Detection and prevalence of inducible clindamycin resistance in staphylococci.

Staphylococcus aureus and coagulase-negative staphylococci (CNS) are recognized as causing nosocomial and community-acquired infections in every region of the world. The resistance to antimicrobial agents among staphylococci is an increasing problem. Clindamycin (CL) is considered to be one of the alternative agents in these infections. This study demonstrates a simple, reliable method (double-disc diffusion test) for detecting inducible resistance to CL in erythromycin-resistance (ER-R) isolates of S. aureus and CNS. A total of 883 (52.3%) isolates of S. aureus and 804 (47.7%) isolates of CNS were selected from recent (2003-2005) clinical isolates recovered in the laboratory of the authors; duplicate isolates were not included. A total of 214 (12.6%) S. aureus and 308 (18.3%) CNS isolates were selected based on ER-R and CL sensitivity using standard National Committee for Clinical Laboratory Standards disc diffusion testing. A total of 1687 staphylococcal isolates were included, consisting of 27.5% meticillin-resistant S. aureus, 24.8% meticillin-sensitive S. aureus, 36.1% meticillin-resistant CNS and 11.6% meticillin-sensitive CNS isolates: 30.9% of staphylococcal isolates (214 S. aureus and 308 CNS) that were erythromycin resistant and CL sensitive were tested for inducible resistance using the D-test. A D-shaped zone around the CL was observed for 70.9% of staphylococcal isolates (81.8% of S. aureus isolates and 63.3% of CNS isolates) with an ER-R and a clindamycin-sensitive (CL-S) phenotype. The organism was positive for inducible clindamycin resistance (CL-R). There was a 21.9% level of inducible macrolide-lincosamide-streptogramin B resistance phenotype among all the staphylococcal isolates. When the S. aureus and CNS strains among all the staphylococcal isolates were compared statistically, inducible CL-R in CNS strains was determined to be 23% more positive (P=0.028, odds ratio 0.77, 95% confidence interval 0.61-0.98). When a statistical comparison was performed among ER-R but CL-S staphylococcal isolates inducible CL-R in S. aureus strains was determined to be 2.6 times more positive (P=0.000, odds ratio 2.6, 95% confidence interval 1.68-4.04). A simple, reliable method of detecting inducible resistance to CL in ER-R isolates of S. aureus and CNS is described. Clinical microbiology laboratories should use the double-disc diffusion test as standard practice with all ER-R staphylococci. CL should not be used in patients with infections caused by inducibly resistant staphylococcal isolates. Therapeutic failures may thus be avoided.