Niveta Aravind, MD
PGY III Resident in Neurology
Wayne State University/Detroit Medical Center
Editor
Wazim Mohamed, MD
Associate Professor of Neurology and Neurosurgery
Department of Neurology, Wayne State University
Original Article: Miano, Todd A et al. “Association of vancomycin plus piperacillin-tazobactam with early changes in creatinine versus cystatin C in critically ill adults: a prospective cohort study.” Intensive care medicine 48.9 (2022): 1144-1155.
Background/Methods
Vancomycin (VN) and Piperacillin-Tazobactam (PT) are commonly used antibiotics for acutely ill patients with sepsis. A major safety concern for this combination is its link to acute kidney injury (AKI), which has been demonstrated in multiple prior studies. VN can cause oxidative stress and form obstructive tubular casts resulting in acute tubular necrosis. There are rare cases of acute interstitial nephritis associated with PT use, however there is no true evidence that PT causes nephrotoxicity. Serum creatinine is a standard kidney function biomarker that is secreted in the proximal tubule. Both VN and PT bind to renal transporters that assist with creatinine secretion. Therefore, creatinine-defined AKI may be caused by disturbance of creatinine tubular secretion rather than true toxicity on kidney parenchyma. Cystatin C (Cys-C) is another primary biomarker of kidney function, however in contrast to creatinine, Cys-C is unaffected by tubular secretion. Blood Urea Nitrogen (BUN) is a secondary biomarker of glomerular filtration that is also not affected by tubular secretion.
The purpose of this study was to identify the association of VN+PT with acute kidney injury versus pseudotoxicity due to its effects on creatinine secretion. This was a prospective cohort study part of the Molecular Epidemiology of SepsiS in the ICU (MESSI) Project, which enrolls patients admitted to the ICU with severe sepsis or septic shock meeting sepsis-2 criteria. This study identified MESSI patients treated with VN+PT or VN+Cefepime (CP) for ≥ 48 hours, with each drug initiated within 48 hours of ICU admission. The primary outcomes were kidney function biomarkers (creatinine, Cys-C, BUN) measured on the day of antibiotic initiation and two days later. Secondary outcomes included creatinine-defined AKI through day 14, the need for dialysis, and 30-day mortality. The study was adjusted for confounding factors by using inverse probability of treatment weights (IPTW), calculated from the propensity score for VN+PT treatment.
Results
There were a total of 3303 patients enrolled in the MESSI Project between 2008 and 2020, of which 1293 received antibiotics within 48 hours of ICU admission. Of these 1293 patients, 739 received either VN+PT (n=297) or VN+CP (n=442), among whom 192 had plasma samples for Cys-C analysis. VN+PT was associated with significantly higher average creatinine concentrations compared to VN+CP (19.79% difference) and a higher frequency of creatinine increases of ≥ 50% at day 2 (rate ratio 2.33). However, VP+PT was associated with non-significantly lower average Cys-C concentrations (1.26% difference) and a similar frequency of Cys-C increases of ≥ 50% at day 2 (rate ratio 1.37) compared to VN+CP. There was no significant difference in BUN levels or in the rate of ≥ 50% increases of BUN through day-14 between VN+PT and VN+CP groups. Both Cys-C:Cr and BUN:Cr ratios were significantly lower in the VN+PT group compared to VN+CP, indicating that creatinine is increased to a significantly greater extent than either Cys-C or BUN by day 2. In addition, VN+PT was associated with a significantly higher rate of creatinine-defined AKI at day 14, but there was no association in the rate of dialysis or mortality after IPTW analyses.
Commentary
Given the abundant use of VN+PT and multiple studies linking VN+PT with creatinine-defined acute kidney injury, it is imperative to determine if VN+PT truly causes acute kidney injury or rather pseudotoxicity due to effects on creatinine secretion. This study showed that patients receiving VN+PT had increased creatinine concentrations at day 2 and an increased rate of creatinine-defined AKI at day 14. However, VN+PT was not associated with changes in Cys-C or BUN at day 2, nor were there higher rates of dialysis or mortality associated with VN+PT. The authors in this study conclude that VN+PT is associated with an increased risk of creatinine-defined AKI but that there is no association with changes in other kidney function biomarkers (Cystatin C, BUN), indicating that VN+PT causes a pseudotoxicity rather than a true kidney injury. Therefore, VN+PT can safely be used in adult patients with suspected sepsis or septic shock without drug combination safety.
From a side effect profile, it can be argued that PT may be more favorable than CP, especially if it does not increase the incidence of kidney injury when combined with VN. CP can cause neurotoxicity in patients with renal dysfunction due to its ability to cross the blood–brain barrier and cause concentration-dependent ϒ-aminobutyric acid (GABA) antagonism, increased excitatory amino acid release, as well as cytokine release. Studies show that approximately 10% of serum CP crosses the blood-brain barrier, but this fraction can increase to up to 45% in patients with renal impairment. The elevated CNS concentrations of CP can result in decreased consciousness, encephalopathy, aphasia, myoclonus, seizures, and even coma. In addition, CP is associated with an increased risk of unexplained mortality in hospitalized neutropenic patients when compared with other antibiotics. As there may not be a significant difference in rates of kidney injury between VN+PT and VN+CP, one can consider the use of PT over CP in select patients.
One of the limitations of this study is that Cys-C concentrations can be affected by other factors such as body mass index, corticosteroids, thyroid activity, cancer, diabetes mellitus, and solid organ transplant. However, this study attempted to mitigate potential confounding factors through IPTW analysis and by comparing VN+PT to VN+CP, which was given for the same indications. Another limitation of this study was that the Cys-C concentrations were only available in a subset of patients, making the study sample smaller. In addition, the incidence of dialysis initiation was low, suggesting an underrepresentation of this outcome. Finally, the study was only conducted at a single center, which may limit its generalizability. Further studies with a larger cohort sampled from different centers can be used to validate this study’s conclusions.
In conclusion, vancomycin + piperacillin-tazobactam can potentially be safely used without concern for acute kidney injury. In addition, educating physicians regarding the limitations to creatinine-defined AKI and the importance of using other primary kidney function biomarkers such as Cystatin-C would be beneficial in determining kidney damage.
#LiteratureWatch