Risk factors for BK polyomavirus nephritis in renal allograft recipients
Copyright ª Blackwell Munksgaard 2004
Clin Transplant 2004 DOI: 10.1111/j.1399-0012.2004.00191.x
Risk factors for BK polyomavirus nephritis
in renal allograft recipients1
Rocha PN, Plumb TJ, Miller SE, Howell DN, Smith SR. Risk factors for Paulo N Rochaa, Troy J Plumba,
BK polyomavirus nephritis in renal allograft recipients. Sara E Millerb, David N Howellb
Clin Transplant 2004 DOI: 10.1111/j.1399-0012.2004.00191.x and Stephen R Smitha
ª Blackwell Munksgaard, 2004
Departments of a Medicine and b Pathology,
Duke University Medical Center, Durham, NC,
Abstract: Recurrent episodes of acute rejection (AR) and/or the intense USA
immunosuppression used for their treatment have been proposed as risk
factors for BK nephritis (BKN; BK refers to the initials of the ﬁrst patient
from whom this polyomavirus was isolated). To further examine the rela-
tionship between AR and BKN, we analyzed all kidney transplants per-
formed at our center between January 1999 and August 2001 (n ¼ 286).
After a mean follow-up of 737 ± 22 d, we identiﬁed nine cases of BKN
(3.1%). The mean time to diagnosis of BKN was 326 ± 56 d. No patient
with BKN had a prior history of AR. During the same period, 62 patients
were diagnosed with AR (22%). The mean time to diagnosis of AR was
197 ± 40 d (p ¼ 0.01 vs. time to diagnosis of BKN). Despite aggressive
Key words: immunosuppression – kidney
therapy with methylprednisolone and, in some cases, anti-lymphocyte
transplantation – kidney/pancreas
antibody, none of these patients with AR developed BKN. We compared
transplantation – polyomavirus – rejection
the baseline characteristics of patients in both groups and found that BKN
patients were more likely to be white people (78 vs. 44%, p ¼ 0.05) and Corresponding author: Stephen R. Smith MD,
male (89 vs. 53%, p ¼ 0.04). Moreover, the mean tacrolimus (TAC) levels MHS, Duke University Medical Center,
before diagnosis were higher in BKN than in AR patients (11.7 ± 0.5 vs. PO Box 3014, Durham, NC 27710, USA.
6.5 ± 0.6 ng/mL, p < 0.001). In summary, our study shows that BKN Tel.: 919 660 6858; fax: 919 684 4476;
often occurs in the absence of prior episodes of AR. In addition, our e-mail: email@example.com
ﬁndings suggest that white males exposed to higher TAC levels are at
greater risk of developing BKN. Accepted for publication 19 December 2003
Over the last decade, renal transplantation has the more powerful modern immunosuppressants
become the treatment of choice for most patients are a predisposing factor (5).
with end-stage renal disease. The advent of potent The work of Atencio et al. provided an alter-
new immunosuppressants such as tacrolimus native to the traditional view that viral reactivation
(TAC) and mycophenolate mofetil (MMF) has results from immunosuppression (6). Because kid-
contributed to a reduction in the incidence of acute neys of newborn – but not adult – mice are highly
rejection (AR) episodes. However, many centers permissive for polyomavirus infection (7), the
have noted a concomitant rise in the incidence of authors proposed that ongoing cellular diﬀerenti-
opportunistic infections caused by BK virus (1). ation is required for viral reactivation. They
Although BK virus was initially described in the showed that chemical or ischemic renal injury
early 1970s (2), the ﬁrst case of BK nephritis promoted high levels of polyomavirus replication
(BKN) was only recognized in 1995 (3). Since then, in adult mouse kidneys in the absence of immuno-
many cases have been reported, and BKN is suppression (6).
currently a growing problem in kidney transplan- As several reported cases of BKN have been
tation (4). The precise reasons behind this phe- preceded by recurrent episodes of AR (5, 8–10),
nomenon are unknown, but it is well accepted that some authors (11) have united both theories and
proposed the following scheme: (a) the cycle of renal
This work was presented in a plenary session at the American injury and repair caused by an episode of AR would
Transplant Congress, Washington DC, on June 2, 2003. lead to ongoing cellular diﬀerentiation; (b) this
Rocha et al.
would promote replication of latent BK virus in trough TAC level for each patient and averaged the
kidney tissue; (c) the intense immunosuppression mean levels for all patients in each group. In the
used for the treatment of AR would further facilitate BKN group, 205 levels were measured prior to
viral replication, and BKN would ultimately devel- diagnosis (mean ¼ 26 data points/patient) and 87
op; (d) the reduction in immunosuppression em- post-diagnosis (mean ¼ 11 data points/patient). In
ployed for the treatment of BKN would, in turn, the AR group, 688 (mean ¼ 13 data points/patient)
predispose to further episodes of AR. and 1402 (mean ¼ 27 data points/patient) levels
The clinical experience with BKN, however, were measured pre- and post-diagnosis, respectively.
shows that the correlation between AR and BKN
is very inconsistent. Although in some reports AR
preceded all cases of BKN (5, 9, 10), this time
course was not observed in many others (12–14). In Renal allograft biopsy specimens were routinely
fact, in the largest series to date, AR preceded processed and stained by the hematoxylin and
BKN in only 12% of cases (15). To further eosin, periodic acid-Schiﬀ, periodic acid-methen-
examine the relationship between AR (and its amine silver, and Masson trichrome methods.
treatment) and BKN, we conducted a retrospective Acute cellular rejection was diagnosed and graded
analysis of all kidney transplants performed at our according to the Banﬀ 97 working classiﬁcation of
institution between January 1999 and August 2001. renal allograft pathology (16). Acute humoral
rejection was diagnosed using a combination of
histology, C4d staining, and ﬂow cytometry cross-
Materials and methods matching as previously described (17). Immunop-
eroxidase staining for BK polyomavirus was
performed in all histologic specimens using a
We retrospectively reviewed all kidney and kidney– monoclonal primary antibody directed against the
pancreas transplants performed at Duke University BK virus large T antigen (Chemicon, Temecula,
Medical Center from January 1999 until August CA, USA). The avidin–biotinylated peroxidase
2001. Pertinent patient information was obtained staining technique was used, with diaminobenzi-
from electronic medical records. Diagnostic proce- dine as chromogen.
dures such as urine electron microscopy (EM) for
viruses or allograft biopsies were performed at the
EM of urine
discretion of the treating nephrologist to evaluate
clinically signiﬁcant graft dysfunction and not as Electron microscopy was performed as previously
part of a surveillance protocol. Recipients were described (13). Brieﬂy, cells and debris were ﬁrst
classiﬁed into three groups: BKN, AR, or controls. removed from urine by low speed centrifugation,
The diagnosis of BKN rested on a combination of and 1.5–3.0 mL of supernatant was ultracentri-
urine EM and histology in all but one patient, for fuged at 100 000 · g for 50 min (Beckman
whom a positive urine EM in the setting of a TL-Optima; Beckman Instruments, Palo Alto,
compatible clinical presentation was considered CA, USA). The pellet was re-suspended in
suﬃcient by the treating nephrologist. Similarly, approximately 20 lL distilled water, placed on a
the diagnosis of AR was based on allograft biopsy Formvar- and carbon-coated grid, negatively
ﬁndings in all but one patient who was empirically stained with uranyl acetate, and examined by EM
treated with methylprednisolone within the ﬁrst at 60 000–80 000· (Philips EM 300, EM 400, or
month of transplant for presumed cellular rejection. CM 12; FEI-Philips, Hillsboro, OR, USA). Occa-
All other subjects who did not undergo renal biopsy sionally, the pellet had to be re-suspended in
or whose biopsy reports revealed conditions other 200 lL water and washed by ultracentrifugation
than BKN or AR were included in the control to remove contaminating soluble protein (Beck-
group. man Airfuge, Beckman Instruments) prior to
uranyl acetate staining.
Induction and maintenance immunosuppression
We analyzed all trough TAC levels obtained from
the day of transplant until the diagnosis of BKN or Patients who were sensitized or considered to be at
AR. We also computed all levels from diagnosis high immunologic risk received induction therapy as
until last follow-up. A dummy value of zero was per our previously published protocol (17). All other
assigned to undetectable levels (<4 ng/mL). We patients received no induction. Maintenance immu-
then calculated the mean pre- and post-diagnosis nosuppression consisted of a calcineurin inhibitor,
Risk factors for BK nephritis
MMF, and prednisone. TAC was the calcineurin All kidney and Kidney–Pancreas transplants
inhibitor used in 80% of our patients; the target January 1999 to August 2001
trough level was 10–15 ng/mL during the ﬁrst (n = 286)
month and 5–10 ng/mL afterwards. MMF was
started at 1 g twice daily and dose adjusted, if Follow-up
needed, for gastrointestinal or bone marrow toxicity. March 2003
Methylprednisolone was given as a 250 mg i.v. bolus mean 737 ± 22 d
in the operating room and then tapered to 30 mg oral
prednisone by postoperative day 4. The prednisone BKN AR Control
dose was tapered by 5 mg every 2 wk until a (n = 9) (n = 62) (n = 215)
maintenance dose of 5–10 mg/d was achieved. 3% 22% 75%
F/U post diagnosis F/U post diagnosis
Treatment of acute rejection 412 ± 57 d 373 ± 37 d
The treatment of acute cellular rejection consisted
of methylprednisolone pulses; anti-lymphocyte Fig. 1. Study ﬂow-chart.
antibody therapy was reserved for severe cases
(Banﬀ II or greater) or steroid resistance. One were identiﬁed and followed for approximately
patient with mild acute cellular rejection received 2 yr (mean ± SEM, 737 ± 22 d) (Fig. 1). Using a
no therapy. Acute humoral rejection was treated combination of urine EM and/or histology, we
with plasmapheresis and intravenous immunoglob- diagnosed nine cases of BKN for an incidence of
ulin as previously described (17). 3.1% of all transplants. None of these cases had a
prior history of AR. During the same period, 62
patients were diagnosed with AR for an incidence
Treatment of BKN
of 22%. Similarly, none of the AR patients
Maintenance immunosuppression was reduced in subsequently developed BKN during the follow-
all cases. MMF was either reduced or discontinued up period despite aggressive therapy of rejection
while TAC doses were typically tapered to achieve with methylprednisolone pulses (61/62) and anti-
a serum level around 5 ng/mL. TAC was switched lymphocyte antibody therapy (27/62). Moreover,
to cyclosporine in one case and discontinued in BKN and AR did not co-exist in any of the
another. biopsies studied. The remaining 215 patients com-
prised the control group. Most patients in the
control group experienced stable allograft function
throughout the study period and therefore did not
We used ANOVA and the two-tailed, unpaired t- undergo diagnostic studies. Among the biopsied
test to compare continuous variables and the control patients, the most common histologic
chi-square test for comparison of dichotomous ﬁndings were calcineurin toxicity and acute tubular
variables between groups. All calculations were necrosis.
performed using MINITAB for Windows version Compared with AR, the time of presentation of
10.2 (Minitab Ltd, State College, PA, USA). BKN was signiﬁcantly delayed (326 ± 56 vs.
Graphs were created with SigmaPlot 2000 for 197 ± 40 d post-transplant, p ¼ 0.01) (Fig. 2A).
Windows version 6.00 (SPSS, Chicago, IL, USA). Furthermore, BKN was associated with less severe
Figures and tables were created in MicrosoftÒ graft dysfunction at diagnosis, as evidenced by a
PowerPointÒ 2000. Results were expressed as considerably lower mean serum creatinine value
mean ± standard error of the mean (SEM) or (3.2 ± 0.4 vs. 6.1 ± 0.5 mg/dL, p ¼ 0.017)
median and range. Survival analysis was per- (Fig. 2B). We compared the characteristics of
formed with SAS software version 8.2 (SAS Insti- patients who developed BKN with those of
tute Inc., Cary, NC, USA) using the Kaplan–Meier patients who developed AR and found that BKN
method and comparisons between survival curves patients were more likely to be male (89 vs. 53%,
were made with the log-rank test. Statistical p ¼ 0.04) and white people (78 vs. 44%, p ¼ 0.05)
signiﬁcance was deﬁned as p £ 0.05. (Table 1). In addition, BKN patients were less
likely to have a positive PRA. In contrast, the
highest percentage of females, black people and
pre-sensitized patients was found in the AR group.
Between January 1999 and August 2001, 286 Cold ischemia time and incidence of delayed graft
kidney or kidney–pancreas transplant recipients function were similar between groups (data not
Rocha et al.
Serum Cr. (mg/dL)
1000 p = 0.01 p = 0.017 Fig. 2. Clinical presentation. (A) Time
800 12 to diagnosis of BKN and AR. (B) Ser-
600 10 um creatinine at the time of diagnosis
8 of BKN and AR. The 25th percentile
6 (boundary of the box closest to zero);
200 median (line within the box); 75th per-
0 2 centile (boundary of the box farthest
0 from zero); 90th and 10th percentiles
BKN AR BKN AR (whiskers above and below the box).
Table 1. Baseline characteristics 16
p = 0.001 p < 0.0001
BKN AR Control 14 BKN
Variables (n ¼ 9) (n ¼ 62) (n ¼ 215) p-value 12 p = 0.001 AR
Age 44 ± 5 44 ± 2 47 ± 1 0.28 10
Male 89% (8)a 53% (33) 61% (131) 0.11 8
Female 11% (1)a 47% (29) 39% (84) 6
Black people 22% (2) 56% (35) 37% (80) 0.01 4
Non-black people 78% (7)b 44% (27) 63% (135)
Transplant type 2
CAD 78% (7) 74% (46) 65% (140) 0.32
LD 22% (2) 26% (16) 35% (75)
Positive PRA (>10%) 11% (1) 34% (21) 20% (43) 0.05
Tacrolimus 89% (8) 84% (52) 79% (170) 0.576
Cyclosporin A 11% (1) 16% (10) 21% (45) 0.57 Fig. 3. Mean tacrolimus levels. PRE, all levels from transplant
Induction therapy 56% (5) 63% (39) 53% (113) 0.35 until diagnosis; POST, all levels from diagnosis until last
p ¼ 0.04 vs. AR.
p ¼ 0.05 vs. AR.
p-values on the last column represent a comparison between the three groups. immunostaining of a kidney biopsy specimen with
We used ANOVA to compare recipient age and the chi-square test for all other an antibody speciﬁc for BK virus. The same
variables. method was used to conﬁrm the absence of BKN
CAD, deceased donor; LD, living donor; PRA, panel reactive antibody.
in 17/25 patients with negative urine EM.
The treatment of BKN consisted of a decrease in
shown). All patients received prednisone and maintenance immunosuppression. Table 2B separ-
MMF; TAC use was no diﬀerent across groups ates BKN patients into two groups according to the
(Table 1). degree of reduction in immunosuppression. The
We then calculated the mean of all trough TAC ﬁrst group is composed of seven patients for whom
levels obtained since the day of transplant until the MMF was completely discontinued. Most of them
day of diagnosis of BKN or AR and found that (ﬁve of seven) also experienced a reduction in the
BKN patients had been exposed to signiﬁcantly dose of the calcineurin inhibitor, and one was
higher TAC levels (11.7 ± 0.5 vs. 6.5 ± 0.6 ng/mL, switched from TAC to cyclosporin A. In all of these
p < 0.001). As part of the treatment of these seven patients, urine EM became negative for virus
conditions, trough TAC levels were reduced in the approximately 5 months after diagnosis and none
BKN group and increased in the AR group. It is lost their grafts. The other group is composed of
important to highlight that these variations in two patients for whom MMF dose was decreased
levels occurred within the therapeutic range but not discontinued. Neither of these patients
(Fig. 3). cleared the viruria, and one ultimately lost the
During the study period, 55 patients had urine allograft to the viral infection (Table 2B). Two-year
EM analysis for BK virus as part of the investiga- graft survival in BKN patients was 89%, a rate no
tion of renal dysfunction; Table 2A depicts 25 of diﬀerent than that observed in controls (Fig. 4A).
these patients for whom a concomitant kidney However, serum creatinine values obtained at last
allograft biopsy was available and could be used as follow-up revealed signiﬁcantly higher levels of
the gold standard. All eight of 25 patients with a graft dysfunction in the BKN group compared with
positive urine EM were conﬁrmed to have BKN by AR or control patients (Fig. 4B).
Risk factors for BK nephritis
Table 2. Performance of urine EM in the
diagnosis and follow-up of patients with BKN (A) Urine EM BKN AR Controls Total
Positive 8 0 0 8
Negative 0 11 6 17
Total 8 11 6 25
Conversion to Time to negative Graft
(B) Groups Additional treatment negative urine EM (%) urine EM (days) loss (%)
D/C MMF (n ¼ 7) Decrease Cl (5) 100 147 (29-257) 0
TAC ﬁ CyA (1)
ﬂ MMF (n ¼ 2) D/C TAC (1) 0 – 50
Time to negative urine EM is presented as median (range).
BKN, BK nephritis; EM, electron microscopy; MMF, mycophenolate mofetil; NEG, negative; D/C, discontinue;
ﬂ, decrease; Cl, calcineurin inhibitor; CyA, cyclosporin A.
p = 0.05 vs. AR
Serum cr. (mg/dL)
p < 0.001 vs. control
BKN AR Control
Months of followup
Fig. 4. Outcomes. (A) Kaplan–Meier survival curves. Renal allograft survival data was calculated from the day of transplant. solid
line, control group; dashed line, BKN group; dotted line, AR group. Values on top of horizontal axis represent the number of
patients at risk. Death with graft function was included as allograft loss. (B) Serum creatinine of patients with graft function at last
follow-up. The 25th percentile (boundary of the box closest to zero); median (line within the box); 75th percentile (boundary of the
box farthest from zero); 90th and 10th percentiles (whiskers above and below the box).
replication of polyomaviruses (6). AR is one cause
of renal injury that is peculiar to allografts and has
The main ﬁndings of this retrospective analysis been shown by some to predate all cases of BKN
were that male gender, white people and exposure (5). The heavy immunosuppression that follows an
to higher TAC levels were associated with BKN. In episode of AR may provide an even more condu-
addition, AR did not precede BKN in any of the cive environment for viral replication. Herein we
cases studied; rather the two complications tended describe nine cases of BKN that were not preceded
to occur in distinct patient groups. Compared with by AR. Our ﬁndings are in agreement with those of
BKN, AR tended to present earlier, with more Ramos et al. who noted a previous episode of AR
severe renal impairment, and in a more sensitized in only eight of 68 (12%) cases of BKN (15). We
group of patients. also did not ﬁnd an association between BKN and
The apparent predilection of BK virus for the cold ischemia time or delayed graft function. Our
kidney allograft over native kidneys of recipients of data suggests that neither overt kidney damage nor
non-renal solid organ transplants has been the treatment of AR is required for the development of
focus of much speculation (18). Transplantation of BKN.
kidneys from BK virus seropositive donors into Why are we not seeing more cases of BKN
seronegative recipients cannot fully explain this involving the native kidneys of non-renal solid
discrepancy as it has been reported that most organ transplant recipients? One possibility is that
recipients who ultimately developed BKN were the Ôalloimmune environmentÕ, perhaps in the form
seropositive before surgery (10). Atencio et al. have of low-level, subclinical rejection, is indeed neces-
shown that the ongoing cellular diﬀerentiation that sary to potentiate viral reactivation. Another
follows kidney injury (and repair) promotes the theory, however, is that these cases are simply
Rocha et al.
underdiagnosed. In this regard, Haririan et al. Mancinelli et al. (21), we postulate that the
recently described a case of BKN involving the increased incidence of BKN among white people
native kidneys of a solitary pancreas transplant maybe related to higher drug (TAC) exposure in
recipient (19). The same group has detected BK this group. To our knowledge, this is the ﬁrst
virus in the urine of four of 38 pancreas alone report to correlate trough TAC levels and BKN.
recipients (20). In their study, higher TAC levels Although we found signiﬁcant diﬀerences
constituted a risk factor for BK viruria. We between BKN and AR patients regarding gender,
speculate that similar cases could be going unde- time to diagnosis, TAC levels, and serum creatinine
tected by transplant physicians and written oﬀ as at presentation, it is critical to recognize that there
Ôcalcineurin nephrotoxicityÕ when renal dysfunction was a signiﬁcant overlap with respect to all of these
develops later. parameters. Moreover, given the low incidence
In the present study, BKN and AR tended to of BKN, a white male presenting with allograft
aﬀect distinct groups of renal allograft recipients. dysfunction is still more likely to have AR than
While we noted a predominance of white males in BKN. Thus, a renal biopsy is often needed to
the BKN group, the highest percentage of black reliably distinguish AR from BKN, especially
people and females were encountered in the AR because the appropriate treatment of these entities
group. In addition, BKN and AR patients were at requires opposite strategies. Several groups have
opposite ends of the spectrum regarding pre- demonstrated interest in identifying a less invasive
sensitization, as deﬁned by a historic panel reactive means to distinguish between these two conditions
antibody (PRA) >10%. The signiﬁcance of these (9, 10, 14, 22, 23). Our data suggests that the urine
ﬁndings remains unclear at present. Female gender, EM might be a dependable method to diagnose (or
black race and pre-sensitization are known risk exclude) BKN. Because of the retrospective nature
factors for AR, while white people and male of our study, diagnostic procedures such as urine
gender have not been consistently associated with EM and allograft biopsies were only performed
BKN. In a recent study, however, investigators when deemed necessary by the treating clinician.
from the University of Maryland found that the Therefore, we do not have concomitant urine EM
percentage of males was signiﬁcantly higher in the and allograft biopsy data for all 286 patients.
BKN group than in controls (79 vs. 65%, Nevertheless, when a kidney biopsy was available
p ¼ 0.02) (15). Whether this reﬂects a true pre- to serve as the gold standard, we showed that the
dilection of BKN for the male gender or simply a urine EM correctly identiﬁed BKN in eight of eight
high incidence of females with AR in their control patients. Similarly, the urine EM correctly exclu-
group is unclear. ded BKN from 17 patients who were found to have
A recent report showed no association between other conditions on allograft biopsy. It is import-
the mean doses of immunosuppressants (predni- ant to underscore that we did not use urine EM as
sone, MMF, TAC, cyclosporin A) and develop- a screening tool but rather to investigate clinically
ment of BKN (15). However, a rigorous signiﬁcant graft dysfunction. When used in this
assessment of medication doses is diﬃcult to setting, the urine EM appears to be highly sensitive
accomplish because doses are frequently adjusted, and speciﬁc for BKN. However, a large, prospect-
and accurate documentation is not always avail- ive study is needed to determine whether the urine
able. In addition, the dose of a given immunosup- EM can be used as a non-invasive means to
pressant may not correlate with true drug distinguish BKN from AR.
exposure. For example, Mancinelli et al. have We also found the urine EM to be a useful tool
demonstrated that serum concentration of TAC for following BKN patients after a reduction in
after oral administration varies signiﬁcantly with immunosuppression. In seven of nine cases,
race, with the highest levels seen in white people immunosuppression was gradually tapered until
and the lowest in black people; the authors clearance of viruria was achieved. All seven
hypothesize that these results may be explained patients remained with graft function until last
by diﬀerences in intestinal CYP3A or P-glycopro- follow-up. Persistence of viruria was associated
tein activities (21). We therefore decided to investi- with graft loss in one of the remaining two
gate the relationship between trough TAC levels patients. Two-year graft survival in the BKN
and BKN as they more closely reﬂect true drug group was 89% and not diﬀerent from controls.
exposure and are readily accessible from electronic However, BKN patients with surviving allografts
medical records. Our data shows that the mean exhibited signiﬁcant renal dysfunction at last
TAC level was signiﬁcantly higher in patients who follow-up as demonstrated by serum creatinine
developed BKN compared with those who devel- values (Fig. 4B). It is probable that the higher
oped AR. In light of the aforementioned data by serum creatinine values in the BKN group repre-
Risk factors for BK nephritis
sents a surrogate marker for a poor outcome and 9. Nickeleit V, Klimkait T, Binet IF et al. Testing for
that a longer follow-up will reveal a much lower polyomavirus type BK DNA in plasma to identify renal-
allograft recipients with viral nephropathy [see comments].
graft survival in these patients. N Engl J Med 2000: 342: 1309.
In summary, our ﬁndings show that BKN often 10. Hirsch HH, Knowles W, Dickenmann M et al.
occurs in the absence of a prior history of AR. In Prospective study of polyomavirus type BK replication
the present study, BKN and AR tended to aﬀect and nephropathy in renal-transplant recipients [see com-
distinct populations. The mean TAC levels were ments]. N Engl J Med 2002: 347: 488.
11. Randhawa PS, Demetris AJ. Nephropathy due to
highest in the group of patients who developed polyomavirus type BK [comment]. N Engl J Med 2000:
BKN, suggesting that the intensity of immunosup- 342: 1361.
pression is critical for viral reactivation. Transplant 12. Randhawa PS, Finkelstein S, Scantlebury V et al.
nephrologists should be aware of this potential Human polyomavirus-associated interstitial nephritis in
relationship between high TAC levels and BKN, the allograft kidney. Transplantation 1999: 67: 103.
13. Howell DN, Smith SR, Butterly DW et al. Diagnosis
especially in white males. A substantial reduction and management of BK polyomavirus interstitial nephritis
in immunosuppression was associated with 89% in renal transplant recipients. Transplantation 1999: 68:
graft survival at 2 yr. However, as other reports 1279.
have shown that BKN may ultimately lead to graft 14. Limaye AP, Jerome KR, Kuhr CS et al. Quantitation
loss in up to 45% of patients despite reduction in of BK virus load in serum for the diagnosis of BK virus-
associated nephropathy in renal transplant recipients [see
immunosuppression (15), there is a need for further comments]. J Infect Dis 2001: 183: 1669.
studies designed to evaluate the eﬃcacy of new 15. Ramos E, Drachenberg CB, Papadimitriou JC et al.
therapies (24, 25) against BKN. Clinical course of polyoma virus nephropathy in 67 renal
transplant patients. J Am Soc Nephrol 2002: 13: 2145.
16. Racusen LC, Solez K, Colvin RB et al. The Banﬀ 97
Acknowledgements working classiﬁcation of renal allograft pathology. Kidney
Int 1999: 55: 713.
The authors would like to thank Dr Thu Le for her 17. Rocha PN, Butterly DB, Greenberg A et al. Beneﬁcial
comments and critical review of this manuscript. There was eﬀect of plasmapheresis and intravenous immunoglobulin
no external source of funding. on renal allograft survival of patients with acute humoral
rejection. Transplantation 2003: 75: 1490.
18. Haririan A, Klassen DK. BK virus infection after
References nonrenal transplantation. Graft 2002: 5: S58.
1. Reploeg MD, Storch GA, Clifford DB. BK virus: a 19. Haririan A, Ramos ER, Drachenberg CB, Weir MR,
clinical review. Clin Infect Dis 2001: 33: 191. Klassen DK. Polyomavirus nephropathy in native
2. Gardner SD, Field AM, Coleman DV, Hulme B. New kidneys of a solitary pancreas transplant recipient.
human papovavirus (B.K.) isolated from urine after renal Transplantation 2002: 73: 1350.
transplantation. Lancet 1971: 1: 1253. 20. Haririan A, Hamze O, Drachenberg CB, Ramos E,
3. Purighalla R, Shapiro R, McCauley J, Randhawa P. Weir MR, Klassen DK. Polyomavirus reactivation in
BK virus infection in a kidney allograft diagnosed by native kidneys of pancreas alone allograft recipients.
needle biopsy. Am J Kidney Dis 1995: 26: 671. Transplantation 2003: 75: 1186.
4. Mylonakis E, Goes N, Rubin RH, Cosimi AB, Colvin 21. Mancinelli LM, Frassetto L, Floren LC et al. The
RB, Fishman JA. BK virus in solid organ transplant pharmacokinetics and metabolic disposition of tacrolimus:
recipients: an emerging syndrome. Transplantation 2001: a comparison across ethnic groups. Clin Pharmacol Ther
72: 1587. 2001: 69: 24.
5. Binet I, Nickeleit V, Hirsch HH et al. Polyomavirus 22. Ding R, Medeiros M, Dadhania D et al. Noninvasive
disease under new immunosuppressive drugs: a cause of diagnosis of BK virus nephritis by measurement of mes-
renal graft dysfunction and graft loss. Transplantation senger RNA for BK virus VP1 in urine. Transplantation
1999: 67: 918. 2002: 74: 987.
6. Atencio IA, Shadan FF, Zhou XJ, Vaziri ND, 23. Hirsch HH, Mohaupt M, Klimkait T. Prospective
Villarreal LP. Adult mouse kidneys become permissive monitoring of BK virus load after discontinuing sirolimus
to acute polyomavirus infection and reactivate persistent treatment in a renal transplant patient with BK virus
infections in response to cellular damage and regeneration. nephropathy. J Infect Dis 2001: 184: 1494.
J Virol 1993: 67: 1424. 24. Kadambi PV. Treatment of refractory BK virus-associated
7. Rochford R, Moreno JP, Peake ML, Villarreal LP. nephropathy with cidofovir. Am J Transplant 2003: 3: 186.
Enhancer dependence of polyomavirus persistence in 25. Vats A, Shapiro R, Singh RP et al. Quantitative viral
mouse kidneys. J Virol 1992: 66: 3287. load monitoring and cidofovir therapy for the manage-
8. Nickeleit V, Hirsch HH, Binet IF et al. Polyomavirus ment of BK virus-associated nephropathy in children and
infection of renal allograft recipients: from latent infection adults. Transplantation 2003: 75: 105.
to manifest disease. J Am Soc Nephrol 1999: 10: 1080.