NASOPHARYNGEAL COLONIZATION WITH STREPTOCOCCUS PNEUMONIAE IN CHILDREN RECEIVING TRIMETHOPRIM-SULFAMETHOXAZOLE PROPHYLAXIS

Abstract

Antibiotic-resistant pneumococcal isolates continue to be reported from different parts of the world.1 A multicenter study reported in 1996 by the Centers for Disease Control showed that among Streptococcus pneumoniae isolates recovered from normally sterile body sites (blood, cerebrospinal fluid, joint fluid, middle ear), 18% were resistant to trimethoprim-sulfamethoxazole (TMP-SMX) and multiply resistant strains represented 9.1% of isolates.2 Hedlund et al.3 noted a significant increase in S. pneumoniae resistance to TMP-SMX in Sweden during a 4-year period. The resistance rate increased from 1.4% in 1987 to 7.1% in 1992. The cause of this increase was not investigated; however, the authors attributed it to the increased use of TMP-SMX in Sweden. This finding may be important because increased use of TMP-SMX might promote further emergence of antibiotic resistance. Nasopharyngeal colonization with S. pneumoniae has been described as a risk factor for pneumococcal disease.4, 5 Studies have demonstrated that nasopharyngeal colonization with multiply resistant pneumococci is enhanced by frequent use of antibiotics.5 TMP-SMX is frequently used for prophylaxis in children with HIV infection, children with vesicoureteral reflux and sometimes children with frequent episodes of otitis media. However, there have been no studies regarding addressing the effect of TMP-SMX prophylaxis on pneumococcal colonization. In the present study we compared nasopharyngeal colonization with S. pneumoniae in children receiving oral TMP-SMX prophylaxis to that of an age-matched control group. Patients and methods. Children ages 2 months to 18 years who were taking oral TMP-SMX prophylaxis and were followed in our hospital clinics were eligible to be enrolled. Patients who were included in the study were receiving daily or every other day TMP-SMX for at least 6 weeks. The indications for TMP-SMX prophylaxis included prophylaxis against Pneumocystis carinii pneumonia given to patients with immunosuppressive disorders caused by HIV infection, renal transplantation or immunosuppressive treatment for autoimmune disorders. Also included were healthy children with underlying genitourinary anomalies who were taking TMP-SMX for urinary tract infection prophylaxis. Compliance with TMP-SMX prophylaxis was verified by interviewing the parent or legal guardian and by reviewing the medical records of each patient. Children who had received other antibiotics during the preceding 2 months were excluded. During the study period nasopharyngeal cultures were also obtained from an age-matched group of healthy children who were seen in the outpatient clinic for regular checkup. The study was conducted between November, 1997, and August, 1998. Informed consent was obtained from the parent or legal guardian of each child. Children who had upper respiratory tract infection with fever, otitis media or pneumonia or who appeared ill or required hospitalization were not enrolled. Children who had received antibiotic treatment during the preceding 2 months were excluded. Culture material was collected from the posterior pharyngeal wall with sterile cotton-tipped wire swabs (Pur Fybr Inc., Munster, IN). Specimens were transported in sterile transport tubes and were inoculated within 0.5 h onto Columbia CNA plates as well as agar plates containing 5% defibrinated sheep blood. Plates were incubated for 24 to 48 h at 35°C in 5% carbon dioxide. Pneumococci were identified by standard methods. Typically 3 to 5 alpha-hemolytic colonies exhibiting morphologic characteristics of S. pneumoniae were subcultured for screening with optochin disks (Bacto; Difco, Detroit, MI). All isolates showed a zone of inhibition of ≥14 mm around the optochin disk and had a positive bile solubility test characteristic of S. pneumoniae. No isolates were questionable. MICs were determined for each isolate using E test strips (AB Biodisk, Solna, Sweden) as previously described.6 MIC values were interpreted according to the current guidelines of the National Committee for Clinical Laboratory Standards.7 MIC breakpoints for susceptible, intermediate and resistant, respectively, were (in micrograms per ml): penicillin ≤0.06, 0.1 to 1.0 and ≥2; cefuroxime ≤0.5, 1.0 and ≥2; cefotaxime ≤0.5, 1.0 and ≥2.0; erythromycin ≤0.25, 0.5 and ≥1.0; TMP-SMX ≤0.5, 1.0 to 2.0 and ≥4.0; and vancomycin-susceptible at ≤1.0. Fisher's exact test was used for analysis of differences between the two groups. Results. Nasopharyngeal cultures were obtained from 148 children: 73 in the prophylaxis group and 75 in the control group. The age range of the children in the prophylaxis group was 5 months to 18 years (median, 90 months) and 2 months to 18 years (median, 60 months) in the control group. There were 42 (58%) male children in the prophylaxis group and 46 (61%) in the control group. Duration of TMP-SMX prophylaxis at the time of obtaining cultures ranged from 6 weeks to several years. Nasopharyngeal colonization with Streptococcus pneumoniae was detected in 11 (15%) of 73 children taking TMP-SMX prophylaxis and in 15 (20%) of the 75 children in the control group (P = 0.52). Colonized children in the prophylaxis group included 8 of 36 children who were receiving TMP-SMX prophylaxis to prevent urinary tract infection secondary to vesicoureteral reflux or other urinary tract anomalies, 1 of 25 children who had renal transplantation and 2 of 9 HIV-infected children. The age range of colonized children was 12 months to 13 years (median, 48 months) in the prophylaxis group and 7 months to 6 years (median, 30 months) in the control group. Although the overall prevalence of colonization was similar in the two groups, resistance to penicillin and multiple antibiotics was identified more frequently in the isolates recovered from children in the prophylaxis group than those in the control group. Of the 15 isolates recovered from the control group, 1 (7%) was intermediately resistant to penicillin (MIC 0.16 μg/ml), 1 (7%) to cefuroxime (MIC 1.0 μg/ml) and 2 (13%) to TMP-SMX (MIC 1.5 to 3.0 μg/ml). None was resistant to erythromycin or cefotaxime. In contrast among the 11 isolates recovered from children taking prophylaxis, 9 (82%) were resistant to penicillin (MIC 0.5 to 8.0 μg/ml), 7 (77%) to cefuroxime (MIC 2.0 to 16 μg/ml), 8 (73%) to cefotaxime (MIC 1.0 to 8.0 μg/ml), 4 (36%) to erythromycin (MIC 1.5 to >256 μg/ml) and 7 (64%) to TMP-SMX (MIC 3.0 to >32 μg/ml). Susceptibility data are summarized in Table 1. All 7 isolates that were resistant to TMP-SMX were also resistant to penicillin, cefuroxime and cefotaxime, and 5 were also resistant to erythromycin. Of the 4 isolates susceptible to TMP-SMX 2 were resistant to penicillin, cefuroxime and cefotaxime; all were susceptible to erythromycin. Multiple drug resistance (resistance to 2 or more antibiotics) was present in 9 (82%) of the 11 isolates in the prophylaxis group compared with none in the control group (P = 0.00001). All pneumococcal isolates from both groups were susceptible to vancomycin.TABLE 1: Susceptibility of Streptococcus pneumoniae isolates recovered from children taking prophylactic TMP-SMX and from control subjects Discussion. Increased rates of NP carriage of resistant pneumococci in children have been reported after treatment with beta-lactam and macrolide antibiotics.5, 8-10 The increase in antibiotic resistance is attributed to the selective effect of these antibiotics in allowing the growth of resistant bacterial strains of the NP flora. Our findings suggest that TMP-SMX prophylaxis is a risk factor for NP colonization with multiply resistant pneumococci. Immunocompromised children may be at increased risk for acquisition of resistant S. pneumoniae.11 Among 37 immunocompromised children in our study only 3 were found to be colonized, compared with 8 of 36 otherwise healthy children who were taking TMP-SMX prophylaxis. This indicates that immunocompromised state may not be the only predisposing condition for colonization with resistant S. pneumoniae. Antimicrobial prophylaxis appears to play a major role. Young age has also been considered a risk factor for colonization with resistant S. pneumoniae. Factors considered to put this age group at risk include day-care center attendance and frequent use of empiric antibiotics.12, 13 Of the 26 colonized children in our study population 25 were 6 years old or younger. The median age of children colonized with S. pneumoniae was 30 months in the control group and 48 months in the prophylaxis group. The median age of children colonized with multiple antibiotic-resistant pneumococci was 48 months. This is consistent with previous reports that colonization with S. pneumoniae and antibiotic-resistant pneumococci is more common in younger children. Colonization with resistant S. pneumoniae in our study population was associated with both young age and intake of TMP-SMX. Additionally 4 of 11 isolates recovered from the prophylaxis group were susceptible to TMP-SMX. This indicates that TMP-SMX in the doses used for prophylaxis may not result in adequate NP concentration to eradicate susceptible strains in some children. All 7 S. pneumoniae isolates that were resistant to TMP-SMX in the prophylaxis group had high level resistance (MIC ≥ 4) and all were also resistant to penicillin, cefuroxime and cefotaxime. The majority of isolates resistant to penicillin, cefuroxime and cefotaxime also had high level resistance. Penicillin resistance has been associated with changes in the penicillin-binding proteins of the bacterial cell membrane.14 Resistance to sulfonamides and to trimethoprim may result from alteration of dihydropteroate synthetase and dihydrofolate reductase, respectively.15 Multiple drug resistance in pneumococci has been shown to be transferred by transposable chromosomal conjugative elements.16 However, the exact mechanism of multiple drug resistance as seen in our study patients remains to be determined. Because nasopharyngeal colonization is a risk factor for pneumococcal disease9 our findings make empiric treatment of invasive pneumococcal infections in patients receiving TMP-SMX prophylaxis challenging. Treatment failures have been reported with extended spectrum cephalosporins in meningitis caused by pneumococcal strains intermediately or highly resistant to cefotaxime or ceftriaxone.17 Resistance of pneumococci to penicillin, cefuroxime, erythromycin and TMP-SMX may also have implications in the management of otitis media and sinusitis. The use of TMP-SMX in the prophylaxis of recurrent otitis media may be ill-advised because of potential increase in colonization with resistant pneumococci. Surveillance of nasopharyngeal cultures may provide useful information on the prevalence of resistant strains in a certain community provided that subgroups of patients receiving prophylactic antibiotics are included in these studies. Additional studies with more children are needed to further clarify the role of prophylactic TMP-SMX effect on pneumococcal NP colonization, drug resistance and disease. Nahed Abdel-Haq, M.D. Walid Abuhammour, M.D. Basim Asmar, M.D. Ronald Thomas, Ph.D. Shermine Dabbagh, M.D. Ricardo Gonzalez, M.D. Divisions of Infectious Diseases (NAH, WA, BA) and Nephrology (SD); Urology Department (RG); Children's Hospital of Michigan Children's Research Center of Michigan (RT) Detroit, MI