The great pretender

Abstract

In November, 2009, a 40-year-old man presented to us because of diffi culty in walking, facial asymmetry, and numbness in both legs. His symptoms had begun 6 weeks earlier with a sensation of pins and needles in both thighs spreading distally. At the same time he noticed a left facial droop, with involvement of the right side soon after. There was no change in bladder or bowel function. He had diabetes mellitus and was receiving insulin. He had no history of signifi cant exposure to toxins or alcohol misuse. There was no family history of neurological problems. He had recently been found HIV negative on testing after starting a new heterosexual relationship and reported no symptoms of sexually transmitted infections. Physical examination showed no rash or lymphadenopathy. His left conjunctiva was injected; both pupils reacted normally. On neuro logical examination he had bilateral lower motor neuron facial weakness and mild proximal leg weakness with 4– power in hip fl exion, 4+ at the knees, with full power elsewhere. Tone was normal but refl exes were globally absent with fl exor plantar responses. There was no muscle wasting or fasciculations. Loss of pain sensation was elicited in both legs in a high stocking distribution extending to the upper thigh. Position sense was intact. A clinical assessment of subacute sensorimotor polyneuropathy was made, and the diagnosis of GuillainBarre syndrome was considered. Although diabetes can present with chronic predominantly sensory peripheral neuropathy, cranial mononeuropathy, and radiculoplexus neuropathy, we thought the last less likely in view of the bilateral facial weakness and the absence of pain. Cerebrospinal fl uid (CSF) analysis showed a high protein concentration and lymphocyte count (fi gure). Because Guillain-Barre syndrome is rarely associated with CSF pleiocytosis in immunocompetent patients, we did a VDRL test on CSF; the result was positive, as was that for serum (titre 1 to 128), confi rming an unexpected diagnosis of neuro syphilis. Our patient was treated with intravenous benzyl penicillin 5 mU every 6 h. After 14 days of treatment his sensory symptoms had resolved, and within a month after discharge he had regained full power, including of the facial muscles. At fi nal follow-up in November, 2010, a repeat HIV ELISA was negative and he had made a full recovery. He had disclosed the diagnosis to his sexual partner who underwent screening and treatment. Neurosyphilis can be divided into early and late forms. In our patient, a cellular response in the CSF, rapid symptom onset, and strongly positive serum VDRL favoured early disease with meningeal involvement. Late neurosyphilis manifests as tabes dorsalis and syphilitic paresis resulting from brain and cord parenchymal infection. There is evidence that the clinical spectrum of neurosyphilis has changed in the post-antibiotic era, with early forms now predominating. Diagnosis of neuro syphilis can be challenging and requires a combination of tests. Although CSF-FTA assays are reactive in more cases than the CSF-VRDL in all stages of syphilis, their specifi cities are less well established and they are not regarded as a valid basis for the diagnosis of neurosyphilis in isolation. By contrast, VDRL is highly specifi c in the CSF and is regarded as the standard serological test, but it reportedly has lower sensitivity. Immunocompetent patients with positive serum treponemal tests and compatible clinical signs should be treated for neurosyphilis if the CSF shows a raised protein concentration or leucocyte count. Syphilis is a great mimicker in clinical medicine, and neurosyphilis can occur early after infection, in many cases without accompanying peripheral manifestations. The diagnosis should be considered in patients presenting with symptoms and signs resembling Guillain-Barre syndrome and CSF pleiocytosis.