Varicella-zoster meningitis in an immunocompetent young man presenting with a painless erythematous rash

Abstract

A 32-year-old man presented with a 7-day history of generalised headache, intermittent fever, emesis and diarrhoea. Four days after symptom onset, he developed a vesicular rash on his medial left thigh, without associated pain, paraesthesia or pruritus. He had no significant past medical history, and no HIV risk factors. He was presumed to have enteroviral meningitis and was commenced on supportive therapy. Lumbar puncture was performed and cerebrospinal fluid (CSF) analysis revealed a lymphocytic pleocytosis. While awaiting CSF serology, the formation of a new vesicle was noted at the site of the rash and was swabbed. Results for both the CSF and vesicle swab returned positive for varicella-zoster virus (VZV) confirming concurrent VZV meningitis with atypical painless herpes zoster in a young immunocompetent patient. He was initiated on intravenous acyclovir and made a full recovery after 2 weeks of treatment.

Background

Herpes zoster (HZ) is a common condition that affects approximately 25% of the general population within their lifetime.¹ It is caused by reactivation of latent varicella-zoster virus (VZV), which remains dormant within sensory nerve ganglia following primary exposure, typically from early childhood. HZ is classically characterised by eruption of a vesicular exanthem localised within a dermatomal boundary and is ordinarily accompanied by intense neuralgic pain.² These distinctive features allow clinicians to generally diagnose HZ based on clinical findings alone.

Besides dermatological manifestations, VZV reactivation can result in a number of less common but more serious complications.³ These include various neurological sequelae, ocular and visceral disorders, as well as vasculopathies. Diagnosis of such complications may be straightforward when a characteristic zoster rash is present. However, the absence or atypical presentation of rash can create diagnostic uncertainty, delay prompt initiation of treatment and lead to prolonged morbidity.⁴ Indeed, VZV reactivation without rash is particularly insidious. Here, VZV dissemination can persist unbeknown for several years and cause chronic radicular pain (zoster sine herpete) in addition to many of the complications listed above.⁵

We present a case of VZV meningitis in which diagnosis was delayed due to atypical cutaneous features and hope that it informs fellow clinicians in their future practice.

Case presentation

A 32-year-old man presented to our emergency department with a 7-day history of a generalised dull headache with absence of aura. The headache was associated with intermittent fever (39°C), rigors and myalgia. He also reported ongoing nausea with a single episode of emesis and occasional loose bowel motions. He denied any photophobia, limb weakness or neck stiffness. There was no history of primary headaches and no coryzal symptoms.

Five days earlier he had presented to his general practitioner where he was diagnosed with a benign viral illness and recommended analgesia and rest. After an additional 2 days without improvement, he had presented to another local hospital where he underwent a CT scan of the head and various blood testing, from which there were no pertinent findings. That evening, following discharge from that hospital, he noted a rash developing on his inner left thigh. Further questioning revealed that he had remained asymptomatic from the rash, specifically denying any pain, paraesthesia or pruritus.

The patient’s past medical history included varicella (chickenpox) in early childhood, with no previous episodes of HZ, meningitis or other infections. He had no allergies and did not take any regular medication. He was a lifelong non-smoker with occasional alcohol consumption and no known illicit substance use. He worked in the financial sector and denied high risk behaviours. He had no sick contacts at home where he lived with his wife and two children. His last travel history was 2 years ago, during which time he had migrated to Australia from his native country of China. Immunisations were complete including the seasonal influenza vaccination. There was no relevant family history.

On physical examination the patient was alert, remained haemodynamically stable and was afebrile. He had a supple neck, with no signs of meningism. Neurological, cardiorespiratory and abdominal examination were normal. Inspection of the skin demonstrated a 5cm × 4cm vesicular rash overlying an erythematous base on the medial left thigh across the L2 dermatome (figure 1). The vesicles appeared crusted and there were no sensory changes or pain overlying the area. No oral or genital lesions were observed.

Figure 1.

Vesicular rash in the L2 dermatome of the left thigh at time of presentation.

Based on the presenting symptomology a diagnosis of viral meningitis was made, but suspicion for VZV remained low. The presence of a painless rash and VZV meningitis in a normally fit and otherwise immunocompetent young adult was considered too peculiar. This created uncertainty as to whether the rash was part of the constellation of clinical findings or rather a separate and unrelated occurrence. Given that the patient remained clinically stable and in the absence of serological confirmation, initiating empiric antiviral therapy was not deemed essential. This decision was made in discussion with the Infectious Diseases team at our hospital.

Investigations

The patient’s blood panels including full blood count, biochemistry and inflammatory makers were all within normal limits. Blood cultures remained negative after 48 hours of incubation. Multiplex nasopharyngeal PCR testing was negative for respiratory viruses. On MRI of the brain and whole spine with gadolinium there was no abnormal finding or enhancement.

A lumbar puncture was performed and cerebrospinal fluid (CSF) analysis revealed a normal glucose level of 3.1 mmol/L (normal range: 2.5–4.5 mmol/L), significantly elevated protein at 2.64 g/L (normal range: 0.15–0.45 g/L) and raised leucocyte count (0.75×10⁹/L) which were entirely (100%) mononuclear. Serum glucose was measured as 5.3 mmol/L with a CSF/serum glucose ratio of 0.6. The subsequent CSF culture was negative. CSF pathogen PCR was requested for cryptococcus, enterovirus, tuberculosis (TB), herpes simplex virus (HSV) types 1 and 2, Streptococcus pneumoniae, Neisseria meningitidis and VZV. CSF flow cytometry was consistent with a reactive process and there were no malignant cells noted on cytology.

Initial attempts at collecting a swab specimen of the exanthem were unsuccessful given that all vesicles had crusted. While awaiting the results of the CSF PCR, we noted a new vesicle emerge and this was then swabbed. PCR results for both CSF and vesicle swab returned positive for VZV confirming concurrent HZ and VZV meningitis. Subsequently, the patient was investigated for possible underlying immunodeficiency, but tested negative for HIV infection.

Differential diagnosis

Our clinical suspicion of viral meningitis was validated by the finding of CSF lymphocytic pleocytosis. Given the patient was young and seemingly immunocompetent and in context of associated emesis and bowel changes, we anticipated enterovirus as being the causative pathogen. Had we received no diagnostic yield from our investigations and if patient symptoms were ongoing, we were also prepared to conduct further testing to rule out other causes of undifferentiated fever and headache including TB screening, particularly in light of the patient’s recent migration status. With regard to the exanthem, we considered a dermatology consult in order to exclude alternative causes of painless vesicular appearing rash. This was ultimately not necessary.

Treatment

On admission, we commenced the patient on supportive therapy including intravenous hydration, antiemetics and simple analgesia. Given the PCR processing at our institution is conducted off-site with 3 to 5 days’ turnaround, there was some delay in receiving virological confirmation of VZV infection. During this time the patient remained clinically stable, with his headache improving in severity, although not entirely resolving. He also remained asymptomatic from the exanthem although new vesicle formation was noted. Following laboratory confirmation of VZV detection, the patient was commenced on intravenous acyclovir at a dose of 10 mg/kg administered three times per day for a total of 14 days. Given marked clinical improvement after the first 48 hours of antiviral therapy, the patient was discharged under the care of ‘hospital at home’ services and completed the remaining duration of treatment as an outpatient.

Outcome and follow-up

At 2-week follow-up the patient had made a complete recovery. There were no ongoing symptoms of meningism. Physical examination was unremarkable including the HZ rash which had resolved and was no longer visible. Treatment was considered complete and continuation with a course of oral antiviral therapy was deemed unnecessary.

Discussion

A number of features from our case are unique and noteworthy. First, although the latency period of VZV is lifelong following primary exposure, the frequency of reactivation is considerably greater in immunocompromised individuals given their weakened cell-mediated immunity.⁶ For this reason, HZ more typically affects the elderly (in context of immunosenescence) or those individuals who are in an immunosuppressed state (eg, malignancy, corticosteroids, alcohol dependence). In addition, neurological complications such as meningitis resulting from VZV are considered a rare entity. In a previous study the incidence of VZV meningitis among 859 adults with HZ was found to be 0.5%.⁷ As such, the collective picture of a young immunocompetent patient with concurrent VZV meningitis and HZ is an unusual presentation.

Second, several characteristics of our patient’s zoster rash can be considered atypical or uncommon. Acute pain is considered a distinct feature of HZ and is attributed to the necrosis of sensory neurons as newly synthesised VZV particles are transported along their central and distal axons during reactivation.⁸ In fact, chronic pain is widely known to be the most common complication associated with HZ.⁹ Hence, in our patient, the occurrence of HZ with complete absence of any preherpetic or postherpetic neuralgia is perplexing. At present, only a few cases of painless HZ have been described in the literature.^10–13

Furthermore, neurological complications such as VZV meningitis normally occur in the ‘posteruptive’ phase. Here, an HZ infection is first established in skin cells creating a high viral load, which then reaches the central nervous system (CNS) via direct axonal spread or through the bloodstream.¹⁴ In our patient the onset of VZV meningitis oddly preceded the eruption of HZ. The lack of an external tissue viral source suggests a direct spread of viral particles to the CNS following reactivation in sensory ganglia. Interestingly, in this model the viral load remains low which may explain why the subsequent eruption of HZ in our patient was only mild.¹⁴ Finally, the localisation of our patient’s HZ in the L2 dermatome was again unusual given that the most common sites are the trunk (60%) followed by the head and neck (26%).¹⁵

In view of the presented case, there are some clinical implications that should be considered. Although VZV is believed to be a rare cause of aseptic meningitis, physicians should remain alert and recognise it as a potential cause even among demographic groups typically considered to be at less risk. Clinical diagnosis can be difficult given that features of VZV meningitis are essentially the same across all forms of meningitis regardless of the causative pathogen. Additionally, supportive findings such as HZ can present atypically and create further diagnostic uncertainty. As such, performance of rapid CSF serology remains increasingly important in guiding management.

Moreover, the optimal treatment for VZV meningitis remains ill-defined and global guidelines are yet to be established.¹⁶ Most clinicians tend to favour intravenous acyclovir for a period of 10–14 days, although oral valacyclovir has also been reported to be efficacious.¹⁷ Although enterovirus and HSV are by far the most common causes of aseptic meningitis and largely self-limiting with supportive care, definitive clinical diagnosis is not always possible.¹⁶ As such, empiric use of antiviral therapy in all patients with normal renal function while awaiting CSF serology may be reasonable particularly given that these agents are generally safe and well tolerated.

Learning points.

• Varicella-zoster meningitis can occur in all age groups regardless of immunological status.

• Cutaneous eruptions of varicella-zoster virus can precede, arise concurrently or follow neurologic symptoms and the classic features of neuralgia and dysaesthesia are not always present.

• Identifying the causative pathogen of meningitis is difficult based on clinical assessment and prompt cerebrospinal fluid analysis is required to guide management.

• Empiric antiviral treatment should be considered in all cases and ceased as appropriate once serology results become available.