Early Detection of Epstein-Barr Virus as a Risk Factor for Chronic High Epstein-Barr Viral Load Carriage at a Living-donor–dominant Pediatric Liver Transplantation Center : Transplantation

Journal Logo

Original Clinical Science—Liver

Early Detection of Epstein-Barr Virus as a Risk Factor for Chronic High Epstein-Barr Viral Load Carriage at a Living-donor–dominant Pediatric Liver Transplantation Center

Yamada, Masaki MD, PhD1,2,3; Fukuda, Akinari MD, PhD4; Ogura, Miyuki MS1; Shimizu, Seiichi MD, PhD4; Uchida, Hajime MD, PhD4; Yanagi, Yusuke MD, PhD4; Ishikawa, Yuriko PhD1; Sakamoto, Seisuke MD, PhD4; Kasahara, Mureo MD, PhD4; Imadome, Ken-Ichi PhD1

Author Information
Transplantation: November 22, 2022 - Volume - Issue - 10.1097/TP.0000000000004429
doi: 10.1097/TP.0000000000004429
  • Free
  • SDC
  • PAP
  • Infographic



Liver transplantation (LT) has become a successful, life-saving procedure for children with end-stage liver diseases. The prognosis for LT has been improving because of standardized surgical procedures, newly developed immunosuppressive agents, and strategies to prevent infectious complications. However, infections remain a major complication for children undergoing LT.1

Epstein-Barr virus (EBV) is a ubiquitous virus that infects >90% of the adult population worldwide.2 EBV infection after pediatric LT accounts for significant morbidity and mortality because of its ability to immortalize and transform the infected cells, resulting in EBV-associated posttransplant lymphoproliferative disorders (PTLDs).3 Surveillance measurement of EBV DNA loads as measured by quantitative polymerase chain reaction (PCR) in peripheral blood has been implemented since the early 2000s, as EBV DNAemia usually precedes the development of PTLD.4 A “high” EBV load, which is usually determined per each assay, is not diagnostic for PTLD per se but is known to precede the development of PTLD.5-7 Accordingly, a reduction of immunosuppression (RIS) is recommended when new EBV DNAemia is identified, and the EBV loads reach “high” to keep the EBV loads at a low level. With this approach, the overall incidence of PTLD appears to decrease.8,9 However, several prior studies have identified a group of patients who maintained EBV loads at a high level despite RIS, and they are described as chronic high EBV load (CHL) carriers.10,11 Importantly, a very high incidence of PTLD in CHL carriers has been reported in heart recipients but is not obvious in kidney and liver recipients.10,12-15 Hence, the CHL carrier state is believed to be a potential indicator of a condition of EBV-infected cells transitioning to the development of PTLD. Although RIS is an effective therapeutic intervention, minimizing immunosuppression may increase the risk of rejection. Hence, titration of immunosuppressants to manage CHL with the risk of rejection can be challenging.

Generally, pediatric LT recipients have a higher risk of EBV-associated complications than their adult counterparts because EBV serologies are more frequently “mismatched” (R−/D+) in pediatric LT; some donors are adults, who are typically EBV seropositive (D+), and the recipients tend to be young infants or toddlers, who are mostly EBV seronegative (R−). However, few studies have investigated EBV complications and CHL carriers among pediatric living-donor LT recipients.16,17

Therefore, the primary aims of this study are (1) to investigate the EBV load kinetics and epidemiology of LT recipients in each EBV condition, with a particular focus on CHL carriers, and (2) to identify risk factors for CHL carriage at a living-donor–dominant pediatric LT center.


Patient Enrollment

This retrospective study was conducted using an electronic medical record and transplantation database at the National Center for Child Health and Development, Tokyo, Japan. A total of 525 patients undergoing LT between January 1, 2006, and December 31, 2018, were identified in the initial screening. Of those, patients older than 18 y, undergoing multiorgan transplantation, or undergoing multiple LT were excluded. The data collection was performed until the end of 2020; thus, all the study participants were followed up at least 2 y post-LT.

This study protocol conformed to the ethical guidelines of the 2013 Declaration of Helsinki and was approved by the Institutional Review Board (#2021-073).

Immunosuppressive Regimens

Standard immunosuppression protocols during the study periods were as follows. The first-line induction therapy was high-dose methylprednisolone with several tapering protocols that have changed over the decade. No patients received lymphocyte-depleting inductions, such as antithymoglobulin (ATG). ATG was administered only for the treatment of steroid-refractory acute T cell–mediated rejection. Pretransplant plasma exchange and rituximab were administered for the ABO-incompatible LT recipients older than 2 y before and 18 mo after protocol modification in 2017.

Maintenance therapy consisted of tacrolimus monotherapy, or less commonly, cyclosporin monotherapy. Tacrolimus trough levels were aimed between 10 and 12 ng/mL during the first month and gradually decreased to 3 to 5 ng/mL after 6 mo post-LT. For the patients with a history of rejection, prednisone and mycophenolate mofetil were added depending on their pathological diagnosis and clinical course.

EBV Load Monitoring Protocol

EBV monitoring was performed per clinical protocol, typically biweekly during the first 3 mo and monthly until the end of the first year post-LT, when the clinical course was uneventful. Patients with newly quantifiable EBV loads (>10 copies/μg DNA) and a recent rejection episode with intensive immunosuppression were monitored more closely in every outpatient visit, typically monthly or bimonthly until the EBV loads stabilized or cleared. Additional tests, including computed tomography, were indicated when a patient develops symptoms consistent with EBV infection and PTLD (eg, fever, lymphadenopathy, failure to thrive, and night sweat). EBV DNA loads were reported in copies/microgram DNA in blood cells with a conversion formula to the World Health Organization (WHO) standards.18 The equation for conversion to international units (IU) was calculated using international standard reagents purchased from WHO as follows19:


With this formula, 1000 copies/μg DNA were equivalent to 742 IU/μg DNA.

Definition of High EBV Loads and Management

The method for EBV DNA amplification is described elsewhere.18 For statistical analysis on a logarithmic scale, negative and positive results below the quantifiable range were expressed as 1 and 10 copies/μg DNA, respectively. Seventy-five percent of all the EBV load monitoring among transplant recipients was 650 copies/μg DNA. EBV loads exceeding 1000 copies/μg DNA in blood were considered “high” in our clinical protocol because the previous data consistently showed 75% being between 500 and 1000 copies/μg DNA. CHL was defined when the high EBV load persisted for 6 mo or longer with at least 2 data point measurements. All the other quantifiable detections of EBV loads were considered “low.”

When EBV loads exceed 1000 copies/μg DNA, the second and third immunosuppressive agents, such as mycophenolate mofetil and prednisone, were tapered or discontinued per clinical protocol. If high EBV loads persisted with those interventions, reducing calcineurin inhibitor by 25% to 50% was attempted and monitored monthly until the load trends were stabilized. CHL developed in some cases even with RIS, and in these cases, the goal of tacrolimus trough level was set as low as 1 to 3 ng/mL whenever clinically appropriate.

EBV and Cytomegalovirus Serology

EBV serologies of both recipients (R) and donors (D) were routinely measured before LT. When the results were either unrecorded or equivocal, we assigned those serostatuses to the highest EBV/cytomegalovirus (CMV) risk groups, as clinical guidelines recommend (eg, donor serologies were treated as D+, and recipient serologies were treated as R− when D+).7 Similarly, positive serology among children <1 y of age was considered R− unless PCR confirms pretransplant EBV/CMV infection (eg, EBV hepatitis as a primary disease resulting in fulminant liver failure) to avoid false-positives due to transplacental transmission of those antibodies and extrinsic blood products.

We have been using a preemptive strategy to prevent CMV disease20; thus, the antiviral prophylaxis was not provided routinely.

Statistical Analysis

Categorical data were expressed as numbers (%), and continuous variables were expressed as the median with interquartile ranges (IQRs). The chi-square test and Fisher exact test were used for categorical variables; 1-way ANOVA and the Kruskal-Wallis test, to compare continuous variables, were used for univariate analysis where appropriate.

Cumulative incidence was calculated using the time to EBV DNAemia, with death and retransplantation as competing risks. After removing confounding factors with significant correlation, a multivariate logistic regression analysis was conducted if the model fitting was appropriate. A P value <0.05 was considered statistically significant. JMP 16 (SAS Institute Inc) and Prism 8 (GraphPad) were used for statistical analysis and graph creation.



Initial screening identified 525 children undergoing LT between January 1, 2006, and December 31, 2018. Of those, 20 patients older than 18 y of age and 16 patients who had undergone previous organ transplantation were excluded. Additionally, 95 LT recipients with fewer than 5 EBV load measurements were excluded to assign EBV load categories appropriately. The most common reasons for too few EBV load measurements were early post-LT death and incomplete monitoring, mainly at the beginning of the protocolized monitoring before 2010. As a result, 394 isolated LT recipients were included for further analyses, of which 313/394 (79%) recipients were R− (Figure 1).

Patient enrollment. A total of 525 pediatric patients underwent liver transplantation between January 1, 2006, and December 31, 2018. Of those, 20 patients older than 18 y and 16 patients with multiorgan or multiple liver transplantation were excluded. After collecting the data, the patients who only had a few EBV load data points were also excluded because of the inability to define CHL. Finally, 394 LT recipients were enrolled in this study, of which 313 (79%) were EBV seronegative or unknown at the time of LT. CHL, chronic high Epstein-Barr viral load; EBV, Epstein-Barr virus; LT, liver transplantation; PCR; polymerase chain reaction; R±, recipients’ negative/positive EBV serology.

Patient Demographics

The demographics of 394 LT recipients are summarized in Table 1. The median age of LT recipients was 0.83 y (IQR: 0.58–2.83), which is younger than in most previous publications on pediatric LT worldwide.21,22 The primary diagnoses resulting in LT were biliary disease in 214 of 394 (54%), metabolic diseases in 69 of 394 (18%), fulminant hepatitis in 59 of 394 (15%), and others in 52 of 394 (13%). A total of 381 of 394 (97%) donors were living-related donors, and 75 of 394 (19%) underwent ABO-incompatible LT. Regarding the pre-LT EBV/CMV serologies, 308 of 394 recipients (78%) were EBV mismatched (D+ R−), and 257 of 394 (65%) were CMV D+. The EBV and CMV seropositivity rates of donors were consistent with other domestic reports among healthy adults.

Table 1. - Patient demographics
Number of patients N = 394
Age at LT, years, median, (IQR) 0.83 (0.58–2.83)
Male sex, n (%) 181 (46)
BW, kg, median, (IQR) 8.1 (6.6–12.7)
Height, cm, median, (IQR) 68.6 (62.5–89.1)
Primary liver disease, n (%)
 Biliary disease 214 (54)
 Metabolic disease 69 (18)
 Fulminant hepatitis 59 (15)
 Others 52 (13)
Pre-LT ICU admission, n (%) 94 (24)
Donor type, n (%)
 Living donor 381 (97)
 Deceased donor 13 (3)
Blood type compatibility, n (%)
 Identical 224 (57)
 Compatible 95 (24)
 Incompatible 75 (19)
GRWR, %, median (IQR) 2.68 (1.89–3.45)
Blood loss, L/kg, median (IQR) 62.8 (35.7–102.6)
Donor age, years, median (IQR) 34.5 (30–39)
EBV serology, n (%)
 R+/D+ 78 (20)
 R+/D– 3 (1)
 R–/D+ 308 (78)
 R–/D– 5 (1)
CMV serology, n (%)
 R+/D+ 41 (10)
 R+/D– 15 (4)
 R–/D+ 216 (55)
 R–/D– 122 (31)
BW, body weight; CMV, cytomegalovirus; EBV, Epstein-Barr virus; LT, liver transplantation; IQR, interquartile range; GRWR, graft-to-recipient weight ratio; ICU, intensive care unit.

Samples of EBV DNA Load Monitoring

There were 5827 EBV PCR results analyzed from 394 patients, and EBV loads were measured approximately 15 times per patient on average. EBV DNA has been detected in 2783 of 5827 (47.8%) samples. The EBV load values ranged from undetectable to 1 500 000 (IQR: undetectable to 650) copies/μg DNA (Figure 2A). A total of 5024 of 5827 (86%) EBV DNA PCR assays were performed on the samples from R− patients. The majority (1158/1244, 93%) of the high EBV load values above 1000 copies/μg DNA was seen in the samples from R− patients (Figure 2B).

Range and distribution of EBV DNA load in peripheral blood from pediatric LT recipients. A, The overall EBV load data obtained from 394 LT patients and time post-LT are depicted. When the results were either negative or less than the quantifiable range, those values were treated as 1 and 10 copies/μg DNA, respectively. B, Number of samples in each viral load range and EBV serostatus are summarized. The majority of viral load measurements were from EBV R− patients. EBV, Epstein-Barr virus; LT, liver transplantation; R±, recipients’ negative/positive EBV serology.

Frequent Post-LT EBV DNAemia in Both EBV R+ and R− Patients

The timing of the first EBV DNAemia separated by each EBV serostatus was analyzed between the primary acquisition of EBV (the first EBV DNAemia in R−) and reactivation of EBV (the first EBV DNAemia in R+) using a cumulative incidence function test (Figure 3A). Overall, EBV DNAemia occurred in approximately 80% of the study cohort during the first 2 y post-LT. Cumulative incidences of EBV DNAemia showed no difference between R+ and R− patients. EBV DNAemia developed in all R+ patients during the long-term follow-up, although the values were relatively low (data not shown).

Cumulative incidence of EBV DNAemia after pediatric LT. A, Frequencies of EBV DNAemia were calculated by a cumulative incidence function with death or retransplantation as competing risks. There was no difference between R− and R+ patients. B, EBV loads in each time period. The data were categorized every quarter year until 3 y out, every half year between 3 and 5 y out, and thereafter. The individual values in each time category were analyzed to assess the differences between R+ and R− patients. EBV loads in R− patients were constantly higher in most of the period between 6 mo and 3 y post-LT, but the difference became negligible after 3 y and thereafter with few exceptions. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. EBV, Epstein-Barr virus; LT, liver transplantation; R−/+, recipients’ negative/positive EBV serology.

Interestingly, on average, the patients who developed the first EBV DNAemia within the first 6 mo post-LT had higher maximum EBV loads than those who developed the first EBV DNAemia later (Figure S1, SDC,https://links.lww.com/TP/C630).

Different Kinetics of EBV Loads Between R+ and R− Patients

To assess the different EBV load kinetics between the R+ and R− patients, we categorized the EBV loads by time post-LT and compared their distributions. Between 6 mo and 3 y post-LT, the overall EBV loads of R− were significantly higher. However, this difference between R+ and R− became negligible 3 y post-LT and later (Figure 3B).

Demographic Characteristics in Each EBV Load Category

Next, we divided the patients into the following 3 categories based on the EBV loads: (1) low (184/394, 47%), whose EBV loads were always below 1000 copies/μg DNA; (2) high, whose EBV loads occasionally exceeded 1000 copies/μg DNA but were not persistent (149/394, 38%); and (3) CHL (61/394, 15%). We analyzed the demographic differences among these 3 groups (Table 2). Although the age at the time of LT did not reach statistical significance, body weight (BW) and height were significantly lower in CHL carriers than in the other 2 groups (P = 0.047 and 0.041, respectively), indicating that the CHL carriers tended to be younger and smaller. The pre-LT intensive care unit (ICU) admission rate was significantly lower (6/61, 10%) in CHL carriers than in the others. The graft-versus-weight ratio was significantly higher in CHL carriers, although this was likely confounded by low BW. In addition, there were fewer male patients (26/61, 43%) and more mismatched EBV serology (55/61, 90%) among CHL carriers. Accordingly, EBV R− recipients developed CHL more frequently (56/313, 18%) than EBV R+ recipients (5/81, 6%).

Table 2. - Patient characteristics in each EBV load category
Category, number of patients, n (%) Low, N = 184 (47) High, N = 149 (38) CHL, N = 61 (15) P
Age at LT, years, median (IQR) 0.83 (0.58–5.69) 0.83 (0.58–2.50) 0.75 (0.58–1.21) * 0.10
Male sex, n (%) 98 (53) 57(38) 26 (43) 0.02
BW, kg, median (IQR) 9.0 (6.5–16.3) 8.0 (6.6–12.3) 7.6 (6.6–9.6) * 0.047
Height, cm, median (IQR) 70.0 (62.5–104.1) 67.5 (62.8–87.5) 66.7 (62.0–74.3) * 0.041
Primary liver disease, n (%) <0.001
 Biliary disease 81 (44) 95 (64) 38 (62)
 Metabolic disease 28 (15) 27 (18) 14 (23)
 Fulminant hepatitis 38 (21) 17 (11) 4 (7)
 Others 37 (20) 10 (7) 5 (8)
Pre-LT ICU admission, n (%) 56 (30) 32 (21) 6 (10) 0.003
Donor type, n (%) 0.71
 Living donor 173 (96) 144 (97) 60 (98)
 Deceased donor 7 (4) 5 (3) 1 (2)
Blood type compatibility, n (%) 0.77
 Identical 99 (53.8) 87 (58.4) 38 (62.3)
 Compatible 47(25.5) 36 (24.2) 12 (19.7)
 Incompatible 38 (20.7) 26 (17.5) 11 (18.0)
Pre-LT use of rituximab for ABO-I 5 (3) 1 (1) 0 (0) 0.17
GRWR, %, median (IQR) 2.49 (1.65–3.32) 2.71 (1.96–3.48) 2.89 (2.30–3.52) 0.025
Blood loss, L/kg, median (IQR) 61.8 (32.9–104.8) 65.5 (39.3–102.3) 60.5 (31.4–105.8) 0.70
Donor age, years, median (IQR) 35 (30–39) 34 (30–39) 34 (30–37) 0.61
EBV serology, n (%) 0.0296
 R+/D+ 49 (27) 24 (16) 5 (8)
 R+/D– 2 (1) 1 (1) 0 (0)
 R–/D+ 132 (72) 121 (81) 55 (90)
 R–/D– 1 (1) 3 (2) 1 (2)
EBV disease, n (%) 0 (0) 3 (2) 2 (3) 0.08
CMV serology, n (%) 0.0201
 R+/D+ 28 (15) 12 (8) 1 (2)
 R+/D- 8 (4) 3 (2) 4 (7)
 R–/D+ 94 (51) 90 (60) 32 (52)
 R–/D– 54 (29) 44 (30) 24 (39)
ABO-I, ABO incompatibility; BW, body weight; CHL, chronic high load; CMV, cytomegalovirus; D, donor serology; EBV, Epstein-Barr virus; GRWR, graft-to-recipient weight ratio; ICU, intensive care unit; IQR, interquartile range; LT, liver transplantation; R, recipient serology.

In contrast, donor type, blood type compatibility, use of rituximab, amount of blood loss (mL/kg), and donor age were not statistically different. Also, the incidence of biopsy-proven rejection and use of ATG within 1 mo and the biopsy-proven rejection rate after 1-y post-LT were not different (Table S1, SDC,https://links.lww.com/TP/C630).

Analysis to further evaluate the onset of CHL carrier state revealed that high EBV load typically starts early; 80% of CHL carrier states had begun within 1-y post-LT (Figure S2, SDC,https://links.lww.com/TP/C630).

Risk Factors for CHL

To evaluate the risk factors for the development of CHL carrier state, univariate and multivariate analyses were conducted using several demographic and clinical characteristics as follows: early development of EBV DNAemia within 6 mo post-LT, history of biopsy-proven rejection within 1-mo post-LT, and use of ATG before the development of CHL carrier state, in addition to other demographic factors described in Table 2.

Younger age, low BW, short height, EBV R−, CMV D−, and the first EBV DNAemia within 6 mo post-LT were all associated with the increased risk of CHL development in univariate analysis. In contrast, pre-LT ICU admission appeared to reduce the risk of CHL development (Table 3).

Table 3. - Risk factors for chronic high EBV load after pediatric liver transplantation.
Univariate analysis Multivariate analysis
OR 95% CI P OR 95% CI P
Age (years) <5 years 8.14 1.94-34.1 0.0007 8.20 1.80-37.4 0.007
Male sex (vs. female) 0.85 0.0.49-1.48 0.57 1.03 0.56-1.90 0.91
BW (kg) 0.91 Odds/UNIT 0.0002 NA*
Height (cm) 0.98 Odds/UNIT 0.0003 NA*
Primary diagnosis 0.08 0.78
 Cholestasis Reference Reference
 Metabolic disorder 1.18 0.60-2.34 0.64 1.42 0.65-3.09 0.38
 Fulminant hepatitis 0.34 0.12-0.99 0.04 1.80 0.32-10.11 0.50
 Others 0.49 0.18-1.32 0.15 1.19 0.37-3.84 0.77
Pre-LT ICU admission 0.30 0.13 -0.73 0.005 0.26 0.06-1.05 0.06
Living donor (vs. Deceased donor) 2.24 0.29-17.57 0.43 0.88 0.09-8.80 0.91
Blood type compatibility 0.61 0.39
 Identical Reference Reference
 Compatible 0.71 0.35-1.42 0.33 0.77 0.36-1.64 0.49
 Incompatible 0.84 0.41-1.74 0.64 0.58 0.25-1.32 0.20
GRWR >3% 1.49 0.84-2.51 0.19 0.99 0.56-1.90 0.97
EBV R− (vs. R+) 3.31 1.28 -8.56 0.009 2.93 0.98-8.73 0.05
EBV D+ (vs D−) 1.23 0.16-10.66 0.81 2.67 0.27-26.17 0.39
CMV R− (vs R+) 1.19 0.66-2.12 0.56 1.0 0.52-1.92 0.99
CMV D− (vs D+) 1.74 1.00-3.03 0.047 2.42 1.26-4.66 0.008
First EBV DNAemia <6 mo post-LT 3.78 1.73-8.22 0.0004 4.25 1.84-9.78 0.0007
Biopsy-proven rejection <1 mo post-LT 1.32 0.74-2.35 0.35 1.39 0.74-2.62 0.31
Use of ATG <1 mo post-LT 1.00 0.33-3.02 0.99 0.72 0.21-2.45 0.59
P value for whole model test was <0.0001.
Body weight and height were significantly correlated with age (years) (R2= 0.91 and 0.92, respectively) and had to be excluded from the multivariate analysis.
ATG, antithymocyte globulin; BW, body weight; CI, confidence interval; CMV, cytomegalovirus; D, donor serology; EBV, Epstein-Barr virus; GRWR, graft-to-recipient weight ratio; ICU, intensive care unit; LT, liver transplantation; OR, odds ratio; R, recipient serology.

Because age, BW, and height showed collinearity with significant correlation, BW and height were removed from multivariate analysis, and the age was changed to categorical data (younger or older than 5 y).

As a result, age younger than 5 y (odds ratio [OR]: 8.20; 95% confidence interval [95% CI]: 1.80-37.4.0; P = 0.007), CMV D− (OR: 2.42; 95% CI: 1.26-4.66; P = 0.008), and the early detection of EBV within 6 mo post-LT (OR: 4.25; 95% CI: 1.84-9.78; P = 0.0007) were found to be significant risk factors for CHL development.

Notably, pre-LT ICU admission was found to have reduced risk and EBV R− to have increased risk for CHL in univariate analyses (OR: 0.30; 95% CI: 0.13-0.73; P = 0.005, OR: 3.31; 95% CI: 1.28-8.56; P = 0.009, respectively); however, those did not reach significance in multivariate analysis (OR: 0.26; 95% CI: 0.06-1.05; P = 0.06, OR: 2.93; 95% CI: 0.98-8.73; P = 0.05, respectively).

Incidence of EBV-associated PTLD and EBV-associated Diseases

Although 15% (61/394) of patients met the definition of CHL carrier (Table 2), no patient developed EBV-associated PTLD in this study. However, 2 PTLD cases were identified during the study period; both were excluded because of the second LT following severe rejection and PTLD developed only after the second LT. Similarly, only a few EBV-associated diseases were diagnosed in the entire cohort, 0 of 184 (0%) in low, 3 of 149 (2%) in high, and 2 of 61 (3%) in CHL, respectively,17,23,24 and this difference did not reach statistical significance (P = 0.08) (Table 2).


This is a single-center, retrospective observational study describing the kinetics of EBV loads among EBV seronegative and seropositive patients and the epidemiology and risk factors for being a CHL carrier in the context of a large, living-donor–dominant pediatric LT center in Japan.

Our study cohort has the following unique demographic characteristics: very young patient age, very high rate of living-donor LT from patients’ relatives, and a high proportion of non biliary diseases.21,25 Because of the first 2 characteristics, the EBV mismatch rate was very high, and the increased risk of EBV-associated complications was hypothesized. Consistent with many previous reports, the EBV R− state was associated with overall high EBV loads, resulting in more frequent EBV load measurements and a longer duration of monitoring and developing the CHL carrier state.12,26

We also demonstrated that the difference in the levels of EBV loads between R+ and R− became negligible, and the EBV loads in R− decreased below 1000 copies/μg DNA after 3 y post-LT. Although the frequency of EBV load measurement varies based on the patient’s condition, this observation indicated that the higher risk of EBV infection in R− patients might decrease to the level of what R+ recipients possess after several years post-LT. This observation parallels the incidence of EBV-associated PTLD that decreases after 1 to 2 y post-LT.3 Therefore, intensive EBV monitoring may be tapered off after 3 y post-LT.

Of interest, time to the first EBV DNAemia, which has been analyzed only in a few studies, did not differ between R+ and R− (Figure 3A). However, the EBV loads in R+ recipients were low overall and rarely exceeded 1000 copies/μg DNA, a threshold considered “high” in our assay. This indicated that low-level EBV reactivation among R+ recipients occurs as early and frequently as EBV infection in R− recipients, whereas EBV loads in R− recipients trended almost exclusively higher than those in R+ recipients.

Besides that, the patients who developed the first EBV DNAemia early (within 6 mo post-LT) showed significantly higher maximum EBV load values in each patient category. Considering the need for intensive immunosuppression within the first 6 mo post-LT, it is plausible that the early EBV DNAemia results in a higher set point of EBV loads and their persistence. In fact, the multivariate analysis confirmed that the first EBV DNAemia <6 mo post-LT was found to be an independent risk factor for CHL development.

Although young age (<5 y) has been a well-accepted risk factor for EBV-associated complications and PTLD, this study revealed several unreported factors that can be applied for better risk stratification. For example, in a previous study, CMV D− status was reported as a potential risk factor for PTLD27 and has now been rediscovered as a risk factor for CHL development. Given that most post-LT EBV infections occur via graft and CMV reactivation in the graft is known to cause bystander inflammation, the CMV D+ graft may prevent the increase of EBV loads because of a CMV-driven inflammatory milieu. This phenomenon can be understood by the indirect effect of CMV infection, which may even increase the risk of rejection for the same reason.28,29

Pre-LT ICU admission was only proven to reduce the risk of CHL development in univariate analysis, although not in multivariate analysis (Table 3). Of interest, most patients with severe end-stage liver disease in the ICU require treatment by hemodialysis, plasmapheresis, and intravenous gamma globulin infusion, which includes both EBV and CMV antibodies. Although the detailed treatment records before 2014 could not be recovered in this study, and thus we could not evaluate this possibility, the presence of EBV and CMV antibodies might have contributed to better EBV control. In fact, intravenous gamma globulin infusions used to be an optional treatment or prophylaxis against posttransplant CMV and EBV infections, as indicated in a few studies.30,31

EBV R− status, believed to be the most important factor contributing to EBV complications after organ transplantation, was not firmly demonstrated to be a risk factor in multivariate analysis (P = 0.05). The definition of EBV R− in this study and the very high proportion of R− likely explain this. Because we adopted the clinical definition determining the EBV serology and because the recipients younger than 1 y were categorized as R− unless proven otherwise, the R− state was intrinsically confounded with age, which is also a known risk factor for CHL and PTLD. Accordingly, the very high R− rate (313/394, 79%) might have decreased the statistical power. Thus, as other studies have reported, we still suggest considering EBV R− an important risk factor for CHL development.32

The limitations of this study come from the nature of the retrospective study in a single center. This includes less heterogeneous viral strains and donor-recipient HLA patterns. However, the robust dataset from a large number of pediatric living-donor LT recipients and the uniform clinical practice procedures might have contributed to the discovery of new findings. In addition, regarding the role of EBV monitoring, the risk of being a CHL carrier in the light of PTLD development was not assessed because of the extremely low incidence of PTLD (0/394 and 2/525). Likewise, the impact of being a CHL carrier on EBV-associated disease was not fully evaluated because of the low numbers. Although this is consistent with previous CHL studies, particularly in LT recipients,10 the low incidence of PTLD might have been achieved because of our unique situation: because parental LT, which is the most common form in Japan, may have higher immunotoleratogenicity than LTs from unrelated donors,33,34 we were able to reduce the levels of immunosuppression more aggressively without remarkable adverse events such as rejection in maintenance phase.

Nonetheless, the current clinical practice appears effective in preventing PTLD development.

In conclusion, although EBV DNAemia frequently occurred similarly in both R− and R+ in the early post-LT period, the EBV loads were higher in R− recipients in the first 3 y and became indistinguishable thereafter, suggesting the particular importance of EBV load monitoring within the first 3 y. The LT recipients with young age, negative EBV serology, and early EBV detection within 6 mo post-LT who received an organ from CMV seronegative donors should be closely monitored as to developing the CHL carrier state and should be treated proactively to prevent PTLD.


The authors thank technical staff, Fuyuko Kawano, Etsuko Takahashi, Yuki Yoshikawa, Ayaka Shiraga, Yuya Hayashi, Yuko Sakimoto, and Dr. Michael Green and Dr. Marian Michaels at the Children’s Hospital of Pittsburgh for constructive article review.


1. Green MD, Michaels MG. Opportunistic infections in liver and intestinal transplantation. In: Cherry JDMDM, Harrison GJMD, Kaplan SLMD, Steinbach WJMD, Hotez PJMDP, eds. Feigin and Cherry’s Textbook of Pediatric Infectious Diseases. 2019:705–715.e702.
2. Knipe DM, Howley PM. Fields Virology. 6th ed. Philadelphia, PA: Wolters Kluwer/Lippincott Williams & Wilkins Health; 2013.
3. Yamada M, Nowalk A, Green M. Epstein-Barr virus and posttransplant lymphoproliferative disorders. In: Emerging Transplant Infections. 2020:1–29.
4. Wadowsky RM, Laus S, Green M, et al. Measurement of Epstein-Barr virus DNA loads in whole blood and plasma by TaqMan PCR and in peripheral blood lymphocytes by competitive PCR. J Clin Microbiol. 2003;41:5245–5249.
5. Kanakry J, Ambinder R. The biology and clinical utility of EBV monitoring in blood. Curr Top Microbiol Immunol. 2015;391:475–499.
6. Kanakry JA, Hegde AM, Durand CM, et al. The clinical significance of EBV DNA in the plasma and peripheral blood mononuclear cells of patients with or without EBV diseases. Blood. 2016;127:2007–2017.
7. Allen UD, Preiksaitis JK, Practice ASTIDCo. Post-transplant lymphoproliferative disorders, Epstein-Barr virus infection, and disease in solid organ transplantation: guidelines from the American society of transplantation infectious diseases community of practice. Clin Transplant. 2019;33:e13652.
8. Narkewicz MR, Green M, Dunn S, et al.; Studies of Pediatric Liver Transplantation Research Group. Decreasing incidence of symptomatic Epstein-Barr virus disease and posttransplant lymphoproliferative disorder in pediatric liver transplant recipients: report of the studies of pediatric liver transplantation experience. Liver Transpl. 2013;19:730–740.
9. Lee TC, Savoldo B, Rooney CM, et al. Quantitative EBV viral loads and immunosuppression alterations can decrease PTLD incidence in pediatric liver transplant recipients. Am J Transplant. 2005;5:2222–2228.
10. Green M, Soltys K, Rowe DT, et al. Chronic high Epstein-Barr viral load carriage in pediatric liver transplant recipients. Pediatr Transplant. 2009;13:319–323.
11. Kullberg-Lindh C, Saalman R, Olausson M, et al. Epstein-Barr virus DNA monitoring in serum and whole blood in pediatric liver transplant recipients who do or do not discontinue immunosuppressive therapy. Pediatr Transplant. 2017;21.
12. Bingler MA, Feingold B, Miller SA, et al. Chronic high Epstein-Barr viral load state and risk for late-onset posttransplant lymphoproliferative disease/lymphoma in children. Am J Transplant. 2008;8:442–445.
13. Lau AH, Soltys K, Sindhi RK, et al. Chronic high Epstein-Barr viral load carriage in pediatric small bowel transplant recipients. Pediatr Transplant. 2010;14:549–553.
14. Yamada M, Nguyen C, Fadakar P, et al. Epidemiology and outcome of chronic high Epstein-Barr viral load carriage in pediatric kidney transplant recipients. Pediatr Transplant. 2018;22:e13147.
15. Moudgil A, Martz K, Moore T, et al. Significance of asymptomatic persistent Epstein-Barr viral load in pediatric renal transplant recipients: North American pediatric renal trials and collaborative studies report. Open Urol Nephrol J. 2014;7:123–128.
16. Kamei H, Ito Y, Kawada J, et al. Risk factors and long-term outcomes of pediatric liver transplant recipients with chronic high Epstein-Barr virus loads. Transpl Infect Dis. 2018;20:e12911.
17. Shigeta T, Imadome K, Sakamoto S, et al. Epstein-Barr virus infection after pediatric living-related liver transplantation--management and risk factors. Transplant Proc. 2010;42:4178–4180.
18. Imadome K, Fukuda A, Kawano F, et al. Effective control of Epstein-Barr virus infection following pediatric liver transplantation by monitoring of viral DNA load and lymphocyte surface markers. Pediatr Transplant. 2012;16:748–757.
19. Fryer JF, Heath AB, Wilkinson DE, et al. A collaborative study to establish the 1st WHO International Standard for Epstein-Barr virus for nucleic acid amplification techniques. Biologicals. 2016;44:423–433.
20. Saitoh A, Sakamoto S, Fukuda A, et al. A universal preemptive therapy for cytomegalovirus infections in children after live-donor liver transplantation. Transplantation. 2011;92:930–935.
21. Kasahara M, Sakamoto S, Fukuda A. Pediatric living-donor liver transplantation. Semin Pediatr Surg. 2017;26:224–232.
22. Kwong A, Kim WR, Lake JR, et al. OPTN/SRTR 2018 annual data report: liver. Am J Transplant. 2020;20:193–299.
23. Yokoyama S, Kasahara M, Fukuda A, et al. Epstein-Barr virus-associated erythema nodosum after living-donor liver transplantation: a case report. Liver Transpl. 2009;15:446–448.
24. Yokoyama S, Kasahara M, Fukuda A, et al. Evans syndrome after successful living-donor liver transplantation for neonatal giant cell hepatitis. Transplantation. 2007;84:798–799.
25. Kwong AJ, Kim WR, Lake JR, et al. OPTN/SRTR 2019 annual data report: liver. Am J Transplant. 2021;21:208–315.
26. Allen UD, Preiksaitis JK, Practice ASTIDCo. Post-transplant Lymphoproliferative disorders, EBV infection and disease in solid organ transplantation: guidelines from the American society of transplantation infectious diseases community of practice. Clin Transplant. 2019; 33:e13652.
27. Green M, Michaels MG, Katz BZ, et al. CMV-IVIG for prevention of Epstein Barr virus disease and posttransplant lymphoproliferative disease in pediatric liver transplant recipients. Am J Transplant. 2006;6:1906–1912.
28. Kotton CN, Kumar D, Caliendo AM, et al. The third international consensus guidelines on the management of cytomegalovirus in solid-organ transplantation. Transplantation. 2018;102:900–931.
29. Razonable RR, Humar A. Cytomegalovirus in solid organ transplant recipients-Guidelines of the American Society of Transplantation Infectious Diseases Community of Practice. Clin Transplant. 2019;33:e13512.
30. King SM, Superina R, Andrews W, et al. Randomized comparison of ganciclovir plus intravenous immune globulin (IVIG) with IVIG alone for prevention of primary cytomegalovirus disease in children receiving liver transplants. Clin Infect Dis. 1997;25:1173–1179.
31. Krampe K, Briem-Richter A, Fischer L, et al. The value of immunoprophylaxis for cytomegalovirus infection with intravenous immunoglobulin in pediatric liver transplant recipients receiving a low-dose immunosupressive regimen. Pediatr Transplant. 2010;14:67–71.
32. Kim WR, Lake JR, Smith JM, et al. OPTN/SRTR 2017 annual data report: liver. Am J Transplant. 2019;19:184–283.
33. Kim MH, Akbari O, Genyk Y, et al. Immunologic benefit of maternal donors in pediatric living donor liver transplantation. Pediatr Transplant. 2019;23:e13560.
34. Feng S, Ekong UD, Lobritto SJ, et al. Complete immunosuppression withdrawal and subsequent allograft function among pediatric recipients of parental living donor liver transplants. Jama. 2012;307:283–293.

Supplemental Digital Content

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.