Results of single-center screening for chronic hepatitis E in children after liver transplantation and report on successful treatment with ribavirin


Norman Junge, Children's Hospital, Hannover Medical School, Pediatric Gastroenterology and Hepatology OE 6728, Carl Neuberg Street 1, 30625 Hannover, Germany

Tel.: +49 176 1532 8630

Fax: +49 511 532 3294



RNA screening for HEV in 22 liver-transplanted children with chronic graft hepatitis out of a cohort of 267 liver-transplanted children detected a single patient with chronic HEV infection. Although this patient remained viremic for 33 months, anti-HEV-IgG was not detectable with MP assay but with Wantai assay. We present the first case of successful ribavirin therapy in an immunosuppressed child with chronic HEV infection. In conclusion, chronic HEV infection in immunosuppressed children may not be detectable employing serological assays. Therefore, the most reliably screening method is screening for HEV-RNA. Chronic HEV infection in children can successfully be treated with ribavirin.


acute cellular rejection


biliary atresia


chronic hepatitis due to HEV


hepatitis E virus




international accepted fibrosis score


orthotopic liver transplantations


polymerase chain reaction


patient with chronic hepatitis E virus infection.

Presently, four human pathogen genotypes of HEV are known [1]. In developed countries, genotype 3 with assumed zoonotic transmission is identified as the main cause for autochthonous infections [1]. Cases of cHEV were first described in immunocompromised adults in 2008 [2]. The first case report of an immunocompromised child with cHEV was published in 2011 [3], and a second case was revealed by a recent study [4]. However, in this series, serological analyses for anti-HEV IgG and IgM were carried out by a commercial immunoblot assay (RecomLine), and only anti-HEV-IgG positive children (n = 4) were tested for presence of HEV-RNA. Significant discrepancies between different anti-HEV IgG assays and a greater sensitivity of Wantai assay compared with immunoblot assay were described [5].

Therefore, in our study, we used HEV-RNA detection independently of serological results to determine the frequency of cHEV in our pediatric liver transplant population. Because the aim of our study was to detect clinical significant cHEV rather than to evaluate seroprevalence or disease prevalence, we analyzed HEV-RNA only in patients with clinical and laboratory signs of active hepatitis. HEV IgG and IgM antibodies were determined only in patients shown to be viremic. Furthermore, we describe the first case of successful treatment with ribavirin for cHEV in this population.


  1. Medical charts of 267 liver-transplanted children were screened retrospectively for unexplained chronic elevation of liver enzymes and recurrent, chronic, or refractory rejection. Patients thus identified were screened for HEV-RNA by PCR as described previously [6, 7]. Stored frozen plasma samples were used for analysis in a few cases.
  2. In the single-identified HEV-RNA positive patient, seven stored frozen plasma samples from before and six samples since diagnosis of cHEV were tested for HEV-RNA to characterize the clinical course of infection. Testing for HEV-specific antibodies was carried out by MP assay (MP Biomedicals, formerly Genelabs Diagnostics, Singapore) and additionally for three samples by Wantai Assay (Wantai Biologic Pharmacy Enterprise, Beijing, China).


Screening results

Twenty-two patients (11 female) fulfilled the above-mentioned criteria. Median age at screening was 6.7 yr (1.4–17.2). Patients had undergone a total of 28 OLT at a median age of 1.4 yr (0.4–16.7). Median follow-up was 50.3 months (4.22–144). Patients had experienced a total of 52 acute and two chronic rejections. Liver enzymes were elevated within a wide range during follow-up (Table 1). Of 22 patients, one patient with cHEV and over 33-month persistent HEV viremia (genotype 3) was identified.

Table 1. Clinical data and liver enzyme profile of study patients (n = 22)
Patient No.DiseaseFollow-up in monthsAmount OLTAcute rejectionChronic rejectionALT meanAST meanGLDH meanISHAK-F
  1. Follow-up, retrospectively from two wk after OLT to last visit in our clinic; ALT mean, mean alanine transaminase levels in μkatal/L during follow-up; AST mean, mean aspartate aminotransaminase levels in μkatal/L during follow-up; GLDH mean, mean glutamate dehydrogenase levels in μkatal/L during follow-up; ISHAK F, international accepted fibrosis score (F0, no fibrosis; F5, incomplete cirrhosis; F6, definite cirrhosis); n.e., not evaluated; CF, cystic fibrosis; PSC, primary sclerosing cholangitis; ALF, acute liver failure; Hepatobl., hepatoblastoma; PiZZ, alpha 1 antitrypsin deficiency; PFIC, progressive familiar intrahepatic cholestasis; P-HEV during cHEV, mean AST, ALT, GLDH levels of P-HEV from assumed time of infection until disappearing of viremia.

3PFIC II16.662503.651.980.232
22 P-HEVBA42.683402.632.880.125
22P-HEV during cHEV 34   4.624.70.08 
Overall mean 
Overall median 
Overall s.d.     2.382.40.68 
Pat.1-21 mean (without P-HEV)     1.751.580.39 
Pat.1-21 median (without P-HEV) 
Pat.1-21 s.d. (without P-HEV)


The P-HEV is a 10-yr-old boy with BA. He underwent liver transplantation at eight months of age and was retransplanted at age seven yr for chronic graft failure, followed by a third transplantation shortly afterward for primary graft failure of unknown origin. His IS contained tacrolimus, mycophenolate mofetil, and pre-dnisolone (for details, please see Data S1). After a difficult post-transplant course following the third OLT, liver enzymes normalized within three months. One month later, liver enzyme levels increased and stayed elevated (Fig. 1). Several liver biopsies showed features of ACR. Immunosuppression was steadily increased to include high-dose prednisolone, sirolimus and finally Campath (for details, please see Data S1). However, no stabilization of transaminase levels was achieved. In subsequent liver biopsies, features compatible with chronic viral hepatitis rather than ACR became increasingly prominent and hepatic fibrosis increased to an ISHAK score of F5. Diagnostic workup for chronic graft hepatitis was repeated. Anti-HEV-IgG (assessed by MP assay) remained negative, but HEV-RNA analysis revealed HEV viremia, which was confirmed in control samples and by positive HEV-RNA in stool. As a first therapeutic approach, IS was reduced. However, both viremia and raised liver enzyme levels persisted over the next two months. We therefore decided to start ribavirin at a dose of 400 mg (15 mg/kg bodyweight) per day for six months. From 42 days after initiation of ribavirin therapy, both stool and blood were tested negative for HEV-RNA, and liver enzymes decreased to normal levels (Fig. 1).

Figure 1.

Course of aspartate aminotransaminase (AST), alanine transaminase (ALT), and glutamate dehydrogenase (GLDH) levels, viremia and HEV-specific antibodies in patient with chronic HEV infection. @anti-HEV IgG assessed by MP assay, in contrast to these results, we could detect anti-HEV IgG using Wantai assay in three randomly and retrospectively tested samples from 2010 to 2011. n.t., non-tested; pos, positive; pos*, positive in retrospective screening; neg, negative.

Overall, therapy with ribavirin was tolerated well by P-HEV. The main side effects reported for ribavirin are anemia caused by hemolysis, neutropenia secondary to myelotoxicity, nephrotoxicity, and hypo- as well as hyperthyroidism. Our patient showed a decrease in hemoglobin from 9.6 g/dL to a minimum of 6.7 g/dL during therapy, but without any obvious signs of hemolysis (LDH 3.12–3.95 μkatal/L; haptoglobin 0.26 g/L; free hemoglobin 37 mg/L), blood transfusion became necessary. As reticulocyte count was already elevated (46/nL before ribavirin treatment, 175/nL under treatment), we did not initiate a treatment with erythropoietin. P-HEV already had pancytopenia secondary to hypersplenism before starting ribavirin. Leukocyte count remained stable at around 900/μL during therapy. Kidney function of P-HEV was impaired because third OLT with a glomerular filtration rate calculated based on cystatin C of slightly above 50 mL/min. During treatment, we could observe a short drop down to 40 mL/min, but with fast recovery to old values without any change in ribavirin treatment. Levels of thyroid stimulating hormone remained unchanged in P-HEV.

Anti-HEV-IgG (assessed by MP assay) was detectable for the first time six months after initiation of ribavirin therapy. Ribavirin treatment was discontinued three months ago, and up to now, we could not observe any relapse or recurrence of viremia. Since discontinuation, hemoglobin increased without treatment to 14.2 g/dL, white blood cell count to 2500/μL, and kidney function improved to 72 mL/min.

To determine the time of infection and way of transmission, we retrospectively analyzed stored frozen plasma samples (Fig. 1). Presence of HEV-RNA could first be demonstrated six months after the third OLT, eight days after discharge from hospital. Retrospectively, liver enzymes started to increase one month before discharge. This therefore appears to be the most likely time of infection. Within this time, P-HEV received 105 blood products. As blood samples from blood donors were stored for three yr only, a retrospective analysis for HEV-RNA was not possible in these samples. Transmission from the second liver graft can be ruled out based on retrospective screening results from P-HEV. In contrast, transmission from the third liver graft could not be excluded, as no back up blood samples from the organ donor exist, and the left liver lobe was allocated to another center, but was not transplanted or histological examined.

Interestingly, although the patient remained viremic for HEV in blood and stool for more than 33 months, his immunocompetent healthy parents have never shown any signs of hepatitis and have tested negative for anti-HEV-IgG and IgM, indicating that person-to-person transmission did not occur.

Retrospective testing of three serum samples (from 2010 and 2011) from P-HEV with Wantai assay revealed anti-HEV-IgG positivity in all of them.


The screening of 267 liver-transplanted children from our center for clinical and laboratory signs of active chronic graft hepatitis without known cause identified 22 patients. Screening of these 22 patients for HEV-RNA could detect one patient with chronic HEV.

Our study demonstrates the significance of cHEV as a rare, but important differential diagnosis for chronic graft hepatitis, because it represents a treatable disease, which could lead to graft cirrhosis and graft loss, if untreated. As histological differentiation between cHEV and ACR can be difficult, blood tests for HEV should be considered. As we and other authors [1, 5, 6] could show screening for HEV-RNA appears mandatory in immunocompromised patients given that HEV antibodies could be insufficient to exclude HEV infection because significant discrepancies between commercial available test assays exist. To our knowledge, the presented case is the longest published course of cHEV without detectable anti-HEV-IgG tested by MP assay. This phenomenon is in line with previous reports describing PCR as superior to serology for detection of acute HEV infection and chronic HEV infection in immunosuppressed patients [8, 9]. As it is shown that different anti-HEV assays can have significant discrepancies [5], we retrospectively tested serum from P-HEV additionally with Wantai assay for anti-HEV-IgG and could detect positivity. This phenomenon is in line with a recent study from Rossi-Tamisier [5] and suggests that the Wantai assay is superior to the MP assay. To prove this hypothesis, further studies are necessary. Independently of this, HEV-RNA screening is always superior to all anti-HEV assays. Even though HEV-RNA screening in stool is more sensitive than in blood, to our knowledge, there is no described case of a solid organ transplant recipient having HEV without detectable HEV-RNA in blood. Also our experiences with 36 adult organ transplant recipients having HEV show no patient with negative HEV-RNA PCR in blood.

We describe a clinical course of cHEV with viremia over 33 months, associated with histological signs of hepatitis and progressive fibrosis, but without detectable anti-HEV-IgG assessed by MP assay. In this patient, treatment with ribavirin led to clearance of viremia within 42 days.

Reduction of IS is described as reliable therapy for cHEV [1]. In our patient, this treatment failed to achieve improvement. In adults with cHEV, the efficacy of therapy with ribavirin has been demonstrated [7, 10, 11]. We therefore initialized treatment with ribavirin in our patient and could observe impressive success with dis-appearance of viremia 42 days after initiation of therapy, seroconversion to HEV IgG positivity (assessed with MP assay) six month later, and relapse-free follow-up over three months after discontinuing ribavirin treatment.

To date, it is still unsure whether giving more than a three-month therapy can improve the virological response, but because ribavirin treatment was tolerated well in our patient and given that cases with relapse of HEV infection after a three-month course of ribavirin treatment [11] and cases of development of treatment resistance were described [7], we decided to treat our patient with ribavirin for six months.

Three months after discontinuing ribavirin treatment, blood count and kidney function have improved to values even better than before ribavirin treatment. Therefore, we assume that blood count and kidney function were also impaired by cHEV infection. HEV-associated glomerulonephritis as an HEV-related extrahepatic manifestation was previously described by Kamar et al. [12].

One question we were unable to clarify is how the patient acquired HEV in the first place. As P-HEV was hospitalized due to third OLT at the presumed time of infection, a direct zoonotic transmission is unlikely. Fecal-oral transmission between humans can almost be excluded in industrialized nations. Therefore, foodborne or bloodborne transmission is most likely. As 10% of fresh frozen plasma pools in Germany contain HEV-RNA [13], transmission from a blood product out of 105 blood products P-HEV received is likely, but given that P-HEV favors minced meat products a zoonotic foodborne transmission is also conceivable.


In our cohort of liver-transplanted children (n = 267), chronic HEV is a rare but important differential diagnosis for graft hepatitis in children. Clarification should be sought with a high level of suspicion, because chronic HEV can lead to cirrhosis and graft loss if not treated. Screening for HEV-RNA appears mandatory, as testing for HEV antibodies does not reliably rule out the disease in immunocompromised patients as significant discrepancies between commercially available test assays exist. In cases where reduction of IS is insufficient for viral clearance, ribavirin seems to be a reliable alternative.


We thank A. Heim, Institute of Virology, Medical School Hannover, for providing frozen plasma samples, and we thank J. Schlue, Institute of Pathology, Medical School Hannover, for discussing P-HEV's liver biopsies.

Authors' contributions

Sven Pischke, MD: Concept/Design; Ulrich Baumann, MD: Critical revision of article; Imeke Goldschmidt, MD: Critical revision of article; Michael Manns, MD: Providing laboratory facilities; Heiner Wedemeyer, MD: Critical revision of article; Eva-Doreen Pfister, MD: Concept/Design.