Antimicrobial Use and Serotype Distribution of Nasopharyngeal Streptococcus pneumoniae Isolates Recovered from Greek Children Younger than 2 Years Old

Transcription

1 MAJOR ARTICLE Antimicrobial Use and Serotype Distribution of Nasopharyngeal Streptococcus pneumoniae Isolates Recovered from Greek Children Younger than 2 Years Old George A. Syrogiannopoulos, George D. Katopodis, Ioanna N. Grivea, and Nicholas G. Beratis Department of Pediatrics, Division of Infectious Disease, University of Patras, School of Medicine, General University Hospital, Patras, Greece The serotype distribution of 781 nasopharyngeal pneumococcal isolates recovered from 2448 unselected children aged 2 23 months was studied. Only 3.9% of the children for whom cultures were performed attended day care centers. The proportions of pneumococcal isolates that belonged to serotypes related to the 7-, 9- and 11-valent conjugate pneumococcal vaccine were 65%, 66%, and 70%, respectively. The pneumococcal carriage rate among untreated children was 34%; the rates among children treated with antibiotics during the periods 1 30 or days before the time of nasopharyngeal sampling were 25% and 36%, respectively. There was a significant positive association between antimicrobial use and carriage of antibiotic-resistant pneumococci, which belonged mainly to vaccine-related serotypes. The proportion of isolates that belonged to vaccine-related serotypes in untreated carriers was 72%; however, the proportions in carriers treated 1 30 days or days before sampling were 66% and 56%, respectively. In the nasopharynx, antimicrobial use selects for antibiotic-resistant pneumococci, mainly of vaccine-related serotypes, whereas it may promote an increase in the frequency of colonization with nonvaccine serotypes. Streptococcus pneumoniae is a major etiologic agent of childhood bacteremia, meningitis, otitis media, pneumonia, and sinusitis [1, 2]. Children!2 years old experience the greatest burden of pneumococcal infections [1, 3 5]. The nasopharynx is the main reservoir of S. pneumoniae, and carriage usually precedes and, in most cases, parallels pneumococcal infection [1, 6, 7]. The increasing occurrence of infection with antibi- Received 9 January 2002; accepted 18 July 2002; electronically published 21 October Presented in part: 3rd International Symposium on Pneumococci and Pneumococcal Diseases, Anchorage, Alaska, May 2002 (abstract P-03, 52A). Financial support: Wyeth Vaccines. Reprints or correspondence: Dr. George A. Syrogiannopoulos, Dept. of Pediatrics, Div. of Infectious Disease, University of Patras, School of Medicine, Rion, Patras, Greece (syrogian@med.upatras.gr). Clinical Infectious Diseases 2002;35: by the Infectious Diseases Society of America. All rights reserved /2002/ $15.00 otic-resistant pneumococci, a significant percentage of which are multidrug resistant, challenges the outcome of treatment and the prevention measures for pneumococcal diseases [8 13]. The optimal approach to the problem of pneumococcal infections, including those caused by antibiotic-resistant pneumococci, is to prevent them by enhancement of the antipneumococcal immunity. Recently, 7-, 9-, and 11-valent protein conjugate pneumococcal vaccines, which are immunogenic in children!2 years old, have been developed [14 20]. The profile of pneumococci colonizing the nasopharynx of children reflects the strains currently circulating in the community and causing infections, particularly infections of the respiratory system [7, 21 23]. The profile of pneumococcal serotypes carried by children and causing infections has been shown to be associated with age and geographic conditions [24 27]. There is little information in the literature regarding the impact of antibiotic use on the nasopharyngeal 1174 CID 2002:35 (15 November) Syrogiannopoulos et al.

2 carriage of S. pneumoniae in relation to the serotypes of these isolates, especially in children!2 years old who were not attending day care centers. Recent studies have been focused mainly on the carriage rate of antibiotic-resistant pneumococci among children attending day care centers or schools [11, 28 35]. In addition, in most of the observational studies that investigated the impact of antimicrobials on the carriage of antibiotic-resistant S. pneumoniae for a time period of 130 days, this period was evaluated as a single time interval [36]. The development of conjugate pneumococcal vaccines emphasizes the need to extend our knowledge about the carriage of S. pneumoniae serotypes, particularly in children!2 years old. The Hellenic Antibiotic-Resistant Respiratory Pathogens (HARP) Study is a 2-year surveillance of the nasopharyngeal flora of Greek children 2 23 months old. Nasopharyngeal samples were screened for S. pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis [37, 38]. The present study aimed to investigate the serotype distribution of nasopharyngeal S. pneumoniae isolates cultured from children, particularly of isolates that exhibited resistance to antibiotics; the percentage of the identified serotypes that are related to the 7-, 9-, and 11-valent conjugate pneumococcal vaccines; and the influence of recent antimicrobial therapy on the colonization with antibiotic-resistant pneumococci in relationship to the serotypes of these isolates. SUBJECTS, MATERIALS, AND METHODS Study population. Between 10 February 1997 and 10 February 1999, nasopharyngeal samples for culture were obtained from 2448 children, 2 23 months old, living in different areas of central and southern Greece located between latitudes 36 N and 40 N. None of the enrolled children had received any pneumococcal vaccines. Samples of nasopharyngeal secretions for culture were obtained from all the children who came to the outpatient department of the University Hospital of Patras (Patras, Greece). In addition, nasopharyngeal samples were obtained from children enrolled randomly from the outpatient clinics of 3 smaller provincial hospitals, as well as from the private offices of 14 practicing pediatricians who participated in the study. One culture was performed for each child who participated in the study. The research protocol was approved by the Ethics Committee of the General University Hospital of Patras. Informed consent was obtained from one of the parents of each participant. Parents responded to an interviewer-administered questionnaire that collected data on demographic characteristics, day care attendance, and the number and type of courses of antimicrobial therapy administered during the previous 90 days. The questions regarding antimicrobial drug use were focused on the time periods 1 30 days, days, and days before the time that nasopharyngeal samples were obtained. In addition to the information recalled by parents, the interviewing pediatrician collected all available information about each child s medical history, including prescriptions recorded in the child s health booklet, which is kept by all children in Greece. For children enrolled in the study by private pediatricians, the information was validated by reviewing the pediatricians own records. Responses were recorded on structured case report forms. The findings of the physical examination at the time of enrollment, with detailed descriptions of the findings regarding the respiratory tract, were recorded on the same forms. Urban areas were classified as cities and towns with 15,000 inhabitants. Laboratory procedures. Specimens of nasopharyngeal secretions were obtained pernasally with sterile swabs on flexible shafts with calcium alginate fiber tips (Fisher Scientific). Swabs were placed in Amies transport medium (Sanofi Diagnostic Pasteur) after sampling and were sent to the Laboratory of the Division of Pediatric Infectious Disease of the University of Patras, where isolation, identification, and susceptibility testing of the S. pneumoniae isolates were performed. The maximum delay between collection and cultivation was 72 h. The swabs were plated on Columbia agar plates supplemented with 5% defibrinated horse blood, 10 mg colistin sulfate, and 15 mgof nalidixic acid per milliliter. The plates were incubated at 35 C in an atmosphere supplemented with 5% CO 2 for h. Isolates were confirmed to be S. pneumoniae by inhibition testing (optochin; Difco) and by bile solubility testing [39]. Susceptibility testing was performed on Mueller-Hinton agar supplemented with 5% defibrinated horse blood. Susceptibility to penicillin and erythromycin was determined by the Etest (AB Biodisk) method, and susceptibility to clindamycin, chloramphenicol, tetracycline, and trimethoprim-sulfamethoxazole (TMP-SMZ) was determined by the disk diffusion method, as described elsewhere [37]. Capsule serotyping. All the antibiotic-resistant pneumococci recovered between 10 February 1997 and 31 May 1998, a total of 177 isolates, were serotyped at the Centre National de Référence du Pneumocoque, Créteil, France, by latex agglutination using specific antisera prepared by the investigators of the Centre. A total of 604 isolates were serotyped at our laboratory in Patras by the capsular swelling method with use of pooled antisera (Pneumotest and Pool sera G and I). The isolates of serogroups 6, 19, and 23 were subtyped with use of pneumococcal factor antisera. The pooled and factor antisera were purchased from Statens Serum Institut (SSI). All isolates belonging to other serogroups were sent to SSI for serotyping within the groups. In addition, all the pneumococci that did not react with the available antisera were confirmed by SSI as nontypeable. In a validation study of pneumococcal typing, the degree of agreement between the findings of our laboratory and Antimicrobial Use and Pneumococcal Serotypes CID 2002:35 (15 November) 1175

3 those of SSI was 93%, which is similar to a 95% mean agreement observed between the findings of 10 other reference laboratories and SSI [40]. Statistical analysis. The data were analyzed by the x 2 test with Yates correction and by Fisher s exact test, as well as by means of 95% CIs for the OR. The analysis of antibiotic use was based on courses of treatment administered within the periods 1 30 days and days before the time of sampling. The percentage of pneumococcal isolates belonging to vaccinerelated serotypes that were recovered from untreated and treated carriers according to the semester of the study was analyzed by logistic regression analysis. The effect of treatment group, time interval, and their interaction was examined by testing the changes in deviance between different models against the x 2 distribution (SPSS software, release 11.0). Two-tailed P.05 was considered to be significant. All percentages 110% were rounded to the closest integral number. RESULTS Population and samples. Of the 2448 children enrolled in the study, 95 (3.9%) attended a day care center, 474 (19%) had 2 siblings, and 77 (3.1%) had 2 siblings!5 years old. Of these 2448 children, 766 (31%) were colonized with S. pneumoniae. A single pneumococcal isolate was recovered from each carrier, with the exception of 15 children (2%) from whom 2 different pneumococci were identified. The 766 carriers had the following epidemiologic characteristics: 362 (47%) were 112 months old, 426 (56%) were boys, 455 (59%) were enrolled between November and April, and 431 (56%) were living in an urban area. At the time of sampling, 50% of carriers 12 months old (201 of 404) and 55% of carriers 112 months old (198 of 362) had signs and symptoms of a respiratory tract infection. Serotype distribution. Of the 781 nasopharyngeal isolates, 738 belonged to 42 different serotypes, and 43 (5.5%) were nontypeable (table 1). The 6 most frequently isolated serotypes 6B, 23F, 19F, 6A, 14, and 11A accounted for 57% of the isolates. The proportions of the recovered serotypes related to that is, included in or partially covered by the 7- and 9- valent conjugate pneumococcal vaccine were 65% and 66%, respectively. For the 11-valent vaccine, this proportion increased to 70%. Of the 781 isolates, 404 (52%) and 377 (48%) were recovered from carriers with and without a respiratory tract infection, respectively. Two hundred eighty-two (70%) of the 404 recovered from carriers with a respiratory tract infection and 260 (69%) of the 377 isolates recovered from carriers without a respiratory tract infection belonged to vaccine-related serotypes (OR, 0.96; 95% CI, ; P p.860). Antimicrobial resistance. Resistance (including both intermediate and full resistance) to 1 antimicrobial agent was found in 239 (31%) of the 781 S. pneumoniae isolates. One hundred twenty-two (16%) of the isolates demonstrated decreased susceptibility to penicillin. Of the 122 penicillin-nonsusceptible isolates, 50 (41%) had MICs of penicillin of 2 mg/ ml. An additional 117 (15%) of the 781 S. pneumoniae isolates were penicillin susceptible and resistant to 1 non b-lactam agent. Of the 781 isolates, 137 (18%) were resistant to erythromycin, 96 (12%) were resistant to clindamycin, 136 (17%) were resistant to tetracycline, 86 (11%) were resistant to chloramphenicol, and 151 (19%) were resistant to TMP-SMZ. Resistance to 3 antibiotics was found in 129 (17%) of the 781 pneumococci. Serotype distribution of antibiotic-resistant isolates. Pneumococci belonging to vaccine-related serotypes constituted most of the antibiotic-resistant isolates (table 2). In addition, 82% of the 129 isolates resistant to 3 antibiotics belonged to vaccine-related serotypes. Rates of carriage of antibiotic-resistant pneumococci and the serotype distribution of antibiotic-resistant isolates did not vary significantly by age group. Nasopharyngeal carriage in relation to history of antimicrobial use. A history of 1 course of orally administered antibiotic therapy during the previous 2 months was reported for 941 children. The antibiotics administered were known for 907 (96%) of those 941 children. The antibiotics used were a penicillin in 45% of the courses, a cephalosporin in 32% of the courses, a macrolide in 18% of the courses, and TMP-SMZ in 5% of the courses. Children treated with multiple courses of antibiotics during the previous 60 days were included in the analysis only if all courses had been administered during the periods 1 30 or days before the time of sampling. Of the 2448 children enrolled in the HARP Study, 1507 had not received any antibiotic therapy during the 60-day period before the time of sampling. Of these 1507 children, 509 (34%) carried S. pneumoniae. Of the children who had received antibiotic therapy only during the periods 1 30 or days before the time of nasopharygeal sampling, 25% (159 of 632) and 36% (73 of 205), respectively, were colonized with pneumococci. The rate of pneumococcal carriage among children treated during the previous 30 days was significantly lower than that among untreated children (OR, 1.52; 95% CI, ; P p.0001) and that among those treated with antibiotics days before the time of sampling (OR, 0.61; 95% CI, ; P p.005). Antibiotic treatment had been administered during the previous 60 days to 32% of carriers 12 months old (128 of 404) and 36% of carriers 112 months old (129 of 362). Antibiotic use and serotype distribution of pneumococcal isolates. Table 3 shows the proportions of antibiotic-resistant S. pneumoniae isolates belonging to vaccine-related and nonvaccine serotypes, according to whether or not oral antibiotics had been administered before the time nasopharygeal samples 1176 CID 2002:35 (15 November) Syrogiannopoulos et al.

4 Table 1. Distribution of 781 nasopharyngeal Streptococcus pneumoniae isolates according to antimicrobial susceptibility, serotype, and relation of the serotype to protein conjugate vaccines. Serotype, by vaccine relation No. (%) All isolates (n p 781) Cumulative percentage No. (%) Isolates resistant to 1 antibiotic (n p 239) Cumulative percentage 7-Valent Included 6B 127 (16) (20) 20 23F 97 (12) (26) 46 19F 91 (12) (12) (5.9) (8.8) 67 18C 19 (2.4) 48 2 (0.8) 68 9V 13 (1.7) 50 6 (2.5) Partially covered 6A 54 (6.9) 57 5 (2.1) 72 19A 24 (3.1) (5.0) 77 23A/B 20/5 (3.2) 63 3/0 (1.3) 78 9N 7 (0.9) A/B 1/3 (0.5) Valent a (0.4) 66 2 (0.8) Valent b 3 17 (2.2) F/C 14/1 (1.9) Nonvaccine 11A 33 (4.2) 74 9 (3.8) 83 15A/B/C 16/20/11 (6.0) 80 1/4/3 (3.3) 87 10A 16 (2) 82 2 (0.8) 88 33F 10 (1.3) 83 1 (0.4) 88 24F 9 (1.2) 84 1 (0.4) 88 8/20/21/22F 4/5/7/7 (2.9) 87 1/1/1/1 (1.7) 90 Other c 58 (7.4) Nontypeable 43 (5.5) (10) 100 Total a In addition to the serotypes in the 7-valent vaccine and partially covered. b In addition to the serotypes in the 9-valent vaccine and partially covered. c Serotype (no. of isolates): 34 (18), 13 (7), 16F (6), 35F (6), 38 (5), 31 (4), 35B (3), 17F (2), 37 (2), 10F (1), 12F (1), 15F (1), 35A (1), and 35C (1). were obtained. A significant association was observed between antimicrobial use during the periods 1 30 or days before the time of nasopharygeal sampling and carriage of antibioticresistant pneumococci, which belonged mainly to vaccinerelated serotypes ( P.01). Conversely, receipt of antibiotics was associated with a decrease in the frequency of carriage of isolates belonging to vaccine-related serotypes. The proportion of isolates belonging to vaccine-related serotypes recovered from untreated carriers was 72% (374 of 522 isolates); the proportions recovered from carriers treated during the periods 1 30 or days before the time of nasopharygeal sampling were 66% (107 of 161 isolates) and 56% (41 of 73 isolates) (table 4). The possibility that this decrease in the frequency of carriage of vaccine-related serotypes in treated children was only a seasonal effect was evaluated for 6-month intervals during the 2-year study period (figure 1). It was found that the Antimicrobial Use and Pneumococcal Serotypes CID 2002:35 (15 November) 1177

5 Table 2. Proportion of the 239 antibiotic-resistant nasopharyngeal Streptococcus pneumoniae isolates that belonged to vaccine-related serotypes. Agent to which isolates were resistant No. of vaccine-related resistant isolates/total no. of resistant isolates (%) Penicillin 99/122 (81) Erythromycin 114/137 (83) Clindamycin 77/96 (80) Tetracycline 110/136 (81) Chloramphenicol 84/86 (98) TMP-SMZ 126/151 (83) NOTE. Vaccine-related serotypes are those included or partially covered by the 7-, 9-, and 11-valent vaccines. Antibiotic-resistant serotypes include both intermediately resistant and resistant isolates. TMP-SMZ, trimethoprimsulfamethoxazole. percentage of isolates belonging to vaccine-related serotypes depended only on the treatment group and not on the time of year. DISCUSSION The HARP Study is, to our knowledge, the largest investigation of nasopharyngeal pneumococcal carriage among children!2 years old who were not recruited among day care center attendees. Our study population differs from those of other studies because the children studied in this investigation had less exposure to other children [11, 28, 32 35]. Of the children we studied, only 3.9% attended day care centers, and 19% had 2 siblings. Greece is a European country with a temperate Mediterranean climate, without excessively humid or cold weather. In this study, 781 pneumococcal isolates belonging to 42 serotypes were obtained. The proportions of the recovered serotypes related to that is, included in or partially covered by the 7- and 9-valent conjugate pneumococcal vaccine were 65% and 66%, respectively; the proportion was higher (70%) for the 11-valent vaccine. Serotypes that were not included in the vaccines but that belonged to the same serogroup as a serotype included in the vaccine were considered to be partially covered. In the present study, most of the antibiotic-resistant pneumococci belonged to vaccine-related serotypes. The pneumococcal serogroups or serotypes most frequently isolated from young children in industrialized countries are 6, 14, 19, and 23, for both nasopharyngeal and invasive isolates [15, 41]. These serotypes were among the most common in our study as well, but serotype 11A was also commonly isolated. Serogroups 7 and 18, which cause serious diseases but are less frequently recovered from carriers [11, 14, 35], were relatively rare among our isolates. In addition, serotype 3 was uncommon in the present series. This serotype is identified in nasopharyngeal, middle-ear-fluid, and invasive isolates recovered more frequently from older children and adults than from infants and young children [15, 35, 41, 42]. The fact that this observational, community-based study included a large number of carriers enrolled over a 2-year period provided the opportunity to evaluate the impact of antibiotic use on the prevalence of nasopharyngeal carriage of S. pneumoniae in relation to the serotypes of the isolates recovered. Unlike earlier observational studies [36], our study focused on specific 30-day intervals and investigated carriers who received courses of antibiotics only within these restricted time periods. This helped us avoid the effect of overlapping periods of treat- Table 3. Serotypes of antibiotic-resistant (AR) pneumococci colonizing untreated children and children who were treated with antibiotics during the time periods 1 30 or days before the time nasopharygeal samples were obtained. Class of serotypes, by vaccine relation No. of AR isolates/total no. of isolates (%), according to antibiotic treatment received before sampling Treatment only during the period 1 30 days before Treatment only during the period days before Statistical comparison, by treatment received 1 30 Days before vs. none Days before vs. none No treatment during the 60 days before OR (95% CI) P OR (95% CI) P 7-Valent a Included 47/80 (59) 17/27 (63) 95/269 (35) 2.61 ( ) ( ).009 Included and partially covered 52/102 (51) 20/38 (53) 107/347 (31) 2.33 ( ) ( ) Valent a Included 48/81 (59) 18/28 (64) 95/270 (35) 2.68 ( ) ( ).005 Included and partially covered 53/103 (51) 21/39 (54) 107/348 (31) 2.39 ( ) ( ) Valent a Included 48/85 (56) 18/30 (60) 95/295 (32) 2.73 ( )! ( ).004 Included and partially covered 53/107 (50) 21/41 (51) 107/374 (29) 2.45 ( )! ( ).005 Nonvaccine 17/54 (31) 7/32 (22) 24/148 (16) 2.37 ( ) ( ).61 a Serotypes are listed in table CID 2002:35 (15 November) Syrogiannopoulos et al.

6 Table 4. Serotypes of Streptococcus pneumoniae isolates in relation to antimicrobial treatment received before nasopharygeal samples were obtained. Serotype, by vaccine relation Vaccine related a No. (%) of isolates recovered, according to antimicrobial treatment received before sampling No treatment during the 60 days before Treatment only during the period 1 30 days before Treatment only during the period days before 6B 90 (17) 27 (17) 8 (11) 19F 67 (13) 15 (9.3) 3 (4.1) 23F 59 (11) 20 (12) 11 (15) (5.7) 12 (7.5) 2 (2.7) 18C 15 (2.9) 3 (1.9) 1 (1.4) 9V 8 (1.6) 3 (1.9) 2 (2.7) 6A 34 (6.5) 12 (7.5) 6 (8.2) 9N 5 (1.0) 1 (0.6) 0 18A/B 1/2 (0.6) 0/1 (0.6) 0 19A 16 (3.1) 5 (3.1) 3 (4.1) 23A/B 17/3 (3.8) 1/2 (1.9) 2/0 (2.7) 1 1 (0.2) 1 (0.6) 1 (1.4) 3 13 (2.5) 3 (1.9) 1 (1.4) 7F/C 12/1 (2.5) 1/0 (0.6) 1/0 (1.4) Subtotal 374 (72) b 107 (66) 41 (56) b Nonvaccine 11A 21 (4.0) 6 (3.7) 6 (8.2) (2.5) 3 (1.9) 2 (2.7) 15A 13 (2.5) 1 (0.6) 1 (1.4) 15B 14 (2.7) 5 (3.1) 1 (1.4) 10A 8 (1.5) 3 (1.9) 5 (6.8) 15C 6 (1.1) 4 (2.5) 1 (1.4) 33F 5 (1.0) 3 (1.9) 2 (2.7) 24F 5 (1.0) 2 (1.3) 2 (2.7) Other 41 (7.9) 11 (6.8) 7 (9.6) Nontypeable 22 (4.2) c 16 (9.9) c 5 (6.8) Subtotal 148 (28) 54 (34) 32 (44) Total a Serotypes included in or partially covered by the 3 conjugate vaccines are listed in table 1. b OR, 1.97; 95% CI, ; P p.01. c OR, 0.40; 95% CI, ; P p.01. ment on the particular time intervals investigated. Reported interventional studies, although adherent to specific time intervals, frequently have a short follow-up period or provide limited information about the serotypes of isolates recovered after the end of treatment [43 46]. We found a significant reduction in the rate of carriage of pneumococci among children treated with antibiotics within the 30 days before the time nasopharyngeal samples were obtained (25%), compared with the rate among untreated children (34%) and children treated with antimicrobials days before sampling (36%). This observation is in accordance with recent reports indicating that antimicrobial treatment reduces the carriage rate temporarily [7, 47]. There was a strong positive association between antimicrobial use during the 30 days before sampling and carriage of antibiotic-resistant pneumococci, particularly carriage of isolates belonging to serotypes related to those in the 3 conjugate vaccines. It is notable that the selective effect of antimicrobials on colonization with antibiotic-resistant pneumococci of vaccinerelated serotypes remained significant even if the samples were Antimicrobial Use and Pneumococcal Serotypes CID 2002:35 (15 November) 1179

7 Figure 1. Percentage of pneumococcal isolates belonging to vaccinerelated serotypes among carriers who had received no antibiotic treatment (Rx) and carriers who received antibiotics during the periods 1 30 or days before the time nasopharygeal samples were obtained, according to 6-month periods of the study. The effect of the treatment 2 group was statistically significant ( x p 7.51; P p.023); the effect of 2 the time period was not significant ( x p 2.23; P p.53). obtained days after treatment. Some courses of antibiotics administered days before sampling might have not been recorded because of the time lapsed since treatment. However, such underrecording should not affect the trend of the findings, because, if treated children were misclassified with the untreated ones because of recall bias, this would have resulted in an underestimation of the demonstrated effect of antibiotic treatment on an increased carriage rate of antibiotic-resistant pneumococci. Thus, such a misclassification would reduce the level of statistical significance without altering the trend of the association observed in this study [48]. The findings indicate that antimicrobial use selected for antibiotic-resistant pneumococci that mainly belonged to vaccinerelated serotypes whereas it appeared to promote a decrease in the frequency of colonization with vaccine-related serotypes. The proportion of isolates belonging to vaccine-related serotypes among untreated carriers was 72%; the proportions among carriers treated during the periods 1 30 or days before the time of sampling were 66% and 56%, respectively. This finding cannot be explained with certainty at the present time. The possibility of a seasonal effect was excluded because data from the 2-year study period was evaluated according to 6-month intervals. We hypothesize that treatment with antimicrobials temporarily reduces pneumococcal carriage, which is followed by new colonization with isolates that are more likely to belong to nonvaccine serotypes. This result could be explained by the development of a partial immunologic response, due to previous carriage, against pneumococci belonging to the most prevalent vaccine-related serotypes. The presence of secretory IgA might inhibit recolonization with the same serotypes. It has been reported that both nasopharyngeal pneumococcal carriage and otitis media induce local immune response, accompanied by production of IgA [49 51]. Mucosal immunity to bacterial polysaccharides matures earlier in life than systemic immunity and has been observed in children aged 6 months [51]. A factor contributing to the preferential recolonization with nonvaccine serotypes could be the fact that the children in this study were not exposed to a day care center environment. A second hypothesis is that use of antibiotics unmasked pneumococci of nonvaccine serotypes in the nasopharynx that were present before the administration of therapy and that would not otherwise be detected because of the abundant presence of other isolates belonging to vaccine-related serotypes. Clearly, further studies are warranted to investigate this issue. Leach et al. [47] found that the rate of pneumococcal carriage was lower (29%) 2 3 weeks after administration of a single dose of azithromycin than it was before treatment (68%); at the 2-month follow-up visit, the carriage rate had rebounded to an even higher level (78%) than had been observed before treatment. In the same study, it was reported that the profile of the pneumococcal serotypes isolated changed after treatment. Most of the children were recolonized with pneumococci belonging to serotypes that differed from those initially carried, and the most prevalent serotypes in the community before treatment were not isolated at the 2-month follow-up visit. Our findings indicate that antimicrobial use selects for antibiotic-resistant pneumococci in the nasopharynx that mainly belong to vaccine-related serotypes and that persist for up to 2 months, whereas antimicrobial use may promote an increase in the frequency of colonization with nonvaccine serotypes. Immunization and judicious use of antibiotics are necessary for young children to reduce the rate of colonization with antibiotic-resistant pneumococci in this population and the subsequent spread of such strains to the community. References 1. Klein JO. The epidemiology of pneumococcal disease in infants and children. Rev Infect Dis 1981; 3: Musher DM. Streptococcus pneumoniae. In: Mandell GL, Bennett JE, Dolin R, eds. Principles and practice of infectious diseases. 5th ed. New York: Churchill Livingstone, 2000: Gray BM, Converse GM 3rd, Dillon HC Jr. Epidemiologic studies of Streptococcus pneumoniae in infants: acquisition, carriage, and infection during the first 24 months of life. J Infect Dis 1980; 142: Eskola J, Takala AK, Kela E, Pekkanen E, Kalliokoski R, Leinonen M. Epidemiology of invasive pneumococcal infections in children in Finland. JAMA 1992; 268: Sleeman K, Knox K, George R, et al. Invasive pneumococcal disease in England and Wales: vaccination implications. J Infect Dis 2001; 183: CID 2002:35 (15 November) Syrogiannopoulos et al.

8 6. Faden H, Duffy L, Wasielewski R, et al. Relationship between nasopharyngeal colonization and the development of otitis media in children. J Infect Dis 1997; 175: Syrjänen RK, Kilpi TM, Kaijalainen TH, Herva EE, Takala AK. Nasopharyngeal carriage of Streptococcus pneumoniae in Finnish children younger than 2 years old. J Infect Dis 2001; 184: Radetsky MS, Istre GR, Johansen TL, et al. Multiply resistant pneumococcus causing meningitis: its epidemiology within a day-care centre. Lancet 1981; 2: Duchin JS, Breiman RF, Diamond A, et al. High prevalence of multidrug-resistant Streptococcus pneumoniae among children in a rural Kentucky community. Pediatr Infect Dis J 1995; 14: Syrogiannopoulos GA, Grivea IN, Beratis NG, et al. Resistance patterns of Streptococcus pneumoniae from carriers attending day-care centers in Southwestern Greece. Clin Infect Dis 1997; 25: Yagupsky P, Porat N, Fraser D, et al. Acquisition, carriage, and transmission of pneumococci with decreased antibiotic susceptibility in young children attending a day care facility in Southern Israel. J Infect Dis 1998; 177: Doern GV, Pfaller MA, Kugler K, Freeman J, Jones RN. Prevalence of antimicrobial resistance among respiratory tract isolates of Streptococcus pneumoniae in North America: 1997 results from the SENTRY antimicrobial surveillance program. Clin Infect Dis 1998; 27: Ball P, Baquero F, Cars O, et al. Antibiotic therapy of community respiratory tract infections: strategies for optimal outcomes and minimized resistance emergence. J Antimicrob Chemother 2002; 49: Sniakack DH, Schwartz B, Lipman H, et al. Potential interventions for the prevention of childhood pneumonia: geographic and temporal differences in serotype and serogroup distribution for sterile site pneumococcal isolates from children implications for vaccine strategies. Pediatr Infect Dis J 1995; 14: Butler JC, Breiman RF, Lipman HB, Hofmann J, Facklam RR. Serotype distribution of Streptococcus pneumoniae infections among preschool children in the United States, : implications for development of a conjugate vaccine. J Infect Dis 1995; 171: Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunogenecity of heptavalent pneumococcal conjugate vaccine in children. Pediatr Infect Dis J 2000; 19: Dagan R, Sikuler-Cohen M, Zamir O, Janco J, Givon-Lavi N, Fraser D. Effect of a conjugate pneumococcal vaccine on the occurrence of respiratory infections and antibiotic use in day-care center attendees. Pediatr Infect Dis J 2001; 20: Wuorimaa T, Dagan R, Eskola J, et al. Tolerability and immunogenicity of an eleven-valent pneumococcal conjugate vaccine in healthy toddlers. Pediatr Infect Dis J 2001; 20: Eskola J, Kilpi T, Palmu A, et al. Efficacy of a pneumococcal conjugate vaccine against acute otitis media. N Engl J Med 2001; 344: Whitney CG, Farley MM, Hadler J, et al. Decline in invasive pneumococcal disease in the US in 2000: an effect of pneumococcal conjugate vaccine? [abstract G-2041]. In: Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy (Chicago). Washington, DC: American Society for Microbiology, 2001: Klugman KP, Koorhof HJ, Kuhnle V. Clinical and nasopharyngeal isolates of unusual multiply resistant pneumococci. Am J Dis Child 1986; 140: Klugman KP, Friedland IR. Antibiotic-resistant pneumococci in pediatric disease. Microbial Drug Resist 1995; 1: Cohen R, Bingen E, Varon E, et al. Change in nasopharyngeal carriage of Streptococcus pneumoniae resulting from antibiotic therapy for acute otitis media in children. Pediatr Infect Dis J 1997; 16: Dagan R, Melamed R, Muallem M, Piglansky L, Yagupsky P. Nasopharyngeal colonization in southern Israel with antibiotic-resistant pneumococci during the first 2 years of life: relation to serotypes likely to be included in pneumococcal conjugate vaccines. J Infect Dis 1996; 174: Scott JAG, Hall AJ, Dagan R, et al. Serogroup-specific epidemiology of Streptococcus pneumoniae: associations with age, sex, and geography in 7,000 episodes of invasive disease. Clin Infect Dis 1996; 22: Hausdorff WP, Bryant J, Paradiso PR, Siber GR. Which pneumococcal serogroups cause the most invasive disease: implications for conjugate vaccine formulation and use, part I. Clin Infect Dis 2000; 30: Hausdorff WP, Bryant J, Kloek C, Paradiso PR, Siber GR. The contribution of specific pneumococcal serogroups to different disease manifestations: implications for conjugate vaccine formulation and use, part II. Clin Infect Dis 2000; 30: Reichler MR, Allphin AA, Breiman RF, et al. The spread of multiply resistant Streptococcus pneumoniae at a day-care center in Ohio. J Infect Dis 1992; 166: Ussery XT, Gessner BD, Lipman H, et al. Risk factors for nasopharyngeal carriage of resistant Streptococcus pneumoniae and detection of a multiply resistant clone among children living in the Yukon-Kuskokwim Delta region of Alaska. Pediatr Infect Dis J 1996; 15: Arnold KE, Leggiadro RJ, Breiman RF, et al. Risk factors for carriage of drug-resistant Streptococcus pneumoniae among children in Memphis, Tennessee. J Pediatr 1996; 128: Boken DJ, Chartrand SA, Moland ES, Goering RV. Colonization with penicillin-nonsusceptible Streptococcus pneumoniae in urban and rural child-care centers. Pediatr Infect Dis 1996; 15: Arason VA, Kristinsson KG, Sigurdsson JA, Stefansdottir G, Molstad S, Gudmundsson S. Do antimicrobials increase the carriage rate of penicillin resistant pneumococci in children? Cross sectional prevalence study. BMJ 1996; 313: Melander E, Mölstad S, Persson K, Hansson HB, Söderström M, Ekdahl K. Previous antibiotic consumption and other risk factors for carriage of penicillin-resistant Streptococcus pneumoniae in children. Eur J Clin Microbiol Infect Dis 1998; 17: Guillemot D, Carbon C, Balkau B, et al. Low dosage and long treatment duration of b-lactam: risk factors for carriage of penicillin-resistant Streptococcus pneumoniae. JAMA 1998; 279: Kellner JD, Ford-Jones EL, Members of the Toronto Child Care Study Group. Streptococcus pneumoniae carriage in children attending 59 Canadian child care centers. Arch Pediatr Adolesc Med 1999; 153: Lipsitch M. Measuring and interpreting associations between antibiotic use and penicillin resistance in Streptococcus pneumoniae. Clin Infect Dis 2001; 32: Syrogiannopoulos GA, Grivea IN, Davies TA, Katopodis GD, Appelbaum PC, Beratis NG. Antimicrobial use and colonization with erythromycin-resistant Streptococcus pneumoniae in Greece during the first 2 years of life. Clin Infect Dis 2000; 31: Stutzmann Meier P, Troller R, Grivea IN, Syrogiannopoulos GA, Aebi C. The outer membrane proteins UspA1 and UspA2 of Moraxella catarrhalis are highly conserved in nasopharyngeal isolates from young children. Vaccine 2002; 20: Ruoff KL, Whiley RA, Beighton D. Streptococcus. In: Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken PH, eds. Manual of clinical microbiology. 7th ed. Washington, DC: American Society for Microbiology, 1999: Konradsen HB. Validation of pneumococcal typing in Europe [abstract P-01, 39A]. In: Program and abstracts of the 3rd International Symposium on Pneumococci and Pneumococcal Diseases (Anchorage). Washington, DC: American Society for Microbiology, 2002: Kellner JD, McGeer A, Cetron MS, et al. The use of Streptococcus pneumoniae nasopharyngeal isolates from healthy children to predict features of invasive disease. Pediatr Infect Dis J 1998; 17: Joloba ML, Windau A, Bajaksouzian S, Appelbaum PC, Hausdorff WP, Jacobs MR. Pneumococcal conjugate vaccine serotypes of Streptococcus pneumoniae isolates and the antimicrobial susceptibility of such isolates in children with otitis media. Clin Infect Dis 2001; 33: Dagan R, Leibovitz E, Greenberg D, Yagupsky P, Fliss DM, Leiberman A. Dynamics of pneumococcal nasopharyngeal colonization during the Antimicrobial Use and Pneumococcal Serotypes CID 2002:35 (15 November) 1181

9 first days of antibiotic treatment in pediatric patients. Pediatr Infect Dis J 1998; 17: Dabernat H, Geslin P, Megraud F, et al. Effects of cefixime or coamoxiclav treatment on nasopharyngeal carriage of Streptococcus pneumoniae and Haemophilus influenzae in children with acute otitis media. J Antimicrob Chemother 1998; 41: Cohen R, Navel M, Grunberg J, et al. One dose ceftriaxone vs. ten days of amoxicillin/clavulanate therapy for acute otitis media: clinical efficacy and change in nasopharyngeal flora. Pediatr Infect Dis J 1999; 18: Ghaffar F, Muniz LS, Katz K, et al. Effects of amoxicillin/clavulanate or azithromycin on nasopharyngeal carriage of Streptococcus pneumoniae and Haemophilus influenzae in children with acute otitis media. Clin Infect Dis 2000; 31: Leach AJ, Shelby-James TM, Mayo M, et al. A prospective study of the impact of community-based azithromycin treatment of trachoma on carriage and resistance of Streptococcus pneumoniae. Clin Infect Dis 1997; 24: Rothman K. Modern epidemiology. Boston: Little Brown, Virolainen A, Jero J, Käyhty H, Karma P, Leinonen M, Eskola J. Nasopharyngeal antibodies to pneumococcal capsular polysaccharides in children with acute otitis media. J Infect Dis 1995; 172: Nieminen T, Virolainen A, Käyhty H, et al. Antibody-secreting cells and their relation to humoral antibodies in serum and in nasopharyngeal aspirates in children with pneumococcal acute otitis media. J Infect Dis 1996; 173: Simell B, Korkeila M, Pursiainen H, Kilpi TM, Käyhty H. Pneumococcal carriage and otitis media induce salivary antibodies to pneumococcal surface adhesin A, pneumolysin, and pneumococcal surface protein A in children. J Infect Dis 2001; 183: CID 2002:35 (15 November) Syrogiannopoulos et al.