CYTOMEGALOVIRUS INFECTION AND GRAFT REJECTION IN RENAL TRANSPLANTATION

Abstract

Background. 

Cytomegalovirus (CMV) infection and CMV disease have been associated with acute and chronic graft rejection. The introduction of the sensitive CMV antigenemia pp65 assay for detection of CMV infection allowed us to study the time course of CMV infection and acute rejection and the long-term outcome in renal transplant recipients with and without a CMV risk constellation.

Methods. 

Prospective single center study including 48 renal transplant recipients at risk for CMV infection (donor and/or recipient CMV seropositive) and a control group of 36 CMV seronegative recipients of CMV seronegative kidney donors. Evidence of CMV infection was monitored by the CMV antigenemia pp65 assay every 1 to 2 weeks and compared with the occurrence of acute rejection in the posttransplant period and graft function at 5 years.

Results. 

CMV infection developed in 83% (40/48) of patients of the CMV risk group within 4 months posttransplant. A total of 18 of patients experienced an acute rejection episode (control group 16/36;P =0.65). In 12/18 CMV infection followed rejection and in three patients antigenemia preceded the diagnosis of rejection. In three patients CMV antigenemia remained negative. Five-year follow up: Patient survival (44/48 vs. 31/36;P =0.48), graft survival (38/48 vs. 27/36;P =0.79), number of patients with at least one acute rejection episode: CMV risk group: 42.1%, control group 51% (P =0.46), serum creatinine: CMV risk group:130±66 μmol/liter, control group: 126±37 μmol/liter (P =0.56), proteinuria: CMV risk group: 0.02±0.02 g/mmol creatinine, control group: 0.02±0.02 g/mmol creatinine (P =1.0).

Conclusion. 

CMV infection within 4 months posttransplant, as defined by a positive antigenemia assay was not found to be a risk factor for acute graft rejection or chronic graft dysfunction at 5 years.

INTRODUCTION

The impact of cytomegalovirus (CMV) infection on acute rejection and on long-term outcome after transplantation remains controversial. Studies performed in animals suggested a role in graft rejection ⁽¹⁾. Although some clinical studies indicate an association of CMV infection with acute rejection episodes ^(2–7) or chronic graft loss ^{(8, 9)}, others failed to show this correlation ^{(10, 11)}.

The aim of this study was to evaluate, if the onset of a CMV infection precedes the time of a first acute rejection episode, and may therefore induce it, or if it follows the rejection episode, as a result of an enhanced immunosuppression.

PATIENTS AND METHODS

Study design.

Prospective single center study. Primary endpoint: incidence of a first acute rejection episode in time correlation to the onset of CMV infection, as determined by a positive CMV antigenemia. Secondary endpoints: 5-year patient and graft survival and renal function at 60 months.

Study population.

From March 1993 to September 1994 all patients receiving a renal allograft were prospectively studied. Patients were stratified into two groups according to the serological CMV status at the time of transplantation. CMV risk group: recipient and/or donor with a positive CMV serology at the time of transplantation. Control group: CMV seronegative patients receiving a CMV seronegative graft.

Immunosuppression.

All patients received induction therapy with antithymocyte globulin (ATG) (ATG Fresenius) for 7 days. Primary immunosuppressive therapy consisted of cyclosporine, azathioprine, and prednisone.

CMV antigenemia.

CMV antigenemia, using the pp65 assay was measured in the CMV risk group every 1 to 2 weeks within the first 4 months and quantified as number of infected leukocytes per 250,000 cells ^{(12, 13)}.

Definition of CMV Infection and CMV disease.

CMV infection was considered to be present if CMV antigens were detected in blood leukocytes, as quantified by the number of infected cells per 250,000 leukocytes, in the absence of clinical or laboratory signs of a symptomatic CMV infection. CMV disease: the diagnosis of CMV disease required CMV detection combined with clinical or laboratory signs including fever, leukopenia, or organ involvement (hepatitis, pneumonitis, colitis, gastritis, chorioretinitis).

Prophylactic antiviral therapy.

No prophylactic therapy against CMV, Pneumocystis carinii, or other viral or bacterial germs was used with the exception of antibiotic prophylaxis for the first 48 hr after transplantation. Patients at risk for CMV infection (donor and/or recipient CMV seropositive) who received antilymphocyte globulins due to acute rejection were treated prophylactic with ganciclovir i.v. every other day, totally nine doses. Patients with CMV disease were treated with ganciclovir i.v. for 14 consecutive days.

Diagnosis and grading of acute rejection.

Criteria for the clinical diagnosis of acute rejection were used as previously reported ⁽¹⁴⁾. Acute rejection was treated with 0.5 g methylprednisolone given i.v. on 3 consecutive days. The plasma creatinine peak had to fall ≥25% within 3 days after the initiation of empiric anti-rejection treatment. Otherwise a renal biopsy was performed.

Biopsy data.

All biopsies were analyzed according to the Banff criteria ⁽¹⁵⁾ by light microscopy and immunofluorescence staining technique. Acute rejection episodes were histologically divided into three types: interstitial rejection (moderate: type IA, severe: type IB), vascular rejection (i.e., transplant endarteritis without necrosis; type II), and vascular rejection with necrosis and/or hemorrhage and/or thrombosis (type III).

Biopsy-proven type IA interstitial acute rejection was treated with another three steroid pulses, whereas type IB interstitial acute rejection or vascular (type II or III) rejection were treated with either OKT-3 or ATG for 7 days.

Statistical analysis.

Statistical analysis was done either by the Mann-Whitney U test or Fisher’s exact test.

RESULTS

Eighty-four patients underwent transplantation during the study period. A total of 48 patients were stratified to the CMV risk group, and 36 patients to the control group, respectively. There was no significant difference regarding the type and the number of transplants. The mean age of patients of the control group was 51 years (range 24–70 years) and significantly higher than the mean age of the CMV risk group (46 years; range 18–71 years;P =0.05). Donor and recipient were both serologically CMV positive in 22 cases in the CMV risk group (46%), in 16 cases the donor was CMV positive and the recipient CMV negative (33%), and in 10 cases a CMV positive recipient received a CMV negative graft (21%). Results at 4 months indicated that there were no significant differences observed between the two groups concerning graft survival, patient survival, number of patients with a first clinically or biopsy proven rejection, time of diagnosis of first rejection, and the number of biopsy proven rejection episodes (Table 1). One patient in the CMV risk group died due to posttransplant non-Hodgkin’s lymphoma. Two other grafts in this group were lost due to a hemolytic uremic syndrome and a transplant artery stenosis. Three grafts in the control group were lost. Two due to therapy resistant rejection, and one due to an acute tubular necrosis.

Table 1: Patient and graft survival, and number of rejection episodes within the first 4 months after transplantation in patients at risk for CMV infection and controls showed no differences

A total of 40 of 48 patients (83%) in the CMV risk group developed a positive CMV antigenemia and experienced either asymptomatic CMV infection (25/40; 62.5%) or CMV disease (15/40; 37.5%). Eight patients (17%) remained CMV antigenemia negative during the first 4 months. A first acute rejection episode occurred in 18 patients in the CMV risk group (14 biopsy proven). Nine patients were treated for type IB interstitial or vascular (type II or III) rejection with either OKT-3 or ATG in combination with steroid pulses, although the other nine received only steroid pulses. A positive CMV antigenemia preceded the first acute rejection episode in three patients. In 12 of these 18 patients a positive CMV antigenemia followed the first episode of acute rejection (see Fig. 1). In three patients with antirejection treatment no CMV antigenemia occurred within the 4 months period. Nine of the 18 patients (50%) of the CMV risk group experienced a second acute rejection episode. Eight patients were already CMV antigenemia positive after their first rejection episode, although one patient remained CMV antigenemia negative after the first and second acute rejection.

Figure 1: Onset of a positive CMV antigenemia and CMV disease in 15 patients with acute rejection and CMV infection in correlation to the time of the diagnosis and treatment of the first acute rejection episode (day 0).

CMV disease developed in 15 (31%) of 48 patients of the CMV risk group. The onset of CMV disease was at a mean of 49 days (range 23–77 days) after transplantation. Six of the 15 patients with CMV disease had also suffered from biopsy proven rejection, but in all six patients the acute rejection episode preceded the onset of CMV disease, as shown in Figure 1. All but one of the patients with CMV disease was treated with ganciclovir as described.

Sixteen serologically CMV negative recipients had received a graft from a serologically positive donor. Patient and graft survival in this high-risk group was 100% within the first 4 months after transplantation. Eleven of the 16 recipients (69%) became CMV positive as diagnosed by a positive CMV antigenemia and 6 of them developed CMV disease. Only 3 of 16 patients (19%) experienced a first acute rejection episode (2 of them biopsy proven).

Acute rejection developed in 16 of 36 (44%) patients of the control group including six biopsy-proven cases. Three of these patients were treated with either OKT-3 or ATG in combination with steroid pulses, although the other 13 patients received at least three steroid pulses.

Results at 5-year follow-up.

Patient and graft survival was not significantly different between the two groups at 5-years follow-up as shown in Table 2. Renal function and the amount of excreted urine protein (expressed as protein − creatinine ratio in the urine) were similar in both groups.

Table 2: Results at 5 yr follow-up comparing patients at risk for CMV infection and a control group

The number of patients with at least one clinically or biopsy-proven rejection and the total number of clinically or biopsy proven acute rejection episodes during the 5-year follow-up were not significantly different between the two groups (CMV risk group: 35 episodes in 38 patients, control group: 32 episodes in 27 patients;P =0.48). The mean number of applied immunosuppressive agents in both groups was also not significantly different.

DISCUSSION

One of the controversial issues regarding the development of acute rejection and chronic graft loss in solid organ transplantation is the role of CMV infection. There is strong evidence that inflammatory processes are triggered by CMV infection ^(16–20). Different CMV strains seem to be associated with different expression patterns of growth factors ⁽²¹⁾. CMV infection was therefore proposed as an important risk factor for development of rejection and subsequent graft loss. However, because patients with rejection are treated with enhanced immunosuppression CMV might be a consequence of treatment of rejection and not the underlying cause.

Our results show that more than 80% of renal allograft recipients at risk for CMV infection experience CMV infection as diagnosed by the highly sensitive CMV pp65 antigenemia assay. However, a positive CMV antigenemia follows the acute rejection episodes in most cases, independent of an asymptomatic CMV infection or symptomatic CMV disease indicating that CMV infection or disease is most likely the result of an enhanced immunosuppression due to antirejection therapy. In comparison to the control group CMV viremia did not influence the incidence of acute rejection episodes, short- and long-term graft survival, and had no effect on graft function.

How can the controversial results in the literature be explained? The comparison of the clinical studies is difficult because of the use of different definitions as well as different prophylactic and therapeutic regimens with antiviral drugs. It is not excluded that some drugs, such as ganciclovir, might influence the immune response and therefore decrease the incidence of rejection, independent of their antiviral activity ^{(4, 22)}. This might explain the beneficial effect of CMV prophylaxis on graft function shown in some studies. The time course of CMV infection in relation to the time point of an acute rejection is often not delineated in detail. With the introduction of highly sensitive diagnostic methods it became possible to detect CMV infections at an early stage. The time correlation of CMV infection to an acute rejection episode might therefore be better evaluated. Some studies compared the number of rejection episodes and the outcome in patients after the introduction of CMV prophylaxis with historical controls, however, a recent analysis showed an improvement in graft survival over the last few years ⁽²³⁾. Therefore, a comparison with historical patient groups might give better results for the recently transplanted patients irrespective of CMV infection or prophylaxis. Finally, before effective therapeutic agents against CMV disease were available, immunosuppression with azathioprine in the case of CMV disease had often to be reduced or even stopped by leukopenia, thus inducing acute rejection.

In conclusion, we found no correlation of CMV infection in renal transplant recipients with the subsequent development of graft rejection and chronic graft failure at 5 years.