Native kidney BK virus nephropathy, a systematic review

Abstract

Background:

There is a growing base of literature describing BK nephropathy (BKVN) in patients outside of the setting of kidney transplant. Previous systematic reviews of the literature have been limited by methodology or by the scope of patients included.

Study Design and Methods:

Systematic Review (Prospero # CRD42018088524).

Setting & Population:

Patients without kidney transplant who had biopsy-proven BKVN.

Selection Criteria for Studies:

Full-text articles that describe native BKVN patient cases.

Analytical Approach:

Descriptive synthesis.

Results:

The search identified 630 unique articles of which 51 were included in the final review. Sixty-five cases (including two new cases presented in this review) were identified, all but one occurred in the setting of known immunosuppression.

Limitations:

The primary limitation was the exclusion of studies that did not fulfill the stringent review criteria. We excluded reports with only a clinical diagnosis of BKVN, such as those with viruria and/or viremia without biopsy.

Conclusions:

As of May 2018, there are 65 reported cases of BKVN in native kidneys. This represents the most comprehensive description of biopsy-proven BKVN in native kidneys to date. Evaluation for BK nephropathy should be considered in immunocompromised patients who exhibit unexplained renal failure.

1 |. INTRODUCTION

BK virus, a member of Polyomaviridae family, was first discovered in 1971 in the urine of a kidney transplant recipient.¹ BK is a non-enveloped, circular, and double-stranded DNA virus.² There are four subtypes of BK virus, I-IV, with subtype I being the most prevalent and consisting for four subgroups (Ia, Ib-1, Ib-2, and Ic).^3,4 Primary BK virus is typically acquired in childhood with seroprevalence rates in adults reaching 80%.^5,6 After primary infection, the virus establishes latency in the renourinary epithelium and is usually asymptomatic in immunocompetent hosts.⁷ The clinical spectrum of BK virus infection in immunocompromised patients includes nephropathy, ureteral stricture, and hemorrhagic cystitis.⁸ Less commonly, systemic vasculitis, pneumonia, pneumonitis, retinitis, and hepatitis have been described.^9–13 BK nephropathy (BKVN) are most common after kidney transplantation. Routine screening after kidney transplant leads to the diagnosis of BK viremia in 10%-20% of patients and BKVN in 2%-12% of patients.^14–17 The best-studied intervention is the careful reduction of immunosuppression, and no there are no definitively effective antiviral prophylaxis or treatments.¹⁸

BKVN is rare in non-renal transplant recipients, but there is a growing literature of case reports and case series describing BKVN outside of the setting of kidney transplantation.¹⁹ The largest description to date of BKVN in non-kidney transplant recipients was published in 2016 and described 39 patients. However, previous publications on this topic have been limited by search methodology, inclusion of patients who have not undergone kidney biopsies, inclusion of patients with BK cystitis, and/or by limiting the scope of patient populations reviewed. In this study, we aimed to provide the most comprehensive description of this entity to date by presenting two cases and performing a systematic review of the literature of biopsy-proven BKVN in native kidney.

1.1 |. Case 1

A 65-year-old Caucasian man with ischemic cardiomyopathy, hypertension, and hyperlipidemia underwent an orthotopic heart transplant. His maintenance immunosuppression was tacrolimus, mycophenolate mofetil (MMF), and prednisone. His pre-transplant serum creatinine (SCr) was 1.5-1.7 mg/dL (estimated glomerular filtration rate 42-49 mL/min/1.73 m² by the Chronic Kidney Disease Epidemiology (CKD-EPI) Collaboration equation) without hematuria or proteinuria on urinalysis, and his kidneys measured 10.0 and 10.9 cm in maximal longitudinal dimension on kidney ultrasound. His SCr remained stable after transplant. He developed post-transplant diabetes mellitus which resolved within 1 year after transplant. He had stable cardiac allograft function with an ejection fraction of 65% and no episodes of rejection.

Three years post-transplant, he developed acute kidney injury (AKI) with a SCr rising to 3.8 mg/dL over 5 months. His physical exam at that time was only remarkable for trace lower extremity edema, without evidence of intravascular volume overload. Tacrolimus trough levels were 5-7 µg/L. His urinalysis was negative for blood and albumin, urine microscopy showed no red blood cells (RBCs) or white blood cells per high powered field (hpf), and a random urine protein:creatinine ratio (UPCR) was 0.2 g/g. An ultrasound showed both kidneys to be 10.5 cm without evidence of obstruction.

He then underwent a kidney biopsy. The light microscopy showed acute tubular injury and tubulointerstitial nephritis with moderate interstitial fibrosis. There were focal tubular epithelial cells with glassy nuclear inclusions. Immunohistochemical staining for polyomavirus using SV40 and Pab597 antibodies was positive in focal tubular epithelial cell nuclei. A serum BK polymerase chain reaction (PCR) was positive (296 075 copies/mL - linear range of assay 325 to 100 million copies/mL).

His MMF was discontinued with improvement in his serum BK viral load to 6500 copies/mL, and his SCr remained stable at 3.5 mg/dL for 2 years, at which time his SCr began to rise again. Repeat serum BK PCR testing was not performed. He progressed to endstage kidney disease (ESKD). Six weeks after beginning hemodialysis, he developed a malignant pleural effusion and was subsequently found to have metastatic pancreatic cancer. He transitioned to hospice care and expired.

1.2 |. Case 2

A 34-year-old African American man with Hodgkin’s lymphoma had been treated 7 years prior to his recent presentation with cyclophosphamide, hydroxydaunorubicin, vincristine, and prednisone. He was in remission when he presented to the hospital with diarrhea, weakness, and volume depletion. He was diagnosed with recurrence of his lymphoma as well as Staphylococcus aureus bacteremia that was treated with intravenous vancomycin. Laboratory testing was notable for AKI with a SCr of 5.8 mg/dL (increased from baseline 1.2 mg/dL) which did not improve with intravenous fluids. A urinalysis showed 500+ mg/dl albumin and small blood with 1-2 RBCs/hpf on microscopy. UPCR was 13.0 g/g.

He underwent a kidney biopsy which was notable for focal segmental glomerulosclerosis, nuclear inclusions in focal tubular epithelial cell nuclei, and immunohistochemistry for SV40 (clone Pab416) showed positive staining of focal tubular epithelial cell nuclei. These findings fulfill Banff 2018 criteria for BK nephropathy.²⁰ At the time of this biopsy, his serum BK was negative by qualitative PCR (lower limit of detection not reported).

He required intermittent hemodialysis for two weeks with subsequent recovery of his kidney function (SCr 1.2 mg/dL). Because his Reed-Sternberg cells expressed CD20, he was treated with rituximab. One month after discontinuation of hemodialysis, his SCr worsened again, peaking at 6.5 mg/dL, along with transaminitis (aspartate aminotransferase 134, alanine aminotransferase 61 IU/L). Urine and serum BK PCR were undetectable, urinalysis showed moderate blood with 500+ mg/dL albumin with 1-2 RBCs/hpf on microscopy and UPCR was 3.3 g/g. A second kidney biopsy showed tubular epithelial cells with widespread degenerative and regenerative changes, frequent epithelial mitoses and intranuclear inclusions. Immunostaining for SV40 was positive in focal tubular epithelial cell nuclei.

He was subsequently diagnosed with hemophagocytic lymphohistiocytosis due to his lymphoma and died of multi-organ system failure.

1.3 |. Systematic review methods

We searched MEDLINE and EMBASE from inception to May 2018 using predefined keywords and/or medical subject headings (MeSH) to discover biopsy-proven cases of native kidney BKVN (Prospero # CRD42018088524). The detailed search strategy can be found in the Supplement (Tables S1 and S2). Two authors independently screened the titles and abstracts of identified references, and those meeting the inclusion criteria were retrieved for dual independent review of full-text. Searches were performed for relevant primary studies that were restricted to English but not by date. References of all papers meeting inclusion criteria were manually reviewed for identification of additional cases.

Primary studies were included for full-text articles that describe native BKVN patient cases. Exclusion criteria were (a) patients who underwent kidney transplant; (b) non-BK virus infection (eg JC virus); (c) BK viremia/viruria without immunohistochemical biopsy-proven BKVN; (d) non-human or in vitro models; and (e) cases described only in posters and abstract form. Previously published systematic reviews were only used to identify the relevant primary literature and were not included in this systematic review.

2 |. RESULTS

The search identified 630 unique articles of which 51 were included in the final review. Sixty-five cases (including our two patients) met our search criteria (Figure 1). The relevant medical histories for these patients include solid organ (heart, lung, liver, pancreas) transplant (n = 21), hematopoietic cell transplant (HCT, n = 19), human immunodeficiency virus (HIV) infection (n = 10), hematologic malignancy (n = 10), autoimmune disease (n = 1), primary immunodeficiency (n = 1), prostate cancer (n = 1), and pulmonary tuberculosis and diabetes (n = 1). One patient had no identified immunodeficiency (Table 1; Figure 2).

FIGURE 1.

Article screening process for systematic review

TABLE 1.

Individual cases of native kidney BK virus nephropathy included in our case series and systematic review of the literature

Reference	Age/ gender	Relevant medical history	Time to diagnosisa	Therapy attempted	Therapy effect	Dialysis	Outcome
HCT
Aksenova43	10 F	Allo-HCT	13 mo	Add IVIG, leflunomide, cidofovir, and ciprofloxacin.	Leflunomide/Cipro/IVIG simultaneously had no effect, but addition of cidofovir resulted din resolution of viremia and improved Cr		CKD
Bruno44	Nr	Allo-HCT	NR	Not reported	Not reported	N/A	Death
Burbach45	45 F	Allo-HCT	46 mo	Reduced immunosuppression.	Viremia remained elevated.	Yes	Death
Lekakis46	51 M	Allo-HCT	5 mo	Add leflunomide, cidofovir, and ciprofloxacin.	Viruria improved, Viremia not reported	Yes	Death
Maximova47	15 F	Allo-HCT	3 mo	Stopped immunosuppression	Viremia improved. Persistent viremia.	No	Death
O’Donnell48	41 F	Matched-unrelated donor HCT	30 mo	Add leflunomide.	Viremia not changed.	Yes	ESRD
Papanicolaou49	58 M	Allo-HCT	16 mo	Add brincidofovir	Viremia improved. Persistent viremia. Cr stable.	No	Death
Shapiro50	14 M	Unrelated cord blood transplant	1 mo	Reduced immunosuppression. Add cidofovir.	Received too few doses. Died of multiorgan failure shortly after 2nd dose of cidofovir.	No	Death
Sharma51	17 M	CML, haploidentical BMT complicated by GVHD	2 y	Add leflunomide and IVIG.	Viremia improved. Persistent viremia. Cr worsened.	No	Death
Sharma51	16 M	AML, Allo-HCT	4 mo	Diagnosed on autopsy		N/A	Death
Sharma51	58 M	CLL, Allo-HCT complicated by GVHD	24 mo	Add cidofovir. Add sirolimus due to GVHD.	Viremia improved. Persistent viremia. Cr stable.	No	CKD
Stracke52	28 F	Haploidentical T-cell depleted peripheral blood SCT	2 y	Add cidofovir	Viremia not improved. Cr not improved	Yes	ESRD
Van der Bij53	57 M	Allo-HCT with leukemic invasion	21 mo	Add leflunomide.	Cr worsened to ESRD, started on HD.	Yes	ESRD
Verghese54	10 M	unrelated cord blood transplant	3.5 y	Add cidofovir and ciprofloxacin.	No effect to cidofovir, resolution of viremia to cipro. Cr improved	No	CKD
Verghese54	13 M	Allo-HCT	2 y	Add cidofovir.	Viremia unchanged. Cr worsening.	No	Death
Vigil42	30 M	Allo-HCT	Approx. 4 y	Add ciprofloxacin.	Viremia worsened.	No	Death
Limaye55	41 M	Autologous Peripheral blood cell transfusion	6.2 y	Add cidofovir.	Viremia resolved.	Yes	ESRD
O’Donnell48	36 F	Auto-HCT	9 mo	Add leflunomide.	Viremia not changed.	Yes	Death
Sanchez-pinto56	10 F	Auto-HCT	5 mo	Add cidofovir and ciprofloxacin.	Viremia improved. Persistent viremia. Cr worsened.	Yes	Death
Hematologic malignancy
Filler57	10 M	ALL	3 y	Add IVIG	Viremia improved. Persistent viremia	No	CKD
Inaba58	5 F	ALL	21 mo	Reduce chemotherapy dose. Add IVIG, cidofovir, leflunomide	Viremia improved. Persistent viremia. Cr stable.	No	Unclear
Collett59	56 M	CLL	NR	Add IVIG and ciprofloxacin.	Viremia did not respond.	No	CKD
Mccrory60	73 M	CLL	1 y	Stopped IVIG, started leflunomide	Viremia improved. Persistent viremia.	No	CKD
Sangala61	72 M	CLL	19 mo	Short course of methylprednisolone, added IVIG and ciprofloxacin.	Cr worsened to ESRD requiring regular HD	Yes	Unclear
Sharma51	66 M	CLL	10 y	Add IVIG and cidofovir	Cr worsened to ESRD	Yes	ESRD
Sharma51	73 M	CLL	1 y	Add IVIG and cidofovir	Cr remained elevated.	No	CKD
De Silva62	6 mM	Cartilage-hair hypoplasia and Hodgkin’s disease	26 mo	Not reported		No	Death
Sharma51	53 M	Non-Hodgkin Lymphoma	4 y	Add cidofovir	Viremia improved. Persistent viremia.	No	Death
Shah (this paper)	34 M	Non-Hodgkin lymphoma		No specific therapy		Yes	Death
HIV
Bratt63	26 M	HIV: CD4 20 × 106/L	N/A	Diagnosed on autopsy		No	Death
Crum-Cianflone64	38 M	HIV: CD4 < 50*106/L but at time of renal failure was 0*106/L	N/A	No specific therapy	Required hemodialysis. Died 1 y later du to CNS lymphoma.	Yes	ESRD
Cubukcu-Dimopulo65	14 M	HIV: CD4 = 0	N/A	Diagnosed on autopsy		No	Death
Jung66	49 M	HIV: CD4 4 *106/L | VL 96800 copies/mL	N/A	No specific therapy	Worsened viremia. Cr worsened. Died of nonrenal reasons.	No	CKD
Manabe67	32 M	HIV: CD4 3 *106/ L | VL 4400 copies/mL	N/A	Diagnosed on autopsy		No	CKD
Mouratoff68	49 M	HIV: CD4 3 × 106/L	N/A	No specific therapy	Renal function improved with antiretrovirals for HIV, but worsened within 2 mo of biopsy.	No	CKD
Nebuloni69	31 M	HIV: CD4 0.01 × 109/L	N/A	No specific therapy	2 mo later, no clinical worsening.	No	CKD
Smith70	36 M	HIV: CD4 < 50 × 106/L	N/A	No specific therapy	Left hospital abruptly and died.	Yes	ESRD
Sukov71	43 M	HIV: CD4 2.0 × 109/L	N/A	No specific therapy	Became HD-dependent. No comment on viremia.	Yes	ESRD
Vallbracht10	27 M	HIV: n/a (autopsy)	N/A	Diagnosed on autopsy		No	Death
Solid organ transplant
Ali72	12 M	Heart transplant	21 mo	Reduce immunosuppression, add leflunomide and cidofovir	Effect unclear	No	CKD
Barber73	26 M	Heart transplant	18 mo	Reduce immunosuppression, add cidofovir	Viremia resolved with cidofovir 1 mg/kg	Yes	ESRD
Butts74	9 F	Heart transplant	8 y	Reduce immunosuppression, add leflunomide	Viremia reduced. Stable viremia.	No	CKD
Joseph75	60 M	Heart transplant	3 y	Reduce immunosuppression, add ciprofloxacin and IVIG	Viremia N/A. Cr worsened.	Yes	ESRD
Joseph75	43 M	Heart transplant	2 y	Reduce immunosuppression, add ciprofloxacin	Stable GFR. Viremia reduced. Stable viremia.	No	CKD
Limaye55	59 M	Heart transplant	6.8 y	Diagnosed on autopsy		Refused	Death
Lorica76	14 M	Heart transplant	6 mo	Reduce immunosuppression, add IVIG, cidofovir, and ciprofloxacin	Viremia reduced. Stable viremia.	Yes	Death
Maddirala77	54 M	Heart transplant	3.5 y	Reduce immunosuppression	Viremia N/A. Cr not improving.	Yes	Death
Menahem78	59 F	Heart transplant	16 mo	Reduce immunosuppression, add cidofovir	Viruria remained positive. GFR had no improvement	Yes	ESRD
Pereira79	5 F	Heart transplant	2 y	Reduce immunosuppression, add IVIG and cidofovir	Viral loads decreased at cidofovir 1 mg/kg but remained positive. Died of BKV rhombo-encephalomyelitis	Yes	Death
Sahney80	8 M	Heart transplant	10 mo	Reduce immunosuppression, add IVIG and cidofovir	Viremia reduced. Stable viremia.	Yes	ESRD
Schmid81	57 M	Heart transplant	29 mo	Reduce immunosuppression, add cidofovir	Cr improved. Viremia improved. Persistent viremia.	Yes	ESRD
Shah (this paper)	70 M	Heart transplant		Reduce immunosuppression	Cr worsened. Viremia improved. Persistent viremia.	Yes	Death
Dufek82	9 M	Lung transplant	2 y	Reduce immunosuppression, add cidofovir	Viremia persisted	Yes	Death
Egli83	72 F	Lung transplant	5 y	Reduce immunosuppression, add leflunomide	Viremia resolved. Cr improved.	No	CKD
Kuppachi84	63 M	Lung transplant	CKD at 2 y, biopsy at 4 y	Reduce immunosuppression, add cidofovir and leflunomide	Cr stabilized and improved. Persistent viremia	No	CKD
Schwarz85	40 M	Lung transplant	15 mo	Add cidofovir and leflunomide	Resolution of viremia but relapsed after therapy discontinued	Yes	ESRD
Sharma51	30 M	Lung transplant	2 y	Reduced immunosuppression	Cr worsened. Viremia improved. Viremia persistent.	No	CKD
Vigil42	70 M	Lung transplant	2 y	Reduce immunosuppression, add IVIG and leflunomide	Viremia improved. Viremia persistent.	No	CKD
Zeng86	59 M	Liver transplant	7 y	Not reported	Not reported	None reported	Not reported
Haririan87	54 M	Pancreas transplant	9 mo	Not reported	Not reported	No	CKD
Other
Go88	79 M	Prostate cancer	N/A	Diagnosed on nephrectomy	Not reported	No	CKD
Krystel-Whittemore89	55 M	Rheumatoid Arthritis	6 y	Reduce immunosuppression, add IVIG, ciprofloxacin, leflunomide	Viremia improved. Persistent viremia.	Temporary HD	CKD
Park90	34 M	None	N/A	Not reported	Not reported	No	CKD
Rosen91	6 M	Hyper-IGM	5 y	Not reported	Not reported	Yes	Death
Sharma51	66 M	Pulmonary tuberculosis	9 mo	Not reported	Not reported	Yes	Death

ALL, acute lymphoblastic leukemia; CKD, chronic kidney disease; CLL, chronic lymphocytic leukemia; ESRD, end stage renal disease; F, female; HCT, hematopoietic stem cell transplantation; HCT, hematopoietic cell transplant; HD, hemodialysis; HIV, human immunodeficiency virus; IVIG, intravenous IgG; M, male.

^aTime to diagnosis is defined as the time from a known immunologic inult to diagnosis of BKVN. In practice, this refers to the time from transplant/diagnosis of illness to diagnosis of BKVN. HIV was not included in this because the time to development and diagnosis of BKVN is not a meaningful metric when the underlying HIV can be well-controlled on antiretroviral therapy, yielding long time frames or highly variable times.

FIGURE 2.

This diagram represents a breakdown of the primary medical issue faced by the native-kidney

2.1 |. Solid organ transplantation

Twelve cases of kidney biopsy-proven BKVN in heart transplant recipients were already reported in the literature and our case increased the total to 13. Ten of 13 cases occurred in males, with a mean age of 36.6 years and 38.5% of cases occurring in children. Of these cases, 9 (69%) required renal replacement therapy and mortality at time of publication was 38.5%.

Six cases of native kidney biopsy-proven BKVN were identified in patients with lung transplants (one pediatric case, five adults). The mean age at diagnosis was 47.3 years. Of these cases, two (33.3%) required renal replacement therapy and mortality at time of publication was 16.7%.

A single case of native kidney biopsy-proven BKVN was described in a liver transplant recipient whose immunosuppressive regimen consisted of tacrolimus (trough 10-12 µg/L), sirolimus (trough 8-12 µg/L), MMF, and prednisone. A single case of native kidney biopsy-proven BKVN has been noted in a pancreas transplant recipient whose immunosuppression regimen included tacrolimus (trough 12-15 ng/dL), MMF, and prednisone.

The average time from transplant to diagnosis of BKVN for the collective solid organ transplant population is 2.88 years.

2.2 |. Hematopoietic cell transplantation

A total of 19 cases of native kidney biopsy-proven BKVN were found in the literature for patients who had previously undergone HCT. The majority of cases (16/19) occurred in patients after allogeneic HCT. We excluded three cases (Iwamoto et al,²¹ Hoefele et al,²² Gagneux-Brunon et al²³) which have previously been included in other native BKVN literature reviews because they did not perform immunohistochemistry to confirm the diagnosis. The mean age at diagnosis was 30.6 years with 42.1% of cases occurring in the pediatric population. In cases with demographic information, 58% were males. Half of these patients (8/16) required renal replacement therapy and 63.2% died during follow-up. A number of cases in the original search described BK virus-induced hemorrhagic cystitis, but these studies were excluded from our systematic review as these patients did not have BKVN.

The average time from transplant to diagnosis of BKVN for the hematopoietic stem cell transplant population is 1.91 years.

2.3 |. Hematologic malignancy

Including our second case, 10 cases of native kidney biopsy-proven BKVN in the setting of hematologic malignancy have been reported. Nine of 10 patients were males. The mean age was 44.8 years with 30% of cases occurring in children. Of these cases, 3 (30%) required renal replacement therapy and three patients died. The underlying pathology varied and outcome data is minimal in this cohort of patients. The average time from diagnosis of hematologic malignancy to diagnosis of BKVN is 3.06 years.

2.4 |. Human immunodeficiency virus

A total of 10 cases of native kidney biopsy-proven BKVN were reported in patients with HIV infection. All reported cases occurred in males, and the mean age was 34.5 years, with 10% of cases occurring in a pediatric population. Of these cases, 3 (30%) required renal replacement therapy and mortality at time of publication was 30%. The median CD4 count at diagnosis was 4 × 10⁶/L (Interquartile Range (IQR): 1.5 × 10⁶-35 × 10⁶).

2.5 |. Other

Individual cases of native biopsy-proven BKVN were reported in the setting of rheumatoid arthritis, Hyper IgM immunodeficiency syndrome (with BK virus infection diffusely in both kidneys and the urothelium on autopsy), pulmonary tuberculosis and diabetes mellitus, and prostate cancer. A final case reported a 34-year-old male with no relevant medical history in whom long-term follow-up and workup for immunocompromised states were not reported.

3 |. DISCUSSION

BK virus infection is a well-described complication of kidney transplantation. BK viremia is associated with impaired graft function ²⁴ and prevalence is estimated at 10%-20% while BKVN occurs in approximately 2%-12% of kidney transplant recipients and is associated with allograft loss.^14,17 There is limited literature for non-renal transplant patients diagnosed with BKVN. Here, we have presented two cases from our institution and performed a systematic review of the literature describing published cases of native kidney BKVN. Our systematic review is comprised of 65 native BKVN cases in diverse populations including solid organ transplant recipients, HCT recipients and those with hematologic malignancy. Of these 65 cases, 14 (21.5%) resulted in chronic requirement for dialysis and 26 (40%) of patients died during follow-up. This study represents the most comprehensive description of BK VN in the native kidney to date.

Our systematic review suggests that BKVN should be included in the differential diagnosis for immunosuppressed or immunocompromised patients with native kidney dysfunction, and evaluation for BK nephropathy in these patients should be considered. Of note, BKV serum PCR screening is limited by lack of a universal standard assay, and the majority of BKV PCR assay primers are based on the genotype I sequence. However, if the patient’s clinical course or biopsy-findings are inconsistent with BKV serum PCR results, sample re-analysis at a lab that utilizes different BKV genotype sequences is recommended.^19,25 The utility of screening immunosuppressed/immunocompromised patients with renal insufficiency for BK viremia, and the predictive of BK viremia for finding BK nephropathy on kidney biopsy in this patient population, remain to be studied.

Twelve of 65 (18.5%) patients identified in our review were children. Prior studies in pediatric kidney transplant have demonstrated an increased risk with primary infection but serostatus was unavailable in the two cases we presented.^26,27

While effective anti-viral therapy is not currently available, the experience in kidney transplant recipients suggests that careful reduction of immunosuppression, when possible, can be effective in treating BKVN. In immunocompetent patients or those in whom immunosuppression reduction is not possible, alternative therapies can be considered. There is an extensive literature in the renal transplant population detailing use of mammalian target of rapamycin (mTOR) inhibitors, leflunomide, flouroquinolones, cidofovir, and intravenous immunoglobulin (IVIg) for the treatment of BKVN, but none have been shown to be particularly effective.^28,29 Further research is ongoing, with newer agents such as brincidofovir, virus-specific T lymphocytes, and IVIg being currently studied.^30–32

Noting the limitation of publication bias, a number of different treatment strategies were reported in the cases we review. Monotherapy with cidofovir resulted in undetectable viral loads in 3/12 patients reviewed, ciprofloxacin resulted in undetectable viral loads in 1/3 patients reviewed, and leflunomide in 1/6. Prospective trials are needed to determine efficacy of these agents used alone or in concert with others to better understand how to treat this entity.

Several authors have observed that BKVN is more common among heart transplant recipients than other solid organ transplant populations.^33–35 While the reason for this is unknown, this disparity persists in our systematic review. It has been hypothesized that this may be related to their cumulative immunosuppressive burden or due to the effect of cardiac allograft dysfunction on renal perfusion, in accordance with a two-hit hypothesis.^36–39 Conversely, only one patients with BKVN after liver transplant has been described, in a patient on fourdrug immunosuppression. Previous literature has described a lower prevalence of BK in liver transplant recipients compared to heart or kidney transplant recipients, which could reflect the lower cumulative immunosuppression exposure that liver transplant patients.⁴⁰

It should also be noted that there is a large base of literature on the incidence of viremia and viruria in patients with systemic lupus erythematosus and lupus nephritis, but our systematic review did not find any cases of biopsy-proven BKVN in this population.⁴¹ Animal studies have shown that infection with BK can virus can lead to the formation of anti-dsDNA antibodies, resulting in speculation that the BK virus may have a pathogenic role in the development of systemic lupus.⁴¹

While our study is purely descriptive, we submit that the methodology and criteria for our systematic review represents the most comprehensive description of BKVN outside of the kidney transplant setting to date. These results show that our search methodology was more sensitive than prior reviews. Compared to the largest prior review, published in 2016 and which found 39 patients,⁴² we have added 26 cases to the literature. Taking into account our 2 cases, only 3 cases in our review were published in 2016 or after. This indicates that the majority of the cases we have added were found due to our search methodology, the strength of which include our robust review of the primary and secondary literature as well as the rigorous inclusion and exclusion criteria applied to studies identified by our search.

Our series is also more precise and specific than previous reviews, as we only included patients with biopsy-proven BKVN. This stringent inclusion criterion ensured high-quality data and minimized crossover with other BK complications such as hemorrhagic cystitis or asymptomatic viruria/viremia without biopsy-proven BKVN. A number of non-biopsy proven cases have been included in prior reviews based on elevated SCr in the setting of BK hemorrhagic cystitis, high levels of viruria, and high levels of viremia.

The primary limitation of our study in assessing the true burden of BKVN outside of the kidney transplant setting is the exclusion of studies that did not fulfill the stringent review criteria. We excluded reports with only a clinical diagnosis of BKVN, such as those with viruria and/or viremia without biopsy. As a kidney biopsy are not always pursued in immunocompromised patients with renal insufficiency, the total number of native-kidney BKVN patients may be greater than what is encompassed in our systematic review. As such, some patients that were excluded from our search may have had BKVN. The lack of a PCR standard for detecting BK viremia may have lead to discordant results between biopsy findings and blood testing. Finally, the retrospective nature of our review precludes us from establishing the incidence and prevalence of the disease.

4 |. CONCLUSION

This case series and systematic review of the literature provides the most comprehensive description of BKVN outside of the kidney transplant setting to date. While the incidence of BKVN in these patient populations remains to be determined evaluation for BKVN should be considered in immunocompromised patients with unexplained renal insufficiency.