Changes in the incidence of Vancomycin Associated-Acute Kidney Injury (VA-AKI), risk factors and applicability of studies are questionable in recent times. This review aims to evaluate the incidence, risk factors and the epidemiological errors of findings on VA-AKI. 

All studies related to vancomycin associated acute kidney injury in the past 20 years were identified from PubMed, MEDLINE, Google scholar and UWIlinC. During the past decade, the incidence of VA-AKI varied between 7.3% to 15.8%. The lowest incidence was found among the patients who received pharmacist intervention prior to the initiation of therapy, and the highest incidence was among elderly patients over 65-year-old. Notable risk factors were old age, high trough ≥15 μg/mL, AUC/MIC ratio > 650 mg × hour/L, concomitant use of aminoglycosides, broad-spectrum antibiotics, diuretics and vasopressor, using vancomycin without pharmacist intervention, hyperuricemia, respiratory failure, cardiac failure, chronic kidney disease, mechanical ventilation, obesity and black race. Common epidemiological errors in these studies were generalization and selection bias. The disparity of facilities among healthcare services is a concern. 

Knowing modifiable risks, using vancomycin as monotherapy and advance consultation with pharmacists seem valuable preventive tools for VA-AKI regardless of resource settings. 

A surge of reports on VA-AKI was noted after the infectious disease guidelines recommended to use sufficient initial and maintenance vancomycin doses to prevent treatment failure in in 2005 and in 2009 [1-4]. The nephrotoxicity related to a higher trough level from 1995 to 2012 was checked extensively and found out that the incidence of VA-AKI varied from 5-43% [1]. A longitudinal study from 2016 to 2019 stated that therapeutic drug monitoring was able to do on only 32.3% among the patients exposed to vancomycin [2]. The concerned on the recommended higher trough level in 2005 for the treatment of pneumonias associated with hospital, healthcare services and ventilators was expressed [3]. The most immediate response to the recommended higher trough level was produced based on the results between 2006 to 2008 which revealed an alarming outcome as 81.8% of patients with the trough level exceeded more than 35 mg/L developed VA-AKI [4]. Fortunately, a recent report in 2022, the incidence declined to 7.3% among patients who received pharmacist intervention prior to intravenous vancomycin therapy [5]. In the past decade, the incidence reached as high as 15.8% among the elderly hospitalized patients in 2016-2017 [6]. Despite being well-recognized as a nephrotoxin, the incidence of VA-AKI remained un-earthed in certain regions, probably due to lack of vancomycin level monitoring facilities. A complete picture can be seen when the incidence and risk factors are well-generalized.  We aimed to identify the recent changes in incidence, reports on risk factors and to explore epidemiological errors related to the VA-AKI.

Renal affinity of vancomycin

The accumulation of vancomycin occurred two times higher in kidneys than in plasma and its toxicity was dose related.  The vancomycin concentration in the kidney increased by 3.5-fold with high dose vancomycin (600 mg/Kg) [7]. Up to 90% of vancomycin unchanged form are filtered to the proximal convoluted tubules (PCT), resulting in accumulation in the PCT lining cells where it was metabolized [8]. Murphy E and Barreto E declared that vancomycin is a nephrotoxin and alternative therapy should be considered, for high-risk patients, if possible [9]. 

Pathophysiology of VA-AKI

The pathological manifestation of VA-AKI includes acute tubular necrosis (ATN), acute tubulointerstitial nephritis (ATIN) and intratubular crystal obstruction. A pooled renal biopsy reported that a combination of acute tubular necrosis (ATN) and acute tubulointerstitial nephritis (ATIN) were the most common findings. The ATIN was found to have a significantly higher risk of permanent renal dysfunction [HR: 5.08, 95% CI: 91.05–24.50)] [10]. The accumulation of vancomycin molecules in the PCT cells triggers inflammatory reactions, oxidative stress, and complement activation. Subsequent mitochondrial dysfunction, and apoptosis of the renal tubular epithelial cells led to acute tubular necrosis [11]. Htike, N. L et al stated that the pathophysiology of ATIN was thought to be mediated by a type IV delayed hypersensitivity reaction involving T-cells [12]. A smaller subset of patients developed VA-AKI due to intratubular crystal obstructions in the distal convoluted tubules, where casts are formed by uromodulin and vancomycin precipitates in a low urinary acidity (pH Less than 5.5) state [13].

Search strategy and selection criteria

PubMed, MEDLINE, Google Scholar and UWIlinC databases were selected as search databases. The initial search terms were “vancomycin”, “acute kidney injury” or “acute renal failure” or “aki” and “nephrot* and “vancomycin trough” or “vancomycin auc/*”. The search included both animal and human studies over the past 20 years. All the articles are centered on the incidence and/or risk factors of VA-AKI. The articles detailing acute kidney injury, drug pharmacodynamics, clinical outcomes, publications in English language, studies focused on adults and studies accessible in full-text form are included. The exclusion criteria are duplicate publications, AKI not related to vancomycin, studies lacking adequate data (abstracts with no full text available), publications older than 20 years, and non-intravenous use of vancomycin. 

Records identified were 81 from PubMed, 57 from MEDLINE, 2610 from Google Scholar and 1401 from UWIlinC (total n= 4158). Records removed before the screening were those articles which did not include vancomycin (n= 4082) and duplicate articles (n= 44). Of the remaining 32 records, articles with no renal association (n=5), study with sample size Less than 400 (n=5), article based on vancomycin via intra-articular spacers use (n=1), articles on pediatric studies (n=6) and textbook article published beyond the past 20 years (n=1) were removed. The final selected studies in this review were 14 in which 7 articles focused on both incidence and risk and 7 articles mainly focused on risk of VA-AKI. The range of study designs were 7 systemic reviews with meta-analyses, 2 systematic reviews, 3 cross sectional studies, 1 cohort study, and 1 case-control study. Most studies used serum creatinine levels as the primary biomarker for diagnosing AKI according to AKIN, RIFLE or KDIGO criteria. (Figure 1).

                                                      Figure 1: PRISMA flow diagram showing search strategy for literature review of VA-AKI

A systematic review and meta-analysis by van Hal SJ et al., (2012) stated that the incidence of VA-AKI from 1995 to April 2012 was between 5 – 43% [1]. A longitudinal study by Kunming P et al., (2021), revealed that the incidence of VA-AKI was 14.3%. In addition, these patients had longer hospital stays (23 vs. 20 days, p Less than 0.001) and a higher 30-day mortality rate (8.8% vs. 1.5%, p Less than 0.001) [2]. Among ICU patients with pneumonia, VA-AKI was as high as 15.4% and the independent risk factors were initial vancomycin trough levels ≥15 mg/L (OR, 5.2, 95% CI, 1.9-13.9; P = 0.001), concomitant aminoglycoside use (OR, 2.67; 95% CI, 1.09-6.54; P = 0.03), and duration of vancomycin therapy (OR for each additional treatment day, 1.12; 95% CI, 1.02-1.23; P = 0.02) [3].  A study by Horey A et al., (2012) revealed that the prevalence of vancomycin associated nephrotoxicity was 12.6% with maximum trough concentrations (OR 1.14; 95% CI 1.09 to 1.20) and documented hypotension (OR 4.7; 95% CI 1.3 to 16.4) were with higher risk [4]. A systematic review and meta-analysis in 2022 revealed that the incidence of VA-AKI without pharmacist intervention was 9.6% and with pharmacist intervention was 7.3%. Hence, the incidence of VA-AKI was 2.3% less among the patients who received pharmacist intervention in advance (OR 0.52, 95% CI 0.41, 0.67, P Less than .001) [5]. In terms of age-related risk, a cross-sectional study by Pan K et al (2018) on elderly patients over 65-years-old revealed that the incidence of VA-AKI was 15.8% [6]. The studies which included incidence as well as risk ratios were summarized in the table (Table 1).

                        Table 1: Eligible Journal Articles on incidence and risk factors of Vancomycin Associated Acute Kidney Injury in adults.

VA-AKI, Vancomycin Associated Acute Kidney Injury; USA, United States of America; ICU, Intensive care unit; HAP, Hospital Acquired Pneumonia; VAP, Ventilator Associated Pneumonia; HCAP, Healthcare Associted Pneumonia; PTZ, Piparacillin-Tazoactam.

VA-AKI risk factors 

A pooled results of 7 RCTs and 7 cohort studies on 4033 patients by Ray AS et al., (2016) concluded that the risk of nephrotoxicity was 2.5-fold higher in patients who received intravenous vancomycin [14]. Kim J.Y et al. (included 53 studies in a meta-analysis (n = 50-3719), revealed a higher risk of Vancomycin Associated AKI (VA-AKI) among black and obese population. (OR 1.47, 95% CI: 1.16–1.87 and OR 1.46, 95% CI: 1.12–1.90, respectively) [15]. Pan K et al expressed independent associations between VA-AKI and hyperuricemia (OR 3.045; 95% CI 1.834, 5.057; P = Less than0.001), mechanical ventilation (OR 1.906; 95% CI 1.098; 3.310, P = 0.022) and concomitant vasopressor therapy (OR 1.919; 95% CI 1.078, 3.418; P = 0.027) [6]. A prospective cohort by Lodise, T (2008), found that the prevalence of AKI among patients who received vancomycin >4 G/day, Less than4/day and linezolid was noted as 34.6%, 10.9%, and 6.7%, respectively (P = 0.001) [16]. A systematic review by Elyasi et al., reported that high dose vancomycin >4 G/day, trough level >20 mg/dL and longer duration of therapy > 7 days were associated with higher prevalence of VA-AKI compared to the conventional dose (30-40% vs 10-20%) [17]. 

Among the critically ill patients, backwards logistic regression analysis by Hanrahan T.P (2015) revealed serum vancomycin concentration and APACHE II score as independent positive predictors (OR= 1.174, P= 0.024 and OR= 1.141, P= 0.012, respectively) [18]. A meta-analysis done in 2022 focused on the first 24 hours (0-24 hours) and second 24 hours (24-48 hours) vancomycin troughs. AKI was significantly less when trough level area under curve (AUC) was less than 650 mg×h/L [19]. Abdelmessih, E et al (2022) augmented another meta-analysis which showed that the VA-AKI was significantly lower in the AUC-guided dosing strategies than trough-guided dosing strategies (OR 0.625, 95% CI (0.469–0.834), p = 0.001) [20]. In 2021, Tsutsuura M et al found a significantly lower rate of treatment failure in patients with the trough level ≥ 15 µg/mL ((OR 0.63, 95% CI 0.47–0.85) but when trough concentrations ≥20 μg/mL, the rate of AKI was two times higher than trough concentrations between 15-20 μg/mL (OR 2.39, 95% CI 1.78–3.20) and target cut-off 600 ± 15% was associated with a higher risk of VA-AKI (OR 2.10, 95% CI 1.13–3.89) [21]. A retrospective cohort by Pitcock CT et al (2023) on prolonged vancomycin therapy (> 14 days) demonstrated a higher incidence of AKI (45.6% vs 28.4%, p Less than 0.001) with higher mortality (12.9% vs 8.3%, p = 0.078) in the trough level monitoring group compared to the AUG guided group and the latter had 54% less incidence of AKI (OR 0.46, 95% CI [0.31–0.69) [22]. Retrospective cohort by Ishigo T et al., (2024) supported with a comparison analysis which showed that high-AUC group, intermediate-AUC group and low-AUC group had different AKI rates (42.9%, 28.0%, 6.5%, respectively) [23]. A meta-analysis by Yang W et al (2024) revealed that vancomycin trough concentration (Ctrough) were indicators for nephrotoxicity (OR 2.193; 95% CI 1.582–3.442, p Less than 0.001) and Ctrough 10–20 mg/L was equivalent with a mean AUC24 within 400–600 mg·h/L in most patients (92.3%). The conclusion was that the observed trough level, Ctrough, should remain a beneficial tool of monitoring for VA-AKI [24].

Kim T et al (2015) reassured with their findings that vancomycin monotherapy in hemodynamically stable non-critically ill patients had significantly lower nephrotoxicity (OR 0.14, 95 % CI 0.04–0.52, p = 0.004) compared to the combined Vancomycin/Piperacillin-Tazobactam group and the overall incidence was 11.8% [25]. In 2018, Luther MK et al declared that Piperacillin-Tazobactam and vancomycin combined therapy. had a significantly higher risk of VA-AKI than vancomycin monotherapy (OR 3.40; 95% CI, 2.57–4.50 [26]. A comprehensive review by Blair M et al., (2021) proposed the potential pathogenesis of Vancomycin Piperacillin-Tazobactam (VPT) induced AKI as multifactorial including tubular toxicity, oxidative stress, cast formation, acute interstitial nephritis and inhibition of tubular creatinine secretion [27]. Apart from Piperacillin-Tazobactam, concomitant use of cephalosporin (OR 1.55, 95% CI 1.08-2.21, p = 0.017), carbapenems (OR 1.46, 95% CI 1.11-1.91, p = 0.006) with vancomycin also had an increased risk of VA-AKI, according to a longitudinal study [2]. Documented factors with their odds for VA-AKI are demonstrated collectively in the Forest plot [Fig. 2]. A matched case-control study by Gyamlani G et al (2019) (n=33,527) revealed similar or even less odds ratio in the vancomycin group with trough level ≤20 mg/L compared to the linezolid/daptomycin group. However, when vancomycin levels > 20 mg/L, a 4-fold higher risk of AKI was observed, compared to vancomycin levels 2Less than 10 mg/L [28]. The studies which mainly focused on the risk of VA-AKI are collectively shown in the table (Table 2) and epidemiologic errors of the selected studies are summarized in table 1 & 2.

                                                             Table 2: Eligible Journal Articles which mainly focused on the risk of VA-AKI

VA-AKI, Vancomycin Associated Acute Kidney Injury; USA, United States of America, VCM, Vancomycin; AUC, Area Under Curve; MIC, Minimum Inhibitory Concentration, AUC24, The area under the curve over 24 hours; Ctrough, trough concentrations; MRSA, Methicillin Resistant Staphylococcus Aureus.

                                 Figure 2: Forest plot demonstrating the odds ratio and factors associated with AKI after intravenous vancomycin usage.

</>

The incidence of Vancomycin Associated Acute Kidney Injury varies according to age, race, pharmacokinetics of vancomycin and personal fitness. Reports on VA-AKI surged around the time of 2012 [1,3,4,17], which was thought to be due to higher target levels of Vancomycin. [29] The trend of the VA-AKI had apparently increased with recommended trough levels or AUC/MIC ratios. The incidence was as high as 15.8% among elderly hospitalized patients in China [6] and 15.4% among ICU patient in USA [3]. Based on these reports, the vancomycin consensus guidelines committee revised in 2020 with a recommendation of target AUC/MIC ratio of 400–600 mg*hour/L to prevent treatment failure and to ensure the safety for the treatment in severe MRSA [30]. However, countries with limited-resource settings were not able to adhere to these guidelines. Although Kim J.Y et al [15]. reported a higher risk of VA-AKI among blacks, reports from regions of black preponderance, such as Africa and the Caribbean, were scarce which indicates the study had a reduced external validity with possible selection bias and confounding by disparities in healthcare access or different baseline health conditions.

Many reports recognized initial trough level ≥15 mg/L or > 4 G/day as risk factors of VA-AKI. Further analysis revealed that the trough (>20 µg/mL) or target AUC/MIC ratio value cut-off > 600 mg·h/L were the strong risk factors for VA-AKI [2, 16-24]. On the other hand, lower trough level and AUC were associated with lower risk of VA-AKI. Suggestion by Bruniera, F. R et al (2015) to use adequate dosage for a shorter duration of therapy was quite agreeable to minimize treatment failure and toxicity [31]. The statement in 2019 by Gyamlani G et al, in which the risk of AKI with intravenous vancomycin monotherapy was not higher than other antimicrobial combined therapy unless vancomycin trough level reached beyond 20 µg/dL, reassured vancomycin prescribers [28]. Collectively, findings suggested that risk of VA-AKI could be minimized with lenient monotherapy. 

It is important to be aware that more risk factors have been emerging. A retrospective study in 2014 mentioned that hypertension (74 vs. 51 %, p = 0.0009, OR 2.74, 95 % CI 1.5-5.0), furosemide use (65 vs. 39 %, p = 0.0009, OR 2.91, 95 % CI 1.64-5.15) and trough concentration ≥16.2 μg/mL (OR 2.33, 95 % CI 1.25-4.44) were associated with VA-AKI independently [32]. More risk factors were introduced by Qin, X et al (2020) in a retrospective cohort in which the different trough levels [OR of 15.1~20.0 level: 2.50 (95%CI 2.13, 2.93), OR of > 20.0 level: 3.89 (95%CI 3.34, 4.53, respectively), respiratory failure (OR 1.38; 95% CI 1.22, 1.58), chronic renal failure (OR 3.17; 95% CI 2.72, 3.70) and congestive heart failure (OR 1.56; 95% CI 1.37,1.79, concomitant diuretics (OR 1.71; 95% CI 1.51,1.94), PTZ (OR 1.39; 1.24, 1.57) and meropenem (OR 1.29; 95% CI 1.16,1.45), were found to have higher risk of VA-AKI [33]. Categorizing the risk factors into modifiable and non-modifiable factors by Kan WC et al in 2022 was also beneficial [34]. Focus on modifiable risk factors like intravascular volume depletion and acute severe illnesses should be maximized. Unfortunately, non-modifiable risks such as older age, female gender, black race, drug allergy, pre-existing comorbidities, end organ failures (cardiac, renal, liver), diabetes, immunocompromised state and obesity would remain untouchable.

Concerns should be raised for the countries with limited-resource settings where recommended trough levels or AUC/MIC ratios were not applicable and the vulnerability of those patients on conventional therapy remained questionable. The safety of vancomycin at these healthcare services should be explored more. To our knowledge, these institutions relied on the website calculator https://clincalc.com/Vancomycin/ [35], drug prescribing insert or infectious disease specialist guidance to use intravenous vancomycin. The suggestion made by Kunming P et al (2022) to seek pharmacist intervention prior to the intravenous vancomycin is quite reasonable for resource-limited countries. 

Studies and systematic reviews have been reported mainly from resource-rich countries. Recommendations were impressive but the generalization was limited, and selection bias were there. We recommend the availability of Vancomycin trough levels for any users in clinical practice especially for high-risk patients.

The incidence of VA-AKI varies widely from 7.3% to 15.8%, according to the samples selected and associated risk factors. Some findings and recommendations had selection bias and decreased generalization. Reports were mainly from resource-rich countries. Correction of modifiable risks, vancomycin monotherapy, consulting infectious disease specialists and pharmacists in advance seem useful tools to control VA-AKI regardless of resource settings.