Cytomegalovirus (CMV) infection most commonly occurs in solid organ transplantation patients within the first 3 months following surgery, usually due to mismatched CMV serostatus, ie, a CMV-positive donor and a CMV-negative recipient (CMV D+/R−).1, 2 More than 75% of solid organ transplant recipients are newly infected with CMV or exhibit reactivation of a latent CMV infection after transplantation.3 CMV infection has adverse effects on graft and patient survival.2 In particular, graft survival in liver transplant recipients is significantly lower in patients with a history of CMV infection compared to those without.
In the absence of any preventive therapy, 30%-75% of transplant recipients develop CMV infection, and the reported incidence of CMV disease in liver transplant recipients is between 8% and 30%.4 Antiviral prophylaxis is associated with reduced biopsy-proven graft rejection and improved patient survival.5–7 However, the widespread use of prophylactic therapy to prevent CMV infection has resulted in an increase in late CMV infections in solid organ transplant recipients.8 However, little is known about CMV infections in CMV donor-positive/recipient-positive (CMV D+/R+) adult liver transplant cases without antiviral prophylaxis.
We compared CMV-positive and CMV-negative patients and determined the incidence of early and late-onset CMV infections among CMV D+/R+ Korean patients, regardless of antiviral prophylaxis. We also investigated mortality and graft failure risk factors in CMV D+/R+ adult liver transplant recipients.
CI, confidence interval; CMV, cytomegalovirus; CTP, Child-Turcotte-Pugh; DDLT, deceased donor liver transplantation; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HR, hazard ratio; LDLT, living donor liver transplantation; LD-PC, leukocyte-depleted platelet concentrate; LD-RBC, leukocyte-depleted red blood cells; MELD, model for end-stage liver disease; MMF, mycophenolate mofetil.
MATERIALS AND METHODS
We reviewed 653 medical records of consecutive adult liver transplant cases between January 1996 and January 2009 at Samsung Medical Center, Seoul, Korea. Patients who died within 1 month of liver transplantation, retransplant cases due to graft failure, and 1 patient with a negative donor/positive recipient status were excluded. All medical records of all patients were reviewed for epidemiologic and clinical characteristics. All patients were followed until death or the end of the study in October 2009. The following data were collected before CMV infection detection: patient demographics, preoperative diagnosis, Child-Turcotte-Pugh (CTP) scores, Model for End-Stage Liver Disease (MELD) scores, type of liver transplantation, amount of leukocyte-depleted red blood cell (LD-RBC) transfusion, amount of fresh frozen plasma, amount of leukocyte-depleted platelet concentrate (LD-PC), cryoprecipitate during liver transplantation, postoperative bleeding, reoperation, biliary complication, hepatic dysfunction, renal dysfunction, infection (positive bacterial or fungal culture in blood, sputum, ascites, and urine), acute rejection, recurrence of hepatitis B virus (HBV) or hepatitis C virus (HCV), use of antilymphocyte agents (antithymocyte globulin or muromonab-CD3), highest number of pp65-staining cells, preemptive therapy, CMV syndrome, tissue-invasive CMV disease, and date of death.
Virologic Follow-Up and CMV Infection Treatment
All patients were routinely tested for CMV after liver transplantation with the use of a CMV pp65 antigenemia assay. We did not routinely use antiviral prophylactic therapy in the liver transplant recipients. For the first month after transplantation, ethylene diamine tetraacetic acid–treated blood samples were examined weekly for the presence of CMV. Testing was performed 3 times a week if the patient had a known CMV infection. In the absence of symptoms, patients were routinely monitored for CMV once a month. If the patient had an unexplained fever or if a CMV infection was clinically suspected, a CMV antigenemia assay was conducted. If the CMV antigenemia assay showed more than 10 pp65 antigen–positive cells per 4 × 105 leukocytes, the patient was admitted to the hospital and preemptive treatment was initiated, regardless of clinical manifestations. CMV assays were conducted 3 times a week until a negative result was obtained. Preemptive antiviral therapy consisted of intravenous ganciclovir and was administered daily for 10-14 days until a negative CMV assay was confirmed. The dose of intravenous ganciclovir was adjusted according to patient creatinine clearance. Patients who developed acute rejection received anti-CMV prophylaxis during the treatment course of antithymocyte globulin or muromonab-CD3.
Tacrolimus, steroids, and mycophenolate mofetil (MMF) were the primary agents used for immunosuppression after liver transplantation. All transplant recipients were given 500 mg intravenous methylprednisolone during the anhepatic phase until postoperative day 2, which was tapered to 60 mg/day for a period of 5 days, and then administered at 8 mg, twice per day, for 1 month starting on postoperative day 8. Tacrolimus treatment was started on postoperative day 3, and the optimal blood level was adjusted to maintain a trough plasma concentration of 10-15 ng/mL during the first month and was reduced to 5-10 ng/mL thereafter. MMF was used in combination with tacrolimus and steroids. Starting on postoperative day 1, 750 mg MMF was administered twice a day. Cyclosporin (plasma concentration adjusted to 100-200 ng/mL) was used in the event of tacrolimus toxicity or tacrolimus refractory rejection, and was given orally twice a day. A liver biopsy was performed if acute rejection was clinically suspected. Methylprednisolone (500 mg) was administered intravenously every day for 3 days if acute rejection was confirmed by biopsy and was tapered to 60 mg/day over a period of 4 days thereafter.
All patients with HBV infection or recipients without hepatitis B surface antigen who received liver allograft with hepatitis B core antibody were given 10,000 units of hepatitis B immunoglobulin (Green Cross Corp., Yongin, Korea) intravenously during the anhepatic phase, which was followed by a 7-day intravenous course of 10,000 units hepatitis B immunoglobulin per day. Patients received 10,000 units intravenously every month to maintain anti–hepatitis B surface antibody titers at ≥200 IU/mL. Before 2008, patients who were reinfected with HBV received only lamivudine (100 mg/day) for treatment. After January 2008, patients received a combination of entecavir (0.5 mg/day) and hepatitis B immunoglobulin for HBV prophylaxis.
Definition of CMV Infection and Disease
CMV infection was defined as a CMV pp65 antigen-positive cell number greater than 1 positive cell per 400,000 white blood cells. Patients whose infections were detected in the first 3 months after liver transplantation were defined as early-onset CMV, with late-onset CMV defined as infections detected after the third month. Early and late-onset CMV cases were divided by onset time of the first CMV infection episode. CMV disease presented either as CMV syndrome or as tissue-invasive CMV disease. CMV syndrome was defined as a positive antigenemia assay with more than 1 of the following symptoms or signs: unexplained fever (>38.3°C), constitutional symptoms such as fatigue or general myalgia, leukopenia (white blood cell count <3000/mm3), or thrombocytopenia (platelet count <100,000/mm3). Tissue-invasive CMV disease was defined as the presence of hepatitis, pneumonitis, retinitis, or gastroenteritis, confirmed by biopsy.9
Categorical data were compared using a chi-square or Fisher's exact tests, and the Mann-Whitney U test weere used for continuous variables. Overall graft and patient survival between groups were determined by Kaplan-Meier methods, and differences were assessed by the log-rank test. We confirmed the assumption that hazard ratio in graft and patient survival curve between groups was equivalent prior to analysis. Risk factors for mortality and graft failure in adult liver transplant recipients were identified forward, using the Cox proportional hazard regression model. The Statistical Package for the Social Sciences (SPSS) version 17 for Windows was used for all tests. P values less than 0.05 were considered statistically significant.
Characteristics of CMV-Infected and Uninfected Cases
Of the 654 liver transplants that took place during the study period, 21 patients were excluded due to death unrelated to CMV infection within 1 month of the transplant surgery. All liver transplant donors and recipients were CMV seropositive except for 1 patient who was CMV D−/R+. Fourteen retransplant cases were also excluded. A total of 618 patients had only 1 transplant and were enrolled in the study.
The epidemiologic and clinical characteristics of the adult liver transplantation patients are summarized in Table 1. CMV infection was found in 344 patients (55.7%) who had more than 1 antigen-positive cell in the CMV antigenemia assay.9 The cumulative incidence of CMV infection in adult liver transplant recipients with CMV-seropositive status is shown in Fig. 1. The highest occurrence of CMV infection was within the first 6 weeks after transplantation resulting in a cumulative incidence of approximately 80%.
Table 1. Epidemiologic and Clinical Characteristics of Adult Liver Transplant Cases
Data are presented as the mean ± standard deviation.
Positive bacterial or fungal culture in blood, sputum, ascites, and urine.
Reoperation (P = 0.027), hepatic dysfunction (P = <0.001), renal dysfunction (P = <0.001), and infection (P = 0.002) were frequently found in patients with CMV infection (Table 1). Among CMV-infected patients, there were increased numbers of fulminant hepatic failure cases and higher MELD scores in the preoperative period than in uninfected patients (P = <0.001 and P = <0.001, respectively). Biliary complications (P = 0.002), acute rejection (P = 0.016), and HBV recurrence (P = <0.001) were more frequently found in uninfected compared to CMV-infected patients.
However, preoperative CTP scores, sex, age, type of liver transplantation, amount of transfusion during liver transplantation, postoperative bleeding, HCV recurrence, hepatocellular carcinoma (HCC) recurrence, use of antilymphocyte agents such as antithymocyte globulin or muromonab-CD3, and retransplantation showed no significant difference between the 2 groups.
A total of 154 (24.5%) patients received intravenous ganciclovir as a preemptive therapy due to more than 10 pp65 antigen-positive cells per 4 × 105 leukocytes on the CMV antigenemia assay. Thirty-four (5.5%) patients progressed to CMV disease, including 32 (5.2%) patients with CMV syndrome and 11 (1.8%) patients with tissue-invasive CMV disease [liver (n = 2), lung (n = 1), gastrointestinal tract (n = 6), and skin (n = 2)]. CMV infection was not detected in 2 cases with tissue-invasive CMV disease.
Patient and Graft Survival
A total of 131 (21.2%) patients died during the median follow-up (median time, 44.5 months; range, 2-156 months). Uninfected and CMV-infected patients were compared for graft failure and mortality (Fig. 2). The differences in graft failure and patient survival rates were not significant between the groups (P = 0.973 and P = 0.707, respectively), although there was a statistically significant difference between uninfected patients and patients with CMV disease (Fig. 3). For uninfected patients, survival rates at 1, 3, and 5 years were 86.9%, 81.0%, and 78.8%, respectively. The corresponding survival rates for patients with CMV disease were 65.6%, 62.0%, and 57.6%. Patients with CMV disease had lower survival rates than uninfected patients (P = 0.005). The graft survival rates of patients with CMV disease were also lower than those of uninfected patients (P = 0.030).
Early and Late-Onset CMV Infection
CMV-infected patients were divided into early-onset and late-onset groups (Table 2) and analyzed for survival rates. Male sex (P = 0.032), biliary complications (P = 0.003), acute rejection (P = <0.001), and the use of antilymphocyte agents such as antithymocyte globulin or muromonab-CD3 (P = 0.008) were more frequent in patients with late-onset CMV infection. Recipients with early-onset CMV infection were transfused with smaller amounts of LD-RBCs and fresh frozen plasma (P = <0.001 and P = <0.001) than those with late-onset CMV infection. Other variables were not significantly different between the 2 groups.
Table 2. Early and Late-Onset CMV Infections
Data are presented as the mean ± standard deviation.
Positive bacterial or fungal culture in blood, sputum, ascites, and urine.
There were 27 (8.6%, 27/313) and 7 (22.6%, 7/31) patients of early-onset and late-onset CMV disease, respectively (P = 0.023). All early-onset tissue-invasive CMV disease cases (8 in total) were diagnosed with CMV syndrome, although 2 cases among patients with late-onset tissue-invasive CMV disease were not accompanied by CMV syndrome. Late-onset CMV disease cases (n = 7) consisted of CMV syndrome (n = 4), tissue-invasive CMV disease (n = 2), and a combination of CMV syndrome and tissue-invasive CMV disease (n = 1). Of the 154 patients treated preemptively, 14 received preemptive therapy due to CMV recurrence (9.1%). Three patients suffered from CMV infection 3 times during the study period. There were 8 patients with early-onset CMV infection that developed into late-onset CMV infection. Of these, 5 died.
Patient and graft survival rates are shown in Fig. 4. For the early-onset CMV-infected patients, survival rates at 1, 3, and 5 years were 87.9%, 83.2%, 78.0%, respectively, whereas the corresponding graft survival rates were 88.2%, 83.2%, and 77.5%. However, in the late-onset CMV-infected patients, the corresponding yearly survival rates at 1, 3, and 5 years were 83.9%, 64.5%, and 60.9%, with graft survival rates of 87.1%, 67.0%, and 63.3%, respectively. There were no statistically significant differences in patient and graft survival between late-onset and early-onset CMV infection groups (P = 0.068 and P = 0.158).
Risk Factors for Mortality and Graft Failure in Adult Liver Transplantation
The associations between multiple variables and the graft failure and mortality of adult liver transplant recipients (n = 618) are outlined in Tables 3 and 4. Using univariate and multivariate analyses by Cox regression analysis, HBV recurrence (P = 0.002 and P = 0.003), HCC recurrence (P = <0.001 and P = <0.001), hepatic dysfunction (P = <0.001 and P = <0.001), infection which is defined by positive bacterial or fungal culture in blood, sputum, ascites, and urine (P = <0.001 and P = 0.001), CMV disease (P = 0.007 and P = 0.021), and highest pp65-staining cells (P = 0.027 and P = 0.035) were found to be risk factors for mortality and graft failure; early and late-onset CMV infection were not risk factors.
Table 3. Univariate and Multivariate Analysis Showing Risk Factors for Graft Failure in Adult Liver Transplant Recipients
Positive bacterial or fungal culture in blood, sputum, ascites, and urine.
In this study, we asked how clinical characteristics and mortality might differ based on CMV infection status and time of onset in a CMV-endemic area of Korea and in a situation where all donors and recipients were CMV-seropositive, regardless of antiviral prophylaxis. We investigated the epidemiologic and clinical characteristics of adult liver transplant recipients. Our results showed a high incidence (55.7%, 344/618) of CMV infection, but a low incidence of (5.5%, 34/618) of CMV disease, and we showed that CMV disease had adverse effects on graft and patient survival.
Many articles have reported that CMV infection is associated with significant morbidity and mortality. The present study shows that patient and graft survival do not differ based on the presence of CMV infection. However, patient and graft survival rates in patients with CMV disease were lower than those in uninfected patients. These results hold in CMV D+/R+ recipients with preemptive therapy. CMV disease was a risk factor for patient and graft survival, but CMV infection was not.
A total of 154 (24.9%) patients received intravenous ganciclovir as a preemptive therapy after showing more than 10 pp65 antigen-positive cells per 4 × 105 leukocytes on a CMV antigenemia assay. Thirty-four patients (5.5%) developed CMV disease, 4 of whom did not receive previous preemptive therapy. The incidence of patients with CMV disease among patients who received preemptive therapy was 19.5% (30 of 154 patients).
The most significant risk factor for CMV infection after liver transplantation was mismatched serostatus (D+/R−). However, CMV D+/R− serologic mismatch was rarely found in adult Korean patients because nearly all adult recipients were CMV-seropositive. Our center experienced only 1 case of CMV D−/R+ in adult liver transplantation. The occurrence of CMV infection in adult liver transplant recipients with CMV D+/R+ was due to reactivation of a latent virus. The use of antiviral prophylaxis in CMV D+/R+ cases is recommended after transplantation because these patients are usually at higher risk for developing CMV disease.9 However, the use of antiviral prophylaxis in Korea has been limited by the National Health Insurance Corporation (NHIC). Therefore, preemptive therapy is considered important for the prevention of CMV disease when the use of antiviral prophylaxis is limited. Our study revealed that the incidence of CMV infection in recipients with CMV D+/R+ was approximately 80% the first 6 weeks after adult liver transplantation. There was no statistically significant difference of graft and patient survival rates between uninfected and CMV-infected patients who received preemptive therapy due to diagnosis with positive CMV antigenemia assay.
Some studies have demonstrated the efficacy of preemptive therapy for CMV disease prevention,10–12 and 3 meta-analyses have confirmed the benefits of preemptive therapy in the prevention of CMV disease.13–15 CMV disease developed in CMV-infected patients despite preemptive therapy. Only 2 patients developed tissue-invasive CMV disease without positive CMV antigenemia more than 3 months after liver transplantation. Because CMV disease has been shown as a risk factor for graft and patient survival, continuous monitoring for the detection and prevention of CMV disease should be required even 3 months after transplantation.
We also compared the effect of early and late-onset CMV infection. Late-onset CMV-infected patients were mostly male and showed biliary complications and acute rejection. We examined patient survival rates based on the time of CMV infection for the exclusion of lead-time bias. The patient and graft survival rates for patients with late-onset CMV infection tended to be lower than those of the early-onset patients; however, patient and graft survival rates of early and late-onset CMV infection recipients were not statistically significant. The data show that 3-year patient and graft survival rates abruptly decreased to approximately 60% with the presence of late-onset CMV infection. We examined the effect of biliary complications and acute rejection in late-onset CMV infection cases and found that these problems were significantly associated with mortality and graft failure.
This was expected, because the amount of allogenic blood transfused during transplantation increased patient immunosuppression, which explains why many patients experienced early CMV infection. However, patients with late-onset CMV infection had increased levels of LD-RBCs and fresh frozen plasma transfused during liver transplantation than patients with early-onset CMV infection. These issues are in need of further study.
We found that CMV disease was a risk factor for mortality and graft failure in adult liver transplant recipients as patient and graft survival rates in patients with CMV disease were lower than in those of uninfected patients. We also found that the state of CMV infection is not an important factor in graft survival and mortality. The occurrence of CMV disease was more serious in liver transplant recipients with CMV D+/R+, and CMV disease developed most commonly when the number of pp65-staining cells in the CMV antigenemia assay was high. Previous studies have shown that viral load is a significant risk factor for the development of CMV disease,16 and it has been shown to be a useful prognostic indicator for recipients of solid organ transplants.17–19 CMV replication is highly dynamic once an active infection has been established, which leads to a rapid increase in viral load, the development of CMV viremia, and an elevated risk of invasive disease.18, 20 The probability of developing CMV disease increases logarithmically with viral load, which indicates that preemptive therapy may prevent the progression of CMV disease by effectively decreasing CMV titer.
There are several problems in the present study. First, the CMV-PCR assay was not conducted in patients with CMV infections. Second, for optimal preemptive therapy, there was a strong consensus that liver transplants should be monitored weekly by CMV antigenemia for 3 months after transplantation.9 Usually, our liver transplant patients stayed in the hospital for 3 weeks after transplantation, revisited 1 week after discharge, and were subsequently followed every month. We could not encourage patients to visit every week until the first 3 months after liver transplantation because most patients lived far from the hospital and the patients' primary care physicians themselves could not care for liver transplant recipients. Therefore, there was no choice but to have patients visit the hospital every month after discharge. As such, we were not able to detect asymptomatic CMV infection in the second and third months after transplantation. Third, our study was limited to investigating the role of preemptive therapy because our patients were not classified with a case and control group. All patients who showed more than 10 pp65 antigen-positive cells per 4 × 105 leukocytes on a CMV antigenemia assay received preemptive therapy; therefore, we could not extract the true effect of preemptive therapy in the present study. Finally, most patients in this study had an HBV infection, although the prevalence of HBV infection was not as high as in other countries. Therefore, it is uncertain whether the results of this study can be generalized to other populations. Nevertheless, we believe that the results may be generalized to other populations as long as the patients are classified as CMV D+/R+.
In conclusion, patient and graft survival rates did not differ according to the presence of CMV infection, but these rates significantly lowered in patients with CMV disease. Patient and graft survival rates were lower in patients with late-onset CMV infection compared to those with early-onset CMV infection. However, late-onset CMV infection was not a risk factor for patient or graft survival in multivariate analyses. The risk factors of mortality and graft failure in adult liver transplant recipients with CMV D+/R+ were recurrent HBV infection, recurrent HCC, hepatic dysfunction, infection, high numbers of pp65-staining cells, and CMV disease. The occurrence of CMV disease was a risk factor for mortality and graft failure in adult liver transplant recipients with CMV D+/R+, but asymptomatic CMV infection was not.