LI Dan, WANG Qian, SU Meng, WANG Ling Ling, ZHANG Qing, WANG Chun Yang,WANG Jiang Li, DU Luan Ying, LIU Jian Ying, and XIE Guang Cheng,#

Community-acquired pneumonia (CAP) is an acute lung infection that is caused by several different pathogens and is associated with significant morbidity and mortality. The high global incidence of CAP poses a heavy disease and economic burden to patients, especially children. Respiratory illnesses such as pneumonia and influenza are the fourth leading cause of death in China[1]. The top 3 etiologic pathogens of CAP in the Asia-Pacific region areStreptococcus pneumoniae,Haemophilus influenzae(H. influenzae), andMycoplasma pneumoniae(M.pneumoniae). A study of CAP-associated pathogens from 2001 to 2013 showedM. pneumoniaeas the most commonly isolated pathogen, especially in children.H. influenzaeinfection was most common(20.6%) between 2001 and 2003, but data onH.influenzaewere missing from 2010 because the focus of CAP etiology had turned toM. pneumoniaeand respiratory viruses[2].

H. influenzaeis a Gram-negative bacterium that can be divided into typeable and non-typeableH.influenzae(NTHi) based on the presence or absence of a polysaccharide capsule. TypeableH. influenzaeare further divided into 6 serotypes (a to f) based on the capsular polysaccharide antigen, withH.influenzaetype b (Hib) representing the most common cause of bacterial pneumonia. While the incidence of Hib has declined dramatically since the widespread introduction of Hib-conjugate vaccines,the incidence of NTHi has increased significantly.Globally, NTHi are recognized as the most common causative pathogens in all ages, and they are confirmed pneumonia pathogens in children[3].Therefore, the aim of this study was to investigate the molecular epidemiology and clinical features of CAP caused byH. influenzaeamong hospitalized children in Chengde, China.

A total of 333 children aged 5 months to 14 years who were hospitalized with CAP in the Department of Pediatrics of a Clinical Teaching Hospital affiliated to Chengde Medical University were enrolled in this study from November 2017 to May 2018. We collected throat swabs from all patients and reviewed medical records for information on age,sex, residence, peak and duration of fever, as well as hospitalization length. Laboratory test data including white blood cell (WBC), neutrophil, and lymphocyte count as well as C-reactive protein (CRP) and lactate dehydrogenase (LDH) concentrations were also recorded. This study was approved by the Ethics Committee of Chengde Medical University with the reference number 2017020 and was performed in accordance with the principles set forth in the Declaration of Helsinki. The collection of samples used in this study was authorized by the guardians of the hospitalized children with prior informed consent.

Bacterial DNA and viral nucleic acids (DNA or RNA) were extracted from throat swabs using QIAamp DNA Mini Kit (QIAGEN, Hilden, Germany) or Viral Nucleic Acid Extraction Kit II (Geneaid, New Taipei City, Taiwan, China) according to manufacturer instructions. ThebexAgene ofH.influenzaewas selected as the target gene and the prevalence ofH. influenzaewas detected using realtime PCR. The forward and reverse primers(TGCGGTAGTGTTAGAAAATGGTATTATG and GGACAAACATCACAAGCGGTTA, respectively) used in this study were described previously[4]. Real-time PCR was prepared using 2 ×TansStart® Top Green qPCR SuperMix (TRANS, Beijing, China) in a final volume of 20 μL containing5 μL of template DNA and conductedviathe MyiQ2 real-time PCR detection system (Bio-Rad, California, USA). Realtime PCR detection was conducted according to the following protocol: initial denaturation at 95 °C for 5 min, followed by 40 cycles of denaturation for 30 seconds at 95 °C and annealing for 30 seconds at 55 °C. In order to assess co-infection, atypical pathogens such asM. pneumoniae,Chlamydophila pneumoniae(C. pneumoniae), andLegionella pneumophila(L. pneumophila); bacterial pathogens including group A Streptococcus (GAS),Klebsiella pneumoniae(K. pneumoniae),Staphylococcus aureus(S. aureus), andPseudomonas aeruginosa(P.aeruginosa); and viral pathogens including Influenza A/B/C viruses, parainfluenza virus (PIV) 1, 2, and 3,Adenovirus (AdV), human bocavirus (HBoV), human rhinovirus (HRV), human metapneumovirus (hMPV),respiratory syncytial virus (RSV), and human coronavirus (HCoV) were also detected. Co-infection was defined as testing positive forH. influenzaeand presenting with at least one bacterial or viral pathogen.

Categorical and continuous variables were described as counts (percentages) or median(interquartile range, IQR). Categorical variables were compared using the Chi-squared or Fisher′s exact test, and continuous variables were compared using the Mann-Whitney U-test. All statistical analyses were performed using SPSS v.19.0 (IBM Corp.,Armonk, NY). Results were considered statistically significant ifP< 0.05.

Among the 333 children hospitalized with CAP,209 were positive forH. influenzae.Of these, 86(41.1%) were singly infected withH. influenzaeand 123 (58.9%) were co-infected (98, 22, and 3 children had double, triple, and quadruple bacterial infections, respectively). A total of 125 children were infected with GAS, 32 of whom were singly infected,and 64 of whom were co-infected with GAS andH.influenzae. Of 66 children infected withK.pneumoniae,only 10 were singly infected, 30 were co-infected withK. pneumoniaeandH. influenzae,and 4 were co-infected withK. pneumoniaeand GAS.Finally, of 20 children infected withS. aureus, only 1 child was singly infected, and 4 were co-infected with bothS. aureusandH. influenzae. Our results show that co-infections—especially double and triple bacterial infections—were common events in hospitalized children with CAP (Table 1).

Table 1. Detection and co-infections of H. influenzae, GAS, K. pneumoniae and S. aureus among hospitalized children with CAP

AlthoughH. influenzaewas detected throughout the study period of November 2017 to May 2018,the months of December 2017 and January 2018 had the highest detection rates, with 57 and 40 children diagnosed, respectively (Figure 1A). The prevalence ofK. pneumoniae, GAS, andS. aureuswas concentrated from November 2017 to January 2018(Figure 1A), and 107 males and 102 females were infected withH. influenzaebetween November 2017 and May 2018 (Figure 1B). These children mainly resided in urban regions (n= 175) (Figure 1C).

We divided the 333 hospitalized children into two groups according to age: < 5 years old (n= 168)and ≥ 5 years old (n= 165).Results showed that 109 children infected withH. influenzaewere aged < 5 years and 100 were ≥ 5 years (P= 0.821). Out of 125 children infected with GAS, 71 were < 5 years old and 54 were ≥ 5 years old, and children < 5 years old were the main target for infection with GAS (P<0.0001). Out of 66 children infected withK.pneumoniae, 32 were < 5 years old and 34 were ≥ 5 years old, with children ≥ 5 years old acting as the main targets forK. pneumoniaeinfection (P= 0.039).There was no difference in the distribution ofS.aureusbetween two the groups (P= 0.400)(Supplementary Table S1 available in www.besjournal.com).

We further divided the 333 children into 2 groups according to disease severity. Overall, there were 293 mild cases and 40 severe cases. Further stratification showed that 229 (78.2%) mild cases and 31 (77.5%) severe cases were positive for at least one of the following pathogens:H. influenzae,GAS,K. pneumoniaeandS. aureus. We found 78 children in the ‘mild case’ group and 8 in the ‘severe case’ group who were singly infected withH.influenzae.Co-infection was also common in mild and severe case groups, with 115 and 16 children affected, respectively (Supplementary Table S2 available in www.besjournal.com).

We compared co-infection withH. influenzaeand atypical pathogens (M. pneumoniae and C.pneumoniae), bacterial pathogens (H. influenzae,GAS,K. pneumoniae,andS. aureus), and viruses(RSV, hMPV, HBoV, HCoV, and AdV) across age groups. Our results show that 15 (13.8%) children < 5 years old and 11 (11.0%) children ≥ 5 years old were singly infected withH. influenzae, 39 (18.7%) were co-infected with atypical pathogens (22 children with a percentage of 20.2 were < 5 years old), 83 (39.7%)were co-infected with at least one bacterial pathogen (44 children with percentage of 44.0 were≥ 5 years old), 16 (7.7%) were co-infected with at least one viral pathogen, and a total of 45 (21.5%)children were co-infected with at least one bacterial and one viral pathogen (26 children with percentage of 23.9 were < 5 years old) (Supplementary Table S3 available in www.besjournal.com).

There were no differences across sex between severe and mild case groups (11vs.96;P= 0.576).Children with severe cases infected withH.influenzaetended to be older than mild cases[median, 7 (IQR, 6–8) yearsvs.4 (IQR, 3–6) years;P<0.0001]. There was no difference in the peak of fever between the two groups [39.5 (IQR, 39.0–40.0) °Cvs.39.3 (IQR, 38.8–39.7) °C;P= 0.159]; however, the duration of fever in the severe case group was longer than that of the mild case group [7 (IQR,4.5–10.0) d;P= 0.029]. There was no difference in the number of days of cough (P= 0.116) and rales(P= 0.912) between the two groups. The patients in the severe case group had a longer duration of hospitalization [9 (IQR, 7.5–12.0) d;P< 0.0001], a higher ratio of neutrophils [65.1 (IQR, 60.4%–70.4%);P= 0.005] and concentration of CRP [2.1 (IQR,1.3–14.8) mg/L;P= 0.041], and lower ratio of lymphocytes [24.0 (IQR, 20.6%–30.3%);P= 0.010]than the mild case group. There was no difference in WBC count [7.6 (IQR, 6.4–8.8) ×109cells/Lvs.7.8(IQR, 5.7–11.5) × 109cells/L;P= 0.360] or LDH concentration [249 (IQR, 218–304) U/Lvs.264 (IQR,228–320) U/L;P= 0.505] between the two groups(Table 2).

In a previous study of 156 children, 107 were positive for at least one pathogen, and 64 (41%)were infected withH. influenzaein Romania from June 2016 to May 2017[5]. Another study of 2,970 patients found 176 cases ofH. influenzaeand 49H.influenzaeco-infections in the German prospective cohort study CAPNETZ from October 2002 to July 2013[6].H. influenzaeis the dominant pathogen in UK adults with non-severe CAP and chronic lung disease[7]. In our study, 209 hospitalized children were infected withH. influenzae, but only 86 (41.1%)of those were singly infected withH. influenzae. We also examined 3 atypical pathogens, 5 bacterial pathogens, and 8 viral pathogens in these specimens, and our results show thatM.pneumoniae(n= 221) andH. influenzaewere the dominant pathogens. Regarding viral pathogens, 80 children were positive for hMPV and 22 children were positive for HBoV[8], with co-infection accounting for a higher proportion among these detected pathogens. This result indicates that coinfection is a common cause of pneumonia.

A total of 8,571 invasiveH. influenzaecases were reported in England and Wales during 2000–2013,but only 1,585 cases were aged between 1 month and 10 years old. Among 362H. influenzaeinvasive isolates during 2009–2013, there were 214 NTHi, 25 Hib, and 52 non-type bH. influenzae(36 Hif, 14 Hie,1 Hia, and 1 Hic) cases[9]. This result indicated that while NTHi was the dominant pathogen, other nontype bH. influenzaewere also present, warranting surveillance in clinical cases. ThebexAgene was used to determine the presence of a polysaccharide capsule, and the type ofH. influenzae(a–f) was confirmed via PCR and slide agglutination test using capsule-specific primers and monovalent a–f antisera, respectively. NTHi should be PCR-negative forbexAand Hib-specific targets without a detectable capsule[9]. The previous study also confirmed that NTHi isolates lacked bothbexAand capsule-specific genes[10]. In our study, we used thebexAgene as a target to screen for the prevalence ofH. influenzaeand confirmed positivity in 209 children. However, one limitation of our study is that we could not determine the serotypes ofH.influenzae. Therefore, further studies are needed to determine the exact bacteriologic characteristics of theseH. influenzaestrains.

Table 2. Clinical features of hospitalized children infected with H. influenzae according to disease severity

Our results showed the prevalence ofH.influenzaeandM. pneumoniaeis higher in hospitalized children with CAP in Chengde between 2017 and 2018, and these results indicate current pneumonia surveillance should focus not only on viral pathogens but should also establish a bacterial pathogen profile. Additionally, strengthening the surveillance of non-type bH. influenzaeand NTHi in CAP would provide important information in controlling and preventingH. influenzae-associated CAP.

The authors declare that they have no competing interests.

&These authors contributed equally to this work.

#Correspondence should be addressed to XIE Guang Cheng, E-mail: xieguangcheng123@126.com; LIU Jian Ying,E-mail: liujianying1160@sina.com

Biographical notes of the first authors: LI Dan, female,born in 1987, Masters, majoring in infectious diseases;WANG Qian, female, born in 1990, Masters, majoring in pediatric infectious diseases.

Received: January 16, 2020;

Accepted: July 7, 2020