Although occasional patients have been described that show both central and peripheral nervous system (CNS and PNS) demyelination on magnetic resonance (MRI) or electrophysiology studies, clinically evident disease is usually confined to one compartment, not both. Patients with combined CNS and PNS demyelination have been previously reported in the literature, but have been given various names for their disorder, based on the part of the nervous system in which symptoms predominated. For example, "peripheral neuropathy in MS" (8) or "chronic demyelinating peripheral neuropathy associated with multifocal central nervous system demyelination" (13) are diagnoses usually assigned to these combined CNS and PNS myelinopathy cases. These variable appellations make it quite difficult to search the literature for similar cases but have arisen in an attempt to work around the thorny issue of whether this is a single, unique condition distinct from MS or CIDP or co-associated CIDP and MS. These descriptive designations are also based on the fact that while neuroimaging abnormalities in central white matter are well described among patients with CIDP, most have no symptoms referable to these findings, making it difficult to consider them definitive MS patients (7). Similarly, patients with clinically-evident MS and subclinical peripheral nerve dysfunction have been described.(8) Significant symptomatology referable to both CNS and PNS demyelinative lesions in the same patient has been reported but is distinctly unusual.(9, 10, 11, 13) The present case provides another example of clinically evident demyelinating disease involving both compartments, although it is obviously unclear whether CNS or PNS symptoms predominated in life.
Although details of the patient's clinical features were sketchy, while alive, our patient was diagnosed at various times with Friedreich's ataxia, MS, and GBS. The cerebral, cerebellar, brainstem, and spinal cord demyelinative lesions found at autopsy manifested classic histological appearance and distribution for MS. The degree of peripheral nervous system involvement was also pronounced, corroborating the diagnosis of peripheral neuropathy in life. Occasionally, the nosology of GBS and CIDP may be ambiguous, though chronically progressive or relapsing and remitting nerve dysfunction eventually excludes the diagnosis of GBS. The similarity of initial clinical symptoms may be further confounded by shared histopathological features, such as segmental demyelination, perivenular foci of lymphocytic infiltrates in ventral spinal roots and dorsal root ganglia, and macrophage associated demyelination (2, 3, 6). However, true onion-bulb formation, as detected focally in spinal nerve roots and much more conspicuously in peripheral nerve segments in our case, are much more commonly seen in CIDP than in GBS (3). In our case, the persistent inflammatory infiltrates and the chronic nature of changes in spinal nerve roots (such as formation of Schwann cell proliferations around meningeal and spinal cord parenchymal vessels secondary to nerve damage) provided further evidence that the patient's condition more closely resembled CIDP, rather that a monophasic disorder such as GBS.
One of the most interesting aspects of the case was the microscopic onion bulb formation. Onion bulb formation has been well-described at various levels of the PNS, including proximal spinal nerve roots, nerve trunks and even distal sural nerve in patients with combined CNS and PNS myelinopathy. What is considerably less appreciated is that Schoene et al. described, and beautifully illustrated, onion bulb formation in cases of MS associated with hypertrophic neuropathy even within the CNS (11). These CNS onion bulbs were within MS plaques and could be highlighted by trichrome or LFB-PAS stains. Schoene et al. made the careful observation that "Luxol fast blue preparations showed that the myelin within the 'onion bulbs' in the CNS was dark blue suggesting that it was of peripheral type"(11). Electron microscopy further verified that true onion bulb formation did occur in the CNS and that the origin of the myelin in these CNS bulbs was from Schwann cells, not oligodendrocytes (11). Although we did not have EM on our case, on careful re-review we did identify the small patches of darker blue myelin staining with concentric appearance within the most severely demyelinated plaque in mid-thoracic spinal cord. These intraparenchymal collections of CNS onion bulbs also showed faint trichrome staining. Of note, neither we nor Schoene et al. felt that the CNS onion bulbs were in direct continuity with PNS onion bulbs, peripheral nerve roots, or the surface of the spinal cord (11).
Also identical to our case, Schoene et al. noted, and illustrated, the presence of schwannosis around vessels in several of his cases (11). Schwannosis can be seen in various types of long-standing spinal cord lesions, including tumors, syringomyelia, traumatic scarring, cord or nerve root compression, and severing of nerve roots. They represent aberrant regenerated axons surrounded by Schwann cells. Our case was also identical to the cases of Schoene et al. in that the spinal cord contained the most extensive demyelinative plaques and that the optic nerves showed demyelination (11). The optic system may have a particular proclivity for involvement, given the fact that in one clinical study, prolonged visual evoked potential latencies were found in 47% of patients with CIDP, even in some cases in the absence of high signal intensity lesions in the brain on T2-weighted MRI scans (12).
Observation of occasional patients who have both CNS and PNS symptomatology has led to speculation that central and peripheral demyelination may share common pathogenic mechanisms. Previous investigators (5) have observed electrophysiological similarities between GBS and the experimental model of chronic experimental allergic neuritis (EAN) described by Waksman and Adams (14). An analogous model in the central nervous system of experimental autoimmune encephalomyelitis (EAE) has been described which shares many clinical and pathological features with MS. While the experimental models and clinical disorders are generally distinct, a few observations in animal models are of interest. Peripheral nerve P1 protein sensitization in some species may produce a disease resembling EAE, possibly due to shared epitopes with CNS myelin basic protein.(1) Yonezawa noted that serum from guinea pigs and rabbits sensitized with a combination of large amounts of complete Freund's adjuvant, tubercule bacilli, and peripheral nerve, presumably containing P1 protein, may produce a combined CNS and PNS syndrome that has similarities to human cases such as ours.(15) In Lewis rats, Clark noted that T-lymphocytes involved in EAN and EAE may share common V region receptors, suggesting a possible ligand common to both CNS and PNS.(4) Yet, Lewis rats sensitized with peripheral nerve myelin only develop a PNS syndrome (16). At least in some species, shared CNS and PNS epitopes alone appear insufficient to cause disease in both compartments. The generally distinct PNS and CNS syndromes in human beings with occasional cases of overlap have long resulted in speculation of a possible relationship between MS and CIDP. However, while the experimental models suggest potential mechanisms, the implications of these findings for human PNS and CNS overlap syndromes such as we describe here remain uncertain.
Contributed by Dianna Quan, MD, B.K. Kleinschmidt-DeMasters, MD