Objectives: Because of the controversy regarding the effects of BK virus on non-renal solid-organ transplant, we detected the BK virus via different methods and its effect on clinical findings, liver and kidney functions, and graft dysfunction in liver transplant recipients.
Materials and Methods: This prospective cohort study comprised patients over the age of 18, who consecutively received liver transplant from January 1 to December 31, 2011. The patients were examined once, every 2 weeks, for the first 3 months after transplant. Clinical findings were evaluated on each examination; blood and urine samples were collected, BK virus DNA was assessed with real-time polymerase chain reaction, and the presence of decoy cells (which are epithelial cells with large nuclei and large basophilic inclusions) in the urine was investigated. Patients were followed-up for 1 year to see if rejection occurred.
Results: Five of 39 patients (12.8%) showed BK viremia; 11 patients (28.2%) showed BK viruria, and 13 (33.3%) showed decoy cells. No statistically significant differences were found between BK virus positive and negative groups, respecting demographic variables, kidney and liver functions, and graft survival. BK virus DNA positivity in blood was the standard, while decoy cell assessment in urine and BK virus polymerase chain reaction test sensitivity in urine was 40%.
Conclusions: No matter the method used to detect BK virus in the urine, the negativity of the tests is more valuable than their positivity. Although no statistically significant difference was found between the groups, we concluded that BK virus is a factor that should be considered when unexplained deterioration in kidney and liver function tests is observed in liver transplant recipients. Prospective studies with larger numbers of patients are warranted.
Key words : Liver transplant, Polymerase chain reaction, Viremia, Viruria
While immunosuppressants reduce the rate of rejection, they increase the prevalence of opportunistic bacterial, fungal, and viral infections. The BK virus, referred to as Polyomavirus hominis 1, is one of these opportunistic viral agents. It was first isolated in the urine of a kidney transplant recipient by Gardner in 1971.1 The effects of BK virus are significant to persons with renal transplant and hematologic malignancies, but it has not been clearly established in liver transplant recipients. Like other opportunistic viral infections, BK virus infections also are observed when immunosuppressants are used intensively within the first 1 to 6 months after transplant (specifically during the 3 months after transplant).2 This study sought to evaluate BK virus assessed in blood and urine samples collected within the first 3 months after transplant by different methods, and its relation to clinical findings, liver and kidney functions, and graft dysfunction in liver transplant recipients.
Materials and Methods
Study design and patient selection
This is a prospective cohort study comprising volunteer patients, older than 18 years, who consecutively received a liver transplant from January 1 to December 31, 2011, at the Department of General Surgery at our center. The study received approval by the ethical committee for clinical research in the Ege University Faculty of Medicine, issued on December 12, 2010; approval number B.30.2.EGE.0.20.05.00/OY/54-40. All protocols conformed with the ethical guidelines of the 1975 Helsinki Declaration. Informed consent was obtained from all subjects.
Patients were examined once every 2 weeks within the first 3 months after transplant. Demographic variables (age, sex, and presence of comorbid illness, cause of liver failure, donor type, and presence of rejection) were recorded on their first visit. Additionally, date of visit, results of physical examination, type and dose of immunosuppressant agent, complaints (if any), results of a complete blood count (white blood cell count, and percentages of polymorphonuclear leukocytes, monocytes, and lymphocytes) and biochemical (creatinine, aspartate aminotransferase, alanine aminotransferase) were recorded on each visit. Blood and urine samples were collected. Decoy cells were investigated in the urine. Blood and urine samples were coded and retained at -20°C and -40°C until assessment of BK virus DNA was done with real-time polymerase chain reaction (PCR).
In the blood and urine samples, BK virus DNA via PCR, and decoy cells (which are epithelial cells with large nuclei and large basophilic inclusions) in urine, were investigated.
Method of decoy cell investigation
Urine samples were centrifuged for 5 minutes at a rate of 2000 rpm. If the volume of the resulting sediment was < 0.2 cc, it was spun in the cytometer for 5 minutes at 800 rpm, if it was > 0.2 cc, the smear was prepared directly from the sediment. Preparations were stained by hematoxylin and eosin, and presence of the decoy cells was investigated under a light microscope.
Real-time polymerase chain reaction method
Viral DNA extraction method
High Pure PCR Template Preparation Kit (Roche Diagnostics Corporation, Roche Applied Science, Indianapolis, IN, USA, Version 16.0) was used for viral DNA extraction. The blood and urine samples on which the extraction procedure was completed, were retained at -20°C and -40°C until the PCR was done.
Real-time polymerase chain reaction method
Real-time polymerase chain reaction was used because it provides rapid results, has a low rate of contamination and cross-reaction with John Cunningham virus, and it is more sensitive than nested PCR.3
LightMix BK/JC Polyomavirus Detection kit (TIB MOLBIOL, Berlin, Germany, Ver 110401) and Light Cycler 1.5 device (Roche Diagnostics Corporation, Roche Applied Science) were used. The BK virus data were viewed on 498 to 640 channels on the quantitation mode. While the negative control did not emit a signal, internal control standards and positive sample data were viewed on channel 640 which was the detection channel. BK virus DNA positive results were recorded in copies per mL.
Definition of rejection
Patients were followed for 1 year regarding presence of rejection. Histopathologic rejection is not always accompanied by clinical findings.
Definition of histopathologic rejection
The histopathologic diagnosis of acute rejection is based on 3 histopathologic features: (1) the presence of mixed, but predominantly mononuclear, portal inflammation containing blastic or activated lympho-cytes plasmocytes, macrophages, neutrophils, and eosinophils; (2) inflammatory cell infiltration and damage in interlobular bile duct epithelium; and (3) subendothelial inflammation of portal vein branches in portal tracts and/or terminal hepatic veins. At least 2 of the above histopathologic features along with biochemical evidence of liver damage were considered diagnostic.4
Definition of clinical rejection
Clinical rejection required the presence of 1 or more of the following criteria: fever, fatigue, abdominal pain, loss of appetite, hepatosplenomegaly, reduction in the amount of bile coming from the T-tube, or an increase in liver enzyme activity.5
Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 16.0, IBM Corporation, Armonk, NY, USA). The Mann-Whitney U test was used for qualitative data, and the Fisher exact test was used for quantitative data. P values less than .05 were considered significant.
Study design and patient data
Liver transplant was performed on 46 patients in the general surgery department of our hospital during 2011. Among these patients, 5 of them (10.2%) were under 18 years of age and were not included in the study. Two (4.8%) of the 41 patients over 18 years of age, died before they could be included in the study. Four (10.2%) of 39 patients (84.7%) who met inclusion criteria for the study died after their first visit. One patient (2.5%) dropped out after the second visit, and 1 patient (2.5%) dropped out after the third visit. The remaining 33 patients (84.6%) turned up to participate in each visit for the following 3 months and completed 6 visits.
Blood and urine samples were collected on each visit from all patients. The total number of visits completed was 207, and a total of 414 samples (207 blood and 207 urine) was collected. Demographic characteristics are presented in Table 1. When the causes of liver failure were investigated in 39 cases, 16 of them (41.0%) were found to have hepatitis B, which was determined to be the most common cause. While viral infections were the cause of liver failure in 29 of these patients (74.3%), in 10 (25.7%) the cause was noninfectious.
No statistically significant association was found between the decoy cells and BK virus DNA positivity (urine and blood) and sex, age, presence of comorbidity, type of immunosuppressant treatment, and cause of liver failure (viral/non-viral). When the most common symptoms and the results of physical examination were evaluated separately for each visit in cases in which BK virus DNA was detected positive in blood and urine, these cases were asymptomatic and normal.
The effect of BK virus DNA positivity in blood and/or urine on hematologic and biochemical test results were investigated, and BK virus DNA positive and negative cases were compared respecting white blood cell count, and percentages of polymorphonuclear leukocytes, monocytes, and lymphocytes and biochemical assays (creatinine, aspartate aminotransferase, alanine amino-transferase), no significant difference was found between groups. On the first visit (day 15), the mean level of aspartate aminotransferase of positive cases was greater than mean level of aspartate aminotransferase of negative cases (P = .01). On third visit (day 45) just the opposite was observed: the mean level of aspartate aminotransferase of negative cases was found to be greater than mean level of aspartate aminotransferase of positive cases (P = .006).
Laboratory data collection (decoy cell and blood/urine real-time polymerase
At least once in 5 of 39 patients (12.8%) corresponding to 7 of the 207 (3.3%) blood samples, BK virus DNA was found to be positive by real-time polymerase chain reaction. At least once in 11 patients (28.2%) corresponding 23 of the total 207 urine samples (11.1%) tested, BK virus DNA was found to be positive by real-time polymerase chain reaction. Four patients presented with persistent viruria, 1 of whom tested positive for BK virus DNA in urine on all of the visits. In 13 of the patients (33.3%), decoy cells was found to be present at least once in the corresponding 26 of the 207 urine samples (12.5%). Decoy cells belonging to these patients are presented in Figure 1. In most patients, only 1 type of test (blood BK virus DNA, urine BK virus DNA, or decoy in urine) was positive for BK virus; however, 2 patients tested positive in both blood and urine BK virus DNA on separate visits.
BK virus DNA positivity ranged between 10 and 103 copies per mL and was found to be 104 copies per mL in 1 urine sample. Two other patients were found to have both BK virus DNA positivity in blood and decoy cells in urine on separate visits. The numbers of patients and samples which tested positive in the 3 methods are presented in Table 2. The distribution of the numbers of cases testing positive in blood/urine PCR and decoy cell screening, according to the date of each visit, is presented in Figure 2. Given BK virus DNA positivity in blood as the standard, the sensitivity and specificity of investigating decoy cells in urine and of BK virus PCR test in urine, and their positive and negative predictive values, are presented in Table 3.
Rejection was observed in 6 of the 39 patients (15.3%) evaluated respecting occurrence of rejection within 1 year after transplant (Table 1). All of the 6 patients in which rejection occurred were men (P = .304). Persistent viruria was observed in 2 of these 6 patients (P = .104) (number of patients with persistent viruria = 4). BK virus DNA was found in the blood or urine of 4 of these 6 cases in which rejection occurred (P = .163).
Rejection was observed in 2 of the 13 cases with decoy positivity; however, this was not statistically significant. One man received retransplant because of rejection related to hepatic artery thrombosis.
BK virus may lead to serious clinical manifestations if transported to a seronegative recipient via tissue from a seropositive donor, or if reactivated due to the use of immunosuppressive medication in a once seropositive recipient.6 Ninety percent of the world’s adult population is BK virus seropositive.7 In a study of 1123 healthy individuals, BK virus seroprevalence in Turkey was found to be 78.5%.8 Especially in renal transplant recipients, BK virus may lead to nephropathy, hemorrhagic cystitis, and urethral stricture in immunosuppressed patients. Nephropathy caused by BK virus is important because it is one of the causes of graft loss in kidney transplant recipients.7 However, its role has not been clearly established in non-renal solid-organ transplant recipients. There are studies on BK virus; however, there is no study of liver transplant groups in Turkey. To the best of our knowledge, this is the first study in our country that investigates BK virus in an liver transplant group.
No statistically significant association was found between demographic characteristics of patients and BK virus positivity. While the decoy cells in the urine were not present in any of our patients diagnosed with diabetes, none of the 13 patients testing positive in urine decoy was found to be diabetic. In a literature search, no study was found investigating the association between diabetes and decoy cells in liver transplant groups. However, in their study on renal transplant recipients, Hirsch and associates9 reported older age, male sex, presence of diabetes, and white race to be risk factors for BK virus-related nephropathy. This discrepancy may be due to the type of organ transplant. The significance of BK virus in renal transplant recipients is clear; however, there is a need for studies done in liver transplant groups. In our study, no complaint or physical examination result was found to indicate BK virus infection among BK virus positive patients. Munoz and associates10 reported not being able to find any symptom or indication that could be related to BK virus infection among liver transplant recipients in which they found BK viruria.
In our center, the most frequently used immuno-suppressive regimen in liver transplant recipients was cyclosporine plus mycophenolate mofetil plus corticosteroids. Studies report various agents such as antirejection treatment regimens with pulse steroid,11 steroid use and antithymocyte globulin induction therapy,12 high level of tacrolimus,13,14 and mycophenolate mofetil,14,15 as being responsible for the emergence of BK viremia and viruria. In our study, no statistically significant association was found between the type of immunosuppressant and BK virus positivity. Likewise, Loeches and associates16 did not find a significant association between immunosuppressant type and BK virus positivity. The fact that each study associates a different type of immunosuppressant with BK virus positivity is the reason why a consensus is not reached on this matter. Yet considering that immunosuppressant medication is generally delivered to patients in combinations of 3, it would seem to be impossible to evaluate this association respecting only one type of medication.
When the biochemical findings of our study were examined, the mean level of aspartate aminotransferase in BK virus DNA positive cases was found to be greater than the mean level of aspartate aminotransferase in BK virus DNA negative cases on the first visit; but on the third visit just the opposite was observed, the mean level of aspartate aminotransferase and BK virus DNA negative cases was found to be greater than that of BK virus DNA positive cases. Because many various factors such as medication, surgical technique used, and rejection status in this patient group may be the cause of fluctuations in aspartate aminotransferase values, the statistical significance found in our study was not concluded to be resulting solely from BK virus positivity.
The rate of viruria (28.2%) was higher than the means found in other studies conducted on liver transplant recipients. In similar series of studies conducted in liver transplant groups, Splendiani and associates17 reported 6.2%, Munoz and associates10 7.8%, Marshall and associates18 8%, Doucette and associates19 12%, and Loeches and associates16 reported a 21% rate of BK viruria; while Barton and associates15 found no viruria or viremia in their study of 8 patients with liver transplant. Varying viruria rates were considered to be related to study design and investigation of BK virus in urine at different times.
In our study, viruria was found in 5 patients (12.8%). Both viremia and viruria were found in the same patient on separate visits (2 of 5 patients); however, in neither of them was viruria found at the same time in urine on the visit that viremia was found. Among studies conducted by Mitterhofer and associates,20 which found viruria and viremia at the same time in candidates for liver transplant, there are many studies in which viruria without viremia or viremia without viruria occurred.10,16,21-24
Additionally, there are studies that show that viruria and viremia levels and/or persistency may be important in developing nephropathy.16,21,25,26 In our study, BK virus DNA positivity was found to range between 10 and 103 copies per mL with a single urine sample of 104 copies per mL. In their study comprising a group with renal transplant and hematologic malignancy, Rota and associates25 observed DNA ≥ 107 copies per mL in 24.7% of the urine samples. Compared with the liver transplant recipient group, the BK virus may be considered to be found with a greater number of copies in the renal transplant group. The American Society of Transplantation accepts that if polyomavirus DNA is found to be ≥ 107 copies per mL in urine and ≥ 104 copies per mL in plasma, this may lead to the development of clinical nephropathy in renal transplant.26 BK virus DNA was found to range between 10 and 104 copies per/mL in our study, and it is thought that because of these low copy numbers, nephropathy was not observed during follow-up. In support of this hypothesis, in a study conducted on a pediatric liver transplant recipients group, the average BK viral load in urine was found to be 102 copies per mL and kidney functions were reported to be unimpaired in this group.21 On the other hand, Loeches and associates,16 in their study on liver transplant recipients, showed that BK viral load, on average, may be as high as 105 copies per mL in blood and 106 copies per mL in the urine. These results imply that when investigating BK virus-associated nephropathy, BK virus should be simultaneously screened for in both samples (blood and urine), and that the observation of viremia and viruria levels may be important.
In their study of 62 patients, in which they screened only for BK virus in liver transplant recipients, Loeches and associates16 determined viremia within the first 3 months in 10 of 11 patients that had viremia. One year later, viremia was not found in any of the patients. Likewise, in our study, in 3 out of 5 patients with viremia, it was detected within the first month after transplant. These results show that viremia should be screened for, especially within the first week and months after transplant. BK viruria, on the other hand, may emerge, after primary BK virus infection, 10 days to 6 weeks after transplant. In case of reactivation or reinfection via BK virus strain from donor in those who are BK virus seropositive before transplant viruria may emerge 5 weeks to 17 months after transplant and may persist for weeks or years.27 This shows that viruria should be monitored for a longer time.
In our study, while BK virus was investigated via real-time polymerase chain reaction, decoy cells also were simultaneously screened for in urine; however, decoy cell positivity was not correlated with BK virus PCR results. The investigation of decoy cells is a screening method that is simpler and less expensive than PCR. However, the fact that decoy cells also may emerge in other virus infections (John Cunningham virus, Simian virus 40, cytomegalovirus, adenovirus),28 and that they may be discharged with urine in asymptomatic individuals and may be detected in urine in the presence of malignity,29 renders its value questionable in diagnosing BK virus nephritis. This is supported by the fact that Salama and associates,30 in their study comprising only liver transplant recipients, reported finding decoy cells in 9.7% of the patients without detecting viruria in any of them. Because BK virus may be discharged in urine in asymptomatic individuals even if nephritis is not clinically present, no matter which method is used for screening, the diagnostic value of viruria in urine is lower than that of BK viremia in diagnosing BK virus infection.31 The detection of BK virus DNA in the blood is more valuable in diagnosing BK virus-associated nephropathy. Nickeleit and associates32 found BK viremia in all patients with BK virus-associated nephropathy. Furthermore, in 50% of the patients, BK virus was detected in blood 16 to 33 weeks before any clinical evidence of nephropathy emerged. The detection of BK virus DNA in the blood by PCR is a more-sensitive and more-specific test compared with the other 2 methods. The sensitivity and specificity of the BK virus PCR test in blood have been reported as 100% and 88%.11 In our study, while the sensitivity and positive predictive value of BK virus PCR and decoy screening tests in the urine was lower than BK virus PCR test in blood, their negative predictive values were found to be 89.2% and 88.4% (Table 3). Likewise, while evaluating the decoy screening results that were not supported by PCR in their study, Randhawa and associates33 emphasize that the low sensitivity of this test should not be disregarded.
When the relation between BK virus and rejection is investigated, rejection occurred in 15.3% of the patients, and in 4 of these, BK virus was positive in either blood or urine. In 1 patient, rejection was diagnosed by a biopsy taken 3 months after the visit in which viruria was detected. In another patient, rejection was diagnosed by a biopsy taken because of liver function impairment 4 days after the visit in which BK virus DNA tested positive in blood. Six days after this visit, BK viruria was detected in this patient. Diagnosed with rejection because of hepatic artery thrombosis, the patient received a retransplant. No association between BK virus and hepatic artery thrombosis was encountered in our literature research. In a liver transplant recipients presenting with viremia for at least 3 months, Loeches and associates16 report both renal failure and findings of liver biopsy supporting viral infection, in the development of acute rejection. These results led us to consider that BK virus positivity may be a factor facilitating rejection in liver transplant recipients.
For this study to be cost-effective, the BK virus seropositivity of patients before transplant could not be investigated; patients were monitored for only 3 months after transplant, and the possible presence of other accompanying viral factors such as cytomegalovirus could not be screened for; these factors contribute to the limitations of this study. The evaluation of seropositivity and the monitoring of patients for a long time, especially in case of viruria, and the screening for accompanying cytomegalovirus DNA will provide more detailed data for future prospective studies.
As a result, no specific symptom and/or clinical finding could be detected unique to the diagnosis of BK virus in liver transplant recipients. Although no statistically significant result was seen related to routine blood count measures and kidney-liver function tests, the BK virus was considered to be an significant factor worth investigation when unexplained impairments are detected in kidney and liver functions. However, the joint evaluation of tests for screening of the virus in urine and blood samples was found to be important, because no matter which method is used for the screening of BK virus in urine, the negativity of the tests was found to be more valuable than the positivity. Therefore, these tests may be used routinely for excluding BK virus infections; however, when positivity is detected, the results should be evaluated together with BK virus PCR in blood. Prospective studies with more patients are needed to investigate the role of BK virus in the development of rejection.
Volume : 12
Issue : 5
Pages : 429 - 436
DOI : 10.6002/ect.2013.0224
From the Department of 1Infectious Diseases and Clinical Microbiology,
Surgery, and 3Pathology, Dokuz Eylul University Faculty of Medicine, Izmir,
Acknowledgements: None of the authors reported any conflict of interest. This study was supported by The Dokuz Eylül University Scientific Research Projects Unit fund number 2011.KB. SAG.014.
Corresponding author: Vildan Avkan-Oguz, MD, Prof, Dokuz Eylul University School of Medicine, Department of Infectious Diseases and Clinical Microbiology, 35340 Inciralti, Izmir, Turkey
Phone: +90 232 412 43 06 / +90 532 480 26 35
Fax: +90 232 278 59 54
E-mail: firstname.lastname@example.org email@example.com
Table 1. Demographic Characteristics
Figure 1. H&E stain, original magnification × 100
Table 2. The Newcastle-Ottawa Quality Assessment Scale for Studies Included in the Meta-Analysis
Figure 2. Distribution of Positive Cases According to Date of Visits
Table 3. Specificity and Sensitivity Rates and Negative and Positive Predictive Values of the Methods According to Viremia