Acta Neurol. Scandinav. 52, 161-166, 1975
Department of Neurology, University Hospital, Lund, Sweden, and the Tissue-Typing Laboratory and Blood-Grouping Department, State University Hospital (Rigshospitalet ), Copenhagen, Denmark.
HIA-A HISTOCOMPATIBILITY ANTIGENS I N OPTIC NEURITIS MAGNHILD SANDBERG-WOLLHEIM, P. PLATZ, L. P. RYDER,
I,. STAUB NIELSEN & M. THOMSEN ABSTRACT HL-A and MLC typing in 54 patients with optic neuritis showed increased frequencies of the HL-A3, 7 and LD-7a determinants of approximately the same magnitude as i n patients with multiple sclerosis. The frequencies of the same three determinants were not different in the 11 patients who developed multiple sclcrosis during the period of follow-up compared to the remaining patients. There was a significant increase in the frequency of the HL-A3 determinant in patients with oligoclonal IzG of the cerebrospinal fluid a t the onset of disease. The connection between this determicTant and the occurrence of cerebrospinal fluid changes is not clear. The results suggest that optic neuritis and multiple sclerosis represent different aspects of the same disease entity, rather than two distinct diseases. HL-A and MLC typing did not seem to offer prognostic information as to the later development of mnltiple sclerosis in these patients, but the observation period is still short.
The relationship between acute monosymptomatic optic neuritis and multiple sclerosis ( M S ) is still unclear. A number of authors (for references see Sandberg & Bynke 1973) have reported greatly differing results regarding the frequency with which patients wilh optic neuritis develop MS. Figures between 11 per cent and 76 per cent have been mentioned. Increased frequencies of the serologically defined ( S D ) histocompatibility (HL-A) antigens HL-A3 and 7 in patients with MS have been reported by several groups (for references see Ryder e f al. 1974). As shown by Jersild el al. (1973) the increase of these serologically defined antigens seems to be secondary t o a n increase of the This study was aided by the Foundation for Medical Research in Copenhagen, Greenland and Faeroe Islands, by the Danish and Swedish Multiple Sclerosis Association, the Danish Medical Research Council and by the Medical Faculty, University of Lund.
11
ACTA NEUROL. SCAND.
52,3
mixed lymphocyte culture (MLC) determinant, LD-7a, which is associated with HL-A7. Arnason et al. (1974) recently reported a difference in the HL-A antigen frequency between optic neuritis and MS; however, MLC typing was not performed. We have performed HL-A and MLC typing on 54 patients with acute monosymptomatic optic neuritis. The preliminary results were reported briefly in a letter ( P l a f zet al. 1975).
PATIENT MATERL41, Fifty-four patients with acute monosymptomatic optic neuritis, of t h e kind encountered in MS, were examined. Cases of toxic o r vascular etiology, diabetes, tumour etc. were excluded, as were cases with established o r possible diagnosis of MS. All the patients came from the southern part of Sweden and belong to a prospective study on optic neuritis which started i n 1969. Details on this patient material were presented elsewhere (Sandberg & Bgnke 1973, Sandberg-Wollheim, in preparation). There were 19 men and 35 women. The age a t onset of disease varied between 15 and 46 years (mean 27.6 years). The observation time ranged from 0.5 to 6 years (mean 2.6 years). During this time, 11 of the patients developed symptoms and signs from other lesions of the central nervous system, establishing the diagnosis MS. The cerebrospinal fluid (CSF) was examined in all cases at the onset of optic neuritis. In 47 cases, t h e CSF was examined repeatedly during the follow-up. At onset of symptoms, the CSF was normal i n 21 cases. There was a pleocytosis (mononuclear cell increase above 5 per mm3) and an oligoclonal IgG distribution on agar gel electrophoresis of the CSF in 19 patients. CSF pleocytosis but normal electrophoretic pattern were found in 8 cases. In 6 patients, there was an oligoclonal IgG but a normal cell count. Four of the 11 patients developing MS had a normal CSF a t the onset of optic neuritis; five had both a pleocytosis and a n oligoclonal IgG on electrophoresis, while two had pleocytosis but normal IgG. All but one showed an oligoclonal IgG later. The electrophoretic abnormality appeared before o r after the time of the dissemination of symptoms. Eight more patients, who have not had any symptoms o r signs of dissemination of the disease since the initial attack of optic neuritis, showed changes from normal to oligoclonal CSF IgG during the follow-up.
METHODS HL-A serotyping was performed by the microlymphocytotoxicity test (KissmeyerNielsen & K j e r b y e 1967), with typing for 23 different HL-A antigens. MLC typing was carried out as a one-way Mixed Lymphocyte Culture test with 105 stimulating and 105 responding cells ( T h o m s e n et al. 1974). Individuals with a stabilized relative response against a n LD-7a homozygous test cell below 50 per cent were classified as being LD-7a-positive (Thornsen et al. 1975).
RESULTS
The results of the HL-A and MLC typings are given for all 54 patients, as well as for the patients divided into various groups on the basis of the clinical and CSF findings. Table 1. H L - A antigen frequencies (per cent).
5 4 patients with optic neuritis
1967 controls
HL-A1 HGA2 HL-A3 HL-A9 HL-A10 HLAll HL-A28 w19
25.9 55.6 33.3 14.8 7.4 9.3 7.4 14.8
31.1 53.6 26.9 17.3 9.6 10.1 10.0 17.8
H L-A 5 HGA7 HL-A8 HGAl2 HL-A13 HL-A14 HL-A17 HL-A27 w5 w10 W15 W16 W18 w21 w22
11.1 38.9 18.5 7.4 3.7 0 3.7 7.4 11.1 27.8 24.1 0 13.0 5.6 3.7
10.6 26.8 23.1 25.2 4.3 4.5 7.7 8.6 13.1 17.9 17.9 5.4 7.1 3.5 3.8
Only the decrease of HL-A12 remains probably significant (P = 0.02) after multiplication by two (for one-sidedness of the Fisher's exact test) and 23, the number of antigens investigated.
The HL-A and MLC typing showed increased frequencies of HL-A3, 7 and LD-7a and a decrease of HL-A12 (Tables 1 and 2 ) . Eight patients carried HL-A3, 7 and LD-7a. Ten patients were LD-7a but not HL-A7 positive; while four patients carried the HL-A7 but not the LD-7a determinant. The frequencies of HL-A3 and LD-7a were a little, but not significantly, lower in optic neuritis than in MS. Patients with oligoclonal IgG of the CSF at onset of optic neuritis had a significantly ( P = 0.02) higher frequency of the HL-A3 determinant than patients with normal CSF (Table 3 ) . The frequencies of HL-A7 11'
164 Table 2 . Frequencies of HL-AS, 7 and L l l - 7 ~determinants in controls, patients w i t h multiple sclerosis ( M S ) , and patients w i t h optic neuritis.
Controls
MS
Optic neuritis
Optic neuritis patients who developed MS
No.
1967
233
54
11
HL-A3 % HL-A7 70
26.9 26.8
35 39
33 39
27 36
132
48
54
11
18
60"
50"
45
No.
LD-7a To
results from PIatz et al. (1974). * * - 1.2 x 10-7 P values by Fisher's'exact test : t t + - 1.8 x 10-5 Table 3. Frequencies o f HL-A3, 7 and LD-'la deierminants in 54 patients w i t h optic neuritis. The patients were divided into groups according t o the findings in the cerebrospinal fluid (CSF) at the onset o f disease.
Normal CSF
Pleocvtosis and/or oligoclonal IgG
Normal IgG
No.
21
33
29
25
27
27
HL-A3 ?lo HL-A7% LD-7a70
19 33 48
42 42
17 41 52
52 36 4s
26 30 44
41 48 55
* p
52
Oligoclonal Normal no. Pleocytosis IgG of cells
= 0.01 (Fisher's exact test).
and LD-7a were approximately the same in these groups. Patients with pleocytosis of the CSF had a higher incidence of all three determinants than did patients with a normal cell count; but the difference was not significant . At the time of the HL-A and MLC typing, eleven patients had developed new neurological symptoms, and therefore had MS. The frequency of all lhree determinants was approximate at the same level in these 11 patients as in the whole material (Table 2 ) . DISCUSSION
Patients with optic neuritis run a potential risk of developing MS, but it is not known how great the risk is. A number of retrospective studies
165
on this problem have been published, but the results have differed so much between the various reports that no conclusions can be drawn from them. A11 the patients in this study on HL-A and MLC typing belong to a prospective study on optic neuritis, which was undertaken: (1) to determine the risk of these patients developing MS; and ( 2 ) to find out, if possible, which patients remain well and which acquire new symptoms. The results obtained with HL-A and MLC typing in 54 patients with optic neuritis were very similar to those found in patients with MS. The increases of HL-A3 and 7 were approximately the same as for MS. The frequency of the LD-7a determinant was a litile, but not significantly, lower in optic neuritis Lhan in MS, and significantly increased when compared to normal controls. Eleven of the patients developed MS during the follow-up, and some of them did not carry any of the HL-A3, 7 or LD-?a determinants. HL-A and MLC typing, therefore, did not seem to offer any prognostic information as to the later development of MS. The observation period is still short, however, and it is possible that as more patients develop M S the results will allow firmer conclusions on this point. When the patient material was divided into groups according to the CSF findings at the onset of disease, there was a conspicuous increase in the frequency of the HL-A3 determinant in patients with oligoclonal IgG of the CSF, while there were no significant differences between these groups in the frequencies of the HL-A7 and LD-7a determinants. The increase of HL-A3 may be a chance deviation, and additional studies are needed to clarify whether patients carrying the HL-A3 determinant are more prone to develop oligoclonal IgG in CSF. It was reported by Arnason ef al. (1974) that normal controls as well as patients with MS carrying the HL-A3 determinant had higher serum levels of measles antibodies. Link et al. (1973) found an intrathecal production of antibodies against measles antigen in optic neuritis patients with oligoclonal CSF IgG but not in those with normal CSF. The increase of HL-A3 in optic neuritis and MS may, therefore, be a consequence of the increased frequency of oligoclonal IgG and increased levels of measles antibodies in these patient groups. Arnason ef al. (1974) recently reported the same frequency of HL-A3 in patients with optic neuritis and normal controls. The differences between their results and ours may well be explained by differences in the patient materials studied. The present material was carefully selected so as to include only patients with optic neuritis of the type seen in MS, while the material reported by Arnason ef al. may have included various kinds of optic neuritis. This seems likely since
it was a retrospective study, and since none of the patients had developed MS. ACKNOWLEDGEMENT Dr. A. Svejgaard is thanked for helpful discussion and criticism, and I. Christensen for preparation of the manuscript.
REFERENCES Arnason, B. G. W., T. C. Fuller, J. R. Lehrich & S. H. Wray (1974): Histocompatibility types and measles antibodies i n multiple sclerosis and optic neuritis. J. neurol. Sci. 22, 419-428. JersiId, C., G. S. Hansen, A. Svejgaard, T. Fog, M. Thomsen & B. Dupont (1973) : Histocompatibility determinants i n multiple sclerosis, with special reference to clinical course. Lancet ii, 1221-1225. Kissmeyer-Nielsen, F. & K. E. Kjerbye (1967) : In Histocompatibility Testing, p. 381. Munksgaard, Copenhagen. Link, H., E. Norrby & J. E. Olsson (1973) : Immunoglobulins and measles antibodies in optic neuritis. New Engl. J. Med. 289, 1103-1107. Platz, P., B. Dupont, T. Fog, L. Ryder, M. Thomsen, A. Svejgaard & C. Jersild (1974) : Mixed Lymphocyte Culture Determinants, measles infection and multiple sclerosis. Proc. roy. SOC.Med. 67, 1133-1136. Platz, P., L. P. Ryder, L. Staub Nielsen, A. Svejgaard, M. Thomsen & M. SandbergWollheim (1975) : HL-A and idiopathic optic neuritis. Lancet i, 520-521. Ryder, L. P., I,. Staub Nielsen & A. Svejgaard (1974) : Associations between HL-A histocompatibility antigens and non-malignant diseases. Humangenetik 2, 1-14. Sandberg, M. & H. Bynke ( 1 9 7 3 ) : Cerebrospinal fluid in 25 cases of optic neuritis. Acta Neurol. Scandinav. 49, 443-452. Sandberg-Wollheim, M. (1975) : Optic neuritis : Studies on the cerebrospinal fluid in relation to clinical course in 61 patients (submitted to Acta neurol scand.). Thomsen, M., G. S. Hansen, A. Svejgaard, C. Jersild, J. A. Hansen, R. A. Good & B. Dupont (1974) : Mixed Lymphocyte Culture technique. Tissue Antigens 4, 495-506. Thomsen, M., P. Platz, 0. Ortved Andersen, M. Christy, J. Lyngsee, J. Nerup, K. Rasmussen, L. P. Rgder, L. Staub Nielsen & A. Svejgaard (1975): MLC typing in juvenile diabetes mellitus and idiopathic Addison’s disease. Transplantation Reviews 22, 125-147. Received February 10, 1975.
M . Sandberg-Wollheim, M.D. Department of Neurology University Hospital Lund. Sweden
Department of Neurology, University Hospital, Lund, Sweden, and the Tissue-Typing Laboratory and Blood-Grouping Department, State University Hospital (Rigshospitalet ), Copenhagen, Denmark.
HIA-A HISTOCOMPATIBILITY ANTIGENS I N OPTIC NEURITIS MAGNHILD SANDBERG-WOLLHEIM, P. PLATZ, L. P. RYDER,
I,. STAUB NIELSEN & M. THOMSEN ABSTRACT HL-A and MLC typing in 54 patients with optic neuritis showed increased frequencies of the HL-A3, 7 and LD-7a determinants of approximately the same magnitude as i n patients with multiple sclerosis. The frequencies of the same three determinants were not different in the 11 patients who developed multiple sclcrosis during the period of follow-up compared to the remaining patients. There was a significant increase in the frequency of the HL-A3 determinant in patients with oligoclonal IzG of the cerebrospinal fluid a t the onset of disease. The connection between this determicTant and the occurrence of cerebrospinal fluid changes is not clear. The results suggest that optic neuritis and multiple sclerosis represent different aspects of the same disease entity, rather than two distinct diseases. HL-A and MLC typing did not seem to offer prognostic information as to the later development of mnltiple sclerosis in these patients, but the observation period is still short.
The relationship between acute monosymptomatic optic neuritis and multiple sclerosis ( M S ) is still unclear. A number of authors (for references see Sandberg & Bynke 1973) have reported greatly differing results regarding the frequency with which patients wilh optic neuritis develop MS. Figures between 11 per cent and 76 per cent have been mentioned. Increased frequencies of the serologically defined ( S D ) histocompatibility (HL-A) antigens HL-A3 and 7 in patients with MS have been reported by several groups (for references see Ryder e f al. 1974). As shown by Jersild el al. (1973) the increase of these serologically defined antigens seems to be secondary t o a n increase of the This study was aided by the Foundation for Medical Research in Copenhagen, Greenland and Faeroe Islands, by the Danish and Swedish Multiple Sclerosis Association, the Danish Medical Research Council and by the Medical Faculty, University of Lund.
11
ACTA NEUROL. SCAND.
52,3
mixed lymphocyte culture (MLC) determinant, LD-7a, which is associated with HL-A7. Arnason et al. (1974) recently reported a difference in the HL-A antigen frequency between optic neuritis and MS; however, MLC typing was not performed. We have performed HL-A and MLC typing on 54 patients with acute monosymptomatic optic neuritis. The preliminary results were reported briefly in a letter ( P l a f zet al. 1975).
PATIENT MATERL41, Fifty-four patients with acute monosymptomatic optic neuritis, of t h e kind encountered in MS, were examined. Cases of toxic o r vascular etiology, diabetes, tumour etc. were excluded, as were cases with established o r possible diagnosis of MS. All the patients came from the southern part of Sweden and belong to a prospective study on optic neuritis which started i n 1969. Details on this patient material were presented elsewhere (Sandberg & Bgnke 1973, Sandberg-Wollheim, in preparation). There were 19 men and 35 women. The age a t onset of disease varied between 15 and 46 years (mean 27.6 years). The observation time ranged from 0.5 to 6 years (mean 2.6 years). During this time, 11 of the patients developed symptoms and signs from other lesions of the central nervous system, establishing the diagnosis MS. The cerebrospinal fluid (CSF) was examined in all cases at the onset of optic neuritis. In 47 cases, t h e CSF was examined repeatedly during the follow-up. At onset of symptoms, the CSF was normal i n 21 cases. There was a pleocytosis (mononuclear cell increase above 5 per mm3) and an oligoclonal IgG distribution on agar gel electrophoresis of the CSF in 19 patients. CSF pleocytosis but normal electrophoretic pattern were found in 8 cases. In 6 patients, there was an oligoclonal IgG but a normal cell count. Four of the 11 patients developing MS had a normal CSF a t the onset of optic neuritis; five had both a pleocytosis and a n oligoclonal IgG on electrophoresis, while two had pleocytosis but normal IgG. All but one showed an oligoclonal IgG later. The electrophoretic abnormality appeared before o r after the time of the dissemination of symptoms. Eight more patients, who have not had any symptoms o r signs of dissemination of the disease since the initial attack of optic neuritis, showed changes from normal to oligoclonal CSF IgG during the follow-up.
METHODS HL-A serotyping was performed by the microlymphocytotoxicity test (KissmeyerNielsen & K j e r b y e 1967), with typing for 23 different HL-A antigens. MLC typing was carried out as a one-way Mixed Lymphocyte Culture test with 105 stimulating and 105 responding cells ( T h o m s e n et al. 1974). Individuals with a stabilized relative response against a n LD-7a homozygous test cell below 50 per cent were classified as being LD-7a-positive (Thornsen et al. 1975).
RESULTS
The results of the HL-A and MLC typings are given for all 54 patients, as well as for the patients divided into various groups on the basis of the clinical and CSF findings. Table 1. H L - A antigen frequencies (per cent).
5 4 patients with optic neuritis
1967 controls
HL-A1 HGA2 HL-A3 HL-A9 HL-A10 HLAll HL-A28 w19
25.9 55.6 33.3 14.8 7.4 9.3 7.4 14.8
31.1 53.6 26.9 17.3 9.6 10.1 10.0 17.8
H L-A 5 HGA7 HL-A8 HGAl2 HL-A13 HL-A14 HL-A17 HL-A27 w5 w10 W15 W16 W18 w21 w22
11.1 38.9 18.5 7.4 3.7 0 3.7 7.4 11.1 27.8 24.1 0 13.0 5.6 3.7
10.6 26.8 23.1 25.2 4.3 4.5 7.7 8.6 13.1 17.9 17.9 5.4 7.1 3.5 3.8
Only the decrease of HL-A12 remains probably significant (P = 0.02) after multiplication by two (for one-sidedness of the Fisher's exact test) and 23, the number of antigens investigated.
The HL-A and MLC typing showed increased frequencies of HL-A3, 7 and LD-7a and a decrease of HL-A12 (Tables 1 and 2 ) . Eight patients carried HL-A3, 7 and LD-7a. Ten patients were LD-7a but not HL-A7 positive; while four patients carried the HL-A7 but not the LD-7a determinant. The frequencies of HL-A3 and LD-7a were a little, but not significantly, lower in optic neuritis than in MS. Patients with oligoclonal IgG of the CSF at onset of optic neuritis had a significantly ( P = 0.02) higher frequency of the HL-A3 determinant than patients with normal CSF (Table 3 ) . The frequencies of HL-A7 11'
164 Table 2 . Frequencies of HL-AS, 7 and L l l - 7 ~determinants in controls, patients w i t h multiple sclerosis ( M S ) , and patients w i t h optic neuritis.
Controls
MS
Optic neuritis
Optic neuritis patients who developed MS
No.
1967
233
54
11
HL-A3 % HL-A7 70
26.9 26.8
35 39
33 39
27 36
132
48
54
11
18
60"
50"
45
No.
LD-7a To
results from PIatz et al. (1974). * * - 1.2 x 10-7 P values by Fisher's'exact test : t t + - 1.8 x 10-5 Table 3. Frequencies o f HL-A3, 7 and LD-'la deierminants in 54 patients w i t h optic neuritis. The patients were divided into groups according t o the findings in the cerebrospinal fluid (CSF) at the onset o f disease.
Normal CSF
Pleocvtosis and/or oligoclonal IgG
Normal IgG
No.
21
33
29
25
27
27
HL-A3 ?lo HL-A7% LD-7a70
19 33 48
42 42
17 41 52
52 36 4s
26 30 44
41 48 55
* p
52
Oligoclonal Normal no. Pleocytosis IgG of cells
= 0.01 (Fisher's exact test).
and LD-7a were approximately the same in these groups. Patients with pleocytosis of the CSF had a higher incidence of all three determinants than did patients with a normal cell count; but the difference was not significant . At the time of the HL-A and MLC typing, eleven patients had developed new neurological symptoms, and therefore had MS. The frequency of all lhree determinants was approximate at the same level in these 11 patients as in the whole material (Table 2 ) . DISCUSSION
Patients with optic neuritis run a potential risk of developing MS, but it is not known how great the risk is. A number of retrospective studies
165
on this problem have been published, but the results have differed so much between the various reports that no conclusions can be drawn from them. A11 the patients in this study on HL-A and MLC typing belong to a prospective study on optic neuritis, which was undertaken: (1) to determine the risk of these patients developing MS; and ( 2 ) to find out, if possible, which patients remain well and which acquire new symptoms. The results obtained with HL-A and MLC typing in 54 patients with optic neuritis were very similar to those found in patients with MS. The increases of HL-A3 and 7 were approximately the same as for MS. The frequency of the LD-7a determinant was a litile, but not significantly, lower in optic neuritis Lhan in MS, and significantly increased when compared to normal controls. Eleven of the patients developed MS during the follow-up, and some of them did not carry any of the HL-A3, 7 or LD-?a determinants. HL-A and MLC typing, therefore, did not seem to offer any prognostic information as to the later development of MS. The observation period is still short, however, and it is possible that as more patients develop M S the results will allow firmer conclusions on this point. When the patient material was divided into groups according to the CSF findings at the onset of disease, there was a conspicuous increase in the frequency of the HL-A3 determinant in patients with oligoclonal IgG of the CSF, while there were no significant differences between these groups in the frequencies of the HL-A7 and LD-7a determinants. The increase of HL-A3 may be a chance deviation, and additional studies are needed to clarify whether patients carrying the HL-A3 determinant are more prone to develop oligoclonal IgG in CSF. It was reported by Arnason ef al. (1974) that normal controls as well as patients with MS carrying the HL-A3 determinant had higher serum levels of measles antibodies. Link et al. (1973) found an intrathecal production of antibodies against measles antigen in optic neuritis patients with oligoclonal CSF IgG but not in those with normal CSF. The increase of HL-A3 in optic neuritis and MS may, therefore, be a consequence of the increased frequency of oligoclonal IgG and increased levels of measles antibodies in these patient groups. Arnason ef al. (1974) recently reported the same frequency of HL-A3 in patients with optic neuritis and normal controls. The differences between their results and ours may well be explained by differences in the patient materials studied. The present material was carefully selected so as to include only patients with optic neuritis of the type seen in MS, while the material reported by Arnason ef al. may have included various kinds of optic neuritis. This seems likely since
it was a retrospective study, and since none of the patients had developed MS. ACKNOWLEDGEMENT Dr. A. Svejgaard is thanked for helpful discussion and criticism, and I. Christensen for preparation of the manuscript.
REFERENCES Arnason, B. G. W., T. C. Fuller, J. R. Lehrich & S. H. Wray (1974): Histocompatibility types and measles antibodies i n multiple sclerosis and optic neuritis. J. neurol. Sci. 22, 419-428. JersiId, C., G. S. Hansen, A. Svejgaard, T. Fog, M. Thomsen & B. Dupont (1973) : Histocompatibility determinants i n multiple sclerosis, with special reference to clinical course. Lancet ii, 1221-1225. Kissmeyer-Nielsen, F. & K. E. Kjerbye (1967) : In Histocompatibility Testing, p. 381. Munksgaard, Copenhagen. Link, H., E. Norrby & J. E. Olsson (1973) : Immunoglobulins and measles antibodies in optic neuritis. New Engl. J. Med. 289, 1103-1107. Platz, P., B. Dupont, T. Fog, L. Ryder, M. Thomsen, A. Svejgaard & C. Jersild (1974) : Mixed Lymphocyte Culture Determinants, measles infection and multiple sclerosis. Proc. roy. SOC.Med. 67, 1133-1136. Platz, P., L. P. Ryder, L. Staub Nielsen, A. Svejgaard, M. Thomsen & M. SandbergWollheim (1975) : HL-A and idiopathic optic neuritis. Lancet i, 520-521. Ryder, L. P., I,. Staub Nielsen & A. Svejgaard (1974) : Associations between HL-A histocompatibility antigens and non-malignant diseases. Humangenetik 2, 1-14. Sandberg, M. & H. Bynke ( 1 9 7 3 ) : Cerebrospinal fluid in 25 cases of optic neuritis. Acta Neurol. Scandinav. 49, 443-452. Sandberg-Wollheim, M. (1975) : Optic neuritis : Studies on the cerebrospinal fluid in relation to clinical course in 61 patients (submitted to Acta neurol scand.). Thomsen, M., G. S. Hansen, A. Svejgaard, C. Jersild, J. A. Hansen, R. A. Good & B. Dupont (1974) : Mixed Lymphocyte Culture technique. Tissue Antigens 4, 495-506. Thomsen, M., P. Platz, 0. Ortved Andersen, M. Christy, J. Lyngsee, J. Nerup, K. Rasmussen, L. P. Rgder, L. Staub Nielsen & A. Svejgaard (1975): MLC typing in juvenile diabetes mellitus and idiopathic Addison’s disease. Transplantation Reviews 22, 125-147. Received February 10, 1975.
M . Sandberg-Wollheim, M.D. Department of Neurology University Hospital Lund. Sweden
