Reply to Gandra et al

To the Editor—We thank Gandra et al for their analysis of clindamycin/erythromycin-resistant methicillin-susceptible Staphylococcus aureus (MSSA) across the United States between 1999 and 2011 [1]. The authors observed a steady increase in this resistance profile nationwide, with a particularly steep jump in New York state between 2002 and 2004 and subsequently in the Northeast region around 2007–2008. Based on our study and others [2–4], Gandra et al suggest that this pattern of resistance to erythromycin and clindamycin with preserved susceptibility to tetracycline might serve as a surrogate marker for the emergence and spread of ST398 MSSA in the United States.

In support of this observation, we note that the projected timeline coincides with the detection of ST398 in our studies in northern Manhattan [5, 6] and adjacent regions [7]. In addition, 25% (n = 48/196) of non-ST398 MSSA strains harbored clindamycin/erythromycin resistance in our study, of which 23% (n = 11/48) were also resistant to other antibiotics, in particular levofloxacin and tetracycline. In contrast, we only observed resistance to levofloxacin in 1 ST398 isolate [2].

Although the data provided by Gandra et al provide a comprehensive summary of drug resistance in MSSA across the United States, studies on the molecular epidemiology of MSSA in the United States have so far failed to detect ST398 in sizable numbers across the country [8, 9]. The authors acknowledge that other clones are likely to contribute to the proportion of MSSA that are resistant only to erythromycin and clindamycin. Extrapolating from our studies in northern Manhattan, isolates with a diverse genetic background, including of the CC30 and CC8 lineages, accounted for this susceptibility pattern among the non-ST398 isolates [2].

It would have been interesting to correlate the resistance data presented by Gandra et al with the nationwide usage of macrolides and clindamycin. This could help delineate whether the observed evolving resistance in MSSA is a direct consequence of increased drug pressure, or rather a fingerprint of a clone harboring the erythromycin/clindamycin resistance genotype. Interestingly, in ST398 MSSA, ermT is integrated into the core chromosome rather than carried on a plasmid, providing a potential mechanism for a preferential clonal spread of erythromycin- and clindamycin-resistant ST398 MSSA [3, 6].

Nationwide databases on resistance and drug usage, as accessed by Gandra et al in their study, have the potential to rapidly provide profiles of shifting patterns in drug resistance. These can then be further evaluated in targeted molecular surveillance studies, increasing our ability to recognize emerging pathogenic clones in real time.