Egad!

Blood tests showed a leukocyte count of 17,350/µL, neutrophils 8,720/µL (normal: 1,500–7,800), lymphocytes 6,130/µL, hemoglobin 11.3 g/dL, and platelets 231,000/µL. The basic metabolic panel was normal. Serum total protein was 6.7 g/dL with albumin 3.5 g/dL. Aspartate aminotransferase (AST) was 94 U/L (normal: 0-31), alanine aminotransferase (ALT) 56 U/L (normal: 0-31), alkaline phosphatase 45 U/L, and total bilirubin 1.1 mg/dL. Vitamin B12 was 868 pg/mL. Hemoglobin A1c and thyrotropin levels were normal. Creatine kinase was 3,757 U/L and lactate dehydrogenase (LDH) 435 U/L (normal: 122-220). The syphilis treponemal test and hepatitis B surface antigen were negative. HIV and hepatitis C antibodies were nonreactive. The anti-nuclear antibody screen was negative and complement C3 and C4 were normal.

Neutrophilia likely reflects glucocorticoid-induced demargination, as opposed to an infectious process, given the temporal association with steroid administration. Persistent mild lymphocytosis is nonspecific but more likely to reflect a reactive rather than a clonal process. Elevated LDH and CK, as well as a greater increase of AST relative to ALT, suggest muscle injury, although mild concomitant hepatic injury cannot be excluded. Normal or negative serum studies for TSH, HIV, ANA, peripheral blood smear, and creatinine eliminate many of the systemic causes of her pruritus, but malignancy and associated paraneoplastic etiologies remain considerations.

The initial work-up for SPS includes electromyography (EMG) which would show spontaneous muscle activity. Her poorly localized sensory abnormalities, transient vestibular symptoms, and confusion warrant an MRI of the brain and spine to evaluate for inflammation (eg, encephalomyelitis), which could be consistent with PERM.

An MRI of the brain and cervicothoracic spine without contrast was significantly limited by motion artifact but without obvious intracranial or cord signal abnormalities. Electromyography demonstrated spontaneous muscle activity in both lower extremities with co-contraction of agonist and antagonist muscles (hamstrings and quadriceps as well as medial gastrocnemius and tibialis anterior). Sensory and motor nerve conductions were normal. Cerebral spinal fluid (CSF) contained six leukocytes (96% lymphocytes) and three red blood cells per microliter; glucose was 67 mg/dL and protein 24 mg/dL. There were two oligoclonal bands unique to the CSF. Cytology was negative for malignant cells.

The EMG narrows the differential diagnosis considerably. Co-contraction of opposing flexor and extensor groups (with predominance of extensors) on EMG is a diagnostic criterion for SPS and explains the myalgia and elevated CK. Her normal MRI studies effectively ruled out any focal lesion and did not show signs of encephalitis. Oligoclonal bands in the CSF are a sensitive marker of intrathecal inflammation, although not specific to one diagnosis. The mildly elevated cell count also supports CNS inflammation. In the setting of a lymphocytic pleocytosis and unique oligoclonal bands, it is important to consider infectious, neoplastic, autoimmune, and paraneoplastic causes of neuroinflammatory disorders.

Serum analyses, including antiglutamic acid decarboxylase 65 (GAD65) antibody and anti-amphiphysin antibody, should be ordered. The anti-GAD65 antibody is most commonly elevated in the setting of autoimmune diabetes mellitus; the titer, however, is usually dramatically higher in SPS. The CSF titer of anti-GAD65 antibodies is more specific than the serum titer for SPS. Antibodies against amphiphysin are typically elevated in paraneoplastic SPS, and anti-glycine receptor antibodies are associated with PERM, which commonly does not have elevated anti-GAD65 antibodies.