871. Acute Renal Failure Associated with Vancomycin and Beta-lactams for the Treatment of Osteomyelitis: Piperacillin/Tazobactam Compared to Cefepime
Session: Poster Abstract Session: Bone, Joint, and Soft Tissue Infection
Friday, October 19, 2012
Room: SDCC Poster Hall F-H

Antibiotic therapy for osteomyelitis (OM) in diabetics frequently requires a broad-spectrum regimen, often including vancomycin and an anti-pseudomonal beta-lactam.  Little data is available comparing the nephrotoxic potential of vancomycin when combined with certain beta-lactam antibiotics.


A retrospective cohort study was conducted of all patients with diabetes and OM treated with vancomycin plus either piperacillin/tazobactam or cefepime for at least 72 hours at a VA Medical Center between 1 January 2006 and 31 December 2011.  All patients with a creatinine clearance (CrCl) ≤ 40 mL/min, blood-urea nitrogen to serum creatinine (BUN:SCr) ratio ≥ 20:1, or absolute neutrophil count <500 cells/mm3 were excluded.  Patients could not have received concurrent therapy with intravenous (IV) acyclovir, amphotericin B, an aminoglycoside antibiotic, or any IV vasopressor. 

The primary outcome was development of acute renal failure (ARF)(an increase in SCr by 0.5 mg/dL or an increase of 50%).  Necessity of hemodialysis, total hospital days, time to peak SCr, and recurrence of infections were secondary outcomes.


One hundred thirty-nine patients met inclusion criteria; 109 in the piperacillin/tazobactam group and 30 in the cefepime group.  Thirty-six (25.9%) patients overall developed ARF; average vancomycin troughs were 18.6 mg/dL and 14.4 mg/dL for patients with ARF and without ARF respectively (p=0.051).  In patients receiving piperacillin/tazobactam, 29.3% (32/109) developed ARF compared to 13.3% (4/30) treated with cefepime (p=0.099).  Of patients receiving high-dose therapy (≥ 18 grams of piperacillin/tazobactam per day or ≥ 3 grams of cefepime per day), 37.5% (9/24) receiving piperacillin/tazobactam and 17.6% (3/17) treated with cefepime developed ARF (p=0.29).  A logistic regression analysis, correcting for the covariates of age, weight, total duration of therapy, and average vancomycin troughs, resulted in an odds ratio of 3.45 (95% CI 0.96-12.40; p=0.057) for the development of ARF in the piperacillin/tazobactam group compared to the cefepime group.


The authors were unable to detect a statistically significant difference in ARF between the beta-lactam groups; however, power was not met.  These results suggest further study with a larger patient population is warranted. 

Ryan P. Moenster, Pharm.D.1, Travis W. Linneman, Pharm.D.2, Patrick M. Finnegan, Pharm.D.2, Stephanie Hand3, Zaneta Thomas3 and Jay McDonald, MD4, (1)Pharmacy Practice, St. Louis College of Pharmacy, St. Louis, MO, (2)Pharmacy Services/Medicine Specialty Care, St. Louis VA Medical Center - John Cochran Division, St. Louis, MO, (3)St. Louis College of Pharmacy, St. Louis, MO, (4)St. Louis VA Medical Center, St. Louis, MO


R. P. Moenster, None

T. W. Linneman, None

P. M. Finnegan, None

S. Hand, None

Z. Thomas, None

J. McDonald, None

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 17th with the exception of research findings presented at the IDWeek press conferences.