Clinical and laboratory features of pediatric Mycoplasma pneumoniae Pneumonia with viral versus bacterial co-infection

Abstract

Objective

To analyse the epidemiological, clinical, and laboratory differences among children, in Suzhou, with Mycoplasma pneumoniae (MP) mixed with viral or bacterial pneumonia and to provide evidence for early clinical differentiation and localised prevention.

Methods

This retrospective study included 392 children with Mycoplasma pneumoniae pneumonia (MPP) who were divided into an MPP mixed with viral pneumonia group (n = 254) and a bacterial pneumonia group (n = 138). Epidemiological characteristics, clinical manifestations, immune, coagulation, and inflammatory indicators, pathogen distribution, and bacterial drug susceptibility were compared.

Results

The MPP mixed with viral pneumonia group had older children [6 (3, 8) vs. 4 (2, 6.25) years, P = 0.001], a higher incidence of pneumonia in spring (51.97% vs. 18.12%, P < 0.001), and more crackles and fewer breath sounds (P < 0.05). The MPP mixed with bacterial group was more common in the summer and autumn, with a higher vomiting incidence (P = 0.021), higher C3 abnormal rate (P = 0.049) and greater procalcitonin elevation (P < 0.001). The MPP mixed with viral pneumonia group had higher level of serum amyloid A and D-dimer (P < 0.05). Human rhinovirus was the predominant virus (53.67%), and Streptococcus pneumoniae was the main bacterium (58.99%) with 100% Erythromycin resistance but high sensitivity to Penicillin G and Cefotaxime.

Conclusions

The two groups differed in age, season, clinical features, and several biomarkers. Viral coinfection was associated with higher SAA and D-dimer, while bacterial coinfection with higher PCT. However, the discriminatory value of these markers is limited, and they should be used in conjunction with other clinical data. Local pathogens were dominated by rhinovirus and S. pneumoniae, with the latter exhibiting complete erythromycin resistance, highlighting the need for surveillance and rational antibiotic use.

Introduction

Pneumonia is among the leading causes of high morbidity and mortality among children under five years of age worldwide, and Mycoplasma pneumoniae (MP) is a common pathogen of community-acquired pneumonia in children [1]. Mycoplasma pneumoniae pneumonia (MPP) not only involves pulmonary inflammation but also leads to a series of extrapulmonary complications. Severe cases may result in acute pulmonary embolism or multiple organ damage, among which refractory MPP has attracted increasing attention [2–5]. In recent years, studies have shown that MPP follows a 3–5-year epidemic cycle, and its global infection rate increased in 2023 [6], posing a severe challenge for prevention and control. The pathogenic mechanism of MP differs from that of conventional bacteria and viruses because of its lack of a cell wall. Coinfection may alter the body’s inflammatory response, coagulation function, and immune response patterns, which serve as important targets for clinical evaluation and intervention. Clarifying the specific changes and differences in coagulation function, immune function, and inflammatory indicators in children with MPP mixed with viruses or bacteria can compensate for the shortcomings of long turnaround times and limited positive rates in paediatric aetiological detection, and provide a basis for early differential diagnosis. Although several studies have focused on MP coinfection [7–8], systematic analyses of MPP with different types of pathogens in children in Suzhou, especially comparative studies on clinical manifestations, immune status, coagulation function, and the inflammatory response, remain insufficient because of significant regional and seasonal variations in the epidemic spectrum of pathogens. This study aimed to analyse the clinical data and laboratory characteristics of children with MPP mixed with bacterial or viral pneumonia in Suzhou. To provide data support and theoretical evidence for early clinical differential diagnosis, empirical treatment, and localised prevention and control of respiratory tract infections.

Materials and methods

Study subjects

From July 2021 to July 2023, a total of 1,071 children diagnosed with MPP were initially screened at the Children’s Hospital of Soochow University (CHSU). Among them, 394 cases of single MPP infection were excluded, and 285 cases of MPP coinfected with both viruses and bacteria were also excluded. Ultimately, 392 cases were included in the final analysis, comprising 254 cases of MPP mixed with viral pneumonia and 138 cases of MPP mixed with bacterial pneumonia. The clinical data and laboratory results of the above children were analysed. Children with single MPP and those with MPP mixed with viral and bacterial pneumonia were not included in this study. Most of the enrolled children were from the general paediatric ward. As the largest Grade A tertiary children’s specialised hospital in the region, CHSU admits relatively severe cases from Suzhou and surrounding areas. Therefore, the results of this study mainly reflect the epidemiological and clinical characteristics of MP coinfection among hospitalised children in this region. This study was approved by the Ethics Committee of the CHSU (Approval No.: 2025CS239), and the requirement for the informed patient consent was waived.

Inclusion and exclusion criteria

The inclusion criteria were as follows: (1) aged 0–16 years; (2) hospitalised between July 2021 and July 2023; (3) diagnosed with MPP according to Zhu Futang’s Practical Paediatrics (8th Edition) [9] and the Guidelines for the Management of Community-acquired Pneumonia in Children (2024 revision) [10], including the presence of acute respiratory symptoms and radiological evidence of pneumonia, together with positive MP aetiological test results (positive MP-DNA or MP-IgM); and (3) complete clinical data available.

The exclusion criteria were as follows: (1) pneumonia caused by foreign body aspiration; (2) readmission for pneumonia within the past 1 month; (3) prior antibiotic use before admission; and (4) children with co-morbid conditions (e.g. congenital heart diseases).

Grouping method

This study excluded children with single MPP group (n = 394) and those with MPP mixed with viral and bacterial pneumonia group (n = 285) and focused on the infection characteristics of MP combined with a single class of pathogens. According to the aetiological test results, the children were divided into two groups.

MPP mixed with bacterial pneumonia group: On the basis of the MPP diagnostic criteria, a single dominant pathogenic bacterium was isolated from the qualified sputum specimens, and the isolate was judged as a pathogenic bacterium with clinical manifestations. Qualified sputum was assessed using < 10 squamous epithelial cells/low-magnification field and > 25 white blood cells/low-magnification field. Sputum specimens were examined using the four-part streaking method. The main bacterial pathogens were identified as Streptococcus pneumoniae (S. pneumoniae), Haemophilus influenzae (H. influenzae), Moraxella catarrhalis (M. catarrhalis), and Staphylococcus aureus (S. aureus) (usually considered positive). Even if the growth was 1 + or 2+, as long as the following bacteria were isolated, the result was usually considered clinically significant; gram-negative bacteria (such as Klebsiella pneumoniae (K. pneumoniae), Pseudomonas aeruginosa (P. aeruginosa), and Acinetobacter baumannii (A. baumannii)) had to be combined with the growth rate, whereas 3 + or 4 + growth was considered meaningful. Specimens were collected before antibiotic administration.

MPP mixed with viral pneumonia group: On the basis of the MPP diagnostic criteria, at least one common respiratory viral nucleic acid was detected in respiratory specimens (nasopharyngeal swab/sputum/bronchoalveolar lavage fluid) using multiplex polymerase chain reaction (PCR), with cycle threshold values meeting the positive criteria of the kit, and the pathogenicity was confirmed in combination with clinical manifestations. Figure 1 summarizes the grouping procedure described above.

Fig. 1.

Flow diagram of patient selection and grouping

Aetiological detection methods

Nucleic acid detection of viruses

Nucleic acids were extracted using an automatic nucleic acid extraction system (Jiangsu Shuoshi Biological Technology Co., Ltd., Jiangsu, China). A 13-plex respiratory pathogen multiplex PCR detection kit (Ningbo Health Gene Technology Co., Ltd., Ningbo China) was used to detect the nucleic acids of MP, human rhinovirus (HRV), respiratory syncytial virus (RSV), human parainfluenza virus (HPIV), human coronavirus (HCOV), human metapneumovirus (HMPV), human bocavirus (Boca), adenovirus (ADV), chlamydia (Ch), and influenza viruses (subtypes influenza A (InfA) H1N1, InfA H3N2, and influenza B (InfB)) on a 3500XL-Dx genetic analyser.

Bacterial identification and susceptibility testing

Common bacteria such as S. pneumoniae and H. influenzae were detected by routine culture methods. The specimens were inoculated onto blood agar plates and chocolate agar plates (Zhengzhou Autobio Engineering Co., Ltd., Zhengzhou, China) and incubated at 37 °C under 5% CO₂ for 24–48 h. Matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (Bruker, Germany) was used for bacterial identification. The reliability of identification was determined by a confidence score (e.g., > 90%). Antimicrobial susceptibility testing was performed using the VITEK^® 2 Compact automatic microbial analysis system (bioMérieux, France) and the Kirby–Bauer disk diffusion method. The results were interpreted according to the breakpoints in the CLSI M100-2021 published by the Clinical and Laboratory Standards Institute (CLSI). For S. pneumoniae, the interpretive criteria for penicillin and other β-lactam agents were based on the oral or intravenous breakpoints for non-meningitis isolates. S. pneumoniae (American Type Culture Collection (ATCC) 49619) was used as the quality control strain and was purchased from the National Center for Clinical Laboratories, National Health Commission of the People’s Republic of China.

Detection of MP-IgM antibodies

Serum specimens were collected and tested for MP-IgM antibodies using a chemiluminescence immunoassay kit (Shenzhen YHLO Biotech Co., Ltd., Shenzhen China) on an iFlash 3000 chemiluminescence analyser. In accordance with the manufacturer’s instructions, the kit had a sensitivity of ≥ 90% and a specificity of ≥ 85%. A cut-off index (COI) ≥ 1.0 was defined as positive.

Blood biomarker detection

Venous blood (3–5 mL) was collected. The laboratory indicators used in this study and their detection methods were as follows: Complement C3, C4, and immunoglobulins (IgA, IgM, and IgG) were detected by immunoturbidimetry. Lymphocyte subsets, including CD3+, CD3 + CD4+, and CD3 + CD8+, were detected by flow cytometry. Coagulation parameters including prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen (Fib), thrombin time (TT), antithrombin III (AT-III), D-dimer (D-D), and fibrin degradation products (FDP) were measured as follows: AT-III was measured by the chromogenic substrate method, whereas D-dimer and FDP were measured by immunoturbidimetry. Procalcitonin (PCT) was detected by electrochemiluminescence immunoassay. Heparin-binding protein (HBP) was detected by the immunofluorescence dry quantitative method. Serum amyloid A (SAA) was detected by the colloidal gold method. High-sensitivity C-reactive protein (sCRP) was detected by immunoturbidimetry. These markers are routine clinical evaluation items for children with infections in our hospital and are used to differentiate infections and assess disease severity.

Statistical methods

SPSS 27.0 statistical software was used for data analysis. Nonnormally distributed data were expressed as medians (interquartile ranges, P25, P75) and were compared using the Mann‒Whitney U test. Categorical data were presented as n (%) and were compared using the chi-square (χ2) test. To control for the influence of confounding factors, a binary logistic regression model was used to further analyse the independent predictors of infection type. The results are presented as adjusted odds ratios (OR) with 95% confidence intervals (CI). 95% CI were calculated using the Clopper–Pearson exact binomial method. A two-sided P < 0.05 was considered to indicate statistical significance.

Results

Comparison of general data and clinical characteristics

A total of 392 children were included, including 254 in the MPP mixed with viral pneumonia group and 138 in the MPP mixed with bacterial pneumonia group. Children in the viral pneumonia group were significantly older than those in the bacterial pneumonia group [6 (3–8) years vs. 4 (2–6.25) years, P = 0.001]. A significant difference was observed in the distribution of onset seasons: the viral pneumonia group peaked in spring (51.97%), while the bacterial pneumonia group had higher proportions in summer (33.33%) and autumn (34.78%) (P < 0.001). In terms of clinical signs, the viral pneumonia group had higher rates of crackles (77.95% vs. 67.39%) and decreased breath sounds (18.50% vs. 7.25%, P < 0.05), whereas vomiting was more common in the bacterial pneumonia group (30.43% vs. 20.08%, P = 0.021). No significant differences were found between the two groups in terms of sex, mode of delivery, past medical history, length of hospital stay, or duration of cough (P > 0.05). See Table 1.

Table 1.

Comparison of general data and clinical characteristics between the two groups [n (%), M (P25, P75)]

Clinical characteristics	MP mixed with bacterial pneumonia group (n = 138)	MP mixed with viral pneumonia group (n = 254)	Z/χ2	P
Hospital stay (days, IQR)	8 (7-9.25)	8 (7–10)	-1.469	0.142
Cough duration (days, IQR)	7 (4–10)	6 (4–8)	-0.879	0.379
Age (years, IQR)	4 (2-6.25)	6 (3–8)	-3.298	0.001**
Male, n (%)	62 (44.93)	109 (42.91)	0.148	0.701
Full-term delivery, n (%)	128 (92.75)	242 (95.28)	1.074	0.300
Vaginal delivery, n (%)	90 (65.22)	167 (65.75)	0.011	0.916
History of pneumonia, n (%)	23 (16.67)	39 (15.35)	0.116	0.734
History of wheezing, n (%)	16 (11.59)	20 (7.87)	1.484	0.223
Onset season, n (%)
Spring	25 (18.12)	132 (51.97)	42.679	< 0.001***
Summer	46 (33.33)	41 (16.14)	15.304	< 0.001***
Autumn	48 (34.78)	50 (19.69)	10.87	< 0.001***
Winter	19 (13.77)	31 (12.20)	0.196	0.658
Fever grade, n (%)
Low fever (< 38℃)	9 (6.52)	9 (3.54)	1.811	0.178
Moderate fever (38–39℃)	44 (31.88)	89 (35.04)	0.397	0.529
High fever (> 39℃)	69 (50)	135 (53.15)	0.355	0.551
Symptoms, n (%)
Nasal congestion	50 (36.23)	105 (41.34)	0.975	0.323
Rhinorrhea	64 (46.38)	118 (46.46)	0	0.998
Chest retractions	9 (6.52)	8 (3.15)	2.451	0.117
Vomiting	42 (30.43)	51 (20.08)	5.299	0.021*
Lung signs, n (%)
Crackles	93 (67.39)	198 (77.95)	5.215	0.022*
Wheezes	42 (30.43)	72 (28.35)	0.189	0.664
Decreased breath sound	10 (7.25)	47 (18.50)	9.119	0.003**
Radiological findings, n (%)
Atelectasis	3 (2.17)	8 (3.15)	0.312	0.576
Pulmonary consolidation	5 (3.62)	16 (6.30)	1.263	0.261
Complications, n (%)
Gastrointestinal dysfunction	9 (6.52)	29 (11.42)	2.448	0.118
Hypokalemia	10 (7.25)	28 (11.02)	1.457	0.227
Hyponatremia	5 (3.62)	7 (2.76)	0.227	0.634
Myocardial injury	3 (2.17)	5 (1.97)	0.019	0.891

Note: The Chest retractions refers to subcostal, intercostal, and suprasternal retractions. The conclusion of myocardial injury was based on abnormalities in cardiac enzyme profiles and electrocardiogram. Hypokalemia (serum potassium < 3.5 mmol/L) and Hyponatremia (serum sodium < 135 mmol/L). *P < 0.05, **P < 0.01, ***P < 0.001

Multivariate logistic regression analysis

To control for confounding factors and identify independent predictors of MPP mixed with viral pneumonia, variables with P < 0.05 according to univariate analysis were entered into a binary logistic regression model. The results are shown in Table 2. Older age (OR = 1.097, P = 0.015), onset in spring (OR = 3.028, P = 0.003), crackles (OR = 1.849, P = 0.021), and decreased breath sounds (OR = 2.270, P = 0.036) were independent predictors of MPP mixed with viral pneumonia. See Table 2 for details.

Table 2.

Multivariate logistic regression analysis of risk factors for mixed viral pneumonia

Variable	B	SE	Wald χ2	P	OR	95% CI
Age	0.093	0.038	5.886	0.015	1.097	1.018–1.182
Spring	1.108	0.376	8.704	0.003	3.028	1.450–6.321
Summer	-0.628	0.377	2.778	0.096	0.534	0.255–1.117
Autumn	-0.362	0.367	0.97	0.325	0.697	0.339–1.430
Vomiting	-0.361	0.267	1.832	0.176	0.697	0.413–1.176
Crackles	0.615	0.266	5.338	0.021	1.849	1.098–3.115
Decreased breath sound	0.82	0.392	4.385	0.036	2.270	1.054–4.890
Constant	-0.427	0.439	0.944	0.331	0.653	-

Note: 95% CI stands for 95% confidence interval. Dependent variable coding: MP mixed viral pneumonia = 1, MP mixed bacterial pneumonia = 0

Comparison of immune indicators

The percentage of patients with abnormal complement C3 levels in the bacterial pneumonia group (9.85%) was significantly greater than that in the viral pneumonia group (4.60%) (P = 0.049). There were no significant differences in other immune indicators (C4, immunoglobulins, lymphocyte subsets, and so forth) between the two groups (P > 0.05). See Table 3.

Table 3.

Comparison analysis of abnormal results of immune function

Lab test	MP mixed with bacterial pneumonia group (n = 138)	MP mixed with viral pneumonia group (n = 254)	χ2	P
C3 (> 1.8 g/L, < 0.9 g/L)	9.85% (13/132)	4.60% (11/239)	3.868	0.049*
C4 (> 0.38 g/L, < 0.16 g/L)	55.30% (73/132)	49.37% (118/239)	1.197	0.274
IgA (> 2.22 g/L, < 0.38 g/L)	26.52% (35/132)	20.50% (49/239)	1.755	0.185
IgG (> 14.3 g/L, < 5.9 g/L)	16.67% (22/132)	17.15% (41/239)	0.014	0.905
IgM (> 2.08 g/L, < 0.45 g/L)	8.55% (10/117)	15.64% (33/211)	3.324	0.068
CD3+ (> 83%, < 55%)	12.82% (15/117)	10.43% (22/211)	0.431	0.511
CD3 + CD4+ (> 57%, < 28%)	12.82% (15/117)	14.22% (30/211)	0.124	0.725
CD3 + CD8+ (> 39%, < 10%)	2.56% (3/117)	2.84% (6/211)	0.022	0.882
CD4+/CD8+ (> 1.94%, < 0.98)	44.44% (52/117)	34.60% (73/211)	3.094	0.079
CD3-CD19+ (> 19%, < 6%)	51.28% (60/117)	48.34% (102/211)	0.260	0.610
CD3-CD(16 + 56)+ (> 31%, < 7%)	31.62% (37/117)	31.75% (67/211)	0.001	0.981
CD19 + CD23+ (> 9.7%, < 3.8%)	44.44% (52/117)	54.03% (114/211)	2.766	0.096

Note: Some data were missing for certain indicators. Percentages shown in the table were calculated based on the number of patients who completed the corresponding tests. *P < 0.05

Comparison of coagulation function

The rate of D-D increase in the viral pneumonia group was greater than that in the bacterial pneumonia group (44.96% vs. 33.06%, P = 0.029), whereas the rate of prolonged APTT was lower in the viral pneumonia group than in the bacterial pneumonia group (72.80% vs. 61.76%, P = 0.036). The details are shown in Table 4.

Table 4.

Comparison analysis of abnormal results of coagulation function

Lab test	MP mixed with bacterial pneumonia group (n = 138)	MP mixed with viral pneumonia group (n = 254)	χ2	P
PT (> 12.1 s)	91.20% (114/125)	94.12% (224/238)	1.088	0.297
APTT (> 36.5s)	72.80% (91/125)	61.76% (147/238)	4.421	0.036*
Fib (> 3.5 g/L)	84.80% (106/125)	86.97% (207/238)	0.326	0.568
TT (> 21s)	0	0.42% (1/238)	0.527	0.468
AT-III (> 120%, < 80%)	20.80% (26/125)	30.25% (72/238)	3.715	0.054
D-D (> 550ug/L)	33.06% (41/124)	44.96% (107/238)	4.771	0.029*
FDP (> 5000ug/L)	9.26% (5/54)	7.1% (13/183)	0.276	0.599

Note: Some data were missing for certain indicators. Percentages shown in the table were calculated based on the number of patients who completed the corresponding tests. *P < 0.05

Comparison of inflammatory indicators

The rate of increase in PCT in the bacterial pneumonia group was significantly greater than that in the viral pneumonia group (20.99% vs. 5.68%, P < 0.001). In contrast, the rate of increase in SAA in the viral pneumonia group was significantly greater than that in the bacterial pneumonia group (94.29% vs. 68.57%, P = 0.006). No significant differences were observed in the abnormal rates of HBP and sCRP between the two groups (P > 0.05). See Table 5.

Table 5.

Comparative analysis of abnormal results of inflammatory indicators

Lab test	MP mixed with bacterial pneumonia group (n = 138)	MP mixed with viral pneumonia group (n = 254)	χ2	P
PCT (> 0.5 ng/mL)	20.99% (17/81)	5.68% (10/176)	13.821	< 0.001***
SAA (> 10 mg/L)	68.57% (24/35)	94.29% (33/35)	7.652	0.006**
HBP (> 11.4 ng/mL)	86.67% (52/60)	84.85% (56/66)	0.085	0.771
sCRP (> 8 mg/L)	61.03% (83/136)	54.22% (135/249)	1.662	0.197

Note: Some data were missing for certain indicators. Percentages shown in the table were calculated based on the number of patients who completed the corresponding tests. **P < 0.01, ***P < 0.001

Analysis of viral coinfection patterns

Analysis of viral coinfection patterns revealed that a single virus was the main pattern (85%; 217/254) (Fig. 2A). In terms of single viral infection, HRV ranked first (52%, 133/254) (Fig. 2B). In dual viral coinfection, common viral combinations included HRV/HPIV (3%, 7/254) and HRV/RSV (2%, 5/254) (Fig. 2C). Coinfection with three or more viruses is relatively rare.

Fig. 2.

Analysis of viral coinfection patterns

Analysis of the viral spectrum

A total of 300 virus strains were detected in children with MPP. HRV had the highest percentage (53.67%, 161/300), followed by HCOV (11.67%, 35/300) and RSV (11%, 33/300). The number of virus-positive cases peaked in summer and was lowest in winter. Other pathogens were detected sporadically. The details are presented in Fig. 3.

Fig. 3.

Analysis of detected virus in different seasons

Annual composition of the bacterial spectrum

A total of 139 bacterial strains were isolated from 138 children with bacterial coinfection (one child had both S. pneumoniae and H. influenzae detected simultaneously). S. pneumoniae was the absolute dominant pathogen, accounting for 58.99% (82/139), followed by H. influenzae (24/139, 17.27%) and S. aureus (23/139, 16.55%). See Fig. 4.

Fig. 4.

Annual distribution of bacterial strains in different years

Drug susceptibility results of S. pneumoniae

As shown in Table 6, S. pneumoniae was fully resistant to Erythromycin and persistently highly resistant to Tetracycline (84.62%-90.48%) from 2021 to 2023. Trimethoprim‒sulfamethoxazole resistance decreased to 61.90% in 2023. Cefotaxime resistance and amoxicillin resistance peaked at 38.46% in 2022 and decreased in 2023. Penicillin G and Chloramphenicol resistance was less than 10%, while susceptibility to five drugs, including Linezolid and Vancomycin, was 100%. In brief, the strain is highly resistant to macrolides but susceptible to most reserve antibiotics.

Table 6.

Annual analysis of drug resistance rates of S. pneumoniae

Antimicrobial Agents	2021 (n = 34)	2022 (n = 13)	2023 (n = 21)
Erythromycin	100% (34/34)	100% (13/13)	100% (21/21)
Clindamycin	100% (34/34)	100% (13/13)	-
Tetracycline	85.29% (29/34) [95% CI: 68.9–95.0%]	84.62% (11/13) [95% CI: 54.6–98.1%]	90.48% (19/21) [95% CI: 69.6–98.8%]
Sulfamethoxazole-trimethoprim	79.41% (27/34) [95% CI: 62.1–91.3%]	84.62% (11/13) [95% CI: 54.6–98.1%]	61.90% (13/21) [95% CI: 38.4–81.9%]
Quinupristin/dalfopristin	61.76% (21/34) [95% CI: 43.6–77.8%]	84.62% (11/13) [95% CI: 54.6–98.1%]	-
Cefotaxime	11.76 (4/34) [95% CI: 3.3–27.5%]	38.46% (5/13) [95% CI: 13.9–68.4%]	14.29% (3/21) [95% CI: 3.0–36.3%]
Amoxicillin	11.76 (4/34) [95% CI: 3.3–27.5%]	38.46% (5/13) [95% CI: 13.9–68.4%]	28.57% (6/21) [95% CI: 11.3–52.2%]
Chloramphenicol	8.82% (3/34) [95% CI: 1.9–23.7%]	7.69% (1/13) [95% CI: 0.2–36.0%]	4.76% (1/21) [95% CI: 0.1–23.8%]
Penicillin G	0	0	4.76% (1/21) [95% CI: 0.1–23.8%]
Rifampicin	0% (0/34)	0% (0/13)	-
Linezolid	0% (0/34)	0% (0/13)	-
Levofloxacin	0% (0/34)	0% (0/13)	0% (0/21)
Vancomycin	0% (0/34)	0% (0/13)	0% (0/21)
Moxifloxacin	0% (0/34)	0% (0/13)	0% (0/21)

Note: S. pneumoniae was the major pathogen, with 82 strains isolated in total. Among them, 68 strains were dominant strains in pure culture from qualified sputum samples and were included in the final analysis of drug susceptibility results; the remaining 14 strains failed to complete full-set drug susceptibility testing due to culture conditions or specimen quality issues. “-” indicates that the corresponding antibiotic was not tested. 95% CI stands for 95% confidence interval

Discussions

Mycoplasma pneumoniae pneumonia is the leading cause of community-acquired pneumonia in children aged 3–15 years [1]. Coinfection with viruses or bacteria has become a key factor exacerbating disease severity and complicating clinical management. Previous studies have reported that coinfection with viruses or bacteria is associated with more severe clinical symptoms and higher complication rates compared with single MPP infection [11–12]. However, because the present study did not include a single MPP control group, this comparison remains to be verified. This study aimed to further compare the differences in clinical characteristics, immune function, coagulation parameters, and inflammatory indicators between children with MPP mixed with viral pneumonia and those with MPP mixed with bacterial pneumonia, so as to describe the distinct profiles of the two conditions.

The results revealed that children with MPP mixed with viral pneumonia were older than those with MPP mixed with bacterial pneumonia, which is consistent with immune developmental characteristics: older children have a wider range of activities and a higher risk of viral exposure, whereas younger children have immature immune systems and are more susceptible to secondary bacterial infections [13]. In terms of onset seasons, viral pneumonia peaked in spring, whereas bacterial pneumonia predominated in summer and autumn, which may be related to increased viral activity and a variable climate in spring, which reduces respiratory mucosal resistance, as well as suitable temperatures for bacterial proliferation in summer and autumn. These findings are consistent with those reported by Li et al. in 2026, who reported a peak of MPP mixed with viral pneumonia in spring [14]. In terms of symptoms and pulmonary signs, the incidence of vomiting was greater in the MPP mixed with bacterial pneumonia group, whereas crackles and decreased breath sounds were more common in the MPP mixed with viral pneumonia group.

To further control for potential confounders, multivariate logistic regression analysis was performed (Table 2). The results revealed that older age, spring onset, crackles, and decreased breath sounds were independent predictors of MP-mixed viral pneumonia (compared with MP-mixed bacterial pneumonia; all P < 0.05). The independent effect of spring (OR = 3.028) aligns with the high prevalence of HRV in spring (53.67%) in our viral spectrum and with spring viral peaks reported in other regions of China [15]. Notably, vomiting, which differed in univariate analysis, did not reach statistical significance in the regression model (P = 0.176), suggesting that the association between vomiting and bacterial coinfection may be confounded by age or season. Integrating univariate and multivariate findings, age, spring season, and the presence of crackles or decreased breath sounds may help clinicians raise suspicion of viral pneumonia (vs. bacterial pneumonia), although prospective validation is needed.

On the basis of the identified differences in clinical characteristics, we further analysed the immune status of the two groups. The results revealed a greater abnormal rate of complement C3 in the MPP mixed with bacterial pneumonia group, suggesting that early bacterial infection may activate the complement system via the classical or alternative pathway, which is involved in pathogen clearance and inflammatory responses. Abnormal elevation or reduction of C3 may reflect stronger immune activation induced by bacterial infection [15]. However, there were no significant differences in immunoglobulins or T lymphocyte subsets between the two groups. This may be because MPP itself has already been shown to affect children’s immune function, masking the specific immune differences between bacterial and viral pneumonia. These findings also indicate that changes in immune function during MPP coinfection may be more closely related to the MPP itself rather than to the type of secondary pathogen.

Coagulation dysfunction is a common manifestation of MPP, and its mechanism is associated with MPP-induced cytokine release, vascular injury, and increased thrombotic risk [16]. In this study, the rate of prolonged APTT was greater in the bacterial pneumonia group, which may be related to direct endothelial cell damage and activation of the intrinsic coagulation pathway by bacteria. In contrast, D-D elevation was more significant in the viral pneumonia group, suggesting that viral pneumonia is more likely to trigger systemic inflammatory responses and fibrinolytic system activation, increasing the risk of thrombosis. Previous studies have identified D-D as an independent risk factor for thrombosis [17]. Close monitoring of coagulation function may be beneficial for children with MPP, especially those with mixed infections, but further comparisons with single MPP is needed.

Among the inflammatory indicators, PCT and SAA showed significant intergroup differences. PCT was significantly elevated in the bacterial pneumonia group but not obviously increased in the viral pneumonia group, which is consistent with the known biology of PCT [18]. SAA showed a higher elevation rate in the viral pneumonia group compared with the bacterial pneumonia group, in line with the characteristic rapid early elevation of SAA in viral pneumonia group [19]. No significant differences were detected in serum levels of HBP or sCRP between the two groups (P > 0.05). In our univariate comparison, PCT levels were higher in the bacterial pneumonia group, while SAA levels were higher in the viral pneumonia group, indicating different distribution patterns of these two markers between the groups. However, their clinical value, whether used alone or in combination, is limited and should be interpreted alongside other clinical parameters.

Pathogen spectrum analysis revealed that children with MPP were predominantly complicated with single-virus infection, accounting for 85.0% (217/254), while dual or multiple viral coinfections were also common. The most frequent dual combinations were HRV/HPIV (3.0%, 7/254) and HRV/RSV (2.0%, 5/254), indicating the necessity of multiplex pathogen screening for paediatric MPP patients. HRV was the predominant local strain, with 161 positive cases (53.7%), which was detectable year-round and peaked in summer, which was consistent with the findings of a previous study [20], followed by the results of a study of HCOV (11.67%) and RSV (11%). Distinct regional discrepancies in the dominant viral spectrum were observed when compared with published data: ADV ranked first in mixed infections in Liaoning, China [21], HPIV and SARS-CoV-2 were the leading copathogens in Russia [22], and RSV together with HRV represented the most prevalent coinfecting viruses among hospitalised children according to a North American multicentre study [23]. Such geographical disparities may be attributed to the climate of East China, population-based immunity of children and locally circulating pathogen strains, suggesting that respiratory pathogen surveillance should be region-specific and that clinical management strategies cannot be arbitrarily applied across different regions.

Among the bacterial isolates, 75.54% were gram-positive. The top three bacterial pathogens were S. pneumoniae, H. influenzae, and Staphylococcus aureus, in descending order. This finding is consistent with the distribution of common pathogens causing community-acquired pneumonia [24–25]. Susceptibility testing revealed that the S. pneumoniae isolates were 100% resistant to Erythromycin and Clindamycin, precluding empirical treatment with macrolides and lincosamides. Bisetegn et al. reported generally low antimicrobial resistance, with a maximum Tetracycline resistance rate of only 46.27% [26]. In contrast, Tetracycline resistance in our study reached 84.62%-90.48%, which conflicts with the findings of Munson et al., who reported that more than 91% of patients were susceptible to Tetracycline [27]. In 2023, the resistance to Trimethoprim-sulfamethoxazole decreased compared to 2021–2022, which is consistent with the declining epidemiological trend described by Ling et al. [28]. Resistance to Cefotaxime and Amoxicillin fluctuates yearly, peaking at 38.46% in 2022 before decreasing to low-to-moderate levels (14.29% and 28.57%, respectively) in 2023, supporting the ongoing clinical use of β-lactams. Penicillin G maintained good susceptibility, with a 2023 resistance rate of merely 4.76%, which is distinct from the 31.8% overall Penicillin resistance reported in a 2025 study [29]. Discrepancies in resistance profiles can be attributed to regional serotype distribution, local antibiotic consumption patterns, enrolled population characteristics, vaccination coverage, study periods and limited sample size. Rifampicin, Linezolid, Levofloxacin, Vancomycin and Moxifloxacin exhibited 100% susceptibility. These susceptibility results guide empirical antibiotic regimens for paediatric mixed infections. Given the small numbers of isolates in 2022 (n = 13) and 2023 (n = 21), temporal resistance trends should be interpreted cautiously and validated by further large-scale surveillance.

In this study, the in vitro susceptibility results for β-lactam antibiotics, such as Penicillin G, Cefotaxime, and Amoxicillin, indicated that most isolates were susceptible to these agents. Monotherapy with highly resistant macrolides should be avoided in children with suspected MPP complicated by pneumococcal infection, which is consistent with the opinion of Munson et al. [27]. However, individualised medication should be determined on the basis of disease severity, local guidelines, antimicrobial stewardship and risks of bacterial colonisation.

Nevertheless, this study has several limitations. First, this study lacks a control group of a single MPP group. Therefore, our conclusions are limited to the comparisons between the two mixed infection types and cannot be extrapolated to comparisons between mixed infections and single infections. Second, the detection of these markers is costly and not routinely available in all healthcare facilities; therefore, more cost-effective and simplified marker combinations need to be explored in the future. Third, the limited number of isolates obtained for certain common bacterial species precluded reliable resistance profiling, and sputum culture (even with quality criteria and clinical judgment) may not reliably differentiate true lower respiratory tract infection from colonization. These limitations should be addressed in future multicenter studies with larger sample sizes. Furthermore, this study is a retrospective, single-centre analysis that included only hospitalised children and did not cover mild outpatient cases, which may introduce selection bias. These limitations should be considered when interpreting the results. Future research may incorporate comprehensive pathogen detection techniques such as metagenomic next-generation sequencing in multicentre, prospective studies to further elucidate the dynamic changes in coinfections, pathogen interactions, and their impact on prognosis, thereby providing higher-level evidence for the development of individualised treatment and prevention strategies.

In summary, differences exist among children with MPP mixed with bacterial pneumonia and those with MPP mixed with viral pneumonia in terms of age at onset, season of onset, clinical symptoms, immune function, coagulation function, and inflammatory indicators. PCT, SAA, and D-D levels were significantly different between the two groups, but their clinical reference values are limited and should be interpreted in conjunction with clinical manifestations.

Author contributions

YX Z and Y J conceived the study and designed the experiments. X Z provide financial support and collected the data. YX Z and Y J analysed the data and interpreted the results. YX Z and Y J drafted the manuscript, and all authors critically revised the manuscript for intellectual content, and read and approved the final manuscript.

Funding

Suzhou Municipal Health Commission (KJXW2023025), Guided Project of Jiangsu Provincial Natural Science Foundation (Z2025037).

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Declarations

Ethics and consent to participate

The project is a retrospective study of data and only includes specific data. It was reviewed and approved by the Medical Ethics Committee of the Children's Hospital of Soochow University (Ethics batch number: 2025CS239).

Consent for publication

Not applicable. This manuscript does not contain any individual person’s data in any form (including individual details, images or videos).

Conflict of interest

All other authors report no potential conflicts of interest.