Cross-sectional hospital-based investigation on clinical characteristics of pediatric Group A Streptococcus isolates in a Beijing hospital from 2014 to 2023

Abstract

Background

Group A Streptococcus (GAS) is a major bacterial pathogen associated with diverse clinical diseases. Despite its general susceptibility to β-lactam antibiotics, GAS outbreaks remain a significant global health and economic concern. In this study, we retrospectively analyzed the clinical features and antimicrobial resistance patterns of GAS infections in pediatric patients in Beijing over the past decade.

Methods

Data on all GAS infections from 2014 to 2023 were collected from the microbiology department of Beijing Children’s Hospital. Antimicrobial susceptibility testing was performed using the Kirby–Bauer disc diffusion method and interpreted according to the Clinical and Laboratory Standards Institute (CLSI) guidelines.

Results

From 2014 to 2023, a total of 1,445 strains of GAS were isolated. School-age children (6–18 years) had the highest isolation rate (77.7%) among all age groups. Among the isolates, 53.6% were from outpatients (except pediatric gynecology). The top 3 wards with the highest GAS isolation rates were the rheumatology and immunology, pulmonology and dermatology wards. Most GAS infections were superficial (95.8%). Invasive infections were primarily due to bacteremia, which accounted for 40.0% of invasive cases. The isolation rate of GAS peaked in 2017 and remained at a lower level during the SARS-CoV-2 pandemic, specifically from 2019 to 2022. GAS infections also showed seasonal variation, with the highest prevalence in winter. No GAS strains were resistant to penicillin, cephalosporins, vancomycin, or linezolid. The resistance rates to erythromycin and clindamycin were generally above 90%.

Conclusion

In the post-SARS-CoV-2 period, the isolation rate of GAS has increased. GAS infections are more common in winter and summer, so children, especially school-age children, should be cautious. Given the high resistance of GAS to erythromycin and clindamycin, clinicians should limit the use of these two drugs to avoid treatment failure.

Introduction

Group A Streptococcus (GAS), also known as Streptococcus pyogenes (S. pyogenes), is an important human pathogen that can cause a wide range of diseases [1]. It can cause conditions ranging from asymptomatic carriage and superficial infections (such as acute pharyngitis, acute tonsillitis, impetigo, erysipelas, and vaginitis) to deep-seated infections (such as cellulitis, bacteremia, septicemia, pericarditis and meningitis) [2], as well as toxin-mediated diseases (such as toxic shock syndrome, and fulminant purpura). GAS infections can result in autoimmune complications, including rheumatic fever, which may lead to rheumatic heart disease (RHD) and acute post-streptococcal glomerulonephritis (APSGN) [3, 4].

Against the backdrop of increasingly severe bacterial resistance, Streptococcus pyogenes remains susceptible to penicillin and β-lactam antibiotics. Nevertheless, this pathogen continues to pose a major global health threat, causing over 700 million infections annually (including 1.8 million severe cases and 517,000 deaths) [5, 6]. The World Health Organization’s Global Burden of Disease data rank it as the 12th leading cause of death globally [7].

Several studies have indicated that in the mid-20th century, the burden of GAS-related diseases decreased in developed countries. However, over the past 50 years, there have been numerous reports of GAS-related disease outbreaks [8–10]. Recent epidemiological investigations have found outbreaks of scarlet fever in Asia and the United Kingdom, with the outbreak in the United Kingdom coinciding with a surge in invasive infections [11, 12]. At the end of 2022, several countries—including the United Kingdom, the United States, Australia, and several European countries—reported unexpectedly high incidence rates of invasive Group A streptococcal (iGAS) infections and scarlet fever, particularly among children under 10 years old. During this period, the iGAS incidence rate in the United Kingdom was higher than that in the pre-SARS-CoV-2 pandemic years (2017 to 2019) and significantly higher than the rates reported from 2020 to 2021 [1].

The treatment regimen for GAS infections primarily involves the use of appropriate antibiotics. GAS remains highly sensitive to penicillin, while other antibiotics—including cephalosporins, macrolides, and clindamycin—are also commonly used in clinical practice [13]. The increasing resistance to macrolides and clindamycin remains a growing global concern that requires urgent attention [3]. Although penicillin resistance in GAS has not yet been documented, cases of treatment failures to penicillin have emerged [14].

This study, focusing on pediatric patients in Beijing, analyzed the antimicrobial resistance patterns of GAS across all specimen types over a decade, encompassing both the pre-pandemic and post-pandemic periods of SARS-CoV-2. Due to its long-time span and large sample size, particularly in the Beijing region, this research is of great significance for predicting antimicrobial resistance trends, guiding clinical treatment, and formulating strategies to prevent resistance in Beijing.

Materials and methods

Study setting and design

The retrospective analysis in this study centered on all GAS strains from specimens of children admitted to Beijing Children’s Hospital between January 2014 and December 2023. For the same patient, only the first strain of bacteria isolated from the same site was included in the study. Priority should be given to strains isolated from sterile sites (such as blood, cerebrospinal fluid, joint fluid, pleural and peritoneal fluids, etc.). If all specimens were from sterile sites, blood specimens held the highest priority. If all specimens were from non-sterile sites (such as sputum, urine, wound secretions), selection should be made in the order of wound secretions, bronchial lavage fluid, and sputum. Moreover, only one strain of GAS was to be included within a single cycle (30 days). Study-relevant data, including patient age, gender, sample type, and antimicrobial susceptibility test results, were obtained from the hospital’s electronic information system.

Species culture and identification

Clinical specimens were initially inoculated onto Columbia blood agar plates, MacConkey agar plates, and Chocolate agar plates, in accordance with the National Guide to Clinical Laboratory Procedures. Subsequently, these plates were incubated at 37℃ with 5% CO₂ for 24–48 h. GAS identification was performed using the Vitek MS system (bioMérieux France) with daily quality control using Escherichia coli ATCC 8739. The quality control pass rate for strain identification was 100%.

Antimicrobial susceptibility testing

The antimicrobial susceptibility testing of GAS was performed using the Kirby–Bauer agar diffusion method with antimicrobial disks from Oxoid Company (UK). The tested antimicrobial agents included penicillin, ceftriaxone, cefepime, levofloxacin, erythromycin, clindamycin, vancomycin, and linezolid. Mueller-Hinton Agar containing 5% sheep blood was used as the medium (Autobio, China). These plates were incubated at 37℃ with 5% CO₂ for 24 h. Weekly quality control for antimicrobial susceptibility testing was conducted with Streptococcus pneumonia ATCC 49619, following the same protocol as clinical testing. Both quality control and clinical results were interpreted according to the American Clinical & Laboratory Standards Institute (CLSI) M100-S34 [15].

Statistical analysis

WHONET 5.6 software was used for data processing. GraphPad Prism 6 (La Jolla, CA, USA) was employed for statistical evaluations. Statistical significance was determined using the chi-square test and t-test, with a p < 0.05 defined as statistically significant.

Results

Characteristics of pediatric patients

Between 2014 and 2023, a total of 1,445 GAS strains were isolated. The majority of GAS isolates (77.7%) were from school-age children aged 6–18 years. Female patients accounted for 82.5% of the cases. After excluding vaginal discharge specimens (a specimen type exclusive to females), the proportion of female patients was 46.5% (220/473). Among all patients, 53.6% were outpatients (except pediatric gynecology). The GAS proportion of the pediatric gynecology department was 33.8% (489/1,445). The top 3 wards with the highest GAS isolation rates were the rheumatology and immunology, pulmonology and dermatology wards. In sterile body fluid specimens, bacteremia was the most common clinical presentation of GAS infection, comprising 40.0% (24/60) of cases (Table 1).

Table 1.

Demographics of patients, specimen types, and infection classification

Gender		No. of patients	%
	Male	253	17.5
	Female	1192	82.5
	Male (excluding vaginal discharge)	253	53.5
	Female (excluding vaginal discharge)	220	46.5
Age
	Neonate (Newborn up to first 28 days)	3	0.2
	Infant (28 days-below 1 year)	34	2.4
	Toddlers (1 year-below 3 years)	48	3.3
	Preschool (3 years-below 6 years)	237	16.4
	School (6 years-18 years)	1123	77.7
Outpatient/Ward
	Outpatient (except pediatric gynecology)	774	53.6
	Pediatric Gynecology	489	33.8
	Rheumatology and Immunology Ward	35	2.4
	Pulmonology Ward	31	2.1
	Dermatology Ward	28	1.9
	Intensive Care Unit	26	1.8
	ENT Ward	25	1.7
	Infectious Diseases Ward	14	1.0
	Cardiology Ward	12	0.8
	Burns Unit	5	0.3
	Neurology Ward	3	0.2
	Orthopedics	3	0.2
Diagnosis of sterile body fluid
	Bacteremia	24	1.7
	Cellulitis	21	1.5
	Pleurisy	9	0.6
	Sepsis	4	0.3
	Endocarditis	1	0.1
	Arthritis	1	0.1
Specimen type
	Vaginal discharge	972	67.3
	Lower respiratory tract	61	4.2
	Broncho-alveolar lavage	24	1.7
	Sputum	37	2.6
	Upper respiratory tract	286	19.8
	Respiratory secretions	4	0.3
	Nasopharyngeal secretions	282	19.5
	Sterile body fluid	60	4.2
	Ear	44	3.0
	Skin secretions	19	1.3
	Urine	3	0.2
Infection classification
	Superficial	1,385	95.8
	Invasive	60	4.2

ENT Ear, Nose, and Throat

Epidemic characteristics of GAS

The annual, seasonal, and monthly variation patterns of GAS isolation rates

From 2014 to 2017, the GAS isolation rate generally increased year by year, peaking in 2017. Since 2017, the GAS isolation rate declined annually until 2020. Between 2020 and 2022, GAS detection remained stable, with approximately 40 strains isolated each year. However, in 2023, there was a sharp increase, with 155 strains isolated that year. The proportion of GAS among all isolated strains followed the same trend as GAS detection rate (Fig. 1). Over the past decade, the GAS positivity rate fluctuated between 1.1% and 4.6%, as detailed in Table 2.

Fig. 1.

Annual trends in GAS detection counts and proportion of total bacterial isolates (2014–2023). The GAS number represented the number of strains isolated by GAS in the corresponding year, while the GAS percentage represented the percentage of GAS strains isolated in the corresponding year compared to the total number of strains isolated in that year

Table 2.

Yearly and seasonal distribution of GAS isolation cases

Parameters	Total No. of specimens	Group A Streptococcus
		No. of isolates (n)	Percentage (%)	Positive Rate (%)
Total	51,460	1,445	100.0	2.8
Collecting year
2014	5,722	144	10.0	2.5
2015	5,591	186	12.9	3.3
2016	6,314	185	12.8	2.9
2017	6,527	301	20.8	4.6
2018	5,949	225	15.6	3.8
2019	5,386	122	8.4	2.3
2020	3,265	42	2.9	1.3
2021	3,831	49	3.4	1.3
2022	3,262	36	2.5	1.1
2023	5,613	155	10.7	2.8
Collecting season
Spring (Mar-May)		283	19.6
Summer (Jun-Aug)		386	26.7
Autumn (Sep-Nov)		308	21.3
Winter (Dec-Feb)		468	32.4

The monthly distribution of GAS isolates from 2014 to 2023 was shown in Fig. 2. Higher isolation rates were observed in winter (Dec-Feb, 32.4%) and summer (Jun-Aug, 26.7%). Data were presented in Fig. 2; Table 2.

Fig. 2.

Monthly distribution of GAS detection counts (2014–2023)

Distribution of specimen types of GAS

The specimen type with the highest GAS isolation rate was vaginal discharge (67.3%, 972/1,445), followed by upper respiratory tract specimens (19.8%, 286/1,445). Most GAS infections were superficial (95.8%, 1,385/1,445), with the remainder being invasive (4.2%, 60/1,445). Detailed data were shown in Table 1.

Antibiotic resistance phenotypes

All GAS strains exhibited full susceptibility to penicillin, ceftriaxone, cefepime, linezolid, and vancomycin. Fluoroquinolone sensitivity exceeded 96%. In contrast, clindamycin and erythromycin resistance rates were notably high (≥ 90%). Detailed resistance data were presented in Table 3.

Table 3.

Susceptibility of GAS strains to antimicrobial agents, 2014–2023(%)

Antimicrobial agent	2014 (n = 144)		2015 (n = 186)		2016 (n = 185)		2017 (n = 301)		2018 (n = 225)		2019 (n = 122)		2020 (n = 42)		2021 (n = 49)		2022 (n = 36)		2023 (n = 155)
	R	S	R	S	R	S	R	S	R	S	R	S	R	S	R	S	R	S	R	S
Penicillin G	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100
Ceftriaxone	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100
Cefepime	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100
Levofloxacin	0	98.6	1.1	98.4	1.6	96.8	1	96.6	0	100	0.8	98.4	2.4	97.6	2	98	0	100	1.3	98.7
Clindamycin	96.5	3.5	96.2	3.8	95.1	4.9	94.0	6.0	91.1	8.9	84.4	15.6	83.3	16.7	91.8	8.2	97.2	2.8	95.5	4.5
Erythromycin	95.8	2.8	95.1	4.3	95.7	4.3	97.0	2.7	92.0	6.7	90.2	9.8	85.7	11.9	93.9	6.1	97.2	2.8	96.8	3.2
Linezolid	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100
Vancomycin	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100	0	100

We analyzed data on the drug resistance rate of GAS isolated from different specimen types. No significant differences in erythromycin or clindamycin resistance were observed across specimen types. However, levofloxacin resistance showed differences between the respiratory tract group and the vaginal secretion group, and also between the respiratory tract group and the sterile body fluid group. Details were shown in Table 4.

Table 4.

Susceptibility of GAS strains to antimicrobial agents in different specimen types (%)

Antimicrobial agent	Vaginal discharge (n = 972)		Respiratory tract (n = 347)		Sterile body fluid (n = 126)			Total (n = 1445)
	R	S	R	S	R		S	R		S
Penicillin G	0	100	0	100	0		100	0		100
Ceftriaxone	0	100	0	100	0		100	0		100
Cefepime	0	100	0	100	0		100	0		100
Levofloxacin	1.4	98.3	0	100	3		97	1.2		98.5
Clindamycin	93	7	93.3	6.7	95.2		4.8	93.3		6.7
Erythromycin	94.1	5.2	95.6	4.1	96		3.2	94.7		4.8
Linezolid	0	100	0	100	0		100	0		100
Vancomycin	0	100	0	100	0		100	0		100

Discussion

In our ten-year retrospective study, after excluding vaginal secretion specimens (a type exclusive to females), the ratio of GAS infections between males and females was 1.15:1. Of the 1,445 GAS strains isolated, 972 were from vaginal secretions of female pediatric patients. A 2021–2023 research report from Beijing Children’s Hospital showed that among pediatric patients with vaginosis, GAS ranked fourth among all detected bacteria, following Haemophilus influenzae, Staphylococcus aureus, and Streptococcus agalactiae. In this report on bacterial vaginitis, it was found that the isolation rate of GAS in cases of bacterial vaginitis was 7.0%. Since all our data were obtained from samples collected from children diagnosed with bacterial vaginitis, this portion of the data should be considered as representing the pathogen. At the same time, the author also pointed out that the positive rate of bacterial cultures was highest in summer. This was because children were more likely to come into contact with various pathogens when they wear thin clothes and engage in recreational water activities during summer. This situation was also consistent with the phenomenon that GAS has a peak incidence in summer [16]. The pediatric gynecology department prioritized pediatric patients with symptomatic vaginitis, all of whom underwent standardized vaginal secretion cultures. We strictly followed standardized protocols during specimen collection to ensure accurate and reliable vaginal swab sampling from children, thereby maintaining high specimen quality. These measures collectively accounted for the predominance of vaginal secretion-derived strains in this study. Notably, our research identified GAS as a prevalent pathogen causing vulvovaginal infections in prepubertal girls, highlighting the need for enhanced clinical awareness and improved prevention strategies among healthcare professionals.

Regarding specimen types, respiratory specimens accounted for only 24.0%, far less than vaginal secretions. The low proportion of respiratory strains was likely primarily due to the low submission rate of throat swab specimens from patients with pharyngotonsillitis by physicians in our hospital’s internal medicine outpatient clinic, coupled with the frequent use of antibiotics in the outpatient setting, which may have affected the bacterial detection rate. Clegg et al. found that 95% of female children with vulvovaginitis had positive GAS cultures in their throats, suggesting that the nasopharyngeal flora might serve as the source of pathogens causing vulvovaginal infections in female children [17]. Another European study identified bacterial upper respiratory tract infections as the most prevalent risk factor for vaginitis in girls [18]. Further research was required to determine whether the GAS strains isolated from pediatric vaginitis patients in our hospital are homologous with those detected in the upper respiratory tract.

Despite the relatively low incidence of iGAS infections, their clinical impact can be substantial. Studies indicated that 8%–23% of patients with iGAS infections died within seven days [3]. Previous studies in New Zealand and Australia showed that, among these settings, cellulitis posed the highest burden on patient health and the economic cost of the healthcare system among all GAS diseases [19]. In recent years, small-scale outbreaks of iGAS infections occurred in some countries. Reports indicated that the proportion of iGAS cases in the UK exceeded pre-pandemic levels (2017–2019). With ongoing iGAS outbreaks, heightened media coverage, and increased community awareness, clinicians should recognize the broad spectrum of invasive diseases caused by GAS [20–23]. Meanwhile, many national health authorities have recognized the limited availability of longitudinal public health data on iGAS. The significance of GAS infection surveillance in China remains insufficiently emphasized. To effectively monitor GAS infections, particularly iGAS, relevant authorities should enhance surveillance efforts [24].

Notably, the annual number of GAS isolates from 2019 to 2022 was significantly lower than that during 2013–2018, coinciding with the SARS-CoV-2 pandemic in China. This reduction might be attributed to non-pharmaceutical interventions such as school closures, mandatory mask mandates, social distancing, among others [25, 26]. These physical interventions might disrupt or reduce pathogenic microorganism transmission [27, 28], potentially contributing to the decreased GAS prevalence among children observed in our study during the pandemic. A parallel trend was noted in GAS and S. aureus isolation rates among children reported by other studies, further supporting this inference [29, 30].

We also observed a seasonal pattern in GAS prevalence, with higher incidence during winter. A similar seasonal variation has been reported in Europe, which shares a temperate climate [31]. This pattern might result from complex interactions among climatic conditions, behavioral habits, and the prevalence of other respiratory infectious diseases.

GAS can enter host cells via M protein and lipoteichoic acid (LTA), enabling colonization and proliferation within cells, evasion of the immune system, and subsequent transmission to new hosts. The application of antimicrobial agents was the preferred treatment approach for GAS infections. Among the tested antibiotics, GAS exhibited the highest resistance to macrolides, peaking at 97.2%. These rates were comparable to findings from studies in China [32]. However, erythromycin and clindamycin resistance rates were significantly lower in the United States. A study reported that from 2015 to 2019, GAS non-susceptibility to erythromycin increased from 18% to 25%, and to clindamycin from 17% to 24% within four years [33]. GAS resistance to erythromycin and clindamycin also varied regionally. A Northern European multicenter study reported 1%-2% resistance rates to these antibiotics among GAS isolates from necrotizing fasciitis cases [34]. In addition, the data from Spain was consistent with the above results [35]. Taiwan, geographically close to mainland China, exhibited significant variations in erythromycin and clindamycin resistance rates among GAS, though the overall proportion remained above 50% [36]. Data disparities primarily stemmed from geographical variations in mfeA expression and changes in erm expression, the latter being closely linked to clindamycin resistance [32, 33, 37].

Our study data demonstrated that GAS maintained high sensitivity to fluoroquinolones with stable trends over the 10-year period, consistent with prior findings in China [38, 39]. In contrast, in Japan, also a country in East Asia, showed slightly lower sensitivity of GAS to fluoroquinolone compared to China, with a trend of increasing resistance. This was primarily attributed to the dissemination of emm6 and emm11 GAS strains [40, 41]. Notably, GAS strains from respiratory specimens showed significantly higher fluoroquinolone sensitivity than those from genital tracts or sterile body fluids. This unique phenomenon, unreported in other studies, might relate to fluoroquinolone pharmacology or GAS interactions with varying microenvironments, warranting further investigation.

β-lactam antibiotics are the first-line treatment for GAS infections. To date, no penicillin- or cephalosporin-resistant GAS strains have been identified. However, some studies have reported increasing penicillin MICs and treatment failures in GAS strains despite their in vitro susceptibility to penicillin. As early as 1953–1993, GAS pharyngitis cases remained culture-positive after 10 days of penicillin therapy [42]. Such cases have also emerged in recent years. Penicillin has been ineffective in preventing acute rheumatic fever triggered by GAS [43]. Several factors may explain this phenomenon. Firstly, GAS genome mutations could be involved. During a community outbreak in Seattle, certain GAS strains showed an eightfold reduction in penicillin sensitivity, with identified missense T553K substitutions in PBP2x [44]. The second potential factor involves β-lactamase-producing bacteria like H. influenzae. Coexisting strains may release outer membrane vesicles (OMVs) containing β-lactamase (βL-OMVs), which shield GAS from β-lactam antibiotics [45]. Additionally, bacterial growth arrest and efflux pumps also contribute to reduced antibiotic susceptibility in GAS. Martini et al. observed that in clustered states—such as high-density bacterial populations or biofilms on biological/non-biological surfaces—GAS subpopulations can persist despite exposure to high concentrations of β-lactam antibiotics and other antimicrobial agents [46]. The failure of penicillin in treating GAS infections may lead to multiple impacts, including persistent or worsening infections, an increased risk of complications, prolonged disease course, the emergence of drug resistance, and the potential for allergic reactions. The penicillin MIC serves as a key indicator of potential treatment failure, suggesting a shift from the K-B to MIC method in clinical practice. Clinicians should closely monitor GAS cases with poor responses, considering gene sequencing for further analysis while comprehensively evaluating other contributing factors.

Conclusion

In summary, our study offers comprehensive insights into the distribution and antibiotic resistance of GAS isolates in Beijing (2014–2023). School-age children (6–18 years) showed the highest infection incidence, with predominant specimens from vaginal discharge, upper, and lower respiratory tracts. Winter and summer were the most susceptible seasons. GAS maintained high susceptibility to penicillin, ceftriaxone, and cefepime but exhibited significant resistance to erythromycin and clindamycin. Rigorous resistance monitoring and judicious antimicrobial use are therefore essential.

Author contributions

1. [Luwei Wang]: Data Collection: Participated in the collection of raw data from relevant sources.Data Analysis: Led the statistical analysis and interpretation of the data.Writing - Original Draft: Drafted the initial version of the manuscript.Visualization: Created data visualizations (e.g., graphs, tables) to present the findings. 2. [Zhiyong Lyu]: Conceptualization: Assisted in refining the research idea and objectives.Methodology: Developed the methodology for data collection and analysis.Data Collection: Oversaw the collection of data and ensured its accuracy.Writing - Review & Editing: Contributed to the refinement of the manuscript, particularly in the sections related to data analysis and results.

Funding

There is no funding report.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The GAS isolates analyzed in this study were collected from clinical laboratories. Our study was a retrospective study based on the clinical specimens, which were part of the routine hospital laboratory procedures. This study was reviewed and approved by the Ethics Committee of Beijing Children’s Hospital. The ethics committee waived the need for participants’ informed consent because the study was retrospective. The study was conducted following the principle of the Declaration of Helsinki.

Competing interests

The authors declare no competing interests.