Disseminated Vasculomyelinopathy:
An Immune Complex Disease

Louis Reik, Jr, MD

The numerous nervous system abnormalities which follow antecedent infections and immunizations appear to share a common pathogenesis involving the immune system. Pathologically, a small vessel vasculopathy involving arterioles and capillaries as well as venules in both gray and white matter is the earliest and most consistent change. Perivascular demyelination appears to develop subsequently. Delayed hypersensitivity to myelin basic protein may not adequately account for these changes.

Humoral immunity may be involved instead. I postulate that antigen-antibody complexes, formed following the introduction of foreign antigen by infection or inoculation, cause vascular injury with secondary damage to myelin. There is considerable evidence that circulating immune complexes are present in some postinfectious nervous system disorders, as are associated systemic features which suggest immune complex disease. Similar clinical and pathological nervous system changes occur in a variety of disorders in which circulating immune complexes are thought to cause vascular injury.

Reik L Jr:   Disseminated vasculomyelinopathy: an immune complex disease. Ann Neurol 7:291-295, 1980

There is general agreement that the postinfectious and postvaccinal disorders of the nervous system represent an "allergic" phenomenon. Current hypotheses concerning their pathogenesis focus on the cellular immune system and an attack on the myelin sheath with subsequent demyelination [4, 6, 15, 32]. Such hypotheses have developed largely as a result of the clinical and histopathological similarity between these human disorders and experimental allergic neuritis (EAN) and experimental allergic encephalomyelitis (EAE) in laboratory animals, both of which appear to result from delayed hypersensitivity to myelin basic proteins [4, 15,32]. They gain support from the demonstration of both sensitized lymphocytes in patients with Guillain-Barré syndrome or Bell’s palsy capable of demyelinating peripheral nerve in culture [4] and of lymphocyte sensitization to nerve tissue antigen in a variety of postinfectious disorders [6, 15, 35].

These immune alterations may not be causal in humans, however, since not all affected patients possess such sensitized lymphocytes [13, 35]. Although failure to demonstrate sensitized lymphocytes in humans may reflect their appearance in the peripheral circulation in numbers too small to detect or result from inadequacies inherent in in vitro tests, the development of delayed hypersensitivity to myelin may be merely an epiphenomenon resulting from nervous tissue damage, since it also develops after cerebral infarction [35]. Indeed, the Guillian-Barré syndrome has developed in an immunosuppressed patient apparently incapable of demonstrating delayed hypersensitivity [13]. In addition, both EAE and EAN follow sensitization with injections of Freunds adjuyant and nervous system antigen. How a wide variety of infections, immunizations, and vaccinations could induce similar sensitization to myelin basic protein has not been satisfactorily explained. Only the complications following inoculation against rabies with the nerve tissue antirabies vaccine are strictly comparable.

The histopathological features in postinfectious diseases of the nervous system were recently reviewed at length by Poser [32], who noted that vascular changes almost invariably accompany demyelination. To stress the importance of vascular involvement, he used the term disseminated vasculornyelinopathy and suggested that vasculopathy may both precede and initiate the nervous tissue damage itself. Vascular alterations do occur in EAE and may in fact be more prominent, overshadowing demyelination, when the course is fulminant [15].. However, the recent detection of circulating antigen-antibody complexes in serum from patients with a variety of postinfectious disordeis [16, 19, 39] suggests an alternative explanation—one that accounts for a primary vascular lesion, secondary damage to myelin, and a uniform response to a wide variety of precipitating events.

The Hypothesis

Common to all the postinfectious and postvaccinal complications affecting the nervous system is the initial introduction of a foreign antigen, through either invasion or inoculation, followed by a variable period free of nervous system symptoms and then clinical nervous system involvement. During this asymptomatic period, antibodies could form and combine with the foreign antigen to form circulating immune complexes. If such complexes are of appropriate size and contain the correct proportion of antigen to antibody, they can cause systemic liberation of vasoactive substances. An increase in vascular permeability results, complexes are trapped in vessel walls in a focal fashion, complement is activated, and inflammatory cells accumulate and release proteolytic enzymes, causing tissue injury [11]. Diffuse and multifocal involvement of both the central and peripheral nervous systems could result from such a vasculopathy, with the occurrence of limited or localized forms of nervous system injury depending on the size and number of complexes formed, the inherent properties of the vessels, and the chance occurrence of localized deposition. Because perivascular demyelination can result from vascular injury alone [32], the participation of delayed hypersensitivity would not be necessary. The host and its ability to make antibody, not the antecedent illness, would thus be the major factor in development of neurological disease:

the clinical and histopathological features would be the same regardless of the initiating event.

Examination of the clinical, laboratory, and histopathological features of disseminated vasculomyelinopathy and comparison with other human disorders in which immune complexes circulate and cause, or are thought to cause, vascular injury lends support to this hypothesis.

Disseminated Vasculomyelinopathy

The Clinical and Pathological Spectrum

The data collected by Miller, Stanton, and Gibbons [26] establish that a wide variety of clinical disturbances follow such common infections as rubeola, rubella, varicella, scarlet fever, and mumps after a similar latent period of a few days to several weeks. The disease is monophasic and the clinical onset usually abrupt, with signs of multiple lesions throughout the nervous system. In addition to encephalitis, which is frequently multifocal, cerebellar ataxia, aseptic menigitis, chorea, myelitis, cranial neuritis (including optic neuritis), mononeuritis, brachial neuritis, and Guillain-Barré syndrome all arise, either singly or in combination [26, 32]. The occurrence of these abnormalities is apparently independent of the nature, severity, and cause of’ the antecedent infection [26, 32],and with only rare exceptions, the virus has not been isolated from the nervous system in such cases [29]. The. same broad range of nervous system abnormalities occurs after a variety of immunizations and vaccinations [25], and clinically identical illness may follow banal upper respiratory infections and gastrointestinal disturbances, or even occur sporadically without identifiable antecedent infection [32].

The pathological changes in all these cases— postinfectious and postvaccinal as well as apparently sporadic—are identical [10, 29, 32]. Although current textbooks of neuropathology emphasize perivenous lymphocytic infiltration and demyelination in cerebral white matter [10, 29], a wide variety of pathological reactions actually occurs at all levels of the nervous system. Lesions of small blood vessels are consistently present and are more prominent and more frequent than demyelination in patients dying soon after onset [17, 26, 32]. The lesions range from perivascular edema and lymphocytic infiltrates and endothelial proliferation in the usual postinfectious encephalitis to vessel wall necrosis, fibrinoid degeneration, infiltrates of neutrophils and eosinophils, and perivascular hemorrhage in acute hemorrhage leukoencephalitis [10, 17, 26, 29, 32]. Identical lesions occur in arterioles and capillaries as well as venules [10, 32] and are frequently more common in the cerebral cortex, basal ganglia, and gray matter of brainstem and spinal cord than in white matter [10, 17]. The frequency of demyelination may also be overestimated: axons are often substantially affected in addition, and a clear differentiation between perivascular demyelination and necrosis has not always been made [17, 32].

The frequent occurrence of these various syndromes after the same antecedents, often in combination with one another, and their similar pathological features suggest they all may share a common pathogenesis. Conventional wisdom would, however, remove some of them from consideration, especially the Guillain-Barré syndrome. This objection is due, in large part, to the insistence of many clinicians that the term be restricted to those patients in whom only the spinal roots and peripheral nerves are affected. The problem may be largely one of semantics. Transitional cases with central nervous system involvement have been reported frequently, with nervous system abnormalities ranging from isolated Babinski signs, meningismus, or electroe ncephalographic changes to frank encephalomyclitis [9, 18, 33, 36]. Even in cases without clinical central nervous system involvement, lesions, mainly vascular, of the central nervous system are present postmortem in up to 20% and include vascular dilatation in the spinal cord and leptomeninges vith meningeal inflammatory cell infiltrates, perivascular lymphocytic infiltrates in cerebral white matter and subependyma; and petechial hemorrhages in the gray matter of brain and spinal cord [18, 36].

Evidence for Immune Complex Disease

These arguments are of some importance in the case of Guillain-Barré syndrome, since circulating immune complexes have been detected by both the Raji cell assay and binding of the Clq component of complement in up to 93% of affected patients [16, 39]. Although immune complexes have not been localized at sites of vessel injury, some circumstantial evidence suggests they may be involved in pathogenisis. Typical polyneuritis occurs in the course of serum sickness and systemic lupus erythematosus, both disorders thought to be caused by immune coplex vasculitis, as well as in association with a variety of infections and immunizations [4, 14, 25]. Onset is usually one to three weeks after the initiating event in postvaccinal and postinfectious cases; but it is worth noting that polyneuritis following typhoid-pararyphoid immunization usually begins three days following the second injection (its onset sometimes coinciding with the appearance of an urticarial rash [4, 25]), since the "incubation" period for serum sickness is similarly shortened to one to four days by a previous sensitizing injection [11]. The pathogenesis is assumed to be the same in all cases since the pathological features are the same [4]. Vascular changes occur first (vasodilatation, endothelial swelling, perivascular edema, and infiltration of lymphocytes and, occasionally, neutrophils) and are followed only later by perivascular demyelination [4,18].

Some systemic features of the Guillain-Barré syndrome are also consistent with immune complex disease: myalgias and arthralgias are frequent [4, 18], muscle tenderness is common [18], skin rash may occur at onset [4, 25], the sedimentation rate and leukocyte count are elevated [18, 36], and albuminuria and azotemia occur [18, 36]. In some patients with albuminuria, immune complex glomerulonephritis has been present, and epimembranous deposits of 1gM, IgG, and C3 have been identified by immunofluorescence [5, 31]. Pathological changes similar to those of serum sickness also occasionally occcur in other organs: perivascular inflammatory cell infiltrates in skeletal muscle and heart, focal mononuclear and polymorphonuclear infiltrates in liver, and focal phlebitis of the coronary veins have all been described [9, 18].

Some evidence is also available to suggest that circulating immune complexes may be involved in other postinfectious disorders. Complexes have been detected in 45% of patients with monosymptomatic optic neuritis and in single cases of acute disseminated encephalomyelitis and poststreprococcal encephalitis [19, 39] An association between postinfectious encephalitis and renal injury, a frequent manifestation of immune complex disease, has also been noted [7]; about 25% of patients with acute hemorrhagic leukoencephalitis have proteinuria [6] and occasional cases of postvaccinal encephalitis are associated with systemic vasculitis [38].

Human Immune Complex Disease

Serum sickness, the model for immune complex disease in humans, develops experimentally when antigen-antibody complexes-formed in the presence of antigen excess appear in the circulation. Serum complement falls, and vascular lesions develop in which immune complexes containing antigen, antibody, and components of complement can be demonstrated by immunofluorescence [11]. Various lesions occur, ranging from endothelial proliferition, increased vascular permeability, and variable infiltration of vessel walls by neutrophil leukocytes to perivascular mononuclear cell infiltration and even noninflammatory, degeneration, depending on the amount of antigen-antibody interaction. The tissue response is apparently independent of the antigen administered since a wide variety of serum proteins produce the same lesions. The amount of antibody formed is the key to subsequent development of immune complex disease: complexes formed when antibody synthesis is minimal are too small to activate complement or be retarded at vascular filtering surfaces, and complexes formed at marked antibody excess are insoluble. The character of the host as an antibody former is thus paramount in determining the development of immune complex disease, and disease may be more likely in relatively poor antibody formers [11].

Clinically, serum sickness in humans begins seven to ten days after the injection of foreign protein, with fever, malaise, urticaria, joint pains, lymphadenopathy, and immune complex nephritis [11]. Like Guillain-Barré syndrome, it has developed in immunosuppressed renal transplant recipients [21].

Neurological abnormalities are frequent in serum sickness, sometimes occur without accompanying systemic symptoms, and resemble the nervous system abnormalities of disseminated vasculomyelinopathy both clinically and pathologically Like disseminated vasculomyelinopathy, the disease is monophasic, with multilevel involvement throughout the nervous system. Most common is mild meningoencephalitis; headache, nausea, and vomiting are common at onset, 50% of patients have lymphocytic pleocytosis [22], and mild encephalopathy is almost always present [30]. Papilledema, aphasia, hemiplegia, hemianopia, coma, chorea, cerebellar syndromes, cranial neuropathies, radiculoneuritis (especially brachial neuritis), the Guillain-Barré syndrome, and acute transverse myelitis have all been reported in addition [22, 24, 30]. Pathologically, arterioles, venuies, and capillaries in the brain, spinal cord, and meninges are congested and hyalinized their endothelium is swollen, and there is perivascular edema, demyelination, hemorrhage, and round cell infiltration plus areas of focal necrosis in both brain and spinal cord. Similar changes occur in spinal roots, and there is proliferation of interfascicular connective tissue and Schwann cells in peripheral nerve [22 24].

Similar nervous system disease also occurs in a number of other human disorders in which immune complexes circulate and cause vascular injury. In essential mixed cryoglobulinemia the main pathological feature is a small vessel vasculitis with perivascular lymphocytic infiltration [23], and immunoglobulin and complement can be identified at sites of vessel injury [8]. Although immune complexes have not been demonstrated within the nervous system, polyneuritis, mononeuritis multiplex, cranial neuritis, myelopathy, blindness, hemiplegia, and encephalopathy have all been reported [1, 8]. Most cases of essential cryoglobulinemia probably represent cryoglobulinemia secondary to infection, and immune complexes containing the infective agent, antibody to the agent, and complement components constitute the cryoprecipitate. The hepatitis B virus is usually the agent responsible [23], but cryoglobulinemia with encephalopathy has also been reported following streptococcal infection [19]. In addition, serum cryoglobulins containing the virus and antibody directed against it are frequently present in infectious mononucleosis [41]. Multilevel nervous system abnormalities occur as well, usually begin one to three weeks after the onset of systemic symptoms, and have as their base vascular and perivascular changes which resemble those seen following the exanthems [2, 37]. Cryoglobulinemia is also common in cytomegalovirus infection [41], which may precede up to 33% of cases of Guillain-Barré syndrome [12]. In Lyme disease, mixed cryoglobulins are present at the onset of multilevel nervous system disease and disappear as active nervous system disease resolves [34]. Finally, circulating immune complexes are present in up to 100% of patients with systemic lupus erythematosus [40], and multilevel nervous system abnormalities develop in as many as 75% [14, 20] as a result of small vessel vasculopathy histologically similar to disseminated vasculomyelinopathy [20]. Immunoglobulin has been identified in the choroid plexus, cerebrospinal fluid complement is sometimes reduced, and immune complexes have been detected in the cerebrospinal fluid [14]; the immune complexes have not been identified at sites of vessel injury, however.


The numerous postinfectious and postvaccinal nervous system disorders appear to share a common pathogenesis involving the immune system: they follow the same antecedent events by a similar latent period, direct nervous system infection has rarely been demonstrated, and their occurrence depends more on the character of the host than on the severity of the initial infection. In the traditional view, delayed hypersensitivity to myelin develops and results in perivenous demyelination. However, how this hypersensitivity might develop after such diverse antecedents has never been satisfactorily explained, and sensitized lymphocytes are not always demonstrable by laboratory test. Damage to small blood vessels appears to precede demyelination in humans, and although it now seems that vascular damage does occur experimentally in EAE, it is possible that vasculopathy is primary in these disorders and that demyelination is a consequent change.

The detection of circulating immune complexes in sera from some patients with postinfectious nervous system diseases suggests that foreign antigens, introduced by infection or inoculation, may induce the formation of antibodies and, subsequently, of antigen-antibody complexes which fix complement and damage vascular endothelium. Myelin could be damaged secondarily. The attractiveness of this hypothesis lies in its ability to explain: (1) the monophasic character of the illness, (2) the common reaction to a wide variety of antecedents, and (3) the dependence of the reaction on the character of the host (as an antibody former).

These two hypotheses are not mutually exclusive. Even though small vessel injury in itself can eventuate in demyelination, this may be mediated through cellular immune mechanisms. Changes in vascular permeability and perivascular inflammation could either alter the antigenicity of myelin or release antigen previously sequestered by a competent blood-brain barrier. The cell-mediated immune response could then perpetuate the damage initiated by the original vascular injury.

Although some systemic features, both clinical and pathological, of postinfectious nervous system disease resemble those of human immune complex disease, in which similar neurological and neuropathological abnormalities also occur, these constitute the exception rather than the rule. Why the nervous system should be preferentially involved in immune complex—mediated vascular injury is not clear. One could speculate, however, that the consequences of minor vascular injury and alteration in permeability are more serious in the nervous system, where normal function depends on an impenetrable vascular barrier, than in other organ systems in which the vascular bed is already permeable.

Initial immune complex—mediated vascular injury could explain the observed facts in disseminated vasculomyelinopathy; however, only some indirect evidence supports the hypothesis: the detection of circulating complexes in some patients with these disorders, the presence of systemic features compatible with immune complex disease in occasional patients, and the occurrence of similar nervous system abnormalities in other human disorders caused by immune complexes.

Other interpretations of these observations are clearly possible. The detection of circulating immune complexes in some patients with amyotrophic lateral sclerosis [28] suggests that the complexes themselves may represent an epiphenomenon—either a consequence of antecedent infection without pathogenetic role or the result of nervous tissue injury occurring by other mechanisms and resulting in release of antigen. Alternatively, immune complexes might be involved in pathogenesis but exert their influence by interacting with suppressor cells, allowing autoimmunity to develop [3]. In view of their relative infrequency, the systemic features in some patients might indeed result from immune complex disease while the nervous system abnormalities arise by entirely different mechanisms. Finally, the nervous system reactions to injury may be so limited that both sensitized lymphocytes and antigen-antibody complexes cause similar lesions, explaining the apparently similar clinical and pathological features in disseminated vasculomyelinopathy and human immune complex disease.

Testing the Hypothesis

Techniques are now at hand for detecting circulating immune complexes. Cryoprecipitation has long been available and may be most effective in isolating complexes formed in the presence of antigen excess [42]. The Raji cell radioimmunoassay uses complement receptors on the Raji cell surface to detect complement containing antigen-antibody complexes [40] Complexes may also be detected by their ability to bind the Clq component of complement [27]. All these methods should be applied to the serum of patients with the various postinfectious disorders. Circumstantial evidence of a pathogenetic role for these complexes can be provided by demonstrating a fall in serum complement components, but complement consumption and activation may occur without a decrease in serum complement if both its synthesis and catabolism are accelerated [42]. More direct evidence of a pathogenetic role requires immunofluorescent study of tissue obtained postmortem or by biopsy with demonstration of immunoglobulin, complement, and, when it is known, the inciting antigen at sites of vessel injury [27]. Obviously, the chances of demonstrating tissue deposition of immune complexes will be greatest in pathological material obtained soon after the onset of illness since complexes may be eliminated long before their injurious effect on vessels has been repaired.


1.Abramsky O, Slavin S: Neurologic manifestations in patients with mixed cryoglobulinemia. Neurology (Minneap) 24: 245—249, 1974
2. Ambler M, Stoll J, Tzamaloukas A, et al: Focal encephalomyelitis in infectious mononucleosis. Ann Intern Med 75:579—583, 1971~

3. Antel JP, Arnason BGW, Medof ME: Suppressor cell function in multiple sclerosis: correlation with clinical disease activity. Ann Neurol              5:338—342, 1979
4.Arnason BGW: Inflammatory polyradiculoneuropathies, in Dyck PJ, Thomas PK, Lambers EH (eds): Peripheral Neuropachy. Philadelphia,              Saunders, 1975, vol 2, pp 1110—1148
5. Behan PO~ Lowenstein LM, Silmant M, et al: Landry-Guillian-Barré-Strohl syndrome and immune-complex nephritis. Lancet 1:850—854,          1973
6. Behan P0, Moore MJ, Lamarche JB: Acute necrotizing hemorrhagic encephalopathy. Postgrad Med 54:154—160, 1973
7. Brain WR, Hunter D, Turnbull HM: Acute meningoencephalomyelitis of childhood. Lancet 1:221—227, 1929
8. Brouec JC, Clauvel JP, Danon F, cc al: Biologic and clinical significance of cryoglobulins. A report of 86 cases. Am J Med 57:775—788, 1974
9. Cambier J, Schott B: Nosologie des polyradiculonévrites inflammatoires. Rev Neurol (Paris) 115:811—842, 1966
10. Carpenter S, Lampert PW: Post-infectious perivenous encephalitis and acute hemorrhagic leukoencephalitis, in Minckler J (ed): Pathology of          the Nervous System. New York, McGraw-Hill, 1972, vol 3, pp 2260—2269
11 Cochrane CG, Dixon FJ: Immune complex injury, in Samter M (ed): Immunological Diseases. Third edition. Boston, Little, Brown, 1978, Vol 1,          pp 210—229
12. Dowling P, Menonna J, Cook S: Cytomegalovirus complement fixation antibody in Guillain-Barré syndrome. Neurology (Minneap)                      27:1153—1156, 1977
13. Drachman DA, Paterson PY, Berlin BS, et al: Immunosuppression and the Guillain-Barré syndrome. Arch Neurol 23:385—393, 1970
14. Feinglass EJ, Arnett FC, Dorsch CA, et al: Neuropsychiatric manifestations of systemic lupus erythematosus: diagnosis, clinical spectrum,          and relationship to other features of the disease. Medicine (Baltimore) 55:323—339, 1976
15 Field EJ: The brain and nervous system in allergic disease, in Gell PGH, Coombs RRA, Lachmann PJ (eds): Clinical Aspects of Immunology.          Third edition. London, Blackwell, 1975, pp 1545—1586
16. Goust JM, Chenais F, Carnes JE, cc al: Abnormal T cell sub-populations and circulating immune complexes in the Guillain-Barré syndrome          and multiple sclerosis. Neurology (Minneap) 28:421—425, 1978
17. Hart MN, Earle KM: Haemorrhagic and perivenous encephalitis: a clinical-pathological review of 38 cases. J Neurol Neurosurg Psychiatry          38:585—591, 1975
18. Haymaker W, Kernohan JW: The Landry-Guillain-Barré syndrome: a clinicopathologic report of fifty fatal cases and a critique of the                  literature. Medicine (Baltimore) 28:59—14 1, 1949
19 Hodson AK, Doughty RA, Norman ME: Acute encephalopathy, streptococcal infection and cryoglobulinemia. Arch Neurol 35:43—44, 1978
20. Johnson RT, Richardson EP: The neurological manifestations of systemic lupus erythematosus. A clinical-pathological study of 24 cases          and review of the literature. Medicine (Baltimore) 47:337—369, 1968
21. Kashiwagi N, Brantigan CO, Brettschneider L, et al: Clinical reactions and serologic changes after the administration of heterologous                  antilymphocyte globulin to human recipients of renal homografts. Ann Intern Med 68:275—286, 1968
22. Kraus WM, Chancy LB: Serum disease of the nervous system. Arch Neurol Psychiatry 37:1035—1047, 1937
23. Levo Y, Gorevic PD, Kassab HJ, et al: Association between hepatitis B virus and essential mixed cryoglobulinemia. N EnglJ Med                      296:1501—1504, 1977
24 . Miller HG: Clinical manifestations of tissue reaction in the nervous system, in Williams D (ed): Modern Trends in Neurology (Second              Series). London, Butterworth, 1957, pp 164—176
25. Miller HG, Stanton JB: Neurological sequelae of prophylactic inoculation. QJ Med 23: 1—27, 1954
26. Miller HG, Stanton JB, Gibbons JL Para-infectious encephalomyelitis and related syndromes. QJ Med 25:427—505, 1956
27. Nakamura RM: Immunopathology: Clinical Laboratory Concepts and Methods. Boston, Little, Brown; 1974, pp 168— 200
28. Oldstone MBA, Wilson CB, Perrin LH, et al: Evidence for immune-complex formation in patients with amyotrophic lateral sclerosis. Lancet          2:169—172, 1976
29. Oppenheimer DR: Demyelinating diseases, in Blackwood W, Corsellis JAN (eds): Greenfields Neuropathology. Third edition. Chicago, Year          Book, 1976, pp 470—499
30. Park AM, Richardson JC: Cerebral complications of serum sickness. Neurology (Minneap) 3:277—283, 1953
31. Peters DK, Sevitt LH, Direkze M, et a!: Landry-Guillain. Barré-Strohl polyneuritis and the nephrotic syndrome. Lancet 1:1183—1184, 1973
32. Poser CM: Diseases of the myelin sheath, in Baker AB, Raker LH (eds): Clinical Neurology. Hagerstown, MD, Harper & Row, 1978, vol 2, pp          80—104
33. Poser CM, Fowler CW: The nosologic situation of the Landry-Guillain-Barré syndrome. Acta Neurol Scand 39: 187—201, 1963
34. Reik L, Steere AC, Bartenhagen NH, et al: Neurologic abnormalities of Lyme disease. Medicine (Baltimore) 58:28 1— 294, 1979
35. Rocklin RE, Sheremata WA, Feldman RG, et al: The Guillain-Barré syndrome and multiple sclerosis: in vitro cellular responses to nervous              tissue antigens. N Engl J Med 284: 803—808, 1971
36. Schaltenbrand G, Bammer H: La clinique et le traitement des polynévrites inflammatoires ou séreuses aiguës (syndrome de Landry, Guillain,              Barré). Rev Neurol (Paris) 115:783—810, 1966
37. Schnell RG, Dyck PJ, Bowie EJW, et al: Infectious mononucleosis: neurologic and EEG findings. Medicine (Baltimore) 45:51—63, 1966
38. Spillane JD, Wells CEC: The neurology of Jennerian vaccination. Brain 87:1—44, 1964
39. Tachovsky TG, Lisak RP, Koprowski H, et al: Circulating immune complexes in multiple sclerosis and other neurological diseases. Lancec              2:997—999, 1976
40. Theofilopoulos AN, Wilson CB, Dixon FJ: The Raji cell radioimmune assay for detecting immune complexes in human sera. J Clin Invest              57:169—182, 1976
41. Wager O, Rasanen JA, Hagman A, et al: Mixed cryoimmunoglobulinaemia in infectious mononucleosis and cytomegalovirus                                mononucleosis. Inc Arch Allergy AppI Immunol 34:345—361, 1968
42. Wands JR, Dienstag JL, Bhan AK, et al: Circulating immune complexes and complement activation in primary biliary cirrhosis. N Engl J Med          298:233—237, 1978