Cost-Effectiveness of Daptomycin versus Vancomycin and Gentamicin for Patients with Methicillin-Resistant Staphylococcus aureus Bacteremia and/or Endocarditis
- Reprints or correspondence: Dr Sujata M. Bhavnani, Institute for Clinical Pharmacodynamics, Ordway Research Institute, 43 British American Blvd, Latham, NY 12110 (SBhavnani-icpd{at}ordwayresearch.org).
Abstract
Background. Methicillin-resistant Staphylococcus aureus (MRSA) is an increasingly common cause of bacteremia and endocarditis. The cost-effectiveness (CE) of daptomycin was compared with that of vancomycin-gentamicin in patients with MRSA bacteremia with or without endocarditis.
Methods. With use of data from an open-label, randomized, double-blind study comparing daptomycin with vancomycin-gentamicin in the aforementioned patient population, 3 cost strata were considered: (1) study drug acquisition (daptomycin, $0.37/mg; vancomycin, $7/g; and gentamicin, $0.12/mg); (2) stratum 1 plus the cost of therapy for treatment failures and adverse events, therapeutic drug monitoring, and preparation and administration of all medications; and (3) stratum 2 plus hospital bed costs. Drug costs were based on mean wholesale price, with other costs based on those for a typical community hospital. Cost-effectiveness ratios were calculated as cost divided by proportion of successes. Sensitivity analyses were performed by varying the study drug cost.
Results. Forty-five (20 successes) and 44 (14 successes) patients received daptomycin and vancomycin-gentamicin, respectively. The respective median cost-effectiveness ratios for daptomycin and vancomycin-gentamicin for each cost stratum were as follows: $4082 (range, $1062-$13,893) and $560 (range, $66-$1649) for stratum 1 (P<.001); $4582 (range, $1109-$21,882) and $1635 (range, $163-$33,444) for stratum 2 (P=.026); $23,639 (range, $6225-$141,132) and $26,073 (range, $5349-$187,287) for stratum 3 (P=.82). Sensitivity analyses indicated that if the cost of vancomycin was $0, strata 3 cost-effectiveness ratios did not differ ($23,639 and $25,668, respectively; P =.85). Similar results between groups were seen among patients with bacteremia.
Conclusions. When all costs of therapy were considered, the cost-effectiveness of daptomycin and vancomycin-gentamicin was similar, even if the cost of vancomycin was $0.
Treatment options for bacteremia and endocarditis are limited, especially those arising from methicillin-resistant S. aureus (MRSA). For decades, vancomycin has been the mainstay for standard therapy in such cases. Given the increasing prevalence of MRSA, including among community-acquired infections, vancomycin is often the choice of therapy if MRSA is documented or suspected. However, the use of this agent has been associated with suboptimal outcomes [1-4]. Results of a meta-analysis of studies evaluating S. aureus bacteremia demonstrated a significantly higher mortality rate associated with patients with MRSA bacteremia, compared with methicillin-susceptible S. aureus (MSSA) bacteremia [5].
A randomized, clinical trial evaluated daptomycin, a lipopeptide antibiotic that is rapidly bactericidal against MRSA, compared with both semisynthetic penicillins and vancomycin for the treatment of S. aureus bacteremia and infective endocarditis; the trial demonstrated noninferiority of daptomycin [6]. Data from this pivotal trial were the basis for the approval of daptomycin in the United States for treatment of S. aureus bacteremia, including right-side infective endocarditis.
Cost-effectiveness (CE) analyses of different therapeutic agents can provide a source of useful data for decision support. Use of data demonstrating differences in efficacy and/or resource consumption between different therapeutic agents provides an opportunity to assess overall costs above and beyond drug acquisition costs and to potentially reduce institutional cost while maintaining a high standard of patient care. The analyses described herein compared the CE, from the institution's perspective, of daptomycin with the CE of vancomycin-gentamicin in patients with MRSA bacteremia, with or without endocarditis, with use of data from the aforementioned study [6].
Methods
Study design. Data were obtained from an open-label, randomized, clinical trial involving patients with S. aureus bacteremia, with or without endocarditis [6]. In brief, patients were randomized (ratio, 1:1) to receive either daptomycin (6 mg/kg intravenously) once daily or standard therapy with either vancomycin (1 g intravenously every 12 h, with dose adjustment based on renal function) for MRSA infection or an antistaphylococcal penicillin (2 g intravenously every 4 h) for MSSA infection. The duration of therapy was determined by the investigator and was assessed on the basis of the working diagnosis. Patients with uncomplicated and complicated bacteremia received therapy for a minimum of 10–14 and 28–42 days, respectively. Patients with uncomplicated MSSA right-side endocarditis received therapy for a minimum of 14–28 days, whereas those with complicated right-side endocarditis received therapy for a minimum of 28–42 days. Study medication was administered for a minimum of 28–42 days for patients with left-side endocarditis.
All patients assigned to receive standard treatment and patients with left-side endocarditis assigned to receive daptomycin also received gentamicin (1 mg/kg intravenously every 8 h) for the first 4 days.
Patient population. The analyses conducted included all patients with MRSA bacteremia alone and/or with endocarditis who had received at least 1 dose of study medication (modified MRSA intent-to-treat population). Eligible patients included hospitalized patients who were aged ⩾18 years and who had ⩾1 blood culture positive for S. aureus ⩽2 days before they initiated study medication. Patients were excluded if they had a creatinine clearance ⩽30 mL/min, known osteomyelitis, polymicrobial bacteremia, or pneumonia, as described by Fowler et al [6].
Clinical response evaluation. Patients were evaluated at baseline, at the end of therapy and 42 days after the end of therapy (test-of-cure). Clinical response was categorized as success or failure. Failure was defined as clinical failure, microbiological failure, death, failure to perform blood culture, receipt of potentially effective nonstudy antibiotics, or premature discontinuation of the study medication because of clinical failure, microbiological failure, or an adverse event (AE). Nonevaluable patients were considered to have experienced failure.
Cost data. Resource and outcome data, including the following, were extracted for the analysis population: study medication use; clinical response at Test-of-Cure; subsequent antimicrobial therapy administered for treatment failures; therapeutic drug monitoring for vancomycin or gentamicin; AEs with possible or probable association to study drug administration; procedures, laboratory assessments, or medications associated with AE management (ie, diagnosis and/or treatment); and the duration and location of hospital stay. On the basis of these data, 3 strata of costs were computed for each patient. Stratum 1 costs considered the drug acquisition cost for daptomycin, vancomycin, and gentamicin, which were based on unit costs of $0.365/mg, $7.00/g, and $0.12/mg [7], respectively.
Stratum 2 costs included stratum 1 costs plus the cost of antimicrobial therapy administered for treatment failures, therapeutic drug monitoring for vancomycin-gentamicin (during administration of study medication and/or follow-up therapy, including concomitant treatment, for treatment failures), medications or procedures (including laboratory assessments) associated with the management of AEs with possible or probable association with study drug administration, as well as costs associated with intravenous drug admixture and oral drug administration for both study drug and all drugs administered for either treatment failure and/or management of AEs. Drug costs for treatment failure and/or AE management were based on mean wholesale price [7]. The cost of preparation and administration of intravenous and oral agents was assumed to be $7.00 and $2.25 per dose, respectively [8-13]. A cost of $10 was used to represent the combined cost of obtaining a blood sample for assay and determining the drug concentration of that sample for vancomycin or gentamicin. In the event of follow-up therapy with these 2 agents, in which therapeutic drug monitoring data had not been collected, the frequency of monitoring was inferred according to standards of clinical practice. Procedure costs were based on data from a typical community-based hospital. Costs for medications and procedures were considered from the beginning of study medication use to 42 days after the end of therapy or until hospital discharge, whichever occurred first.
Stratum 3 costs included stratum 2 costs plus hospital stay per diem costs, which were based on the duration of hospital stay beginning with the first day of study drug administration and ending with the last day of study drug administration or hospital discharge, whichever occurred first. For those patients who were not discharged after the end of study drug administration and who received medications for treatment failure or medications and/or procedures for AE management, stratum 3 costs also included the hospital days associated with this additional care. Because patient location was not documented for these additional hospital days, costs were based on the duration of stay in a hospital ward. Hospital stay per diem costs, which reflected those of a typical community-based hospital, were as follows: intensive care unit, $1300, step-down, $800, and hospital ward, $550.
Statistical analyses. Comparisons of continuous and categorical data for the groups of patients receiving daptomycin and vancomycin-gentamicin (ie, all patients with MRSA infection and the subset of those with bacteremia) were made using the Mann-Whitney U test and the χ2test (or the Fisher exact test), respectively. The probability of a type I error of 0.05 was used to ascertain statistical significance of all comparisons. CE ratios for each of the 3 cost strata were calculated by dividing each cost level by the proportion of clinical successes associated with either the daptomycin or vancomycin-gentamicin group. To compare groups with respect to CE ratios in a manner accounting for the variability in both cost and clinical response, standard errors of the difference in group medians were estimated using bootstrap resampling. A sensitivity analysis, in which the unit cost of vancomycin was assumed to be negligible (eg, $0/g), was performed.
Results
A total of 124 and 122 patients were randomized to receive daptomycin and standard therapy, respectively. Of these, 120 and 115 patients received at least 1 dose of study medication, respectively, and comprised the modified intent-to-treat population. A total of 45 and 44 patients in each group, respectively, had MRSA bacteremia, with or without endocarditis. As shown in Table 1, these 2 groups were similar with respect to demographic characteristics, including the proportion of patients within each diagnostic group. Among patients with uncomplicated or complicated bacteremia, demographic characteristics were also similar between groups.
Figure 1
Median cost-effectiveness ratios for the whole study population (A) and for patients with bacteremia (B).
Table 1
Comparison of Treatment Group Demographic Characteristics of Patients with Methicillin-Resistant Staphylococcus aureus
Of the patients who received standard therapy, all received vancomycin. All but 4 vancomycin-treated patients with complicated bacteremia received at least 1 dose of gentamicin, with a median duration of therapy of 4.5 days (range, 1–7). No patients in the daptomycin group, including those with left-side endocarditis, received gentamicin. Eight patients (3 who received daptomycin and 5 who received vancomycin-gentamicin) left the hospital during treatment with study drug but returned to hospital before the end of therapy.
The median total duration of study drug therapy was 14 days (range, 3–42 days) for the daptomycin-treated patients and 15 days (range, 2–44 days) for vancomycin-treated patients (P=.56). The median duration of study drug therapy in the hospital was 12 days (range, 3–35 days) and 10 days (range, 2–35 days; P=.53), respectively. During this period, 33% and 18% (P=.15) of patients in the daptomycin and vancomycin-gentamicin groups, respectively, required stay in the intensive care unit. Within 24 h after receipt of the last dose, 19 (42.2%) of 45 daptomycin-treated patients and 16 (36.4%) of 44 vancomycin-gentamicin-treated patients had been discharged from the hospital to home or to a residential care facility.
Although more patients who received daptomycin had a successful outcome, compared with patients who received vancomycin, clinical response at the test-of-cure visit was not statistically significantly different between groups. Successful clinical responses were observed in 20 (44.4%) of 45 and 14 (31.8%) of 44 of daptomycin- and vancomycin-gentamicin-treated patients, respectively (treatment difference 12.6%; 95% confidence interval [CI], −7.38% to 32.6%). Similarly, among patients with bacteremia, successful clinical responses were observed in 16 (50.0%) of 32 and 11 (33.3%) of 33 patients, respectively (treatment difference, 16.7%; 95% CI, −6.97% to 40.3%). A total of 11 and 4 patients in the daptomycin and vancomycin-gentamicin groups, respectively, died before or by the test-of-cure visit.
Follow-up therapy for treatment failure was documented on the case report forms for 15 and 13 patients in the daptomycin and vancomycin-gentamicin groups, respectively. In the daptomycin group, these regimens commonly consisted of vancomycin (n=13) and rifampin (n=8). In the vancomycin-gentamicin group, these regimens commonly consisted of vancomycin (n=7), rifampin (n=6), and linezolid (n=6). The median drug acquisition cost for this therapy was $242 (range, $33-$6535) and $1179 (range, $38-$9740) for the daptomycin and vancomycin-gentamicin groups, respectively (P=.12). In addition, the median duration of hospital stay associated with this follow-up therapy was 20 days (range, 3–60 days) and 26 days (range, 3–74 days) for the daptomycin and vancomycin-gentamicin groups, respectively (P=.51). For 9 and 14 patients in the daptomycin and vancomycin-gentamicin groups, respectively, for whom follow-up therapy was not documented but for whom clinical failure was declared, common reasons for this outcome included death, an AE, and the lack of available cultures to document clearance of bacteremia rather than direct study drug treatment failure. In the remaining 1 and 3 daptomycin- and vancomycin-gentamicin-treated patients, respectively, clinical response was considered indeterminate.
During study drug therapy, 43 patients in the vancomycin-gentamicin group required therapeutic drug monitoring. The median estimated cost for these patients was $30 (range, $10-$220). During follow-up therapy for treatment failure, 13 and 7 patients in the daptomycin and vancomycin-gentamicin groups, respectively, required therapeutic drug monitoring. The median estimated cost among these patients was $40 (range, $10-$100) and $60 (range, $10-$110) for these 2 groups, respectively (P=.42). The median total cost of therapeutic drug monitoring was $0 (range, $0-$100) and $30 (range, $10-$220) among all patients in the daptomycin and vancomycin-gentamicin groups, respectively (P<.001).
A total of 6 and 9 AEs that had possible or probable associations with study drug administration and that were associated with medications and/or procedures for the management of the AEs were observed in the daptomycin and vancomycin-gentamicin groups, respectively. In the daptomycin group, the most common AE requiring management was candidal infection, which occurred in 3 of 6 patients. In the vancomycin-gentamicin group, the AE most frequently requiring management was renal impairment or failure (4 of 9 patients). In these 4 cases, each patient received 4–6 doses of gentamicin. The median cost of diagnosis and/or management of all of these events for each treatment group was $91 (range, $9-$2530) and $439 (range, $4-$1156), respectively (P=.35).
The median cost of the preparation and administration of intravenous and oral agents associated with study drug and follow-up therapy for treatment failures and AEs was $98 (range, $21-$1075) and $194 (range, $21-$1057) for the daptomycin- and vancomycin-gentamicin groups, respectively (P<.001). Unit cost assumptions and the number of units consumed during the clinical trial, which support the aforementioned between-group comparisons, are summarized in Table 2. Descriptive statistics summarizing the cost by stratum for all patients in each treatment group are presented in Table 3. Although median strata 1 and 2 costs statistically favored vancomycin, median stratum 3 costs were not significantly different between the daptomycin and vancomycin-gentamicin groups.
Table 2
Summary of Unit-Cost Assumptions Used in Economic Analyses and the Actual Number of Units Consumed during the Clinical Trial or the Total Cost Associated with Each Resource
Table 3
Costs for All Patients Who Were Treated with Daptomycin or Vancomycin for Each Economic Stratum
Figure 1Ashows the median CE ratios for daptomycin and vancomycin-gentamicin for the 3 cost strata. Sensitivity analyses demonstrated that the median CE ratio between groups for stratum 3 was not statistically different ($23,639 vs $25,668; P=.85) when the cost of vancomycin was $0.
As shown in Table 4and Figure 1B, similar differences between groups in costs and CE ratios were seen for the subset of patients with bacteremia. Sensitivity analyses based on this subset of patients demonstrated that the difference in the median CE ratio between groups for stratum 3 was not statistically different ($20,859 vs $24,244; P=.77) when the cost of vancomycin was $0.
Table 4
Costs for Patients with Bacteremia Who Were Treated with Daptomycin or Vancomycin-Gentamicin for Each Economic Stratum
Discussion
The objective of this analysis was to compare the CE of daptomycin and vancomycin-gentamicin in patients with MRSA bacteremia and/or endocarditis. These data demonstrated that when total inpatient treatment costs (stratum 3) were considered in the modified MRSA intent-to-treat population, vancomycin-gentamicin was not more cost-effective than daptomycin. Given that a typical course of therapy for patients in the vancomycin-gentamicin group (median, $178) was far less costly than that of daptomycin (median, $1814), this finding is somewhat counterintuitive. Although not statistically significant, clinical response rate differences, which favored daptomycin over vancomycin-gentamicin (daptomycin, 44%; vancomycin-gentamicin, 32%), represented one important reason for this finding. Moreover, stratum 2 costs associated with therapy for patients who experienced treatment failure (P=.12) and both the cost of preparation and administration of intravenous and oral agents (P<.001) and therapeutic drug monitoring (P<.001) among all patients in each group also tended to favor daptomycin over vancomycin-gentamicin.
There were 2 reasons that the drug acquisition costs associated with treatment failure tended to be higher in the vancomycin-gentamicin group than in the daptomycin group. First, there was a trend toward a higher cost for follow-up therapy for treatment failures among vancomycin-gentamicin-treated patients than among daptomycin-treated patients. Perhaps, not surprisingly, if a patient experienced treatment failure with vancomycin-gentamicin therapy, the follow-up regimen often included linezolid (6 [46%] of 13 patients), which is a relatively expensive agent. On the other hand, if a patient experienced treatment failure with daptomycin therapy, the follow-up regimen involved primarily vancomycin (13 [87%] of 15 patients), a relatively inexpensive agent. A longer median duration of follow-up therapy by 6 days in the vancomycin-gentamicin, compared with the daptomycin group, represented a second and important contribution to the higher cost of treatment failure in the vancomycin-gentamicin group.
Other stratum 2 costs also tended to be higher in the vancomycin-gentamicin versus the daptomycin group. Because vancomycin is typically administered twice per day and daptomycin once per day, infusion-related costs were greater in the vancomycin-gentamicin cohort. Because therapeutic drug monitoring is not recommended during daptomycin therapy, between-group differences for therapeutic drug monitoring costs favoring daptomycin were expected. Furthermore, the number of renal AEs requiring management was higher in the vancomycin-gentamicin group (4 of 9) than in the daptomycin group (0 of 6).
One goal of an economic analysis is to provide clinical and formulary decision-makers with a framework to make clinical and/or economic decisions. However, because of the differences in institution cost structures, generalizing such information such that it is useful to a specific institution or health care system is challenging. In an effort to provide the reader the ability to generalize both direct and indirect costs to their own institution, certain unit-costs and the associated total number of units consumed for studied patients are presented in Table 2.
There are several potential limitations of these analyses. First, the clinical trial described herein was designed to support regulatory approval efforts and, as such, was powered to support clinical rather than economic end points. Such a study may have been underpowered for the purpose of economic analyses. Second, clinical practice patterns may not have evolved enough to capture the true benefit of the innovator treatment, which, in this case, was daptomycin. In addition, with respect to vancomycin patterns of dosing and monitoring, clinical practice has changed over time. Administration of higher vancomycin doses, which have the potential for both increased efficacy and toxicity [14], may require more rigorous monitoring of vancomycin concentrations. Third, study physicians may have been more conservative in assessing treatment effect in this trial, because it involved a new therapeutic entity, which could have led to longer periods of treatment and hospitalization.
In conclusion, using data derived from a randomized, clinical trial that compared daptomycin with vancomycin-gentamicin in patients with bacteremia and/or endocarditis, we characterized the costs associated with MRSA therapy on 3 different economic strata. These cost data were used to calculate CE ratios, and a bootstrap resampling technique was then used to estimate standard errors of the difference in group medians, thereby accounting for variability in clinical response within groups. These analyses demonstrated that, although the drug acquisition costs of a typical vancomycin-gentamicin course of therapy were much lower than the costs for daptomycin, the vancomycin-gentamicin regimen was not more cost-effective when the total cost of inpatient care was considered. The 2 main reasons for this finding were differences in response rates and costs associated with treatment failures, which favored daptomycin over the vancomycin-gentamicin regimen. These data, which support daptomycin as a first-line agent in this patient population, are of particular interest in the context of the present era in which we see the continued erosion of vancomycin activity with the resultant need for more aggressive dosing and thus, an increased risk of toxicity.
Acknowledgments
We gratefully acknowledge Marilyn Campion, a statistical consultant for Cubist Pharmaceuticals, for her assistance in reviewing the manuscript and with data management.
Financial support. Cubist Pharmaceuticals.
Potential conflicts of interest. All authors: no conflicts.
- Received July 8, 2008.
- Accepted April 20, 2009.
- © 2009 Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
-
- Articles and Commentaries
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
