Primary Human Herpes Virus-6 Causing Recalcitrant Pyrexia after Pilocytic Astrocytoma Resection

• HHV-6 is ubiquitous and can acutely manifest with generalized symptoms of fever, irritability, rash, and lymphadenopathy, but has a predilection for the CNS like other human herpes viruses.

• Viral infection is well documented in the literature, particularly in the immunocompromised host.

• Postoperative fever in pediatric patients after tumor resection can be attributed to fastidious difficult-to-grow Propionibacterium or post-craniotomy inflammatory processes and treated with a course of steroids.

• This is the first report of primary HHV-6 as an agent leading to meningoencephalitis after brain tumor resection.

• FilmArray Meningitis/Encephalitis (FAME)-extended PCR panel is a less commonly used, but potentially important, diagnostic tool to assist in the detection and management of postoperative fever.

• Expeditiously ruling out bacterial pathogenic sources of postoperative fever could decrease shunt internalization delays.

Human herpes virus-6 (HHV-6) is a ubiquitous virus with a tropism for the central nervous system (CNS). Due to its high prevalence with nearly 100% seropositivity in older children and adults, detailed study and accurate diagnosis are problematic. It is proposed that by the age of 12 months, 40% of children have suffered primary HHV-6 infection, and 77% by 24 months [1]. Though common and often benign in its presentation, this double-stranded DNA virus can cause clinically important disease including febrile seizures, meningoencephalitis, paresis, dysautonomia, and coma [2, 3].

Over the past 2 decades, a growing body of literature has suggested an association between viruses and intracranial tumors [4]; however, considerable equipoise remains regarding definitive evidence of HHV-6-induced oncologic transformation [5]. Given the propensity for shunt infections in the acute and subacute period following initial placement, revision, or manipulation, accurate and timely identification of the infective organism in the face of negative cultures is key in long-term management of meningitic sequelae. Here, we report the first known case of pilocytic astrocytoma-associated HHV-6 infection presenting as acute meningitis detected by FilmArray Meningitis/Encephalitis (FAME) panel. We propose a unique diagnostic approach to pediatric fever of unknown origin following neurosurgical intervention. This modified algorithm may reduce the need for long-term CSF diversion, shorten ICU length of stay, and minimize hospital cost burden.

A 2-year-old female patient presented with 1 year of progressive gait ataxia and intranuclear ophthalmoplegia. MRI of the brain with gadolinium demonstrated a large contrast-enhancing 4th ventricular exophytic brain stem mass with cystic components and associated hydrocephalus (shown in Fig. 1). She underwent suboccipital craniotomy and C1 laminectomy and concurrent placement of an external ventricular drain (EVD) with near gross total resection and less than 1 cm residual exophytic disease arising from the cerebellar peduncle. Final tumor pathology was consistent with pilocytic astrocytoma.

Her postoperative course was complicated by persistent hydrocephalus followed by persistent fever and associated CSF leukocytosis despite multiple broad-spectrum antibiotic courses and EVD exchanges (shown in Fig. 2a, b). Initially, she was placed on postoperative prophylactic oxacillin while her EVD was in place to assist with post-tumor resection ventricular decompression. This immediate postoperative fever was not associated with CSF leukocytosis, and she defervesced quickly. On postoperative day (POD) 7, despite prophylactic antibiotic use, she developed a true fever (101.8°F) and was transitioned to even more broad-spectrum antibiotics (vancomycin and gentamycin) following EVD replacement and endoscopic third ventriculostomy (ETV). Given the concern for possible gram-negative rods in the CSF broth, along with ongoing fever, ceftriaxone was added on POD 10 for additional coverage and the EVD was replaced again on POD 12. There was also concern for possible drug fever; however, she demonstrated reassuring inflammatory markers (including C-reactive protein, procalcitonin) and the CSF gram-negative rods deemed a contaminant, so antibiotics were narrowed to oxacillin.

On POD 15, the patient’s fever again trended upward and was associated with high-grade CSF leukocytosis. Pediatric Infectious Disease was consulted who recommended the addition of linezolid and rifampin, and the EVD was exchanged for a third time. Given this recalcitrant fever of unknown origin and CSF leukocytosis despite multiple antibiotic regimens, further work-up including FAME panel was initiated on POD 19. FAME PCR panel was positive for qualitative HHV-6, and additional quantitative CSF analysis confirmed HHV-6 diagnosis (3,239 copies/mL; IgG 1:160, IgM <1:20). Given suspected symptomatic HHV-6 viremia, twice-daily ganciclovir was started on POD 21. Tumor resected at time of surgery was negative for HHV-6 genome suggesting this was a primary infection acquired following her craniotomy. Within a week of ganciclovir initiation, she no longer had fever spikes, CSF leukocytosis resolved, and ICP improved. Over the course of her hospital stay, her ventricles remained with poor compliance, and she ultimately required a ventriculoperitoneal shunt (VPS) for long-term CSF diversion. Nine months postoperatively, she remains free of VPS infections or revisions with complete resolution of meningitis.

HHV-6 infection in the context of intracranial tumor pathology may play an unknown but clinically important role in the diagnosis of postoperative CNS infection. Given the ubiquity of HHV-6 and its description in all organ systems including the CNS, it is possible that HHV-6-related infections are currently underdiagnosed. One property that lends HHV-6 to escaping detection includes its capability to exist in either a latent or active state and integrate into the telomeric region of DNA, demonstrating high viral loads in both states. Further, given the high viral load in either state, differentiating between acute, chronic, and reactivated infections from an immunocompromised state may be challenging. It is prudent to not only consider but detect potential HHV-6-related CNS infections given the propensity to develop deleterious disease such as meningoencephalitis which may be complicated by sequelae such as paresis, ataxia, dysautonomia, chorea, myoclonus, myoclonic eye movements, and even coma [2, 3, 6]. In the pediatric patient, definitive diagnosis of HHV-6 is imperative as it accounts for a significant portion of febrile seizures and hospitalizations which may lead to severe and permanent neurologic deficit without treatment [2, 7]. Our case patient represents the first known pediatric report describing likely primary HHV-6 infection leading to meningoencephalitis following intracranial tumor resection, use of postoperative steroids, and relative immunosuppression.

In addition to primary infection, HHV-6 reactivation is also a theoretical consideration given that 90% of pediatric patients have been infected by 2 years of age [8]. This high infection rate may be in part due to congenital transmission as HHV-6 has the ability to integrate into chromosomal DNA [9], thereby increasing the potential for reactivation in immunocompromised states [10, 11]. Similarly, there are reports of HHV-6 reactivation in cancers such as lymphoma and leukemia [12, 14]. Specific to the CNS, HHV-6 has been linked to variable detection in inheritable glial tumors, but there are no known reports relating HHV-6 and pilocytic astrocytoma [4, 5, 15, 16]. This may be in part due to different variants of HHV-6 (A and B). While these variants express 90% similar homology, they differ in their pathogenicity and tissue selection. For example, variant A has been associated with lymphoproliferative disorders and variant B with exanthem subitem in children [17, 20] which suggests each variant may have selective properties that underly its relation to certain cancers. Given literature supporting a possible link between HHV-6 and brain tumor pathogenesis, a postoperative tumor specimen analysis was performed and showed no HHV-6 DNA integration of the tumor specimen [21].

Serologic evidence of acute infection or HHV-6 reactivation is demonstrated by a significant rise or seroconversion of IgG and IgM titers. High titers relative to healthy controls are suggestive of a recent infection or ongoing chronic infection. Further, HHV-6 is only detected in the serum or plasma during an acute infection or a chromosomally integrated infection [22]. These parameters suggest one of the following etiologies in our patient: (1) a primary infection in the CNS following previous peripheral infection or (2) a reactivation in CNS infection due to isolated genome integration within the peripheral ganglion and not within the brain. Taken together, serum-positive HHV-6 PCR, positive CSF HHV-6, negative HHV-6 tumor integration, a robustly positive serum IgG level, and the extreme rarity of chromosomal integrated HHV-6 infection, this patient’s meningitis was likely due to a recent HHV-6 infection in a postoperative immunocompromised state.

Limited data on HHV-6 meningitis exist, and there is no well-described, pathognomonic CSF profile. In the absence of any confirmatory culture or PCR results, clinical similarities obscure identifying viral versus bacterial meningitis. The pathogenesis can also be highly variable depending on the specific pathogen, the patient’s co-morbidities, medical history, age, and immunocompetency [23]. While both may present with photophobia, neck stiffness, headache, and vomiting, viral meningitis tends to have a biphasic fever profile. A systemic illness first develops accompanied by an initial fever, followed by a second stage of fevers that begins with the onset of meningismus, though neonates may have no signs of meningeal irritation [23, 24]. Bacterial infections conversely have a continuous fever pattern [25]. Viral infections also often have a longer clinical course, lasting up to weeks before full patient recovery [26]. CSF cell count profiles typically show a mononuclear cell predominance with fewer WBCs, less protein, and normal glucose compared to bacterial infection [27] (Table 1). Though fungal infections can also cause meningitis, these tend to be rarer and occur in immunocompromised individuals. In addition, their CSF profiles tend to have more elevated WBC and protein than viral infections but still less than bacterial [28].

Given the putative link between HHV-6 infection and an immunocompromised host state, viruses like HHV-6 should be considered as a potential infective postoperative pathogen. Rapid detection of viral agents using the FAME panel has been recently supported as an effective test to differentiate between bacterial and viral meningitis. This rtPCR assay has the capability to identify and detect up to 14 pathogens in a single sample with an 85% sensitivity for HHV-6 [29]. As demonstrated in our patient, utilization of the FAME panel may expedite identification of the offending pathogen in the absence of confirmed bacterial meningitis. In addition, studies demonstrate that early detection via FAME panel reduces the use of anti-infectives, shortens the hospital course [30, 32], and decreases the overall cost burden [33]. The expense of patient care may also be reflected in those patients whose diagnosis of fever remains elusive for some time, leading to additional procedures and subsequent hospital stay as was the case in our patient who developed poor ventricular compliance and ultimately required a VPS for permanent CSF diversion.

While ideal postoperative management of posterior fossa and/or infectious meningitis would avoid permanent CSF diversion procedures, one must weigh the risk of extended EVD placement and prolonged ICU stay against definitive management with shunt placement. It should also be noted that permanent CSF diversion may not be required in all patients following meningitis. In this specific case, our patient had at least two risk factors for requiring permanent CSF diversion including both posterior fossa tumor and infectious course [34]. If patients eventually require CSF diversion, one must consider the optimal parameters for performing such procedures. To reduce the risk of shunt infection and malfunction, confirmation of clearance of the infectious agent as well as lower protein levels may be considered. However, there is currently no consensus regarding a specific protein level that may reduce shunt malfunctions [35, 36]. Alternative CSF diversion procedures including hardware-free options such as ETV should also be considered [37, 38]. In the specific case of this patient, an ETV was performed; however, the patient remained with symptomatic increased ICP and ultimately required a VPS for definitive management.

Pragmatically, early detection of HHV-6 may not only reduce the need for additional procedures, but also prevent future hospital stays and unnecessary tests given the native risk of VPS infections [39, 40]. When caring for pediatric patients with a fever of unknown origin that persists beyond 48 h, we propose a diagnostic algorithm as shown in Figure 3. This pathway incorporates the diagnostic utility of a FAME panel, infectious disease consultation, and management of anti-infectives which may expedite the detection and treatment of a nonbacterial fever etiology, such as that with HHV-6 infection, and reduce the possibility of devastating neurologic sequelae. Taken together, HHV-6 should be considered as a potential pathogen in pediatric patients with persistent fever of unknown origin following intracranial tumor resection. FAME panel may refine this process by reducing unnecessary medications and procedures, shorten hospital stay, and decrease cost burden over time.

We would like to thank the Pediatric Infectious Disease team for their expertise and guidance regarding use and management of anti-infectives.

Ethical approval is not required for this study in accordance with local or national guidelines. No IRB was indicated for this study as it was a retrospective review of the clinical course, and no identifying information has been presented in this manuscript. The patient’s parents provided their written informed consent for publication of the details of this medical case and any accompanying images.

Julie Lynn Chan has received research materials and funding from MISONIX, Inc. Peyton Nisson, Moise Danielpour, and Jack Green have no conflicts of interest to declare.

No funding was indicated for this retrospective single-case review.

Julie Lynn Chan and Jack Green conceptualized, drafted, and critically revised the manuscript. Peyton L. Nisson drafted the manuscript. Moise Danielpour conceptualized and critically revised the manuscript.

All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.

Senior Authorship: Dr. Moise Danielpour and Dr. Jack Green are co-senior authors.