The impact of early cytomegalovirus infection and disease in renal transplant recipients S. Sagedal 1, A. Hartmann 1,2 and H.

Transcription

1 REVIEW ARTICLE /j x The impact of early cytomegalovirus infection and disease in renal transplant recipients S. Sagedal 1, A. Hartmann 1,2 and H. Rollag 3 1 Department of Internal Medicine, 2 Laboratory for Renal Physiology and 3 Institute of Medical Microbiology, Rikshospitalet University Hospital, Oslo, Norway ABSTRACT Human cytomegalovirus (HCMV) infection is the single most frequent infectious complication in the early period after kidney transplantation. The HCMV load in blood, measured by HCMV PCR or the HCMV pp65 antigen test, is a predictor of HCMV disease in seropositive recipients. However, plasma virus load measurements are of only modest value in predicting the risk of HCMV disease in seronegative recipients of kidneys from seropositive donors. HCMV infection is an independent riskfactor for acute kidney graft rejection. There is also evidence that HCMV is associated with an increased long-term mortality and post-transplant diabetes mellitus. Whether pre-emptive or prophylactic therapy should be the preferred strategy is not yet decided. Some studies indicate that HCMV prophylaxis may reduce the risk of acute rejection, and thereby increase long-term graft survival in seronegative recipients of kidneys from seropositive donors. Keywords Cytomegalovirus disease, human cytomegalovirus, kidney graft rejection, patient survival, review, transplantation Accepted: 30 January 2005 Clin Microbiol Infect 2005; 11: INCIDENCE AND DEFINITION OF HUMAN CYTOMEGALOVIRUS INFECTION AND DISEASE Human cytomegalovirus (HCMV) is the single most frequent cause of infectious complications in the early period after kidney transplantation, with overall incidences of HCMV infection and disease during the first 100 days post-transplantation of c. 60% and 25%, respectively, when no HCMV prophylaxis or pre-emptive therapy is given [1 4]. The frequency of HCMV infection is the same for HCMV IgG-seronegative recipients (R ) with seropositive donors (D+) (primary infection) as for pre-transplant seropositive recipients (R+), whereas HCMV disease is nearly three-fold more frequent in the serostatus group D+ R than for D+/R+ [2,4]. HCMV infection is defined as isolation of HCMV, or detection of HCMV proteins or nucleic Corresponding author and reprint requests: S. Sagedal, Department of Medicine, Rikshospitalet University Hospital, N-0027 Oslo, Norway solbjorg.sagedal@rikshospitalet.no acid, in any body fluid or tissue specimen [5]. HCMV disease is defined as detection of HCMV in a clinical specimen, accompanied by either HCMV syndrome with fever, muscle pain, leukopenia and or thrombocytopenia (other causes excluded), or by organ involvement, such as hepatitis, gastrointestinal ulceration, pneumonitis, retinitis, central nervous system disease, nephritis, myocarditis, cystitis or pancreatitis [5]. DIAGNOSTIC TESTS Conventionally, HCMV infection has been diagnosed by serology, defined as seroconversion of anti-hcmv IgG and IgM antibodies (for primary HCMV infection), and by a significant rise in IgG antibodies (for secondary infection). Today, serological tests are of importance for the assessment of the serological status of organ transplant recipients at the time of transplantation, thus defining the risk for HCMV disease, but have little diagnostic value thereafter. Viraemia is defined by the isolation of HCMV by culture involving the use of either standard cell culture or shell vial culture techniques [5]. Ó 2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases

2 Sagedal et al. Human cytomegalovirus in renal transplant recipients 519 Antigenaemia is defined by the detection of HCMV pp65 antigen in leukocytes, while DNaemia is defined by the detection of HCMV DNA in samples of plasma, whole blood or isolated peripheral blood leukocytes, measured by a nucleic acid amplification technique such as PCR [5]. Tests for detection of antigenaemia or DNaemia are far more sensitive than virus isolation from blood. Finally, RNaemia is defined by the detection of RNA in samples of whole blood, isolated peripheral leukocytes or buffy coat specimens [5]. There is a good correlation [6] between the presence of pp67 mrna and pp65 antigenaemia of > 10 positive cells leukocytes. THE PREDICTIVE VALUE OF HCMV PP65 ANTIGENAEMIA AND HCMV DNA LOAD Kidney graft recipients with HCMV disease have significantly higher maximum HCMV pp65 counts than those who remain asymptomatic [4,7 9]. This is also true when the serostatus groups D+ R+ and D R+ are considered separately, but is not true in the D+ R serostatus group [4]. This indicates that the number of cells positive for HCMV pp65 is a better predictor of HCMV disease in reactivated than in primary HCMV infection. Among HCMV-seropositive recipients, 50% developed HCMV disease with maximum HCMV pp65 antigen counts of at least leukocytes [4] (Fig. 1). In the D+ R serostatus group, > 80% of those with an HCMV pp65 count of leukocytes developed HCMV disease. However, the practical application of measuring peak virus loads is limited because the maximum is often reached after the appearance of HCMV disease. In accordance with the above results, Humar et al. [10] concluded that routine monitoring would have predicted disease in only 38% of seronegative recipients with seropositive donors, indicating that plasma virus load measurements are of only modest value in this patient group. In contrast, Hassan-Walker et al. [8] used data from a much smaller study of 87 renal allograft recipients to underline the importance of virus load in patients with primary infection. Peak HCMV load was one of the covariates in a multivariate logistic regression analysis of riskfactors for HCMV disease, irrespective of whether peak load occurred before or after the initiation of Fig. 1. The percentage of first renal graft recipients (n = 397) who developed human cytomegalovirus (HCMV) disease as a function of the maximum HCMV pp65 antigen count before or during HCMV disease. The upper curve represents seronegative recipients with seropositive donors (D+ R ; n = 72) and the lower curve represents seropositive recipients (R+; n = 286). For instance, in R+ patients with a maximum HCMV pp65 antigen count 1, the risk of HCMV disease is 29%. In D+ R patients with a maximum HCMV pp65 antigen count 1, the risk is 82%. (Taken from the study of Sagedal et al. [4] and printed with permission from Lippincott Williams & Wilkins.) HCMV disease. However, it is not feasible for a peak virus load occurring after the initiation of HCMV disease to be a risk-factor for disease, so the statistics used in this study may be inappropriate. Furthermore, in a study of 52 solid-organ recipients with HCMV disease, Humar et al. [11] found that virus load half-life was significantly longer in patients with recurrence than in those without recurrence of disease, and also found that failure to clear virus was predictive of relapse. HCMV INFECTION AND DISEASE In multivariate analyses, episodes of acute clinical rejection and recipient age are independent riskfactors for HCMV infection, while independent risk-factors for HCMV disease are episodes of acute clinical rejection and the serostatus group D+ R [4,12]. The fact that the serostatus group D+ R is a risk-factor for HCMV disease is accepted widely [13]. The large United States Renal Data System study reported significantly more HCMV disease in D+ R patients compared to all other kidney transplant recipients (including D R ) [14]. However, it seems inappropriate to include D R patients as a control

3 520 Clinical Microbiology and Infection, Volume 11 Number 7, July 2005 group for D+ R patients, since D R patients very seldom become infected by HCMV with development of HCMV disease. Steroid-resistant rejection is also a risk-factor for HCMV disease [15,16]. The risk of HCMV disease associated with anti-thymocyte globulin and muromonab-cd3 (OKT3) is higher when these agents are given as treatment of rejection than when they are used for induction therapy [17]. When evaluating risk-factors for HCMV disease in 87 kidney transplant recipients, Emery et al. [18] found that donor recipient serostatus and receipt of anti-thymocyte globulin for rejection were only of borderline significance in univariate analyses, and both became non-significant after adjustment for initial HCMV load. The authors considered that the classical risk-factors of donor recipient serotatus were explained entirely by virus load. However, logistic regression analysis was used, irrespective of the fact that HCMV disease appeared at different timepoints after transplantation. Initial virus load is also a time-dependent covariate, which can sometimes be measured only after the appearance of HCMV disease. In addition, the authors did not state whether only anti-thymocyte globulin therapy preceding HCMV disease was included. Cox regression analysis would probably be a more appropriate statistical method to use. In a much smaller study [19], a high initial plasma HCMV DNA load, but not the initial HCMV pp65 count, was highly predictive of development of HCMV disease, although multivariate analyses were not performed. Several studies have reported an increased incidence of HCMV disease in patients receiving mycophenolate mofetil (MMF) [14,20,21]. In the USA renal transplant MMF study [22], there was a trend towards a higher incidence of HCMV tissue-invasive disease in the MMF groups given 2 g and 3 g, compared to the group given azathioprine, but the difference was not statistically significant. However, this study was statistically powered to find differences in graft rejection, not in HCMV disease. Also, in the European and tricontinental MMF studies [23,24], the frequency of HCMV tissue-invasive disease was highest in the MMF group given 3 g. Other studies have failed to show any association between maintenance MMF treatment and an increased frequency of HCMV disease [12,25,26]. Interestingly, Giral et al. [25] found that HCMV disease was associated strongly with uncensored graft loss only in the group treated with azathioprine. In this study, all patients with HCMV disease were treated with ganciclovir for 14 days. It has been postulated that a reported MMF-induced enhanced anti-herpes virus activity of ganciclovir could contribute to the favourable effect on graft survival in the MMF-treated group [27]. THE IMPACT OF HCMV INFECTION AND DISEASE ON ACUTE REJECTION EPISODES Several studies have shown that HCMV is a riskfactor for episodes of acute rejection [2,28 30]. In three of these studies, the diagnosis of HCMV infection was based upon positive culture or seroconversion [2,29,30]. The results of clinical studies investigating HCMV as a risk-factor for acute renal graft rejection are summarised in Table 1. Two other studies implicate HCMV infection indirectly as a risk-factor for episodes of acute rejection [30,31]. Lowance et al. [31] showed that Study design Number of patients Results Reference Retrospective 1424 HCMV disease was associated significantly with subsequent biopsy-proven acute rejection Prospective 451 Both HCMV infection (positive pp65 antigen) and HCMV disease were independent predictors of clinical acute rejections. HCMV disease was an independent predictor of tubulointerstitial acute rejection Retrospective 192 HCMV disease, but not asymptomatic HCMV infection, was associated significantly with biopsy-verified acute rejection Prospective 170 HCMV infection, but not HCMV disease, was associated significantly with subsequent episodes of clinical acute rejection Prospective 84 Clinical and biopsy-proven rejections were as frequent in the HCMV risk group (D+ and or R+) during the first 4 months after transplantation and at 5-year follow-up as in the D R risk group [92] [28] [2] [29] [93] Table 1. Clinical studies investigating human cytomegalovirus (HCMV) as a risk-factor for episodes of acute rejection in kidney transplant recipients

4 Sagedal et al. Human cytomegalovirus in renal transplant recipients 521 rejections were reduced by 50% in the D+ R serostatus group by valacyclovir prophylaxis following transplantation for 90 days. Similarly, in a retrospective register study by Opelz et al. [32], an inhibitory effect of HCMV prophylaxis on acute rejection was suggested. Among seronegative recipients with seropositive donors with a functioning graft after 1 year, significantly fewer patients received rejection treatment in the preceding year (26.3% vs. 32.4%; p < ) in the group that received HCMV prophylaxis. However, the duration of HCMV prophylaxis and the antiviral agents used were not stated. HCMV infection in the transplanted kidney may result in HCMV inclusions in capillary endothelial cells and tubular epithelial cells, as seen by light microscopy [26,33]. The HCMV inclusions may or may not be accompanied by interstitial nephritis [26,34]. However, even with elevated serum creatinine and apparent HCMV disease, HCMV inclusions are detected in < 1% of kidney allograft biopsy specimens [26]. The results of sensitive methods, e.g., tissue HCMV PCR, may be positive even in the absence of HCMV inclusions, indicating that HCMV nephropathy is much more common than reported previously [35]. HCMV infection of the transplanted kidney may itself contribute to graft dysfunction with an elevated serum creatinine level, and improvement following treatment with ganciclovir (without treatment for rejection) has been described [26,36]. Moreover, it has been reported that ganciclovir treatment (without specific antirejection therapy) improved renal function in renal transplant recipients with asymptomatic HCMV infection and biopsy-proven late acute rejection [37]. HCMV infection of the kidney may also result in direct endothelial injury, with subsequent HCMV glomerulopathy [36,38]. A prompt decline in serum creatinine levels after institution of intravenous ganciclovir (without anti-rejection therapy) has also been described in patients with HCMV glomerulopathy, but without tubulointerstitial changes [38]. In the study of Boyce et al. [39], no correlation was found between acute transplant glomerulopathy and HCMV infection diagnosed by seroconversion and viraemia. However, immunohistochemical staining of the allograft tissue to confirm the presence of HCMV was not performed. Finally, with HCMV infection, a granular pattern of HCMV antigen may be seen in the mesangium [36]. Thus, although rejection before virus infection is the most frequent occurrence, virus infection may certainly precede rejection. However, Humar et al. [40] showed that HCMV disease was a risk-factor for chronic rejection only in the presence of acute rejection. IMPACT OF HCMV INFECTION AND DISEASE ON LONG-TERM PATIENT SURVIVAL As summarised in Table 2, HCMV disease in the early post-transplant period has a detrimental effect on long-term recipient survival in kidney transplant recipients, and Sagedal et al. [41] demonstrated, for the first time, that asymptomatic HCMV infection in the early post-transplant period was an independent risk-factor for overall long-term mortality. The mechanisms for the increase in overall long-term mortality caused by HCMV are not fully understood. Nephrosclerosis, diabetic nephropathy and amyloidosis causing renal insufficiency are known to be associated with an increased risk of death after transplantation [42 44]. However, the relative prevalences of these diagnoses are the same among patients who develop HCMV infection and those who do not (personal unpublished data). Table 2. Clinical trials studying the impact of human cytomegalovirus (HCMV) infection and disease on long-term patient survival in kidney transplant recipients Study design Number of patients Results Reference Retrospective; register study Hospitalisation for HCMV disease was associated independently with decreased all-cause patient survival Prospective 471 Both asymptomatic HCMV infection and HCMV disease during the first 100 days after transplantation were associated significantly with increased overall patient mortality beyond 100 days in multivariate analysis Prospective 276 Overt HCMV disease was associated significantly with decreased patient survival in univariate analysis Prospective 85 HCMV disease during the first 4 months after transplantation was associated significantly with subsequent increased patient mortality in multivariate analysis [14] [41] [94] [95]

5 522 Clinical Microbiology and Infection, Volume 11 Number 7, July 2005 It has been postulated previously that overt HCMV disease may cause post-transplant diabetes mellitus through both impaired b-cell function and insulin action [30,45,46]. Moreover, asymptomatic HCMV infection is associated significantly with post-transplant diabetes mellitus [47]. This raises the question of whether the occurrence of diabetes mellitus contributes to the increased mortality found in HCMV-infected renal transplant patients. Unfortunately, the oral glucose tolerance test was not used routinely during the above-mentioned study (October 1994 to July 1997) [41], and such a hypothesis remains speculative. However, HCMV viraemia has been associated with death in paediatric lung transplant recipients [48], and HCMV disease is a significant risk-factor for patient death following liver transplantation [49]. The association between HCMV and cardiovascular risk In heart transplant patients, HCMV infection has been associated with accelerated coronary artery atherosclerosis [50,51]. Also, in the native heart, an association between HCMV antibodies and coronary restenosis after atherectomy has been described in many studies, although no association has been found in others [8,29,31,52 55]. Similarly, an association between HCMV seropositivity and native atherosclerosis has been found in some studies, but not in others [56,57]. Different study designs may partly explain these differing results. Interestingly, HCMV seropositivity in relatively young asymptomatic individuals was found to be associated significantly with endothelial dysfunction, but the association with coronary artery calcification was not significant [58]. Thus, the question of whether there is an association between HCMV and native atherosclerosis is still unresolved. IMPACT OF HCMV INFECTION AND DISEASE ON LONG-TERM KIDNEY GRAFT SURVIVAL Clinical trials studying the association between early HCMV infection and long-term kidney graft survival are summarised in Table 3. Only a few trials have investigated HCMV disease as a riskfactor for death-censored graft loss (death with functioning graft is excluded) [40,41,59]. Sagedal et al. [41] were unable to demonstrate any detrimental effect of early asymptomatic HCMV infection or disease on death-censored graft loss. However, Humar et al. [40] found that HCMV disease was a significant risk-factor for deathcensored graft loss, but only in the presence of acute graft rejection. In a retrospective study, Nett et al. [59] found that HCMV disease was Study design Number of patients Results Reference Unknown 1339 HCMV disease increased the risk of death-censored [40] graft loss, but only in the presence of acute rejection Prospective 471 Asymptomatic HCMV infection and HCMV disease [41] were both associated significantly with uncensored, but not censored graft loss Unknown 445 HCMV disease was an independent predictor of [25] uncensored graft loss in patients treated with azathioprine, but not in those treated with mycophenolate mofetil Retrospective 470 HCMV disease had a detrimental effect on uncensored [96] kidney graft survival, but only in recipients with zero HLA-DR matches Prospective 276 Overt HCMV disease was associated significantly with [94] decreased graft survival in univariate analysis. Graft survival was not defined (censored vs. uncensored) Prospective 85 HCMV disease during the first 4 months after [95] transplantation had no detrimental effect on graft survival. Graft survival was not defined (censored vs. uncensored) Prospective 84 Uncensored graft survival was the same in the HCMV [93] risk group (D+ and or R+) during the first 4 months after transplantation and at 5-year follow-up as in the D R risk group Retrospective 2740 kidney and 606 pancreas kidney transplant recipients HCMV disease was associated significantly with an increased censored graft loss, but whether HCMV disease was treated with a reduction in immunosuppressive treatment was not stated [59] Table 3. Clinical trials studying the impact of human cytomegalovirus (HCMV) infection and disease on long-term kidney graft survival HLA-DR, tissue type donor recipient.

6 Sagedal et al. Human cytomegalovirus in renal transplant recipients 523 associated significantly with censored graft loss, but it was not clear whether HCMV disease was treated by a reduction in immunosuppressive treatment, which might possibly explain the increase in graft loss. Several studies have investigated uncensored graft failure (death with functioning graft is included). As shown in Table 3, most studies report a significant association between HCMV disease and uncensored graft loss. Unfortunately, some studies fail to define graft loss. It is obviously important to define whether graft loss is censored or uncensored to make comparisons meaningful. ASSOCIATION BETWEEN HCMV AND OTHER INFECTIOUS DISEASES HCMV is known to predispose to opportunistic superinfection with a range of different microorganisms, including Pneumocyctis jiroveci (carinii), a variety of fungi and Listeria monocytogenes, with a marked decrease in CD4-positive cells and an increase in CD8-positive cells in HCMV-infected organ transplant recipients implying the highest risk of opportunistic superinfection [60]. Another possible mechanism of superinfection could involve the disruption of mucosal surfaces by HCMV, thereby increasing the risk of superinfection [5]. Interestingly, valacyclovir prophylaxis for 100 days (vs. placebo) resulted in a significantly reduced incidence of non-herpes virus infections in seronegative recipients of kidneys from seropositive donors [31]. Infections with Candida were reduced significantly in the serostatus group D+ R receiving valacyclovir, and the reduction in staphylococcal infections in the same patient group nearly reached statistical significance [31]. The association between HCMV and other herpes viruses Human herpes virus 6 (HHV-6) is detected during the first 4 weeks after solid-organ or bone marrow transplantation [61,62]. Based on serology, HHV-6 reactivation after kidney transplantation was associated significantly with primary HCMV infection and HCMV syndrome [61]. However, serological cross-reactions between HHV-6 and HHV-7 exist, and serology alone is unable to discriminate between the roles of these two herpes viruses [63]. Based on detection of HCMV, HHV-6 and HHV-7 DNA in peripheral blood leukocytes, HHV-7, but not HHV-6, was associated significantly with HCMV disease [63]. However, the small number of patients in the study with detectable HHV-6 DNA (n = 7) may explain why no association was found between HHV-6 and HCMV disease. Nevertheless, the result is in accordance with another study of renal transplant recipients in which a significant association between HCMV disease and detection of HHV-7 DNA was found, while that with HHV-6 DNA was inconclusive [64]. During the first 120 days following kidney transplantation, HCMV, HHV-7 and HHV-6 DNaemia were present in 58%, 46% and 23% of patients, respectively [65]. Moreover, the HCMV virus load was significantly higher than that of either HHV-7 or HHV-6, and there was a significant increase in HCMV disease in patients with HCMV and HHV-7 co-infection. In liver transplant recipients, there was a clustering of high HHV-6 and HHV-7 virus loads in the HCMV D+ R serostatus subgroup, indicating that HHV-6 and HHV-7 replication depend on the level of HCMV replication [66]. Moreover, intravenous ganciclovir reduced the load of both HCMV, HHV-6 and HHV-7, confirming the effect of ganciclovir on HHV-6 and HHV-7. In addition, there have been several reports of an association between HCMV and Epstein Barr virus (EBV). In a study of 208 renal transplant recipients followed for 3 years, HCMV and EBV replication often overlapped in time, but HCMV replication mostly preceded EBV replication, indicating that HCMV replication may induce activation of latent EBV infection [67]. EBV infection is associated with development of posttransplant lymphoproliferative disease (PTLD), which occurs in c. 2% of all transplant patients. In pre-transplant EBV-seronegative liver transplant recipients, HCMV disease was the only factor associated significantly with the development of PTLD in multivariate analysis [68]. However, in a large study of renal transplant patients comprising both EBV-seropositive and -seronegative individuals, an EBV-seronegative status at transplant was the only factor associated significantly with development of PTLD, and HCMV disease was not a risk-factor for PTLD in univariate or multivariate analyses [69]. Malouf et al. [70] found that anti-herpes virus prophylaxis in EBV-seronegative lung transplant recipients reduced the incidence of PTLD significantly.

7 524 Clinical Microbiology and Infection, Volume 11 Number 7, July 2005 PREVENTION OF HCMV DISEASE Pre-emptive anti-hcmv therapy With a pre-emptive anti-hcmv therapeutic regimen, an antiviral agent is introduced at the first sign of HCMV antigenaemia, positive HCMV PCR, or positive HCMV viraemia. Frequent monitoring of HCMV in blood is thus an important part of such a regimen. One possible advantage with this therapeutic regimen, compared to prophylaxis, is that fewer patients are exposed to antiviral agents, resulting in less drug resistance. Clinical trials investigating the effect of preemptive anti-hcmv therapy in kidney transplant recipients are summarised in Table 4. There are fewer studies investigating pre-emptive, as opposed to prophylactic, anti-hcmv therapy in renal transplant recipients. Moreover, the HCMV diagnostic techniques and therapeutic regimens differ among various trials, making comparisons somewhat difficult. Some studies on pre-emptive anti-hcmv therapy in liver transplant patients are therefore also discussed to elucidate clinical aspects associated with this therapeutic strategy. In the study by Singh et al. [71] of 47 liver transplant patients, intravenous ganciclovir was given to the experimental group for 7 days following a positive HCMV culture, while the control group received high-dose acyclovir during the first 24 weeks post-transplantation. HCMV disease was reduced significantly in the ganciclovir-treated group, and it was concluded that oral acyclovir was ineffective as prophylaxis against HCMV disease in liver transplant patients. Oral ganciclovir given pre-emptively to kidney transplant patients upon detection of HCMV pp65 antigenaemia during the first 12 weeks post-transplantation (minimum treatment duration of 4 weeks) is also highly effective in preventing HCMV disease [72]. Furthermore, the incidence of late occurrence of HCMV disease was the same in the treatment and control groups, indicating that HCMV disease is prevented and not simply delayed by early pre-emptive therapy. In a prospective, randomised study of 60 liver transplant patients [73], oral ganciclovir was given pre-emptively for 14 days following detection of HCMV pp65 antigenaemia. There was no difference in the incidence of HCMV disease between the treatment group and the control group, although the short treatment period of 14 days may have contributed to the lack of efficacy observed for pre-emptive treatment. In contrast, oral ganciclovir given pre-emptively for a median duration of 49 days in kidney transplant recipients was highly effective in reducing HCMV disease [72]. However, in the study of Rayes et al. [73], 28 (32%) of 88 patients were not randomised because of HCMV disease preceding antigenaemia. This may be explained by the fact that a positive test was defined as at least one HCMV pp65-positive cell leukocytes (compared to at least one HCMV pp65-positive cell leukocytes [72]). This higher cut-off level may explain why HCMV disease preceded HCMV antigenaemia more often in that study (7.5% of all observed patients vs. 1.8% of all observed patients [72,73]). These observations demonstrate that frequent Study design Number of patients Results Reference Retrospective, uncontrolled Prospective, randomised, comparative Prospective, randomised, comparative Prospective, non-randomised, uncontrolled Prospective, randomised, placebo-controlled with no HCMV infection Ganciclovir given for 14 days following detection of positive viraemia in 87 patients resulted in the same uncensored graft loss and patient mortality after follow-up for 5 years as in 82 patients with no HCMV infection 83 Oral ganciclovir given following detection of HCMV pp65 antigenaemia during the first 12 weeks post-transplantation (minimum treatment duration of 4 weeks) is highly effective in preventing HCMV disease 36 Intravenous ganciclovir given for 21 days following detection of HCMV viraemia (15 patients) decreased HCMV disease in the pre-emptively treated group, compared to deferred therapy (21 patients), but the difference was not significant 35 Intravenous ganciclovir was given for 2 3 weeks following detection of HCMV antigenaemia, and no patients developed HCMV disease Ten kidney and two liver transplant recipients Intravenous ganciclovir given for 2 weeks (n = 5) following detection of HCMV antigenaemia reduced HCMV disease significantly [74] [72] [97] [98] [99] Table 4. Clinical trials investigating the effect of pre-emptive anti-human cytomegalovirus (HCMV) therapy in kidney transplant recipients

8 Sagedal et al. Human cytomegalovirus in renal transplant recipients 525 monitoring for HCMV infection is mandatory with a pre-emptive therapeutic regimen. The study of Akposso et al. [74] (Table 4) compared a pre-emptive treatment group of patients with a control group who were not suffering from HCMV infection, so the conclusion that ganciclovir given pre-emptively may reduce HCMV-induced graft loss may be based on false assumptions. Risk adaptive strategies have been used in some studies, in which ganciclovir was administered while the patients also received anti-lymphocyte antibodies for induction therapy or steroid-resistant rejections, irrespective of HCMV viraemia or antigenaemia [75 77]. However, the therapeutic strategy in all these studies involved targeted prophylaxis rather than pre-emptive therapy, and HCMV disease was reduced significantly in the patients treated with ganciclovir. Prophylactic anti-hcmv therapy With a prophylactic anti-hcmv therapeutic regimen, the antiviral agent is introduced to a large patient population before any signs of active HCMV infection, and frequent HCMV monitoring is not mandatory. Antiviral agents used prophylactically against HCMV are acyclovir valacyclovir and ganciclovir valganciclovir. An HCMV protein kinase (a UL97 gene product) phosphorylates ganciclovir to ganciclovir monophosphate, which is converted subsequently by cellular enzymes to ganciclovir triphosphate, the biologically effective compound [78]. Interestingly, acyclovir is phosphorylated far less efficiently than is ganciclovir by the HCMV UL97 enzyme [79]. Thus, acyclovir and valacyclovir are ineffective in inhibiting HCMV replication, and therefore have no place in pre-emptive anti- HCMV therapy or in the treatment of HCMV disease. However, given prophylactically, they reduce HCMV disease significantly in kidney transplant recipients. Selected clinical trials investigating the effect of prophylactic anti-hcmv therapy in kidney transplant recipients are summarised in Table 5. Prophylactic oral valacyclovir yields increased acyclovir exposure compared to oral acyclovir, and reduces HCMV disease effectively [31]. Oral acyclovir has also been shown to be effective in reducing HCMV disease in some studies, but not in others (Table 5). In contrast, HCMV prophylaxis with oral ganciclovir is reported to be superior to oral acyclovir in reducing HCMV disease [13,80]. In one of these studies, the dose of oral acyclovir was only 400 mg three-times-daily, while high-dose oral acyclovir was given (up to 3200 mg day) in the other study. Valganciclovir has now increasingly replaced oral ganciclovir because of its higher bioavailability, and has been shown to be as efficient as oral ganciclovir in reducing HCMV disease in solid-organ recipients [81]. However, only onethird of the patients were kidney transplant recipients, and the study was not designed or powered to show an overall reduction in the incidence of HCMV disease with valganciclovir compared with oral ganciclovir. Interestingly, a meta-analysis of 12 randomised trials showed that acyclovir (including valacyclovir) prophylaxis in solid-organ transplant patients was associated significantly with reductions in HCMV disease and patient mortality [82]. Prophylaxis resulted in Table 5. Selected clinical trials investigating the effect of prophylactic anti-human cytomegalovirus (HCMV) therapy in kidney transplant recipients Study design Number of patients Results Reference Prospective, randomised, placebo-controlled Prospective, randomised, placebo-controlled Prospective, randomised not placebo-controlled Prospective, randomised, not placebo-controlled Prospective, randomised, comparative Prospective, randomised, not placebo-controlled 616: 135 valacyclovir; 120 placebo Valacyclovir prophylaxis (2000 mg four-times-daily) for 90 days post-transplantation delayed and reduced HCMV disease significantly compared to deferred therapy 104: 53 acyclovir; 51 placebo Prophylaxis with oral acyclovir mg day for 12 weeks post-transplantation reduced HCMV disease significantly during the first year 44: 22 ganciclovir; 22 controls Oral ganciclovir (750 mg twice-daily) for 3 months reduced HCMV infection significantly within the first 9 months after transplantation compared to no prophylaxis 32 (only D+ R ):22 acyclovir; 10 controls No beneficial effect of oral acyclovir ( mg day) for the first 12 weeks compared with no prophylaxis 42:19 ganciclovir; 23 acyclovir Oral ganciclovir (1000 mg three-times-daily) for the first 12 weeks reduced HCMV disease significantly compared to oral acyclovir (200 mg twice-daily) 32 (only D+ R ):17 ganciclovir; 15 controls Intravenous ganciclovir from day 14 to day 28 after transplantation delayed the time to HCMV infection significantly, but did not decrease HCMV disease [31] [100] [101] [102] [103] [104]

9 526 Clinical Microbiology and Infection, Volume 11 Number 7, July 2005 a 40% reduction in overall mortality. Moreover, prophylaxis was associated with a significant reduction in opportunistic infections and acute graft rejection. During prophylactic treatment, valacyclovir was superior to high-dose acyclovir in reducing HCMV disease. In another meta-analysis of 13 studies, anti- HCMV prophylaxis was associated significantly with reduced HCMV disease [83]. In contrast to the meta-analysis described above, most of the studies analysed used ganciclovir prophylaxis. Prophylactic treatment reduced graft loss and patient mortality by 22% and 20%, respectively, but statistical significance was not achieved. There was no effect of prophylactic treatment on the incidence of episodes of acute rejection. Snydman et al. [84] found that intravenous HCMV immunoglobulin reduced HCMV disease significantly in seronegative recipients with seropositive donors. However, with the antiviral agents available today, the role of intravenous immunoglobulin is reduced. Seronegative recipients with seropositive donors have been a matter of concern. In a study of 41 renal transplant recipients of the serostatus D+ R group, only 40% of those who developed HCMV disease were HCMV PCR-positive before the disease [85]. Based on the assumptions made, the authors indicated that the deferred strategy and the pre-emptive strategy are slightly less costly than the prophylactic strategy. However, according to European best-practice guidelines, HCMV prophylaxis for kidney transplant recipients in the serostatus D+ R group is recommended [86]. Interestingly, in a retrospective register study by Opelz et al. [32], HCMV prophylaxis resulted in significantly increased censored and uncensored graft survival in seronegative recipients with seropositive donors (D+ R ). Moreover, HCMV prophylaxis resulted in significantly better patient survival in the serostatus D+ R group. However, the antiviral agents used and the duration of HCMV prophylaxis were not stated. Further prospective, randomised studies are needed to fully evaluate the effect of HCMV prophylaxis on acute rejection, graft loss and long-term patient mortality. Ganciclovir resistance A potential drawback with prolonged use of anti- HCMV agents is increased virus drug resistance. Limaye et al. [87] reported that 7% of 240 solidorgan recipients who received oral ganciclovir prophylactically for the first 100 days had ganciclovir-resistant HCMV disease, with all cases occurring in seronegative recipients with seropositive donors. Ganciclovir resistance after valacyclovir prophylaxis in a renal transplant recipient has also been reported [88]. In another study [89], 301 seronegative solid-organ recipients with seropositive donors received either valganciclovir or oral ganciclovir prophylaxis for days after transplantation. The incidence of HCMV UL97 mutations was 1.9% at the end of prophylaxis, and 6.1% for patients with suspected HCMV disease up to 1 year after transplantation in the oral ganciclovir group. No resistance mutations were detected in valganciclovir-treated patients. However, in AIDS patients with HCMV retinitis who received long-term valganciclovir therapy, the cumulative incidence of patients carrying viruses with UL97 mutations was 2.2%, 6.5%, 12.8% and 15.3% after 3, 6, 12 and 18 months of treatment, respectively [90]. Similarly, a high incidence (9%) of ganciclovir-resistant HCMV infection was found in lung transplant recipients given intravenous ganciclovir preemptively [91]. CONCLUSIONS According to many authors, anti-hcmv prophylaxis may be the preferred therapeutic strategy for seronegative recipients with seropositive donors. Whether prophylaxis or pre-emptive therapy should be the preferred strategy for seropositive recipients is still a matter of debate. Some transplantation centres keep the patients for only a few weeks after transplantation, and therefore do not favour pre-emptive therapy because of infrequent HCMV monitoring. Both oral ganciclovir and valacyclovir are superior to oral acyclovir as anti- HCMV prophylaxis. Valganciclovir is at least as effective as oral ganciclovir in reducing HCMV disease, and seems to be associated with far fewer resistance mutations. In seronegative recipients with seropositive donors, there is some evidence that anti-hcmv prophylaxis may reduce acute kidney graft rejection and increase patient and censored uncensored graft survival. On the other hand, the data in the literature on the association between pre-emptive anti-hcmv therapy and patient and graft survival are scarce.

10 Sagedal et al. Human cytomegalovirus in renal transplant recipients 527 Different designs and statistical tests used in the various studies may to some degree account for the divergent results. REFERENCES 1. Yang CW, Kim YO, Kim YS et al. Clinical course of cytomegalovirus (CMV) viremia with and without ganciclovir treatment in CMV-seropositive kidney transplant recipients. Longitudinal follow-up of CMV pp65 antigenemia assay. Am J Nephrol 1998; 18: Toupance O, Bouedjoro-Camus MC, Carquin J et al. Cytomegalovirus-related disease and risk of acute rejection in renal transplant recipients: a cohort study with case-control analyses. Transplant Int 2000; 13: Hokeberg I, Eriksson BM, Zweygberg-Wirgart B, Tufvesson G, Olding-Stenkvist E, Grillner L. Diagnostic markers and risk factors of cytomegalovirus infection and disease in renal allograft recipients. Scand J Infect Dis 1995; 27: Sagedal S, Nordal KP, Hartmann A et al. A prospective study of the natural course of cytomegalovirus infection and disease in renal allograft recipients. Transplantation 2000; 70: Ljungman P, Griffiths P, Paya C. Definitions of cytomegalovirus infection and disease in transplant recipients. Clin Infect Dis 2002; 34: Degre M, Kristiansen KI, Rollag H, Holter E, Nordal KP. Detection of human cytomegalovirus (HCMV) pp67- mrna and pp65 antigenemia in relation to development of clinical HCMV disease in renal transplant recipients. Clin Microbiol Infect 2001; 7: Lo CY, Ho KN, Yuen KY et al. Diagnosing cytomegalovirus disease in CMV seropositive renal allograft recipients: a comparison between the detection of CMV DNAemia by polymerase chain reaction and antigenemia by CMV pp65 assay. Clin Transplant 1997; 11: Hassan-Walker AF, Kidd IM, Sabin C, Sweny P, Griffiths PD, Emery VC. Quantity of human cytomegalovirus (CMV) DNAemia as a risk factor for CMV disease in renal allograft recipients: relationship with donor recipient CMV serostatus, receipt of augmented methylprednisolone and antithymocyte globulin (ATG). J Med Virol 1999; 58: Kim CK, Song JH, Kim SM et al. Clinical usefulness of human cytomegalovirus antigenemia assay after kidney transplantation. Transplantation 2003; 75: Humar A, Paya C, Pescovitz MD et al. Clinical utility of cytomegalovirus viral load testing for predicting CMV disease in D+ R solid organ transplant recipients. Am J Transplant 2004; 4: Humar A, Kumar D, Boivin G, Caliendo AM. Cytomegalovirus (CMV) virus load kinetics to predict recurrent disease in solid-organ transplant patients with CMV disease. J Infect Dis 2002; 186: Sarmiento JM, Munn SR, Paya CV, Velosa JA, Nguyen JH. Is cytomegalovirus infection related to mycophenolate mofetil after kidney transplantation? A case-control study. Clin Transplant 1998; 12: Rubin RH, Kemmerly SA, Conti D et al. Prevention of primary cytomegalovirus disease in organ transplant recipients with oral ganciclovir or oral acyclovir prophylaxis. Transplant Infect Dis 2000; 2: Abbott KC, Hypolite IO, Viola R et al. Hospitalizations for cytomegalovirus disease after renal transplantation in the United States. Ann Epidemiol 2002; 12: Hibberd PL, Tolkoff-Rubin NE, Cosimi AB et al. Symptomatic cytomegalovirus disease in the cytomegalovirus antibody seropositive renal transplant recipient treated with OKT3. Transplantation 1992; 53: Yeung S, Tong KL, Chan HW. Impact of OKT3 and polyclonal antibodies on symptomatic CMV infection in renal transplant recipients. Transplant Proc 2000; 32: Kusne S, Shapiro R, Fung J. Prevention and treatment of cytomegalovirus infection in organ transplant recipients. Transplant Infect Dis 1999; 1: Emery VC, Sabin CA, Cope AV, Gor D, Hassan-Walker AF, Griffiths PD. Application of viral-load kinetics to identify patients who develop cytomegalovirus disease after transplantation. Lancet 2000; 355: Rollag H, Sagedal S, Kristiansen KI et al. Cytomegalovirus DNA concentration in plasma predicts development of cytomegalovirus disease in kidney transplant recipients. Clin Microbiol Infect 2002; 8: Sarmiento JM, Dockrell DH, Schwab TR, Munn SR, Paya CV. Mycophenolate mofetil increases cytomegalovirus invasive organ disease in renal transplant patients. Clin Transplant 2000; 14: ter Meulen CG, Wetzels JF, Hilbrands LB. The influence of mycophenolate mofetil on the incidence and severity of primary cytomegalovirus infections and disease after renal transplantation. Nephrol Dial Transplant 2000; 15: Sollinger HW. Mycophenolate mofetil for the prevention of acute rejection in primary cadaveric renal allograft recipients. US Renal Transplant Mycophenolate Mofetil Study Group. Transplantation 1995; 60: European Mycophenolate Mofetil Cooperative Study Group. Placebo-controlled study of mycophenolate mofetil combined with cyclosporin and corticosteroids for prevention of acute rejection. Lancet 1995; 345: The Tricontinental Mycophenolate Mofetil Renal Transplantation Study Group. A blinded, randomized clinical trial of mycophenolate mofetil for the prevention of acute rejection in cadaveric renal transplantation. Transplantation 1996; 61: Giral M, Nguyen JM, Daguin P et al. Mycophenolate mofetil does not modify the incidence of cytomegalovirus (CMV) disease after kidney transplantation but prevents CMV-induced chronic graft dysfunction. J Am Soc Nephrol 2001; 12: Kashyap R, Shapiro R, Jordan M, Randhawa PS. The clinical significance of cytomegaloviral inclusions in the allograft kidney. Transplantation 1999; 67: Neyts J, Andrei G, De Clercq E. The novel immunosuppressive agent mycophenolate mofetil markedly potentiates the antiherpesvirus activities of acyclovir, ganciclovir, and penciclovir in vitro and in vivo. Antimicrob Agents Chemother 1998; 42: Sagedal S, Nordal KP, Hartmann A et al. The impact of cytomegalovirus infection and disease on rejection episodes in renal allograft recipients. Am J Transplant 2002; 2:

11 528 Clinical Microbiology and Infection, Volume 11 Number 7, July Pouteil-Noble C, Ecochard R, Landrivon G et al. Cytomegalovirus infection an etiological factor for rejection? A prospective study in 242 renal transplant patients. Transplantation 1993; 55: von Willebrand E, Pettersson E, Ahonen J, Hayry P. CMV infection, class II antigen expression, and human kidney allograft rejection. Transplantation 1986; 42: Lowance D, Neumayer HH, Legendre CM et al. Valacyclovir for the prevention of cytomegalovirus disease after renal transplantation. International Valacyclovir Cytomegalovirus Prophylaxis Transplantation Study Group. N Engl J Med 1999; 340: Opelz G, Dohler B, Ruhenstroth A. Cytomegalovirus prophylaxis and graft outcome in solid organ transplantation: a collaborative transplant study report. Am J Transplant 2004; 4: Payton D, Thorner P, Eddy A, Yeger H, Baumal R. Demonstration by light microscopy of cytomegalovirus on a renal biopsy of a renal allograft recipient: confirmation by immunohistochemistry and in situ hybridization. Nephron 1987; 47: Cameron J, Rigby RJ, van Deth AG, Petrie JJ. Severe tubulo-interstitial disease in a renal allograft due to cytomegalovirus infection. Clin Nephrol 1982; 18: Liapis H, Storch GA, Hill DA, Rueda J, Brennan DC. CMV infection of the renal allograft is much more common than the pathology indicates: a retrospective analysis of qualitative and quantitative buffy coat CMV-PCR, renal biopsy pathology and tissue CMV-PCR. Nephrol Dial Transplant 2003; 18: Shaver MJ, Bonsib SM, Abul-Ezz S, Barri YM. Renal allograft dysfunction associated with cytomegalovirus infection. Am J Kidney Dis 1999; 34: Reinke P, Fietze E, Ode-Hakim S et al. Late-acute renal allograft rejection and symptomless cytomegalovirus infection. Lancet 1994; 344: Onuigbo M, Haririan A, Ramos E, Klassen D, Wali R, Drachenberg C. Cytomegalovirus-induced glomerular vasculopathy in renal allografts: a report of two cases. Am J Transplant 2002; 2: Boyce NW, Hayes K, Gee D et al. Cytomegalovirus infection complicating renal transplantation and its relationship to acute transplant glomerulopathy. Transplantation 1988; 45: Humar A, Gillingham KJ, Payne WD, Dunn DL, Sutherland DE, Matas AJ. Association between cytomegalovirus disease and chronic rejection in kidney transplant recipients. Transplantation 1999; 68: Sagedal S, Hartmann A, Nordal KP et al. Impact of early cytomegalovirus infection and disease on long-term recipient and kidney graft survival. Kidney Int 2004; 66: Lim EC, Terasaki PI. Outcome of kidney transplantation in different diseases. Clin Transplant 1990; 4: Fernandez-Fresnedo G, Zubimendi JA, Cotorruelo JG et al. Significance of age in the survival of diabetic patients after kidney transplantation. Int Urol Nephrol 2002; 33: Hartmann A, Holdaas H, Fauchald P et al. Fifteen years experience with renal transplantation in systemic amyloidosis. Transplant Int 1992; 5: Hjelmesaeth J, Hartmann A, Kofstad J et al. Glucose intolerance after renal transplantation depends upon prednisolone dose and recipient age. Transplantation 1997; 64: Hjelmesaeth J, Jenssen T, Hagen M, Egeland T, Hartmann A. Determinants of insulin secretion after renal transplantation. Metabolism 2003; 52: Hjelmesaeth J, Sagedal S, Hartmann A et al. Asymptomatic cytomegalovirus infection is associated with increased risk of new-onset diabetes mellitus and impaired insulin release after renal transplantation. Diabetologia 2004; 47: Danziger-Isakov LA, DelaMorena M, Hayashi RJ et al. Cytomegalovirus viremia associated with death or retransplantation in pediatric lung-transplant recipients. Transplantation 2003; 75: de Otero J, Gavalda J, Murio E et al. Cytomegalovirus disease as a risk factor for graft loss and death after orthotopic liver transplantation. Clin Infect Dis 1998; 26: Grattan MT, Moreno-Cabral CE, Starnes VA, Oyer PE, Stinson EB, Shumway NE. Cytomegalovirus infection is associated with cardiac allograft rejection and atherosclerosis. JAMA 1989; 261: McDonald K, Rector TS, Braulin EA, Kubo SH, Olivari MT. Association of coronary artery disease in cardiac transplant recipients with cytomegalovirus infection. Am J Cardiol 1989; 64: Zhou YF, Leon MB, Waclawiw MA et al. Association between prior cytomegalovirus infection and the risk of restenosis after coronary atherectomy. N Engl J Med 1996; 335: Blum A, Giladi M, Weinberg M et al. High anti-cytomegalovirus (CMV) IgG antibody titer is associated with coronary artery disease and may predict post-coronary balloon angioplasty restenosis. Am J Cardiol 1998; 81: Neumann FJ, Kastrati A, Miethke T et al. Previous cytomegalovirus infection and restenosis after coronary stent placement. Circulation 2001; 104: Hernandez D, Hanson E, Kasiske MK, Danielson B, Roel J, Kasiske BL. Cytomegalovirus disease is not a major risk factor for ischemic heart disease after renal transplantation. Transplantation 2001; 72: Espinola-Klein C, Rupprecht HJ, Blankenberg S et al. Impact of infectious burden on progression of carotid atherosclerosis. Stroke 2002; 33: Ridker PM, Hennekens CH, Stampfer MJ, Wang F. Prospective study of herpes simplex virus, cytomegalovirus, and the risk of future myocardial infarction and stroke. Circulation 1998; 98: Grahame-Clarke C, Chan NN, Andrew D et al. Human cytomegalovirus seropositivity is associated with impaired vascular function. Circulation 2003; 108: Nett PC, Heisey DM, Fernandez LA, Sollinger HW, Pirsch JD. Association of cytomegalovirus disease and acute rejection with graft loss in kidney transplantation. Transplantation 2004; 78: Rubin RH. Impact of cytomegalovirus infection on organ transplant recipients. Rev Infect Dis 1990; 12(suppl): DesJardin JA, Gibbons L, Cho E et al. Human herpesvirus 6 reactivation is associated with cytomegalovirus infection and syndromes in kidney transplant recipients at risk