Routine admission biomarkers for identifying co-infection and stratifying severity in pediatric macrolide-resistant Mycoplasma pneumoniae pneumonia: a retrospective cohort study

Introduction

Mycoplasma pneumoniae (MP) is a well-recognized cause of community-acquired pneumonia (CAP) in children and continues to account for a substantial proportion of pediatric respiratory morbidity worldwide (1-3). Beyond pulmonary disease, MP infection can also give rise to extrapulmonary complications, largely through a combination of direct cytotoxic injury and dysregulated host immune responses. These manifestations may obscure the clinical picture and add to the overall disease burden (4,5). During the past two decades, extensive use of macrolides has coincided with a marked rise in macrolide-resistant Mycoplasma pneumoniae (MRMP), most notably in East Asia, where the A2063G mutation in domain V of the 23S rRNA gene represents the predominant resistance genotype (6-9). Compared with macrolide-sensitive MP infection, MRMP is more frequently associated with persistent fever, slower clinical recovery, prolonged hospitalization, and a greater need for intensified treatment (10-12).

Meanwhile, mixed infection has emerged as a clinically relevant concern in children with MP pneumonia. Earlier studies have reported that the co-detection of additional respiratory pathogens, particularly viruses such as adenovirus, is relatively common and may accompany a more severe disease course (13-17). Among children with MP-related pneumonia, co-infection has been associated with longer fever duration, wider radiographic involvement, increased oxygen requirement, and more frequent use of glucocorticoids, immunoglobulin, or other escalation therapies (16,17). This issue may be especially important in the context of MRMP, since delayed recognition of co-infection can complicate antimicrobial decision-making and may interfere with timely risk assessment as the illness evolves (7,18,19). Although targeted next-generation sequencing (t-NGS) has improved the detection of respiratory pathogens and resistance-related mutations, its cost, technical requirements, and limited immediacy still restrict its value for routine bedside decision-making (20-22).

Against this background, simple and readily obtainable indicators at admission are needed to help clinicians identify children with MRMP pneumonia who are more likely to have co-infection and, within this group, those at greater risk of developing severe disease. To date, however, most available evidence has centered on the overall severity of MP pneumonia or refractory MP pneumonia, while studies specifically addressing early evaluation of pediatric MRMP pneumonia complicated by bacterial and/or viral co-infection remain limited (13-17,23,24). From this perspective, we conducted a single-center retrospective study to examine whether routine admission parameters could support the early identification of co-infection and the stratification of disease severity in children with MRMP pneumonia. We first compared children with MRMP mono-infection with those who had MRMP co-infection involving bacteria and/or viruses, and then further explored factors associated with severe disease within the co-infected subgroup. We present this article in accordance with the TRIPOD reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0155/rc).

Methods

Study design

This single-center retrospective cohort study included pediatric patients with MRMP pneumonia admitted to the Sixth Affiliated Hospital of Harbin Medical University during a large outbreak of MP from August 1 to December 31, 2023. The MR genotype was confirmed as the A2063G point mutation in domain V of the 23S rRNA gene, identified using t-NGS from oropharyngeal/nasopharyngeal swabs or aspirates, induced sputum (IS), lower respiratory tract samples, or bronchoalveolar lavage fluid (BALF). Demographic data including sex, age, height, weight, and body mass index (BMI), as well as clinical parameters such as white blood cell (WBC) count, neutrophil percentage (NEUT%), neutrophil count (NEUT), lymphocyte percentage (LYM%), lymphocyte count (LYM), platelet (PLT), fibrinogen (FIB), D-dimer, alanine aminotransferase (ALT), aspartate aminotransferase (AST), lactate dehydrogenase (LDH), total bilirubin (TBIL), direct bilirubin (DBIL), indirect bilirubin (IBIL), albumin (ALB), blood urea nitrogen (BUN), serum creatinine (SCr), C-reactive protein (CRP), and procalcitonin (PCT) on admission, as well as maximum body temperature and disease severity during hospitalization, were collected from medical records. In the first phase, patients were divided into two groups: MRMP mono-infection and MRMP co-infection with bacteria and/or viruses. This classification was based on a thorough evaluation of medical history, clinical symptoms, physical examinations, laboratory results, imaging findings, and t-NGS data. In the second phase, the co-infection group was further divided into mild and severe groups based on disease severity. Demographic data and clinical parameters on admission and during hospitalization were compared between these groups. The study data were extracted from the hospital medical record system and anonymized by designated research personnel before analysis. This study was approved by the Ethics Committee of The Sixth Affiliated Hospital of Harbin Medical University (IRB number: LC2024-072). Given the retrospective and observational design, the requirement for informed consent was waived by the Ethics Committee. Patient privacy and identifying information were strictly protected throughout the study. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Study population

The inclusion criteria for this study were pediatric patients (aged 28 days to 16 years) admitted to The Sixth Affiliated Hospital of Harbin Medical University between August 1 and December 31, 2023, who were definitively diagnosed with MRMP pneumonia. The exclusion criteria included concomitant non-respiratory infections or other infectious diseases unrelated to the current pneumonia episode, autoimmune diseases, primary immunodeficiency diseases (PID), acquired immune deficiency syndrome (AIDS), chronic organ failure, uncontrolled malignancies, organ transplants or immunotherapy within the past 6 months, critical MRMP pneumonia, and incomplete medical records (Figure 1).

Figure 1 Flowchart of the study participants. MRMP, macrolide-resistant Mycoplasma pneumoniae.

Determination of MRMP pneumonia

In this study, MRMP pneumonia was diagnosed through a comprehensive assessment of medical history, clinical symptoms, physical examination, laboratory results, imaging findings, and t-NGS data (25). While t-NGS was essential, it was not the sole determinant in clinical practice.

Diagnostic criteria for severe and critical MRMP pneumonia

Pediatric patients with severe MRMP pneumonia met at least one of the following criteria: (I) persistent high fever (above 39 ℃) for ≥5 days or fever lasting ≥7 days without a decrease in peak temperature; (II) wheezing, shortness of breath, dyspnea, chest pain, or hemoptysis; (III) extrapulmonary complications without meeting the criteria for critical illness; (IV) oxygen saturation (SpO₂) ≤0.93 while breathing air at rest; (V) involvement of ≥2/3 of a single lobe or consistent high-density consolidation in one or more lobes, with moderate to large pleural effusion and localized bronchiolitis; (VI) diffuse bronchiolitis in one lung or ≥4/5 of both lungs, with bronchitis and atelectasis caused by mucus plug formation; (VII) worsening clinical symptoms with >50% progression of imaging lesions within 24 to 48 hours; (VIII) significant elevation of CRP, LDH, or D-dimer (25). Critical MRMP pneumonia was defined by the presence of respiratory failure and/or life-threatening extrapulmonary complications requiring life support, such as mechanical ventilation (25).

Data collection

Demographic data and clinical parameters on admission and during hospitalization were collected from medical records by dedicated research personnel. Other team members did not have access to the personal information of the enrolled pediatric patients beyond what was necessary for the study.

Statistical analysis

SPSS 24.0 (SPSS, Inc., Chicago, IL, USA) was used for statistical analysis. Continuous data with a normal distribution were expressed as mean ± standard deviation (SD), while those without a normal distribution were presented as median (P25, P75). Categorical data were represented by frequency. An independent sample t-test was used to compare continuous data with a normal distribution between groups, while the Mann-Whitney U test was applied for continuous data without a normal distribution. The Chi-square (χ²) test was used to compare categorical data between groups, and Fisher’s exact test was used if the assumptions of the χ² test were not met. Variables with a significance level of less than 0.1 in inter-group comparisons were included in a multiple logistic regression analysis model and screened using the Wald backward method. Finally, receiver operating characteristic (ROC) curve analysis was performed to calculate the area under the ROC curve, the cut-off value, and the corresponding sensitivity and specificity. A P value <0.05 was considered statistically significant.

Results

Baseline and clinical characteristics of children with MRMP mono-infection and co-infection

A total of 278 pediatric patients with MRMP pneumonia were included in this single-center retrospective cohort study. Of these, 78 were in the MRMP mono-infection group, and 200 were in the MRMP co-infection with bacteria and/or viruses group. In the first comparison, a significant difference was observed on admission in PLT levels between the groups (P=0.03) (Table 1).

Factors associated with MRMP co-infection

No factors associated with MRMP co-infection with bacteria and/or viruses were identified through multiple logistic regression analysis (Table 2).

ROC analysis of PLT for identifying MRMP co-infection

The area under the ROC curve for PLT in classifying pediatric patients with MRMP pneumonia into the MRMP co-infection group was 0.584 (Table 3 and Figure 2). The cut-off value for PLT was 279.00, with a sensitivity of 59% and a specificity of 58%.

Figure 2 ROC analysis of PLT for identifying MRMP co-infection. MRMP, macrolide-resistant Mycoplasma pneumoniae; PLT, platelet; ROC, receiver operating characteristic.

Characteristics of mild and severe MRMP co-infection

In the second part, significant differences were found in several clinical parameters between the mild and severe groups of pediatric patients with MRMP pneumonia co-infected with bacteria and/or viruses. These parameters included NEUT%, LYM%, LYM, D-dimer, LDH, ALB, and PCT on admission, as well as maximum body temperature during hospitalization (P=0.02, P=0.01, P=0.03, P<0.001, P<0.001, P<0.001, P<0.001, P=0.03, respectively) (Table 4).

Factors associated with severe MRMP co-infection

Lower ALB on admission was associated with lower odds of severe disease among children with MRMP co-infection, whereas higher PCT was associated with higher odds of severe disease. Each 1 g/L increase in ALB was associated with lower odds of severe disease [odds ratio (OR) =0.800, 95% confidence interval (CI): 0.710–0.900], whereas each 1 ng/mL increase in PCT was associated with higher odds of severe disease (OR =9.440, 95% CI: 1.110–80.340) (Table 5).

ROC analysis of ALB, PCT, and their combination for identifying severe co-infection

The areas under the ROC curves for ALB combined with PCT, ALB alone, and PCT alone in classifying pediatric patients with MRMP pneumonia co-infected with bacteria and/or viruses into the severe group were 0.720, 0.704, and 0.704, respectively (Table 6 and Figure 3). The cut-off values for ALB and PCT were 40.250 and 0.105, respectively. The sensitivity and specificity for the ROC curves of ALB combined with PCT were 63.4% and 76.1%, for ALB alone were 68.3% and 67.3%, and for PCT alone were 73.2% and 64.8%, respectively.

Figure 3 ROC analysis of ALB, PCT, and their combination for identifying severe co-infection. ALB, albumin; PCT, procalcitonin; ROC, receiver operating characteristic.

Discussion

In this single-center retrospective study, we assessed whether routinely available admission parameters could aid in identifying co-infection and stratifying disease severity in children with MRMP pneumonia. Two findings were particularly noteworthy. First, PLT levels were higher in children with MRMP co-infection than in those with MRMP mono-infection; however, PLT had limited discriminatory value and was not retained as an independent factor in the multivariable analysis. Second, among children with MRMP co-infection, lower ALB and higher PCT levels were independently associated with severe disease, and the combined use of these two markers yielded a slightly higher area under the curve (AUC) than either marker alone. Overall, these results suggest that routine admission biomarkers may provide preliminary value for early assessment in pediatric MRMP pneumonia, although they remain insufficient for independent clinical decision-making. The clinical relevance of this question is closely related to the growing complexity of pediatric MP-related pneumonia in the context of increasing macrolide resistance. Compared with macrolide-sensitive infection, MRMP is more frequently associated with prolonged fever, delayed treatment response, greater treatment intensity, and a higher risk of unfavorable in-hospital outcomes (10-12). At the same time, mixed infection has been increasingly recognized in children with MP pneumonia and may contribute to a heavier inflammatory burden, wider radiographic involvement, and more severe clinical manifestations (13-17). Although t-NGS has improved pathogen detection and resistance profiling, its cost, turnaround time, and technical requirements may restrict its immediate bedside application in many clinical settings (20-22). In this context, simple admission indicators that help clinicians identify children requiring closer attention remain clinically meaningful.

In the first-stage analysis, children with MRMP co-infection showed significantly higher PLT levels than those with MRMP mono-infection, which may partly reflect a more pronounced systemic inflammatory response. PLTs are now understood not only as components of hemostasis but also as active mediators of inflammation and host defense, and reactive thrombocytosis may occur during acute infectious or inflammatory conditions. Even so, this finding requires cautious interpretation. PLT was not retained as an independent factor in the multivariable analysis, and its AUC for distinguishing co-infection was only 0.584. Thus, although PLT differed significantly between groups, its discriminatory ability was weak and does not support its use as an isolated screening marker. From a clinical standpoint, PLT may be more appropriately viewed as a modest signal of increased inflammatory activity rather than a reliable indicator of mixed infection. The more clinically relevant finding emerged from the second-stage analysis. Among children with MRMP pneumonia and co-infection, those in the severe group presented with a more inflammatory laboratory profile at admission, including higher NEUT%, D-dimer, LDH, and PCT levels, along with lower LYM and ALB levels. After multivariable adjustment, ALB and PCT remained independently associated with severe disease. This pattern is biologically plausible. PCT is widely used as an indicator of systemic bacterial inflammatory response and has been associated with the severity of pediatric pneumonia. In MRMP complicated by bacterial and/or viral co-infection, elevated PCT may suggest stronger inflammatory activation or a greater burden of secondary infection (23,24). By contrast, hypoalbuminemia may reflect systemic inflammation, capillary leakage, depletion of nutritional reserve, and the overall stress response to severe illness. From this perspective, low ALB combined with high PCT may capture two complementary dimensions of severity: impaired systemic condition and enhanced inflammatory activation. The ROC analysis provided further support for this interpretation. The combination of ALB and PCT yielded an AUC of 0.720, which was higher than that of either marker alone. Nevertheless, this level of performance should still be regarded as moderate rather than definitive. It may help raise early awareness of risk, but it does not establish a validated bedside prediction tool. Accordingly, our findings are better interpreted as hypothesis-generating rather than practice-changing. In children with MRMP pneumonia and suspected co-infection, abnormal ALB and PCT levels at admission may help identify patients who require closer monitoring, repeated reassessment, and earlier supportive evaluation. These markers, however, should still be interpreted together with clinical status, imaging findings, and microbiological evidence.

This study may also inform the management of pediatric MRMP during epidemic periods. MP epidemics tend to recur cyclically, and recent reports suggest that renewed surges may place considerable pressure on pediatric inpatient services (19,26). In such settings, clinicians often need to distinguish mono-infection from mixed infection while also identifying children at higher risk of deterioration. Our findings indicate that readily available admission parameters may assist early stratification, particularly in high-volume or resource-limited clinical environments. Still, laboratory surrogates should not replace etiologic testing or formal severity assessment. Rather, they may serve as adjunctive indicators within a broader clinical evaluation framework.

Several limitations should be acknowledged. First, this was a single-center retrospective study, which limits both internal robustness and external generalizability. Second, the sample size was relatively modest, and the number of severe cases in the second-stage analysis was small, increasing the possibility of unstable multivariable estimates. Third, co-infection was classified using combined clinical, laboratory, imaging, and t-NGS findings rather than a uniform microbiological gold standard; therefore, misclassification cannot be excluded. Fourth, children with critical MRMP pneumonia were excluded, limiting the applicability of these findings to the sickest patients. Fifth, only apparent model performance was evaluated in a single retrospective dataset, without internal resampling or external validation, leaving room for overfitting and optimistic performance estimates. Finally, although ALB and PCT showed moderate discriminatory value, calibration and model transportability were not assessed. Larger prospective multicenter studies are therefore needed to address these limitations.

In summary, routine admission markers may offer limited but clinically relevant exploratory value in pediatric MRMP pneumonia. PLT differed between mono-infection and co-infection groups, but its ability to discriminate co-infection was poor. In contrast, lower ALB and higher PCT were independently associated with severe disease among children with MRMP co-infection, and their combination modestly improved early risk stratification. Further prospective multicenter studies are required to validate these findings and determine whether they can be incorporated into more robust clinical prediction frameworks.

Conclusions

In conclusion, this single-center retrospective study of children with MRMP pneumonia suggests that PLT count has limited usefulness for distinguishing bacterial and/or viral co-infection. By contrast, lower ALB and higher PCT levels at admission were independently associated with severe disease among co-infected children. When used together, ALB and PCT offered a modest improvement in early severity assessment, although the results should be regarded as exploratory rather than conclusive. Further prospective multicenter studies with more robust validation are needed to confirm these associations and clarify their potential role in future clinical prediction strategies.

Acknowledgments

The authors thank the staff of the Sixth Affiliated Hospital of Harbin Medical University and the clinical laboratory team for their support in data collection and laboratory testing.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0155/rc

Data Sharing Statement: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0155/dss

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0155/prf

Funding: This study was supported by the National Natural Science Foundation of China (No. 82372172), the Key Research and Development Plan Project of Heilongjiang Province (No. GA23C007), the Heilongjiang Province Postdoctoral Start-up Fund (No. LBH-Q20037), and the 2024 Artificial Liver Special Fund (No. iGandanF-1082024-RGG021).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0155/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of The Sixth Affiliated Hospital of Harbin Medical University (IRB number: LC2024-072). Given the retrospective and observational design, the requirement for informed consent was waived by the Ethics Committee. Patient privacy and identifying information were strictly protected throughout the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Xue Y, Wang M, Han H. Interaction between alveolar macrophages and epithelial cells during Mycoplasma pneumoniae infection. Front Cell Infect Microbiol 2023;13:1052020. [Crossref] [PubMed]
2. Meyer Sauteur PM, Beeton ML, Uldum SA, et al. Mycoplasma pneumoniae detections before and during the COVID-19 pandemic: results of a global survey, 2017 to 2021. Euro Surveill 2022;27:2100746. [Crossref] [PubMed]
3. Otheo E, Rodríguez M, Moraleda C, et al. Viruses and Mycoplasma pneumoniae are the main etiological agents of community-acquired pneumonia in hospitalized pediatric patients in Spain. Pediatr Pulmonol 2022;57:253-63. [Crossref] [PubMed]
4. Kumar S, Kumar S. Mycoplasma pneumoniae: Among the smallest bacterial pathogens with great clinical significance in children. Indian J Med Microbiol 2023;46:100480. [Crossref] [PubMed]
5. Poddighe D. Mycoplasma pneumoniae-related hepatitis in children. Microb Pathog 2020;139:103863. [Crossref] [PubMed]
6. Shim JY. Current perspectives on atypical pneumonia in children. Clin Exp Pediatr 2020;63:469-76. [Crossref] [PubMed]
7. Tsai TA, Tsai CK, Kuo KC, et al. Rational stepwise approach for Mycoplasma pneumoniae pneumonia in children. J Microbiol Immunol Infect 2021;54:557-65. [Crossref] [PubMed]
8. Xu M, Li Y, Shi Y, et al. Molecular epidemiology of Mycoplasma pneumoniae pneumonia in children, Wuhan, 2020-2022. BMC Microbiol 2024;24:23. [Crossref] [PubMed]
9. Jiang Y, Dou H, Xu B, et al. Macrolide resistance of Mycoplasma pneumoniae in several regions of China from 2013 to 2019. Epidemiol Infect 2024;152:e75. [Crossref] [PubMed]
10. Lanata MM, Wang H, Everhart K, et al. Macrolide-Resistant Mycoplasma pneumoniae Infections in Children, Ohio, USA. Emerg Infect Dis 2021;27:1588-97. [Crossref] [PubMed]
11. Yan M, Tao R, Li S, et al. Clinical characteristics and logistic regression analysis of macrolide-resistant Mycoplasma pneumoniae pneumonia in children. Eur J Clin Microbiol Infect Dis 2024;43:1825-35. [Crossref] [PubMed]
12. Wang YS, Zhou YL, Bai GN, et al. Expert consensus on the diagnosis and treatment of macrolide-resistant Mycoplasma pneumoniae pneumonia in children. World J Pediatr 2024;20:901-14. [Crossref] [PubMed]
13. Zhou Y, Wang J, Chen W, et al. Impact of viral coinfection and macrolide-resistant mycoplasma infection in children with refractory Mycoplasma pneumoniae pneumonia. BMC Infect Dis 2020;20:633. [Crossref] [PubMed]
14. Shin S, Koo S, Yang YJ, et al. Characteristics of the Mycoplasma pneumoniae Epidemic from 2019 to 2020 in Korea: Macrolide Resistance and Co-Infection Trends. Antibiotics (Basel) 2023;12:1623. [Crossref] [PubMed]
15. Chen Q, Lin L, Zhang N, et al. Adenovirus and Mycoplasma pneumoniae co-infection as a risk factor for severe community-acquired pneumonia in children. Front Pediatr 2024;12:1337786. [Crossref] [PubMed]
16. Zhao MC, Wang L, Qiu FZ, et al. Impact and clinical profiles of Mycoplasma pneumoniae co-detection in childhood community-acquired pneumonia. BMC Infect Dis 2019;19:835. [Crossref] [PubMed]
17. Wei J, Wu S, Jin X, et al. Association of Mycoplasma pneumoniae coinfection with adenovirus pneumonia severity in children. Allergol Immunopathol (Madr) 2022;50:31-6. [Crossref] [PubMed]
18. Gao Y, Wang HL, Zhang ZJ, et al. A Standardized Step-by-Step Approach for the Diagnosis and Treatment of Sepsis. J Intensive Care Med 2022;37:1281-7. [Crossref] [PubMed]
19. Zhang XB, He W, Gui YH, et al. Current Mycoplasma pneumoniae epidemic among children in Shanghai: unusual pneumonia caused by usual pathogen. World J Pediatr 2024;20:5-10. [Crossref] [PubMed]
20. Lin R, Xing Z, Liu X, et al. Performance of targeted next-generation sequencing in the detection of respiratory pathogens and antimicrobial resistance genes for children. J Med Microbiol 2023;72: [Crossref] [PubMed]
21. Yang SL, Gao Y, Han ZY, et al. Successful treatment of near-fatal pulmonary embolism and cardiac arrest in an adult patient with fulminant psittacosis-induced severe acute respiratory distress syndrome after veno-venous extracorporeal membrane oxygenation rescue: A case report and follow-up. Heliyon 2023;9:e20562. [Crossref] [PubMed]
22. Tan J, Chen Y, Lu J, et al. Pathogen distribution and infection patterns in pediatric severe pneumonia: A targeted next-generation sequencing study. Clin Chim Acta 2025;565:119985. [Crossref] [PubMed]
23. Yang S, Lu S, Guo Y, et al. A comparative study of general and severe mycoplasma pneumoniae pneumonia in children. BMC Infect Dis 2024;24:449. [Crossref] [PubMed]
24. Li L, Guo R, Zou Y, et al. Construction and Validation of a Nomogram Model to Predict the Severity of Mycoplasma pneumoniae Pneumonia in Children. J Inflamm Res 2024;17:1183-91. [Crossref] [PubMed]
25. National Health Commission. Guidelines for the Diagnosis and Treatment of Mycoplasma Pneumonia in Children (2023 Edition). Infectious Disease Information 2023;36:291-7. [Crossref]
26. Urbieta AD, Barbeito Castiñeiras G, Rivero Calle I, et al. Mycoplasma pneumoniae at the rise not only in China: rapid increase of Mycoplasma pneumoniae cases also in Spain. Emerg Microbes Infect 2024;13:2332680. [Crossref] [PubMed]