Clinical Therapeutics/Volume 35, Number 6, 2013 Antimicrobial Stewardship

Transcription

1 Clinical Therapeutics/Volume 35, Number 6, 2013 Antimicrobial Stewardship A Stewardship Program s Retrospective Evaluation of Vancomycin AUC 24 /MIC and Time to Microbiological Clearance in Patients with Methicillin-Resistant Staphylococcus aureus Bacteremia and Osteomyelitis Kristen M. Gawronski, PharmD 1 ; Debra A. Goff, PharmD, FCCP 2 ; Jack Brown, PharmD, MS 3,4,5 ; Tina M. Khadem, PharmD 3,5 ; and Karri A. Bauer, PharmD, BCPS 2 1 Department of Pharmacy, Nationwide Children s Hospital, Columbus, Ohio; 2 Department of Pharmacy, The Ohio State University Wexner Medical Center, Columbus, Ohio; 3 School of Pharmacy and Pharmaceutical Sciences, The State University of New York at Buffalo, Buffalo, New York; 4 School of Public Health, The State University of New York at Buffalo, Buffalo, New York; and 5 Department of Pharmacy, University of Rochester Medical Center, Rochester, New York ABSTRACT Background: Current guidelines for the treatment of methicillin-resistant Staphylococcus aureus (MRSA) bacteremia recommend targeting a vancomycin AUC 24 /MIC 400. Data on the association between AUC 24 /MIC and microbiological clearance in patients with concomitant MRSA bacteremia and MRSA osteomyelitis are limited. Objective: The objective of this study is to evaluate the association between the vancomycin AUC 24 /MIC and time to microbiological clearance in patients with concomitant MRSA bacteremia and MRSA osteomyelitis. Methods: Adult inpatients with concomitant MRSA bacteremia and MRSA osteomyelitis treated with vancomycin from January 1, 2007, through December 31, 2011,wereevaluated.Classification and regression tree analysis was used to identify the AUC 24 /MIC associated with time to microbiological clearance. Results: Fifty-nine patients had complete data available for review and were included in the analysis. Classification and regression tree analysis identified an AUC/MIC of 293 as the breakpoint that provides the greatest difference in time to microbiological clearance. On univariate analysis, mean (SD) time to clearance was 2 days shorter when the AUC/MIC was 4293 (4 [2] days vs 6 [3] days, P ¼ 0.01). Mean (SD) infection-related length of stay was 13 (6) versus 18 (14) days in patients with an AUC 24 /MIC ratio 4293 or 293, respectively (P ¼ 0.25). In patients with an AUC 24 /MIC 293, 39% versus 17% (P ¼ 0.09) had recurrent bacteremia and were readmitted compared with patients with an AUC/MIC Only 9% were able to achieve an AUC 24 /MIC 4293 when the vancomycin MIC was 41 μg/ml. Conclusions: We observed a 42.5-fold increase in time to microbiological clearance in patients with concomitant MRSA bacteremia and MRSA osteomyelitis unable to achieve a vancomycin AUC 24 /MIC A 5-day increase in hospital and infection-related length of stay was observed when this target was not achieved. Recurrence of bacteremia and hospital readmissions were higher in the cohort who did not achieve an AUC 24 /MIC Only 9% of patients were able to achieve an AUC 24 /MIC 4293 if the isolate MIC was 41 μg/ml. Trough concentration did not correlate with AUC 24 / MIC. In patients with concomitant MRSA bacteremia and MRSA osteomyelitis treated with vancomycin, stewardship programs should optimize pharmacodynamic parameters, specifically AUC 24 /MIC, or alternative therapies should be considered. (Clin Ther. 2013;35: ) & 2013 Elsevier HS Journals, Inc. All rights reserved. Keywords: infectious diseases, osteomyelitis, pharmacodynamics, vancomycin. INTRODUCTION Methicillin-resistant Staphylococcus aureus (MRSA) infections are associated with significant morbidity, Accepted for publication May 10, /$ - see front matter & 2013 Elsevier HS Journals, Inc. All rights reserved. 772 Volume 35 Number 6

2 K.M. Gawronski et al. mortality, and health care expenditures. 1 4 In recent years, MRSA infection rates across the United States have increased, with approximately 68,000 cases of MRSA bacteremia reported annually. 5,6 Mortality estimates vary, but studies have reported a rate that exceeds 20%. This rate may be even higher in patients with complicated MRSA bacteremia, including endocarditis and osteomyelitis. 1,7 9 MRSA infections are associated with significant health care costs. Various institutions have reported on the economic impact, with median hospital charges per episode of MRSA bacteremia ranging from $26,424 to $113,852. 8,10,11 Despite reports of treatment failures, many still consider vancomycin the empiric antibiotic of choice for the treatment of MRSA bacteremia. Clinical practice guidelines recommend targeting vancomycin steady-state trough concentrations between 15 and 20 μg/ml. 12 Trough concentrations are considered a surrogate marker of clinical efficacy and toxicity. 1,7,12 15 An additional pharmacodynamic parameter, AUC 24 /MIC, has been proposed as the variable associated with improved clinical efficacy. 1,7,12,13,15,16 An AUC 24 /MIC 400 was found to be predictive of positive clinical and bacteriologic outcomes in patients with MRSA pneumonia. 1,17 However, this cutoff may not be widely applicable. In a second study, an AUC 24 /MIC o211wasfoundtobe associated with increased attributable mortality in a cohort of patients with MRSA-complicated bacteremia and infective endocarditis. 18 A significant complication associated with MRSA bacteremia is the development of osteomyelitis. This complication presents a difficult challenge to clinicians because often source control may not be possible; therefore, it is imperative to optimize antimicrobial therapy. Vancomycin bone penetration is variable based on the type of bone tissue, degree of vascularization, and inflammation. 19,20 Considering these variables, we sought to evaluate the association between the vancomycin AUC 24 /MIC and time to microbiological clearance in patients with complicated MRSA bacteremia and associated MRSA osteomyelitis. PATIENTS AND METHODS This study was a retrospective medical record review of all adult inpatients from January 1, 2007, to December 31, 2011, with concomitant MRSA bacteremia and MRSA osteomyelitis who received vancomycin at The Ohio State University Wexner Medical Center (OSUWMC), a 1234-bed, university-affiliated, academic medical center. The study was approved by The Ohio State University Institutional Review Board. The primary outcome was to evaluate vancomycin AUC 24 /MIC and its correlation with time to microbiological clearance. Patients aged 18 to 89 years with complicated bacteremia ( 2 positive blood culture results for MRSA) and osteomyelitis (International Classification of Diseases, Ninth Revision [ICD-9]code for osteomyelitis plus radiographic findings) were included. Patients had to have received 3 doses of vancomycin and have a steady-state trough concentration to be eligible for inclusion. Patients were excluded if they received o3 doses of vancomycin, were transferred from an outside hospital while receiving vancomycin, received dialysis (hemodialysis or continuous renal replacement therapy), or had a polymicrobial bacteremia. A computerized list of patients with MRSA bacteremia was generated by the Clinical Microbiology database, which allowed the identification of potential study patients. Only the first episode per study period was analyzed. The electronic medical records of all patients were reviewed to determine eligibility. Data collection included clinical and microbiological characteristics, including age, sex, weight, hospital service, intensive care unit admission, infectious diseases consult, receipt of prior vancomycin or anti-mrsa therapy (clindamycin, daptomycin, linezolid, trimethoprim-sulfamethoxazole, or tigecycline) within 90 days, time to first positive blood culture result from admission, time to vancomycin administration from blood culture draw, inpatient treatment duration, and vancomycin pharmacodynamics. Clinical outcomes included time to microbiological clearance, length of stay (LOS), infection-related LOS, treatment failure, alteration of therapy from vancomycin, recurrent MRSA bacteremia within 90 days, hospital readmission within 30 days, inpatient all-cause mortality, and nephrotoxicity. Treatment duration was defined as the difference of days between vancomycin administration to discontinuation of vancomycin therapy or hospital discharge. Microbiological clearance was defined as 2 consecutive negative blood culture results. The stewardship program recommended daily blood cultures until blood culture results were negative for 48 hours. The LOS was calculated as the difference between day of hospital admission until the day of hospital discharge or death. Infection-related LOS was defined as the difference from the day the blood June

3 Clinical Therapeutics culture was obtained until the completion of MRSA antimicrobial therapy, hospital discharge, or death. On the basis of institution-specific definition, treatment failure was defined as persistent positive blood cultures for 7 days with therapeutic vancomycin levels of 15 to 20 μg/ml. Nephrotoxicity was defined as an increase of 40.5 mg/dl (or a 50% increase) in serum creatinine above baseline on 2 consecutive days in the absence of an alternative explanation. 12 Microbiological Data The microbiology laboratory performed antimicrobial susceptibility testing of clinical isolates using Etest (AB Biodisk, Solna, Sweden; biomérieux, Durham, North Carolina) (2007 to 2010) or Microscan (Siemens, Malvern, Pennsylvania) (2010 to 2012) according to the manufacturers instructions and the Clinical and Laboratory Standards Institute guidelines. Clinical isolates were not tested using both Etest and Microscan. Testing was completed based on the specified time frames. An isolate with a vancomycin MIC 2 μg/ml was considered susceptible. Pharmacokinetic and Pharmacodynamic Data All patients were initially treated with vancomycin, 15 mg/kg per dose based on actual body weight, up to a maximum of 2 g per dose. Loading doses of vancomycin are not routinely administered at our institution, and no patient received a loading dose in this study. The dosing interval is selected and adjusted based on renal function per an institutional guideline: patients with a creatinine clearance (CrCl) 460 ml/ min, every 12 hours; CrCl of 30 to 59 ml/min, every 24 hours; and CrCl o30 ml/min, every 48 hours. Dosing regimens are adjusted by clinical pharmacists to achieve a serum trough concentration of 15 to 20 μg/ml based on an institutional guideline. Vancomycin trough concentrations were obtained from serum collected 1 hour before administration of a dose at steady-state concentration. Steady-state concentration was considered to be achieved after at least 3 doses of vancomycin. Individual pharmacokinetic parameters were estimated for each study participant by ADAPT 5 software (Biomedical Simulations Resource, Los Angeles, California) based on patient-specific CrCl, vancomycin dosing, and drug concentration history. Vancomycin serum concentrations were fitted to a 2- compartment volume-clearance model using a maximum a posteriori probability bayesian approach to parameterize individual patient pharmacokinetic profiles The mean daily predicted AUC 24 /MIC values were used for the analysis in situations where renal function was changing. Once a patient s pharmacokinetic parameters were estimated based on vancomycin serum concentrations, the steady-state AUC 24 /MIC was computed based on the patient s fitted vancomycin clearance (calculated from the patient s CrCl and body weight) and the isolate MIC. Statistical Analysis The distributions of quantitative variables were examined for normality using Kolmogorov-Smirnov tests. Bivariate analyses were conducted comparing the distribution of covariates among individuals with vancomycin using the Fisher exact test for dichotomous variables and the Mann-Whitney test for continuous variables. Variables with P o 0.20 were considered in multivariable analyses as potential confounders, along with other variables associated with the outcome of interest of time to clearance of bacteremia, including time to vancomycin initiation and time to a therapeutic vancomycin plasma concentration. Time to clearance of bacteremia was modeled as a function of vancomycin AUC 24 /MIC and covariates from the bivariate analysis using a Cox proportional hazards model. Vancomycin AUC 24 /MIC did not indicate a linear relationship with the log (hazard) and was therefore dichotomized using a classification and regression tree analysis (CART) derived breakpoint. CART is a set of analytic methods used to identify subgroups within a particular population that share similar characteristics that influence the dependent variable. 25 Standard model building techniques were used, including backwards and forwards stepping and testing for the assumption of proportional hazards. All statistical analyses were performed in SAS statistical software, version 9.3 (SAS Institute, Cary, North Carolina), except CART, which was performed in SYSTAT statistical software, version 13 (Systat Software, Chicago, Illinois). Unless otherwise specified, statistical tests are 2-sided with α ¼.05. RESULTS During the study period, 131 patients were identified for potential inclusion. Of these, 61 patients were excluded for the following reasons: 24 were receiving dialysis, 16 were transferred from an outside hospital with vancomycin, 10 were duplicate patients, 5 patients did not have radiographic findings supporting 774 Volume 35 Number 6

4 K.M. Gawronski et al. the diagnosis of osteomyelitis, 3 did not receive at least 3 doses of vancomycin, and 3 had polymicrobial bacteremia. Of the remaining 70 patients who met inclusion criteria, 59 had complete data available for review and were included in the analysis. Baseline patient demographic and vancomycin-associated characteristics are listed in Table I. Overall, the mean (SD) age was 54 (16) years, and 59% of the cohort was male, with a mean (SD) Charlson score of 4 (2). Eighty-three percent of patients were receiving care from a medical service, and 27% were admitted to the intensive care unit. Ninety-five percent of patients had an infectious diseases consultation, which was obtained within a mean (SD) of 2 (3) days from the day of the first positive blood culture result. Source removal was achieved in 39% of patients, most commonly by amputation. Mean (SD) AUC 24 /MIC was 301 (157), with a mean serum trough concentration of 19.8 (5.9) μg/ml. A goal serum trough concentration of 15 to 20 μg/ml was achieved within 4 (2) days. Nephrotoxicity was observed in 12% of patients receiving vancomycin, with a mean trough concentration of 19.9 μg/ml. Of the patients who developed nephrotoxicity, 57% had a trough 420 μg/ml; however, no patient had a trough 423 μg/ml. No patient required dialysis. Fifty-eight percent of the cohort had a vancomycin MIC 41 μg/ml. On further investigation, CART analysis identified an AUC 24 /MIC of 293 as the breakpoint that provides the greatest difference in time to clearance of bacteremia. The characteristics and clinical outcomes are listed in Table II. Time to microbiological clearance was 2 days shorter when the target AUC 24 /MIC of 4293 was achieved (4 [2] days vs 6 [3] days; P ¼ 0.01). Mean (SD) hospital LOS was 14 (7) versus 19 (14) days in patients with an AUC 24 /MIC 4293 or 293, respectively (P ¼ 0.22). Mean (SD) infection- Table I. Characteristics of patients with concomitant MRSA bacteremia and MRSA osteomyelitis who received 3 doses of vancomycin. Characteristic Finding* (n ¼ 59) Male, % 59 Age, y 54 (16) Table I (continued) Characteristic Finding* (n ¼ 59) 18 Charlson score 4 (2) Admitted to medical service, % 83 LOS, d 17 (12) Infection-related LOS, d 16 (12) ICU admission, % 27 ICU LOS, d 3 (8) ID consultation, % 95 Time to ID consultation from first 2 (3) positive blood culture result, d Hospital day of first positive blood 1 (2) culture result Time to clearance of bacteremia, d 5 (3) Prior vancomycin exposure (within days), % Other anti-mrsa agent exposure (within 90 days), % Source removal 39 Time from blood culture draw to 11 (16) vancomycin administration, h Trough, g/ml 19.8 (5.9) Trough attainment, % g/ml g/ml g/ml 46 Time to trough of g/ml, d 4 (2) Duration of inpatient vancomycin 14 (8) therapy, d MIC determined by Etest, % 85 MIC 41 g/ml, % 58 MIC 41 g/ml determined by 64 Etest, % AUC 24 /MIC 301 (157) AUC 24 /MIC 4293, % 39 Nephrotoxicity, % 12 ICU ¼ intensive care unit; ID ¼ infectious disease; LOS ¼ length of stay; MRSA ¼ methicillin-resistant Staphylococcus aureus. *Data presented as mean (SD) unless otherwise specified. Anti-MRSA agents included clindamycin, daptomycin, linezolid, tigecycline, and trimethoprimsulfamethoxazole. An increase of 40.5 mg/dl (or a 50% increase) in serum creatinine above baseline. June

5 Clinical Therapeutics Table II. Characteristics of patients with concomitant MRSA bacteremia and MRSA osteomyelitis based on AUC 24 /MIC. AUC 24 /MIC Characteristic* r293 (n ¼ 36) 4293 (n ¼ 23) P Time to clearance of bacteremia, d 6 (3) 4 (2) LOS, d 19 (14) 14 (7) 0.22 Infection-related LOS, d 18 (14) 13 (6) 0.25 Treatment failure, % Required change from vancomycin therapy, % Recurrent MRSA bacteremia and readmission (within 90 days), % All-cause readmission (within 30 days of discharge), % All-cause mortality, % Trough, g/ml 19.7 (6.2) 19.9 (5.3) 0.52 LOS ¼ length of stay; MRSA ¼ methicillin-resistant Staphylococcus aureus. *Data presented as mean (SD) unless otherwise specified. related LOS was 13 (6) versus 18 (14) days in patients with an AUC 24 /MIC 4293 or 293, respectively (P ¼ 0.25). In the cohort with an AUC 24 /MIC 293, we observed a higher percentage of treatment failures (19% vs 9%, P ¼ 0.46), required a change from vancomycin to an alternative anti-mrsa agent (25% vs 13%, P ¼ 0.33), recurrent episodes of MRSA bacteremia and hospital readmission within 90 days (39% vs 17%, P ¼ 0.09), and all-cause 30 day readmissions (11% vs 9%, P ) compared with patients who achieved an AUC24/MIC 4293, although no variable reached statistical significance. All-cause mortality was 2%. In the cohort with an isolate MIC 41 μg/ml, only 9% of patients were able to achieve an AUC 24 /MIC 4293 (P ¼ ). On univariate analysis, mean (SD) time to clearance of bacteremia was shorter when the AUC 24 /MIC 4293 (4 [2] days vs 6 [3] days, P ¼ 0.01). This relationship is consistent with additional univariate analysis determined by the Kaplan-Meier curve illustrated in the Figure (log-rank P ¼ 0.007). Although not statistically significant in univariate analyses, in the AUC 24 /MIC 293 group there was a trend toward older age (57 [17] years vs 50 [14] years, P ¼ 0.14), longer mean days spent in the intensive care unit (4 [10] vs 1 [3], P ¼ 0.19), and a higher percentage of patients with recurrent infections (39% vs 17%, P ¼ 0.09) compared with the cohort who achieved an AUC 24 /MIC On multivariate analysis, after controlling for clinical characteristics associated with time to clearance of bacteremia, AUC 24 /MIC was found to be the only independent predictor of time to clearance of bacteremia (hazard ratio ¼ 2.6; P ¼ ). These results indicate that individuals with an AUC 24 /MIC 4293 clear their bacteremia 4 25 times faster than those with an AUC 24 /MIC 293. DISCUSSION In this study, we aimed to identify the AUC 24 /MIC associated with time to microbiological clearance among Event Probability Kaplan-Meier Plot Time to Microclearance, d AUC/MIC > 293 AUC/MIC < 293 Figure. Kaplan-Meier plot of time to microbiological clearance based on AUC 24 /MIC Volume 35 Number 6

6 K.M. Gawronski et al. patients with complicated MRSA bacteremia and associated MRSA osteomyelitis. Using CART analysis, patients who received a vancomycin regimen that resulted in an AUC 24 /MIC ratio of 293 were at increased risk of delayed microbiological clearance (P ¼ 0.01) and had a 2-day increase in time to microbiological clearance. In the multivariate analysis, AUC 24 /MIC was the only independent predictor of time to clearance of bacteremia (hazard ratio ¼ 2.6; P ¼ ). Hospital and infectionrelated LOS was 5 days longer in the group not able to achieve an AUC 24 /MIC Increases in vancomycin treatment failure, change from vancomycin therapy, recurrent MRSA bacteremia within 90 days, and hospital readmissions within 30 days were higher in the cohort not able to achieve an AUC 24 /MIC To our knowledge, this is the first study to examine the relationship between vancomycin AUC 24 /MIC and time to microbiological clearance in patients with concomitant MRSA bacteremia and MRSA osteomyelitis. Maximizing antimicrobial therapy by targeting the optimal pharmacodynamic parameter is critical in these patients because source removal is not always achievable. In the present study, only 39% of patients were able to have the source of infection removed. Previous investigations have sought to determine the optimal pharmacodynamic parameter in patients with MRSA infections and have identified a potential association between AUC 24 /MIC with microbiological and clinical outcomes. Moise-Broder and colleagues 17 determined that in patients with Saureuspneumonia an AUC 24 /MIC 400 was correlated with clinical and bacteriologic response. In patients who achieved an AUC 24 /MIC 400, median time to bacterial eradication was 10 versus 14 days (P ¼ ) in those with an AUC 24 /MIC o400. Jeffres and colleagues 1 evaluated patients with MRSA health care associated pneumonia to determine vancomycin pharmacokinetic indices associated with mortality. They did not find an association between vancomycin AUC 400 and hospital mortality; however, the optimal AUC 24 /MIC ratio was not evaluated. These results may have limited application because they were conducted only in patients with pneumonia. Brown and colleagues 18 performed a CART analysis to determine the AUC 24 /MIC associated with attributable mortality in patients with complicated MRSA bacteremia or infective endocarditis. They found that a much lower breakpoint, an AUC 24 / MIC o211, was independently associated with attributable mortality. In our evaluation, we found that patients remained bacteremic 42.5 times longer when an AUC 24 /MIC 4293 was not achieved. This is most likely due to 58% of the population having an isolate with a MIC 41 μg/ml. In addition, patients who did not achieve an AUC 24 /MIC 4293 had a 5-day increase in both hospital and infection-related LOS. Although not statistically significant, in the current health care environment, prolonged hospitalization results in significant health care costs. Patients who did not achieve an AUC 24 /MIC 4293 also had more treatment failures, required a change from vancomycin, and had an increase in both recurrent episodes of MRSA bacteremia and readmission within 90 days and hospital readmissions within 30 days. With median hospital charges per episode of MRSA bacteremia in excess of $25,000 and major financial penalties for 30-day hospital readmissions, this is a significant economic burden that emphasizes the importance of selecting appropriate empiric therapy and optimizing dosing strategies. 8,10,11,26 Trough concentration is used as a surrogate marker for vancomycin efficacy. 12 There was not a statistically significant relationship between AUC 24 /MIC and trough concentration on univariate analysis. A mean trough concentration of 19 μg/ml was found in both groups (P ¼ 0.52). This finding is similar to the study published by Patel and colleagues. 27 An AUC should be calculated for all patients receiving vancomycin to ensure optimal dosing. We identified that patients with high MIC isolates 41 μg/ml were less likely to achieve an AUC 24 /MIC Patel and colleagues 27 recently examined the association among various vancomycin dosing regimens, MIC, and the probability of achieving target AUC 24 /MIC 400 using Monte Carlo simulation. The investigators found that even with aggressive vancomycin dosing of 2 g every 12 hours, if the isolate MIC was 2 μg/ml, the probability of achieving an AUC 24 /MIC ratio 400 was only 57%. This is in contrast to near 100% achievement when the isolate MIC 1 μg/ml. 27 A similar trend was observed in our evaluation, with only 9% of patients able to achieve an AUC 24 /MIC 4293 in isolates with an MIC 41 μg/ml. Patients with an MIC 1 μg/ml were able to achieve this target ratio 91% of the time. This finding suggests that in patients with MIC values 41 μg/ml vancomycin may no longer be the preferred agent because of the inability to achieve target pharmacodynamic parameters. 27,28 June

7 Clinical Therapeutics In targeting aggressive pharmacodynamic parameters, high-dose vancomycin therapy has been associated with the development of nephrotoxicity. 29 Patel and colleagues 27 reported that increasing vancomycin total daily doses are directly related to a higher risk of nephrotoxicity, and regimens of 4 g/d were associated with a 35% chance of nephrotoxicity. In our evaluation, we observed a 12% nephrotoxicity rate. No patient who experienced nephrotoxicity had a trough 423 μg/ml or required dialysis. There are limitations associated with our study. This was a single-center retrospective study with a limited sample size. We only included patients with MRSA bacteremia and osteomyelitis, which may limit its application to all patient populations. During the entire study period, vancomycin loading doses were not standard of practice. On the basis of vancomycin therapeutic monitoring guidelines, aggressive dosing strategies are recommended to achieve steady-state trough concentrations of 15 to 20 μg/ml and target AUC 24 /MIC ,15 Although patients in this study received vancomycin according to institutional guidelines, we have since modified the regimen to include a vancomycin loading dose of 20 mg/kg, with a 2-g maximum dose. In addition, pharmacists have recently been granted prescriptive authority at OSUWMC to modify vancomycin regimens and order serum drug levels without prior physician approval. Not all patients had consecutive blood cultures. Finally, MICs were determined by 2 different methods because of changes in the microbiology laboratory during the 5-year study. Importantly, studies have found that MICs determined by Microscan tend to be lower than those determined by Etest. In our study, 85% of MICs were determined by Etest. We recommend the determination of MIC via Etest instead of by automated systems for MRSA blood isolates. CONCLUSION In patients with concomitant MRSA bacteremia and MRSA osteomyelitis, we found a 42.5-fold increase in time to microbiological clearance in the cohort of patients who were unable to achieve a vancomycin AUC 24 /MIC Hospital and infection-related LOS was longer and recurrence of MRSA bacteremia and subsequent hospital readmission within 90 days and 30-day all-cause hospital readmissions were higher in this cohort. Trough concentration was not well correlated with AUC 24 /MIC. In addition, only 9% of patients were able to achieve an AUC 24 /MIC 4293 when the vancomycin MIC was 41 μg/ml. In patients with concomitant MRSA bacteremia and MRSA osteomyelitis treated with vancomycin and a MIC 41 μg/ml, pharmacodynamic parameters, specifically AUC 24 /MIC, should be optimized or alternative therapies should be considered because source removal is rarely able to be achieved. ACKNOWLEDGEMENTS No individuals besides the authors contributed to the writing or revising of the manuscript. No monetary support was provided to the authors. All authors contributed equally. No study sponsors were involved in the study. CONFLICTS OF INTEREST The authors have indicated that they have no conflicts of interest regarding the content of this article. REFERENCES 1. Jeffres MN, Isakow W, Doherty JA, et al. Predictors of mortality for methicillin-resistant Staphylococcus aureus healthcare-associated pneumonia: specific evaluation of vancomycin pharmacokinetic indices. Chest. 2006;130: Maor Y, Hagin M, Belausov N, Keller N, Ben-David D, Rahav G. Clinical features of heteroresistant vancomycinintermediate Staphylococcus aureus bacteremia versus those of methicillin-resistant S. aureus bacteremia. J Infect Dis. 2009;199: National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2003, issued August Am J Infect Control. 2003;31: Chua T, Moore CL, Perri MB, et al. Molecular epidemiology of methicillin-resistant Staphylococcus aureus bloodstream isolates in urban Detroit. J Clin Microbiol. 2008;46: Lodise TP, McKinnon PS. Burden of methicillin-resistant Staphylococcus aureus: focus on clinical and economic outcomes. Pharmacotherapy. 2007;27: Klevens RM, Morrison MA, Nadle J, et al. Invasive methicillin-resistant Staphylococcus aureus infections in the United States. JAMA. 2007;298: Hidayat LK, Hsu DI, Quist R, Shriner KA, Wong-Beringer A. High-dose vancomycin therapy for methicillin-resistant Staphylococcus aureus infections: efficacy and toxicity. Arch Intern Med. 2006;166: Cosgrove SE, Qi Y, Kaye KS, Harbarth S, Karchmer AW, Carmeli Y. The impact of methicillin resistance in Staphylococcus aureus bacteremia on patient outcomes: mortality, length of stay, and hospital charges. Infect Control Hosp Epidemiol. 2005;26: Volume 35 Number 6

8 K.M. Gawronski et al. 9. Corey GR. Staphylococcus aureus bloodstream infections: definitions and treatment. Clin Infect Dis. 2009;48(Suppl 4):S254 S Ben-David D, Novikov I, Mermel LA. Are there differences in hospital cost between patients with nosocomial methicillin-resistant Staphylococcus aureus bloodstream infection and those with methicillin-susceptible S. aureus bloodstream infection? Infect Control Hosp Epidemiol. 2009;30: Bauer KA, West JE, Balada-Llasat J- M, Pancholi P, Stevenson KB, Goff DA. An antimicrobial stewardship program s impact. Clin Infect Dis. 2010;51: Rybak MJ, Lomaestro BM, Rotschafer JC, et al. Therapeutic monitoring of vancomycin in adults summary of consensus recommendations from the American Society of Health- System Pharmacists, the Infectious Diseases Society of America, and the Society of Infectious Diseases Pharmacists. Pharmacotherapy. 2009;29: Rybak MJ. The pharmacokinetic and pharmacodynamic properties of vancomycin. Clin Infect Dis. 2006;42 (Suppl 1):S35 S Berthoin K, Ampe E, Tulkens PM, Carryn S. Correlation between free and total vancomycin serum concentrations in patients treated for Gram-positive infections. Int J Antimicrob Agents. 2009;34: Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the infectious diseases society of america for the treatment of methicillinresistant Staphylococcus aureus infections in adults and children. Clin Infect Dis. 2011;52:e18 e Stevens DL. The role of vancomycin in the treatment paradigm. Clin Infect Dis. 2006;42(Suppl 1):S51 S Moise-Broder PA, Forrest A, Birmingham MC, Schentag JJ. Pharmacodynamics of vancomycin and other antimicrobials in patients with Staphylococcus aureus lower respiratory tract infections. Clin Pharmacokinet. 2004;43: Brown J, Brown K, Forrest A. Vancomycin AUC24/MIC ratio in patients with complicated bacteremia and infective endocarditis due to methicillin-resistant Staphylococcus aureus and its association with attributable mortality during hospitalization. Antimicrob Agents Chemother. 2012;56: Garazzino S, Aprato A, Baietto L, et al. Glycopeptide bone penetration in patients with septic pseudoarthrosis of the tibia. Clin Pharmacokinet. 2008;47: Graziani AL, Lawson LA, Gibson GA, Steinberg MA, MacGregor RR. Vancomycin concentrations in infected and noninfected human bone. Antimicrob Agents Chemother. 1988;32: Rodvold KA, Blum RA, Fischer JH, et al. Vancomycin pharmacokinetics in patients with various degrees of renal function. Antimicrob Agents Chemother. 1988;32: Rotschafer JC, Crossley K, Zaske DE, Mead K, Sawchuk RJ, Solem LD. Pharmacokinetics of vancomycin: observations in 28 patients and dosage recommendations. Antimicrob Agents Chemother. 1982;22: Healy DP, Polk RE, Garson ML, Rock DT, Comstock TJ. Comparison of steady-state pharmacokinetics of two dosage regimens of vancomycin in normal volunteers. Antimicrob Agents Chemother. 1987;31: Golper TA, Noonan HM, Elzinga L, et al. Vancomycin pharmacokinetics, renal handling, and nonrenal clearances in normal human subjects. Clin Pharmacol Ther. 1988;43: Lemon SC, Roy J, Clark MA, Friedmann PD, Rakowski W. Classification and regression tree analysis in public health: methodological review and comparison with logistic regression. Ann Behav Med. 2003;26: Axon R, Williams MV. Hospital readmission as an accountability measure. JAMA. 2011;305: Patel N, Pai MP, Rodvold KA, Lomaestro B, Drusano GL, Lodise TP. Vancomycin: we can t get there from here. Clin Infect Dis. 2011;52: Moore CL, Osaki-Kiyan P, Haque NZ, Perri MB, Donabedian S, Zervos MJ. Daptomycin versus vancomycin for bloodstream infections due to methicillin-resistant Staphylococcus aureus with a high vancomycin minimum inhibitory concentration: a case-control study. Clin Infect Dis. 2012;54: Wong-Beringer A, Joo J, Tse E, Beringer P. Vancomycin-associated nephrotoxicity: a critical appraisal of risk with high-dose therapy. Int J Antimicrob Agents. 2011;37: Address correspondence to: Karri A. Bauer, PharmD, BCPS, The Ohio State University Wexner Medical Center, Department of Pharmacy, Room 368, Doan Hall, 410 W 10th Ave, Columbus, OH Karri. bauer@osumc.edu June