High incidence of the virus among respiratory pathogens in children with lower respiratory tract infection in northwestern China

Abstract

Lower respiratory tract infection (LRTI) is one of the major reasons for childhood mortality that threaten the health of the public. We aimed to investigate the epidemiological pathogens and their infection analysis among children with LRTI. Sputum specimens were collected for polymerase chain reaction detection and microbiological tests to identify the viral infection and bacterial infection. The serological specimens were separated from venous blood using for Mycoplasma pneumoniae and Chlamydia pneumoniae detection. The virus was confirmed in 86.2% of the children. Human rhinovirus (38.3%), respiratory syncytial virus (32.1%), and parainfluenza virus type 3 (27.2%) were the most frequently identified pathogens. Patients with viral and bacterial coinfection showed younger age (p = 0.032), a higher proportion of wheezing rales (p = 0.032), three depressions sign (p = 0.028), and tachypnea (p = 0.038), and more likely associated with severe pneumonia (p = 0.035). Additionally, older children were more susceptible to viral‐atypical bacterial coinfection (p = 0.032). Vomiting (p = 0.011) and fever (p = 0.003) were more likely to occur in children with viral‐atypical bacterial coinfection. Attention should be paid to the virus infection of LRTI, as viral‐bacterial coinfection and viral‐atypical bacterial co‐infection may have a detrimental impact on the gravity of LTRI.

Abbreviations

ADV

adenovirus

ALT

alanine aminotransferase

AST

aspartate aminotransferase

BUN

blood urea nitrogen

CK‐MB

creatine kinase isoenzyme

CoV229E/NL63

human coronavirus 229E/NL63

CoVOC43

human coronavirus OC43

CP

Chlamydia pneumoniae

CRE

creatinine

CRP

C‐reactive protein

Flu A

influenza virus A

Flu B

influenza virus B

HBoV

human bocavirus

HEV

human enterovirus

HGB

hemoglobin

HMPV

human metapneumovirus

HRV

human rhinovirus

LRTI

lower respiratory tract infection

MP

Mycoplasma pneumoniae

NEUT %

neutrophil percentage

PCR

polymerase chain reaction

PCT

procalcitonin

PIV1

parainfluenza virus type 1

PIV2

parainfluenza virus type 2

PIV3

parainfluenza virus type 3

PIV4

parainfluenza virus type 4

PLT

platelet

RBC

red blood cell

RSV

respiratory syncytial virus

SaO₂

oxygen saturation

WBC

white blood cell

1. INTRODUCTION

Lower respiratory tract infection (LRTI) is a severe respiratory disease that includes pneumonia, bronchitis, and bronchiolitis. It causes 2.38 million deaths globally in 2016, with 0.65 million deaths in children under 5, ¹ resulting in a high mortality rate among infants and young children, especially in developing nations. ² Viruses (such as human respiratory syncytial virus [RSV], human rhinovirus [HRV], human metapneumovirus [HMPV], and human parainfluenza virus 3 [PIV3]), ³ , ⁴ , ⁵ bacteria (such as Streptococcus pneumoniae, Haemophilus influenza type b, Klebsiella pneumoniae, and Escherichia coli) ¹ , ⁶ and atypical bacterial (such as Chlamydia pneumoniae [CP], Mycoplasma pneumoniae [MP], Legionella pneumophila, and Coxiella burnetii) ⁷ , ⁸ are the most prevalent causative pathogens reported of LRTI. Other pathogens that cause illness in the lower respiratory tract are fungi and parasites. In recent years, molecular technologies and novel assay capability have promoted marked development in the sensitivity and specificity of respiratory pathogens detection during the epidemiology of LRTI. ⁹ , ¹⁰

The prevalence of pathogens that cause LRTIs has wide variations, and the prevalence of pathogens changed between locations because of the differences in geographical location, climate, vaccination coverage, socioeconomic status, and customs. In addition, the main causative agents in LRTI differ in respect of different clinical diagnoses. It has been reported that viruses were the main etiological agents, with RSV as the most epidemic virus causing pneumonia, and Streptococcus pneumoniae remains the most common bacterial pathogen, and in an atypical bacterial group that was MP. ¹¹ , ¹² , ¹³ Ninety percent of the bronchitis cases were caused by virus, and the common virus including HRV, influenza, RSV, human enterovirus (HEV), PIV, HMPV, and coronavirus. ¹⁴ Bacteria were shown to be responsible for 1%–10% of acute bronchitis cases, however, atypical bacteria played a relatively minor role. ¹⁵ , ¹⁶ Babies within 6 months had a higher risk of getting bronchiolitis, and the most frequently confirmed microorganisms were RSV and MP. ¹⁷ , ¹⁸ These data indicate that viruses were the main epidemic agent of LRTI. Thus, attempts should be made for boosting the range of respiratory pathogens and developing effective vaccine to curb the circulation of pathogens. The purpose of the present study is to investigate the epidemiology of children in Ningxia who were admitted for LRTI treatment, analyze the clinical characteristics, seasonal distribution, and coinfection combinations, and provide a favorable basis for preventing LRTI.

2. MATERIALS AND METHODS

2.1. Patients and specimens

A total of 989 hospitalized children under the age of 12 were in inpatient therapy for LRTI from January 2018 to December 2019 at Yinchuan Women and Children Healthcare Hospital, they were divided into five categories according to age: (<12 months, 12–35 months, 36–71 months, ≥72 months, and ≥10 years), as previously described. ¹⁹ , ²⁰ , ²¹ All patients were examined and diagnosed clinically by the pediatrician, and the clinical criteria of pneumonia, bronchitis, and bronchiolitis are defined according to World Health Organization guidelines of common childhood illnesses including respiratory symptoms and chest radiographic findings. ²² Exclusion criteria included aspiration pneumonia and healthcare‐associated pneumonia in this research.

The basic information and clinical information were recorded respectively under the awareness of guardians and family members of each patient which included: (i) the general information about the name, sex, age, and inpatient days of the patient, (ⅱ) the data of vital signs including body temperature, respiration rate and percutaneous oxygen saturation (SaO₂) of each patient, (iii) the clinical symptoms such as cough, vomiting, rales, pharyngeal hyperemia, three depressions sign (depressions in the supraclavicular fossa, suprasternal fossa, and intercostal space), and wheezing, (ⅳ) the laboratory blood biochemistry indices like white blood cell (WBC), neutrophil percentage (NEUT %), hemoglobin (HGB), platelet (PLT), red blood cell (RBC), procalcitonin (PCT), C‐reactive protein (CRP), blood urea nitrogen (BUN), creatinine (CRE), aspartate aminotransferase (AST), alanine aminotransferase (ALT), lactate dehydrogenase (LDH), and creatine kinase isoenzyme (CK‐MB).

Sputum specimens were drawn out with a negative pressure suction device by professional and skilled nurses. A disposable silicone hose was used during the sputum aspiration, which was connected to the suction device, and the other side from the nasal cavity or oropharynx was inserted into the respiratory tract of the children. Secretions of the respiratory tract were obtained by suction and then immediately diluted (the ratio of 1 to 1) with virus protection solution (Solarbio) and stored at −80°C before use. Sputum specimens with less than 10 squamous epithelial cells and more than 25 leukocytes in low power fields were considered to be qualified.

2.2. Virus DNA/RNA extraction and detection

The collected samples were placed at room temperature, a total of 100 μl liquid volume of sample solution were used after thawing, and 30 μl viral DNA/RNA were extracted from the supernatant by using Quick‐DNA/RNA^TM Viral Kit (ZYMO RESEARCH), the eluted DNA/RNA was used immediately. TransScript® One‐Step GenomicDNA Removal and complementary DNA (cDNA) Synthesis SuperMix (TransGen) was applied to synthesize the cDNA of each sample. All the cDNA samples were stored at −20°C for the following detection steps.

The detection of respiratory viruses was performed with cDNA samples under the guidance of Seeplex® RV15 ACE Detection Kit (Seegene) including human bocavirus (HBoV), HRV, parainfluenza virus 1 (PIV1), parainfluenza virus 2 (PIV2), PIV3, parainfluenza virus 4 (PIV4), RSV, human coronavirus OC43 (CoVOC43), human coronavirus 229E/NL63 (CoV229E/NL63), adenovirus (ADV), influenza virus A (Flu A), Flu B, HEV, and HMPV. The real‐time polymerase chain reaction (PCR) reaction was applied for the amplification of different viruses, and gel electrophoresis was used for identifying the 14 viruses with PCR products. Positive and negative controls were also included for each PCR to avoid and exclude carryover contamination.

2.3. Detection of respiratory bacteria

During their hospital stay, the recruited patients’ specimens were sent to the clinical laboratory of Yinchuan Women and Children Healthcare Hospital for microbiological examination. The gold standard for identifying bacterial microorganisms remains to be culture, and specimens were inoculated on blood, chocolate, MacConkey‐agar plate, or Sabouraud's plate separately. The cultivated strains were identified utilizing the VITEK‐2 compact automated bacterial identification system (Bio‐Merieux company production), and the results were documented by a staff of the hospital.

2.4. Detection of atypical bacteria

Children's venous blood was collected by a skilled nurse and sent to the clinical laboratory for follow‐up testing. After being centrifuged, the serum was separated from the venous blood and diluted to different concentrations with serum diluents, then a Passive Particle Agglutination Kit (SERODIA®‐MYCO II) was employed for detecting the antibody against MP. Samples were tested according to the instructions of the kit, and a single serum antibody titer over 1:40 indicates positive infection for MP.

A total of 10 μl serum or 20 μl venous blood was required for the detection of CP and Rapid Gold Immunochromatography Assay Test for Cpn‐IgM was applied for the detection. Specimens were detected following the instructions of the kit, and two clear bars indicate a positive sample of CP infection.

2.5. Identification of coinfection and single infection

Coinfection was described as the detection of ≥2 bacterial, viral, or atypical bacterial pathogens in any combination, whereas a single infection was defined as the presence of only one virus, one bacterium, or one atypical bacterium in a sample.

2.6. Statistical analysis

Statistical processing and analysis were based on the result of SPSS (version 22.0). The counting data were expressed as the number of cases and percentage (%). The χ ² test or Fisher's exact test was used to compare the differences in the distribution of the tested laboratory variables and clinical characteristics between the groups. A p < 0.05 was considered statistically significant. Measurement data were shown by median (interquartile range) [M (P ₂₅, P ₇₅)], and comparison between groups was performed by Mann–Whitney U test.

3. RESULTS

3.1. Detection of LRTI

A total of 989 eligible participants aged 1 day to 146 months were recruited. The virus was confirmed in 86.2% (853/989) of the children, containing 492 boys and 361 girls, with no significant difference (p = 0.680), but a difference was observed in the age group (p < 0.05), showing that viral infection was more common in 1‐year‐old newborns. Bacteria were identified in 24.6% (243/989) of the participants, and there was a significant age difference (p < 0.001), indicating that children under 1 year old were more susceptible to bacterial infection. The atypical bacteria‐positive population recorded 20.5% (203/989) of the total cases, and children aged 1–6 years old got the highest positive cases (p < 0.001). The common diagnosis of LRTI was pneumonia, bronchitis, and bronchiolitis, 88.7% (878/989) of children had been diagnosed with pneumonia (Table 1). In general, there was no significant difference between the clinical diagnosis in pathogen infections, but children infected with atypical bacteria were more likely associated with pneumonia (p < 0.001).

Table 1.

Detection of pathogens by age, gender, and clinical diagnosis

	Sample tesed (N = 989)	Virus positive (n = 853)	Bacteria positive (n = 243)	Atypical bacteria positive (n = 203)
Age
<12 months	371	318	140	43
12–35 months	290	257	50	62
36–71 months	236	208	32	62
≥72 months	74	56	16	30
≥10 years	18	14	5	6
p		0.037	<0.001	<0.001
Gender
Male	573	492	146	109
Female	416	361	97	94
p		0.680	0.435	0.170
Clinical diagnosis
Pneumonia	878	754	225	196
Bronchitis	104	92	17	6
Bronchiolitis	7	7	1	1
p		0.622	0.080	<0.001

Note: Bold values are statistically significant.

3.2. Pathogens detected in LRTI

Sputum samples were analyzed for the 14 pathogens reported in this study using multiple PCR. In LRTI cases, a total of 265 (26.8%, 265/989) patients were confirmed positive for only one pathogen, including 208 (21.0%, 208/989) cases caused by the virus, 22 (2.2%, 22/989) cases caused by bacteria, and 35 (3.5%, 35/989) cases caused by atypical bacteria. Furthermore, coinfection with two or more pathogens was discovered in 653 (66.0%, 653/989) of LRTI cases, whereas the remaining 71 samples were caused by unknown etiology. Table 2 shows the prevalence of respiratory pathogens. HRV, RSV, and PIV3 were the most prevalent among the 14 viruses, with 38.3% (379/989), 32.1% (317/989), and 27.2% (269/989), respectively. Klebsiella pneumoniae, Escherichia coli, and Streptococcus pneumoniae were the three most frequently isolated bacterial pathogens in children with LRTI. MP was the most epidemically appearing atypical bacteria, with 198 positive cases (20.0%, 198/989).

Table 2.

Identification of single infection and coinfection in children with LRTI

Pathogens	Single infection	Coinfection	Positive n (%)
Viruses
Human bocavirus	7	86	93 (9.4%)
Human rhinovirus	60	319	379 (38.3%)
Parainfluenza virus type 3 (PIV3)	27	242	269 (27.2%)
Respiratory syncytial virus	50	267	317 (32.1%)
Human coronavirus OC43	3	10	13 (1.3%)
PIV1	5	82	87 (8.8%)
Human coronavirus 229E/NL63	2	19	21 (2.1%)
PIV4	5	31	36 (3.6%)
Adenovirus	8	54	62 (6.3%)
PIV2	5	43	48 (4.9%)
Influenza A virus (Flu A)	5	37	42 (4.2%)
Human enterovirus	1	37	38 (3.8%)
Human metapneumovirus	30	126	156 (15.8%)
Flu B	0	5	5 (0.5%)
Bacteria
Klebsiella pneumoniae	10	58	68 (6.9%)
Streptococcus pneumoniae	2	36	38 (3.8%)
Escherichia coli	7	40	47 (4.8%)
Pseudomonas aeruginosa	1	23	24 (2.4%)
Staphylococcus aureus	0	21	21 (2.1%)
Enterobacter aerogenes	1	9	10 (1.0%)
Enterobacter cloacae	0	23	23 (2.3%)
Haemophilus influenzae	1	15	16 (1.6%)
Acinetobacter baumannii	0	8	8 (0.8%)
Citrobacter koseri	0	1	1 (0.1%)
Klebsiella oxytoca	0	6	6 (0.6%)
Enterobacter gergoviae	0	1	1 (0.1%)
Serratia marcescens	0	1	1 (0.1%)
Enterococcus faecium	0	2	2 (0.2%)
Morganella morganii	0	1	1 (0.1%)
Stenotrophomonas maltophilia	0	3	3 (0.3%)
Bordetella bronchiseptica	0	1	1 (0.1%)
Acinetobacter junii	0	1	1 (0.1%)
Raoultella planticola	0	2	2 (0.2%)
Atypical bacteria
Mycoplasma pneumoniae	34	164	198 (20.0%)
Chlamydia pneumoniae	1	4	5 (0.5%)
Total	265 (26.8%)

Abbreviation: LRTI, lower respiratory tract infection.

3.3. Coinfection combinations of the respiratory pathogens

Coinfection of respiratory infections was discovered epidemically in LRTI, according to our findings. To explore the coinfection in LRTI, respiratory pathogens that cause pneumonia, bronchitis, and bronchiolitis with distinct coinfection combinations were shown in Table 3. The most common type of coinfection in children with pneumonia was viral coinfection, and the most prevalent coinfection combination was found in PIV3 + HRV, which accounted for 33 cases. RSV + Escherichia coli and PIV3 + Klebsiella pneumoniae were the most common combinations in viral and bacterial coinfection. In atypical bacterial and viral coinfection in patients with pneumonia, RSV + MP was the most commonly identified complex. Twenty‐seven coinfection combinations were found in viral, bacterial, and atypical bacterial infections in children with pneumonia. PIV3 + HRV was the most prevalent combination in children diagnosed with bronchitis. Viral coinfection, viral and bacterial coinfection accounted for the majority of infections in children diagnosed with bronchitis. In general, viral coinfection was the main contributor among the coinfection cases in our study, but bacterial coinfection and bacterial‐atypical bacterial coinfection contributed to a little population in LRTI.

Table 3.

The coinfection combination of pathogens in LRTI

		Coinfection combination	Number of patients
Pneumonia	Viral coinfection		246
		PIV3 + HRV	33
		RSV + HRV	12
		HRV + HMPV	11
		Other combination	190
	Bacterial coinfection		2
	Virus + bacteria		167
		RSV + Escherichia coli	5
		PIV3 + Klebsiella pneumoniae	5
		HRV + Streptococcus pneumoniae	4
		Other combination	153
	Virus + atypical bacteria		128
		RSV + MP	17
		HRV + MP	13
		PIV3 + MP	10
		Other combination	88
	Bacteria + atypical bacteria		7
		Acinetobacter baumannii + MP	2
		Staphylococcus aureus + MP	2
		Other combination	3
	Virus + bacteria + atypical bacteria		27
	Total		577
Bronchitis	Viral coinfection		54
		PIV3 + HRV	9
		RSV + HRV	4
		PIV3 + HEV	3
		Other combination	38
	Virus + bacteria		15
	Virus + atypical bacteria		2
	Virus + bacteria + atypical bacteria		2
	Total		73
Bronchiolitis		PIV3 + HRV	1
		HRV + Acinetobacter baumannii	1
		RSV + MP	1
	Total		3

Abbreviations: HEV, human enterovirus; HMPV, human metapneumovirus; HRV, human rhinovirus; LRTI, lower respiratory tract infection; MP, Mycoplasma pneumoniae; PIV3, parainfluenza virus type 3; RSV, respiratory syncytial virus.

3.4. Seasonal distribution of pathogens in LRTI

Seasonal distribution of viruses, main bacteria, and atypical bacteria was observed and depicted in Figure 1 to explore the seasonal distribution of infectious pathogens within the recruited patients. Positive instances of pathogens were divided into four categories: spring (March–May), summer (June–August), autumn (September–November), and winter (December–February). Overall, pathogen properties vary by season, and the majority of pathogens showed a marked seasonal variation. Patients with HRV who required hospitalization saw the largest seasonal peak in viral infection, which occurred more frequently in the fall and less frequently in winter. Furthermore, PIV4, CoV229/NL63, and CoVOC43 exhibited similar seasonal patterns with HRV. In terms of bacterial infection and atypical bacterial infection, Klebsiella pneumoniae and MP had the biggest seasonal peak. During the epidemic season, we saw a lot of PIV3 and Pseudomonas aeruginosa activity, with peaks in summer and fall and dips in winter. PIV1, PIV2, ADV, and HEV were active in summer and inactive in spring and winter. HBoV, Flu A, Staphylococcus aureus, Enterobacter cloacae, Haemophilus influenzae, and MP, on the other hand, were found more frequently in winter and less frequently in spring and fall. Klebsiella pneumoniae has a similar seasonality, with its peak occurring in the fall. RSV, HMPV, Streptococcus pneumoniae, Escherichia coli, Enterobacter aerogenes, and CP did not show a consistent seasonal trend, despite the significant rise in several of them. Virus outbreaks occurred primarily in autumn, while bacterial and atypical bacterial outbreaks occurred primarily in winter, according to our findings.

Figure 1.

Seasonal distribution of respiratory pathogens in LRTI. (A) Seasonal distribution of viruses in LRTI. (B) Seasonal distribution of bacteria in LRTI. (C) Seasonal distribution of atypical bacteria in LRTI. CoV229E/NL63, human coronavirus 229E/NL63; CoVOC43, human coronavirus OC43; CP, Chlamydia pneumoniae; Flu A, influenza virus A; HBoV,human bocavirus; HEV, human enterovirus; HMPV, human metapneumovirus; HRV, human rhinovirus; LRTI, lower respiratory tract infection; MP, Mycoplasma pneumoniae; PIV3, parainfluenza virus type 3; RSV, respiratory syncytial virus.

3.5. The clinical features of children in different infection groups

Viruses played an important role in LRTI, and clinical data and behavior of patients with or without viral‐bacterial coinfection and viral‐atypical bacterial coinfection in virus‐positive cases were collected to explore the impact of viral infection in different groups. The clinical features of children with or without viral and bacterial co‐infection in virus‐positive cases were shown in Table 4. Among the LRTI patients, 853 cases were proved to be infected with viruses, patients with viral and bacterial coinfection documented in 25.0% of virus‐positive cases, who showed younger age (p = 0.032), a higher percentage of wheezing rales (p = 0.032), three depressions sign (p = 0.028), and tachypnea (p = 0.038), and more likely to associate with severe pneumonia (p = 0.035). Besides, the level of SaO₂ (p < 0.001), NEUT % (p = 0.005), HGB (p < 0.001), RBC (p < 0.001), CRP (p = 0.04), BUN (p = 0.013), and CRE (p < 0.001) was lower than patients without viral and bacterial coinfection, but the concentration of PLT (p = 0.003), AST (p < 0.001), ALT (p < 0.001), and CK‐MB (p = 0.011) was higher. Likewise, the characteristics of children with or without viral‐atypical bacterial coinfection were displayed in Table 5. We discovered older children were more susceptible to viral‐atypical bacterial coinfection (p = 0.032). Vomiting (p = 0.011) and fever (p = 0.003) were more likely to occur in children with viral‐atypical bacterial coinfection compared to those without viral‐atypical bacterial coinfection. Additionally, higher percentages of neutrophils (p < 0.001), a higher concentration of HGB (p = 0.001), CRP (p = 0.015), and CRE (p < 0.001) were discovered in children with viral‐atypical bacterial coinfection. However, the concentration of WBC (p = 0.018), PLT (p = 0.006), AST (p < 0.001), ALT (p < 0.001), and CK‐MB (p = 0.003) were lower than children without viral‐atypical bacterial coinfection.

Table 4.

Characteristics of groups with or without viral‐bacterial coinfection in virus‐positive patients

	The group with viral‐bacterial coinfection (n = 213)	The group without viral‐bacterial coinfection (n = 640)	p
Malea	131 (61.5%)	361 (56.4%)	0.192
Inpatient daysb	7 (5, 8)	6 (5, 8)	0.151
Age (month)b	15 (6, 39)	22 (8, 43)	0.032
Symptoms and signs
Cougha	204 (95.8%)	612 (95.6%)	0.926
Vomitinga	47 (22.1%)	156 (24.4%)	0.493
Ralesa	62 (29.1%)	151 (23.6%)	0.107
Wheezing ralesa	56 (26.3%)	124 (19.4%)	0.032
Pharyngeal hyperemiaa	166 (77.9%)	486 (75.9%)	0.552
Wheezinga	42 (19.7%)	108 (16.9%)	0.345
Fever (>37.5°C)a	111 (52.1%)	364 (56.9%)	0.226
Three depressions signa	19 (8.9%)	31 (4.8%)	0.028
Tachypneaa	34 (16.0%)	68 (10.6%)	0.038
SaO2 (%)b	96.00 (96.00, 97.00)	96.58 (96.00, 97.00)	<0.001
Respiratory failurea	20 (9.4%)	49 (7.7%)	0.422
Severe pneumoniaa	15 (7.0%)	23 (3.6%)	0.035
Laboratory findings
WBC (109/L)b	9.03 (7.01, 10.83)	9.11 (7.10, 10.59)	0.939
NEUT %b	36.70 (24.92, 48.81)	41.45 (29.63, 52.66)	0.005
HGB (g/L)b	123.75 (116.00, 130.58)	129.00 (121.00, 135.41)	<0.001
PLT (109/L)b	282.00 (339.00, 410.50)	320.00 (265.00, 375.67)	0.003
RBC (1012/L)b	4.76 (4.32, 6.56)	5.08 (4.54, 8.06)	<0.001
PCT (ng/ml)b	1.48 (0.11, 5.07)	2.60 (0.16, 4.80)	0.305
CRP (mg/L)b	3.07 (0.62, 10.19)	4.68 (1.00, 12.08)	0.040
BUN (mmol/L)b	2.50 (1.70, 3.30)	2.70 (2.00, 3.50)	0.013
CRE (µmol/L)b	25.00 (20.00, 30.00)	28.00 (23.00, 33.00)	<0.001
AST (U/L)b	35.30 (28.50, 43.55)	32.25 (25.93, 38.08)	<0.001
ALT (U/L)b	17.30 (12.10, 27.15)	13.35 (9.80, 19.92)	<0.001
LDH (U/L)b	297.00 (260.50, 339.50)	293.00 (259.25, 329.75)	0.244
CK‐MB (U/L)b	28.40 (22.15, 41.80)	27.25 (21.00, 35.18)	0.011

Note: Bold values are statistically significant. p values obtained by comparison of groups with or without viral and bacterial coinfection, significant difference p < 0.05.

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CK‐MB, creatine kinase isoenzyme; CRE, creatinine; CRP, C‐reactive protein; HGB, hemoglobin; LDH, lactate dehydrogenase; NEUT %, neutrophil percentage; PCT, procalcitonin; PLT, platelet; RBC, red blood cell; SaO₂, oxygen saturation; WBC, white blood cell.

^a Comparison of frequency distribution was performed by the χ ² test; Fisher's exact test was applied when at least one expected frequency was <5.

^b Comparison of quantitative variables among the groups was performed using the Mann–Whitney U test.

Table 5.

Characteristics of groups with or without viral‐atypical bacterial coinfection in virus‐positive patients

	The group with viral‐atypical bacterial coinfection (n = 162)	The group without viral‐atypical bacterial coinfection (n = 691)	p
Malea	86 (53.1%)	406 (58.8%)	0.189
Inpatient daysb	6 (5, 8)	6 (5, 8)	0.364
Age (month)b	27 (9, 45)	18 (7, 42)	0.032
Symptoms and signs
Cougha	155 (95.7%)	661 (95.7%)	0.991
Vomitinga	51 (31.5%)	152 (23.8%)	0.011
Ralesa	46 (28.4%)	167 (24.2%)	0.263
Wheezing ralesa	36 (22.2%)	144 (20.8%)	0.698
Pharyngeal hyperemiaa	131 (80.7%)	521 (75.4%)	0.140
Wheezinga	28 (17.3%)	122 (17.7%)	0.911
Fever (>37.5°C)a	107 (66.0%)	368 (53.3%)	0.003
Three depressions signa	11 (6.8%)	39 (5.6%)	0.576
Tachypneaa	16 (9.9%)	86 (12.4%)	0.364
SaO2 (%)b	96.52 (96.00, 97.31)	96.44 (95.00, 97.00)	0.158
Respiratory failurea	9 (5.6%)	60 (8.7%)	0.189
Severe pneumoniaa	9 (5.6%)	29 (4.2%)	0.451
Laboratory findings
WBC (109/L)b	8.59 (6.80, 10.06)	9.20 (7.20, 10.82)	0.018
NEUT %b	45.55 (34.10, 60.73)	38.50 (27.60, 50.00)	<0.001
HGB (g/L)b	130.03 (122.75, 138.00)	127.43 (119.00, 133.00)	0.001
PLT (109/L)b	304.68 (248.25, 358.00)	328.38 (276.00, 393.00)	0.006
RBC (1012/L)b	5.06 (4.52, 7.78)	4.95 (4.50, 7.71)	0.864
PCT (ng/ml)b	0.64 (0.11, 4.62)	2.77 (0.15, 4.90)	0.080
CRP (mg/L)b	5.48 (1.64, 12.84)	4.00 (0.71, 11.90)	0.015
BUN (mmol/L)b	2.60 (1.85, 3.20)	2.70 (2.00, 3.50)	0.261
CRE (µmol/L)b	29.00 (23.00, 36.00)	26.00 (22.00, 32.00)	<0.001
AST (U/L)b	30.05 (23.95, 35.40)	33.50 (26.90, 40.10)	<0.001
ALT (U/L)b	12.20 (8.98, 17.03)	15.00 (10.80, 22.10)	<0.001
LDH (U/L)b	291.00 (251.00, 326.25)	295.00 (261.00, 334.00)	0.105
CK‐MB (U/L)b	24.55 (19.65, 33.64)	27.90 (22.00, 36.48)	0.003

Note: Bold values are statistically significant. p values obtained by comparison of groups with or without viral and atypical bacterial coinfection, with significant difference p < 0.05.

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CK‐MB, creatine kinase isoenzyme; CRE, creatinine; CRP, C‐reactive protein; HGB, hemoglobin; LDH, lactate dehydrogenase; NEUT %, neutrophil percentage; PCT, procalcitonin; PLT, platelet; RBC, red blood cell; SaO₂, oxygen saturation; WBC, white blood cell.

^a Comparison of frequency distribution was performed by the χ ² test; Fisher's exact test was applied when at least one expected frequency was <5.

^b Comparison of quantitative variables among the groups was performed using the Mann–Whitney U test.

4. DISCUSSION

In this study, we explored the prevalence of pathogens such as viruses, bacteria, and atypical bacteria in children who acquired LRTI, which revealed a unique pathogen epidemic pattern in northwestern China. Real‐time PCR was utilized in our investigation to assure viral detection sensitivity, and our results suggest that viruses are the predominant agent (86.2%) over other respiratory pathogens, which was in accordance with a recent study in Changsha (83.2%, 909 out of 1092), with a mean of 20.40 ± 24.60 months. ²³ Besides, we discovered bacteria and atypical bacteria were documented in 24.5% and 20.5%, respectively, which was in line with the result of Shanghai. ²⁴ However, they considered the detection number of atypical bacteria increased by age, and we discovered more than half of atypical bacterial LRTI was in children aged 1–6 years (61.1%, 124/203). Our findings indicate the fact that atypical bacterial infection in our study is trending toward younger children, this unusual association may be related to a higher than normal prevalence of MP during the study period, and further observation is needed to confirm this finding. The highest detection rate of bacterial LRTI was in children <12 months according to our results, whereas a study investigating acute respiratory infections suggested the highest number of bacterial infections was in children aged 1–4. ²⁵ Age and season are important determinants of bacteria colonization, and the bacterial carriage can be influenced by the bacterial and viral synergistic or antagonistic effect, ²⁶ which may explain the differences. Inadequate nutrition, vaccines, and low immunity all seem to be contributing factors to young infants’ susceptibility. As a result, it should be emphasized that young children were at higher risk of being infected with respiratory pathogens.

Our analysis showed that of the 14 viral pathogens studied, the presence of HRV was the most prevalent in pediatric patients (38.3% of LRTI cases), followed by RSV (32.1% of LRTI cases) and PIV3 (27.2% of LRTI cases). Consistent with our results, a Saudi Arabia study also confirmed the highest prevalence of HRV. ²⁰ However, HRV was not examined in other investigations, which identified RSV as the major pathogen, ²⁷ , ²⁸ , ²⁹ , ³⁰ and future research should include HRV in the pathogen panel. In our study, 60% of the hospitalized children were infected by two or more pathogens, the high incidence of multiple infections was also presented in other studies. ²⁰ , ²³ , ³¹ It has been reported that Streptococcus pneumoniae was the predominant factor for LRTI in pediatric patients, what's worse, it contributed to more than 50% of the death worldwide. ¹ But this serious situation might be reversed owing to the increase in vaccination coverage. Other commonly identified bacterial etiologies include Haemophilus influenzae, Staphylococcus aureus, and Klebsiella pneumoniae. In our study, we discovered Klebsiella pneumoniae was the most epidemic bacteria occurring in LRTI, which confirmed that Klebsiella pneumoniae was the only bacteria with an increasing isolation rate from 7.1% to 12.1% at the emergency department due to its high antibiotic resistance. ³² In addition, MP was detected as the most common atypical bacteria in our study and other studies, ³³ and the varied incidence of MP may attribute to the studied population and diagnosis method.

Given the high incidence of coinfections revealed in our investigation, we further demonstrated the specific pattern of codetection pairs in children diagnosed with pneumonia, bronchitis, and bronchiolitis, respectively. In our study, we confirmed viral coinfection as the main contributor in LRTI, and PIV3 + HRV was found to be the most common coinfection combination in the cases of pneumonia and bronchitis, but Brazilian and Canadian researchers had described RSV + HRV as the most popular combination. ²⁸ , ³⁴ , ³⁵ The conflicting results may attribute to viral subtypes, which may have a higher pathogenicity than others. Besides, climatic variations and different demographic characteristics may have an impact on the outcomes. Bacterial and viral coinfection may present more severe clinical outcomes, ³⁶ which accounted for the most common type of coinfection in LRTI. ²⁴ In the current study, the most common combination was RSV + Escherichia coli, different from the previous study. Besides, we speculated the highest proportion of RSV + MP in the viral‐atypical bacterial coinfections, which was in favor of Shanghai's result. ²⁴

Participants coinfected with bacteria and virus, worse clinical results were displayed including a higher incidence of wheezing rales, three depressions sign, tachypnea, a higher percentage of severe pneumonia, and a lower level of SaO₂. In addition, children with viral‐atypical bacterial coinfection were easier to have vomiting and fever. The results revealed that mixed infection could play a negative role in the progression of LRTI. ³⁶ Appropriate clinical treatment requires differentiating various pathogens in LRTI correctly, and some laboratory findings will make sense for the physician's diagnosis. A study that assessed the serum amyloid A (SAA), PCT, CRP, and interleukin 6(IL‐6) in children demonstrated that viral infection can be identified by increasing SAA and CRP. Besides, both bacterial infection and coinfection can be differentiated by PCT, and simultaneous detection of SAA, PCT, IL‐6, and CRP can improve the sensitivity and specificity of various infection forms. ³⁷ However, our data reflect that in virus‐positive cases, the percentage of NEUT, the concentration of HGB, CRP, CRE, AST, ALT, and CK‐MB were higher in children with viral‐atypical bacterial coinfection, but lower in children with viral‐bacterial coinfection than their compared groups, which indicates the severeness of viral‐atypical bacterial coinfection. Despite the severe clinical outcomes and impact of coinfections in LRTI, the controversial results showed that children with codetections had significantly lower concentrations of RBC, BUN, and WBC than the compared groups. Several factors may explain the conflicting results, such as the difference of serotypes in population, research stage, and testing method. Since the children were in hospital therapy, it is hard to avoid the intervening action of drugs, which may have an influence on the results.

A distinct pattern of seasonal distribution of pathogens in LRTI was displayed in this study. Located in the inland region of midlatitude of China, and far away from the ocean, Ningxia has a typical continental semihumid and semiarid climate. The rainy season is mainly from June to September. It has four distinct seasons, and there are drastic temperature changes between day and night, which is responsible for the higher frequency of LRTI. The prevalence of pathogens in dry areas is quite different from rainy areas. ²⁰ , ³⁸ , ³⁹ Studies have reported that the prevalence of respiratory pathogens has seasonal variation, ¹⁹ , ³⁵ , ³⁸ , ³⁹ and it has been supposed that the circulation of the virus was more likely to rise during the colder winter months. ¹⁹ We confirmed the epidemic of the virus occurred primarily in the autumn, while bacterial and atypical bacterial outbreaks occurred primarily in the winter. The most prevalent epidemiological period of HRV in our study was in autumn, which was consistent with another report published in northwestern China. ³⁸ In addition, the incidence of PIV3 occurred highly in summer and autumn, but in America, the highest prevalent season of PIV3 was in late spring and summer. ⁴⁰ This difference may be due to geographical location and climatic variation. The majority of bacteria did not show a regular seasonal pattern, except Staphylococcus aureus and Haemophilus influenzae, with a peak in winter. We discovered MP was more prevalent in the cold season, a similar result was revealed in America. ⁴¹ The seasonal distribution of these pathogens is an important message to pediatricians, and seasonal variation of the pathogens needed further evaluation.

5. CONCLUSIONS

Viruses, bacteria, and atypical bacteria are important pathogens in LRTI. In this study, we confirmed the high incidence of the virus in children with LRTI and displayed distinct epidemiological patterns. Our findings emphasize the relationship between coinfection and the severity of LRTI and provide a reference for the diagnosis of pathogens in LRTIs, which can assist in the development of LRTI prevention and therapy.

AUTHOR CONTRIBUTIONS

Yuning Sun and Fang Li designed and instructed the study. Yan Yan and Jinhan Sun performed the experiment. Yan Yan analyzed the data. Kai Ji and Jianhui Guo provided laboratory technical support. Lei Han participated in data collection. Yan Yan wrote the first version of the manuscript. Yuning Sun and Fang Li revised the manuscript. All authors reviewed and approved the final manuscript.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

ETHICS STATEMENT

This study was approved by the ethics committee of the medical faculty at Ningxia Medical University. All pediatric patients’ parents and guardians received informed verbal consent in the study.

Yan Y, Sun J, Ji K, et al. High incidence of the virus among respiratory pathogens in children with lower respiratory tract infection in northwestern China. J Med Virol. 2022;95:e28367. 10.1002/jmv.28367

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.