Objectives: Human β-herpes viruses, including cytomegalovirus and human herpesvirus 6, can become activated in liver transplant patients. Here, we evaluated the effects of human herpesvirus 6 infection as an independent factor on cytomegalovirus infection and the occurrence of associated diseases in liver transplant patients.
Materials and Methods: In this cross-sectional study, 46 patients who underwent deceased-donor liver transplant at Nemazi Hospital, Shiraz University of Medical Sciences (Shiraz, Iran) were prospectively monitored for cytomegalovirus and human herpesvirus 6 infections during 3 months posttransplant. Taq-man real-time polymerase chain reaction assay as an accurate and rapid test and pp65-anigenemia as the standard test were used to monitor cytomegalovirus infections, whereas human herpesvirus 6 infection was monitored by Taq-man real-time polymerase chain reaction assay. We also followed clinical findings from laboratory data and symptoms of cytomegalovirus infections.
Results: Active cytomegalovirus infection was detected in 23 liver transplant recipients (50%), of which 17 (74%) were diagnosed with cytomegalovirus-related diseases. Active human herpesvirus 6 infection was detected in 25 patients (54%). Thirteen of 17 cytomegalovirus-symptomatic patients had coinfection with human herpesvirus 6. In 10 of the 13 patients with coinfection, human herpesvirus 6 DNAemia appeared significantly earlier by 9 days than cytomegalovirus infection. In the pp65 antigenemia test, the mean number of cytomegalovirus-infected polymorphonuclear cells was 42.47 ± 5.41, which was correlated with incidence of clinical presentation. In symptomatic patients, average serum and polymorphonuclear cell viral loads of cytomegalovirus were 12064.59 copies/mL and 6735 copies/2 × 105 cells, respectively, with significant differences between the loads and cytomegalovirus-attributable disease symptoms. Average human herpesvirus 6 DNA burden in serum samples of symptomatic patients was 11283 copies /mL, which was statistically related to cytomegalovirus-attributable disease symptoms.
Conclusions: We found that human herpesvirus 6 infection is often associated with cytomegalovirus reactivation and cytomegalovirus-attributable disease symptoms.
Key words : Early infections, Immunosuppression, Polymerase chain reaction
Cytomegalovirus (CMV) and human herpesvirus-6 (HHV-6) are members of the beta-herpesvirus subfamily. Generally, these viruses have high seroprevalence and vary based on socioeconomic status. After primary infection, both CMV and HHV-6 infections remain latent in the endoplasmic reticulum system and may be reactivated intermittently depending on the host immune status.
The incidence of CMV and HHV-6 infections among liver transplant patients ranges widely, between 14% and 82%1,2 and 11% and 72%,3,4 respectively. The different variations could be due to different ages, use of different CMV prophylactic regimens, and different sensitivities of diagnostic tests.
The 2 variants of HHV-6 are HHV-6A and HHV-6B, with the latter one being the most prevalent. It has been shown that HHV-6 can accelerate the course of other viral infections, such as HHV-7, human immunodeficiency virus, and CMV.5 Subsequent to HHV-6 immunomodulatory effects and/or direct CMV interactions, CMV-associated diseases can develop.6 Human herpesvirus 6 infections typically occur during the first 2 to 6 weeks after liver transplant, whereas CMV reactivation occurs 3 to 10 weeks after liver transplant, when the level of immunosuppression is the most intense. The most common clinical presentations of both viral agents in posttransplant patients include fever, neutropenia, central nervous system manifestations, and other visceral involvements.7
Reactivation and reinfection of HHV-6 usually occur at earlier stages of immunosuppressive administration, compared with other opportunistic infections.8 Therefore, we evaluated the burden and potency of HHV-6 in reactivation, development of rejection, and the occurrence of diseases due to CMV infection among liver transplant recipients.
Materials and Methods
This prospective study included 46 patients who underwent deceased-donor liver transplant at the Nemazi Hospital (Shiraz University of Medical Sciences, Shiraz, Iran). Our ethics committee approved the study design and procedures, and written informed consent was obtained from guardians or the patients. The study was carried out between April 2015 and April 2017. In total, 493 clotted blood samples (to collect serum samples) and 493 corresponding oxalated samples (to isolate polymorphonuclear [PMN] cells; average of 10.7 samples per patient) were obtained weekly from liver recipients for up to 3 months. Patients who died during the follow-up period were excluded.
Study patients received triple drug therapy of various combinations of tacrolimus or cyclosporine, mycophenolic acid, and corticosteroids as anti-rejection treatment posttransplant. Patients were regularly monitored for CMV and HHV-6 infection during the first 3 months after transplant.
We collected demographic data, time of liver transplant, serum levels of CMV and HHV-6 pretransplant, clinical findings, CMV and HHV-6 viral genome burden, CMV antigenemia results, and drug regimens from electronic patient records during our study period. Clinical presentations and laboratory findings were reviewed, and all HHV-6 and CMV-attributable disease symptoms were considered.
Cytomegalovirus and human herpesvirus 6 serological tests
Serum samples from 46 patients before liver transplant were stored in 2.5-mL aliquots at -70°C; these samples were used for analyses of anti-CMV and anti-HHV-6 using commercially available kits (Diapro, Milan, Italy)
Serum DNA extraction
DNA was extracted from 200 μL of serum using a phenol-chloroform protocol after treatment with lysis buffer containing sodium dodecyl sulfate, followed by DNA precipitation with cold ethanol. The resulting DNA pellet was eluted in 50 μL of Tris-EDTA buffer. We used 10 μL of extracted DNA as templates for the CMV and HHV-6 Taq-man real-time polymerase chain reaction (PCR) assays.
Peripheral blood leukocyte DNA extraction
Stored isolated PMN cells (-70°C) from 3 mL of oxalated blood were incubated for 1 hour with 5 μL (200 mg/μL) of proteinase K at 65°C. This process was used for serum DNA extraction.
Cytomegalovirus and human herpesvirus 6 DNAemia monitoring
Taq-man real-time PCR was used to quantify CMV and HHV-6 DNA duration after liver transplant. Amplification was carried out using a commercial kit from Primerdesign (Camberley, UK) on serum and PMN samples for CMV DNAemia monitoring and serum specimens for HHV-6 DNAemia follow-up, using 10 μL of the extracted DNA. The method was performed in accordance with the manufacturer’s protocol.
For diagnosis of CMV infections, pp65-antigenemia assay (CMV Brite Turbo kit, Immuno Quality Products, Groningen, The Netherlands) was used, according to previously published procedures.9 Results are expressed as number of positive cells per 2 ×105 leukocytes.
Data were collected in patients by direct gene counting. Statistical evaluation was performed with SPSS software (SPSS: An IBM Company, version 19, IBM Corporation, Armonk, NY, USA). The frequencies of data were compared in patients versus controls by chi-square test and t test. P < .05 was considered statistically significant.
Patients and samples
Surveillance of CMV and HHV-6 infections in transplant recipients was performed on consecutive 493 serial blood specimens provided weekly before and after transplant. The age range of the 46 deceased-donor liver transplant recipients included in our study was 1.7 to 65 years (mean 25.39 ± 2.6 y; median 22.5 y). Twenty-five of those in our study group (54.3%) were male. Most of the liver transplant recipients were in the 21- to 25-year and 1- to 5-year age ranges respectively.
Of total study patients, 17 (37%) showed CMV-associated illness (8 males and 9 females); differences in terms of biological sex and illness were not significantly different (P = .21). Most symptomatic CMV-associated patients were in the 1- to 5-year age range (Figure 1).
Cytomegalovirus and human herpesvirus 6 seroprevalence and correlation with
As shown in Table 1, immunoglobulin G (IgG) antibodies to CMV and HHV-6 were detected in 42 (91.3%) and 41 (89.1%) of the 46 liver recipients before transplant, respectively. Nearly 37% (17/46) of patients who underwent liver transplant presented with CMV-related diseases. Among symptomatic patients, 13 (76.5%) were seropositive for both CMV and HHV-6 infections.
Our statistical analysis of pretransplant HHV-6 serostatus in the different age groups showed a positive correlation between lower age and HHV-6 seropositivity (P = .013). In 4 CMV IgG-negative symptomatic patients, a positive correlation was also seen between serostatus and appearance of clinical findings (P = .015)
The pp65-antigenemia test for CMV infection was positive in half of our liver transplant patients (n = 23), with 17 of these patients (73.9%) having CMV-associated diseases and 6 (21.1%) without any symptoms of CMV diseases.
Levels of CMV pp65-antigenemia infected PMN cells were 7 to 72 and 0 to 12 in symptomatic and asymptomatic patients, respectively, showing significant differences between these groups (P = .001). However, no significant correlation was found between CMV pp65-antigenemia positivity and age (P = .35).
Cytomegalovirus viral load in polymorphonuclear cells and serum samples
Serum and PMN cell samples were quantified by the Taq-man real-time PCR assay. Cytomegalovirus DNA was detected in serum and PMN cells of 50% of liver transplant recipients, including 17 symptomatic patients.
The mean CMV load in the serum samples of symptomatic patients was 12 064.59 ± 43 copies/mL, whereas mean CMV load was 215.68 ± 92.51 copies/mL in asymptomatic patients, showing significant differences between these groups (P = .008). The mean values of CMV DNA burden of PMNs were 56 735.56 ± 7797 copies/2 × 105 cells and 195.56 ± 2630 copies/2 × 105 cells in symptomatic and asymptomatic liver transplant recipients, respectively. A positive correlation was seen between clinical signs and antigenemia levels in these patients (P = .001). In addition, a statistically significant relationship was observed between antigenemia burden and serum sample levels (P = .008).
Human herpesvirus 6 viral load in serum samples
Table 2 shows that 25 liver transplant recipients (54.3%) presented with HHV-6 infection after transplant, with 14 of these patients having CMV coinfection.
The mean HHV-6 viral load in symptomatic patients was 11 283.29 ± 2326.48 copies/mL, whereas viral load was 1779.27 ± 680.39 copies/mL in asymptomatic patients.
Our statistical analyses revealed that there was a positive correlation between clinical evidence of exposure and HHV-6 positivity rate (P = .02).
Timeline of cytomegalovirus and human herpesvirus 6 infections after liver
The average time to becoming CMV positive was 12 to 60 days after transplant in symptomatic patients, whereas the average time was 73 days in asymptomatic patients.
Among 10 of 13 symptomatic patients, HHV-6 was reactivated before CMV infection. On average, CMV reactivated at 28.60 ± 4.53 days and HHV-6 reactivated at 19.86 ± 4.59 days posttransplant in these patients. In symptomatic patients, the time range for CMV and HHV-6 positivity after quantitative PCR analyses was 14 to 60 days and 9 to 56 days, respectively, which demonstrated a statistically significant difference (P = .001).
This cross-sectional study revealed that, in liver transplant recipients, HHV-6 infection is associated with an increased risk of CMV recurrence and the development of CMV-associated diseases. In our study, most symptomatic patients ranged from 1 to 5 years old. It could be argued that negative serology played the main role; however, our statistical analysis did not show any significant difference between varying age groups (P = .35).
Given that most of the liver grafts were from donors after brain death from all over Iran, the CMV serostatus of donors and recipients was unclear. However, in our previous studies, we reported that 96% and 100% of donors and recipients, respectively, were CMV IgG positive pretransplant.9,10 We also showed that 91.3% of liver recipients were CMV seropositive, whereas HHV-6 seropositivity rate was 89.1%.
Of note, the HHV-6 seronegativity rate in adult solid-organ transplant recipients is low (3.6%), whereas it is relatively high in pediatric transplant recipients.11,12 Studies have indicated reactivation of HHV-6 in intensely immunosuppressed patients, with HHV-6 reactivation also demonstrated in some sporadic immunocompetent individuals.13,14
To detect HHV-6 genomes, different patient samples can be utilized. In Japan and North America, plasma is used for HHV-6 quantification assays, whereas European countries use quantification assays on whole blood specimens.
On the basis of type of CMV prophylaxis, reactivation of HHV-6 after liver transplant has been shown to vary from 4.3% to 54.4%.15-19 In liver transplant recipients, detection rates of HHV-6 DNA in plasma and PMN cells were the same (33% and 34%), whereas in whole blood samples, rates were somewhat lower (19%).11 We found that 25 of our liver transplant recipients (54.3%) exhibited HHV-6 infection during the immunosuppression period posttransplant, similar to that shown by Humar and colleagues.19
Human herpesvirus 6 viral loads vary based on CMV prophylaxis and type of patient sample. In studies from Feldstein and colleagues and Ono and colleagues, viral loads were 25 copies/mL and 1.4 to 2.8 copies/mL in plasma samples.16,20 In Chevret and associates21 and Buyse and associates,22 viral loads were 58 to 10 528 copies/106 and 2.97 to 32 copies/106 in PMN cells. In our study, the HHV-6 viral loads were 11 283 ± 2326 copies/mL in serum samples (median 9806 copies/mL) of CMV symptomatic patients and 1779 ± 680 copies/mL in serum samples of asymptomatic patients.
Harma and associates6 revealed that CMV-related diseases were diagnosed in 42% (13/31) of liver transplant recipients, which is in agreement with our present study. However, Lautenschlager and colleagues detected CMV-associated symptoms disease in 56% of liver transplant recipients.23 The slight difference may be due to duration of patient monitoring, the type of CMV prophylaxis, and pretransplant CMV serostatus of donors and recipients.
The peaks of CMV viral load were 34 210 and 1020 copies/mL in symptomatic and asymptomatic patients, respectively, according to Harma and associates.6 The corresponding values in our study were 1206 and 215 copies/mL in sera samples, which were lower.
We found that the burden of CMV in PMNs was 56 735 and 2630 copies/2 × 105 cells. The CMV viral load was statistically correlated with clinical presentation (P = .05), as also shown in the study from Harma and colleagues6 (P = .008).
We and other studies revealed that HHV-6 is reactivated before CMV infection.6,18,19 Therefore, it can serve as an independent predicting factor for CMV infection prognosis. We suggest that severe immunosuppression may cause a high risk of reactivation of both HHV and CMV; we also cannot rule out the role of virus-virus interactions in organ transplant.
Our data demonstrated that HHV-6 DNAemia is common in liver transplant recipients and usually occurs before CMV infection. Virus-virus interactions and severe immunosuppression can explain this phenomenon.
DOI : 10.6002/ect.2019.0119
From the Professor Alborzi Clinical Microbiology Research Center, Nemazi
Hospital, Shiraz University of Medical Sciences, Shiraz, Iran
Acknowledgements: This study was financially supported by Grant No. 93-24 awarded by the Clinical Microbiology Research Center, Shiraz University of Medical Sciences, Shiraz, Fars, IR Iran. The authors have no sources of funding for this study and have no conflicts of interest to declare. The authors thank Hassan Khajehei, PhD, for copyediting of the manuscript and the transplantation ward of Namazi hospital affiliated by Shiraz University of Medical Sciences.
Corresponding author: Mazyar Ziyaeyan, Department of Virology, Professor Alborzi Clinical Microbiology Research Center, Nemazi Hospital, Shiraz University of Medical Sciences, PO Box 71937-11351, Shiraz, IR Iran
Phone: +98 7116474304
Table 1. Cytomegalovirus and Human Herpesvirus 6 Seroprevalence and Correlation With Cytomegalovirus-Related Diseases
Table 2. Patients With Human Herpesvirus 6 or Cytomegalovirus Positivity in Serum Samples
Figure 1. Distribution of Cytomegalovirus Symptomatic and Asymptomatic Patients According to Age (in years)