Nephrotoxicity and antibiotics

Abstract

The article Impact of Combining Vancomycin with Piperacillin/tazobactam or with Meropenem on Vancomycininduced Nephrotoxicity [1] by Razan Tookhi and colleagues brings up several interesting thoughts about antibiotic interactions with kidney function. This article was a retrospective study of patients admitted from June 2016 to March 2019 who was administered either Vancomycin (VAN) plus Piperacillin/tazobactam (TZP) or VAN plus meropenem (MEM). It is unclear if this was a single center or multiple facilities. They analyzed data from 158 patients, 77 were administered VAN-TZP and 81 VAN-MEM. More patients in the VANMEM group had acute kidney injury (AKI) but this was not statistically significant. In hospital, mortality was noted to be higher in the VAN-MEM group. Their conclusion was the VAN-MEM did not show a lower rate of AKI over the combination of VAN-TZP. The article specifically utilizes the Kidney Disease Improving Global Outcomes (KDIGO) clinical practice guidelines for defining acute kidney injury at 72 h [2]. The definition of increase in serum creatinine (Scr) by ≥ 0.3 within 48 h fits into the author’s study design of evaluating for acute kidney injury (AKI) over 5 days. However, KDIGO also defines AKI as urine volume < 0.5 ml/kg/h for 6 h or increase in Scr to > 1.5 times baseline known or presumed to have occurred within the prior 7 days. Since the KDIGO guideline release in 2012, this definition has been utilized across trials examining AKI in antibiotic use [3, 4]. In this article, excluding patients receiving less that 72 h of antibiotics allowed for ample time to see AKI based on daily Scr measurements during the 3–5 day window after starting antibiotics and ensured patients received multiple doses of the antibiotics being studied [5]. Patients in this study were excluded with creatinine over 1.5. Several potential confounders may come from this methodology. First the potential for AKI may increase by as much as 10 × with underlying chronic kidney disease (CKD) [6]. The exclusion of patients with creatinine above 1.5 may be eliminating the population you would see increased AKI in and artificially lowering the rate of AKI that would be seen in an average population placed on these medications. Seeing this study repeated in those with low levels of renal dysfunction may give a more generalizable picture. Another confounding factor is that serum creatinine does not always reflect true change in the glomerular filtration rate (GFR). The creatinine can be falsely elevated from things such as increased protein consumption, increased muscle breakdown such as from high-intensity exercise, colorimetric assay errors from substance interaction, and decreased excretion in the renal tubules [7]. On the contrary, creatine levels can be falsely low because of the volume status of the patient [8, 9]. This could result in patient creatinine elevation 2–3 days into a hospitalization that is secondary to renal insult prior to hospitalization and not to medications administered in the hospital. One study estimated that up to 18% of patients would be classified as having AKI or renal insult reclassified once the creatinine was adjusted for fluid status [9]. No discussion of IV fluid hydration was present in the article which may have further altered the onset or diagnosis of AKI. The authors included in the baseline demographics the number of other potentially nephrotoxic medications, specifically non-steroidal anti-inflammatory drugs (NSAIDs), angiotensin converting enzyme inhibitors (ACEi), angiotensin receptor blockers (ARB), aminoglycosides, amphotericin B, and colistin. While the number of specific patients did not vary between groups, there were twice as many patients in the VAN-MEM group on > 1 nephrotoxic medication, which could contribute to the higher rate of AKI seen in that population [4]. No comment was made if these drugs were stopped or were part of the in-hospital treatment plan. Patients who did not have enough data to determine AKI were excluded from this study. The authors did not elaborate what this meant exactly. This could have excluded * Angela P. Cornelius [email protected]