Can interchangeability of lincosamides be assumed in clinical practice? Comparative MICs of clindamycin and lincomycin for Streptococcus pyogenes, Streptococcus agalactiae and Staphylococcus aureus

Sir, Recently in Australia, prescribing patterns for lincosamides have changed. Parenteral lincomycin use now exceeds parenteral clindamycin use in hospitals, including intensive care units. The reasons pertain to availability, cost and the recommendations of the national antibiotic guidelines. The Australian therapeutic guidelines present lincomycin and clindamycin as equivalent treatments for serious infections due to Streptococcus pyogenes, Streptococcus agalactiae and Staphylococcus aureus. As there are no CLSI or EUCAST breakpoints for lincomycin, Australian clinicians infer lincomycin susceptibility from clindamycin susceptibility. Challenging the validity of this approach are reports of ‘L-phenotype’ staphylococci and b-haemolytic streptococci that test clindamycin susceptible but lincomycin resistant. L-phenotype staphylococci and streptococci of both animal and human origin have been identified, mediated by lnu(A), lnu(B), lnu(C), lnu(D) and lnu(F). – 5 The frequency of L-phenotype isolates occurring in Australia is unknown. We compared the MICs of lincomycin and clindamycin for local clinical isolates of S. pyogenes, S. agalactiae and S. aureus to determine whether L-type susceptibility patterns were present, and to compare MIC values. The data will be used as an evidence base for local policies regarding lincosamide choice. Ninety S. pyogenes, 45 S. agalactiae and 100 S. aureus isolates (50 methicillin susceptible and 50 methicillin resistant) were tested for MICs of clindamycin and lincomycin by the CLSI agar dilution method. A 52-pin Steers replicator delivering a 2 mL inoculum was utilized. Erythromycin susceptibility was also determined using agar dilution. S. aureus ATCC 29213 and S. pneumoniae ATCC 29619 were included as controls. CLSI breakpoints were used for erythromycin and clindamycin susceptibility interpretation. The Comité de L’Antibiogramme de la Société Française de Microbiologie (CA-SFM) lincomycin breakpoints for Gram-positive cocci (susceptible ≤2 mg/L, resistant .8 mg/L) were utilized to define lincomycin susceptibility. Erythromycin-resistant, clindamycinsusceptible isolates were tested for inducible clindamycin resistance by ‘D-test’. MICs of clindamycin and lincomycin were compared in Stata v11 (College Station, TX, USA) using the Wilcoxon signed-rank test for matched pairs. All S. pyogenes and S. aureus isolates tested, regardless of their macrolide–lincosamide–streptogramin B phenotype, had concordant susceptibilities for clindamycin and lincomycin (Table 1). Of interest, three of the S. agalactiae isolates had an erythromycinsusceptible, low-level clindamycin-resistant pattern (clindamycin MIC1⁄41 mg/L, CLSI breakpoint: susceptible ≤0.25 mg/L) but were susceptible to lincomycin at the CA-SFM breakpoint (lincomycin MIC1⁄42 mg/L, CA-SFM breakpoint: susceptible ≤2 mg/L). This phenotype of macrolide-susceptible, clindamycin-resistant S. agalactiae has previously been reported. All other S. agalactiae strains had concordant lincosamide susceptibilities. All lincosamide-susceptible isolates, of all the bacterial species tested, displayed lower MICs of clindamycin than of lincomycin (Wilcoxon signed-rank test for pairs P,0.0001). MIC50 values