Risk factors for BK virus viremia and nephropathy after kidney transplantation: A systematic review

Accepted Manuscript Title: Risk factors for BK virus viremia and nephropathy after kidney transplantation: a systematic review Authors: Baptiste Demey, Claire Tinez, Catherine Franc¸ois, Franc¸ois Helle, Gabriel Choukroun, Gilles Duverlie, Sandrine Castelain, Etienne Brochot PII: DOI: Reference:

S1386-6532(18)30257-9 https://doi.org/10.1016/j.jcv.2018.10.002 JCV 4070

To appear in:

Journal of Clinical Virology

Received date: Revised date: Accepted date:

26-7-2018 13-9-2018 8-10-2018

Please cite this article as: Demey B, Tinez C, Franc¸ois C, Helle F, Choukroun G, Duverlie G, Castelain S, Brochot E, Risk factors for BK virus viremia and nephropathy after kidney transplantation: a systematic review, Journal of Clinical Virology (2018), https://doi.org/10.1016/j.jcv.2018.10.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Risk factors for BK virus viremia and nephropathy after kidney transplantation: a systematic review.

IP T

Baptiste Demey1,2, Claire Tinez1,2, Catherine François1,2, François Helle2, Gabriel

SC R

Choukroun3, Gilles Duverlie1,2 Sandrine Castelain1,2 and Etienne Brochot1,2

Department of Virology, Amiens University Medical Center, Amiens, France

2

AGIR Research Unit, EA 4294, Jules Verne University of Picardie, Amiens, France

3

Department of Nephrology, Amiens University Medical Center, Amiens, France

*

CC E

PT

ED

M

Corresponding author: Dr Etienne Brochot Laboratoire de Virologie Centre de Biologie Humaine - CHU Amiens Avenue René Laënnec, Salouel F-80054 Amiens cedex 1-France Phone: +33 322 080 764 Fax: +33 322 087 009 E-mail: [email protected]

A

N

U

1

Word counts for main body of manuscript: 2745

A

Word counts for the abstract: 241

1

ED

M

A

N

U

SC R

IP T

Graphical abstract

Highlights:

BK Polyomavirus (BKPyV) reactivation has become a difficult problem to for

PT



kidney transplant community no prophylactic or specific curative treatment



it is imperative to have close biological monitoring and early identification of

CC E



A

patients at risk



numerous dispersed data in the literature and often contradictory

2

Abstract

In the last 20 years, the management of BK polyomavirus (BKPyV) reactivation in kidney transplant patients has become a true challenge for the transplant community. The only

IP T

treatment option is based on the early identification of at-risk patients. The number of reported risk factors for BKPyV reactivation has increased markedly in the literature last years, although

SC R

they are sometimes in an unclear or contradictory manner.

Our purpose is to provide a systematic review and meta-analysis of risk factors for BKPyV viremia and nephropathy described in multivariate analyses.

U

The PubMed database was searched for prospective or prospectively-based observational

N

studies on risk factors for BKPyV viremia and/or nephropathy. Our qualitative assessment of

A

risk factors was based on the odds ratios and hazard ratios calculated in multivariate regression

M

analyses.

Of the 241 publications screened, 34 were included in the qualitative analysis. In all, 144 and

ED

19 distinct factors were analyzed for BKPyV viremia and for BKPyV nephropathy,

PT

respectively. Our evaluation highlighted eight risk factors for BKPyV viremia: a tacrolimus regimen, a deceased donor, a male recipient, a history of previous transplant, age at

CC E

transplantation, ureteral stent use, delayed graft function, and acute rejection episodes increased the risk of BKV viremia to varying extents. Tacrolimus and acute rejection episodes were also associated with a higher incidence of BKPyV nephropathy.

A

BKPyV reactivation is a serious complication after renal transplantation. With a view to combating this problem, existing data should be published in full, and new prospective international multicenter studies should be performed.

3

Keywords: BK polyomavirus; kidney transplantation; risk factors; nephropathy

IP T

Introduction

In recent years, BK polyomavirus (BKPyV) infection has emerged as a major problem in

SC R

kidney transplant recipients. About 90% of the latter are seropositive for BKPyV(Kean et al., 2009). The virus reactivates periodically with urinary shedding but remains asymptomatic in immunocompetent adults. The immunosuppressive therapy used to prevent rejection after solid

U

organ transplantation allows viral infections to emerge. In particular, BK virus can reactivate

N

after kidney transplantation, and may induce progressive disease in three successive stages:

A

viruria, viremia, and then (if viral replication persists) nephropathy. The current lack of

M

effective prophylactic or curative treatments obliges physicians to close monitor clinical and laboratory parameters, in an attempt to detect reactivation as early as possible(Dall and

ED

Hariharan, 2008). Since the 2000s, many factors have been shown to potentially increase the risk of BKPyV infection in kidney transplant recipients(Pham et al., 2014). These factors can

PT

be used to guide the patient’s care and adapt the monitoring as a function of the risk. Depending

CC E

on the study, the risk analysis has been based on BKPyV viruria, viremia, nephropathy, the treatment, or combination of these. However, the contradictory nature of the literature data on risk factors identified at the different stages of the disease is barely considered in clinical

A

practice(Kuypers et al., 2010). Although a parameter might be reported as a risk factor in one study, it might be irrelevant (or only significant for a different stage of the infection) in another. Moreover, most of the literature studies are descriptive, and seek to identify significant clinical or laboratory characteristics by applying an overly simple statistical test (e.g. Student’s t-test). For some reason, few studies have featured multivariate analyses that are commonly used to 4

assess the association between multiple risk factors and to eliminate confounding variables. The objective of the present study was to provide a systematic review of the risk factors for BKPyV viremia and nephropathy in kidney transplant patients, on the basis of multivariate analyses.

Method

IP T

This systematic review was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)(Moher et al., 2009) guidelines. The

SC R

protocol was not registered at the International Prospective Register of Systematic Reviews (PROSPERO).

U

PICOS

M

- Intervention: kidney transplantation.

A

- Population: kidney transplant recipients.

N

The study was performed using the following PICOS parameters:

ED

- Comparison: patients who developed BKPyV viremia and/or nephropathy following renal transplantation, versus patients who did not.

PT

- Outcomes: BKPyV viremia and/or nephropathy. - Studies: prospective cohort studies and studies based on prospectively recorded

CC E

databases with at least 12 months of follow-up.

A

Inclusion and exclusion criteria Only studies published in international, peer-review journals were included. To be included

in a qualitative analysis, the studies had to be based on databases evaluating risk factors for the development of BKPyV infection (viremia and/or nephropathy) following kidney transplantation. Studies lacking a multivariate regression analysis were excluded.

5

Search strategy The PubMed database was searched on September 30th, 2017, with the following search terms: (BK virus) AND ((risk factors) OR (epidemiologic factors) OR (predictive factors)) AND ((renal) OR (kidney)). The search strategy was devised and implemented by BD and CT.

IP T

There were no restrictions on the publication year but the search was restricted to full-text articles published in English.

To cover the search strategy’s limitations, the reference lists of reviews found in the search

SC R

were also examined for relevant publications. The publications’ eligibility was assessed by BD

and CT. Once a publication’s eligibility for inclusion was confirmed, the full text was obtained

A

M

Definitions of outcome measures

N

U

and analyzed in detail.

A risk factor for BKPyV infection was defined as a preoperative, per-transplantation or

ED

post-transplantation patient-related factor that is thought to increase the risk of onset of BKPyV viremia and/or nephropathy.

PT

In the present study’s qualitative analysis, we considered risk factors for BKPyV viremia

CC E

to be factors that (i) had been studied in at least two multivariate analyses and (ii) were statistically significant (p<0.05) in at least one multivariate analysis. In the present study, factors that were significant in a univariate analysis but never in a multivariate analysis were

A

not considered as risk factors. A significant result in a single multivariate analysis was the sole criterion for risk factors

for BKPyV nephropathy (BKPyVN); Inclusion in at least two multivariate analyses was not required.

6

BKPyV viremia was defined here as the detection of a BKPyV viral load in blood,

IP T

regardless of the latter’s magnitude.

SC R

Risk factors

In our assessment for BKPyV viremia, we found that 144 distinct factors had been analyzed statistically (Figure 1). The publications described 33 factors that significantly increased the

U

risk of BKPyV viremia onset in a multivariate analysis, and 8 that significantly decreased the

N

risk of onset. Eight of the 33 positive risk factors were significant in at least one multivariate

A

analysis and had been studied in at least one other multivariate analysis (regardless of their

M

statistical significance in the latter).

Nineteen distinct factors were analyzed statistically for BKPyVN. All 19 factors

ED

significantly increased the risk of BKPyVN in a multivariate analysis. Four of these factors had

A

CC E

PT

been analyzed in a second analysis.

7

Results Studies included We reviewed 34 prospective cohort studies and studies based on prospectively recorded

IP T

databases. A total of 13135 patients had been evaluated, and 1923 of these (14.6%) presented BKPyV viremia. Six studies featured a multivariate analysis of risk factors for BKPyVN; a total

SC R

of 2821 patients had been evaluated, and 149 of these presented BKPyVN (5.3%).

The identified risk factors fell into various classes: the immunosuppressive regimen, donor or recipient characteristics, surgery, post-transplantation outcomes, viral serology, and the time

A

N

U

course of BKPyV viremia.

M

Immunosuppressive regimens

ED

When considering the various immunosuppressive regimens, only the administration of tacrolimus was classified as a risk factor in our systematic review. This factor was included in

PT

six multivariate analyses(Hasegawa et al., 2014; Huang et al., 2014; Masutani et al., 2013; Moura et al., 2017; Vu et al., 2014; Wunderink et al., n.d.) (Table 1), and was found to

CC E

significantly increase the risk of BKPyV viremia in two of these (5.27- and 2.9-fold, respectively). The hazard ratio (HR) was not statistically significant in the study with the largest

A

sample size.

Donor/recipient characteristics Four risk factors were related to pre-operative donor or recipient characteristics (Table 2).

The corresponding studies showed strong associations between BKPyV viremia on one hand and deceased donor condition(Dogan et al., 2017; Huang et al., 2014; Knight et al., 2013; Moura 8

et al., 2017; Santoveña et al., 2015; Vu et al., 2014) and male recipient gender on the other(Dogan et al., 2017; H H Hirsch et al., 2013; Masutani et al., 2013; Moura et al., 2017; Vu et al., 2014; Wunderink et al., n.d.). Despite clear trends, half of the results were not significant. Of the five articles(Kayler et al., n.d.; Masutani et al., 2013; Thölking et al., 2016; Wingate et al., n.d.; Wunderink et al., n.d.), only Kayler et al. reported a low but statistically significant

IP T

odds ratio (OR: 1.02; p=0.016) for the recipient’s age. Jacobi et al. found that previous transplantation was predictive of BKPyV viremia (OR [95% confidence interval (CI)] = 2.74

SC R

[1.05-7.15](Jacobi et al., 2013) but this factor was not significant in two other studies(Vu et al.,

A

N

U

2014; Wunderink et al., n.d.).

M

Surgery

The four multivariate analyses of the use of a ureteral stent during surgery generated

ED

significant risk ratios ranging from 1.36 to 3, with narrow CIs(Kayler et al., n.d.; Maliakkal et

n.d.) (Table 3).

PT

al., 2017; Wingate et al., n.d.). However, Brennan et al. did not report the CI(Brennan et al.,

CC E

Post-transplantation outcomes Following kidney transplantation, two outcomes were associated with a higher risk of

A

BKPyV reactivation, although a causal relationship was not determined (Table 4). Firstly, Mindlova et al. showed that delayed graft function was associated with an increased risk of BKPyV viremia (OR [95%CI] = 3.38 [1.38-8.24])(Mindlova et al., n.d.). However, this was not confirmed in a larger cohort(Vu et al., 2014). Furthermore, univariate analyses in six studies evidenced a higher incidence of viremia in patients who experienced at least one episode of

9

acute rejection during follow-up(Elfadawy et al., 2013; Hässig et al., n.d.; Jacobi et al., 2013; Masutani et al., 2013; Santoveña et al., 2015; Vu et al., 2014). In three publication, this finding was still significant in a multivariate analysis BKPyV nephropathy

IP T

In the present systematic review, we noted that only six studies performed a multivariate analysis of risk factors for BKPyVN. In each analysis, at least one determining factor was

SC R

nevertheless significantly associated with BKPyVN. A total of 19 different risk factors were described (Table 5). Tacrolimus immunosuppression, polyclonal antibody induction, recipient

age, and acute rejection episodes were each evaluated in more than one study. Like BKPyV

U

viremia, tacrolimus immunosuppression (HR [95%CI] = 3.3 [1.5-7.6])(Hsiao et al., 2016;

N

Prince et al., 2009), recipient age (HR [95%CI] = 1.04 [1-1.07])(Theodoropoulos et al., 2018;

A

Wunderink et al., n.d.) and acute rejection episodes (HR [95%CI] = 3.18 [1.42-7.12])(Hsiao et

M

al., 2016; Theodoropoulos et al., 2018) significantly increased the risk of BKPyVN in one of

ED

the two studies. Interestingly, polyclonal antibody induction greatly increased the risk of BKPyVN, with an OR [95%CI] of 11.04 [2.94-41.52]) in one study(Smith et al., 2007) and an

PT

HR [95%CI] of 6.6 [2.3-18.9]) in another(Hsiao et al., 2016). With regard to treatments, a mofetil mycofenolate regimen(Prince et al., 2009), tacrolimus blood levels(Manitpisitkul et al.,

CC E

2009) and prednisone exposure(Manitpisitkul et al., 2009; Wunderink et al., n.d.) increased the risk of BKPyVN. Donor-recipient ABO incompatibility multiplied the probability of BKPyVN development by a factor of 2.32(Sharif et al., 2012). Along with donor BKPyV seropositivity

A

(HR [95%CI] = 2.89 [1.33-6.29](Wunderink et al., n.d.), African-American ethnicity might be a predictive recipient characteristic for BKPyVN (HR [95%CI] = 2.89 [1.366.14].(Theodoropoulos et al., 2018) The graft’s cold ischemia time was reported to enhance the onset of BKPyVN (HR [95%CI] = 1.18 [1.04-1.35])(Hsiao et al., 2016). Cytomegalovirus infection during follow-up was also related to a higher risk of BKPyVN (HR [95%CI] = 2.72 10

[1.19-6.24])(Theodoropoulos et al., 2018). Lastly, Elfadawy et al. showed that initial BKPyV viremic load above 185,000 copies per ml and a peak BKPyV viremic load of reaching 223,000 copies per ml at any time were strong predictive factors for BKPyVN (ORs = 113.25 and 70.5, respectively)(Elfadawy et al., 2013).

IP T

Discussion The present review found that tacrolimus regimen, deceased donor, male recipient and

SC R

acute rejection episodes were most frequently included in multivariate analyses, with 6 studies each. We found two highly significant results for tacrolimus, and three for the other variables.

Along with the statistical significance of the multivariate analyses, it is important to remember

U

that the factor has to be significant (p<0.05 or p<0.1) in a univariate analysis if it is to be

N

included in a logistic regression analysis. Hence, these four factors appear to be the most robust

A

risk factors for BKPyV viremia. Moreover, tacrolimus and acute rejection were also associated

M

with a greater probability of developing BKPyVN. These findings support (at least in part) our

ED

comment in the introduction part because in the absence of treatment, BKPyVN corresponds to the natural course of the disease after BKPyV viremia in kidney transplant recipients. The use

PT

of a ureteral stent use after transplantation was found to be a constant risk factor for BKPyV viremia, with statistical significance in all four multivariate studies. Thus, the present review

CC E

enabled us to define a panel of five variables that could be used to predict the development of BKPyV infection - especially viremia - after kidney transplantation.

A

Although recipient age met the criteria for inclusion in our qualitative review, only one

study reported a statistically significant OR [95%CI]. The effect of age on BKPyV viremia was very weak, with an OR of 1.02 (p=0.016); in routine clinical practice, this might not be enough to accurately detect an at-risk patient.

11

Previous transplantation was reported to be predictive of BKPyV viremia in one of three studies. However, the study did not specify the reasons for previous graft failure. An earlier BKPyV infection might have skewed the statistics and might explain the contradictory conclusions of the other studies. Delayed graft function was examined in two studies, which reported contrasting

IP T

conclusions. However, it is not clear whether the studies measured exactly the same outcome.

Vu et al. did not define the outcome, and Mindlova defined the outcome as the need for

SC R

hemodialysis in the week following kidney transplantation. Further cohort studies might be able

to establish whether or not delayed graft function parameter is a risk factor for BKPyV viremia. Very few studies identified predictive factors for BKPyVN; this is why we decided to

U

analyze all the risk factors reported in a single multivariate analysis (rather than the two or more

N

analyses required for risk factors for BKPyV viremia). Given the low frequency of BKPyVN,

A

the studies included in the present review had smaller sample sizes and thus lacked statistical

M

power. Interestingly, Manitpisitkul et al. studied a cohort within which one in four of the

ED

patients presented with BKPyVN(Manitpisitkul et al., 2009). This proportion is much higher than the prevalence usually reported (1-10%)(H. H. Hirsch et al., 2013) and could have led to

PT

bias.

The present review focused on variables that have been studied several times in

CC E

multivariate analyses. However, we also found risk factors that were highly significant in a single publication. These factors included Asian ethnicity(Knight et al., 2013), HLA B24 or

A

B25 matching(Dogan et al., 2017), BKPyV seropositivity(Wunderink et al., n.d.), the duration of diabetes prior to transplantation,(Mindlova et al., n.d.) and ABO incompatibility;(Sharif et al., 2012) these factors could also be considered when seeking to evaluate the likelihood of BKPyV reactivation following kidney transplantation. Further evaluation in prospective cohort studies is needed to confirm these factors’ influence on BKPyV reactivation. We remain

12

doubtful with regard to Vu et al.’s report of statistically significant ORs whose CI encompassed the value of 1 in a study of interferon lambda polymorphisms(Vu et al., 2014). Unfortunately, we were unable to perform a quantitative analysis of risk factors for BKPyV viremia. The definitions of outcomes varied from one study to another, which explains the large number of factors identified (144). Given that some studies reported ORs and others reported

IP T

HRs, we were unable to calculate pooled risk ratios. This limitation was exacerbated by a lack of results tables and information on demographic factors.

SC R

Despite our best efforts, a fully objective evaluation of risk factors for BKPyV viremia and nephropathy has yet to be performed. We were able to clarify some points, although contradictory results persist. These contradictions are probably due to differences between

U

monitoring protocols and PCR thresholds, which alter the sensitivity with which infections can

N

be detected.

A

The present review focused on BKPyV viremia, i.e. the stage at which treatment can be

M

initiated. Although BKPyVN is the true cause of graft loss, it might be avoidable if BKPyV

ED

viremia is quickly contained. The medical profession is increasingly well informed and cautious about the BKPyV issue; hence, BKPyVN tends to be less common these days, and viremia is

PT

the main warning sign. It would therefore be interesting to more evaluate the risk factors for BKPyV viruria above a threshold of 7 log10 copies/mL(Brochot et al., 2018), given that (i)

CC E

viruria is the earliest manifestation of infection, and (ii) this threshold is critical for the transition to viremia(H. H. Hirsch et al., 2013).

A

Until a specific treatment is found, BKPyV will remain an insoluble problem. With a view

to prevention in the meantime, it is essential to identify potentially modifiable factors that favor infection. The present systematic review highlighted the most relevant risk factors for BKPyV viremia after kidney transplantation. However, we suggest that the transplantation community should (i) establish consensual definitions of the most frequently evaluated outcomes, and (ii)

13

standardize the statistical methods used. Lastly, international studies and consortia may help to definitively identify the factors that increase the risk of developing a BKPyV infection after kidney transplantation.

IP T

Author Contribution Statement.docx

SC R

BD, SC and EB conceived of the presented idea BD, CT and EB wrote the manuscript with support from

BD, CF and FH performed the analytic calculations and performed statistical analysis

U

GC supervised the project

A

N

All authors discussed the results and contributed to the final manuscript.

ED

M

Funding: not applicable

CC E

PT

Competing of interest: No conflict of interest

References

A

Brennan, D.C., Agha, I., Bohl, D.L., Schnitzler, M.A., Hardinger, K.L., Lockwood, M., Torrence, S., Schuessler, R., Roby, T., Gaudreault‐Keener, M., Storch, G.A., n.d. Incidence of BK with Tacrolimus Versus Cyclosporine and Impact of Preemptive Immunosuppression Reduction. Am. J. Transplant. 5, 582–594. https://doi.org/10.1111/j.1600-6143.2005.00742.x Brochot, E., Descamps, V., Handala, L., Faucher, J., Choukroun, G., Helle, F., Castelain, S., François, C., Duverlie, G., Touzé, A., 2018. BK polyomavirus in the urine for follow-up of kidney transplant recipients. Clin. Microbiol. Infect. 0. https://doi.org/10.1016/j.cmi.2018.07.027 Dall, A., Hariharan, S., 2008. BK Virus Nephritis after Renal Transplantation. Clin. J. Am. Soc. Nephrol. CJASN 3, S68–S75.

14

A

CC E

PT

ED

M

A

N

U

SC R

IP T

Dogan, S.E., Celebi, Z.K., Akturk, S., Kutlay, S., Tuzuner, A., Keven, K., Sengul, S., 2017. Prevalence and Risk Factors of BK Viremia in Patients With Kidney Transplantation: A Single-Center Experience From Turkey. Transplant. Proc. 49, 532–536. Elfadawy, N., Flechner, S.M., Liu, X., Schold, J., Srinivas, T.R., Poggio, E., Fatica, R., Avery, R., Mossad, S.B., 2013. Cmv Viremia Is Associated With a Decreased Incidence of Bkv Reactivation after Kidney and Kidney-pancreas Transplantation. Transplant. J. 96, 1097–1103. https://doi.org/10.1097/TP.0b013e3182a6890d Hasegawa, M., Ito, T., Saigo, K., Akutsu, N., Maruyama, M., Otsuki, K., Aoyama, H., Matsumoto, I., Asano, T., Kitamura, H., Kenmochi, T., 2014. Association of DNA Amplification With Progress of BK Polyomavirus Infection and Nephropathy in Renal Transplant Recipients. Transplant. Proc. 46, 556–559. Hässig, A., Roos, M., Etter, A., Bossart, W., Müller, N., Schiesser, M., Wüthrich, R.P., Fehr, T., n.d. Association of BK viremia with human leukocyte antigen mismatches and acute rejection, but not with type of calcineurin inhibitor. Transpl. Infect. Dis. 16, 44–54. https://doi.org/10.1111/tid.12153 Hirsch, H. H., Randhawa, P., AST Infectious Diseases Community of Practice, 2013. BK polyomavirus in solid organ transplantation. Am. J. Transplant. Off. J. Am. Soc. Transplant. Am. Soc. Transpl. Surg. 13 Suppl 4, 179–188. https://doi.org/10.1111/ajt.12110 Hirsch, H H, Vincenti, F., Friman, S., Tuncer, M., Citterio, F., Wiecek, A., Scheuermann, E.H., Klinger, M., Russ, G., Pescovitz, M.D., Prestele, H., 2013. Polyomavirus BK Replication in De Novo Kidney Transplant Patients Receiving Tacrolimus or Cyclosporine: A Prospective, Randomized, Multicenter Study. Am. J. Transplant. 13, 136–145. https://doi.org/10.1111/j.16006143.2012.04320.x Hsiao, C.-Y., Pilmore, H.L., Zhou, L., de Zoysa, J.R., 2016. Outcomes of renal transplant recipients with BK virus infection and BK virus surveillance in the Auckland region from 2006 to 2012. World J. Nephrol. 5, 497–506. https://doi.org/10.5527/wjn.v5.i6.497 Huang, G., Zhang, L., Liang, X., Qiu, J., Deng, R., Li, J., Chen, G., Dong, Y., Chen, L., 2014. Risk Factors for BK Virus Infection and BK Virus–Associated Nephropathy Under the Impact of Intensive Monitoring and Pre-emptive Immunosuppression Reduction. Transplant. Proc. 46, 3448– 3454. Jacobi, J., Prignitz, A., Büttner, M., Korn, K., Weidemann, A., Hilgers, K.F., Heller, K., Velden, J., Knöll, A., Wullich, B., May, C., Eckardt, K.-U., Amann, K.U., 2013. BK viremia and polyomavirus nephropathy in 352 kidney transplants; risk factors and potential role of mTOR inhibition. BMC Nephrol. 14, 207. https://doi.org/10.1186/1471-2369-14-207 Kayler, L., Zendejas, I., Schain, D., Magliocca, J., n.d. Ureteral stent placement and BK viremia in kidney transplant recipients. Transpl. Infect. Dis. 15, 202–207. https://doi.org/10.1111/tid.12051 Kean, J.M., Rao, S., Wang, M., Garcea, R.L., 2009. Seroepidemiology of Human Polyomaviruses. PLOS Pathog. 5, e1000363. Knight, R.J., Gaber, L.W., Patel, S.J., Devos, J.M., Moore, L.W., Gaber, A.O., 2013. Screening for Bk Viremia Reduces But Does Not Eliminate the Risk of Bk Nephropathy: A Single-center Retrospective Analysis. Transplant. J. 95, 949–954. https://doi.org/10.1097/TP.0b013e31828423cd Kuypers, D.R. j, Bammens, B., Claes, K., Evenepoel, P., Vanrenterghem, Y., 2010. Maintenance Immunosuppressive Agents as Risk Factors for Bk Virus Nephropathy: The Need for True Drug Exposure Measurements. Transplantation 89, 1296–1297. Maliakkal, J.G., Brennan, D.C., Goss, C., Horwedel, T.A., Chen, H., Fong, D.K., Agarwal, N., Zheng, J., Schechtman, K.B., Dharnidharka, V.R., 2017. Ureteral Stent Placement and Immediate Graft Function are associated with Increased Risk of BK Viremia in the First Year after Kidney Transplantation. Transpl. Int. Off. J. Eur. Soc. Organ Transplant. 30, 153–161. Manitpisitkul, W., Drachenberg, C., Ramos, E., Munivenkatappa, R., Philosophe, B., Klassen, D., Haririan, A., 2009. Maintenance Immunosuppressive Agents as Risk Factors for Bk Virus

15

A

CC E

PT

ED

M

A

N

U

SC R

IP T

Nephropathy: A Case-control Study. Transplantation 88, 83–88. https://doi.org/10.1097/TP.0b013e3181aa8d93 Masutani, K., Ninomiya, T., Randhawa, P., 2013. HLA-A2, HLA-B44 and HLA-DR15 are associated with lower risk of BK viremia. Nephrol. Dial. Transplant. 28, 3119–3126. https://doi.org/10.1093/ndt/gft298 Mindlova, M., Boucek, P., Saudek, F., Skibova, J., Jedinakova, T., Lipar, K., Adamec, M., Hirsch, H.H., n.d. Prevalence and risk factors of polyomavirus BK replication in simultaneous pancreas/kidney transplant recipients from a single transplant center. Clin. Transplant. 26, 267–274. https://doi.org/10.1111/j.1399-0012.2011.01488.x Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., Group, T.P., 2009. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLOS Med. 6, 1–6. https://doi.org/10.1371/journal.pmed.1000097 Moura, E.B., Petzhold, S.V., Amaral, A.R., Deboni, L.M., França, P.H.C.D., Moura, E.B., Petzhold, S.V., Amaral, A.R., Deboni, L.M., França, P.H.C.D., 2017. Evaluation of the predisposition and clinical impact of BK virus replication in kidney transplant patients. An. Acad. Bras. Ciênc. 89, 675–684. Pham, Phuong-thu, Schaenman, J., Pham, Phuong-chi, 2014. Bk virus infection following kidney transplantation: an overview of risk factors, screening strategies, and therapeutic interventions. Curr. Opin. Organ Transplant. 19, 401–412. Prince, O., Savic, S., Dickenmann, M., Steiger, J., Bubendorf, L., Mihatsch, M.J., 2009. Risk factors for polyoma virus nephropathy. Nephrol. Dial. Transplant. 24, 1024–1033. https://doi.org/10.1093/ndt/gfn671 Santoveña, A.Z., Meseguer, C.G., Mejía, S.M., Melgar, Á.A., Camblor, C.F., Hijosa, M.M., Carrión, A.P., Román, L.E., 2015. BK Virus Infection in Pediatric Renal Transplantation. Transplant. Proc. 47, 62–66. Sharif, A., Alachkar, N., Bagnasco, S., Geetha, D., Gupta, G., Womer, K., Arend, L., Racusen, L., Montgomery, R., Kraus, E., 2012. Incidence and Outcomes of BK Virus Allograft Nephropathy among ABO- and HLA-Incompatible Kidney Transplant Recipients. Clin. J. Am. Soc. Nephrol. CJASN 7, 1320–1327. https://doi.org/10.2215/CJN.00770112 Smith, J.M., Dharnidharka, V.R., Talley, L., Martz, K., McDonald, R.A., 2007. BK Virus Nephropathy in Pediatric Renal Transplant Recipients: An Analysis of the North American Pediatric Renal Trials and Collaborative Studies (NAPRTCS) Registry. Clin. J. Am. Soc. Nephrol. 2, 1037–1042. https://doi.org/10.2215/CJN.04051206 Theodoropoulos, N., Wang, E., Penugonda, S., Ladner, D.P., Stosor, V., Leventhal, J., Friedewald, J., Angarone, M.P., Ison, M.G., 2018. BK Virus Replication and Nephropathy After AlemtuzumabInduced Kidney Transplantation. Am. J. Transplant. 13, 197–206. https://doi.org/10.1111/j.1600-6143.2012.04314.x Thölking, G., Schmidt, C., Koch, R., Schuette-Nuetgen, K., Pabst, D., Wolters, H., Kabar, I., Hüsing, A., Pavenstädt, H., Reuter, S., Suwelack, B., 2016. Influence of tacrolimus metabolism rate on BKV infection after kidney transplantation. Sci. Rep. 6. https://doi.org/10.1038/srep32273 Vu, D., Sakharkar, P., Shah, T., Naraghi, R., Yasir, Q., Hutchinson, I., Min, D., 2014. Association of Interferon Gamma Gene Polymorphisms With Bk Virus Infection Among Hispanic Renal Allograft Recipients. Transplantation 97, 660–667. https://doi.org/10.1097/01.TP.0000438115.20198.89 Wingate, J.T., Brandenberger, J., Weiss, A., Scovel, L.G., Kuhr, C.S., n.d. Ureteral stent duration and the risk of BK polyomavirus viremia or bacteriuria after kidney transplantation. Transpl. Infect. Dis. 19, e12644. https://doi.org/10.1111/tid.12644 Wunderink, H.F., Meijden, E. van der, Brouwer, C.S. van der B., Mallat, M.J.K., Haasnoot, G.W., Zwet, E.W. van, Claas, E.C.J., Fijter, J.W. de, Kroes, A.C.M., Arnold, F., Touzé, A., Claas, F.H.J., Rotmans, J.I., Feltkamp, M.C.W., n.d. Pretransplantation Donor–Recipient Pair Seroreactivity Against BK Polyomavirus Predicts Viremia and Nephropathy After Kidney Transplantation. Am. J. Transplant. 17, 161–172.

16

A

CC E

PT

ED

M

A

N

U

SC R

IP T

Wunderink, H.F., Meijden, E. van der, Brouwer, C.S. van der B., Mallat, M.J.K., Haasnoot, G.W., Zwet, E.W. van, Claas, E.C.J., Fijter, J.W. de, Kroes, A.C.M., Arnold, F., Touzé, A., Claas, F.H.J., Rotmans, J.I., Feltkamp, M.C.W., n.d. Pretransplantation Donor–Recipient Pair Seroreactivity Against BK Polyomavirus Predicts Viremia and Nephropathy After Kidney Transplantation. Am. J. Transplant. 17, 161–172. https://doi.org/10.1111/ajt.13880

17

Figure legends

Figure 1: Study selection process for the qualitative analysis

A

CC E

PT

ED

M

A

N

U

SC R

IP T

Figure 1

18

Table 1: The effect of tacrolimus use (relative to cyclosporine A) in multivariate analyses for the onset of BKPyV viremia. OR: odds ratio; HR: hazard ratio.

38/229

32/38

71/251

49/71

41/553

32/41

111/407

84/111

SC R

NR

0.68 (0.441.07) 5.27 (1.223.16) 2.9 (1.18.1) 0.99 ( 0.91-1.25) 0.71 (0.242.05) 0.76 (0.481.23)

Distribution of relative effects

0.094 0.027

U

22/265

H R O R H R O R O R H R

p value

0.032 0.19

0 1 2 3 4 5 6

0.53 0.264

A

CC E

PT

ED

Moura et al. 2017 Wunderink et al. 2017

64/102

N

Vu et al. 2014

102/998

Relative effect (95%CI)

A

Masutani et al. 2013 Hasegawa et al. 2014 Huang et al. 2014

Viremic patients exposed to tacrolimu s

M

Reference

Prevalence of BKPyV viremia (viremic patients/tota l population)

IP T

Immunosuppressive regimen : tacrolimus (vs cyclosporine A)

19

Table 2: Relative effects of donor/recipient characteristics in multivariate analyses for the onset of BKPyV viremia. OR: odds ratio; HR: hazard ratio. Deceased donor

Knight et al. 2013

57/349

41/57

O R

Vu et al. 2014

71/251

56/71

O R

38/229

26/38

22/84

16/22

Moura et al. 2017

41/553

32/41

Dogan et al. 2017

29/183

10/29

M

O R

Distribution of relative effects

0.03

0.743 0.071

U

A

O R

N

H R R R

1.96 (1.053.81) 0.78 (0.173.49) 2.1 (0.94.6) 3.2 (1.129.3) 2.47 (0.738.31) 3.65 (1.429.39)

p value

0

1

2

3

4

NR

0.14 <0.00 1

ED

Huang et al. 2014 Santoveña et al. 2015

Relative effect (95%CI)

IP T

Viremic patients exposed to a cadaveric graft

SC R

Reference

Prevalence of BKPyV viremia (viremic patients/tot al population)

Male Recipient

Male viremic patients

102/998

78/102

H R

Hirsch et al. 2013

42/487

35/42

O R

Vu et al. 2014

71/251

64/71

O R

Moura et al. 2017

41/553

30/41

O R

CC E

Reference

PT

Prevalence of BKPyV viremia (viremic patients/tot al population)

A

Masutani et al. 2013

Relative effect (95%CI)

2.38 (1.464.09) 2.49 (1.055.89) 1.42 (0.991.98) 1.77 (0.793.85)

p value

Distribution of relative effects

<0.00 1 0.038

0.033

0

1

2

3

4

0.24

20

Wunderink et al. 2017

111/407

73/111

H R

Dogan et al. 2017

20/174

15/20

O R

1.04 (0.68-1.6) 3.47 (1.1110.86)

0.842 0.03

93/600 102/998

Relative effect (95%CI)

53 (15) / 49 (16) 48 (18) / 49 (17)

O R H R

53.4 (13.2) / 50.4 (14.6)

O R

1.02 (11.04) 0.77 (0.26-2.4) 1.03 (0.991.05) 1 (0.991.05) 1.01 (0.9881.02)

71/145

Wunderink et al. 2017

111/407

54 (11.5) / 53 (11.7)

H R

Wingate et al. 2017

89/403

55.25 (13) / 54.32 (13.9)

M

O R

p value

N

Distribution of relative effects

0.016 3 0.65

U

Thölking et al. 2016

A

Kayler et al. 2013 Masutani et al. 2013

Mean age (SD) of patients (viremic/non viremic)

SC R

Reference

Prevalence of BKPyV viremia (viremic patients/tot al population)

IP T

Recipient Age

0.067 0

2

4

0.665 0.574

ED

History of previous transplant

Viremic patients with at least one previous transplantatio n

Jacobi et al. 2013

48/352

9/48

O R

Vu et al. 2014

71/251

20/71

O R

111/407

11/111

H R

A

CC E

Reference

PT

Prevalence of BKPyV viremia (viremic patients/tot al population)

Wunderink et al. 2017

Relative effect (95%CI)

2.74 (1.057.15) 1.53 (0.624.42) 1.36 (0.682.68)

p value

Distribution of relative effects

0.039

0.07

0.384 0 1 2 3 4 5 6 7 8

21

Table 3: Relative effect of ureteral stent use in multivariate analyses for the onset of BKPyV viremia. OR: odds ratio; HR: hazard ratio.

93/600 171/1318 89/403

IP T 0

2

4

A

CC E

PT

ED

M

A

N

Wingate et al. 2017

23/198

SC R

Brennan et al. 2005 Kayler et al. 2013 Maliakkal et al. 2016

Distribution of relative effects

U

Reference

Prevalence of BKPyV viremia (viremic patients/tota l population)

Surgery: ureteral stent Viremi c patient s Relative effect p expose (95%CI) value d to uretera l stent H 10/23 3 (NR) 0.018 R O 1.65 0.031 NR R (1.05-2.6) 5 H 1.36 76/171 0.024 R (1.05-2.6) 1.92 O 44/89 (1.040.044 R 3.74)

22

Table 4: Relative effects of post transplantation outcomes in multivariate analyses for the onset of BK Virus viremia. OR: odds ratio; HR: hazard ratio. Delayed graft function

Mindlova et al. 2012

36/135

14/35

OR

Vu et al. 2014

71/251

24/71

OR

Relative effect (95%CI)

Elfadawny et al. 2013

163/609

21/163

HR

Jacobi et al. 2013

48/352

11/48

OR

44/102

HR

71/251

50/71

OR

49/152

NR

OR

22/84

4/22

RR

Vu et al. 2014

A

Hässiq et al. 2014 Santoveña et al. 2015

Relative effect (95%CI)

M

ED

PT 102/998

CC E

Masutani et al. 2013

A

Reference

Prevalence of BKPyV viremia (viremic patients/total population)

Viremic patients exposed to risk factor

0.129

0 1 2 3 4 5 6 7 8 9

N

Acute rejection

0.0085

Distribution of relative effects

U

3.38 (1.388.24) 0.737 (0.491.09)

p value

IP T

Viremic patients exposed to risk factor

SC R

Reference

Prevalence of BKPyV viremia (viremic patients/total population)

2.01 (1.0963.685) 3.79 (1.59.58) 0.96 (0.581.56) 0.92 (0.851.02) 2.3 (1.094.8) 1.74 (1 included)

p value

Distribution of relative effects

0.02 0.005 0.89

0 1 2 3 4 5 6 7 8 9 10

0.02 NR NR

23

Table 5: Summary of predictive factors for biopsy-proven BKPyVN identified in

Risk factor identified

Mofetil mycofenolate

39/880

HR

39/880

HR

9/227

Relative effect (95%CI)

Tacrolimus (TAC)

TAC level three months before the biopsy

33/99

M

Prednisone dose three months before the biopsy

OR

OR

33/99

OR

12/407

HR

25/542

OR

39/880

HR

ABO incompatibility

11/62

OR

African-American recipient

31/666

HR

31/666

HR

12/407

HR

ED

PT CC E A

33/99

OR

33/99

Mean prednisone dose over the 3 months before biopsy Intravenous methylprednisolone 1000 mg once daily for 3 days

Donor/recipient characteristics

N

Mean TAC level over the 3 months before biopsy

OR

U

33/99

A

Immunosuppressive regimens

TAC level one month before the biopsy

HR

Polyclonal antibody induction

Recipient age

3.5 (1.67.5) 3.3 (1.57.6) 1.64 (0.446.11) 1.28 (1.011.62) 1.4 (1.11.8) 1.3 (1.021.7) 1.2 (1.041.36) 1.22 (1.041.4) 23.52 (4.57120.99) 11.04 (2.9441.52) 6.6 (2.318.9) 2.32 (NR) 2.89 (1.366.14) 1.04 (11.07) 0.98 (0.921.03)

p value

Reference

0.0013

Prince et al. 2009

IP T

Category

BKPyVN cases / Total population

SC R

multivariate analyses. OR: odds ratio; HR: hazard ratio.

0.0038

Prince et al. 2009

NR

Hsiao et al. 2016

0.04

Manitpisitkul et al. 2009

0.009

Manitpisitkul et al. 2009

0.03

Manitpisitkul et al. 2009

0.009

Manitpisitkul et al. 2009

0.02

Manitpisitkul et al. 2009

<0.001

Wunderink et al. 2017

0.0004

Smith et al. 2007

0.0005

Prince et al. 2009

0.04

Sharif et al. 2012

0.006

Theodoropoulos et al. 2013

0.048

Theodoropoulos et al. 2013

0.424

Wunderink et al. 2017

24

0.009

Hsiao et al. 2016

0.007

Wunderink et al. 2017

0.005

Theodoropoulos et al. 2013

0.051

Hsiao et al. 2016

9/227

HR

Viral serology

Donor BKPyV seroactivity

12/407

HR

31/666

HR

9/227

HR

Acute rejection (rejection score)

39/880

HR

CMV infection

31/666

Acute rejection (≥1 episode)

OR

U

8/163

HR

8/163

OR

0.0021

Prince et al. 2009

0.018

Theodoropoulos et al. 2013

<0.001

Elfadawny et al. 2013

0.0001

Elfadawny et al. 2013

A

CC E

PT

ED

M

A

Time course of BKPyV viremia

First positive BKPyV viremia > 185,000 copies/mL BKPyV peak viral load of 223,000 copies/mL at any time

N

Posttransplantation outcomes

IP T

Cold ischemia time

SC R

1.18 (1.041.35) 2.89 (1.336.29) 3.18 (1.427.12) 4.05 (0.9916.53) 5.1 (1.814.6) 2.72 (1.196.24) 113.25 (17.22744.6) 70.5 (8.08615)

Surgery

25