Static vs Cidal Antibiotics

To the Editor—The article by Wald-Dickler et al [1], a literature review of “static vs cidal” antibiotics for a variety of infection sites, concludes that there is no support for the superiority of bactericidal over bacteriostatic antibiotics. This conclusion is a potentially misleading view regarding antibiotic activity for bacterial endocarditis, a condition in which there is minimal host defense in the infected vegetation and with much of the high bacterial population in a state of “microbial persistence” [2]. The elimination of bacteria at that site relies on the ability of the exogenous antimicrobial agent over a prolonged period to penetrate into the vegetation and eliminate (kill) the bacteria, not just inhibit growth. The authors state that “bacteriostatatic antibiotics do kill bacteria; they just require a higher concentration than bactericidal agents to achieve specific thresholds of bacterial reduction.” Bacteriocidality and bacteriostatic activity are terms that require a better and more complex understanding. In vitro, particular antibiotics demonstrate different rates of killing bacteria over time, with each time point analyzed by counting the density of surviving colony formers (colony-forming units) that have been exposed to the antibiotic in an aliquot removed from the antibiotic-containing vessel. That is well demonstrated in “time-kill” survival curve analyses [3]. Thus, antibiotics with high rates of kill are labeled bactericidal, and those with low rates of kill are labeled bacteriostatic. However, the rate of kill varies not only with the type of antibiotic and its concentration [3] but also with the bacterial species being challenged [4–6], the inoculum density of the bacteria [4], and environmental conditions, such as pH and protein concentration. The ratio of minimum bactericidal concentration (MBC) to minimum bacteriostatic concentration (minimum inhibitory concentration [MIC]) also provides information regarding the degree of bactericidality but does not provide rates of kill [3], which are useful. A narrow ratio of the MBC to the MIC suggests good bactericidality, and a wide ratio suggests poor bactericidality (bacteriostasis) or “tolerance” (a phenomenon not indicating lack of clinical response in treating bacterial endocarditis [7]). For example, the MBC of enterococci to penicillin is not achievable, as is the MIC, and there is substantial clinical evidence that enterococcal endocarditis must be treated with bactericidal combinations [8]. In addition, the authors’ discussion of the in vitro “formal definition” of MBC is not what is recommended by the National Committee for Clinical Laboratory Standards, which should be used by microbiology laboratories. Tables displayed in that document, copied from a publication in Antimicrobial Agents and Chemotherapy [9], provide a more accurate method of determining MBCs. The authors note that linezolid is a bacteriostatic antibiotic that has had some success in treating bacterial endocarditis. However, linezolid varies in its static versus cidal activity under different experimental conditions [5] and is capable of bacterial killing. With all of the aforementioned considerations, in both animal models [10] and clinical applications [8, 11], antibiotics with high rates of in vitro kill against a given bacterial species are likely to and do have better efficacy and lower relapse rates in the treatment of bacterial endocarditis than antibiotics with low rates of kill against the same species. The argument by Wald-Dickler et al [1] that the antibiotics considered bacteriostatic have proved inferior in the treatment of bacterial endocarditis to those considered bactericidal, because they achieved lower levels clinically, is speculative. The authors’ final statement, that “It is time to abandon the notion that bactericidal antibiotics are intrinsically more effective than bacteriostatic agents,” is not true for endocarditis. Therefore, that qualification should have been included in their final conclusions on this important issue. Potential conflicts of interest All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.