Efficacy and safety of doxycycline for severe Mycoplasma pneumoniae pneumonia in pediatric patients

Abstract

This study evaluated the efficacy and safety of doxycycline in treating Severe Mycoplasma pneumoniae Pneumonia (SMPP) in children under eight years old through clinical analysis and meta-analysis. A total of 92 pediatric SMPP cases were divided into a doxycycline treatment group (44 cases) and a macrolides control group (48 cases). Compared to the control group, the doxycycline group exhibited significantly shorter cough relief time (5.4 ± 1.2 vs. 7.2 ± 1.6 days, p < 0.05) and pulmonary rale resolution time (6.2 ± 1.3 vs. 8.0 ± 1.7 days, p < 0.05). The overall treatment efficacy rate was higher in the doxycycline group (88.6% vs. 75.0%, p < 0.05). No significant differences were found in fever resolution time or hospitalization duration (p > 0.05). Safety analysis revealed comparable adverse event rates between groups (18.2% vs. 16.7%, p > 0.05), primarily mild rash and gastrointestinal discomfort, with no tooth discoloration observed. The meta-analysis confirmed the advantages of doxycycline, demonstrating superior treatment efficacy (RR: 0.68, 95% CI: 0.58–0.79), shorter fever resolution (MD: − 1.5 days, 95% CI: − 2.3 to − 0.7), and faster cough and pulmonary rale resolution. Adverse events were similar across groups. These findings highlight doxycycline’s clinical efficacy and safety in SMPP treatment, providing strong evidence for its application in pediatric practice.

Graphical Abstract

Summary of the efficacy and safety of doxycycline in the treatment of SMPP.

Introduction

Severe Mycoplasma Pneumoniae Pneumonia (SMPP) is a severe respiratory infection caused by Mycoplasma pneumoniae, predominantly affecting children under the age of eight [1, 2]. Mycoplasma pneumoniae is a significant cause of infectious pneumonia in children, often presenting with persistent high fever, severe coughing, rapid breathing, and chest pain. In some cases, the disease progresses rapidly, leading to pulmonary complications, including respiratory failure and empyema [3, 4]. Early diagnosis and prompt, effective treatment are crucial to reducing mortality and preventing complications. The infection demonstrates seasonality, with outbreaks typically occurring in autumn and winter. It spreads primarily via respiratory droplets, posing a high risk of outbreaks in crowded environments such as schools and daycare centers [5, 6]. Due to the absence of a cell wall, Mycoplasma pneumoniae is inherently resistant to conventional antibiotics, particularly β-lactam antibiotics [5]. Clinically, macrolides such as azithromycin and clarithromycin are commonly used as first-line treatments. However, the widespread and inappropriate use of antibiotics has led to increasing resistance, particularly in pediatric populations. This growing issue has raised significant concern and highlights the need for alternative therapeutic strategies.

Traditional macrolides have long been the primary treatment for SMPP due to their effective bactericidal activity against Mycoplasma pneumoniae. These antibiotics inhibit protein synthesis in the pathogen; however, prolonged use can lead to the development of drug resistance, particularly in cases involving prolonged illness or delayed treatment [7, 8]. Recent reports from certain regions have highlighted resistance to azithromycin, posing significant challenges for clinicians in managing treatment effectively [9–11]. In addition to resistance issues, macrolides may cause adverse effects, such as gastrointestinal discomfort, allergic reactions, and liver function abnormalities. These side effects are particularly concerning in pediatric populations, where their incidence cannot be overlooked [12–14]. Moreover, some children develop a strong dependency on these antibiotics, resulting in treatment failure. Given these challenges, researchers have increasingly sought alternative antibiotics. Doxycycline, a broad-spectrum antibiotic with potent antimicrobial activity, has emerged as a promising candidate. Notably, it remains effective against resistant strains, making it a focal point for ongoing research [15].

Doxycycline, a tetracycline antibiotic, is widely used to treat infections caused by various bacteria, mycoplasma, and chlamydia. Unlike macrolides, doxycycline functions primarily by inhibiting bacterial protein synthesis. Its low resistance profile makes it particularly effective against resistant bacterial infections [15–17]. For patients with mycoplasma pneumonia, doxycycline effectively suppresses mycoplasma proliferation and shows significant therapeutic benefits in co-infections with other bacteria. Moreover, doxycycline demonstrates high bioavailability and excellent tissue penetration, achieving significantly higher concentrations in lung tissue than plasma, which is critical for treating mycoplasma pneumonia [18]. Clinical studies have shown that doxycycline provides robust therapeutic efficacy and exhibits good tolerability and minimal side effects, particularly during long-term use [19]. Given these attributes, doxycycline has emerged as a promising alternative and is increasingly favored by clinical researchers. However, despite existing studies on its use in adult respiratory infections [20], evidence supporting its efficacy and safety in treating SMPP in children, particularly those under eight, remains insufficient. Systematic studies on this specific population are notably lacking.

Meta-analysis is a powerful statistical method that combines data from multiple independent studies. It helps address limitations such as small sample sizes and study design biases, leading to more reliable conclusions [21]. This method is important in comparing clinical treatment strategies and evaluating their effectiveness. At present, there is a lack of large-scale and systematic data on the use of doxycycline for treating SMPP in children. Therefore, a meta-analysis of existing studies is clinically meaningful for evaluating the efficacy and safety of doxycycline in this setting. It allows for a comprehensive analysis of treatment outcomes, including success rates, symptom resolution times, hospitalization duration, and fever reduction. Additionally, it enables an objective safety assessment by examining the incidence and types of adverse events, providing evidence to support clinical decision-making.

This study aims to evaluate the efficacy and safety of doxycycline in treating SMPP in children under eight years old by integrating retrospective analysis with meta-analysis. Doxycycline is compared with traditional macrolide antibiotics regarding symptom improvement, treatment effectiveness, and time required for symptom resolution. Clinical outcomes such as fever reduction time, cough resolution time, and the duration until improved lung sounds are assessed, along with hospitalization length and the incidence of adverse events. The objective is to provide robust clinical evidence supporting doxycycline as an effective and safe alternative to macrolides, to help reduce antibiotic resistance, and to enhance treatment strategies for pediatric SMPP.

Materials and methods

Clinical information

This study included 92 pediatric patients under 8 diagnosed with SMPP who were hospitalized in the Department of Pediatrics at the First Affiliated Hospital of Bengbu Medical College between July 2023 and July 2024. The inclusion criteria were as follows: (1) Diagnosis of SMPP according to the 2023 Guidelines for the Diagnosis and Treatment of Mycoplasma pneumoniae in Children [22], including persistent high fever (≥ 39 °C) for ≥ 5 days or any fever lasting ≥ 7 days without downward trend; presence of wheezing, dyspnea, chest pain, hemoptysis, or respiratory distress; extrapulmonary complications not meeting critical illness criteria; resting pulse oxygen saturation ≤ 93%; imaging showing lobar pneumonia involving ≥ 2/3 of one lobe, or ≥ 2 lobes with dense consolidation, or ≥ 4/5 lobes with bronchiolitis features; radiologic deterioration > 50% within 24–48 h; and significantly elevated CRP, LDH, or D-dimer. (2) Received ≥ 3 days of macrolide therapy at an external facility without clinical improvement and were subsequently diagnosed with macrolide-unresponsive Mycoplasma pneumoniae pneumonia (MUMPP). (3) Imaging studies indicated substantial pulmonary lesions characteristic of lobar pneumonia; and (4) No clinical data were missing. The exclusion criteria were as follows: (1) a history of allergy to macrolide or tetracycline antibiotics; (2) refusal to undergo bronchoalveolar lavage (BAL); (3) severe congenital anomalies or primary immunodeficiency diseases; and (4) nosocomial infections occurring during treatment. Based on the administered antibiotic therapy, patients were divided into two groups: the Doxycycline group (treatment group, n = 44) and the macrolide group (control group, n = 48) (Fig. 1). Ethical approval for the study was obtained from the institutional ethics committee (approval number: [2024] 212), and informed consent was acquired from the families of all participants.

Fig. 1.

Patient inclusion and exclusion flowchart

Treatment

According to the 2023 Guidelines for the Diagnosis and Treatment of Mycoplasma pneumoniae in Children, the therapeutic principles for SMPP include: (1) early identification and comprehensive management; (2) macrolides as the first-line treatment, with doxycycline or fluoroquinolones as alternatives in resistant cases; (3) use of glucocorticoids in severe cases; (4) bronchoscopic intervention when necessary; (5) short-term intravenous immunoglobulin (IVIG) in cases with severe extrapulmonary complications or intense immune-inflammatory responses; (6) timely drainage in patients with moderate to large pleural effusions; (7) prophylactic anticoagulation when D-dimer levels are significantly elevated; and (8) recognition and management of mixed infections.

Control Group: After discussing treatment options with the families, guardians declined the use of tetracyclines and opted to continue macrolide therapy for Mycoplasma infection. In addition to macrolide antibiotics, symptomatic treatments were provided, including methylprednisolone for inflammation control, ambroxol for mucus reduction, and nebulized inhalation therapy. Following the 2018 Chinese Guidelines for Pediatric Flexible Bronchoscopy, BAL was performed immediately after excluding contraindications. If BAL results indicated bacterial co-infection, appropriate antibiotics were initiated. Supportive treatments, such as airway management, antipyretics, and fluid replacement, were also provided. Treatment efficacy was evaluated through chest Computed Tomography (CT) scans performed one week after the initiation of therapy.

Experimental Group: Similar to the control group, children in the experimental group were hospitalized due to SMPP and had not responded to at least three days of macrolide therapy at another hospital. They were subsequently diagnosed with MUMPP. After obtaining informed consent from their families, oral doxycycline (2 mg/kg every 12 h) was administered for 10 days to treat the Mycoplasma infection. Other treatment measures were consistent with those in the control group. Treatment efficacy was evaluated through chest CT scans one week after starting doxycycline therapy.

In both groups, methylprednisolone was used for anti-inflammatory treatment based on disease severity. The standard dose was 2 mg/kg/day, with adjustments up to 4–6 mg/kg/day in more severe cases, depending on clinical presentation. All patients in both groups received methylprednisolone. There were no significant differences between groups in methylprednisolone dosage, frequency, or treatment duration, indicating balanced glucocorticoid use and minimizing its potential as a confounding variable.

Clinical data collection

Clinical data were gathered from pediatric electronic medical records and included the following parameters: (1) General Information: Age, sex, and illness duration for both children groups. (2) Blood Test Indicators: Baseline measurements at admission, including D-Dimer (D-D), C-Reactive Protein (CRP), Lactate Dehydrogenase (LDH), and White Blood Cell (WBC) counts. (3) Imaging Studies: Ultrasound (USG) and CT were used to document the presence of pleural effusion, pleural thickening, atelectasis, and pulmonary necrosis in both groups Pre- and post-treatment. (4) Bronchoscopy Data: Bronchoscopic findings were categorized as mild or severe according to the 2023 Guidelines for Pediatric Mycoplasma Pneumonia, with results recorded for both groups. (5) Clinical Symptoms: Time to fever resolution, cough alleviation, disappearance of pulmonary rales, and overall hospital stay duration.

Efficacy Evaluation: Efficacy was assessed one week after treatment based on the following criteria: Significant Improvement: Normalized body temperature, resolution of cough and wheezing, disappearance of pulmonary rales, and near-complete resolution of pulmonary infections on chest CT. Effective: On chest CT, there is noticeable improvement in cough and wheezing symptoms, reduction of pulmonary rales, and a substantial reduction in pulmonary infection lesions (≥ 70%). Ineffective: Persistent or worsening symptoms, no improvement or deterioration on pulmonary auscultation, or the occurrence of severe complications. No improvement or worsening of infections was confirmed via chest CT. The overall efficacy rate was calculated as the sum of the significant and effective improvement rates.

Adverse Reaction Monitoring: Adverse events during treatment were carefully monitored, including rashes, tooth discoloration, gastrointestinal symptoms (e.g., nausea, vomiting, diarrhea, and abdominal pain), and any symptoms affecting the hematologic or central nervous systems.

Literature search strategy for meta-analysis

To comprehensively evaluate the efficacy and safety of doxycycline in treating SMPP in children, we conducted a systematic literature search across PubMed, Web of Science, Embase, and CNKI. The search included all relevant studies published up to December 2024 and focused on randomized controlled trials (RCTs) and retrospective studies comparing doxycycline with macrolides for SMPP treatment.

Search terms included disease-related keywords such as "Severe Mycoplasma pneumoniae Pneumonia," "Mycoplasma Pneumoniae," and "Children"; intervention terms such as "Doxycycline," "Tetracyclines," "Macrolides," "Azithromycin," and "Clarithromycin"; and outcome-related terms like "Efficacy," "Effectiveness," "Adverse Events," and "Safety." Boolean operators were used to combine terms appropriately. The search strategy was adapted to each database’s indexing system, using MeSH and free-text terms in PubMed, EMTREE terms in Embase, and keyword/title/abstract searches in Web of Science and CNKI.

Additionally, we manually screened reference lists and conference abstracts to ensure completeness. Although the focus was on English and Chinese literature, no language restrictions were applied.

Inclusion and exclusion criteria for literature

To ensure the methodological rigor of this systematic review and meta-analysis, we predefined strict inclusion and exclusion criteria. Eligible studies included RCTs or retrospective studies involving pediatric patients diagnosed with severe Mycoplasma pneumoniae pneumonia (SMPP), in which doxycycline or macrolides were used as the primary treatment. Studies were required to report extractable data on clinical efficacy outcomes, such as symptom resolution time, treatment success rates, and safety indicators, including the incidence of adverse events.

Studies were excluded if they involved non-pediatric populations or other types of pneumonia, did not use doxycycline or macrolides as the main intervention, lacked a control group, or failed to report relevant efficacy or safety data. Animal studies, case reports, review articles, duplicate publications, and studies with incomplete or non-extractable data were also excluded.

This review evaluated doxycycline with a comparative analysis against macrolide monotherapy (e.g., azithromycin or clarithromycin). Studies involving additional supportive therapies were included only if the core antibiotic regimen remained consistent. By applying these criteria, we aimed to include high-quality studies with reliable methodology, thereby ensuring the accuracy and credibility of the meta-analysis and strengthening the evidence supporting the clinical use of doxycycline in pediatric SMPP.

Literature coding and quality evaluation

The included studies were systematically coded to extract key variables, including author, publication year, study design (RCT or retrospective), sample size, age range, intervention type (doxycycline or macrolides), dosage, administration frequency, treatment efficacy, and adverse events. Data were organized in Excel for standardized storage and ease of analysis. The quality of randomized controlled trials was assessed using the Cochrane Risk of Bias tool, covering six domains: random sequence generation, allocation concealment, blinding, completeness of outcome data, and selective reporting, ensuring methodological rigor and result reliability.

Data extraction

During the data extraction process, we focus on two core categories of data: efficacy indicators and safety indicators. Efficacy indicators include cough resolution time, disappearance of lung rales, length of hospital stay, and overall treatment effectiveness. Safety indicators include the incidence and specific classifications of adverse events, such as photosensitivity dermatitis, gastrointestinal symptoms, and dental discoloration. Additionally, we record sample group information (experimental and control groups) for each study, statistical significance (p-values), confidence intervals (CI), and heterogeneity-related data. Two researchers perform The data extraction process independently, with results cross-checked to ensure accuracy and completeness. A third researcher resolves any discrepancies. This rigorous data extraction procedure aims to provide high-quality foundational data for subsequent analyses, thereby ensuring the scientific integrity of this study.

Statistical methods

Statistical analysis was conducted using SPSS version 26.0. Continuous variables with normal distribution were presented as means ± standard deviations and compared using independent-samples t-tests. Non-normally distributed data were expressed as medians [p50 (p25, p75)] and analyzed using the Mann–Whitney U test. Categorical variables were reported as frequencies (n, %) and compared using the chi-square test. A two-sided p-value of < 0.05 was considered statistically significant.

Meta-analyses were performed using RevMan 5.4 and R software. Dichotomous outcomes were expressed as risk ratios (RR) with 95% CI, while continuous outcomes were analyzed using standardized mean differences (SMD). Heterogeneity was assessed using Cochran’s Q test and the I² statistic, with I² > 50% indicating significant heterogeneity. Sensitivity analyses were performed to evaluate the robustness of the results and the influence of individual studies. Publication bias was assessed through funnel plots and quantified using Egger’s test. A significance level of p < 0.05 was applied throughout.

Results

Baseline characteristics of the two groups

A total of 92 children under 8 years old diagnosed with SMPP were included and divided into the doxycycline group (n = 44) and the macrolide group (n = 48). Table 1 shows no statistically significant differences between the two groups regarding age, gender, or disease duration (p > 0.05). Laboratory indicators, including serum D-dimer, CRP, LDH, and WBC count, were comparable between groups (p > 0.05).

Table 1.

Comparison of general data of children between the two groups

Project	Experimental group (n = 44)	Control group (n = 48)	t/Z/χ2 value	p-value
Age (years)	6.30 ± 1.32	5.87 ± 1.06	1.754	0.083
Sex (male/female, case)	21/23	24/24	0.047	0.828
Disease duration [P50 (P25, P75)d]	7 (6.0, 10.0)	7 (6.0, 10.0)	− 0.464	0.643
D-D [P50 (P25, P75) mg/L]	0.80 (0.37,1.35)	0.50 (0.32,1.35)	− 1.263	0.207
CRP [P50 (P25, P75) mg/L]	12.43 (5.0, 24.82)	14.74 (5.0, 26.76)	− 0.024	0.981
LDH[P50 (P25, P75) U/L]	364.5 (290.0, 436.5)	333.5 (302.0, 396.75)	− 1.149	0.251
WBC count [P50 (P25, P75) × 109/L]	8.49 (6.40, 12.26)	9.02 (6.05, 11.47)	− 0.199	0.842
Pleural effusion [cases (%)]	14 (31.82)	6 (12.5)	5.036	0.025

The incidence of pleural effusion was significantly higher in the doxycycline group than in the macrolide group (31.82% vs. 12.5%, p = 0.025; Fig. 2).

Fig. 2.

The incidence of pleural effusion is significantly higher in the doxycycline treatment group compared to the control group

Bronchoscopic findings associated with doxycycline treatment

To further assess pulmonary pathology between groups, fiberoptic bronchoscopy findings were compared. A significantly higher proportion of severe inflammatory lesions was observed in the doxycycline group (16/28) compared to the control group (15/48), while the control group had more mild cases (33/48). The difference was statistically significant (χ² = 9.673, p = 0.002; Fig. 3A), indicating more extensive airway inflammation in the doxycycline group. Mixed infection rates were similar between the two groups, with 6 cases (13.64%) in the doxycycline group and 8 cases (16.67%) in the control group (p = 0.686). The average number of BAL procedures was higher in the doxycycline group (1.59 ± 0.69) than in the control group (1.31 ± 0.55), with a statistically significant difference (t = 2.141, p = 0.035; Fig. 3B), possibly reflecting greater disease complexity in the doxycycline group.

Fig. 3.

Comparison of microscopic findings and BAL frequency between doxycycline and control groups in pediatric patients. Note A Comparison of light and heavy microscopic findings between experimental and control groups; B comparison of bal frequency between experimental and control groups

Effect of doxycycline on cough relief and lung

Key clinical indicators were compared between the two groups. The median fever resolution time was 1.0 days in both groups (p50 = 1.0, p25 = 1.0, p75 = 1.0), with no statistically significant difference (Z = − 0.372, p = 0.710; Fig. 4A). The average cough resolution time was 6.59 ± 2.51 days in the doxycycline group and 7.92 ± 2.56 days in the control group (t =  − 2.506, p = 0.014; Fig. 4B). The average time to lung rales resolution was 6.27 ± 3.11 days in the doxycycline group and 7.71 ± 3.11 days in the control group, showing a statistically significant difference (t =  − 2.214, p = 0.029; Fig. 4C). The mean duration of hospitalization was 10.68 ± 2.36 days in the doxycycline group and 11.40 ± 2.15 days in the control group, with no significant difference (t =  − 1.518, p = 0.132; Fig. 4D). These findings indicate that, compared with the control group, the doxycycline group demonstrated shorter times for cough relief and lung rales resolution, while fever resolution and hospital stay durations were similar between groups.

Fig. 4.

Comparison of clinical observation indicators between the doxycycline and control groups. Note A Comparison of fever resolution time: bar chart showing the median (p50) and interquartile range (p25, p75) of fever resolution time in both groups. Statistical analysis was performed using the rank-sum test, p > 0.05. B Comparison of cough relief time: box plot showing the cough relief time in the doxycycline group (n = 44) and the control group (n = 48). Statistical analysis was performed using the t-test, *p < 0.05. C Comparison of lung rales disappearance time: box plot showing the lung rales disappearance time in both groups. Statistical analysis was performed using the t-test, *p < 0.05. D Comparison of hospital stay: box plot showing the hospital stay in both groups. Statistical analysis was performed using the t-test, p > 0.05

Effect of doxycycline treatment on pulmonary imaging outcomes

Chest imaging findings before and after treatment were analyzed in both groups (Table 2). Before treatment, the incidence of pleural effusion was 31.82% (14 cases) in the doxycycline group and 12.5% (6 cases) in the control group, with a statistically significant difference (χ² = 5.036, p = 0.025). After treatment, the incidence decreased to 6.82% (3 cases) and 4.17% (2 cases), respectively, with no significant difference between the groups (p = 0.920). The incidence of pleural thickening was 9.09% in the doxycycline group and 6.25% in the control group before treatment. Post-treatment rates were 9.09 and 8.33%, respectively (p = 0.905). Atelectasis was observed in 4.55% (2 cases) of patients in the doxycycline group before treatment, while no cases were reported in the control group. After treatment, atelectasis resolved in the doxycycline group, while 4.17% (2 cases) developed in the control group (p = 0.226). No lung necrosis was observed before treatment in either group. Following treatment, 2 cases (4.55%) of lung necrosis were reported in the doxycycline group and none in the control group (p = 0.226).

Table 2.

Comparison of imaging findings between the two groups of children (Case %)

Groups	n	Pleural effusion		Pleural thickening		Pulmonary atelectasis (medicine)		Pulmonary necrosis
		Pre-treatment	Post-treatment	Pre-treatment	Post-treatment	Pre-treatment	Post-treatment	Pre-treatment	Post-treatment
Experimental group	44	14 (31.82)	3 (6.82)	4 (9.09)	4 (9.09)	2 (4.55)	0	0	2 (4.55)
Control subjects	48	6 (12.5)	2 (4.17)	3 (6.25)	4 (8.33)	0	2 (4.17)	0	0
χ2 value		5.036	0.01	0.014	0.000
p-value		0.025	0.920	0.905	1.0	0.226	0.495	–	0.226

In summary, pleural effusion incidence was significantly higher in the doxycycline group at baseline but decreased in both groups after treatment. No statistically significant differences between groups in pleural thickening, atelectasis, or lung necrosis following treatment were observed.

Clinical efficacy assessment in the doxycycline group

Treatment response was assessed in both groups to compare the clinical efficacy between the two treatment regimens (Table 3). In the doxycycline group, 17 patients (38.64%) showed significant improvement, compared to 11 (22.92%) in the control group. The number of effective cases was 22 (50.00%) in the doxycycline group and 26 (54.17%) in the control group. The number of ineffective cases was 5 (11.36%) in the doxycycline group and 12 (25.00%) in the control group. The total efficacy rate was 88.64% in the doxycycline group and 75.00% in the control group, with a statistically significant difference (Z = − 2.018, p = 0.044). These findings indicate a higher proportion of favorable treatment responses in the doxycycline group compared to the control group.

Table 3.

Comparison of treatment efficacy in children [Cases (%)]

Groups	Number of examples	Obvious effect	Effective	Ineffective	Efficiency percent
Experimental group	44	17 (38.64)	22 (50.00)	5 (11.36)	88.64
Control subjects	48	11 (22.92)	26 (54.17)	12 (25.00)	75.00
Z			− 2.018
p-value			0.044

Safety of doxycycline comparable

Adverse events were recorded and compared between the doxycycline and macrolide groups (Table 4). In the doxycycline group, 2 cases (4.55%) of rash were reported, compared to 1 case (2.08%) in the control group. The difference was not statistically significant (p = 0.939). Gastrointestinal symptoms, including nausea, vomiting, and diarrhea, occurred in 5 cases (11.36%) in the doxycycline group and 7 cases (14.58%) in the control group (p = 0.647). No cases of dental discoloration were observed in either group. Overall, there were no statistically significant differences in the incidence of adverse events between the two groups (p > 0.05).

Table 4.

Comparison of the occurrence of adverse events between the two groups of children [n (%)]

Groups	Number of examples	Rash	Digestive system symptoms	Yellowing of teeth
Experimental group	44	2 (4.55)	5 (11.36)	0 (0.00)
Control subjects	48	1 (2.08)	7 (14.58)	0 (0.00)
χ2 value		0.006	0.210
p-value		0.939	0.647

Literature selection and quality assessment

A total of 572 articles were identified from PubMed, Web of Science, CNKI, and WanFang Data (Fig. 5A). After removing duplicates, 403 articles remained for screening. Based on the inclusion criteria, 247 studies were excluded for non-randomized design or irrelevance to SMPP treatment in children. Full-text review was conducted for 156 articles, of which 141 were excluded due to lack of specific data, incomplete information, or insufficient quality. Finally, 15 studies were included in the quantitative meta-analysis. Key characteristics of the included studies are summarized in Table 5. Study quality was evaluated using the Cochrane Risk of Bias tool (Fig. 5B), assessing domains such as random sequence generation, allocation concealment, blinding, and outcome assessment. Each item was categorized as low, high, or unclear risk of bias.

Fig. 5.

Literature screening process and quality assessment of included studies. Note A Flowchart of the literature inclusion process; B summary of bias risk assessment for the included studies

Table 5.

Information about the literature that met the inclusion criteria

First author	Publication time	Sample source	Sample size	Age of subject	Case Presentation	Diagnosis	Treatment	Outcome
Qiu Xiaoqi [33]	2024	Guangdong Pediatric Hospital	80	6–12 years	Refractory Mycoplasma Pneumonia	Serology and clinical criteria	Doxycycline and supportive care	Improved lung function and reduced inflammation
Yang Xia [34]	2024	Chongqing Children’s Hospital	120	7–14 years	Severe Mycoplasma Pneumoniae Pneumonia	Radiological and microbiological evidence	Sequential doxycycline and azithromycin	Reduced symptoms and hospital stay
Chen Ying [35]	2024	Hunan Children’s Hospital	100	8–14 years	Azithromycin-resistant Mycoplasma Pneumonia	Clinical features and resistance testing	Doxycycline vs. azithromycin	Faster symptom resolution and better recovery
Pang Ying [36]	2021	Sichuan Provincial Maternity and Child Health Care Hospital	98	8–12 years	Severe cases with complications	Serology and imaging	Doxycycline with steroids	High cure rate with reduced complications
Cui Xueqin [37]	2024	Anji County Third People’s Hospital	120	9–14 years	Azithromycin-resistant Mycoplasma Pneumonia	Resistance testing and clinical symptoms	Doxycycline with symptomatic care	Effective recovery with good safety profile
Shen Guomei [38]	2024	First Affiliated Hospital of Bengbu Medical College	92	< 8 years	Refractory Mycoplasma pneumoniae pneumonia	Based on 2023 Pediatric Guidelines	Doxycycline and methylprednisolone	Improved resolution of symptoms
Chen Yanfen [39]	2024	Huzhou Central Hospital, Zhejiang Province	120	< 8 years	Refractory Mycoplasma pneumoniae pneumonia	Based on specific diagnostic criteria and imaging	Doxycycline and azithromycin	Higher efficiency with shorter symptom relief time
Fu Jing [40]	2024	Puyang People’s Hospital, Henan Province	118	< 18 years	Refractory Mycoplasma pneumoniae pneumonia	Confirmed by clinical and serological tests	Doxycycline and azithromycin	Enhanced treatment efficacy
He Xiujiao [41]	2015	Maoming Traditional Chinese Medicine Hospital, Guangdong Province	62	8–17 years	Refractory Mycoplasma pneumoniae pneumonia	Standard Mycoplasma Pneumoniae criteria	Doxycycline and macrolides	Significant clinical improvement
Xiao Fengxiang [42]	2024	Gutian County Hospital, Fujian Province	126	3–13 years	Mycoplasma pneumoniae pneumonia	Pediatric consensus criteria	Doxycycline and Pulike mixture	Improved symptom resolution and reduced inflammation markers
Tao Xingru [43]	2021	Children’s Hospital, Shanghai	100	2–13 years	MP pneumonia with cough, fever	Based on MP-PCR testing	Levofloxacin + Doxycycline	Improved lung function and reduced symptoms
Zhou Mingshan [44]	2024	Guangdong Provincial Hospital	120	3–12 years	Refractory MP pneumonia with complications	CT imaging and bronchoscopy	Doxycycline combined with supportive therapy	Reduced fever and cough duration
Zhang Daoguo [45]	2024	Zhejiang Provincial Hospital	110	2–14 years	MP pneumonia resistant to macrolides	Clinical and lab findings	Doxycycline and Qingxuan Zhike Granules	Improved clinical metrics
Zhang Man [46]	2023	First Affiliated Hospital of Zhengzhou University	120	8–14 years	MP pneumonia with inflammatory symptoms	Symptoms and lab testing	Huaiqihuang Granules + Doxycycline	Higher clinical efficacy and reduced inflammation
Feng ChangJie [47]	2023	Zhangjiakou Maternal and Child Health Hospital	116	2–13 years	Chronic MP pneumonia with lung complications	Symptoms and clinical history	Qingxuan Zhike Granules + Doxycycline	Improved pulmonary function and reduced inflammation

Meta-analysis of treatment efficacy and symptom resolution

This meta-analysis evaluated the efficacy and safety of doxycycline in treating SMPP, compared with macrolide therapy (e.g., azithromycin or clarithromycin), based on pooled data from multiple clinical studies (Fig. 6). For treatment efficacy, the pooled analysis showed that doxycycline was superior to macrolides, with low heterogeneity across studies (I² = 25%) and a combined risk ratio (RR) of 0.68 (95% CI: 0.58–0.79) (Fig. 6A). Regarding fever resolution time, the combined mean difference (MD) was − 1.5 days (95% CI: − 2.3 to − 0.7), favoring the doxycycline group, with low heterogeneity (I² = 22%) (Fig. 6B). Cough resolution time was significantly shorter in the doxycycline group, with a pooled MD of − 2.1 days (95% CI: − 3.5 to − 0.7) and I² = 20% (Fig. 6C). The meta-analysis indicated a shorter resolution time for lung rales in the doxycycline group, with a combined MD of − 1.8 days (95% CI: − 2.6 to − 1.0) and I² = 18% (Fig. 6D). The results across all indicators showed low heterogeneity and consistent findings across studies.

Fig. 6.

Forest plot of the meta-analysis. Note A Forest plot of the meta-analysis on the treatment efficacy of doxycycline for children with SMPP; B forest plot of the meta-analysis on fever resolution time in children treated with doxycycline for SMPP; C forest plot of the meta-analysis on cough resolution time in children treated with doxycycline for SMPP; D forest plot of the meta-analysis on lung rales resolution time in children treated with doxycycline for SMPP

Sensitivity analysis of doxycycline treatment outcomes

Sensitivity analysis was conducted by sequentially excluding individual studies to assess their influence on the pooled effect size for four primary indicators: treatment efficacy, fever resolution time, cough resolution time, and lung rale resolution time (Fig. 7). For treatment efficacy, the combined odds ratio (OR) was 4.14 (95% CI: 1.72–9.95) with heterogeneity (I²) of 76.0%. The exclusion of any single study did not substantially alter the overall effect size or confidence interval, indicating the stability of the results (Fig. 7A). For fever resolution time, the combined MD was − 1.64 days (95% CI: − 2.21 to − 1.07), with heterogeneity of 98.9%. The results remained consistent across all exclusions (Fig. 7B). For cough resolution time, the combined MD was − 2.16 days (95% CI: − 3.15 to − 1.18), with I² = 99.5%. No significant fluctuations in effect size were observed when studies were excluded (Fig. 7C). For lung rale resolution time, the combined MD was − 1.98 days (95% CI: − 2.87 to − 1.08), with I² = 98.1%. Sensitivity analysis showed that the overall estimate remained consistent regardless of individual study exclusion (Fig. 7D). The sensitivity analyses across all four indicators demonstrate that the treatment outcomes associated with doxycycline are stable and robust.

Fig. 7.

Sensitivity analysis forest plots. Note A Sensitivity analysis of treatment efficacy for doxycycline in children with SMPP; B sensitivity analysis of fever resolution time for doxycycline in children with SMPP; C sensitivity analysis of cough relief time for doxycycline in children with SMPP; D sensitivity analysis of lung rales resolution time for doxycycline in children with SMPP

Publication bias assessment of meta-analysis results

Publication bias was assessed using funnel plots for the key clinical outcomes of doxycycline treatment for SMPP (Fig. 8). The outcomes evaluated included treatment efficacy, fever resolution time, cough resolution time, and lung range resolution time.

Fig. 8.

Publication of bias test results for meta-analysis. Note A Publication bias test results for treatment efficacy of doxycycline in children with SMPP; B publication bias test results for fever resolution time in children with SMPP treated with doxycycline; C publication bias test results for cough relief time in children with SMPP treated with doxycycline; D publication bias test results for lung rales resolution time in children with SMPP treated with doxycycline

For treatment efficacy, the funnel plot showed an approximately symmetrical distribution of data points (Fig. 8A), with no significant evidence of publication bias. While some studies deviated from the funnel center, the overall distribution was consistent, and the combined effect size and confidence interval remained stable.

For fever resolution time, the funnel plot demonstrated good symmetry between effect size and standard error (Fig. 8B). Despite some heterogeneity among studies, most studies were concentrated near the funnel center, indicating no significant publication bias.

For cough resolution time, the data points also exhibited a generally symmetrical distribution (Fig. 8C). Minor deviations were observed, but the overall plot remained balanced, with no substantial indication of bias affecting the pooled results.

For lung rale resolution time, the funnel plot again showed a symmetric distribution (Fig. 8D). Most studies were near the center of the plot, suggesting that publication bias was not present and that the results were robust.

Discussion

MPP is a common respiratory infection caused by Mycoplasma pneumoniae, particularly prevalent in children [23, 24]. SMPP represents a more severe manifestation of the disease, often characterized by prolonged fever, cough, rapid breathing, and lung rales. It can rapidly progress to respiratory failure and even pose a life-threatening risk [25, 26]. Due to the specific etiology of the disease, conventional antibiotic treatments are often limited in their ability to alleviate symptoms, and some antibiotics face resistance issues [27, 28]. Consequently, identifying more effective and safer treatment options has become a critical clinical research focus. In recent years, doxycycline, a broad-spectrum antibiotic, has demonstrated promising efficacy in treating adult respiratory infections; however, its effectiveness in treating pediatric SMPP remains to be further validated [29, 30]. This study combined clinical observations and meta-analysis to systematically evaluate the efficacy and safety of doxycycline in treating SMPP in children under 8 years of age. The results demonstrated that doxycycline significantly improves treatment outcomes and shortens fever duration, cough resolution time, and lung rale disappearance time, with a favorable safety profile. These findings provide strong evidence supporting its clinical use.

Although this study demonstrates the advantages of doxycycline in treating SMPP, several limitations should be acknowledged. First, the incidence of pleural effusion before treatment was significantly higher in the doxycycline group, which may reflect more severe baseline disease. As this was a retrospective study, patient allocation was based on clinical treatment decisions rather than randomized assignment, possibly leading to baseline differences between groups. Additional statistical analyses were conducted to adjust for potential confounding variables, and both groups were evaluated using standardized imaging methods (USG and CT) to ensure consistency. Despite the baseline difference, the post-treatment incidence of pleural effusion significantly decreased in both groups and showed no statistical difference, suggesting comparable therapeutic effects.

Second, chest CT was used as the primary imaging tool to assess treatment outcomes due to its higher sensitivity in identifying lesion extent and severity than traditional X-rays. However, routine CT scanning may not be feasible in clinical settings with limited resources. Moreover, this study lacked baseline CT images, limiting direct pre- and post-treatment comparisons. Future research should aim to perform CT scans both before and after treatment for a more objective assessment, while also exploring more accessible imaging methods such as chest ultrasound or portable X-ray in resource-limited settings.

Third, treatment efficacy was evaluated one week after treatment initiation, based on standard protocols for pneumonia management. While this allows for comprehensive symptom and imaging assessment, using a single time point may not capture the full dynamics of treatment response. Future studies should include multiple follow-up time points to enhance the robustness of outcome evaluation.

Additionally, most patients experienced rapid defervescence within one day of treatment initiation, which may be closely associated with the concurrent use of corticosteroids such as methylprednisolone. In patients with pleural effusion, corticosteroids may have contributed significantly to symptom relief. Methylprednisolone, as an immunomodulator, may influence fever resolution and symptom improvement, potentially confounding the assessment of antibiotic efficacy. Although no significant differences in corticosteroid use were observed between the two groups in this study, the potential impact should be carefully considered, and future research should further control for this variable.

It is worth noting that fever resolution time and length of hospital stay are objective clinical indicators with important implications in evaluating SMPP treatment. This study found no significant differences between the doxycycline and control groups in these two measures, suggesting comparable effectiveness in reducing fever and hospitalization duration. These metrics are objective and reliable and directly reflect patient recovery and healthcare resource utilization. While cough resolution and lung range disappearance times provide more detailed symptom-related information, fever duration and hospital stay offer more robust clinical endpoints. Future studies should continue emphasizing these objective measures when evaluating antibiotic efficacy for pediatric SMPP.

Finally, this study did not assess macrolide resistance rates in Mycoplasma pneumoniae, which is a notable limitation. Recent epidemiological studies in China report high resistance rates to macrolides, with some regions showing rates as high as 70% to 90% [31, 32]. The widespread resistance to macrolides may have a significant impact on treatment outcomes. Although this study could not provide specific resistance data, this factor should be addressed in future research to better understand its influence on clinical efficacy.

In summary, the results of this study demonstrate that Doxycycline significantly improves treatment efficacy, accelerates symptom relief, and maintains good safety in the treatment of SMPP in children under 8 years of age. Its notable effectiveness in treating cough, lung rales, and other clinical symptoms provides a safe and effective option for pediatric SMPP treatment. Future research should further investigate the effects of Doxycycline in different stages of the disease and among specific patient groups, optimizing treatment protocols to provide more comprehensive scientific evidence for clinical practice.

Abbreviations

BAL

Bronchoalveolar lavage

CI

Confidence interval

CT

Computed tomography

CRP

C-reactive protein

D-D

D-dimer

LDH

Lactate dehydrogenase

MD

Mean difference

MUMPP

Macrolide-unresponsive Mycoplasma pneumoniae pneumonia

OR

Odds ratio

RCT

Randomized controlled trial

RR

Risk ratio

SMP

Severe Mycoplasma pneumoniae pneumonia

SMD

Standardized mean difference

WBC

White blood cell

Author contributions

Qian Zhao, Xu Sang, and Ding Gao contributed equally to this work and share first authorship. Qian Zhao and Xu Sang were responsible for data collection, clinical assessments, and drafting the manuscript. Ding Gao conducted the meta-analysis and contributed to data interpretation. Zhen Zhang and Aili Xuan assisted in patient recruitment and data management. Wansheng Peng conceived and supervised the study, performed critical revisions of the manuscript, and approved the final version for submission. All authors read and approved the final manuscript.

Funding

This study was supported by Bengbu Medical University Natural Science Program (2023byzd076).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethical approval

This study was approved by the Clinical Ethics Committee of The First Affiliated Hospital of Bengbu Medical University (No. [2024] 212).