We included a total of 55 adult KT recipients who received ATG for the treatment of steroid resistant rejection in Seoul St. Mary’s Hospital between January 2010 and April 2016. TCMR was diagnosed based on Banff classification [9]. According to the anti-rejection treatment protocol of Seoul St. Mary’s Hospital, TCMR was treated with 3 to 5 daily boluses of intravenous methylprednisolone (500 mg/day), followed by a 5 to 7 day oral steroid taper. However, when the serum creatinine level failed to decrease within 5 days, we regarded it as steroid resistant rejection and ATG was applied [10]. All recipients received ATG injection (1.5 mg/kg/day) for a week.

CMV prophylaxis protocol in KTRs who received ATG as anti-rejection therapy was as follows (Fig. 1): In all 55 KTRs, we used intravenous ganciclovir during ATG injection at a dose of 2.5 mg/kg twice a day, and the dose was modified according to the allograft function of each patient [11,12]. After the completion of ATG therapy, plasma CMV DNA real-time polymerase chain reaction (PCR) was checked on D7, 1, 3, 6, 9, and 12 months, and when the CMV copy number exceeded 2,000 copies/mL, we used intravenous ganciclovir therapy.

Before July 2013, we used preemptive therapy for CMV infection; in brief, the recipients received intravenous ganciclovir therapy only when CMV infection was detected without the use of anti-viral agent prophylaxis. These 34 KTRs belonged to the historic control group (CONT, n = 34). After July 2013, we used maintenance valacyclovir at a dosage of 2 g four times daily by mouth during 1 month after the completion of ATG therapy, in addition to the previous prophylaxis protocol [8]. Twenty-one patients who received this revised protocol belonged to the valacyclovir study group (VAL, n = 21). We compared the clinical outcomes and parameters between the CONT and VAL groups. This retrospective study was approved by the Institutional Review Board of our institution (KC140NMI0460), which waived the requirement for informed consent.

We collected the baseline characteristics of the recipients, including sex and age. Data about transplantation included the number of human leukocyte antigen mismatches, pretransplant panel reactive antibody, positivity of cross match study, pretransplant CMV serostatus, and posttransplant immunosuppressant regimen. Clinical information, such as pathologic findings of rejection, time interval from transplantation to ATG treatment, duration of ATG administration, total dose of ATG, and duration of valacyclovir administration, was assessed.

The primary outcome of this study was the incidence of CMV infection after ATG treatment for steroid resistant TCMR. CMV infection was defined as CMV viremia (≥ 500 copies/mL) using quantitative CMV PCR, irrespective of the symptoms. The secondary outcomes of this study were CMV disease, subsequent acute rejection, graft function assessed as estimated glomerular filtration rate by the Chronic Kidney Disease Epidemiology Collaboration formula, adverse events, death-censored allograft survival, and patient survival rate [13]. CMV disease was defined as symptomatic infection. Clinical symptoms included CMV syndrome (fever, and one or more of the following findings: constitutional symptoms, leukopenia, thrombocytopenia, or liver enzyme elevation) or tissue-invasive CMV disease [14].

Statistical analyses were performed using SPSS version 23.0 (IBM Co., Armonk, NY, USA). Data are presented as mean ± standard deviation or counts and percentages, depending on the data type. For continuous variables, means were compared using Student t test. For categorized variables, Pearson chi-square test and Fisher exact test were used. Patient, graft and CMV-free survival rates were calculated using Kaplan-Meier analysis, and we used the log-rank method to compare the survival rates between the VAL and CONT groups. All tests were two-tailed, and the results were considered statistically significant when the p value was less than 0.05.

The comparison of baseline characteristics between the VAL group and the CONT group is presented in Table 1. Demographic characteristics such as patient’s mean age at the time of TCMR diagnosis and sex ratio were not different between the two groups. The pretransplant immunologic status and CMV serostatus of donors and recipients, and the time interval from KT to ATG administration did not differ significantly between the two groups (p > 0.05, respectively) (Table 1). At the time of allograft biopsy, immune suppression regimens, pathologic findings at biopsy, allograft function and the duration or administered amount of ATG also showed no differences. The allograft function at biopsy also did not differ between the two groups. The mean duration of valacyclovir administration was 28.3 ± 5 days in the VAL group.

The incidence of CMV infection was significantly lower in the VAL group (2/21, 9.6%) compared to the CONT group (23/34, 67.6%; p < 0.001) (Fig. 2A). With respect to the interval between ATG administration and CMV infection, it was significantly shorter in the CONT group (34.8 ± 19 days) than in the VAL group (101 ± 33 days, p < 0.001 vs. CONT group). In the CONT group, all CMV infections occurred within the first 3 months after ATG treatment. On the other hand, in the VAL group, all cases of CMV infection occurred after 1 month. Therefore, the 1-year CMV-free survival rate was significantly higher in the VAL group compared to the control group (80% vs. 38.6%, p = 0.009) (Fig. 2B). In the VAL group, all cases of CMV infection were limited to CMV viremia. In contrast, in the CONT group, three cases of CMV disease were detected; one case of CMV nephritis, proven by kidney biopsy, and two cases of CMV syndrome.

The cumulative incidence of leukopenia (< 4.0 × 10⁹/L) (CONT, 32% [11/34] vs. VAL, 23% [5/21]; p = 0.556), anemia (hemoglobin < 9 g/dL) (CONT, 24% [8/34] vs. VAL, 33% [7/21]; p = 0.272), and thrombocytopenia (< 100 × 10⁹/L) (CONT, 9% [3/34] vs. VAL, 14% [3/21]; p = 0.664), did not differ between the CONT and VAL groups (Table 2). This indicates that valacyclovir did not increase the risk of bone marrow suppression, and it had no additional effect. In six patients, valacyclovir was withheld within 1 month because of deterioration of allograft function. However, there were no clinically important side effects associated with valacyclovir such as hallucination and/or confusion.

There was no significant difference in subsequent acute rejection after ATG treatment between the two groups (Fig. 3A). The incidence of subsequent acute rejection after ATG treatment was 41.1% (14/34) in the CONT group, whereas the incidence of subsequent acute rejection after ATG treatment was 33.3% (7/21) in the VAL group (p = 0.776). The allograft survival rate after allograft rejection also did not differ between the two groups (p = 0.508) (Fig. 3B). In the CONT group, graft failure occurred in 70.5% of the cases (24/34), while in the VAL group, graft failure occurred in 47.6% of the cases (10/21) (p = 0.152). Chronic rejection was the main cause of graft failure in both groups (CONT, 87.5% [21/24] vs. VAL, 100% [10/10]; p = 0.067). Allograft function showed a lower tendency in the CONT group after ATG treatment; however, the differences were not significant (Fig. 3C). There was no patient death in the VAL group, and there were two cases of death in the CONT group. In the CONT group, one patient died of urothelial cancer of the kidney and the other patient died of atypical pneumonia. Patient survival rate showed no difference between the two groups (p = 0.480).