Analysis of the Sputum Culture Spectrum of Children with Respiratory Tract Infection in Wuxi Between 2014 and 2021

Introduction

The immune resistance of children is lower, making respiratory tract infection one of the more common paediatric diseases.¹ This type of infection seriously affects children’s health and has become a global public health concern.² Identifying the type of the pathogenic bacteria is vital to its treatment. Children’s susceptibility to respiratory tract infection is mainly caused by the anatomical, physiological and immune characteristics of their respiratory systems, resulting in the reduced specific and non-specific immune function of the respiratory tract, especially in infants. As children age, the growth environment of pathogenic bacteria changes. The pathogenic bacteria of the respiratory tract may vary with age, sex and season, with these factors varying almost every year.³ An analysis of the sputum culture profile in children with respiratory tract infection can accurately display the latest types and trends of pathogenic bacteria in these patients in real time, and the timely and accurate detection of pathogenic bacteria and selection of appropriate antimicrobial agents have key clinical implications for the treatment of childhood bacterial pneumonia.^4,5 The aim of this study was to analyse the characteristics, distribution and variation of pathogens isolated from the sputum culture of children with respiratory tract infection in our hospital between 2014 and 2021. We retrospectively analysed these clinical data to determine related influencing factors to better prevent and treat children’s respiratory tract infection.

Materials and Methods

General Information

The clinical data of children (aged between 28 days and 15 years) with respiratory tract infection admitted to Affiliated Hospital of Jiangnan University between January 2014 and December 2021 were retrospectively collected. All specimens collected were from patients with lower respiratory tract infections. The inclusion and exclusion criteria were based on the diagnostic criteria for nosocomial infection issued by the Ministry of Health of China ([2001]2). To prevent hospital data breaches, all case data have been transcribed manually with patient consent and have not been copied directly from the computer. We used negative suction pressure to help suction deep throat sputum, and sputum specimens were sent to the hospital microbiology laboratory for testing. Affiliated Hospital of Jiangnan University approved the study after ethical review (IRB number: LS2021088). Informed consent was waived by our institutional review board due to the retrospective nature of our study, and data were anonymized and kept confidential, and the study met the ethical standards of the World Medical Association’s Declaration of Helsinki.

Obtaining Clinical Test Parameters

Sputum samples were collected within 24 h of admission, and pathogenic bacteria were isolated. Children under 6 years received a water rinse to clean their mouths. The children were placed in a semi-recumbent position. After local anaesthesia, a sterile closed sputum suction tube was inserted and connected to a negative pressure suction device, which collects secretions from deep in the airway. The tip of the suction tube was cut to collect the culture. After gargling with saline, children over 6 years were given a 3% saline nebulised inhalation to induce expectoration (appropriate local anaesthesia was given when the child was unable to tolerate the procedure). All the samples were sent for examination within 1 h (microscopic observation of sputum samples: white blood cells >25 × 10 ^ 9/L, squamous epithelial cells <10 × 10 ^ 9/L) and monitored continuously for 2 to 3 days. Each pathogen was cultured twice.

Inducing sputum is a more non-invasive method but can cause infection if the equipment is not thoroughly cleaned; negative pressure sputum provides better specimens but can damage the trachea, leading to injury and inflammation in patients. The degree of patient cooperation influences which of the 2 sampling methods is chosen. In practice, doctors use the age of 6 years as the dividing line between the 2 methods to unify standards.

Bacterial/Fungi Culture and Identification

The samples were inoculated into a chocolate medium, blood medium, MacConkey medium (Zhengzhou Renfa có Sai Biotechnology, Henan, China) and Sabour’s medium for bacterial isolation and culture. The chocolate medium was cultured at 35°C in a 5% carbon dioxide incubator for 18–24 h, and the blood, MacConkey and Sabour mediums were cultured at 35°C for 18–24 h in the incubator. According to the basic technical standards of an NHC clinical microbiological examination in China, if fungal colonies are found, mycelia or spores must be identified under a microscope before fungal growth is confirmed. The dominant and absolute dominant bacteria were selected for pure culture. Microbial identification testing using a VITEK^®2 Compact automatic microbiology analyser (Meyrié Diagnostic Products, Shanghai, China) continued for approximately 18–24 h.

Statistical Methods

The statistical software SPSS 26.0 was used in this study, and statistical data were expressed as a rate. The difference between ages, sexes, seasons and years was tested using the chi-square test. Fisher’s exact test was used for analysis. When p < 0.05, the difference was considered statistically significant.

Results

Distribution of Pathogens

A total of 4610 sputum specimens were obtained, and 508 strains of pathogenic bacteria were detected by eliminating the duplicate strains through the practice of retaining the first strain of the same bacteria from the same patient. Of the 4610 sputum specimens, 508 (11.02%) were positive. Among the positive samples, 40.2% (204/508) were gram-positive and 59.8% (304/508) were gram-negative. The gram-negative bacteria were composed of Haemophilus (18.3%, n = 55), Enterobacteriaceae (58.6%, n = 178) and non-fermenting gram-negative bacteria (23.3%, n = 71). The positive rate of neonatal specimens was 27.75% (136/490), which is significantly higher than that of paediatric internal medicine [9.03% (372/4120) (X² = 156.61, p < 0.01)]. Of the 508 positive specimens, there were 112 strains of Staphylococcus aureus, 59 of Klebsiella pneumoniae, 59 of Escherichia coli, 55 of Haemophilus influenzae (Among them, 13 BLPAR and 10 BLNAR strains were detected), 41 of Acinetobacter baumannii, 26 of Staphylococcus haemolyticus, 22 of Streptococcus pneumoniae, 22 of Enterobacter cloacae subspecies and 17 of Pseudomonas aeruginosa.

Distribution of Bacteria by Sex

The distribution of pathogens between sexes is detailed in Tables 1 and 2; 36.7% (109/297) were gram-positive bacteria and 57.58% (171/297) were gram-negative bacteria. The most common pathogens were S. aureus (24.92%), E. coli (11.11%) and K. pneumoniae (10.44%). Among girls, gram-positive bacteria accounted for 29.86% (63/211) and gram-negative bacteria 63.03% (133/211), and the most common pathogens were S. aureus (18.01%), H. influenzae (17.54%) and K. pneumoniae (13.27%). The constituent ratio of gram-negative bacteria in children was significantly higher than that of gram-positive cocci (p < 0.01). The constituent ratio of H. influenzae bacteria in girls was significantly higher than that in boys; the difference was significant (p < 0.01).

Table 1 Distribution and Constituent Ratio (N, %) of the Main Pathogens in Sputum Culture of the Respiratory 2021 by Sex and Age Rom 2014 to 2021

Table 2 Distribution and Composition Ratio of Gram-Positive Bacteria, Gram-Negative Bacteria and Fungi in Sputum Culture of Children with Respiratory Tract Infections by Gender and Age from 2014 to 2021 (n, %)

Strain Changes by Year

The changes in strains in different years are shown in Tables 3 and 4. In 2014, 172 (33.86) strains of gram-positive bacteria were detected. In 2021, the constituent ratio showed a downward trend. A total of 304 (59.84) strains of gram-negative bacteria were detected, increasing annually. In the past 8 years, the proportion of S. aureus has peaked, but this proportion has been declining annually. In 2014–2015, S. haemolyticus occupied second place, and its constituent ratio was significantly higher than that of the other 3 groups (X² = 18.22, 19.87 and 25.21, p < 0.01). The constituent ratio of H. influenzae in 2014–2015 was significantly lower than that in the other 3 groups (X² = 7.38, 12.93 and 11.73, p < 0.01). In 2016–2017, 2018–2019 and 2020–2021, the proportion of gram-negative bacteria was significantly higher than that of gram-positive bacteria (2021: X² = 16.68, 32.63 and 34.30, p < 0.01). The constituent ratio of gram-negative bacteria in 2014–2015 was significantly lower than that in the other 3 groups (X² = 4.241, 4.62 and 7.49, p < 0.05), and the proportion of gram-negative bacteria increased annually.

Table 3 Distribution and Composition Ratio of Main Pathogenic Bacteria in Sputum Culture of Children with Respiratory Tract Infections in Different Seasons and Years from 2014 to 2021 (n, %)

Table 4 Distribution and Composition Ratio of Gram-Positive Bacteria, Gram-Negative Bacteria and Fungi in Sputum Culture of Children with Respiratory Tract Infections in Different Seasons and Years from 2014 to 2021 (n, %)

Distribution of Bacteria by Age

The distribution of bacteria in different age groups is shown in Tables 1 and 2. The most common pathogens in children less than 1 month old were S. haemolyticus (19.12%, 26/136), E. coli (19.12%, 26/136) and K. pneumoniae (16.18%, 22/136). The most common pathogens in children between the ages of 1 month and 3 years were S. aureus (27.21%, 77/283), H. influenzae (13.70%, 35/283) and E. coli (12.33%, 31/283). Among children aged 3–6, the main pathogens were S. aureus (25.81%, 16/62), H. influenzae (19.35%, 12/62) and A. baumannii (9.68%, 6/62).

Among children over 6 years old, the main pathogens were H. influenzae (22.2%, 6/27), A. baumannii (22.2%, 6/27) and S. aureus (14.8%, 4/27). Gram-positive bacteria accounted for 40.32% (25/62) in children aged 3 to 6, whereas gram-negative bacteria accounted for 81.48% (22/27) in children older than 6. Except for the age group of 3–6 years old, the detected Gram negative bacteria were significantly higher than Gram positive bacteria (X² = 21.24, 35.84 and 24.03, p < 0.01). Gram-negative bacteria accounted for 81.48% in children over 6 years. Over time, fewer gram-positive bacteria existed in children under 1 month old. Between 1 month and 3 years, the number of gram-positive bacteria increased over time; gram-negative bacteria also increased over time in this age group (Figure 1). S. haemolyticus was only identified in the neonatal period of less than 1 month; its proportion was the highest and was not found in any age group older than 1 month. The proportions of H. influenzae in the 1-month, 1-month–3-years, 3–6-years and >6-years age groups were 1.47%, 13.7%, 19.35% and 22.22%, respectively. The composition ratio of E. coli in the <1-month and 1-month–3-years age groups was significantly higher than that of the 3–6-years age group, and the difference was statistically significant (χ² = 11.08, 5.274, p < 0.05).

Figure 1 Distribution of bacteria and fungi in children of different ages in different years and seasons (stacked bar chart).

Distribution of Pathogens by Season

The distribution of pathogens by season is shown in Tables 3 and 4. In spring and summer, gram-positive bacteria accounted for 29.5% (59/200) and gram-negative bacteria for 67.5% (135/200). The top 3 pathogens were S. aureus at 16.5% (33/200), E. coli at 13% (26/200) and K. pneumoniae at 13% (26/200). In autumn and winter, gram-positive bacteria accounted for 36.69% (113/308) and gram-negative bacteria for 54.87% (169/308), with the top 3 being S. aureus at 25.65% (79/308), E. coli at 10.71% (33/308) and K. pneumoniae at 10.71% (33/308). The composition ratio of gram-negative bacteria was significantly higher than that of gram-positive bacteria in both spring and summer and autumn and winter, and the difference was statistically significant (p < 0.01, Figure 1); the composition ratio of gram-negative bacteria in spring and summer was significantly higher than that in autumn and winter, and the difference was statistically significant (p < 0.01). The gram-negative bacteria were composed of Haemophilus (7.4%, n = 10), Enterobacteriaceae (74.1%, n = 100) and non-fermenting gram-negative bacteria (18.5%, n = 25) in spring and summer; the gram-negative bacteria were composed of Haemophilus (26.6%, n = 45), Enterobacteriaceae (46.2%, n = 78) and non-fermenting gram-negative bacteria (27.2%, n = 46) in autumn and winter. The composition of S. aureus in autumn and winter was significantly higher than that in spring and summer, and the difference was statistically significant (p < 0.05).

Discussion

In this study, the detection rate of pathogens in the sputum culture of children with respiratory tract infection was consistent with that of previous reports, but the proportion of pathogenic bacteria was different.^6–8 In total, 508 strains of pathogenic bacteria were detected in 4610 sputum samples from children with respiratory tract infection. The positive rate was 11.01%, and the proportion of children under 3 years was 82.48%, which is consistent with the 76.05% reported by Wang.⁹ The top 5 strains were S. aureus, K. pneumoniae, E. coli, H. influenzae and A. baumannii, of which 59.84% were gram-negative; gram-positive bacteria accounted for 33.86%. Gram-negative bacteria therefore dominated. Understanding the distribution of respiratory tract strains is useful in guiding the clinical medication and treatment of respiratory tract infection in children; it can reduce the abuse of antibiotics and drug resistance of strains, shorten the course of treatment and reduce damage to children and the economic burden on families.¹⁰

However, most studies on the distribution of pathogens in children’s respiratory tract infection have lasted for only between 1 and 5 years, with few lasting 8 years or more. The longer the time span, the more accurately research can reflect the changes in pathogenic bacteria in children’s respiratory tract infection in a specific area. The results revealed no significant differences between gram-positive bacteria and gram-negative bacteria in 2014–2015, but the proportion of gram-negative bacteria has increased significantly with time, especially in the last 2 years. The data suggest that attention should be paid to gram-negative bacteria infection in children with respiratory tract infection.^11,12 In the past 2 years, gram-negative bacteria in children’s respiratory tract infection have mainly consisted of H. influenzae and E. coli.⁹ Studies have shown that the aetiology of severe pneumonia in children is also dominated by H. influenzae and other gram-negative strains.¹ This study showed that the percentage of respiratory tract infection involving S. aureus in children is currently the highest since 2014, which is consistent with the findings of related studies.^3,13 However, this study also revealed a decreasing trend in the proportion of respiratory tract infection in children, although S. aureus remains a key pathogen of necrotising pneumonia in children.¹⁴ These results indicate that we should consider gram-negative bacilli such as H. influenzae when encountering severe pneumonia in children and gram-positive bacteria such as S. aureus when observing necrotising pneumonia in children. It is particularly important to understand the yearly variations in children’s respiratory tract infection to inform the most appropriate choice of drugs for treatment.

The distribution of respiratory tract infection in children has seasonal characteristics; understanding the distribution of pathogenic bacteria is therefore essential. There are differences in the pathogenic bacteria of respiratory tract infection among children during specific seasons.¹⁵ Different seasons have different temperatures and levels of humidity, and viral respiratory infection is more common in winter and spring.¹⁶ Children with congenital heart defects who have community-acquired pneumonia in winter are more susceptible to S. aureus and E. coli than children with simple community-acquired pneumonia. Our results showed that the constituent ratio of gram-negative bacteria in children with respiratory tract infection was significantly higher than that of gram-positive bacteria in spring, summer, autumn and winter, and gram-negative bacteria were significantly more common in spring and summer than in autumn and winter. The constituent ratio of gram-positive S. aureus was higher in autumn and winter than in spring and summer. Gram-positive bacteria did not differ seasonally among children less than 1 month old. In children aged 1 month to 3 years, gram-negative bacteria were significantly more common than gram-positive bacteria. In autumn and winter, when there is a high incidence of respiratory infection in children, clinicians should be on the alert for S. aureus.¹⁷

The distribution of pathogenic bacteria in children’s respiratory tract infection among different age groups also varied. Newborns were an especially vulnerable group and experienced environmental changes both inside and outside the uterus. The pathogenic bacteria also had age characteristics. The age threshold is one of the criteria used to define the high-risk infection group. The pathogens exhibited a consistent age pattern, with children less than 3 years of age at the highest risk. Reports have shown that K. pneumoniae, E. coli, A. baumannii, E. cloacae and H. influenzae are the predominant gram-negative bacteria isolated from the lower respiratory tract of newborns.^18,19 The results of this study were similar to those reported above.

Understanding the distribution of pathogenic bacteria in respiratory tract infection of children of all ages can provide a key basis for treatment. Community-acquired pneumonia in children aged 4 months to 5 years is mainly caused by respiratory viral infections, with a higher proportion of viral infections occurring at younger ages and bacterial infections present at almost all ages.¹³ As in previous reports, S. aureus was the most common cause of respiratory tract infection in children aged from 1 month to 3 years as well as in the 3–6 age group. We suggest that the possibility of S. aureus infection should be considered when evaluating respiratory tract infections in infants and preschool children. The study also suggested that H. influenzae in the <1-month, 1-month–3-years, 3–6-years and >6-years age groups was gradually increasing. We postulate that H. influenzae is the main pathogen of respiratory tract infection in school-age children, which is consistent with the reported result that H. influenzae is the most prevalent pathogen, accounting for 22.58% of sputum culture in severe pneumonia. However, the above report did not study the constituent ratio of H. influenzae and its changing trends within different age groups. In this study, we found that the distribution of H. influenzae in the respiratory tract had been increasing until recent years and reflects the age difference. Moreover, this study suggests that the constituent ratio of E. coli in the 3–6-years group was significantly lower than that of the younger children’s groups. This may be because of the immaturity of immune function in infants and young children, making them more prone to E. coli and other opportunistic bacterial infections. It may also be related to the use of nappies in infants and young children.

Sex may also be a factor in the presence of different pathogens in children with respiratory tract infection, but there are few related studies. A study tested 12,359 children (who had acute respiratory infection) from Guangzhou, China, for 9 pathogens (Mycoplasma, Chlamydia, Legionella pneumophila, Rickettsia and influenza viruses), revealing that the infection rates of Mycoplasma antibody-IgM, influenza B virus and parainfluenza virus in girls were significantly different from those in boys.²⁰ The results showed that the distribution of pathogenic bacteria in children’s respiratory tracts differed between sexes. The constituent ratio of gram-negative bacteria was higher than that of gram-positive bacteria in both boys and girls. It should be noted that girls are more susceptible to H. influenzae than boys.

The lack of patient baseline characteristics, classification of respiratory infection, consideration of the involvement of viruses and mycoplasma, classification of gram-negative rods and drug susceptibility data, as well as the difficulty in distinguishing whether the detected bacteria represent true infections or are just contaminants and bacteria that cannot grow in these media are limitations of our study. Because of the methods we used to collect specimens, our results need to be further verified. As for the limitations of retrospective analysis, multi-centre studies or more data are needed to refine the results.

Conclusion

Between 2014 and 2021, gram-negative bacteria were the predominant pathogens among school-age children, especially during spring and summer. The proportion of S. aureus in children has been decreasing yearly, whereas the proportion of H. influenzae in children has increased annually over the past 4 years. Clinicians should pay attention to changes in the pathogenic spectrum and improve the monitoring and management of drug resistance.