NEONATAL SOCIETY ABSTRACTS
Hetero-resistance to vancomycin in Coagulase-negative Staphylococci (CoNS) isolated from neonatal blood cultures
Presented at the Neonatal Society 2003 Summer Meeting (programme).
Saraswatula A1, Miles T1,2, Want S3, Holmes A2, Modi N1, Bamford K2
1 Division of Paediatrics, Obstetrics & Gynaecology,
Hammersmith & Queen Charlotte's Hospital and Chelsea & Westminster Hospital,
Imperial College, London
2 Department of Infectious Diseases, Faculty of Medicine, Hammersmith & Queen Charlotte's Hospital and Chelsea & Westminster Hospital, Imperial College, London
3 Department of Microbiology, Hammersmith Hospital, London
Background: Coagulase-negative Staphylococci (CoNS) are commonly isolated from neonatal blood cultures. Though often contaminants, CoNS are also the leading cause of genuine bloodstream infection in preterm neonates. This has led to increased use of vancomycin and concern regarding the risk of emergence of vancomycin resistance, particularly in strains demonstrating hetero-resistance1. Some isolates that fulfil routine laboratory minimum inhibitory concentration (MIC) criteria for susceptibility contain sub-populations of cells at low frequencies that grow at a higher concentration of the antibiotic (i.e. have an elevated MIC). This heterogeneous expression of resistance to the antibiotic is termed hetero-resistance. It is not known if hetero-resistance is associated with greater clinical morbidity.
Aims: We aimed 1) to determine if vancomycin hetero-resistance can be identified in isolates of CoNS from neonatal blood cultures and 2) if there is a predominance of hetero-resistant isolates among CoNS associated with clinical bloodstream infection.
Experimental methods: Archived isolates of CoNS from neonatal blood cultures obtained in 2001 were examined. An isolate of CoNS was considered to be indicative of clinical bloodstream infection if it was associated with 3 or more of 12 predefined clinical signs suggestive of systemic infection and either a rise in C-reactive protein or an acute fall in platelet count. Either a manual2 or a semi automated technique using a Don Whitley spiral plater (WASP®) was used to enumerate the sub-populations within each isolate that were able to grow in 2-fold incremental concentrations of vancomycin. Population analysis profiles3 (PAPs) were then constructed. The results of routine laboratory vancomycin testing (BSAC standardised disc susceptibility test) were noted. PAPs were interpreted using the MIC criteria of the British Society for Antimicrobial Chemotherapy, namely =8 mg/mL, and the United States National Committee for Clinical Laboratory Standards, namely 8 -16 mg/mL (intermediate) and =32 mg/mL (resistant). In our study, sub-populations able to grow at =12 mg/mL of vancomycin were considered hetero-resistant.
Results: PAPs were constructed for 37 out of 79 CoNS blood culture isolates (26 using the WASP technique and 11 manually). All CoNS isolates were sensitive to vancomycin on routine laboratory testing. However hetero-resistance was demonstrated. Sub-populations were identified that were able to grow at = 8 mg/mL in 75% (28/37) of isolates, at =12 mg/mL in 68% (25/37), and at =32 mg/mL in 21% (8/37). Thirteen of the 37 (35%) CoNS isolates were associated with clinical bloodstream infection. There was no significant difference in the distribution of hetero-resistance in strains associated with genuine bloodstream infection and those that were considered contaminants (Fisher's exact test).
Table: Comparison of hetero-resistance status of CoNS isolates with their clinical relevance
Conclusions: There is a high percentage of hetero-resistance to vancomycin in CoNS isolated from neonatal blood cultures believed to represent genuine bloodstream infection, as well as isolates considered to be contaminants. This is not revealed by routine antibiotic susceptibility testing. Further work is needed to determine if hetero-resistance affects the response to vancomycin therapy and if antibiotic pressure is an influence upon the emergence of this phenomenon.
1. Van Der Zwet WC, Debets-Ossenkopp YJ, Reinders E et al. J Clin Microbiol 2002; 40(7): 2520-5
2. Sieradzki K, Villari P, Tomasz A. Antimicrob Agents Chemother 1998; 42(1): 100-7
3. Wootton, M., Howe RA, Hillman R et al. J Antimicrob Chemother 2001; 47(4): 399-403