TEXT

Top Abstract Text References

Members of the genus Clostridium are a major part of the anaerobic microflora of humans and are a potential cause of human infections. Clostridium innocuum belongs to the normal intestinal flora of human infants and adults and is one of the species which have been reported to cause human infections, such as intra-abdominal sepsis, bacteremia, and endocarditis (5, 9, 12). However, its identification by commercial kits remains unsatisfactory and requires, therefore, the use of gas-liquid chromatography and additional conventional tests (1). Clostridia are usually considered susceptible to vancomycin and teicoplanin (8). Unexpectedly, we observed that most C. innocuum strains isolated from clinical specimens examined in the Laboratory of Bacteriology, University Hospital Center, Nancy, France, had diminished inhibition zone diameters (range, 14 to 16 mm) with vancomycin disks (30 µg) as determined by the agar diffusion method. Thus, we wished to determine the in vitro activities of vancomycin and teicoplanin against these strains.

We studied 28 C. innocuum strains consecutively isolated form 28 patients hospitalized at the University Hospital Center of Nancy between 1993 and 1995 (blood, n = 3; intra-abdominal pus, n = 25). All these isolates have been identified as C. innocuum on the basis of morphological and biochemical characteristics. Identification of three blood culture isolates was confirmed by determination of the nucleotide sequence of a portion of the 16S rRNA sequence. From genomic DNA, 311-bp fragments were amplified by PCR with universal 16S rRNA oligonucleotides and sequenced (10). The nucleotide sequences of the three clinical isolates and of C. innocuum NCIB 10674 displayed 99% identity and were distantly related to those of 33 other Clostridium species determined by Collins et al. (3). Strains were stored at 80°C in brucella broth supplemented with 15% glycerol prior to assay. Clostridium perfringens ATCC 13124, Clostridium difficile ATCC 9689, and C. innocuum NCIB 10674 were included as controls. In addition, the in vitro activities of teicoplanin and vancomycin against 37 other clinical Clostridium isolates, including C. perfringens (n = 12), C. difficile (n = 11), Clostridium clostridioforme (n = 5), Clostridium butyricum (n = 4), Clostridium tertium (n = 3), and Clostridium ramosum (n = 2), were determined.

MICs of vancomycin (Lilly, St. Cloud, France) and teicoplanin (Marion Merrell Dow, Levallois-Perret, France) were determined, as recommended by the National Committee for Clinical Laboratory Standards (11), by the agar dilution method on Wilkins-Chalgren agar (Difco Laboratories, Detroit, Mich.). A final inoculum of 10⁵ CFU per spot was delivered with a multipoint inoculator. Cultures were then incubated at 35°C for 48 h in an anaerobic chamber (Don Whitley Scientific Ltd., Shipley, United Kingdom). MICs were defined as the lowest concentrations of each antibiotic that inhibited visible growth on agar and were interpreted in accordance with the guidelines of the Comité de l'Antibiogramme de la Société Française de Microbiologie (4).

A PCR assay was used to detect the presence of vanA, vanB, vanC1, and vanC2 resistance genes in the three clinical C. innocuum strains isolated from blood as well as in C. innocuum NCIB 10674. The PCR assay was performed as described previously for the detection of glycopeptide resistance genotypes in enterococci (6). Enterococcus faecium BM 4147 (vanA), Enterococcus faecalis V 583 (vanB), Enterococcus gallinarum BM 4174 (vanC1), and Enterococcus casseliflavus ATCC 25788 (vanC2) were used as controls.

Teicoplanin showed excellent activity against the C. innocuum strains tested, inhibiting all isolates at concentrations ranging from 0.25 to 1 µg/ml (MIC at which 90% of the isolates are inhibited [MIC₉₀], 0.5 µg/ml). In contrast, the MICs of vancomycin ranged from 8 to 16 µg/ml (MIC₉₀, 16 µg/ml), classifying all isolates as intermediately resistant to this antibiotic, although 46% (13 of 28) of strains had inhibition zone diameters as determined by the disk diffusion method which were 17 mm. This confirms the poor correlation existing between inhibition zone diameters and MICs for glycopeptide susceptibility testing in gram-positive bacteria (2, 14). Vancomycin was at least eight times more active against the other Clostridium strains tested (Table 1).

No amplification product was observed with DNA from the four C. innocuum strains tested by PCR, suggesting that the low-level vancomycin resistance may be due to genetic determinants other than those identified in enterococci.

Few previous studies have reported Clostridium species to be intermediately resistant or even strongly resistant to vancomycin (1, 7, 13). In two of these studies (7, 13), varied clostridial species were tested as one group of uniform organisms, i.e., C. innocuum was not distinguished from the other species. Alexander et al. (1) have noted that 50 and 90% of the 21 strains of C. innocuum that they tested were inhibited by 4 and 8 µg of vancomycin per ml, respectively. However, these authors did not compare the inhibitory activities of vancomycin and teicoplanin. Except for those for some strains which appeared to be susceptible to vancomycin (MICs = 2 to 4 µg/ml), their results are similar to ours since the difference in MIC₉₀s was less than 2 dilutions. These differences could be explained in part by the use of different susceptibility testing media.

Thus, to our knowledge, this report is the first description of low-level vancomycin resistance without cross-resistance to teicoplanin in C. innocuum. The fact that this phenomenon was observed in all isolates tested suggests the presence of intrinsic low-level vancomycin resistance, the genetic basis and biochemical mechanism of which remain to be investigated. Moreover, documenting low-level resistance to vancomycin may aid in characterization of this species, identification of which may be difficult (1).