In a serotype 11A clone of erythromycin-resistant pneumococci isolated from young Greek carriers, we identified the nucleotide sequence of erm(A), a methylase gene previously described as erm(TR) in Streptococcus pyogenes. The erm(A) pneumococci were resistant to 14-and 15-member macrolides, inducibly resistant to clindamycin, and susceptible to streptogramin B. To our knowledge, this is the first identification of resistance to erythromycin in S. pneumoniae attributed solely to the carriage of the erm(A) gene.Resistance of Streptococcus pneumoniae to erythromycin and the other macrolides is increasing in many parts of the world (1,5,7,18). Strains resistant to erythromycin are also resistant to azithromycin, clarithromycin, and roxithromycin (25). Recently, it has been shown that pneumococci resistant to erythromycin have mainly one of two distinct resistance determinants, erm(B) or mef (A)

We evaluated nasopharyngeal colonization with erythromycin-resistant Streptococcus pneumoniae during the first 2 years of life in central and southern Greece. Of 2448 children studied from February 1997 to February 1999, 766 (31%) carried 781 pneumococcal isolates. Ninety-five (3.9%) of the children attended day care centers. Eighteen percent of the pneumococci were resistant to erythromycin (minimal inhibitory concentration 1 to >128 microg/mL), with 67.9% of them carrying the erm(B) gene and 29.2% mef(A) gene products. Four strains possessed neither the erm(B) nor the mef(A) gene. Multidrug resistance occurred in 97% and 40% of isolates carrying the erm(B) and mef(A) gene, respectively. An association was found between the erm(B) gene and serotypes 6B and 23F and between the mef(A) gene and serotypes 14 and 19F. A significant relationship existed between carriage of erythromycin-resistant pneumococci and use of macrolides or beta-lactams in the previous 3 months; the association was strongest when macrolide therapy was administered during the last month (odds ratio, 5.92; P=.0001). The findings indicate the necessity of a judicious use of both macrolides and beta-lactams in young children to reduce the colonization with erythromycin-resistant pneumococci and the subsequent spread of such strains to the community.

No abstract

The current high (38%) prevalence of erythromycin-resistant S. pyogenes in Central and Southern Greece requires continuous surveillance and careful antibiotic policy.

IntroductionIn Greece, fusidic acid and clindamycin are commonly used for the empiric therapy of suspected staphylococcal infections.MethodsThe medical records of children examined at the outpatient clinics or admitted to the pediatric wards of the University General Hospital of Larissa, Central Greece, with community-associated staphylococcal infections from January 2003 to December 2009 were reviewed.ResultsOf 309 children (0-14 years old), 21 (6.8%) had invasive infections and 288 (93.2%) skin and soft tissue infections (SSTIs). Thirty-five patients were ≤30 days of age. The proportion of staphylococcal infections caused by a community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) isolate increased from 51.5% (69 of 134) in 2003-2006 to 63.4% (111 of 175) in 2007-2009 (P = 0.037). Among the CA-MRSA isolates, 88.9% were resistant to fusidic acid, 77.6% to tetracycline, and 21.1% to clindamycin. Clindamycin resistance increased from 0% (2003) to 31.2% (2009) among the CA-MRSA isolates (P = 0.011). Over the 7-year period, an increase in multidrug-resistant CA-MRSA isolates was observed (P = 0.004). One hundred and thirty-one (93.6%) of the 140 tested MRSA isolates were Panton-Valentine leukocidin-positive. Multilocus sequence typing of 72 CA-MRSA isolates revealed that they belonged to ST80 (n = 61), ST30 (n = 6), ST377 (n = 3), ST22 (n = 1), and ST152 (n = 1). Resistance to fusidic acid was observed in ST80 (58/61), ST30 (1/6), and ST22 (1/1) isolates.ConclusionIn areas with high rate of infections caused by multidrug-resistant CA-MRSA isolates, predominantly belonging to the European ST80 clone, fusidic acid and clindamycin should be used cautiously as empiric therapy in patients with suspected severe staphylococcal infections.

Erythromycin-resistant Streptococcus pneumoniae isolates from young carriers were tested for their antimicrobial susceptibility; additionally, inducibility of macrolide and clindamycin resistance was investigated in pneumococci carrying erm(A), erm(B), or mef(A). Of 125 strains tested, 101 (81%) were multidrug resistant. Different levels of induction were observed with erythromycin, miocamycin, and clindamycin in erm(B) strains; however, in erm(A) strains only erythromycin was an inducer. Induction did not affect macrolide MICs in mef(A) strains.The predominant mechanisms of resistance to erythromycin and the other macrolides in Streptococcus pneumoniae are through target site modification by methylation that prevents the binding of the antibiotic to its ribosomal target, encoded by the erm(B) gene (16), or through efflux of the antibiotic, mediated by the mef(A) gene (13). Methylation of the ribosomal target of the antibiotics leads to cross-resistance to macrolides (M), lincosamides (L), and streptogramin B (S B ), the so-called MLS B phenotype. Erm methylase synthesis can be inducible or constitutive (10, 11). Alterations in ribosomal proteins L4 and L22 or 23S rRNA have been reported to cause resistance in S. pneumoniae (1). Antibiotic efflux confers resistance only to the 14-and 15-member macrolides (1, 4, 5, 7).Recently, S. pneumoniae isolates resistant to erythromycin due to carriage of the erm(A) gene have been reported (15). The macrolide resistance determinant erm(A) gene was previously described as erm(TR) in S. pyogenes (12) and now is reclassified as erm(A) (9).The present study was undertaken to compare differences in antimicrobial susceptibility between S. pneumoniae isolates harboring erm(A) and those carrying either the erm(B) or mef(A) gene and to investigate the effect of induction on the expression of macrolide resistance in erm(A)-, erm(B)-, or mef(A)-positive pneumococci. The presence of genes coding for macrolide resistance was studied by PCR as described previously (8,12,13).(This work was presented in part at the 42nd Interscience Conference on Antimicrobial Agents and Chemotherapy, San Diego, Calif., 27 to 30 September 2002 [abstract no. C2-1977).One hundred twenty-five erythromycin-resistant S. pneumoniae isolates carrying the erm(A), erm(B), or mef(A) gene, recovered from 2,448 carriers younger than 2 years between February 1997 and February 1999 (14, 15), were evaluated for their susceptibility to 10 antibiotics representing different classes of antimicrobial agents. These agents were 14-member (erythromycin and clarithromycin), 15-member (azithromycin), and 16-member ring macrolides (miocamycin), lincosamides (clindamycin), streptogramins (quinupristin/dalfopristin), ketolides (telithromycin), penicillin G, tetracycline, and chloramphenicol. Antimicrobial susceptibility tests were performed by the agar dilution method as previously described (3). Plates were incubated in ambient air at 37°C. Antibiotics were obtained from their respective manufacturers. Miocamycin was from Meiji Sei...

The prevalence, resistance patterns and serotypes of antibiotic-resistant Streptococcus pneumoniae strains recovered from Greek carriers under 24 months of age were studied. From February 1997 to April 1998, nasopharyngeal cultures were performed in 1,269 children (ages 2-23 months, median 11 months) living in various areas of central and southern Greece. Resistance (including both intermediate and resistant isolates) to one or more antimicrobial agents was found in 132 of the 421 (31%) Streptococcus pneumoniae isolates, as follows: penicillin, 9% intermediate, 7.6% resistant; cefotaxime, 5.2% intermediate, 0.5% resistant; erythromycin, 0.7% intermediate, 18.1% resistant; clindamycin, 0.2% intermediate, 12.4% resistant; tetracycline, 0.7% intermediate, 16.4% resistant; chloramphenicol, 12.4% resistant; and trimethoprim-sulfamethoxazole, 3.8% intermediate, 14.3% resistant. The MICs of penicillin for 66% of the penicillin-nonsusceptible pneumococci were 1-4 microg/ml. Multidrug resistance was found in 64% of penicillin-nonsusceptible and 37% of penicillin-susceptible strains. Sixty-two percent of the penicillin-susceptible, multidrug-resistant strains belonged to serotype 6B and were resistant to all five non-beta-lactam agents tested. This notable serotype 6B resistance pattern was described for the first time in a previous study performed from December 1995 to February 1996 in the city of Patras, southwestern Greece. Seventy-two percent of antibiotic-resistant isolates belonged to serotypes 6B, 9V, 14, 18C, 19F and 23F. These results document the spread of resistant pneumococcal strains in central and southern Greece, many of which are multidrug resistant.

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 31-60 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 31-60 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.