Nephrotoxicity of piperacillin-tazobactam combined with vancomycin: Should it be a concern?
The combination of piperacillin/tazobactam (TZP) and vancomycin (VAN) is commonly used as empirical antibiotic therapy 1. Recent literature has suggested that the concomitant use of TZP and VAN is associated with a higher risk of acute kidney injury (AKI) compared to the use of VAN alone 1. The specific mechanism by which a combination of TZP and VAN may cause AKI is not well understood, but it is postulated that there is an additive effect of interstitial nephritis and direct cellular necrosis when TZP and VAN are used in combination 2.
Balci and colleagues3 performed a single-centre, retrospective cohort study to evaluate the risk of AKI across different patient groups, i.e. patients receiving (i) both TZP and VAN, (ii) VAN alone, (iii) TZP alone and (iv) both meropenem (MER) and VAN. Patients admitted to the ICU, patients developing septic shock, pregnant women and patients with a baseline serum creatinine > 2 mg/dL were excluded from analysis. An increase of ≥50% in serum creatinine or an absolute increase of ≥0.3 mg/dL from baseline during antibiotic therapy or in the 72 hours after discontinuation of the treatment was described as AKI. Both univariate analysis and multivariable regression was used to study the impact of the four different antibiotic regimens on the risk of AKI. Possible predictors of AKI for the multivariable analysis were selected based on a p-value<0.20 in univariate analysis.
A total of 402 patients were included in the study, 76 patients developed AKI.
- The incidence of AKI was significantly higher in in the TZP-VAN group compared with all other groups (41.3% vs 10.1%, 15.7% and 16% in the MER-VAN group, the VAN group and the TZP group respectively).
- The risk of developing AKI in patients receiving TZP-VAN was 3.5 times higher compared to patients receiving VAN alone.
- The risk of developing AKI was increased 1.76 times with the concomitant administration of potentially nephrotoxic agents.
- The observed AKI was usually mild and self-limiting.
The strength of this study is that it has compared patients receiving TZP-VAN with patients receiving MER-VAN. TZP-VAN is usually administered in patients who are more severely ill compared to patients receiving TZP alone. Disease severity may impact the risk of AKI and this may confound the data. However, this limitation is well handled by also comparing TZP-VAN with MER-VAN, as patients in the MER-VAN group are likely to be at least equally as ill as patients in the TZP-VAN group.
The main limitations of this study are its retrospective nature, the risk of overfitting when selecting variables for regression based on a univariate analysis and finally, the large number of predictors with respect to the numbers of observation 4.
Take home message
The large majority of studies supporting an increased risk of AKI in patients receiving TZP-VAN when compared to TZP alone are based on retrospective data. Nevertheless, it is said that there is no smoke without fire and considering this a potential modifiable risk factor for AKI, further research is certainly warranted.
This article review for the ESICM Journal Review Club was provided by Dr. Sofie Dhaese and Dr. Barbara Borgatta, on behalf of the Antimicrobial Use Working Group.
1. Hammond DA, Smith MN, Li C, Hayes SM, Lusardi K, Bookstaver PB. Systematic Review and Meta-Analysis of Acute Kidney Injury Associated with Concomitant Vancomycin and Piperacillin/tazobactam. Clin Infect Dis 2017; 64: 666–674.
2. Elyasi S, Khalili H, Dashti-Khavidaki S, Mohammadpour A. Vancomycin-induced nephrotoxicity: mechanism, incidence, risk factors and special populations. A literature review. Eur J Clin Pharmacol 2012; 68: 1243–55.
3. Balcı C, Uzun Ö, Arıcı M, Hayran SA, Yüce D, Ünal S. Nephrotoxicity of piperacillin-tazobactam combined with vancomycin: Should it be a concern? Int J Antimicrob Agents 2018; published online April 1. DOI:10.1016/j.ijantimicag.2018.03.024.
4. Babyak M. What You See May Not Be What You Get: A Brief, Nontechnical Introduction to Overfitting in Regression-Type Models. Psychosom Med 2004; 66: 411–421.