-line drug therapy in children with MRMP. The area under the curve (AUC) values were 0.699 for age, 0.659 for hospital stay, 0.827 for CRP, and 0.860 for the combined model of all three factors, with higher values of each parameter associated with increased likelihood of requiring second-line drug therapy. Notably, the combined model of the three factors did not show a significant difference compared to CRP alone (Z/P = 1.332/0.183). (Table 3, Figure 3).
Table 3.
Predictive performance of Age, hospital stay, and C-reactive protein for antibiotic switching in macrolide-resistant Mycoplasma pneumoniae pneumonia.
| Variable | AUC (95%CI) | Cut-off | Sensitivity | Specificity | Youden J | P |
|---|---|---|---|---|---|---|
| Age | 0.699 (0.611–0.786) | 4.085 | 0.923 | 0.444 | 0.368 | <0.001 |
| Hospital stay | 0.659 (0.569–0.749) | 7.500 | 0.892 | 0.361 | 0.253 | <0.001 |
| CRP | 0.827 (0.752–0.901) | 12.785 | 0.877 | 0.778 | 0.655 | <0.001 |
| Age-Hospital stay-CRP | 0.860 (0.795–0.925) | 0.393 | 0.908 | 0.750 | 0.658 | <0.001 |
AUC, area under the curve; CI, confidence interval; CRP, C-reactive protein; MRMP, macrolide-resistant Mycoplasma pneumoniae; Age-Hospital stay-CRP refers to the combined model.
Figure 3.
ROC curves for predicting antibiotic switching in children with macrolide-resistant Mycoplasma pneumoniae. The curves show the predictive performance of age (cyan), C-reactive protein (CRP; yellow), hospital stay (orange), and the combined model of all three factors (blue). The diagonal dashed line represents the reference line (AUC = 0.50). ROC, Receiver operating characteristic; AUC, area under the curve; 95% CI, 95% confidence interval.
Discussion
Global macrolide susceptibility among M. pneumoniae isolates has declined progressively, with antimicrobial resistance emerging as a major public health concern—particularly in the Asia-Pacific region (28). Previous studies have documented the emergence of MRMP as the dominant clinical lineage in China, accounting for more than 90.0% of all documented infections (29, 30). Our study conducted in Nanchong identified the A2063G transition in domain V of the 23S rRNA gene in 91.3% of children with MPP, aligning closely with the national average and reaffirming the high prevalence of MRMP across diverse geographical regions in China.
Although all mutant strains exhibited macrolide resistance, more than 50.0% of the children still achieved a good response to azithromycin, with only 47.4% (65/137) switching to doxycycline. Multivariable logistic regression analysis demonstrated that age, serum C-reactive protein (CRP) concentration, and length of hospital stay were independent determinants governing the decision to escalate from azithromycin to doxycycline. Among these covariates, CRP exhibited strong discriminative power; its continuous elevation not only reflected the intensity of inflammation but also constituted a quantifiable and readily available biomarker that critically guides antimicrobial upgrading in children with MRMP.
This study was based on targeted next-generation sequencing detection, covering known drug resistance sites such as A2063G, A2064G, A2067G, C2617G and L4/L22. The A2063G mutation was the only resistance gene detected, consistent with findings by Zhou et al. (31). Their study recommended switching to doxycycline upon detection of this gene locus. In contrast, we found that a genotype-positive result alone was insufficient to independently guide antibiotic selection, a conclusion aligning closely with data published by Deng et al. (19). This discrepancy may be attributed to several factors. The A2063G mutation increases the minimum inhibitory concentration (MIC) of azithromycin against M. pneumoniae by at least 1,000-fold. However, satisfactory clinical efficacy may still be achieved when the mutant strain exhibits a low MP-DNA load or effectively stimulates the host's immune clearance capacity. This mechanism explains the observed phenomenon of “genetic resistance ≠ clinical resistance” (32, 33). Furthermore, the cohort enrolled by Zhou et al. comprised adults, whereas our investigation exclusively involved children. In children under 8 years of age, tetracycline-class antibiotics have a series of well-documented adverse reactions, including enamel hypoplasia, permanent tooth discoloration, allergic reactions, and gastrointestinal dysfunction (34, 35). Therefore, switching to tetracycline antibiotics in children warrants careful consideration even with positive resistance gene testing.
Among children uniformly carrying the A2063G resistance-associated substitution, no statistically significant difference in clinical outcome was observed between the switch and maintenance groups (P > 0.05). Specifically, 52.6% (72/137) of the children with resistance genes remained on azithromycin, all of whom achieved satisfactory recovery. This finding validates previous research, confirming that macrolides retain durable therapeutic value even in cases of documented MRMP (26, 36, 37). Notably, although all children carried the resistance gene, nearly half (45.8%) of the maintenance group had severe MPP (SMPP), yet still achieved good clinical outcomes with azithromycin alone (P > 0.05). Even in children with SMPP who were positive for drug resistance, the decision to upgrade to second-line antibiotics should be based on individualized assessment rather than automatically triggered by resistance gene results or disease severity alone.
As children grow older, the risk of resistant M. pneumoniae infections increases, prolonging fever duration and reducing the efficacy of macrolides (26, 38, 39). As pediatric age increases, the contraindication risk associated with tetracyclines declines, weakening the age-related threshold for antibiotic escalation and partly explaining why chronological age serves as an independent predictor of switching to second-line therapy. Previous studies have shown that patients with MRMP have significantly higher CRP levels than those with macrolide-susceptible Mycoplasma pneumoniae (MSMP), which are positively correlated with MP-DNA copy numbers and closely related to disease severity (15, 39–42). This correlation suggests that CRP may serve as a clinically accessible surrogate for bacterial burden in MRMP, partially explaining its strong predictive performance (AUC 0.827) for antibiotic switching in our cohort—even in the absence of direct MP-DNA measurement. Consistent with our findings, higher CRP levels in the switch group indicate a higher likelihood of severe infection and clinical resistance. In the study by Deng et al. (19), elevated CRP and PCT were identified as independent risk factors for switching to second-line antibiotics in MRMP patients. We found that CRP has a higher predictive value for switching antibiotics than age and hospital stay, but the difference in predictive efficacy when combined with these three factors is not significant, indicating that CRP is the core information source driving the decision to switch antibiotics. The classic bacterial marker PCT can guide antibiotic use (43, 44). Unlike the study by Deng et al. (19), PCT did not emerge as an independent risk predictor in our analysis, possibly due to inter-study heterogeneity in sampling time points and baseline characteristics. Importantly, about 40.0% of the switch group in Deng's study had MSMP, while our study excluded this confounder and only included MRMP patients, making our finding more illustrative of the risk factors for switching in MRMP specifically. In M. pneumoniae infections, serum PCT levels in patients are usually not significantly elevated. This is because M. pneumoniae lacks a cell and does not contain lipopolysaccharide (LPS), primarily activating the immune response via toll-like receptor 2 (TLR2) rather than the toll-like receptor 4 (TLR4) pathway, which has a relatively limited capacity to induce the release of cytokines such as interleukin-6 (IL-6), resulting in low PCT synthesis in extrathyroidal tissues (45, 46). This further confirms that PCT did not enter the switch prediction model, which is consistent with the biological characteristics of mycoplasma infections.
Hospitalization duration directly reflects disease severity (47). Children with SMPP and positive resistant strains exhibit more complex antibiotic treatment needs compared to general M. pneumoniae pneumonia (GMPP). We found a higher proportion of children with SMPP in the switch group compared to children with GMPP (P < 0.05). Disease severity influences clinical decision-making. Studies by Yan et al. (48) and Liu et al. (49) indicated that MRMP leads to higher bacterial loads and exacerbated immune responses, significantly increasing the likelihood of SMPP when clinical isolates show macrolide resistance. Moreover, SMPP has higher inflammatory markers and is more likely to be complicated by pleural effusion, lung consolidation and necrosis, pulmonary embolism, and diffuse alveolar hemorrhage (50, 51). A meta-analysis indicated that selecting second-line drugs for alternative therapy in SMPP can improve clinical efficacy and alleviate clinical symptoms (52). However, length of hospital stay, while associated with the need for switching treatments, is essentially a post-hoc measure reflecting the cumulative impact of disease severity and treatment response. Its limited utility for early prediction is consistent with its lower discriminant value (AUC 0.659), rendering it more an evolutionary marker than an actionable early predictor. This association with hospital stay, while significant, is of retrospective descriptive value only and should not inform early decision-making. Therefore, children with mild disease or low inflammatory markers can continue azithromycin treatment under close monitoring, while older children and those with elevated CRP levels should be prioritized for switching to sensitive antibiotics. As a routine laboratory biomarker, CRP exhibits robust predictive utility for antimicrobial selection; its early elevation enables clinicians to promptly detect clinically relevant MRMP infections and to switch antibiotics without delay, thereby abbreviating illness duration and reducing the risk of complications.
This investigation is subject to several inherent limitations. Its retrospective design predisposes the cohort to both selection bias and prescription bias. Furthermore, the study did not measure bacterial load (MP-DNA), which is a known correlate of resistance. Follow-up was limited to discharge, lacking data on the side effects of tetracycline-class drugs and long-term pulmonary function monitoring. Future research should conduct multicenter prospective randomized controlled trials to verify the efficacy and cost-effectiveness of switching antibiotics guided by immediate detection of resistance genes, establish a combined prediction model incorporating bacterial load and inflammatory markers, and conduct long-term follow-up on children's tooth development and pulmonary function.
Conclusions
In the context of high resistance rates, the predominant resistance type of MPP in the Nanchong area is macrolide resistance. Age and CRP are early-accessible predictors for guiding initial antibiotic decisions in children with MRMP, while prolonged hospital stay serves as an evolutionary marker that retrospectively confirms the need for switching. CRP exhibits the strongest discriminative performance and should be prioritized in clinical assessment. These findings are likely generalizable to other regions with high MRMP prevalence. Implementing this prediction algorithm enables a precision-switch strategy that reduces unnecessary tetracycline exposure while improving the efficacy and safety of antimicrobial therapy.
Acknowledgments
The authors gratefully acknowledge the clinical teams of the Department of Pediatrics at the Affiliated Hospital of North Sichuan Medical College and the People's Hospital of Nanbu County for their direct contributions to patient enrollment and data curation. We specifically thank Dr. Hao Wei for her direct intellectual contributions to the study design, interpretation of the results, and critical revision of the manuscript.
Funding Statement
The author(s) declared that financial support was not received for this work and/or its publication. The data were collected from the hospital's electronic medical records, and no financial support was obtained for this study.
Footnotes
Edited by: Samuel Otido, Nairobi, Kenya
Reviewed by: Norma Velazquez-Guadarrama, Research Laboratory in Microbiology and Antimicrobial Resistance. Hospital Infantil de México FEderico Gómez, Mexico
Desiree Caselli, Azienda Ospedaliero Universitaria Consorziale Policlinico di Bari, Italy
Abbreviations MP, Mycoplasma pneumoniae; CAP, Community-Acquired Pneumonia; MPP, Mycoplasma pneumoniae Pneumonia; SMPP, Severe Mycoplasma pneumoniae Pneumonia; GMPP, General Mycoplasma pneumoniae Pneumonia; MSMP, Macrolide-Susceptible Mycoplasma pneumoniae; MRMP, Macrolide-Resistant Mycoplasma pneumoniae; T-NGS, Targeted Next-Generation Sequencing; WBC, White Blood Cell; CRP, C-Reactive Protein; PCT, Procalcitonin; ALB, Albumin; ALT, Alanine Aminotransferase; TBA, Total Bile Acids; TBIL, Total Bilirubin; LDH, Lactate Dehydrogenase; ROC, Receiver Operating Characteristic; AUC, Area Under the Curve; OR, Odds Ratio; CI, Confidence Interval; MIC, Minimum Inhibitory Concentration; MP-DNA, Mycoplasma pneumoniae DNA; LPS, Lipopolysaccharide; TLR2, Toll-like Receptor 2; TLR4, Toll-like Receptor 4; IL-6, Interleukin-6.
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving humans were approved by Medical Ethics Committees of the Affiliated Hospital Of North Sichuan Medical College in Nanchong and the Nanfang County People's Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants' legal guardians/next of kin because Informed consent was waived by the ethics committee of the Affiliated Hospital Of North Sichuan Medical College in Nanchong and the Nanfang County People's Hospital,given that the research involved only the analysis of anonymized, pre-existing clinical data without any additional intervention or risk to the patients.
Author contributions
MZ: Writing – review & editing, Conceptualization, Methodology, Writing – original draft, Software, Investigation, Visualization, Data curation, Validation, Resources. QL: Writing – review & editing, Software, Visualization, Validation. HW: Formal analysis, Writing – review & editing, Supervision, Conceptualization. AW: Conceptualization, Writing – review & editing, Project administration, Investigation. JW: Investigation, Writing – review & editing. HL: Writing – review & editing, Resources. XJ: Resources, Writing – review & editing, Investigation.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fped.2026.1785723/full#supplementary-material
References
- 1.Lind K, Benzon M, Jensen J, Clyde WJ. A seroepidemiological study of Mycoplasma pneumoniae infections in Denmark over the 50-year period 1946–1995. Eur J Epidemiol. (1997) 13:581–6. 10.1023/a:1007353121693 [DOI] [PubMed] [Google Scholar]
- 2.Atkinson TP, Balish MF, Waites KB. Epidemiology, clinical manifestations, pathogenesis and laboratory detection of Mycoplasma pneumoniae infections. FEMS Microbiol Rev. (2008) 32(6):956–73. 10.1111/j.1574-6976.2008.00129.x [DOI] [PubMed] [Google Scholar]
- 3.Chi J, Tang H, Wang F, Wang Y, Chen Z. Surge in Mycoplasma Pneumoniae infection and respiratory viruses co-infection in children with community-acquired pneumonia in the post-pandemic. Pediatric Health Med Ther. (2024) 15:279–88. 10.2147/phmt.S473669 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Urbieta AD, Barbeito Castiñeiras G, Rivero Calle I, Pardo Seco J, Rodríguez Tenreiro C, Suárez Camacho R, et al. Mycoplasma pneumoniae at the rise not only in China: rapid increase of Mycoplasma pneumoniae cases also in Spain. Emerg Microbes Infect. (2024) 13(1):2332680. 10.1080/22221751.2024.2332680 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhang X-B, He W, Gui Y-H, Lu Q, Yin Y, Zhang J-H, et al. Current Mycoplasma pneumoniae epidemic among children in Shanghai: unusual pneumonia caused by usual pathogen. World J Pediatr. (2024b) 20(1):5–10. 10.1007/s12519-023-00793-9 [DOI] [PubMed] [Google Scholar]
- 6.Zhao S, Qian S, Chen Z, Gao H, Liu H, Zhang H, et al. Guidelines for the diagnosis and treatment of Mycoplasma pneumoniae pneumonia in children (2023 edition). Electron J Emerg Infect Dis. (2024) 9(1):73–9. 10.19871/j.cnki.xfcrbzz.2024.01.015 [DOI] [Google Scholar]
- 7.Li L, Liang M, Zhen H, Huang L. Current status and research progress on macrolide-resistant Mycoplasma pneumoniae infection. Chin J Appl Clin Pediatr. (2025) 40(7):557–60. 10.3760/cma.j.cn101070-20240817-00521 [DOI] [Google Scholar]
- 8.Bébéar CM, Pereyre S. Mechanisms of drug resistance in Mycoplasma pneumoniae. Curr Drug Targets Infect Disord. (2005) 5(3):263–71. 10.2174/1568005054880109 [DOI] [PubMed] [Google Scholar]
- 9.Li L, Ma J, Guo P, Song X, Li M, Yu Z, et al. Molecular beacon based real-time PCR p1 gene genotyping, macrolide resistance mutation detection and clinical characteristics analysis of Mycoplasma pneumoniae infections in children. BMC Infect Dis. (2022) 22(1):724. 10.1186/s12879-022-07715-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wang N, Xu X, Xiao L, Liu Y. Novel mechanisms of macrolide resistance revealed by in vitro selection and genome analysis in Mycoplasma pneumoniae. Front Cell Infect Microbiol. (2023) 13:1186017. 10.3389/fcimb.2023.1186017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Liu Y, Ye X, Zhang H, Xu X, Li W, Zhu D, et al. Characterization of macrolide resistance in Mycoplasma pneumoniae isolated from children in Shanghai, China. Diagn Microbiol Infect Dis. (2010) 67(4):355–8. 10.1016/j.diagmicrobio.2010.03.004 [DOI] [PubMed] [Google Scholar]
- 12.Chironna M, Sallustio A, Esposito S, Perulli M, Chinellato I, Di Bari C, et al. Emergence of macrolide-resistant strains during an outbreak of Mycoplasma pneumoniae infections in children. J Antimicrob Chemother. (2011) 66(4):734–7. 10.1093/jac/dkr003 [DOI] [PubMed] [Google Scholar]
- 13.Morozumi M, Tajima T, Sakuma M, Shouji M, Meguro H, Saito K, et al. Sequence type changes associated with decreasing macrolide-resistant Mycoplasma pneumoniae, Japan. Emerg Infect Dis. (2020) 26(9):2210–3. 10.3201/eid2609.191575 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Rivaya B, Jordana-Lluch E, Fernández-Rivas G, Molinos S, Campos R, Méndez-Hernández M, et al. Macrolide resistance and molecular typing of Mycoplasma pneumoniae infections during a 4 year period in Spain. J Antimicrob Chemother. (2020) 75(10):2752–9. 10.1093/jac/dkaa256 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jiang T-t, Sun L, Wang T-y, Qi H, Tang H, Wang Y-c, et al. The clinical significance of macrolide resistance in pediatric Mycoplasma pneumoniae infection during COVID-19 pandemic. Front Cell Infect Microbiol. (2023) 13:1181402. 10.3389/fcimb.2023.1181402 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wang Y-S, Zhou Y-L, Bai G-N, Li S-X, Xu D, Chen L-N, et al. Expert consensus on the diagnosis and treatment of macrolide-resistant Mycoplasma pneumoniae pneumonia in children. World J Pediatr. (2024) 20(9):901–14. 10.1007/s12519-024-00831-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhao J, Xu M, Tian Z, Wang Y. Clinical characteristics of pathogens in children with community-acquired pneumonia were analyzed via targeted next-generation sequencing detection. PeerJ. (2025) 13:e18810. 10.7717/peerj.18810 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Pereyre S, Goret J, Bébéar C. Mycoplasma pneumoniae: current knowledge on macrolide resistance and treatment. Front Microbiol. (2016) 7:974. 10.3389/fmicb.2016.00974 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Deng B, Dong W, Cao J, Zhou J. Influence of Mycoplasma resistance genes on pediatric Mycoplasma pneumonia treatment and determinants for second-line antimicrobial adjustment. Comb Chem High Throughput Screening. (2026) 29(5):909–17. 10.2174/0113862073364492250507095636 [DOI] [PubMed] [Google Scholar]
- 20.Kawai Y, Miyashita N, Kubo M, Akaike H, Kato A, Nishizawa Y, et al. Therapeutic efficacy of macrolides, minocycline, and tosufloxacin against macrolide-resistant Mycoplasma pneumoniae pneumonia in pediatric patients. Antimicrob Agents Chemother. (2013) 57(5):2252–8. 10.1128/aac.00048-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Cross R, Ling C, Day NPJ, McGready R, Paris DH. Revisiting doxycycline in pregnancy and early childhood—time to rebuild its reputation? Expert Opin Drug Saf. (2016) 15(3):367–82. 10.1517/14740338.2016.1133584 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Yang TI, Chang TH, Lu CY, Chen JM, Lee PI, Huang LM, et al. Mycoplasma pneumoniae in pediatric patients: do macrolide-resistance and/or delayed treatment matter? J Microbiol Immunol Infect. (2019) 52(2):329–35. 10.1016/j.jmii.2018.09.009 [DOI] [PubMed] [Google Scholar]
- 23.Dekyi, Xiao Y, Wang X, Feng S, Wang Y, Liao L, et al. Predominance of A2063G mutant strains in the Mycoplasma pneumoniae epidemic in children: a clinical and epidemiological study in 2023 in Wuhan, China. Int J Infect Dis. (2024) 145:107074. 10.1016/j.ijid.2024.107074 [DOI] [PubMed] [Google Scholar]
- 24.Huang R, Duan X, Rui Y. Genomic analysis and drug resistance mechanisms of multidrug-resistant Citrobacter freundii. Int J Lab Med. (2024) 45(7):785–9. 10.3969/j.issn.1673-4130.2024.07.005 [DOI] [Google Scholar]
- 25.Guo W, Cui X, Wang Q, Wei Y, Guo Y, Zhang T, et al. Clinical evaluation of metagenomic next-generation sequencing for detecting pathogens in bronchoalveolar lavage fluid collected from children with community-acquired pneumonia. Front Med (Lausanne). (2022) 9:952636. 10.3389/fmed.2022.952636 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lu Y, Li W-j, Wang X-x, Huang A-q, Cheng H. Risk factors and prescription patterns analysis for macrolide-resistant Mycoplasma pneumoniae pneumonia in children. iScience. (2024) 27(12):111503. 10.1016/j.isci.2024.111503 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Zhang W, Wang Y, Cai L. Analysis of multidrug resistance gene mutations and clinical significance in children with Mycoplasma pneumoniae pneumonia. Chin J Contemp Pediatr. (2024a) 26(11):1176–81. 10.7499/j.issn.1008-8830.2407129 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Oishi T, Ouchi K. Recent trends in the epidemiology, diagnosis, and treatment of macrolide-resistant Mycoplasma pneumoniae. J Clin Med. (2022) 11(7):1782. 10.3390/jcm11071782 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Xin D, Mi Z, Han X, Qin L, Li J, Wei T, et al. Molecular mechanisms of macrolide resistance in clinical isolates of Mycoplasma pneumoniae from China. Antimicrob Agents Chemother. (2009) 53(5):2158–9. 10.1128/aac.01563-08 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zhao F, Liu G, Wu J, Cao B, Tao X, He L, et al. Surveillance of macrolide-resistant Mycoplasma pneumoniae in Beijing, China, from 2008 to 2012. Antimicrob Agents Chemother. (2013) 57(3):1521–3. 10.1128/aac.02060-12 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zhou Z, Li X, Chen X, Luo F, Pan C, Zheng X, et al. Macrolide-resistant Mycoplasma pneumoniae in adults in Zhejiang, China. Antimicrob Agents Chemother. (2015) 59(2):1048–51. 10.1128/aac.04308-14 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Li Y, Zhong L-L, Zhang B, Huang H, Ding X-F, Lin L, et al. Clinical characteristics of children with Mycoplasma pneumoniae pneumonia carrying the 23S rRNA A2063G mutation. J Clin Pediatr. (2018) 36(8):569. 10.3969/j.issn.1000-3606.2018.08.001 [DOI] [Google Scholar]
- 33.Oishi T, Hattori N, Yoshioka D. Novel knowledge of macrolide resistance in Mycoplasma pneumoniae by azithromycin exposure. Microorganisms. (2024) 12(1):218. 10.3390/microorganisms12010218 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sloan B, Scheinfeld N. The use and safety of doxycycline hyclate and other second-generation tetracyclines. Expert Opin Drug Saf. (2008) 7(5):571–7. 10.1517/14740338.7.5.571 [DOI] [PubMed] [Google Scholar]
- 35.Xiong C-Y, Yang Y-M, Zhou Y, He T-S, Luo X-W, Wang J, et al. Systematic analysis of the adverse effects of commonly used clinical tetracycline drugs based on the FAERS database. Expert Opin Drug Saf. (2024) 24:949–57. 10.1080/14740338.2024.2392865 [DOI] [PubMed] [Google Scholar]
- 36.Suzuki S, Yamazaki T, Narita M, Okazaki N, Suzuki I, Andoh T, et al. Clinical evaluation of macrolide-resistant Mycoplasma pneumoniae. Antimicrob Agents Chemother. (2006) 50(2):709–12. 10.1128/aac.50.2.709-712.2006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Matsubara K, Morozumi M, Okada T, Matsushima T, Komiyama O, Shoji M, et al. A comparative clinical study of macrolide-sensitive and macrolide-resistant Mycoplasma pneumoniae infections in pediatric patients. J Infect Chemother. (2009) 15(6):380–3. 10.1007/s10156-009-0715-7 [DOI] [PubMed] [Google Scholar]
- 38.Chinese Medical Association Subspecialty Group of Respiratory Medicine, C.J.o.P.E.B., China Medical Education Association Pediatric Professional Committee. Guidelines for the management of community-acquired pneumonia in children (2024 revision). Chin J Pediatr. (2024) 62(10):920–30. 10.3760/cma.j.cn112140-20240728-00523 [DOI] [PubMed] [Google Scholar]
- 39.Leng M, Xu L, Dong Z, Yang J. The clinical significance of pathogen loads and macrolide resistance levels for macrolide-resistant Mycoplasma pneumoniae pneumonia in children. J Inflamm Res. (2024) 17:7339–46. 10.2147/jir.S491981 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Sun H, Li S, Wang T, Chen Z. Mycoplasma Pneumoniae infection and persistent wheezing in young children: a retrospective case-control study. Front Pediatr. (2022) 10:811086. 10.3389/fped.2022.811086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Chen Q, Hu T, Wu L, Chen L. Clinical features and biomarkers for early prediction of refractory Mycoplasma Pneumoniae pneumonia in children. Emerg Med Int. (2024) 2024:9328177. 10.1155/2024/9328177 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Li P, Wang W, Zhang X, Pan J, Gong L. Observational retrospective clinical study on clinical features of macrolide-resistant Mycoplasma pneumoniae pneumonia in Chinese pediatric cases. Sci Rep. (2024) 14(1):5632. 10.1038/s41598-024-55311-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Li Y, Min L, Zhang X. Usefulness of procalcitonin (PCT), C-reactive protein (CRP), and white blood cell (WBC) levels in the differential diagnosis of acute bacterial, viral, and mycoplasmal respiratory tract infections in children. BMC Pulm Med. (2021) 21(1):386. 10.1186/s12890-021-01756-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Schuetz P. How to best use procalcitonin to diagnose infections and manage antibiotic treatment. Clin Chem Lab Med. (2023) 61(5):822–8. 10.1515/cclm-2022-1072 [DOI] [PubMed] [Google Scholar]
- 45.Shimizu T, Kimura Y, Kida Y, Kuwano K, Tachibana M, Hashino M, et al. Cytadherence of Mycoplasma pneumoniae induces inflammatory responses through autophagy and toll-like receptor 4. Infect Immun. (2014) 82(7):3076–86. 10.1128/iai.01961-14 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Flynn CM, Garbers Y, Lokau J, Wesch D, Schulte DM, Laudes M, et al. Activation of toll-like receptor 2 (TLR2) induces interleukin-6 trans-signaling. Sci Rep. (2019) 9(1):7306. 10.1038/s41598-019-43617-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Bader MS, Yi Y, Abouchehade K, Haroon B, Bishop LD, Hawboldt J. Community-Acquired pneumonia in patients with diabetes Mellitus: predictors of complications and length of hospital stay. Am J Med Sci. (2016) 352(1):30–5. 10.1016/j.amjms.2016.02.032 [DOI] [PubMed] [Google Scholar]
- 48.Yan C, Xue G, Zhao H, Feng Y, Li S, Cui J, et al. Molecular and clinical characteristics of severe Mycoplasma pneumoniae pneumonia in children. Pediatr Pulmonol. (2019) 54(7):1012–21. 10.1002/ppul.24327 [DOI] [PubMed] [Google Scholar]
- 49.Liu S, Zhang L, Dai L, Li J, Li D. Prediction model for severe mycoplasma pneumoniae pneumonia and analysis of macrolide-resistance in children: a case-control study. Ital J Pediatr. (2025) 51(1):180. 10.1186/s13052-025-02039-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Yang S, Lu S, Guo Y, Luan W, Liu J, Wang L. A comparative study of general and severe mycoplasma pneumoniae pneumonia in children. BMC Infect Dis. (2024) 24(1):449. 10.1186/s12879-024-09340-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Zhang X, Sun R, Jia W, Li P, Song C. A new dynamic nomogram for predicting the risk of severe Mycoplasma pneumoniae pneumonia in children. Sci Rep. (2024c) 14(1):8260. 10.1038/s41598-024-58784-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Wang X, He H, Zheng J, Wang J, Zheng H, Zhang B. A comparison of efficacy and safety of complementary and alternative therapies for severe mycoplasma pneumonia in children: a protocol for systematic review and meta-analysis. Medicine (Baltimore). (2021) 100(7):e23959. 10.1097/md.0000000000023959 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.