| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Department of Pharmacy and Therapeutics, University of Pittsburgh School of Pharmacy Department of Medicine, Division of Infectious Diseases, University of Pittsburgh Medical Center, Pittsburgh, PA
Correspondence to: Brian A. Potoski, PharmD, Falk Medical Building, Suite 3A, 3601 Fifth Ave, Pittsburgh, PA 15213; e-mail: potoskiba{at}upmc.edu
To the Editor:
We read with interest the study by Jeffres and colleagues (October 2006)1 investigating the relationship between vancomycin pharmacokinetic (PK) indices and outcome in patients with confirmed Staphylococcus aureus pneumonia. This study did not find a correlation between vancomycin area under the curve (AUC) 
400 µg/h/mL and hospital mortality in patients with S aureus pneumonia. While the authors acknowledge that previous clinical studies show an association between vancomycin PK indices and patient outcomes, they do not further clarify that the AUC/minimum inhibitory concentration (MIC) ratio and not AUC alone, as evaluated in the current study, has correlated better with patient outcomes in those studies.23 Although this correlation is not evidenced in all studies45 evaluating this relationship, the AUC/MIC ratio is the best parameter predicting vancomycin activity against S aureus in an animal infection model.6 The authors in the current study acknowledge the lack of isolate MIC values from study patients and further state that although Kirby Bauer zone sizes from isolates between survivors and nonsurvivors did not differ, this may not mean the MIC values were similar.7 In short, the MIC distribution in this study is not known. Undetermined MICs, however, could have resulted in significant AUC/MIC differences between groups that were not appreciated by study methodology.
The authors also question the recommendation of achieving vancomycin trough concentrations 15 µg/mL as a predictor of outcome, yet trough concentrations are a reasonable surrogate marker for AUC as reinforced by the authors’ calculations showing a significant correlation between trough levels and AUC values. Trough levels therefore may be of value in predicting an AUC/MIC ratio, again, provided the isolate MIC is known.
In conclusion, this study is severely limited by lack of MIC data that preclude the determination with certainty that an appropriate index has no correlation to mortality. We do agree with the authors’ assessment that there is a need for a prospective randomized study to assess vancomycin exposure on mortality in ICU patients. This should be conducted using a more appropriate and previously identified index, the AUC/MIC ratio.
Footnotes
Dr. Paterson is a member of the Speaker’s Bureau for Roche, AstraZeneca, Pfizer, Cubist, Merck, and Elan Pharmaceuticals, and has received research funding from AstraZeneca, Pfizer, Merck, and Elan Pharmaceuticals. Dr. Potoski is a member of the Speaker’s Bureau for Wyeth and Pfizer pharmaceuticals and has received research funding from Pfizer Pharmaceuticals. There are no grants or sources of support for either author that are related to the topic of this letter.
The authors have no conflicts of interest to disclose.
References
St. Anthony’s Medical Center Barnes-Jewish Hospital Washington University School of Medicine., St. Louis, MO
Correspondence to: Meghan Jeffres, PharmD, BCPS, Department of Pharmacy, St. Anthony’s Medical Center, 10010 Kennerly Rd, St. Louis MO 63128;email: jeffmn{at}samcstl.org
To the Editor:
We would like to thank Dr. Moine and colleagues for their interest in our article focusing on the outcomes of patients with bronchoscopically proven methicillin-resistant Staphylococcus aureus (MRSA) pneumonia treated with vancomycin. We appreciate their important concern regarding the limitations of the disk-diffusion method to identify vancomycin-intermediate S aureus (VISA) or heteroresistant VISA strains.1 Additionally, we also acknowledge the reports23 associating poor outcomes in patients with infections caused by MRSA that have a vancomycin minimum inhibitory concentration (MIC) > 1.5 µg/mL. However, the point that MRSA isolates with MIC values 2 µg/mL are highly prevalent varies from report to report and likely from institution to institution. The SENTRY Antimicrobial Surveillance Program database evaluated 35,458 S aureus isolates from 1998 to 2003 and determined both the vancomycin MIC of 50% of isolates and vancomycin MIC of 90% of isolates to be 1 µg/mL.1 The percentage of isolates with an MIC of 2 µg/mL was < 8% in each year evaluated, and this value was stable from year to year. More recently, the Surveillance Network data for 2005 found 16.4% of approximately 240,000 S aureus isolates had an MIC 2 µg/mL.4 The patient populations in the studies associating high vancomycin MICs with poor outcomes were small, isolated samples with heterogenous baseline demographics, and as such likely do not represent the patient population of an academic medical center as described in our study.
Targeting pharmacodynamic end points such as unbound trough concentrations four to five times the MIC or an area under the curve (AUC):MIC ratio of 400 to achieve microbiological eradication or successful clinical outcome is also subject to debate, particularly if the MRSA isolate has an MIC 1.5 µg/mL. First, the data supporting achievement of a vancomycin AUC:MIC ratio 350 is based on a sample of 33 patients with lower respiratory tract infections caused by MRSA that was isolated from sputum samples.5 Eradication of MRSA from a sputum culture may or may not be related to hospital mortality in patients with pneumonia. Furthermore, all isolates in this study had MICs of 0.5 µg/mL and 1.0 µg/mL, and therefore this association should not be made for strains with an MIC 1.5 µg/mL. More recently, clinical response rates were significantly lower in a group of patients with MRSA infections caused by isolates with MICs 1.5 µg/mL compared to a group with MICs < 1 µg/mL (62% vs 85%, p = 0.02) despite achieving a trough concentration of four to five times the MIC.2 We clearly acknowledge in the discussion of our article that variability in the MIC distribution could have occurred between survivors and nonsurvivors thus impacting our findings. Interestingly, Monte Carlo simulation of our data reveals 0% probability of achieving an AUC:MIC ratio 400 with either low-dose (trough < 15 µg/mL) or high-dose (trough 15 µg/mL) in MRSA isolates with an MIC of 2 µg/mL.6 Further, based on this simulation, optimizing the vancomycin AUC:MIC ratio may only be of consequence in isolates with an MIC of 1 µg/mL.
We agree with the statement Drs. Potoski and Paterson identifying a correlation between trough serum concentrations and AUC values; however, neither parameter was correlated with a successful outcome in our patient population. Dr. Hall and Ms. Adams-Huet postulate that increasing the serum vancomycin trough concentration in sicker patients may put them on equal footing regarding risk of death as less sick patients receiving lower serum trough concentrations. We attempted to answer this question through logistic regression analysis. If the above question were true, serum trough concentrations and or APACHE (acute physiology and chronic health evaluation) II scores would have been identified as predictors of outcome in this patient population. The two variables associated with mortality were COPD and vasopressor administration.
It is becoming increasingly apparent that targeting vancomycin pharmacodynamic end points in patients with pneumonia caused by an MRSA isolate with and MIC 1.5 µg/mL is a strategy that may not improve outcomes and could lead to increased toxicity.27 Vancomycin may still be an option in MRSA pneumonias that have an MIC 
0.5 µg/mL; however, alternative agents such as linezolid or tigecycline, and in the near future ceftobiprole, dalbavancin, and telavancin, should be considered if the MIC is > 1 µg/mL.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
