Serotypes and Vaccine Coverage of Streptococcus Pneumoniae Colonization in the Nasopharynx of Thai Children in Congested Areas in Chiang Mai

Streptococcus pneumoniae causes around 10% of all deaths in children younger than five years of age. This study aimed to examine the serogroups/serotypes of S. pneumoniae colonization and vaccine serotype coverage of this organism among Thai children. Nasopharyngeal swabs of children less than or equal to 15 years of age were obtained in congested areas in Chiang Mai from 1 February 2013 to 1 August 2013. The serotyping of S. pneumoniae isolates was performed using the ImmuLex™ kit and the vaccine serotype coverage for this organism was evaluated. A total of 292 children were enrolled. One hundred and thirty children (44.5%) had nasopharyngeal colonization with Streptococcus pneumoniae. Eighty-seven (66.9%) isolates were from children younger than five years of age, seventeen (13.1%) were from children aged 6–10 years, and twenty-six (20%) were from children aged 11–15 years. The five most common serogroups/serotypes isolated were 6 (6A, 6B, 6C) (46.1%), 23 (23F, 23A, 23B) (14.6%), 19 (19F, 19A, 19B, 19C) (8.5%), 15 (15F, 15A, 15B, 15C) (6.9%), and 14 (6.1%). Vaccine serotype coverages in pneumococcal conjugate vaccines (PCV):PCV7, PCV10, and PCV13 were 79.1%, 83.6%, and 85.9%, respectively. There were significant increases in coverage between PCV7 and PCV10 (from 79.1% to 83.6%, p < 0.001), PCV7 and PCV13 (from 79.1% to 85.9%, p < 0.001), and PCV10 and PCV13 (from 83.6% to 85.9%, p < 0.001). The majority of pneumococcal serogroup/serotype colonization in the nasopharynx of Thai children in the studied areas was included in the current licensed pneumococcal conjugated vaccines (PCVs). PCV vaccination should be considered for high-risk children to reduce the incidence of invasive pneumococcal disease among Thai children.


Introduction
Streptococcus pneumoniae is a gram-positive bacterium that can cause bloodstream infections in both immunocompetent and immunocompromised patients. S. pneumoniae bacterial infection complications include meningitis, arthritis, and heart diseases. S. pneumoniae bacteremia is known as invasive pneumococcal disease (IPD) [1].
Before the implementation of the pneumococcal conjugated vaccine (PCV), the incidence of IPD in the United States among children younger than five years of age was approximately 17,000 cases per year. This included 700 cases of meningitis and 200 deaths [1]. When the United States added the PCV to the vaccination program, the incidence of IPD decreased by 60 to 90 percent in children younger than two years of age [2][3][4][5][6][7].
The polysaccharides capsule on the outer surface of pneumococci that protects the bacteria from phagocytosis is the most important virulence factor [8]. The disease invasiveness classification and pneumococcal vaccine formation are based on a high heterogeneity of capsular polysaccharides [9]. To date, more than 90 pneumococcal serotypes have been identified [10], and the prevalence of different serotypes among colonization and diseases has been described [11].
Recently, there are three types of PCV available on the market: PCV7, PCV10, and PCV13. PCV7 was licensed in 2000 and contains seven serotypes which cause IPD in North America (4, 6B, 9V, 14, 18C, 19F, and 23F). In 2008, PCV10 was licensed in Canada, Australia, and European countries. It contains 10 serotypes (1, 5, 7F, and all serotypes in PCV7). PCV13, which contains 13 serotypes (3, 6A, 19A, and all serotypes in PCV10), was licensed in Chile and European countries in 2009. However, since pneumococcal conjugated vaccines were developed, only a few countries have added PCV to their national immunization programs. In October 2006, the World Health Organization (WHO) made a recommendation for the serotype composition of pneumococcal conjugated vaccines to be used in resource-poor developing countries [12]. IPD serotypes vary according to geographic region and year. Thus, regional studies are necessary to assist in the decision of developing a new vaccine [13,14].
Many studies showed that children who lived in childcare centers were nasopharyngeal carriers of pneumococci [34][35][36][37][38][39]. The data from a 1999 surveillance in a Portuguese day care center indicated that the carriage rate of S. pneumoniae has increased continuously since 1996, from 47% to 63% over 4 years [36]. Sá-Leão et al. found in a one-year longitudinal study with 11 samplings of nasopharyngeal carriage of pneumococci among 47 children who attended a single day care center that 61.4% contained pneumococci [38].
The immune system of children, which is not fully developed, and poor hygiene can increase risk of colonization and infection with S. pneumoniae [34][35][36][37][38][39]. According to our knowledge, there are epidemiological data on pneumococcal colonization among children in congested areas in northern Thailand. This study was conducted to examine pneumococcal serotypes among Thai children in congested areas in Chiang Mai city, the capital of the northern region and Thailand's second largest province, using epidemiological data.

Results
From 1 February 2013 to 1 August 2013, 292 children who lived in congested areas were recruited into this study. Of these, 52.1% were males. Most children were in the 11-15 years old age group (42.8%), followed by the less than 5 years old age group and the 6-10 years old age group (39.0% and 18.2%, respectively). The mean age was about 8 years old. The mean weight was approximately 26 kg. Approximately 1.4% of all children had chronic lung disease and congenital heart disease.

Antimicrobial Susceptibility
Antimicrobial susceptibility of S. pneumoniae, H. influenzae, M. catarrhalis, and S. aureus to 13 antimicrobial agents are shown in Table 2. All S. pneumoniae isolates were susceptible to levofloxacin. However, approximately 69.2% and 61.5% of the isolates were susceptible to clindamycin and erythromycin, and only 16.9% and 10.8% were susceptible to penicillin with the oxacillin disk diffusion method and co-trimoxazole. The susceptibility of S. pneumoniae to cefotaxime was not assessed since disk diffusion testing of cephalosporin was found unreliable [35][36][37][38][39][40]. The susceptibility rates of H. influenzae and M. catarrhalis to cefotaxime and ciprofloxacin were 100%. All S. aureus isolates were susceptible to clindamycin, vancomycin, and fusidic acid. Approximately 98.6% of S. aureus isolates were susceptible to oxacillin; thus, 1.4% were identified as methicillin-resistant S. aureus (MRSA).

Serotypes of Nasopharyngeal Pneumococcal Colonization
A total of 130 pneumococcal-colonized children were detected from 292 children in congested areas in Chiang Mai. Eighty-seven (66.9%) S. pneumoniae isolates were from children younger than five years of age, seventeen (13.1%) were from children aged 6-10 years old, and twenty-six (20%) were from children aged 11-15 years.

<5 Years 6-10 Years 11-15 Years Total
Vaccine Types Pathogens 2020, 9, x FOR PEER REVIEW 7 of 13 also poorly immunogenic in children younger than two years of age, who had the highest incidence of invasive pneumococcal disease [51][52][53][54][55]. In this study, the percentage of non-vaccine type S. pneumoniae was low, so the existing vaccine formulation may be proper for children in Chiang Mai in order to prevent and control invasive disease. This study indicated that at least 16.9% of S. pneumoniae isolates were susceptible to penicillin by using a 1-µ g oxacillin disk screening test. Pongsamart et al. [25] found that 69.6% of pneumococcal isolates were resistant to penicillin. There was an increase of penicillin resistance from 58.3% in 2000 to 88.2% in 2005 [25]. Higher rates of penicillin resistance have been described in other countries. This was illustrated in a review of 685 S. pneumoniae isolates from 14 centers in 11 Asian countries that were collected from January 2000 to June 2001 [56]. Overall, 52% were resistant to penicillin, with rates of resistance as high as 74% in Vietnam [31,[57][58][59][60]. In the United States, approximately 85% of pneumococci were susceptible to penicillin, 10% were intermediately resistant, and 5% were highly resistant [61,62]. In the present study, 1-µ g oxacillin disk diffusion was used to test the susceptibility of S. pneumoniae to penicillin; the result "susceptible" was reported when the zone size was equal to  Using a multivariate analysis by logistic regression, the risk factors associated with S. pneumoniae colonization were upper respiratory tract infection one month before collection of specimen (OR = 68.91; 95% CI 6.48-732.87; p = 0.001) and abnormal skin lesions (OR = 4.01; 95% CI 1.51-10.64; p = 0.005) ( Table 4).
Moraxella catarrhalis was the most common bacterium found (58.9%). S. pneumoniae (44.5%) was the second most common organism found. Staphylococcus aureus including MRSA, coagulase-negative Staphylococcus, and non-type B H. influenzae were found in 29.8%, 27.4%, and 16.1% of all isolates, respectively. However, H. influenzae type B was not detected in this study.
The most common serogroups for Streptococcus pneumoniae nasopharyngeal colonization of children under the age of five years who lived in congested areas in Chiang Mai were 6, 23, 19, 15, and 14 (in rank order). In children aged 6-15 years the most prevalent serogroups were 6, 23, 19, and 3.
A recent study of the prevalence of S. pneumoniae among healthy children in Thailand was conducted in Nakhon Phanom province (the northeastern region) and Sa Kaeo province (the eastern region) [41]. This multi-country case-control study under the Pneumonia Etiology Research for Child Health Project (PERCH) reported nasopharyngeal colonization with Streptococcus pneumoniae in children aged 1-59 months (from January 2012-February 2014) of 62.5% in community control and 54.5% in cases (severe/very severe pneumonia cases). However, a lower percentage of colonizing isolates from cases and community control that were the serotypes included in PCV10 (70.0% and 61.8%, respectively) and PCV13 (76.7% and 67.9%, respectively) was observed [42].
The previous work conducted in Thailand reported a lower nasal S. pneumoniae carriage rate of 16% in children aged 2-10 years from four schools in three different districts in Phitsanulok province, the middle part of Thailand [41]. Resistances to clindamycin, erythromycin, and co-trimoxazole were found in 18.4%, 21.1%, and 78.9% of all isolates, respectively. The high rate of resistance to commonly prescribed antibiotics correlated to the results of this study while 30.8%, 38.5%, and 89.2% of all isolates were resistant to clindamycin, erythromycin, and co-trimoxazole, respectively [43,44]. A cross-sectional study in Iceland also reported the correlation of the carriage of drug-resistant pneumococci in children with risk factors such as recent antibiotic use, living in an area with high consumption of antibiotics, and use of co-trimoxazole [45].
A study in Peru revealed high rates of colonizing S. pneumoniae in healthy children, 92% (467/506) in 2009 and 89% (451/509) in 2011 [46]. In 2009, 23F, a serotype included in PCV7, was the only type identified as a persister, and 6A, 15B, and 19A were identified as recolonizer serotypes. In 2011, 6B and 7C were persister serotypes, and 13 was a frequent recolonizer serotype. The prevalence of nasopharyngeal carriage of S. pneumoniae serotypes among children aged 2-59 months in India was studied in Palwal District, Haryana, from December 2016 to July 2017, before the introduction of pneumococcal conjugate vaccines [47]. The colonization rate of S. pneumoniae was 74.7% and 54.5% among children with clinical pneumonia and community children, respectively. The prevalence of PCV13 vaccine-type colonization was similar between children with clinical pneumonia (31.9%) and community children (28.0%; p = 0.46). The most predominant serotypes were 6A, 6B, 14, 19A, 19F, and 23F, all of which are included in the PCV13 vaccine product. Antimicrobial resistance to at least one drug was similar between isolates from children with clinical pneumonia (66.1%) and community children (61.5%; p = 0.49), while resistance to at least two drugs was more common among isolates from children with clinical pneumonia than those from community children (25.8% vs. 16.4%; p = 0.08) [48]. The prevalence and distribution of S. pneumoniae serogroups/serotypes colonization varied widely among distinct geographic locations since it depends heavily on a range of host and environmental factors [48,49].
This study supports the results from previous studies reporting on pneumococcal isolates causing IPD in Thailand; the most common vaccine serotypes in children under the age of five years were 6B, 23F, and 14, and in children aged five years and above, 23F, 19F, and 6B were most common [23][24][25][26][27][28][29][30][31][32][33]. A study from Southeast Asian countries found that the top six (in rank order of frequency) pneumococcal serogroups/serotypes causing IPD were 19, 6, 23, 14, 1, and 3 [23]. This study, however, found that the most predominant serogroup that colonized in the nasopharynx of children was serogroup 6. Serogroup 19 ranked the third highest in the order of detected serogroups. According to a pneumococcal global serotype project, the most common global serotypes (in rank order) are 14, 6B, 1, 23F, 5, 19F, 6A, and 19A [15]. A review of published studies of IPD in North American, European, and Latin American children from 1995-1999 by Hausdorff [50] showed that, across the world, the most commonly found serogroups/serotypes that caused serious diseases and complication resulting in hospitalization were 1, 5, and 7 [50]. However, the present study found small numbers of serogroups 1 and 7, and none of serogroup 5. This different finding might be a result of the virulence of the serotypes. This work studied the serogroups/serotypes of S. pneumoniae that colonized in the nasopharynx, not the serotypes of clinical S. pneumoniae isolates that caused IPD. A review by Jauneikaite et al. [23] found that vaccine coverage of S. pneumoniae in Thailand was 44% for PCV7, 50% for PCV10, and 63% for PCV13 [24]. In this study, the serotype coverages were higher since the serotype coverages from PCV7, PCV10, PCV13, and 23-valent pneumococcal polysaccharide vaccine, were 79.1%, 83.6%, 85.9%, and 92.8%, respectively (Figure 1). The higher coverage rates of vaccine in this study might be due to the high rates of S. pneumoniae colonization in the studied population. In addition, this population was not given the pneumococcal vaccine prior to the study, so the serotypes not included in the vaccine (termed "replacement strains") were low. Although the coverage rate by 23-valent pneumococcal polysaccharide vaccine seemed to be higher than that of the PCV, it could not decrease nasal colonization by S. pneumoniae [40]. Polysaccharide vaccines were also poorly immunogenic in children younger than two years of age, who had the highest incidence of invasive pneumococcal disease [51][52][53][54][55]. In this study, the percentage of non-vaccine type S. pneumoniae was low, so the existing vaccine formulation may be proper for children in Chiang Mai in order to prevent and control invasive disease.
This study indicated that at least 16.9% of S. pneumoniae isolates were susceptible to penicillin by using a 1-µg oxacillin disk screening test. Pongsamart et al. [25] found that 69.6% of pneumococcal isolates were resistant to penicillin. There was an increase of penicillin resistance from 58.3% in 2000 to 88.2% in 2005 [25]. Higher rates of penicillin resistance have been described in other countries. This was illustrated in a review of 685 S. pneumoniae isolates from 14 centers in 11 Asian countries that were collected from January 2000 to June 2001 [56]. Overall, 52% were resistant to penicillin, with rates of resistance as high as 74% in Vietnam [31,[57][58][59][60]. In the United States, approximately 85% of pneumococci were susceptible to penicillin, 10% were intermediately resistant, and 5% were highly resistant [61,62]. In the present study, 1-µg oxacillin disk diffusion was used to test the susceptibility of S. pneumoniae to penicillin; the result "susceptible" was reported when the zone size was equal to or larger than 20 mm [63]. However, the penicillin resistance test was not performed in non-susceptible S. pneumoniae strains that presented zone sizes equal to or less than 19 mm by the minimum inhibitory concentration (MIC) test. Moreover, the resistance to penicillin and cephalosporin using the MIC test was not carried out in penicillin non-susceptible isolates. Hence, MIC tests of penicillin and cephalosporin should be further investigated.
This study is the first study to explore the prevalence and the serogroups/serotypes frequency of S. pneumoniae nasopharyngeal colonization among children who live in congested areas in Chiang Mai. Because nasopharyngeal colonization of S. pneumoniae could increase risk of IPD [64], the results reported here are useful for consideration of currently available vaccines. However, there were still some limitations in this study. Some S. pneumoniae isolates could not be identified at the serotype level due to the limitation in budget and test kit availability. Here, serotyping of S. pneumoniae was performed using the ImmuLex™ test (Statens Serum Institute Diagnostica, Hovedstaden, Denmark). This latex agglutination-based test was shown to identify 47.8% of S. pneumoniae in 67 normally sterile clinical samples (blood, cerebrospinal fluid (CSF), and pleural fluid) only to the level of serogroup (6, 7, 9, 10, 11, 12, 15, 19, 22, 23, or 33) [65]. Although there was no discrepancy between serogroups determined by the reference (PCR) and latex methods, the latex test did not allow the maximum discrimination (to the serotype level) [65]. Nevertheless, the main serotypes or serogroups that are included in the pneumococcal vaccine were identified in all strains isolated from children in this study. Therefore, vaccine serotype coverage reported in the present study might be overestimated. Nevertheless, immunologic cross-reactivity among serotypes in the same serogroup might result in cross-protection, such as 19F and 19A, and 6B and 6A [66][67][68][69][70].

Materials and Methods
This prospective cohort study was performed on children less than or equal to 15 years old in congested areas in Chiang Mai. These children are at risk of infection from S. pneumoniae since they live in congested areas and have recurrent infections.
Congested areas in this study are defined as dormitories and orphanages in Chiang Mai, Thailand. Data were collected from 1 February 2013 to 1 August 2013 (7 months). The data were analyzed for 5 months, and the study was completed in January 2014. Children aged less than or equal to 15 years were recruited in congested areas in Chiang Mai, including one dormitory and three orphanages. This study was approved by the ethics committee on human research of the Faculty of Medicine, Chiang Mai University (Research ID: 1267/Study Code PED-12-1267-EX). The caretakers of the children were informed of study details, and consent was asked from the participants in this study.

Sample Size Calculation
Power and sample size calculated by testing for the proportion of patients in the previous studies showed that the rate of nasopharyngeal carriers for S. pneumoniae was about 21% in developed countries and 90% in developing countries [71]. Since Thailand is a developing country, 90% of incidence was chosen to calculate the sample size.
The sample size was 138 children, so approximately 150 specimens were needed. However, a preliminary study showed that only 50% of nasopharyngeal swabs were positive; this study therefore performed nasopharyngeal swabs on 300 children in order to achieve the needed sample size.

Data Collection
The data of each child were recorded, including age, sex, weight, height, ethnicity, address, immunization history, history of illness in the past one month, underlying disease, and medication.

Specimen Collection and Isolation of S. pneumoniae
A nasopharyngeal swab was obtained, plated directly onto 5% blood agar and chocolate agar, and then incubated at 35-37 • C with 5% CO 2 atmosphere for 24-48 h. The typical colonies of S. pneumoniae with a zone of alpha-hemolysis (green) were selected for biochemical identification.

Bacterial Identification and Characterization
S. pneumoniae isolates were identified using Gram stain, catalase, optochin, and bile solubility tests [42]. Other recovered bacteria were also identified according to standard microbiological tests.
Antimicrobial susceptibility testing (AST) of S. pneumoniae, H. influenzae, M. catarrhalis, and S. aureus was performed using disk diffusion method based on the clinical and laboratory standards institute (CLSI) guideline [72].
The 1-µg oxacillin disk diffusion was used to test susceptibility of S. pneumoniae to penicillin. Serotyping of S. pneumoniae was performed using the ImmuLex™ test as described by the packaging inserts (Statens Serum Institute Diagnostica, Hovedstaden, Denmark).

Statistical Analysis
Statistical analysis was performed using Stata software, version 14 (Stata-Corp, College Station, TX, USA). Descriptive statistics were used to report the results. Mean (±standard deviation, SD) and median (range) were used for continuous data, while counts and percentages were used for nominal data. Continuous variables with normal distribution were compared by student t-test. All tests were two-sided, and statistical significance was set at a p value of ≤0.05. Univariate analysis was performed by logistic regression to compare categorical variables as appropriate. Furthermore, multivariate analysis by logistic regression was used to estimate the risk factors associated with S. pneumoniae colonization. The results of the logistic regression were expressed as the adjusted odds ratio (aOR) and 95% confidence interval (CI).

Conclusions
In conclusion, this study showed that the majority of pneumococcal serogroups/serotypes colonizing the nasopharynx of children who lived in congested areas were included in the current licensed PCV. Giving this vaccine to these high-risk children therefore appears to be useful and should be supported by the government. In the future, the prevalence of pneumococcal serogroups/serotypes of the post-PCV vaccination period should be investigated in order to identify changes in regional serotypes that may occur.