Eosinophilic Neuropathy — Is It In Your Skin?
Abstract & Commentary
By Norman Latov, MD,PhD, Professor of Neurology and Neuroscience, Weill Medical College of Cornell University. Dr. Latov is a consultant for Quest Diagnostics and a stockholder in Therapath.
Synopsis: Patients with eosinophilic neuropathy, associated with primary eosinophilia or the Churg-Strauss syndrome, exhibit cutaneous vasculitis and reduced epidermal nerve fiber density.
Source: Chao, CC, et al. Skin Denervation and Cutaneous Vasculitis in Eosinophilia-Associated Neuropathy. Arch Neurol 2007; 64: 959-965.
Peripheral neuropathy is common in patients with the Churg-Strauss syndrome that present with eosinophilia, asthma, and small vessel vasculitis. Neuropathy also occurs in primary eosinophilia, in the absence of Churg-Straus syndrome, but a nerve and muscle biopsy is often required in such cases to determine whether it is caused by vasculitis.
Chao and colleagues examined skin biopsies from 12 patients with neuropathy and concomitant eosinophilia, 6 with primary eosinophilia, and 6 with the Churg-Strauss syndrome. Normal appearing skin was obtained by punch biopsies from the distal calves, and epidermal nerve fiber density was determined by immunocytochemistry, using a monoclonal antibody to the neuronal gene product 9.5. A diagnosis of cutaneous vasculitis was made if there was evidence for both perivascular inflammation and vascular injury. Disruption of vascular integrity was evidenced by the presence of infiltrating leukocytes with discontinuity of the vascular wall, or extravasation of red blood cells.
Mononeuritis multiplex was the most frequent presentation, occurring in 5 of 6 patients in each of the 2 groups, with the others presenting with distal symmetric neuropathy. All 12 improved after therapy with steroids, alone, or in combination with cyclophosphamide or plasmapheresis.
Epidermal nerve fiber density was reduced in 10 patients, 5 with primary eosinophilia and 5 with the Churg-Strauss syndrome, consistent with the loss of small nerve fibers. All 6 patients with Churg-Strauss syndrome, and 3 of the 6 with primary eosinophilia, fulfilled pathological criteria for definite vasculitis; one primary eosinophilia patient had borderline vasculitis. Perivascular eosinophils were seen in 3 of 6 patients with Churg-Strauss syndrome, and in 2 with primary eosinophilia.
The authors conclude that the neuropathy of primary eosinophilia is commonly associated with small vessel vasculitis, similarly to that seen in the Churg-Strauss syndrome. They also note that pathological analysis of punch skin biopsy may spare patients from a nerve and muscle biopsy, if the analysis confirms the presence of cutaneous vasculitis.
The observed reduction of epidermal nerve fiber density in the patients described is not particularly surprising, as all had abnormal EMG and nerve conduction studies. Determination of epidermal nerve fiber density is most useful for diagnosing neuropathy in patients with small fiber disease, or with insufficient involvement of the large fibers to be picked up by electrodiagnostic studies.
Comparison of the nerve fiber density at the distal calf, to that proximally at the thigh, can, in addition, help distinguish between multifocal and distal/symmetric neuropathies. In distal/symmetric neuropathy the nerve fiber loss is more severe at the calf, whereas in multifocal neuropathy or sensory neuronopathy the thigh may be more severely affected. This type of analysis was not done in the current study, but might have revealed more severe reduction of nerve fiber density at the thigh, including in the 2 patients with normal distal nerve fiber densities.
Pathological analyses of skin biopsies can provide clues to the pathogenesis of the neuropathy. Amyloid deposits, as example, can sometimes be seen in skin biopsies from patients with neuropathy and systemic amyloidosis, and the authors previously reported the presence of cutaneous vasculitis in patients with neuropathy and systemic lupus erythematosus. Pathological analysis of skin biopsy may thereby spare the patient from a more invasive nerve and muscle biopsy, in patients suspected to have vasculitis.
The current paper provides support for the hypothesis that eosinophilic neuropathy is caused by vasculitic injury rather than eosinophilic toxicity, although the latter may also contribute to the vasculitis. Primary eosinophilia is a diverse disease, with recent studies implicating such mechanisms as constitutive activation of cellular tyrosine kinases, or secretion of IL5 by clonally expanded T-cells. Potential therapies that more specifically target the underlying mechanisms include mepolizumab, an anti-IL5 antibody, or imatinib mesylate (Gleevec), an inhibitor of tyrosine kinase. It remains to be seen whether these might be more effective therapies for the neuropathies associated with eosinophilia.