**Lessons learned and insights deduced from the published literature and our case experience**

To generate a typical profile of patients suffering from Van-AKI, we attempted to generalize from the group of our six patients on whom we have a complete set of clinical and laboratory data. When the serial renal function data and vancomycin levels for all 6 patients were plotted as group means on the same figure as a function of the time of vancomycin treatment, a clinical profile of Van-AKI and its recovery become apparent (Fig 7). Table 6 summarizes the group data as mean ± SE on their demographic and hematologic data (Table 6A), serial renal function through the 60th and last day of follow-up (Table 6B), vancomycin dosages (cumulative and average daily dose), duration of treatment, and serial vancomycin levels (Table 6 C). The following statistical statements could thus be made, providingsome insights and lessons on the issue of Van-AKI.

1. For our cohort of 6 patients, the mean duration of vancomycin administration was 28.2 ± 12.7 [SE] days (Table 6 C). Our group of patients, with more delayed diagnosis and

Vancomycin-Induced Nephrotoxicity 209

5. Indeed, over these interim 4.2 days, serum creatinine had increased further from 1.85 to 4.1mg/dl (∆ = 2.24 ± 0.74 mg/dl, p <0.04; Table 6 B and fig 7). There was also an

6. Serum creatinine peaked on the average 28.8 ± 9.9 days into vancomycin treatment, which occurred ~ 2 to 3 days after serum vancomycin level had reached its peak. Over this interval, there was a further increase in serum creatinine, climbing from 4.1 to 5.93 ± 1.23 mg/dl (∆= 1.83 ± 0.73 mg/dl, p<0.05; Table 6 B), associated with a further drop in CrCl from the 29.5 to 21.7 ± 5.1 ml/min (∆= - 7.9 ± 3.0 ml/min, p<0.05; Table 6 B). 7. Estimated CrCl fell from the mean baseline of 114.2 ± 15.3 to a nadir value of 21.7 ± 5.1 ml/min, p <0.005 (Fig 7; Table 6 B). This was accompanied by a significant and marked increase in serum creatinine from the baseline of 0.96 ± 0.13 mg/dl to 5.93 mg/dl, a 6 fold increase (∆ = 4.97 ± 1.12 mg/dl, p <0.01). Accordingly, there was a total loss of CrCl of 92.6 ± 13.6 ml/min, p<0.005 at the time of the worst (or peak) serum creatinine. Serum creatinine peaked 28.8 days after initiating vancomycin therapy, which was temporally 2.9 ± 1.4 days after the last dose. It is noteworthy that although vancomycin level reached the highest value 9.5 ± 4.9 hours after the last dose, serum creatinine did not reach its peak until 68.7 ± 32.4 hours after the last dose, indicating a delay of ~ 60 hours between reaching peak drug level and the full impact of functional impairment. 8. After vancomycin was stopped, serum creatinine returned towards the baseline, falling to a nadir value of 1.57 ± 0.22 mg/dl (p<0.02 vs. peak creatinine of 5.93 mg/dl, ∆ = 4.36 ± 1.21 mg/dl; Table 6 B). This was associated with an estimated CrCl of 72.3 ± 12.4 ml/min at the point of maximal functional recovery (∆ = 50.7 ± 13 ml/min, p <0.01 compared to the lowest value at the worst time of the ARF; Table 6 B). The maximal functional improvement was noted 30.8 ± 10.4 days after the last dose, or 59.0 ± 15.7

9. Despite 2 months since vancomycin had been initiated and 1 month after the last dose of vancomycin, once AKI had developed, the functional recovery was unfortunately incomplete even at the time of maximal improvement in the serum creatinine. In the short-term follow-up of ~ 31 days after the last dose or after the onset of ARF, there remained a residual elevation of serum creatinine (∆ = 0.61 ± 0.2 mg/dl, p< 0.04), associated with what appeared to be an irreversible decline of CrCl of 41.9 ± 12.9 ml/min, p <0.03 (Table 6 B & Fig 7). Compared to the baseline CrCl of 114.2 ml/min, this represented a 37 % residual loss of renal function. It is possible that with longer

10. Vancomycin level was ordered and monitored in the recovery period based on decisions by the individual clinicians without any discernible or uniform pattern. In this cohort of 6 patients, the lowest drug level (17.5 ± 7.5 mg/L) was observed 8.0 ± 2.5 days after the last dose (Table 6 C). It is noteworthy that this concentration of vancomycin, 8 days after stopping vancomycin, still fell into if not exceeded most

11. Compared to the duration of vancomycin administration of 28.2 days, the average recovery time for maximal return of renal function was 3.2 ± 1.7 fold longer, indicating significant clinical morbidities (and associated financial burdens) posed by a

additional loss of CrCl of 49.7 ± 19.3 ml/min, p <0.05; Table 6 B and Fig 7).

mg/L as a risk threshold (Ingram et al, 2008).

days after the first dose of vancomycin (Table 6 C).

recommended therapeutic ranges.

preventable medication complication.

period of follow up further return of renal function may ensue.

(Goetz & Sayer, 1993); (b) mean serum creatinine > the 1.7 mg/dl postulated as a risk thresh-hold (Pritchard et al, 2008); and (c) mean steady state vancomycin levels > 28

more severe renal failure, certainly confirmed the general impression from both prospective and retrospective studies that the chronicity of administration beyond 7 to 21 days posed independent risks for nephrotoxicity (Goetz & Sayer, 1993; Hidayat et al, 2006; Pritchard et al, 2008). The cumulative dose was 59.3 ± 23.6 g, yielding an average daily dose of 2.4 ± 0.6 g/day (Table 6 C). Since mean body weight (BW) was 85.5 ± 2.8 kg (Table 6A), the average dose was 28 mg/kg BW per day. The last dose given averaged 1.33 ± 0.17g (Table 6 C), equivalent to 15.5 mg /kg body weight. In retrospect if not prospectively, this dose should be considered excessive and probably unnecessary because at the time of this last dose, both serum creatinine (4-6 fold of baseline) and vancomycin level (30 to 70 mg/L) for that day were known to have been significantly elevated (Table 6 B & C and Fig 7). We believe this last dose could have been much reduced or simply skipped (had greater restraints been exercised and more circumspection applied), considering the fore-knowledge of a significantly elevated serum creatinine (mean being 4.1 mg/dl, Fig 7) and high vancomycin levels of 65-70mg/L (Fig 7Fig 7). Existing literature suggests that both pre-existing renal insufficiency (as denoted by serum creatinine >1.7 mg/dl) (Pritchard et al, 2008) and elevated steady state serum vancomycin levels > 28 mg/ (Ingram et al, 2008) or elevated trough levels >14 mg/L (Rybak et al 1990) are independent risk factors for ARF during vancomycin therapy (Hidayat et al, 2006; Pritchard et al, 2008; Lodise et al, 2008).


2. Peak vancomycin levels (group mean 70 ± 10 mg/L), which we defined here as the individual patient's highest value at any time during therapy and irrespective of when it was last given, was observed 26.2 ± 11.1 days after the first dose and 9.5 ± 4.9 hours after the last dose (Fig 7, Table 6 C). Some patients received their last dose, despite in retrospect already exhibiting a very high peak vancomycin level a few hours earlier (Table 6 C). At the point when vancomycin was discontinued, drug level was 65.1 ± 12.5 mg/L (Table 6 C). We would therefore propose that vancomycin orders be written daily (if not every 12 h), similar to coumadin orders in the titration phase of trying to achieve certain target levels or attaining a steady-state concentration, instead of one generic scheduled order for several days at a time. Needless to say, in retrospect, both the peak vancomycin levels for individual patients and the level on the day of stopping vancomycin (65.1 mg/L; Fig 7) were substantially higher than the recommended steady-state cut-off level of 28 mg/L (Ingram et al, 2008) or the trough cut-off level of 14

3. On the day of peak vancomycin level (day 26.2 of therapy), average serum creatinine had already risen to 4.1 ± 0.78 mg/dl (∆= 3.14 ± 0.67 mg/dl, p<0.005 vs. baseline; Table 6 B) and the corresponding creatinine clearance (CrCl) had fallen to 29.5 ± 5.5 ml/min,

4. Prior to the actual peak serum vancomycin levels documented on day 26, an earlier serum vancomycin had also been obtained, on ~ day 22 (± 10) or ~ 4.2 ± 2 days earlier. This "pre-peak" vancomycin level (30 ± 9.8 mg/L) was already significantly elevated (Table 6 C & Fig 7). On that day, serum creatinine (1.85 ± 0.48 mg/dl) was numerically higher than baseline (∆= 0.89 ± 0.4, n.s.). This corresponded to an estimated CrCl of 79.2 ± 21.1 ml/min, or a drop of 35 ± 14 ml/min from baseline, though just shy of significance (p=0.06) (Table 6 B). It is noteworthy that in retrospect, three independent forewarning signs for nephrotoxicity had already emerged on this 22nd day of therapy: (a) duration of administration > the 3rd week previously suggested as a risk factor

(∆ = - 84.7 ± 11.2 ml/min, p<0.001 vs. baseline; Table 6 B).

et al, 2008).

mg/L (Pitchard et al, 2008).

more severe renal failure, certainly confirmed the general impression from both prospective and retrospective studies that the chronicity of administration beyond 7 to 21 days posed independent risks for nephrotoxicity (Goetz & Sayer, 1993; Hidayat et al, 2006; Pritchard et al, 2008). The cumulative dose was 59.3 ± 23.6 g, yielding an average daily dose of 2.4 ± 0.6 g/day (Table 6 C). Since mean body weight (BW) was 85.5 ± 2.8 kg (Table 6A), the average dose was 28 mg/kg BW per day. The last dose given averaged 1.33 ± 0.17g (Table 6 C), equivalent to 15.5 mg /kg body weight. In retrospect if not prospectively, this dose should be considered excessive and probably unnecessary because at the time of this last dose, both serum creatinine (4-6 fold of baseline) and vancomycin level (30 to 70 mg/L) for that day were known to have been significantly elevated (Table 6 B & C and Fig 7). We believe this last dose could have been much reduced or simply skipped (had greater restraints been exercised and more circumspection applied), considering the fore-knowledge of a significantly elevated serum creatinine (mean being 4.1 mg/dl, Fig 7) and high vancomycin levels of 65-70mg/L (Fig 7Fig 7). Existing literature suggests that both pre-existing renal insufficiency (as denoted by serum creatinine >1.7 mg/dl) (Pritchard et al, 2008) and elevated steady state serum vancomycin levels > 28 mg/ (Ingram et al, 2008) or elevated trough levels >14 mg/L (Rybak et al 1990) are independent risk factors for ARF during vancomycin therapy (Hidayat et al, 2006; Pritchard et al, 2008; Lodise (Goetz & Sayer, 1993); (b) mean serum creatinine > the 1.7 mg/dl postulated as a risk thresh-hold (Pritchard et al, 2008); and (c) mean steady state vancomycin levels > 28 mg/L as a risk threshold (Ingram et al, 2008).


Vancomycin-Induced Nephrotoxicity 211

c. Patients infected by MRSA requiring a high vancomycin MIC and therefore high trough levels of ~ 15-20 mg/L and urgent attainment of such high levels by rapid escalation. Only daily vancomycin and daily creatinine level would allow achievement of such target levels without undue risks for unrecognized renal toxicity (also see point # 4

Pre-emptive renal consultation at the earliest signs of potential nephrotoxicity may prevent

4. Scheduled vancomycin dosing should be discouraged, and if necessary, written no longer than every 2 to 3 consecutive days because of the known narrow therapeutic window of vancomycin. The duration should be further limited to one day at a time in

a. Elevated serum creatinine, whether before or during vancomycin therapy (e.g. cases 3, 4

By routinely refraining from a multi-day scheduled order, physicians could use the latest serum creatinine and vancomycin data to adjust the next dose to avoid further damage which otherwise could easily happen with a standing order. The danger of the latter approach was statistically and pictorially shown in our six patients by the delayed stoppage despite theoretically prior knowledge of vancomycin levels of 70 mg/L and creatinine elevation to 4-fold of the baseline 2 days earlier (Fig 7, Tables 6 B & C). The rationale and justification are identical to writing daily coumadin orders in similar non-steady states like dose titration or escalation. The trade off for the inconvenience and extra though manageable work load would be a far lower incidence of AKI [(and perhaps fewer cases of

5. Based on inferences from our statistical analyses, we recommend three simple practicable thresholds for drastic dose reduction or complete stoppage of vancomycin.

We urge serious considerations for an immediate stoppage if any 2 of these 3 criteria are present, at least temporarily withholding vancomycin until additional tests show stable or

6. We would re-emphasize the observation from classical renal physiology that serum creatinine is a very insensitive index of renal function in terms of detecting early decline in GFR. It is also grossly inaccurate in quantifying the loss of renal function, especially when the absolute values are below 1.5 mg/dl or when the changes occur between 0.5 and 2 mg/dl. Sole reliance on the increases of serum creatinine or the absolute values as

c. 10- to 12-h trough levels ≥ 20-25 mg/L (Table 6 C, Fig 7) (Ingram et al, 2008).

b. Those in the early non-steady state of initiating therapy (e.g. cases 3 and 5), and c. Those requiring rapid dose escalations to meet certain target levels (cases 1, 4, and 6). In cases 1, 4, and 6, for instance, toxic trough vancomycin levels were created and found because of the rapid escalation without attaining a relative steady state at each incremental step. Our experience amply confirmed the virtually identical experience reported in the one case by Barraclough et al in 2007. We would support and re-emphasize their caution regarding the need not only to monitor very tightly and frequently during rapid dose

below).

e. Pre-existing CKD.

d. Protracted infusion (>2 weeks).

costly AKI and y hospitalization.

f. ARF or those with fluctuating serum creatinine.

three particularly susceptible patient cohorts.

escalation but also ordering one dose at a time.

a. A doubling of baseline serum creatinine,

and 6), since vancomycin excretion is already impaired,

chronic kidney disease (CKD)] and reduced health care expenses.

improved renal function and significant decline in vancomycin levels.

b. A serum creatinine ≥ 1.5 mg/dl for any adult patients,
