Vancomycin-induced acute kidney injury after liver transplantation

Xio-Ping Shi ,Dong-Hui Lo ,Qing Xu ,Min Zhng ,Yun-Hong Lu ,Yu Gong ,Ting Wng,*

a Department of Pharmacy, Zhongshan Hospital, Fudan University, Shanghai 20 0 032, China

b Department of Pharmacy Services, Boston Medical Center, Boston, MA 02118, USA

c Department of Critical Care Medicine, Zhongshan Hospital, Fudan University, Shanghai 20 0 032, China

To the Editor:

Multidrug-resistant Gram-positive bacteriaStaphylococcusaureusandEnterococcusspecies emerge as major pathogens after liver transplantation (LT).Although vancomycin remains the best choice of drug for treating those infections,it is associated with significant nephrotoxicity.Acute kidney injury (AKI) is a common complication in liver recipients due to a combination of factors related to the recipient,donor graft,intraoperative and posttransplant events.At present,no liver recipients have been enrolled in any study evaluating the nephrotoxicity of vancomycin.It remains mostly unexplored whether the special pathophysiological state during the perioperative period along with the concurrent use of nephrotoxic agents,affects vancomycin-induced AKI(VI-AKI).Moreover,the relationship between the initial trough concentration and the incidence of nephrotoxicity in liver recipients is uncertain.Therefore,it is necessary to explore the relationship between the initial vancomycin trough concentration and the incidence of nephrotoxicity in LT patients and analyze the extent to which the use of concurrent nephrotoxic medications contributes to VI-AKI.

This was a retrospective cohort study of VI-AKI conducted on liver recipients in an academic teaching hospital in China.Serum vancomycin trough concentration was determined by chemiluminescent microparticle immunoassay (ARCHITECT PLUS i2000SR,Abbott,Irving,TX,USA).Blood samples were draw within 30 minutes before the 4th or 5th dose at steady state.VI-AKI was defined as developing AKI during therapy or within 7 days after the withdrawal of vancomycin.The definition of AKI was based on the Kidney Disease Improving Global Outcomes (KDIGO) guidelines [1] .The severity of VI-AKI was categorized according to KDIGO’s staging system.Primary outcomes were 28-day and 90-day all-cause mortality (including withdrawal of treatment by patients or family members),receipt of renal replacement,and renal outcome at discharge.The renal outcome was categorized into recovery [serum creatinine (SCr) level returned to baseline],improvement (SCr level decreased by 25% or more),and no recovery (lack of improvement in the SCr level) during hospitalization [2] .All recipients were divided into VI-AKI group and non-AKI group.Statistical analyses were conducted using the SPSS for windows (version 19.0,IBM Corp.,Armonk,NY,USA).Normally and non-normally distributed continuous variables were presented as mean ± SD or median (IQR) and were compared by independent samplet-test and Mann-WhitneyUtest,respectively.Categorical variables were compared by Chi-square test or Fisher’s exact test.A forward logistic regression model was adopted to analyze the independent risk factors of VI-AKI.AP＜0.05 was considered statistically significant.

From January 2016 to March 2019,53 recipients of cadaveric LT received intravenous vancomycin treatment were enrolled and 5 recipients with insufficient clinical data or received renal replacement therapy before vancomycin administration were excluded from further analysis.Ultimately,48 liver recipients were analyzed in the present study.The average age was 52.8 ± 8.1 years old,and the average weight was 64.2 ± 11.9 kg.The median acute physiology and chronic health evaluation II (APACHE II) score was 16.0(IQR:10.0-20.0),and the median baseline SCr level was 69.5 (IQR:56.5-90.3)μmol/L (Table 1).Moreover,77.1% (37/48) of recipients used vancomycin within 1 month after the LT.The average single dose was 13.2 ± 3.7 mg/kg,and the average daily dose was 28.4 ± 6.4 mg/kg.The median duration was 10.0 (IQR:5.0-14.0)days (Table 2).

Table 1.Baseline characteristics of enrolled liver transplantation recipients.

Table 2.Comparison of infection and the use of vancomycin between VI-AKI and non-AKI groups.

Table 3.Concomitant medications.

Therapeutic drug monitoring was performed on 156 blood samples.After reaching the steady state,35.4% (17/48) of the recipients had an initial trough concentration＞20 mg/L.Regardless of whether the dose was kept the same or reduced,the subsequent trough concentrations in 35.4% (17/48) of liver recipients were greater compared with their initial trough concentrations.Compared with the liver recipients with stable trough concentration,the length of therapy was longer (13 vs.8 days,P=0.023)in recipients with elevated trough concentration.Besides,the incidence of vancomycin trough concentration rising in liver recipients with primary biliary cirrhosis was higher compared with that of the other recipients,but the difference was not statistically significant (100% vs.32.6%,P=0.051).

During the treatment of vancomycin,13 recipients developed VI-AKI,and the incidence was 27.1%,of which 69.2% (9/13) occurred within 7 days.The initial trough concentration of the VI-AKIgroup was significantly higher compared with that of the non-AKI group (22.0 vs.17.0 mg/L,P=0.005),and the maximal steady state trough concentrations between the two groups were significantly different (22.0 vs.20.0 mg/L,P=0.016).The incidence of VI-AKI was 16.1% (5/31) when the trough concentration was ≤20 mg/L and 47.1% (8/17) when the trough concentration was＞20 mg/L(P=0.021).

Within 7 days before vancomycin administration,nine liver recipients had AKI and received vancomycin after recovery,and only two recipients developed VI-AKI.During vancomycin therapy,30 recipients were on the tacrolimus regimen,and 208 with therapeutic drug monitoring were performed.The average trough concentration of tacrolimus was 5.9 ± 2.7 mg/L in the VI-AKI group and 6.6 ± 2.5 mg/L in the non-AKI group (P=0.0 6 6).More than 60% of the recipients were also on concomitant therapy with mycophenolate mofetil,methylprednisolone,entecavir,carbapenems,acyclovir/ganciclovir,furosemide/torsemide,β-blocker,aspirin,or low-molecular-weight heparin (Table 3).

The 28-day all-cause mortality of liver recipients treated with vancomycin was 6.3% (3/48),and the 90-day mortality was 12.5%(6/48).The causes of death included liver failure,multiple organ failure,and hemorrhagic shock.There was no significant difference between the VI-AKI group and the non-AKI group in 28-day or 90-day mortality.The length of the ICU stay in the VI-AKI group was significantly longer compared with that of the non-AKI group (33.0 vs.17.0 days,P=0.040).According to the severity of VI-AKI,stage 1,stage 2,and stage 3 accounted for 76.9% (10/13),15.4% (2/13),and 7.7% (1/13),respectively.In the end,nine patients recovered,one patient did not improve on the day of hospital discharge,and three patients died due to other reasons.

Logistic regression analysis showed that diabetes mellitus(OR=18.386,95% CI:1.335-253.240,P=0.030),the initial steady state trough concentration (OR=1.123,95% CI:1.014-1.243,P=0.026),and concomitant therapy with fluconazole/voriconazole(OR=7.304,95% CI:1.107-48.193,P=0.039) were independent risk factors of VI-AKI.

The incidence of VI-AKI varies in different studies with diverse subjects,the severity of the disease,clinical diagnosis,and treatment.The incidence of VI-AKI of liver recipients (27.1%) in this study was slightly higher compared with the general patients(14.6%) [3] .In comparison it was lower than the incidence in critically ill patients (39.3%) [4] .Vancomycin is safe for liver recipients when the trough levels do not exceed the recommended upper limit of 20 mg/L.

When the trough concentration exceeds 20 mg/L,the risk of VI-AKI is increased by four-fold [3],which is consistent with the results of our study.Therefore,keeping the trough concentration of vancomycin within the target range is an important way to reduce the incidence of VI-AKI.However,many factors can affect the trough concentration,including age,sex,dosage,SCr,and fluid balance [5] .In this study,although the initial dosing met the guideline recommendation,35.4% of the recipients had initial trough concentrations＞20 mg/L.Higher than target concentration may be related to the overestimation of renal function,which is estimated using empiric formulas,including the Cockcroft Gault equation [6] .The results showed that the SCr-based formula overestimated the renal function of liver recipients compared with the estimation of creatinine clearance using 24 h urine creatinine.Liver recipients tend to have a lower SCr level due to malnutrition at the endstage of liver disease,loss of protein during surgery,and lack of mobility after surgery [7] .Therefore,accurate assessment of renal function is critical to ensure that the trough concentration of vancomycin is within the target range and to reduce the incidence of VI-AKI.

In a previous study,46.7% of the recipients with normal renal function before surgery develop AKI within 1 week after LT,mostly within 3 days [8] .In order to investigate the occurrence of VI-AKI comprehensively,more clinical factors and drugs after LT in the real-world were included in this study.Interestingly,days from transplant to vancomycin initiation,occurrence of AKI within 7 days before vancomycin,and combination therapy with tacrolimus were not related to the occurrence of VI-AKI.In our liver transplant center,intraoperative hemodynamic instability leading to kidney hypoperfusion is the predominant cause of postoperative AKI [9] .The majority of AKI cases are reversible,and 77% of recipients recover within 1 month [9] .Antibody induction with basiliximab and immunosuppressant regimen of mycophenolate mofetil and methylprednisolone are routinely used in our center to ensure delayed introduction of reduced-dose tacrolimus,which is associated with less nephrotoxicity compared with the standard dose of tacrolimus.Besides,maintaining a trough concentration of tacrolimus at a relatively low level (mean trough concentration＜7 mg/L) could also be one reason for the reduced risk of VI-AKI.

Recipients with pre-existing diabetes mellitus and postoperative concomitant administration with fluconazole/voriconazole are at higher risk of VI-AKI.Vancomycin should be avoided or used with caution for those recipients.Diabetic patients are more likely to have a renal parenchymal injury after transplant surgery.In a previous study,some liver recipients with undetected diabetic nephropathy are more prone to renal injury after surgery [10] .It is imperative to identify the recipients with a renal injury before using vancomycin to prevent the worsening of renal function.

The present study has several limitations.Firstly,it was a single-center,retrospective study,which was subject to selection bias and relied on medical record’s accuracy.Secondly,a small sample size might affect the results of the multivariate regression analysis.Thirdly,many factors could lead to AKI in liver recipients.It was difficult to distinguish the real causes of AKI.New vancomycin guideline recommends an area under the concentrationtime curve to the minimum inhibitory concentration (AUC/MIC) ratio of 400 to 600 as a pharmacokinetic/pharmacodynamic (PK/PD)target to achieve clinical efficacy and improve patient safety.Due to limitations in targeting AUC/MIC in clinical practice,we chose trough concentration as the surrogate marker for AUC/MIC.Collectively,AUC-guided dosing could be more accurate and safer than trough-based dosing.In the future,it is necessary to perform a large-scale,multi-center prospective study of vancomycin therapy,using AUC/MIC and trough concentration as monitoring parameters,to investigate the relationship between vancomycin exposure and the probability of AKI in liver recipients.

In summary,vancomycin therapy was safe for treating multidrug-resistant Gram-positive bacterial infection after LT with close monitoring.Keeping the trough concentration within 20 mg/L was a critical way to reduce the incidence of VI-AKI.Besides,it is necessary to avoid renal injury or strengthen renal protection in liver recipients with diabetes mellitus,and concurrent use of fluconazole/voriconazole.

Acknowledgments

We thank Jie Fang (Ruijin Hospital,Shanghai,China) and Ying-Fei Peng (Zhongshan Hospital,Shanghai,China) for providing feedbacks on the measurement and monitoring of vancomycin therapy.

CRediTauthorshipcontributionstatement

Xiao-PingShi:Conceptualization,Data curation,Funding acquisition,Writing -original draft.Dong-HuiLao:Data curation,Formal analysis,Writing -original draft.QingXu:Conceptualization,Methodology,Project administration,Resources,Writing -original draft.MinZhang:Writing -review &editing.Yun-HongLu:Formal analysis.YuGong:Data curation.TingWang:Conceptualization,Methodology,Resources,Writing -review review &editing.

Funding

This study was supported by a grant from the Foundation for Young Scientists of Zhongshan Hospital,Fudan University (No.2018ZSQN-28).

Ethicalapproval

This study was approved by the Ethics Committee of Zhongshan Hospital,Fudan University (No.b2019-219R).

Competinginterest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.