Correlation between CMV Infection and NODAT

Purpose: New-onset diabetes mellitus after transplantation (NODAT) is a well-known complication of transplantation. Materials and methods: Retrospectively, we detected CMV replication (PCR) in every month after transplantation of kidney in the first 12 months after transplantation in patients in a homogenous group from the aspect of immunosuppresion. Results: In the group of 167 patients (control group: n = 103, NODAT group: n = 64), the average value of CMV viremia was without any significant difference between the NODAT group and the control group (P = 0.9285). In the 10th month after kidney transplantation, we recorded significantly higher CMV viremia in the NODAT group (p < 0.0001), however, in the multi variant analysis, that difference was not confirmed. Thus, in our group, CMV is of no relevance with the development of NODAT in the monitored period. The survival of patients and graft was 12 months after kidney transplantation without any statistically significant difference between the monitored groups (P = 0.6113 survival of the patient; P = 0.5381 – survival of the graft). Conclusion: Our analysis shows that in regular monitoring of CMV viremia and applying chemoprophylaxison the risk recipeints, CMV is not the risk factor for NODAT.


Introduction
New-onset diabetes mellitus after transplantation (NODAT) is a well-known complication of transplantation and its development is associated with lower graft function and survival and reduced longterm patient survival mainly because of cardiovascular events [1][2][3]. Kidney transplant recipients who develop NODAT have variably been reported to be at increased risk of fatal and nonfatal cardiovascular events and other adverse outcomes including infection, reduced patient survival, graft rejection, and accelerated graft loss compared with those who do not develop diabetes. Identification of high-risk patients and implementation of measures to reduce the development of NODAT may improve the long-term patient and graft outcome [4]. In 2003, the International Expert Panel consisting of experts from both the transplant and diabetes fields set forth the International Consensus Guidelines for the diagnosis and management of NODAT [5,6]. It was recommended that the definition and diagnosis of NODAT should be based on the definition of diabetes mellitus and impaired glucose tolerance (IGT) described by the World Health Organization (WHO) [6,7]. The American Diabetes Association (ADA) guidelines for the diagnosis diabetes mellitus are provided in Table 1 [4].
Cytomegalovirus (CMV) is one of the most important infections in renal transplant recipients [8][9][10][11][12]. Exposure to the virus, as indicated by presence of detectable immunoglobulin G (IgG) anti-CMV antibodies in the plasma, increases with age in the general population and is present in more than two-thirds of donors and recipients prior to transplantation [8]. It is therefore common for the donor and/or recipient to be CMV-positive at the time of transplantation.
CMV can be transmitted from the donor either by blood transfusion or by the transplanted kidney; the concurrent administration of immunosuppressive drugs to prevent rejection further increases the risk of clinically relevant CMV disease, with induction therapy principally being associated with an increased risk of disease [13,14]. Thus, both the recipient and the donor are routinely tested for anti-CMV antibodies prior to transplantation. CMV disease may manifest as a nonspecific febrile syndrome (e.g. fever, leukopenia, and atypical lymphocytosis) or tissue-invasive infections (e.g. hepatitis, pneumonitis, and enteritis). Tissue-invasive CMV disease is defined as CMV disease and CMV detected in tissue with histology, NAT or culture [15].
The link between cytomegalovirus (CMV) infection and the development of NODAT was first reported in 1985 in a renal transplant recipient [16]. Limited studies suggested that both asymptomatic CMV infection and CMV disease are independent risk factors for the development of NODAT. In a study consisting of 160 consecutive nondiabetic renal transplant recipients who were prospectively monitored for CMV infection during the first three months after transplantation, Hjelmesaeth and colleagues found that asymptomatic CMV infection was associated with a four-fold increased risk of new-onset diabetes (adjusted RR = 4.00; p = 0.025) [17]. Patients with active CMV infection had significantly lower median insulin release compared to their CMV negative counterparts, suggesting that the impaired pancreatic β cell insulin release may be involved in the pathogenic mechanism of CMV-associated NODAT. It is speculated that CMV-induced release of proinflammatory cytokines may lead to apoptosis and functional disturbances of pancreatic β-cells [18]. Randomized controlled trials have demonstrated that the incidence of CMV disease can be reduced by prophylaxis and preemptive therapies in solid-organ transplant recipients [19][20][21].
According to the recommendations of KDIGO, CMV chemoprophylaxis is indicated (except when donor and recipientboth have negative CMV serologies) by applying the oral ganciclovir or valganciclovir for the period of minimum 3 month after kidney transplantation and for the period of 6 weeks after the kidney transplantationin case of T-cell-depleting antibody therapy [15].
In our department, we apply chemoprophylaxis (valganciclovir) in case of seronegative recipient or seropositive donor (R+/D-) 100 days after transplantation. In case of applying antithymocyte globulin, we apply the prophylaxis for the period of 6 weeks. However, CMV viremia (by using the polymerase chain reaction -PCR) is monitored regularly in all recipients, except for R+/D-, as follows: the first 6 months after transplantation every 2 weeks, and from the 6 th -12 th month after transplantation 1x per month.

Materials and Methods
In the retrospective analysis, we monitored CMV viremia as the risk factor for NODAT in the group of patients who underwent primary transplantation of kidney from a deceased donor in the Transplantation Center in Martin in the years 2009-2013. The patients with diabetes mellitus type 1 and 2 and the patients who had not finished 12 months from kidney transplantation were excluded from monitoring. The patients who had the mTOR inhibitor or cyclosporin A in their immunosuppresive regime were also excluded from monitoring because of prevent of results distortion by immunosuppression. In each patient, we recorded the age at the time of transplantation, sex, we identified recipients with risky HLA for NODAT and with polycystic kidney diseases, we recorded the number of HLA mismatches and the type of donor (ECD). In each patient, we identified CMV viremia (by PCR) as customary in our department: 2x per month during the first 6 months from kidney transplantation and 1x per month from the 6 th -12 th month after kidney transplantation. Retrospectively, we identified the patients who had a symptomatic CMV disease in the monitored period. The patients were divided into two sub-groups according to the development of NODAT in the monitored period -the control group and the NODAT group. NODAT was diagnosed in a standard way according to the ADA criteria. The groups were compared from the aspect of development of NODAT and CMV viremia during the entire monitored period, during the first 6 months after transplantation and the next following 6 months after transplantation, and we also compared CMV viremia in every month after kidney transplantation. In the end, we compared the function of the graft 1 year after kidney transplantation (by the estimate of glomerular filtration -eGFR by applying the CKD-EPI creat 2009 formula) and we compared the 12-month survival of the graft (censored for death) and the patients. By the correlation coefficient, we identified whether CMV viremia affects the function of the graft 12 months after kidney transplantation. All risk patients in the monitored group, i.e. the seronegative recipients who received the organ from a seropositive donor and the recipients who received the T-cell-depleting antibody were administered chemoprohylaxis. In case of a seronegative recipient, it was 100 days from kidney transplantation and in case of the T-cell-depleting antibody, it was 6 weeks from administration.
We used a certified statistical program MedCalc version 13. 1. 2. for statistical evaluation and we used the following statistical analyses: Student's t-test, chi-square test, correlation coefficient, Logistic regression, Cox proportional hazard model, Kaplan-Meier curves of survival. We consider the value p < 0.05 to be statistically significant.

Results
The group was composed of 167 patients, including 103 (61.7%) patients who were included in the control group and 64 (38.3%) patients in the group with development of NODAT in the monitored period. The average level of tacrolimus (during the monitored 12 months from kidney transplantation) was in both groups without any statistically significant difference (P = 0.5592), and similarly was the average dose of prednisone/day (P = 0.0877). The average dose of mycophenolate mofetil/dayor mycophenolate sodium was also without any statistically significant difference between the monitored groups (P = 0.0919 -mycophenolate mofetil and P = 0.1734 -mycophenolate sodium). In view of that, both groups were homogenous from the aspect of the applied immunosuppression, and the individual monitored parameters were not distorted by the applied immunosuppresion ( Table 2). The characteristics of the group are given in Table 3. The patients in the NODAT group were significantly older than the patients in the control group, and during the monitored 12 months, the patients in the NODAT group received a statistically significant higher dose of methylprednisolone. However, in the multivariant analysis, the dose of methylprednisolone as an independent risk factor for NODAT was not identified (   The average value of the CMV viremia (cop/ml) in the NODAT group and in the control group was without any statistically significant difference. We compared replication of CMV in the first 6 months after kidney transplantation with replication during the second halfyear after transplantation, and we recorded in both groups significantly higher replication of CMV infection in the first half-year after transplantation. However, upon comparing the control group versus the NODAT group, no difference in replication in the first and in the second half-year after kidney transplantation was recorded ( Figures  1-4).
CMV viremia in individual months after kidney transplantation is presented in Table 5. In the 10 th month after kidney transplantation, we recorded significantly higher CMV viremia in the NODAT group, however, such difference was not confirmed in the multivariant analysis (Tables 5 and 6). In view of the foregoing, in our group, CMV is of no relevance to development of NODAT in the monitored periodin the first 12 months from kidney transplantation. We discovered that significantly more patients (70%) had diagnosed NODAT in the first 6 months after transplantation (P < 0.001) ( Figure 5).
In the whole group, we identified the patients who developed the symptomatic form of CMV infection. In the whole group, it was only 6% patients. In the NODAT group, 10.9% patients had the symptomatic CMV infection, and in the control group it was 2.9% (0.0741).
The value of creatinine 12 months after transplantation was comparable in both groups, and also the eGFR (the limit of statistical significance) ( Table 7). By the correlation coefficient we discovered that the higher number of copies CMV/ml worsens the function of the graft (characterized eGFR) 12 months after kidney transplantation (Figures 6 and 7).

Discussion
Many risk factors have been found to have influence on the development of NODAT. In 1985 Lehr et al. reported a case of cytomegalovirus (CMV) induced NODAT in a kidney recipient patient, after that the role of CMV infection in NODAT has been an area of interest to researchers [16]. Since then, other studies have supported [17,22] the relationship between them whilst other studies [23,24] have failed to prove this association. However, the influence of CMV infection on developing NODAT still remains a question. If the impact of CMV infection on higher incidence of NODAT is proven, initiating prophylaxis against CMV infection after transplantation will be strongly suggested [25]. In meta-analysis of authors Einollahi et al., it was discovered that the risk of NODAT in kidney transplants with CMV infection was 1.94 fold more as compared to individuals without CMV infection using adj ORs from the studies [26]. This significant relationship was proved by overall pooled OR using un-adj ORs. There was a difference in the result of evaluated studies in term of CMV induced NODAT. Though, three studies [27,28] reported no significant relationship between CMV infection and NODAT; other studies [17,23,29] detected CMV infection as the risk factor for NODAT. In         addition, Valderhaug et al. [30] only found the relationship between CMV infection and NODAT in univariate analysis whilst a multivariate analysis, adjusted for age, prednisolone, type of cohort, HLA-B27 phenotype and BMI did not support this association.
According to the above mentioned metaanalysis of the authors Einollahi et al., the studies used different criteria to identify CMV infection. Isolation of the CMV virus and detection of viral proteins or nucleic acid are different ways to recognize CMV infection [27]. In addition, active systemic CMV infection can be diagnosed as CMV-DNA in plasma by polymerase chain reaction methods or by detection of CMV-antigenemia in leucocytes (i.e., CMVpp65) (18). Four from seven works in analysis did not report a criterion for identifying CMV infection [22,[27][28][29]. Three remaining studies used different criterion to recognize CMV infection; Hjelmesaeth et al. [17] defined CMV infection as one or more CMV pp65 antigen-positive cells per 100.000 leucocytes, Marin et al. [24] defined it as more than 50 infected cells per 200,000 leucocytes using the pp65 assay or isolation of CMV antigenemia or fourfold increase in the baseline IgG and Valderhaug et al. [30] diagnosed it by CMV-pp65 antigen in leucocytes or CMV-DNA in plasma, but they did not report any details. Thus using various criteria and methods with different sensitivity and specificity can lead to an overestimate or may in fact underestimate CMV infection in the studies. The studies which determine CMV viremia by PCR may expain the relationship between CMV and NODAT.
In our group, we had not detected any relationship between replication of CMV and development of NODAT. The group was homogenous from the aspect of immunosuppresion. The results of our analysis and the low occurrence of symptomatic CMV infection is, in our opinion, related to the intensive monitoring of CMV viremia (PCR) after transplantation (the first 6 months, CMV viremiais determined every 1 month, in the second half-year every 6 weeks). In the risk patients (seronegative donor and seropositive recipient), we monitor CMV viremia also in the second year after transplantation, every 2 months. The patients who were treated by T-cell-depleting antibody, have in our center monitored CMV viremia every 1 month for 3 months after the end of therapy. All recipients with the increased risk of CMV infection were administered chemoprophylaxis according to the KDIGO recommendations of 2009 [15].
Prospective observational cohort study of authors Smedbråten et al. is an extension of the previous study reporting effects of cytomegalovirus (CMV) on the graft and patient survival in 471 patients who underwent kidney transplantation between 1994 and 1997. None of the patients received CMV prophylaxis or preemptive treatment. CMV infection was an independent significant risk factor for mortality in multivariate analysis (HR = 1.453, 95% CI 1.033-2.045, p = 0.032) [31]. This observed association between CMV infection and long-term graft and patient outcome may be altered by prophylaxis or preemptive CMV therapy. In a study by Kliem et al., oral ganciclovir prophylaxis was compared to intravenous preemptive CMV therapy [32]. Compared to preemptive therapy, prophylaxis was found to be significantly associated with improved long-term (4 yr) uncensored graft survival, with the greatest benefit observed in the donor +/ recipient + CMV serostatus group. Moreover, when analyzing the death-censored graft survival, prophylaxis significantly improved graft survival in the donor +/recipient + CMV serostatus group. Opelz et al. reported from analyses of register data that CMV prophylaxis was significantly associated with improved graft survival both censored and uncensored for death; but in both cases only in the donor +/recipient -CMV serostatus group [33]. In our group, we identified relationship between CMV viremia and function of the graft 12 months after kidney transplantation. With the increasing number of CMV copies/ml, eGFR is worsened one year after kidney transplantation.
CMV replication after transplantation may contribute to reduced graft function and survival through the associated inflammation and       cytokine release [34]. Uncontrolled replication of CMV triggers direct and/or indirect effect in transplant recipients [35]. When CMV is reactivated under immunosuppressive conditions, it has both direct effects, such as development of CMV disease, and indirect effects on transplantation, including increased incidence of allograft rejection [36].
In our study, survival of the patients (censored for death) as well as survival of the graft is numerically worse in the NODAT group, no statistically significant difference was confirmed. We assume that the survival of patients with NODAT is significantly worse from the longterm aspect (10 and more years). Intensive monitoring of glycaemia and early diagnostics and treatment of NODAT, as well as check-up of the other affectable risk factors for NODAT are able to significantly improve survival of both the patients and the graft and to decrease the cardiovascular mortality and morbidity.

Conclusion
Our analysis suggests that by regular monitoring of CMV viremia and applying chemoprophylaxis for risk recipients (in our case, seropositive donor/seronegative recipient or recipient after T-celldepleting antibody therapy) is not the CMV risk factor for NODAT. The results of the analysis (or CMV viremia or development of NODAT) are not distorted by the administered immunosuppressive treatment -from the aspect of immunosuppression, the group was homogenous. Regular monitoring of CMV viremia and chemoprophylaxis may affect not only development of NODAT, but it is possible (as in our group) to eliminate the number of patients with symptomatic CMV infection. Frequent monitoring of CMV viremia may increase the costs on care about the recipient after kidney transplantation, however, we eventually decrease the costs on treatment of later complications arising from CMV infection. In the patients after kidney   transplantation, the most important risk factor for NODAT is the age at the time of transplantation (more than 50 years of age), prediabetes before transplantation, positive family history of diabetes mellitus type 2, and the body mass index of more than 30 kg/sqm at the time of transplantation. Regular weight and waist circumference control in patients after kidney transplantation leads to identification of risk patients for NODAT. Screening of risk factors for diabetes mellitus should be done even before placing the patient on the waiting list and it is advisable to carry out the oral glucose tolerance test (oGTT) also in patients with physiological levels of fasting glycemia [37,38].

Funding
This paper was supported by VEGA grant 1/0160/16 Diagnostic -prognostic relevance of adipocytokine network and glucose homeostasis assessment in cachexia.