An 18-year-old presented to the emergency department (ED) with a 1-day history of severe headache of sudden onset and vomiting. He denied fever, neck pain, and photophobia. There was no history of head trauma, ill contacts, tick bite, or rash. The patient did not have a history of headaches, but his brother had migraines.
In the ED, the patient was alert and oriented but appeared to be in pain. He rated his headache as an 8 on a scale of 1 to 10 (10 being the most severe pain). His oral temperature was 37.2°C (99°F); pulse rate, 54 beats per minute; respiratory rate, 24 breaths per minute; and blood pressure, 144/64 mm Hg. Examination of the head, eyes, ears, nose, and throat showed no evidence of trauma, cranial bruits, or photophobia. Optic disc margins were sharp, and there was no neck stiffness or pain. Neurologic assessment revealed normal cranial nerve function, deep tendon reflexes, and Babinski reflex. The patient did not have ataxia or cerebellar signs. He had no rash and his skin had good turgor and brisk capillary refill. Findings from the lung, cardiovascular, and abdominal examinations were normal.
CT scans of the brain with contrast showed no abnormalities. White blood cell (WBC) count was 12,700/µL, with 75% neutrophils, 15% lymphocytes, and 9% monocytes; hemoglobin, 15.7 g/dL; platelet count, 350,000/µL. Electrolyte concentrations were within normal limits.
The patient was treated with intravenous ketorolac( and metoclopramide( and observed in the ED for several hours. His headache significantly improved with these measures but did not resolve completely. The patient was discharged with a diagnosis of migraine and instructed to follow up with his pediatrician if his symptoms persisted or worsened.
The patient felt much worse the following day and again saw his pediatrician. He had taken a dose of his brother's migraine medication, which had no effect. The patient now had a fever, neck pain, and photophobia. He was sent back to the ED for lumbar puncture and admission for probable meningitis.
The lumbar puncture confirmed the diagnosis of meningitis. The cerebrospinal fluid (CSF)WBC count was 771/µL, with 1% polymorphonuclear leukocytes, 96% lymphocytes, and 3% monocytes; the red blood cell count was 5. CSF glucose and protein levels were normal at 48 mg/dL and 136 mg/dL, respectively. No organisms were seen on the Gram stain. The CSF was sent for culture, and a bacterial antigen panel and polymerase chain reaction (PCR) assay for enterovirus were ordered. Serum Lyme titers were also requested because the patient lived in an area in which Lyme disease is highly endemic.
The patient was admitted to the inpatient unit, where therapy with intravenous ceftriaxone(, 1 g q12h, was initiated. Although his vomiting had subsided, he was in severe pain because of the headache. Intravenous ketorolac and morphine( afforded him some relief. On the second hospital day, results of the CSF bacterial antigen panel were negative. The patient was feeling better but complained of an itchy rash on his back (Figure). The rash initially appeared as a small cluster of pink, blanching, 3- to 4-mm papules located adjacent to the 12th thoracic vertebra. It was neither vesicular nor painful and was thought to be a contact dermatitis. No treatment was ordered.
Results of the CSF culture and PCR assay for enterovirus were negative at 48 hours. The patient was discharged on the third hospital day. He was afebrile and had no meningeal symptoms. He still had a mild headache that was relieved with ibuprofen(. Serum Lyme titers were unavailable for another 24 hours. The patient was given a parting dose of ceftriaxone to cover for Lyme disease and advised to follow up with his pediatrician the next day for results of his Lyme test.
The patient's headache was nearly gone the following day, but the rash on his back had worsened. It had become papulovesicular with a dermatomal distribution consistent with herpes zoster. The hospital laboratory was asked to run PCR assays on the remaining CSF samples for varicella-zoster virus (VZV) and herpes simplex virus. Results confirmed a diagnosis of VZV meningitis.
A review of the patient's history indicated that he had chickenpox when he was 5 years old. Although there was no history that suggested an immune disorder or HIV risk factors, tests for immunoglobulin levels and for HIV were conducted. Results of these tests were normal.
The patient was treated for 14 days with oral acyclovir, even though his symptoms were essentially gone by the time he began this therapy (4 to 5 days after the onset of his illness). No recommendations could be found regarding treatment of VZV meningitis in the literature.
Questions about VZV Meningitis
This case raises several interesting questions. How common is VZV aseptic meningitis in immunocompetent children? Is treatment with antiviral medications warranted? Will we be seeing more zoster and unusual presentations of VZV infection with waning immunity after varicella vaccination?
Those of us whose practice of pediatric medicine predates the varicella vaccine remember well the neurologic manifestations of acute chickenpox--in particular, encephalitis and ataxia. While herpes zoster in children has never been common, CNS complications of herpes zoster are even rarer and have been thought to occur only in immunocompromised patients.1,2
An interesting feature of this case is that symptoms of meningitis preceded the development of the zoster rash. This could be considered a form of zoster sine herpetica--CNS zoster infection without cutaneous manifestations. We found only 2 case reports of zoster sine herpetica in immunocompetent children. One child had meningitis and the other had encephalitis. Both these children had chickenpox before they were 2 years old, which may have placed them at higher risk for VZV reactivation.3,4
In our patient's case, the diagnosis of VZV infection was made by PCR testing in a matter of hours. We had not routinely ordered PCR testing in cases of suspected aseptic meningitis. We now know that herpes zoster virus may be a more common cause of aseptic meningitis than previously thought. This virus can affect immunocompetent children, and we may not have cutaneous clues to aid our diagnosis.5-7
The question about whether treatment is necessary for patients with VZV meningitis remains unanswered. We could not find any data that indicated that treatment shortened the course of the illness or reduced morbidity.8-11
Although acute chickenpox may be a disease of the past, VZV is still alive and well.
Aseptic meningitis is often attributed to a "virus," which is never identified. However, with new techniques such as PCR assays, we may be able to identify causes of non-bacterial meningitis that were previously undetectable.
Further study is needed to determine whether herpes zoster virus is a common cause of meninigitis in older children or adolescents--and whether treatment with acyclovir would alter the course of this illness.