Multiplex qPCR facilitates identification of betaherpesviruses in patients with acute liver failure of unknown etiology

Acute liver failure (ALF) is a condition whereby healthy liver deteriorates rapidly, resulting in jaundice, encephalopathy and coagulopathy. According to the recently stablished diagnostic criteria, ALF may be defined when the interval between the appearance of jaundice and the development of coagulopathy and encephalopathy is up to 8 weeks [1]. The etiology of ALF cases is often unknown and some cases are not positive for hepatotropic viruses or other established potential factors [2]. Previous studies have reported the presence of viruses of the family Herpesviridae in cases of ALF, including the betaherpesviruses human cytomegalovirus (HCMV), human herpesvirus 6 (HHV-6), and human herpesvirus 7 (HHV-7) [3, 4].

HCMV triggers an infection similar to mononucleosis syndrome while HHV-6A/B and HHV-7 cause roseola infantum, febrile seizures and other febrile syndromes in children [4, 5]. After infection, betaherpesviruses remain latent until conditions in the immunocompromised host favor reactivation. HHV-6A/B and HHV-7 display genetic and biological homology and commonly manifest as sudden rash [6]. Primary HCMV and/or HHV-6A/B infections can cause mild and self-limiting hepatitis in immunocompetent patients. However, in immunocompromised cases, in particular, organ transplant recipients, HCMV infection is associated with increased morbidity [7]. Primary infection with HHV-6 is a frequent cause of ALF, necessitating antiviral therapy during early post-transplantation procedures. A number of studies have disclosed the presence of HHV-6 in liver samples from pretransplant cases [8, 9]. Complications resulting from HHV-7 infection have also been identified as a risk in organ transplantation [10]. On the other hand, although have been rarely described, there are increasing evidences of association between some betaherpesviruses and ALF in immunocompetent patients [11‐13].

Poor prognosis of hepatitis evolving from herpesvirus infection is associated with late diagnosis and delayed specific antiviral therapy, highlighting the urgent need for early detection. Currently, while several protocols exist for identification of the herpesvirus genome [14],multiple PCR assays are required for the process. In recent years, real-time PCR has been increasingly applied to simplify the overall procedure and reduce performance time in detecting and quantifying virus in clinical samples [15, 16]. Multiplex qPCR has been suggested as an efficient alternative for simultaneous detection of HCMV, HHV-6 and HHV-7 within the same reaction [9, 17]. In the current study, the incidence of betaherpesvirus infection in patients with ALF of unknown etiology was investigated using the multiplex qPCR approach.

Poor prognosis of hepatitis evolving from herpesvirus infection is associated with late diagnosis and consequently, delayed initiation of antiviral therapy. One major reason for delayed diagnosis is the lack of specific symptoms. Typical mucocutaneous lesions are absent in half of the patients developing hepatitis [3, 17], highlighting the need for early diagnosis for optimizing therapy. In this study, 23.6% samples from ALF patients displayed positivity for at least one of the betaherpesviruses. Our collective results support the utility of qPCR in detecting and quantifying HCMV, HHV-6 and HHV-7 in ALF samples of unknown etiology.

Recent years have seen an increased use of real-time PCR assays to simplify and reduce performance time and quantify virus in a single reaction [20]. In our experiments, RNase P as an internal reference control demonstrated the absence of false-negative results related to detection and quantification of betaherpesviruses in ALF samples with no defined etiology, supporting the accuracy of the results. In addition, multiplex qPCR was specific for betaherpesviruses, since we observed no amplification of other viruses of the Herpesviridae family examined, including HSV-1, HSV-2 and EBV. This result confirmed the specificity of the regions used for betaherpesvirus amplification as well as exclusion of false-positive results potentially caused by the interference of substances or organisms with similar sequences [14]. In the multiplex assays used in this study three fluorescence dye combinations were used to label the probes (FAM, VIC and NED). It allowed that the detection of three types of betaherpesvirus (CMV, HHV-6 and HHV-7). Furthermore, all samples identified as positive via qPCR were confirmed using Pan herpesvirus PCR and sequencing analyses. However, the multiplex is sufficient for screening ALF patients. Although hepatitis viruses are the most common cause of ALF, herpesviruses have also been associated with incidence of hepatitis [16, 21, 22].

In this study, three samples were detected as positive for HCMV. This virus has been reported in cases of liver transplantation in addition to being associated with induction of hepatitis [23, 24]. The association of HCMV replication with ALF suggests a condition that increases the occurrence of HCMV infection after transplantation [24], leading to a requirement for prophylactic antiviral treatment to prevent HCMV infection in transplant recipients [25]. To date, studies have revealed that HCMV is not the main cause of ALF but rather the most common hepatic manifestation in immunocompromised patients [26].

Higher presence of HHV-6 was observed in the ALF samples tested. Among the samples positive for betaherpesviruses, 66% were HHV-6-positive. The greater positivity for HHV-6 corroborates with previous results showing that HHV-6 is more prevalent and pathogenic in ALF cases [9, 13]. In another study, ALF was associated with primary HHV-6 infection in one child [27]. Several other HHV-6-associated cases have been reported, mainly in infants [8, 12]. Earlier cases of ALF have similarly been attributed to HHV-6, although the virus is not yet classified as a causative agent of liver disease [4]. In a previously study, was showed that 12 of 15 patients with disease from an unknown cause and four of 17 patients with ALF from known causes contained HHV-6 antigens in the explanted liver [8], along with a Brazilian case of ALF with HHV-6B [9].

Notably, positivity for HHV-7 was only detected in the presence of other betaherpesviruses. Two cases of dual infection (HHV-7/HHV-6 and HHV-7/HCMV) were identified within our patient population. Co-infection between roseoloviruses (HHV-6 and HHV-7) has been described in a previous studies on post-transplant patient liver samples [10]. Although, in this study was not possible demonstrate that HHV-6 precursor infection can induce subsequent HHV-7 infection, this co-infection should be better investigated.

Some evidence suggest that the molecular mechanisms of viral latency and reactivation are shared among these viruses. HHV-6B is reactivated from latency after coinfection with HHV-7 [28], and HCMV disease is frequently associated with concurrent HHV-6 and HHV-7 reactivation in transplant patients [29, 30]. Studies have suggested an association between HHV-6 and HHV-7 reactivation and increased risk of CMV disease among kidney and liver recipients [31‐33], but it is unclear if HHV-6 and 7 infection truly potentiates CMV disease or if the presence of these viruses represent more immunosuppression and attendant risk of CMV [34].

Evaluation of alanine transaminase (ALT) and aspartate transaminase (AST) enzyme levels disclosed values > 1000 U/L in ALF samples positive for betaherpesviruses. Destruction or injury of liver cells releases these enzymes into the circulation [35] and injury-related levels can reach up to 100 times the upper limit reference values of ALT and AST in cases of ALF. Values higher than 1000 U/L are usually reported in cases of viral hepatitis or hepatitis induced by drugs [35, 36]. Another biochemical parameter of liver function determined was INR of 5.25. The elevation time of coagulation represents low prothrombin production by hepatocytes.

Alkaline phosphatase (ALP) is not a hepatocyte-specific marker. However, ALF patients presented elevated levels of ALP of up to 200 mg/dL. As determined previously, ALP displays small elevations in normal levels in cases of infectious hepatitis [35]. However, biochemical results disclosed no significant differences in values in hepatitis cases due to the limited sampling populations analyzed.

In our study, 55.5% positive patients died, consistent with earlier literature showing that ALF often progresses to multiorgan failure, resulting in death or transplantation in 57% cases [37] A significant proportion (12%) of ALF cases worldwide are of unknown origin, often termed indeterminate [37]. In total ~ 2000 cases per year of liver failure are diagnosed in the United States [38], averaging ~ 1 to 6 cases per million worldwide. In Brazil, 2% liver transplants are attributable to hepatic failure [39]. In the absence of information on etiology of the disease, late therapy is initiated in many cases, leading to ineffective treatment and ultimately death.

The incidence of betaherpesviruses (HCMV, HHV-6 and HHV-7) is common worldwide. HCMV has a prevalence of 70–80% in adults [7]. HHV-6 and HHV-7 are ubiquitous, with higher prevalence in adults of up to 90% [40]. The high prevalence of betaherpesviruses can favor opportunistic infections, as seen in ALF, potentially due to virus reactivation as a result of decreased host immune system function [41, 42]. Betaherpesviruses-associated diseases may be decrease by prophylactic treatments with antiviral drugs [43]. Therefore, considering the interval among the appearance of early specific symptoms and the development of ALF, establishing the presence of active infection by herpesvirus in ALF patients is important to quickly start therapy with antiviral for prevent extensive liver damage.

Although we have not showed a clear association with betaherpesvirus and ALF, this is considered a rare cause. However, several cases of hepatitis highlighting herpesviruses as the most probable cause of this disease have been described in the literature. In keeping with previous findings, data from our study found betaherpesviruses in ALF cases and co-infection of HHV-7 with HHV-6 and CMV. Therefore, diagnosis should be made with caution after excluding classic hepatotropic viruses as another potential cause of disease. Evaluation of betaherpesviruses should be included for differential diagnosis in cases of ALF in order to improve specificity of treatment and minimize patient death. Thus, the multiplex qPCR used in this study simultaneously detected betaherpesviruses and may be useful for rapid and differential diagnosis in cases of ALF with unknown ethiolog.