ABSTRACT
Vancomycin-induced kidney injury is common, and outcomes in humans are well predicted by animal models. This study employed our translational rat model to investigate temporal changes in the glomerular filtration rate (GFR) and correlations with kidney injury biomarkers related to various vancomycin dosing strategies. First, Sprague-Dawley rats received allometrically scaled loading doses or standard doses. Rats that received a loading dose had low GFRs and increased urinary injury biomarkers (kidney injury molecule 1 [KIM-1] and clusterin) that persisted through day 2 compared to those that did not receive a loading dose. Second, we compared low and high allometrically scaled vancomycin doses to a positive acute kidney injury control of high-dose folic acid. Rats in both the low- and high-dose vancomycin groups had higher GFRs on all dosing days than the positive-control group. When the two vancomycin groups were compared, rats that received the low dose had significantly higher GFRs on days 1, 2, and 4. Compared to low-dose vancomycin, the KIM-1 was elevated among rats in the high-dose group on dosing day 3. The GFR correlated most closely with the urinary injury biomarker KIM-1 on all experimental days. Vancomycin loading doses were associated with significant losses of kidney function and elevations of urinary injury biomarkers. In our translational rat model, both the degree of kidney function decline and urinary biomarker increases corresponded to the magnitude of the vancomycin dose (i.e., a higher dose resulted in worse outcomes).
KEYWORDS: biomarkers, dosing, nephrotoxicity, vancomycin
INTRODUCTION
Vancomycin (VAN) is a glycopeptide antibiotic that remains the treatment of choice for methicillin-resistant Staphylococcus aureus (MRSA) infections. Unfortunately, vancomycin-induced kidney injury is a common adverse effect that occurs at an attributable rate of at least 10% (1). In the current guidelines for vancomycin dosing for Staphylococcus aureus infections, loading doses are recommended for patients who are critically ill, require dialysis or renal replacement therapy, have serious MRSA infections, or are receiving vancomycin continuous-infusion therapy (2). However, it is unknown whether vancomycin loading doses contribute to increased kidney injury. Clinical studies that underlie this recommendation assume an increased efficacy, are mostly retrospective, contain few patients, and are based on previous trough goals for vancomycin monitoring (3–6). As a result, there is a need to investigate the nephrotoxic potential of vancomycin loading doses using sensitive and specific biomarkers for kidney function and injury. The goals of this study were thus 2-fold. First, we examined kidney injury and function in the settings of a loading dose and no loading dose. Second, we defined the dose-response relationship of standard-dose and high-dose vancomycin with a positive control of acute kidney injury. To do so, we employed our translational rat model using iohexol clearance as a surrogate for kidney function and urinary kidney injury biomarkers (kidney injury molecule 1 [KIM-1], osteopontin [OPN], and clusterin) (7–10). Our previous work using this model showed that exposures and outcomes related to vancomycin-induced kidney injury are well predicted by the rat model and are closely linked to the those in prospective human studies (11). In addition, increases in specific urinary biomarkers of kidney injury precede declines in the glomerular filtration rate (GFR) in response to nephrotoxins (10, 12). The goal of this study was to investigate the temporal changes in glomerular function and urinary injury biomarkers related to variable vancomycin doses.
RESULTS
Characteristics of the animal cohort.
A total of 34 male Sprague-Dawley rats were assigned to two distinct studies (investigation of the impact of vancomycin [VAN] loading doses [study 1] and investigation of variable vancomycin doses with an acute kidney injury [AKI] positive control [study 2]), with dosing group assignments shown in Fig. S1 and S2 in the supplemental material. One animal provided only terminal plasma samples due to occluded catheters; all other animals contributed complete data. The mean baseline weight of the rats was 274.9 g, with a standard deviation (SD) of 9.4 g. The mean weight change was not significantly different over all experimental days among rats in the loading dose experiment. In the AKI positive-control experiment, rats in the folic acid (AKI positive-control) group experienced significant mean weight loss by day 3 compared to rats in the low- and high-vancomycin dose groups (−26.8 g versus −1.2 g versus −4.0 g [P < 0.0001]).
Kidney function changes over time.
The baseline (i.e., prior to therapy) mean GFRs were not significantly different between rats enrolled into the VAN no-load and those in the VAN loading dose groups (0.45 [95% confidence interval {CI}, 0.42 to 0.49] mL/min/100 g of body weight versus 0.46 [95% CI, 0.41 to 0.50] mL/min/100 g of body weight [P = 0.76]). After receipt of the VAN loading dose on the morning of day 1 (day 1-AM), rats in the VAN loading dose group experienced a significant decline in their GFR compared to the those in the group with no VAN loading dose (−0.16 mL/min/100 g of body weight [95% CI, −0.24 to −0.08 mL/min/100 g of body weight] [P < 0.001]) (Fig. 1). A significantly lower GFR was also observed among the VAN loading dose group rats after receipt of the maintenance VAN dose in the evening of day 1 (day 1-PM) (−0.12 mL/min/100 g of body weight [95% CI, −0.19 to −0.04 mL/min/100 g of body weight] [P = 0.005]), which persisted through day 2 (−0.11 mL/min/100 g of body weight [95% CI, −0.19 to −0.03 mL/min/100 g of body weight] [P = 0.007]).
FIG 1.
Iohexol GFR comparison for rats in the loading dose arm, between treatment groups and across dosing days. Comparing the treatment groups by experimental day, the GFR was significantly lower on day 1 in the morning (a, −0.16 mL/min/100 g of body weight [95% CI, −0.24 to −0.08 mL/min/100 g of body weight] [P < 0.001]), day 1 in the evening (b, −0.12 mL/min/100 g of body weight [95% CI, −0.19 to −0.04 mL/min/100 g of body weight] [P = 0.005]), and day 2 (c, −0.11 mL/min/100 g of body weight [95% CI, −0.19 to −0.03 mL/min/100 g of body weight] [P = 0.007]) for rats that received a vancomycin loading dose. No significant changes in the GFR from the baseline (day 0) were identified among rats that did not receive a vancomycin loading dose.
In the second experimental arm, the baseline GFR (i.e., prior to drug administration) was not significantly different for rats that were assigned to any of the treatment groups: the folic acid (AKI positive-control), low-dose VAN, and high-dose VAN groups (0.42 [95% CI, 0.45 to 0.50], 0.43 [95% CI, 0.43 to 0.51], and 0.43 [95% CI, 0.39 to 0.52] mL/min/100 g of body weight, respectively [P = 0.99]). Compared to the AKI positive-control group, the GFR was significantly higher among rats in the low-dose VAN group on day 1 (0.33 mL/min/100 g of body weight [95% CI, 0.22 to 0.44 mL/min/100 g of body weight] [P < 0.001]), day 2 (0.34 mL/min/100 g of body weight [95% CI, 0.23 to 0.45 mL/min/100 g of body weight] [P < 0.001]), day 3 (0.29 mL/min/100 g of body weight [95% CI, 0.18 to 0.40 mL/min/100 g of body weight] [P < 0.001]), and day 4 (0.39 mL/min/100 g of body weight [95% CI, 0.29 to 0.51 mL/min/100 g of body weight] [P < 0.001]) (Fig. 2). Compared to the AKI positive-control group, the GFR was also significantly higher among rats in the high-dose VAN group on day 1 (0.21 mL/min/100 g of body weight [95% CI, 0.09 to 0.32 mL/min/100 g of body weight] [P < 0.001]), day 2 (0.22 mL/min/100 g of body weight [95% CI, 0.11 to 0.33 mL/min/100 g of body weight] [P < 0.001]), day 3 (0.24 mL/min/100 g of body weight [95% CI, 0.13 to 0.35 mL/min/100 g of body weight] [P < 0.001]), and day 4 (0.26 mL/min/100 g of body weight [95% CI, 0.15 to 0.37 mL/min/100 g of body weight] [P < 0.001]). In a direct comparison of the low- and high-dose VAN groups, the GFR was significantly higher among rats in the low-dose VAN group on day 1 (0.13 mL/min/100 g of body weight [95% CI, 0.02 to 0.24 mL/min/100 g of body weight] [P = 0.025]), day 2 (0.12 mL/min/100 g of body weight [95% CI, 0.01 to 0.23 mL/min/100 g of body weight] [P = 0.037]), and day 4 (0.14 mL/min/100 g of body weight [95% CI, 0.03 to 0.25 mL/min/100 g of body weight] [P = 0.012]).
FIG 2.
Iohexol GFR comparison for rats in the variable vancomycin dose arm, between treatment groups and across dosing days. Using the AKI positive-control group as the reference group, the GFR was significantly higher in both the low- and high-VAN-dose group rats on all dosing days (a, 0.33 mL/min/100 g of body weight [95% CI, 0.22 to 0.44 mL/min/100 g of body weight] [P < 0.001]; b, 0.34 mL/min/100 g of body weight [95% CI, 0.23 to 0.45 mL/min/100 g of body weight] [P < 0.001]; c, 0.29 mL/min/100 g of body weight [95% CI, 0.18 to 0.40 mL/min/100 g of body weight] [P < 0.001]; d, 0.39 mL/min/100 g of body weight [95% CI, 0.29 to 0.51 mL/min/100 g of body weight] [P < 0.001]; e, 0.21 mL/min/100 g of body weight [95% CI, 0.09 to 0.32 mL/min/100 g of body weight] [P < 0.001]; f, 0.22 mL/min/100 g of body weight [95% CI, 0.11 to 0.33 mL/min/100 g of body weight] [P < 0.001]; g, 0.24 mL/min/100 g of body weight [95% CI, 0.13 to 0.35 mL/min/100 g of body weight] [P < 0.001]; h, 0.26 mL/min/100 g of body weight [95% CI, 0.15 to 0.37 mL/min/100 g of body weight] [P < 0.001]).
Urine output and injury biomarkers.
The baseline urine output (prior to drug administration) was not significantly different between the treatment groups of the loading dose arm (8.9 mL/24 h [95% CI, 5.7 to 12.4 mL/24 h] versus 12.7 mL/24 h [95% CI, 6.5 to 18.0 mL/24 h] [P = 0.054]) (Table S1). The urine output was significantly higher on day 3 among rats that received a VAN loading dose (5.1 mL/24 h [95% CI, 1.2 to 9.0 mL/24 h] [P = 0.01]) (Table 1).
TABLE 1.
Marginal differences versus the VAN no-load group as a referencea
| Parameter on dosing day | Value for VAN loading dose group |
|---|---|
| Mean GFR difference (mL/min/100 g of body wt) (95% CI), P value | |
| Baseline | −0.01 (−0.09 to 0.07), 0.79 |
| Day 1-AM | −0.16 (−0.24 to −0.08), <0.001 |
| Day 1-PM | −0.12 (−0.19 to −0.04), 0.005 |
| Day 2 | −0.11 (−0.19 to −0.03), 0.007 |
| Day 3 | −0.07 (−0.15 to 0.01), 0.08 |
| Day 4 | −0.06 (−0.14 to 0.03), 0.18 |
| Mean urinary KIM-1 difference (ng) (95% CI), P value | |
| Baseline | 1.56 (−93.8 to 96.9), 0.97 |
| Day 1-AM | 213.9 (118.5 to 309.2), <0.001 |
| Day 1-PM | 71.8 (−23.6 to 167.1), 0.14 |
| Day 2 | 163.6 (68.2 to 258.9), 0.001 |
| Day 3 | 50.7 (−44.6 to 146.1), 0.29 |
| Day 4 | 30.0 (−65.3 to 125.4), 0.54 |
| Mean urinary clusterin difference (ng) (95% CI), P value | |
| Baseline | 7,073 (−3,144 to 17,290), 0.18 |
| Day 1-AM | 16,493 (6,277 to 26,710), 0.002 |
| Day 1-PM | 5,106 (−5,110 to 15,323), 0.33 |
| Day 2 | 13,397 (3,180 to 23,614), 0.01 |
| Day 3 | 8,317 (−1,900 to 18,534), 0.11 |
| Day 4 | 5,835 (−4,382 to 16,051), 0.26 |
| Mean urinary OPN difference (ng) (95% CI), P value | |
| Baseline | 0.96 (−5.28 to 7.20), 0.76 |
| Day 1-AM | 12.5 (6.24 to 18.7), <0.001 |
| Day 1-PM | 4.72 (−1.52 to 10.9), 0.14 |
| Day 2 | 3.95 (−2.29 to 10.2), 0.21 |
| Day 3 | 0.57 (−5.67 to 6.81), 0.86 |
| Day 4 | −1.68 (−7.92 to 4.56), 0.60 |
| Mean urine output difference (mL) (95% CI), P value | |
| Baseline | 3.86 (−0.07 to 7.79), 0.054 |
| Day 1-AM | 3.31 (−0.63 to 7.24), 0.09 |
| Day 1-PM | 2.13 (−1.80 to 6.06), 0.29 |
| Day 2 | 3.38 (−0.55 to 7.31), 0.09 |
| Day 3 | 5.11 (1.18 to 9.04), 0.01 |
| Day 4 | 0.43 (−3.51 to 4.36), 0.83 |
Boldface type indicates significantly different values.
Compared to the baseline, rats in the VAN loading dose group showed significantly elevated urinary kidney injury molecule 1 (KIM-1) after receipt of the VAN loading dose on the morning of day 1 (213.9 ng/24 h [95% CI, 118.5 to 309.2 ng/24 h] [P < 0.001]) and on day 2 (163.6 ng/24 h [95% CI, 68.2 to 258.9 ng/24 h] [P = 0.001]) (Fig. 3a). Rats that received a VAN loading dose also had significantly elevated urinary clusterin on the morning of day 1 (16,493 ng/24 h [95% CI, 6,277 to 26,710 ng/24 h] [P = 0.002]) and on day 2 (13,397 ng/24 h [95% CI, 3,180 to 23,614 ng/24 h] [P = 0.01]) (Fig. 3b). Significant elevations in urinary osteopontin (OPN) were observed after receipt of the VAN loading dose on the morning of day 1 (12.5 ng/24 h [95% CI, 6.24 to 18.7 ng/24 h] [P < 0.001]) (Fig. 3c).
FIG 3.
Comparison of urinary injury biomarkers for rats in the loading dose arm, between treatment groups and across dosing days. The vancomycin no-load group was used as the comparator. (a) Significant differences in urinary KIM-1 were observed on the morning of day 1 (*, 213.9 ng/24 h [95% CI, 118.5 to 309.2 ng/24 h] [P < 0.001]) and on day 2 (#, 163.6 ng/24 h [95% CI, 68.2 to 258.9 ng/24 h] [P = 0.001]) among rats in the vancomycin load group. (b) Significant differences in urinary clusterin were observed on the morning of day 1 (†, 16,493 ng/24 h [95% CI, 6,277 to 26,710 ng/24 h] [P = 0.002]) and on day 2 (§, 13,397 ng/24 h [95% CI, 3,180 to 23,614 ng/24 h] [P = 0.01]). (c) A significant difference in urinary OPN was observed on the morning of day 1 (‡, 12.5 ng/24 h [95% CI, 6.24 to 18.7 ng/24 h] [P < 0.001]).
In the investigation of variable vancomycin doses with an AKI positive-control arm, the baseline urine output before treatment was not significantly different among the treatment groups (8.9 mL/24 h [95% CI, 7.0 to 10.0 mL/24 h] versus 12.6 mL/24 h [95% CI, 9.3 to 12.6 mL/24 h] versus 11.3 mL/24 h [95% CI, 8.6 to 13.3 mL/24 h] [P = 0.35]) (Table S2). Compared to the group that received folic acid (AKI positive control), rats in the low-dose VAN group had a significantly lower urine output on day 2 (−16.1 mL/24 h [95% CI, −21.7 to −10.6 mL/24 h] [P < 0.001]) and day 3 (−10.3 mL/24 h [95% CI, −15.8 to −4.8 mL/24 h] [P < 0.001]) (Table 2). Rats in the high-dose VAN group also had a significantly lower urine output on day 2 (−10.3 mL/24 h [95% CI, −15.9 to −4.8 mL/24 h] [P < 0.001]) and day 3 (−9.8 mL/24 h [95% CI, −15.3 to −4.2 mL/24 h] [P = 0.001]) than the group that received folic acid.
TABLE 2.
Marginal differences versus the AKI positive control (folic acid) as the reference groupa
| Parameter on dosing day | Value for treatment group |
|
|---|---|---|
| Low VAN dose | High VAN dose | |
| Mean GFR difference (mL/min/100 g of body wt) (95% CI), P value | ||
| Baseline (day 0) | 0.01 (−0.11 to 0.12), 0.86 | 0.01 (−0.10 to 0.12), 0.86 |
| Day 1 | 0.33 (0.22 to 0.44), <0.001 | 0.21 (0.09 to 0.32), <0.001 |
| Day 2 | 0.34 (0.23 to 0.45), <0.001 | 0.22 (0.11 to 0.33), <0.001 |
| Day 3 | 0.29 (0.18 to 0.40), <0.001 | 0.24 (0.13 to 0.35), <0.001 |
| Day 4 | 0.39 (0.29 to 0.51), <0.001 | 0.26 (0.15 to 0.37), <0.001 |
| Mean urinary KIM-1 difference (ng) (95% CI), P value | ||
| Baseline | 0.53 (−149.5 to 150.6), 0.99 | −1.78 (−151.8 to 148.2), 0.98 |
| Day 1 | −215.6 (−365.6 to −65.6), 0.005 | −85.0 (−235.1 to 64.9), 0.27 |
| Day 2 | −344.1 (−494.1 to −194.1), <0.001 | −149.1 (−299.1 to 0.92), 0.051 |
| Day 3 | −239.4 (−389.4 to −89.3), 0.002 | −154.4 (−304.4 to −4.33), 0.044 |
| Day 4 | −154.3 (−304.3 to −4.24), 0.044 | 43.4 (−106.8 to 193.3), 0.57 |
| Mean urinary clusterin difference (ng) (95% CI), P value | ||
| Baseline | 227 (−14,038 to 14,491), 0.98 | 339 (−13,925 to 14,603), 0.96 |
| Day 1 | −12,984 (−27,248 to 1,281), 0.07 | 47 (−14,218 to 14,311), 0.99 |
| Day 2 | −20,652 (−34,917 to −6,388), 0.005 | 5,221 (−9,044 to 19,485), 0.47 |
| Day 3 | −14,021 (−28,285 to 244), 0.054 | 2,058 (−12,207 to 16,322), 0.78 |
| Day 4 | −15,682 (−29,947 to −1,418), 0.031 | −16,037 (−31,018 to −1,057), 0.036 |
| Mean urinary OPN difference (ng) (95% CI), P value | ||
| Baseline | 0.32 (−19.6 to 20.3), 0.98 | −0.68 (−23.0 to 21.7), 0.95 |
| Day 1 | −38.5 (−54.7 to −22.2), <0.001 | −37.0 (−52.6 to −21.4), <0.001 |
| Day 2 | −52.1 (−68.4 to −35.8), <0.001 | −48.6 (−64.2 to −33.0), <0.001 |
| Day 3 | −33.9 (−51.1 to −16.7), <0.001 | −37.0 (−52.6 to −21.4), <0.001 |
| Day 4 | −57.6 (−73.2 to −42.0), <0.001 | −60.2 (−78.8 to −41.6), <0.001 |
| Mean urine output difference (mL) (95% CI), P value | ||
| Baseline | 3.7 (−1.8 to 9.2), 0.19 | 2.4 (−3.2 to 7.9), 0.39 |
| Day 1 | −2.9 (−8.4 to 2.7), 0.31 | −2.4 (−7.9 to 3.1), 0.39 |
| Day 2 | −16.1 (−21.7 to −10.6), <0.001 | −10.3 (−15.9 to −4.8), <0.001 |
| Day 3 | −10.3 (−15.8 to −4.8), <0.001 | −9.8 (−15.3 to −4.2), 0.001 |
| Day 4 | −0.9 (−6.5 to 4.6), 0.75 | −0.1 (−5.7 to 5.4), 0.97 |
Boldface type indicates significantly different values.
Compared to the group that received folic acid (AKI positive control), rats in the low-dose VAN group had significantly lower urinary KIM-1 on day 1 (−215.6 ng/24 h [95% CI, −365.6 to −65.6 ng/24 h] [P = 0.005]), day 2 (−344.1 ng/24 h [95% CI, −494.1 to −194.1 ng/24 h] [P < 0.001]), day 3 (−239.4 ng/24 h [95% CI, −389.4 to −89.3 ng/24 h] [P = 0.002)], and day 4 (−154.3 ng/24 h [95% CI, −304.3 to −4.24 ng/24 h] [P = 0.044]) (Fig. 4a). Rats in the low-dose VAN group also had significantly lower urinary clusterin on day 2 (−20,652 ng/24 h [95% CI, −34,917 to −6,388 ng/24 h] [P = 0.005]) and day 4 (−15,682 ng/24 h [95% CI, −29,947 to −1,418 ng/24 h] [P = 0.031]) (Fig. 4b). The urinary OPN was also significantly lower among the low-dose VAN group rats on day 1 (−38.5 ng/24 h [95% CI, −54.7 to −22.2 ng/24 h] [P < 0.001]), day 2 (−52.1 ng/24 h [95% CI, −68.4 to −35.8 ng/24 h] [P < 0.001]), day 3 (−33.9 ng/24 h [95% CI, −51.1 to −16.7 ng/24 h] [P < 0.001]), and day 4 (−57.6 ng/24 h [95% CI, −73.2 to −42.0 ng/24 h] [P < 0.001]).
FIG 4.
Comparison of urinary injury biomarkers for rats in the variable vancomycin dose arm, between treatment groups and across dosing days. The folic acid group (AKI positive control) was used as the comparator. (a) Among rats in the low-VAN group, significant differences in urinary KIM-1 were observed on day 1 (a, −215.6 ng/24 h [95% CI, −365.6 to −65.6 ng/24 h] [P = 0.005]), day 2 (b, −344.1 ng/24 h [95% CI, −494.1 to −194.1 ng/24 h] [P < 0.001]), day 3 (c, −239.4 ng/24 h [95% CI, −389.4 to −89.3 ng/24 h] [P = 0.002]), and day 4 (d, −154.3 ng/24 h [95% CI, −304.3 to −4.24 ng/24 h] [P = 0.044]). Among rats in the high-VAN group, significant differences in urinary KIM-1 were observed on day 3 (e, −154.4 ng/24 h [95% CI, −304.4 to −4.33 ng/24 h] [P = 0.044]). (b) Among rats in the low-VAN group, the urinary clusterin was significantly lower on day 2 (f, −20,652 ng/24 h [95% CI, −34,917 to −6,388 ng/24 h] [P = 0.005]) and day 4 (g, −15,682 ng/24 h [95% CI, −29,947 to −1,418 ng/24 h] [P = 0.031]). Among rats in the high-VAN group, the urinary clusterin was significantly lower on day 4 (h, −17,866 ng/24 h [95% CI, −32,131 to −3,602 ng/24 h] [P = 0.014]). (c) Among rats in the low-VAN group, the urinary OPN was significantly lower on day 1 (k, −38.5 ng/24 h [95% CI, −54.7 to −22.2 ng/24 h] [P < 0.001]), day 2 (m, −52.1 ng/24 h [95% CI, −68.4 to −35.8 ng/24 h] [P < 0.001]), day 3 (n, −33.9 ng/24 h [95% CI, −51.1 to −16.7 ng/24 h] [P < 0.001]), and day 4 (p, −57.6 ng/24 h [95% CI, −73.2 to −42.0 ng/24 h] [P < 0.001]). Among rats in the high-VAN group, the urinary OPN was significantly lower on day 1 (q, −37.0 ng/24 h [95% CI, −52.6 to −21.4 ng/24 h] [P < 0.001]), day 2 (r, −48.6 ng/24 h [95% CI, −64.2 to −33.0 ng/24 h] [P < 0.001]), day 3 (s, −37.0 ng/24 h [95% CI, −52.6 to −21.4 ng/24 h] [P < 0.001]), and day 4 (t, −60.2 ng/24 h [95% CI, −78.8 to −41.6 ng/24 h] [P < 0.001]).
Compared to the folic acid group, rats in the high-dose VAN group had significantly lower urinary KIM-1 on day 3 (−154.4 ng/24 h [95% CI, −304.4 to −4.33 ng/24 h] [P = 0.044]) (Fig. 4a). Rats in the high-dose VAN group had significantly lower urinary clusterin on day 4 (−16,037 ng/24 h [95% CI, −31,018 to −1,057 ng/24 h] [P = 0.036]). The urinary OPN was also significantly lower among the high-dose VAN group rats on day 1 (−37.0 ng/24 h [95% CI, −52.6 to −21.4 ng/24 h] [P < 0.001]), day 2 (−48.6 ng/24 h [95% CI, −64.2 to −33.0 ng/24 h] [P < 0.001]), day 3 (−37.0 ng/24 h [95% CI, −52.6 to −21.4 ng/24 h] [P < 0.001]), and day 4 (−60.2 ng/24 h [95% CI, −78.8 to −41.6 ng/24 h] [P < 0.001]).
In a direct comparison of the low- and high-dose VAN groups, rats in the low-dose VAN group had significantly lower urinary KIM-1 on day 2 (−194.9 ng/24 h [95% CI, −331.9 to −58.0 ng/24 h] [P = 0.005]) and day 4 (−197.5 ng/24 h [95% CI, −334.5 to −60.6 ng/24 h] [P = 0.005]). Rats in the low-dose VAN group also had significantly lower urinary clusterin on day 2 (−25,873 ng/24 h [95% CI, −40,137 to −11,608 ng/24 h] [P < 0.001]) and day 3 (−16,078 ng/24 h [95% CI, −30,343 to −1,814 ng/24 h] [P = 0.027]). No significant differences in urinary OPN were seen between the low- and high-dose VAN groups.
Correlation between urinary injury biomarkers and the GFR.
Spearman’s rank correlations between the GFR and urinary kidney injury biomarkers in the loading dose experimental arm are listed in Table 3. Among rats that received a vancomycin loading dose, urinary KIM-1 was significantly correlated with decreasing GFRs on the morning of day 1 (Spearman’s rho, −0.94 [P < 0.0001]), the evening of day 1 (Spearman’s rho, −0.66 [P = 0.008]), day 2 (Spearman’s rho, −0.72 [P = 0.002]), day 3 (Spearman’s rho, −0.53 [P = 0.042]), and day 4 (Spearman’s rho, −0.66 [P = 0.007]) (Fig. 5a). Urinary clusterin was significantly correlated with decreasing GFRs on the morning of day 1 (Spearman’s rho, −0.89 [P < 0.0001]), day 2 (Spearman’s rho, −0.58 [P = 0.022]), and day 4 (Spearman’s rho, −0.78 [P = 0.0006]) (Fig. 5b). Urinary OPN was significantly correlated with decreasing GFRs on the morning of day 1 (Spearman’s rho, −0.64 [P = 0.01]) (Fig. 5c).
TABLE 3.
Summary of urinary biomarker correlations with the GFR in the loading dose experimenta
| Dosing day | KIM-1 |
Clusterin |
OPN |
|||
|---|---|---|---|---|---|---|
| Spearman’s rho | P value | Spearman’s rho | P value | Spearman’s rho | P value | |
| Baseline (day 0) | 0.05 | 0.86 | 0.01 | 0.97 | 0.53 | 0.09 |
| Day 1-AM | −0.94 | <0.0001 | −0.89 | <0.0001 | −0.64 | 0.01 |
| Day 1-PM | −0.66 | 0.008 | −0.42 | 0.12 | −0.34 | 0.22 |
| Day 2 | −0.72 | 0.002 | −0.58 | 0.022 | −0.59 | 0.08 |
| Day 3 | −0.53 | 0.042 | −0.28 | 0.3 | −0.06 | 0.82 |
| Day 4 | −0.66 | 0.007 | −0.78 | 0.0006 | −0.22 | 0.43 |
Boldface type indicates significantly different Spearman correlation values.
FIG 5.
Spearman correlations of the GFR with 24-h urinary injury biomarkers (biomarkers are shown on a logarithmic scale) across all experimental days for rats in the loading dose study. Urinary KIM-1 correlations are shown in panel a, clusterin correlations are shown in panel b, and OPN correlations are shown in panel c. LOWESS, locally weighted scatterplot smoothing.
Spearman’s rank correlations between the GFR and urinary kidney injury biomarkers from the variable vancomycin doses with an AKI positive-control experimental arm are listed in Table 4. Among rats in the AKI positive-control group, urinary KIM-1 was significantly correlated with decreasing GFRs on day 1 (Spearman’s rho, −0.56 [P = 0.016]), day 2 (Spearman’s rho, −0.73 [P = 0.0006]), day 3 (Spearman’s rho, −0.52 [P = 0.03]), and day 4 (Spearman’s rho, −0.63 [P = 0.005]) (Fig. 6a). Urinary clusterin was significantly correlated with decreasing GFRs on day 2 (Spearman’s rho, −0.53 [P = 0.0248]) (Fig. 6b). Urinary OPN was significantly correlated with decreasing GFRs on day 1 (Spearman’s rho, −0.75 [P = 0.0005]), day 2 (Spearman’s rho, −0.85 [P < 0.0001]), day 3 (Spearman’s rho, −0.69 [P = 0.003]), and day 4 (Spearman’s rho, −0.63 [P = 0.012]) (Fig. 6c).
TABLE 4.
Summary of urinary biomarker correlations with the GFR in the AKI-positive control experimenta
| Dosing day | KIM-1 |
Clusterin |
OPN |
|||
|---|---|---|---|---|---|---|
| Spearman’s rho | P value | Spearman’s rho | P value | Spearman’s rho | P value | |
| Baseline | −0.01 | 0.99 | 0.21 | 0.41 | −0.26 | 0.49 |
| Day 1 | −0.56 | 0.016 | −0.37 | 0.13 | −0.75 | 0.0005 |
| Day 2 | −0.73 | 0.0006 | −0.53 | 0.0248 | −0.85 | <0.0001 |
| Day 3 | −0.52 | 0.03 | −0.18 | 0.48 | −0.69 | 0.003 |
| Day 4 | −0.63 | 0.005 | −0.31 | 0.22 | −0.63 | 0.012 |
Boldface type indicates significantly different Spearman correlation values.
FIG 6.
Spearman correlations of the GFR with 24-h urinary injury biomarkers (biomarkers are shown on a logarithmic scale) across all experimental days for rats in the variable vancomycin dose with an AKI positive control study. Urinary KIM-1 correlations are shown in panel a, clusterin correlations are shown in panel b, and OPN correlations are shown in panel c.
DISCUSSION
In this study, we found that receipt of a vancomycin loading dose resulted in a significant decline in glomerular function (as assessed by iohexol clearance) immediately after dosing, which persisted for 24 h until function was recovered to baseline values in most rats. This represented a mean 31% relative drop from the baseline before treatment and from matched controls. Urinary injury biomarker changes paralleled functional changes; significant increases in the urinary injury biomarkers KIM-1 and OPN occurred immediately after receipt of the vancomycin loading dose, which correlated with GFR declines immediately after receipt of the loading dose. Among the assessed urinary biomarkers, KIM-1 was most closely correlated with declining GFRs over the course of the experiment. Notably, neither significant changes in the GFR nor significant elevations in urinary injury biomarkers were observed among rats that did not receive a vancomycin loading dose. These findings of increased kidney injury and loss of function in response to receipt of a vancomycin loading dose are both novel and highly relevant to clinical practice. It should be noted that we studied a healthy-rat model. Kidney insults to those with baseline compromise may be even more significant.
Our findings represent the first preclinical analysis, to our knowledge, of the safety of vancomycin loading doses using specific biomarkers for kidney function and injury. In this study, rats received allometrically scaled doses of vancomycin corresponding to a guideline-recommended loading dose of 35 mg/kg of body weight/dose in humans and a maintenance dose of 15 mg/kg/dose in humans. The findings of an immediate, significant decline in the GFR and elevations of urinary KIM-1 after receipt of a vancomycin loading dose are highly relevant. Many patients who qualify for vancomycin loading doses are critically ill and have preexisting renal dysfunction. Due to the known impact of AKI on inpatient mortality, all efforts should be made to characterize and subsequently minimize potential nephrotoxic dosing (13, 14). Our preclinical findings require further investigation and validation in other animal models and humans. However, at minimum, these preliminary results provide evidence that vancomycin loading doses cause early kidney injury and loss of function in an animal model. Thus, loading doses should be considered only for those patients for whom expected gains are anticipated to outweigh the potential for increased kidney injury. The recently updated guidelines for the therapeutic monitoring of vancomycin in serious MRSA infections recommend the use of loading doses in critically ill patients with suspected or documented serious MRSA infections in order to improve efficacy (2). Clinical studies underpinning this recommendation are limited by their retrospective nature, their use of previous trough goals, and their use of nonspecific serum creatinine as a marker for both kidney function and injury (3, 4, 15–17). While multiple studies have found that the administration of a loading dose results in higher serum vancomycin levels and faster achievement of the designated pharmacokinetic-pharmacodynamic (PKPD) targets for efficacy, few have assessed the potential nephrotoxicity associated with these doses (4, 16, 18). A recent retrospective cohort study by Flannery et al. investigated the efficacy and safety of vancomycin loading doses in critically ill patients with MRSA infections (19). In terms of nephrotoxicity, that study found no difference in AKI rates among patients who received a vancomycin load and those who did not. However, serum creatinine was used to assess kidney injury, which is well known to be an insensitive and delayed marker of AKI (20, 21).
In our investigation of variable vancomycin doses with an AKI positive control, we found that receipt of intraperitoneal folic acid (AKI positive-control group) resulted in substantial kidney injury, with a nearly complete loss of function. This was characterized by a 45% increase in the urinary output and an 88% decrease in the GFR compared to baseline values over the course of the study. Folic acid, a methodologic positive control known to induce oxidative damage in proximal tubular epithelial cells (22), was demonstrated to induce much more significant kidney injury and loss of function than in rats that received low and high doses of vancomycin. These results are useful for understanding the relative damage that vancomycin can do. The urinary injury biomarker KIM-1 was most elevated in the AKI positive-control group on all experimental days, except for day 4, when the KIM-1 was most elevated in the high-dose vancomycin group. This finding shows that folic acid induced a higher degree of acute proximal tubular damage than vancomycin at the administered doses. The finding of a KIM-1 elevation on day 4 among rats in the high-dose vancomycin group is notable and may reflect cumulative kidney injury from repeated dosing or biomarker depletion because of kidney cells that are already dead. At the level of the individual rat, 3 of the 6 rats experienced an increase in urinary KIM-1 from day 3 to day 4. The measured urinary KIM-1 in the low- and high-dose vancomycin groups reached maxima of 28 and 79%, respectively (on day 2 for all groups), of the KIM-1 seen in the AKI positive-control group. Similar to the results from the loading dose arm, KIM-1 was the biomarker that most correlated with declining GFRs over the course of the experiment.
Clusterin was most elevated among rats in the high-dose vancomycin group, with the second highest amount among rats in the AKI positive-control group, while rats in the low-dose vancomycin group experienced only mild elevations from the baseline. The clusterin in the low- and high-dose vancomycin groups were 39 and 108%, respectively, of the clusterin seen in the AKI positive-control group. The observed trends in urinary clusterin may be related to its localization at both proximal and distal tubules (23, 24). Notably, vancomycin causes direct cellular toxicity to proximal tubule cells but also forms intratubular crystalline-protein complexes, leading to tubular obstruction (25). When paired with the GFR findings for this experimental arm, we see that rats in the high-dose vancomycin group experienced a mean GFR decrease of 43% from the baseline to day 1, while rats in the AKI positive-control group experienced a mean GFR decrease of 89% over the same time period. Consequently, the observation of the highest degree of urinary clusterin expression among the high-dose vancomycin group rats is not reflective of the highest degree of kidney injury or loss of function. Rather, clusterin may represent localized damage and not capture the full impact on the kidney (12). This is in contrast to folic acid-induced kidney injury, which also causes damage to the proximal tubules but does so via a slightly different mechanism involving crystal formation within and the oxidation of the tubular lumens (26); accordingly, damage may occur at other kidney sites, which is reflected in our osteopontin findings. These results underscore the need for following various biomarkers depending on the kidney insult and simultaneously understanding changes to function.
Osteopontin is a more general marker of damage throughout the nephrons and was 5 times higher, on average, among rats that received folic acid than among those that received vancomycin. When urinary injury biomarkers are displayed in the order of ascending total administered vancomycin doses (see Fig. S6 in the supplemental material), we begin to see trends where the amounts of all three biomarkers increase with increasing cumulative vancomycin doses. In addition, worse kidney injury/function can be observed across all three biomarkers among rats that received a vancomycin loading dose. Consequently, KIM-1, clusterin, and osteopontin can all aid in identifying both early/acute and longer-term kidney injury due to vancomycin. Our results help in the further development and characterization of a translational rat model, specifically regarding severe drug-induced kidney injury and the toxicodynamics of antibiotics in an animal model (6, 8, 11, 12, 27).
There are several considerations for our study. First, due to technical difficulties, two animals did not have adequate plasma collected on one day for each animal. These animals were still included in our analysis because the statistical methodology that we utilized is flexible and does not require dropping the data (such as with analysis of variance [ANOVA] methods). Both animals provided adequate plasma data on all other experimental days, in addition to daily urine and terminal kidney samples. Second, although the doses were allometrically scaled to human doses, the infusion scheme was not fully humanized. Allometrically scaled doses effectively create isometric area under the concentration-time curve (AUC) profiles, but they result in higher maximal concentrations with rapid clearance. A fully humanized model will be required to determine if similar findings are obtained with completely “humanized” maximal concentrations. It is notable that such experiments are much more technically complex and require fully tethered animal models to deliver the drug continuously to the animals and account for the increased clearance of smaller-order mammals. Humanizing by rendering the kidney anephric (as is done in infectious diseases therapeutic efficacy studies) is not possible because the primary target of this study is kidney injury. Third, we measured total urinary volumes in the rats, thereby capturing the total amount of excreted biomarkers (versus the quantification of spot samples), and standardized these to 24-h excretion. Although our findings do not change significantly when analyzed as 24-h concentrations of biomarkers, more exact methods available for laboratory studies should be considered when comparing these results to those of clinical studies where 24-h urine collection is logistically more difficult. Future preclinical studies should evaluate the GFR and urinary biomarkers over a longer period of time (i.e., 7 to 14 days) in order to understand if kidney function and injury can recover to baseline levels after the receipt of a vancomycin loading dose. Clinical studies evaluating the safety of vancomycin loading doses should be conducted in a prospective manner, employing improved markers of kidney function and injury such as the ones utilized in this study.
MATERIALS AND METHODS
Experimental design and animals.
The experimental methods were similar to those that we reported previously (7–10). All experiments were conducted at Midwestern University in Downers Grove, IL, in compliance with the guidelines of the National Institutes of Health Guide for the Care and Use of Laboratory Animals, 8th ed. (28), and were approved under Institutional Animal Care and Use Committee protocol number 3080. In brief, male Sprague-Dawley rats (n = 34; age, 8 to 10 weeks; mean weight, 274.9 g) were housed in a light- and temperature-controlled room for the duration of the study and allowed free access to water and food (see Fig. S1 and S2 in the supplemental material). All animals were placed into metabolic cages (Lab Products, Inc., Aberdeen, MD, USA) for 24-h urine collection, starting prior to dosing (day 0), and sampled every day for a period of 4 days. Animals were assigned to two experimental arms: (i) investigation of the impact of vancomycin loading doses and (i) investigation of variable vancomycin doses with an AKI positive control.
In the first arm, rats were assigned to one of two treatment groups in which they received either VAN at 220 mg/kg intravenously over 2 min followed by VAN at 100 mg/kg 12 h later, for a total dose of 320 mg/kg over 24 h (n = 9) (VAN loading dose group), or VAN at 100 mg/kg intravenously over 2 min followed by VAN at 100 mg/kg 12 h later, for a total dose of 200 mg/kg over 24 h (n = 7) (no-VAN-loading-dose group) (Fig. S3). No additional vancomycin doses were given for the remainder of the study days (i.e., days 2 to 4).
In the second arm, animals were assigned to one of three treatment groups in which they received either intraperitoneal folic acid at 250 mg/kg on day 1 followed by folic acid at 100 mg/kg on days 2 to 4 (n = 6); VAN at 150 mg/kg/day intravenously over 2 min on 4 consecutive days, for a total dose of 600 mg/kg over 96 h (n = 6); or VAN at 250 mg/kg/day intravenously over 2 min on 4 consecutive days, for a total dose of 1,000 mg/kg over 96 h (n = 6) (Fig. S4). Rats in both VAN groups also received intraperitoneal injections of normal saline (NS) at volumes equivalent to those for the folic acid group.
All animals received iohexol at 51.8 mg/day intravenously over 1 min starting at the baseline (day 0) and on all study days. Timed plasma samples were also drawn on each experimental day. Vancomycin doses were selected to approximate human doses allometrically scaled for rats (100 mg/kg in rats is equivalent to ~15 mg/kg in humans, and 220 mg/kg in rats is equivalent to ~35 mg/kg in humans). Following the completion of the dosing protocol, all rats were euthanized and underwent nephrectomies.
Chemicals and reagents.
Rats were administered clinical-grade vancomycin (lot number 167973; Fresenius Kabi, Lake Zurich, IL, USA), folic acid (lot number WXBD4723V; Sigma-Aldrich, St. Louis, MO, USA), iohexol (Omnipaque) (lot number 15025174; GE Healthcare, Inc., Marlborough, MA, USA), and normal saline for injection (Hospira, Lake Forest, IL, USA). Vancomycin was prepared by weighing and dissolving the powder in normal saline to achieve a final concentration of 100 mg/mL. Folic acid was prepared by weighing and dissolving the powder in 0.3 mM sodium bicarbonate to achieve a final concentration of 50 mg/mL. Analytical-grade iohexol (lot number LRAC5648; Sigma-Aldrich, St. Louis, MO, USA) and iohexol-d5 (lot number 28540; Cayman Chemical, Ann Arbor, MI, USA) were used for liquid chromatography-tandem mass spectrometry (LCMS) analyses of plasma samples.
Blood, urine, and kidney sampling.
Double jugular vein catheters were surgically implanted 72 h prior to the initiation of the protocol. One catheter was dedicated to blood sample draws, while drug dosing occurred via the other catheter. Blood samples were obtained from the catheter at prespecified time points (0, 30, 60, and 240 min after iohexol dosing). Each sample (0.2 mL/aliquot) was replaced with an equivalent volume of NS for the maintenance of euvolemia. Blood samples were prepared as plasma with EDTA (Sigma-Aldrich Chemical Company, Milwaukee, WI, USA) and centrifuged at 3,000 × g for 10 min (Thermo Fisher Scientific, Waltham, MA, USA). The supernatants were collected and frozen at −80°C until the time of batch analysis by LCMS.
Urine samples were collected, and the volume was measured starting from day 0. Urine collections at the day 1 morning, day 1 evening, and day 4 time points represented 12- and 4-h urine collections, respectively. The measured volumes were multiplied in order to project the full 24-h urine production volume. All other time points (i.e., day 0, day 2, and day 3) represented 24-h urine collections. Samples were centrifuged at 400 × g for 5 min at 4°C, and the resulting supernatant was collected, aliquoted, and stored at −80°C until batch analysis of renal biomarkers. Following the completion of the dosing protocol, rats were sacrificed.
GFR measurement.
The GFR was assessed by iohexol clearance, as described in “Model building,” below. Iohexol was administered intravenously as an undiluted solution on each dosing day after the administration of the study drug treatment. Rats received once-daily doses of iohexol at 51.8 mg/0.22 mL, given over 1 min, on experimental days 0 through 4.
Calibration curves in rat plasma.
Stock solutions of iohexol were prepared at a concentration of 1 mg/mL, while iohexol-d5 was prepared at a concentration of 100 μg/mL. All drugs were dissolved in purified water. An iohexol standard curve was created by diluting the stock solution with water to obtain concentrations of between 0.5 and 100 μg/mL. Iohexol-d5 was used as the internal standard and was added to each sample to obtain a final concentration of 10 μg/mL.
For each standard curve concentration, the iohexol dilution (4 μL each) was added to 36 μL of blank rat plasma. Iohexol-d5 (4 μL) was added to each standard curve concentration as an internal standard. Each standard curve concentration was then mixed with 140 μL of 0.1% formic acid in methanol, vortexed, and centrifuged at 16,000 × g for 10 min. The resulting supernatant was then transferred to LCMS vials for analysis.
Sample preparation.
For the preparation of samples, 4 μL of iohexol-d5 was added to 40 μL of sample rat plasma. Plasma samples at 30 and 60 min were diluted 1:10 with blank rat plasma. Each sample was then mixed with 136 μL of 0.1% formic acid in methanol, vortexed, and centrifuged at 16,000 × g for 10 min. The resulting supernatant was then transferred to LCMS vials for analysis.
LCMS methods.
An Agilent 1260 series liquid chromatography system paired with a 6420 triple quadrupole mass spectrometer (Agilent Technologies, Santa Clara, CA) was used to analyze plasma samples. The column temperature was maintained at 20°C. Mobile phases consisted of 0.1% formic acid in water (mobile phase A) and acetonitrile (mobile phase B) for both methods. For the analysis of iohexol, a Poroshell 120 analytical column was used (2.7 μm, 100 by 3.0 mm, part number 695975-302; Agilent Technologies). A gradient method was used to separate the analytes with solvent compositions of 5% mobile phase A–95% mobile phase B (0 to 4 min) and 97% mobile phase A–3% mobile phase B (4.1 to 7 min), at a mobile phase flow rate of 0.6 mL/min. Multiple-reaction monitoring mode was used for analyte detection, and the transitions monitored were m/z 821.8 to 803.8 and m/z 826.8 to 808.8 for iohexol and iohexol-d5, respectively.
Model building.
In order to describe iohexol clearance, a two-compartment model was created for samples obtained in the postdistribution phase (Monolix 2021R1; Lixoft, Antony, France). The evaluated covariates included log-transformed weight and treatment group. To capture daily clearance changes, each experimental day was considered a separate occasion with clearance that could vary. The selection of the final model was based on the Akaike information criterion (AIC), the between-subject variability of the population estimates, goodness-of-fit plots for observed versus predicted values, and the rule of parsimony. Empirical Bayes estimates (i.e., individual Bayesian posteriors) were utilized for individual animal parameters.
Determination of urinary biomarkers of AKI.
Microsphere-based Luminex xMAP technology was used for the determination of urinary concentrations of KIM-1, clusterin, and OPN, as previously detailed (10, 12, 27). In brief, urine samples were allowed to thaw at ambient room temperature, aliquoted into 96-well plates, and mixed with Milliplex MAP rat kidney toxicity magnetic bead panel 1 (EMD Millipore Corporation, Charles, MO, USA). For each 96-well plate, a separate standard curve was prepared and run, according to the manufacturer’s instructions. Results were analyzed and urinary biomarker concentrations were determined using five-parameter (linear and logarithmic scale) curve-fitting models (Milliplex Analyst 5.1; VigeneTech, Carlisle, MA, USA). Urinary biomarker concentrations were then normalized to 24-h totals based on individual measured urine volumes.
Statistical analysis.
A mixed-effects, restricted maximum likelihood estimation regression was used to compare urine outputs, mean weight losses, GFRs, and urinary biomarkers among the treatment groups, with repeated measures occurring over days; measures were repeated at the level of the individual rat (Stata version 16.1; StataCorp LLC, College Station, TX, USA). Margins were calculated for a full factorial of the variables, i.e., the main effects for each variable and interaction. The reference groups were the pretreatment baseline values (and folic acid in the dose gradation study). Simple effects from joint tests of drug treatment groups within each level of treatment day are reported. Spearman’s rank correlation coefficient with a Bonferroni correction was used to assess correlations between kidney injury (e.g., KIM-1, clusterin, and OPN) and function (e.g., GFR) by treatment day. Urinary biomarkers were log transformed as needed for relationship exploration. All tests conducted were two tailed, with an a priori level of statistical significance set at an α value of 0.05.
ACKNOWLEDGMENTS
The work performed in this study was supported by a pilot research award from Midwestern University (J.C.).
M.H.S. reports ongoing research contracts with Nevakar and SuperTrans Medical as well as having filed patent US10688195B2. All other authors have no other related conflicts of interest to declare.
Footnotes
Supplemental material is available online only.
REFERENCES
- 1.Aljefri DM, Avedissian SN, Rhodes NJ, Postelnick MJ, Nguyen K, Scheetz MH. 2019. Vancomycin area under the curve and acute kidney injury: a meta-analysis. Clin Infect Dis 69:1881–1887. 10.1093/cid/ciz051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Rybak MJ, Le J, Lodise TP, Levine DP, Bradley JS, Liu C, Mueller BA, Pai MP, Wong-Beringer A, Rotschafer JC, Rodvold KA, Maples HD, Lomaestro BM. 2020. Therapeutic monitoring of vancomycin for serious methicillin-resistant Staphylococcus aureus infections: a revised consensus guideline and review by the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, the Pediatric Infectious Diseases Society, and the Society of Infectious Diseases Pharmacists. Am J Health Syst Pharm 77:835–864. 10.1093/ajhp/zxaa036. [DOI] [PubMed] [Google Scholar]
- 3.Ocampos-Martinez E, Penaccini L, Scolletta S, Abdelhadii A, Devigili A, Cianferoni S, de Backer D, Jacobs F, Cotton F, Vincent JL, Taccone FS. 2012. Determinants of early inadequate vancomycin concentrations during continuous infusion in septic patients. Int J Antimicrob Agents 39:332–337. 10.1016/j.ijantimicag.2011.12.008. [DOI] [PubMed] [Google Scholar]
- 4.Truong J, Levkovich BJ, Padiglione AA. 2012. Simple approach to improving vancomycin dosing in intensive care: a standardised loading dose results in earlier therapeutic levels. Intern Med J 42:23–29. 10.1111/j.1445-5994.2011.02459.x. [DOI] [PubMed] [Google Scholar]
- 5.Brown M, Polisetty R, Gracely EJ, Cuhaci B, Schlecht HP. 2011. Weight-based loading of vancomycin in patients on hemodialysis. Clin Infect Dis 53:164–166. 10.1093/cid/cir322. [DOI] [PubMed] [Google Scholar]
- 6.Li J, Udy AA, Kirkpatrick CM, Lipman J, Roberts JA. 2012. Improving vancomycin prescription in critical illness through a drug use evaluation process: a weight-based dosing intervention study. Int J Antimicrob Agents 39:69–72. 10.1016/j.ijantimicag.2011.08.017. [DOI] [PubMed] [Google Scholar]
- 7.O’Donnell JN, Rhodes NJ, Lodise TP, Prozialeck WC, Miglis CM, Joshi MD, Venkatesan N, Pais G, Cluff C, Lamar PC, Briyal S, Day JZ, Gulati A, Scheetz MH. 2017. 24-hour pharmacokinetic relationships for vancomycin and novel urinary biomarkers of acute kidney injury. Antimicrob Agents Chemother 61:e00416-17. 10.1128/AAC.00416-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Avedissian SN, Pais GM, O’Donnell JN, Lodise TP, Liu J, Prozialeck WC, Joshi MD, Lamar PC, Becher L, Gulati A, Hope W, Scheetz MH. 2019. Twenty-four hour pharmacokinetic relationships for intravenous vancomycin and novel urinary biomarkers of acute kidney injury in a rat model. J Antimicrob Chemother 74:2326–2334. 10.1093/jac/dkz167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pais GM, Liu J, Avedissian SN, Hiner D, Xanthos T, Chalkias A, d’Aloja E, Locci E, Gilchrist A, Prozialeck WC, Rhodes NJ, Lodise TP, Fitzgerald JC, Downes KJ, Zuppa AF, Scheetz MH. 2020. Lack of synergistic nephrotoxicity between vancomycin and piperacillin/tazobactam in a rat model and a confirmatory cellular model. J Antimicrob Chemother 75:1228–1236. 10.1093/jac/dkz563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Chang J, Pais GM, Valdez K, Marianski S, Barreto EF, Scheetz MH. 2022. Glomerular function and urinary biomarker changes between vancomycin and vancomycin plus piperacillin-tazobactam in a translational rat model. Antimicrob Agents Chemother 66:e02132-21. 10.1128/aac.02132-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Scheetz MH, Pais GM, Lodise TP, Tong SYC, Davis JS, O’Donnell JN, Liu J, Neely M, Prozialeck WC, Lamar PC, Rhodes NJ, Holland T, Avedissian SN. 2021. Of rats and men: a translational model to understand vancomycin pharmacokinetic/toxicodynamic relationships. Antimicrob Agents Chemother 65:e01060-21. 10.1128/AAC.01060-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Pais GM, Avedissian SN, O’Donnell JN, Rhodes NJ, Lodise TP, Prozialeck WC, Lamar PC, Cluff C, Gulati A, Fitzgerald JC, Downes KJ, Zuppa AF, Scheetz MH. 2019. Comparative performance of urinary biomarkers for vancomycin-induced kidney injury according to timeline of injury. Antimicrob Agents Chemother 63:e00079-19. 10.1128/AAC.00079-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Cerdá J, Lameire N, Eggers P, Pannu N, Uchino S, Wang H, Bagga A, Levin A. 2008. Epidemiology of acute kidney injury. Clin J Am Soc Nephrol 3:881–886. 10.2215/CJN.04961107. [DOI] [PubMed] [Google Scholar]
- 14.Clec’h C, Gonzalez F, Lautrette A, Nguile-Makao M, Garrouste-Orgeas M, Jamali S, Golgran-Toledano D, Descorps-Declere A, Chemouni F, Hamidfar-Roy R, Azoulay E, Timsit J-F. 2011. Multiple-center evaluation of mortality associated with acute kidney injury in critically ill patients: a competing risks analysis. Crit Care 15:R128. 10.1186/cc10241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.De Waele JJ, Danneels I, Depuydt P, Decruyenaere J, Bourgeois M, Hoste E. 2013. Factors associated with inadequate early vancomycin levels in critically ill patients treated with continuous infusion. Int J Antimicrob Agents 41:434–438. 10.1016/j.ijantimicag.2012.12.015. [DOI] [PubMed] [Google Scholar]
- 16.Wang J-T, Fang C-T, Chen Y-C, Chang S-C. 2001. Necessity of a loading dose when using vancomycin in critically ill patients. J Antimicrob Chemother 47:246. 10.1093/jac/47.2.246. [DOI] [PubMed] [Google Scholar]
- 17.Roberts JA, Taccone FS, Udy AA, Vincent JL, Jacobs F, Lipman J. 2011. Vancomycin dosing in critically ill patients: robust methods for improved continuous-infusion regimens. Antimicrob Agents Chemother 55:2704–2709. 10.1128/AAC.01708-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mohammedi I, Descloux E, Argaud L, Le Scanff J, Robert D. 2006. Loading dose of vancomycin in critically ill patients: 15 mg/kg is a better choice than 500 mg. Int J Antimicrob Agents 27:259–262. 10.1016/j.ijantimicag.2005.11.009. [DOI] [PubMed] [Google Scholar]
- 19.Flannery AH, Wallace KL, Rhudy CN, Olmsted AS, Minrath RC, Pope SM, Cook AM, Burgess DS, Morris PE. 2021. Efficacy and safety of vancomycin loading doses in critically ill patients with methicillin-resistant Staphylococcus aureus infection. Ther Adv Infect Dis 8. 10.1177/20499361211005965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Bellomo R, Kellum JA, Ronco C. 2012. Acute kidney injury. Lancet 380:756–766. 10.1016/S0140-6736(11)61454-2. [DOI] [PubMed] [Google Scholar]
- 21.Doi K, Yuen PST, Eisner C, Hu X, Leelahavanichkul A, Schnermann J, Star RA. 2009. Reduced production of creatinine limits its use as marker of kidney injury in sepsis. J Am Soc Nephrol 20:1217–1221. 10.1681/ASN.2008060617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Yan L-J. 2021. Folic acid-induced animal model of kidney disease. Animal Model Exp Med 4:329–342. 10.1002/ame2.12194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hor K, Diefenbach E, Anderson PF, Nankivell BJ, Chapman JR, Murphy B, O’Connell PJ. 2012. Clusterin and other urinary proteins as predictive biomarkers of long term fibrosis and function in kidney recipients. Transplantation 94(10S):325. 10.1097/00007890-201211271-00603.22814328 [DOI] [Google Scholar]
- 24.Guo J, Guan Q, Liu X, Wang H, Gleave ME, Nguan CY, Du C. 2016. Relationship of clusterin with renal inflammation and fibrosis after the recovery phase of ischemia-reperfusion injury. BMC Nephrol 17:133. 10.1186/s12882-016-0348-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Luque Y, Louis K, Jouanneau C, Placier S, Esteve E, Bazin D, Rondeau E, Letavernier E, Wolfromm A, Gosset C, Boueilh A, Burbach M, Frère P, Verpont M-C, Vandermeersch S, Langui D, Daudon M, Frochot V, Mesnard L. 2017. Vancomycin-associated cast nephropathy. J Am Soc Nephrol 28:1723–1728. 10.1681/ASN.2016080867. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Perales-Quintana MM, Saucedo AL, Lucio-Gutiérrez JR, Waksman N, Alarcon-Galvan G, Govea-Torres G, Sanchez-Martinez C, Pérez-Rodríguez E, Guzman-de la Garza FJ, Cordero-Pérez P. 2019. Metabolomic and biochemical characterization of a new model of the transition of acute kidney injury to chronic kidney disease induced by folic acid. PeerJ 7:e7113. 10.7717/peerj.7113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Rhodes NJ, Prozialeck WC, Lodise TP, Venkatesan N, O’Donnell JN, Pais G, Cluff C, Lamar PC, Neely MN, Gulati A, Scheetz MH. 2016. Evaluation of vancomycin exposures associated with elevations in novel urinary biomarkers of acute kidney injury in vancomycin-treated rats. Antimicrob Agents Chemother 60:5742–5751. 10.1128/AAC.00591-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.National Academies Press. 2011. Guide for the care and use of laboratory animals, 8th ed. National Academies Press, Washington, DC. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material. Download aac.01276-22-s0001.pdf, PDF file, 0.6 MB (575.4KB, pdf)