When the initial cases of vancomycin-intermediate and, later, vancomycin-resistant Staphylococcus aureus were reported more than a decade ago, few were surprised to see a link with end-stage renal disease (ESRD) emerge. 1-3 In the chronic hemodialysis population, the ease of vancomycin administration is one of several factors that have contributed to widespread use, often on an empiric basis. Yet to date, few studies have characterized the appropriateness of antimicrobial use in this population. In this issue, Snyder and colleagues 4 present the results of an observational study of antimicrobial usage in 2 outpatient dialysis centers. Their findings are both instructive and novel and offer important insight into the nature of antimicrobial-prescribing practices in this understudied but critical population. During the 12-month prospective portion of their study, the investigators followed 278 patients (2,549 patientmonths), among whom 89 (32%) received at least one dose of antimicrobials. A total of 1,003 doses were recorded, and an indication for administration was available for 926 (92%). Among these, 276 (30%) doses were classified as inappropriate by the authors, more than half of which simply did not meet criteria for infection. Another quarter represented a failure to choose a more narrow-spectrum agent. The most common inappropriate-use scenarios included administration of vancomycin to treat a single positive blood culture for coagulase-negative staphylococci in the absence of any clinical signs or symptoms, and failure to deescalate from vancomycin to a b-lactam once culture susceptibilities became available. These examples, as well as errors in perioperative prophylaxis, are highlighted by the authors as targets for stewardship. Although this work helps inform future interventions, several questions remain. As presented, the results are quite broad, and additional granularity is required to better understand patterns of use, particularly surrounding vancomycin, given its unique significance. For example, it is unclear how much of the overall antimicrobial use (and more specifically inappropriate use) was driven by a handful of outliers. Although the patient populations appeared relatively similar, significant prescribing differences existed between the 2 dialysis centers, calling the generalizability of the results into question. Despite these limitations, there is little doubt that large amounts of inappropriate antimicrobial use is occurring at many dialysis centers, and the authors clearly identify a meaningful opportunity for improvement. In order to design interventions that curb inappropriate antimicrobial use, we must first consider the conditions that contribute to this situation. While specific reasons were not explored in the Snyder study, it is again worth highlighting prescribing differences between the 2 study sites, suggesting that individual practice patterns play a major role. In our clinical experience, key factors that contribute to antimicrobial administration among hemodialysis patients include the following. ESRD patients are a high-risk population. The chronic hemodialysis population is perhaps the most medically complex group of patients outside an acute care setting. Recent national data show that incident hemodialysis patients have a mortality rate of 225 per 1,000 patient-years and that the probability of survival at 5 years is only 35%. 5 Infection is the second-leading cause of mortality in this group. While no one wakes up in the morning and says, “I think I’ll misuse vancomycin today,” most nephrologists have a low threshold for initiating empiric antimicrobials for even soft signs of infection. In many regards, the most remarkable aspect of Snyder’s findings is not that 30% of antimicrobial use was inappropriate but rather that 70% was appropriate. Administering antimicrobials in hemodialysis patients is convenient, especially vancomycin. Hemodialysis patients are a captive audience with readily available vascular access. Be
Read full abstract
Apr 23, 2024
Ibrahima Lyra Sarr 
Open Access
