Results of studies examining the occurrence of acute kidney injury (AKI) from the concomitant administration of vancomycin and piperacillin/tazobactam (V + P/T) have been mixed, with some demonstrating an increased risk of nephrotoxicity and others showing no evidence of harm. Numerous factors account for this finding, including variations in the definitions of AKI, differences in the timeframe used to evaluate the onset of AKI following exposure, and focusing on use of the drug in critically ill populations. However, little is known about the risk of iatrogenic kidney injury from this antibiotic combination in the general population.

Researchers conducted a 2-year (January 1, 2018-December 31, 2019), multicenter, retrospective, propensity-matched cohort study at a large health system to compare the incidence of AKI among adult general medicine patients receiving either V + P/T compared with vancomycin with either cefepime (V+C) or meropenem (V + M). The purpose of this study was to characterize the development of AKI and the recovery patterns following the administration of these antibiotic combinations.

Inclusion criteria were age 18 years or older; vancomycin administration in combination with either P/T, C, or M; duration of combination therapy of at least 48 consecutive hours; baseline serum creatinine drawn within 24 hours of admission; and the availability of one or more vancomycin levels while the patient was on combination antibiotic therapy. Patients were excluded from the study if they developed AKI 48 or more hours after the initiation of combination therapy; P/T, C, or M were administered in the same hospital admissions; end-stage renal disease was present at baseline; or if they were pregnant.

Antibiotics were dosed by pharmacists using standardized protocols. For V, pharmacists were to modify initial orders to provide 20 to 30 mg/kg, up to a maximum of 2 g as a loading dose, if needed. Pharmacists made dose adjustments based on body weight and renal function with a target V serum concentration of 10 to 20 mcg/mL (or 15-20 mcg/mL as indicated).

Pharmacists were also authorized to order V trough levels prior to the third to fifth dose and again as needed not to exceed 5 to 7 days since the last level. The initial P/T dose was 4.5 grams infused over 30 minutes, regardless of renal function. Subsequent doses consisted of 3.375 grams every 8 hours (administered over 4 hours) in patients with a creatinine clearance of 20 mL/min or greater or this dose was administered every 12 hours in those on hemodialysis or with a creatinine clearance of <20 mL/min. Pharmacists could customize the doses of P/T, C, and M, which were administered over 30 minutes, based on patient and clinical features.

The investigators defined the presence of AKI as a maximum serum creatinine increase of 0.3 or greater mg/dL within 48 hours or any serum creatinine during the follow-up period that was >1.5 times than baseline. Using modified KDIGO staging criteria (in which urine output was excluded), the researchers defined stage I AKI as a serum creatinine increase 1.5 to 1.9 times greater than baseline or serum creatinine increases of 0.3 mg/dL or greater; stage II AKI as serum creatinine increases of 2.0-2.9 times greater than baseline; and stage III AKI as a threefold increases in serum creatinine from baseline, a serum creatinine 4.0 mg/dL or greater or initiation of renal replacement therapy. AKI recovery patterns were classified as <3 days (fast), 3 to 7 days (intermediate), or >7 days (slow). Sepsis was defined as a Sequential Organ Failure Assessment score of 2 or higher. A list of nephrotoxic medications was also developed.

The primary outcome of this study was the incidence of AKI associated with concomitant V + P/T versus V + C/ M. Secondary endpoints included the severity of AKI, incidence and duration of renal replacement therapy if needed, time to the onset and recovery of AKI, hospital mortality, and hospital length of stay. Propensity score matching was performed based on age >65 years, contrast dye administration, administration of three or more nephrotoxic medications, sepsis, baseline serum creatinine, African-American race, female gender, and the presence of chronic kidney disease.

A total of 3,199 patients met the inclusion criteria, of which over two-thirds (67.1%) received V + P/T and the rest (32.9%) received V + C/M. At baseline, more patients in the V+ C/M group had significantly more severe disease (as evidenced by a higher Charlson Comorbidity Index score; 4.3 vs. 3.6 in the V+ C/M compared with the V + P/T, respectively), sepsis, and chronic kidney disease. Vancomycin trough levels were similar between both treatment groups.

Researchers found that the incidence of AKI was significantly higher in the V + P/T group compared with the V + C/M group (16.4% versus 8.7%, respectively). AKI occurred significantly earlier by 1.8 days in the V + P/T group than in the V + C/M group (4.2 days vs. 6.0 days, respectively). There was no significant difference in the occurrence of stage III AKI; however, stage I AKI occurred significantly more often in the V + C/M group than in the V + P/Z group (85.9% vs. 70.0%, respectively). The reverse was seen for more severe AKI with significantly higher incidences of stage II and stage II or III in the V + P/T compared with the V + C/M groups (for stage II, 16.4% vs. 2.3%, for stages II or III, 30.0% vs. 14.1%, respectively).

There was no significant difference in AKI recovery patterns between the groups, with over 90% recovering in less than 3 days. In the V + P/T group, 0.5% needed renal replacement therapy compared with 0.2% in the V + C/M; this difference was not significant. V + P/T was found to be an independent risk factor for the development of AKI as was the duration of beta-lactam use. In-hospital mortality, although low, was significantly higher in the V + P/T group versus the V + C/M group (0.3% vs. 1.0%, respectively). Length of stay was reversed, being significantly longer in the V + C/M group than in the V + P/T group (8.2 days vs. 7.3 days, respectively).

This first-of-its kind large, multicenter, propensity-score matched study conducted exclusively in general medical patients provides pharmacists with valuable information regarding the potential risk of nephrotoxicity associated two common vancomycin regimens. Pharmacists can apply these findings to take a more proactive approach to assure the optimal and safe use of these antibiotics.

The content contained in this article is for informational purposes only. The content is not intended to be a substitute for professional advice. Reliance on any information provided in this article is solely at your own risk.

« Click here to return to Infusion Pharmacy Update.