J Neurol Neurosurg : first published as 10.1136/jnnp.50.5.635-a on 1 May 1987. Downloaded from

Journal of Neurology, Neurosurgery, and Psychiatry 1987;50:635-643 Department of Neurology, the H6pital Neurologique, Lyon since its Letters Middlesex Hospital, inauguration in 1965. We found 28 patients. Mortimer Street, Short REM latency was recorded by Cigarette smoking, Parkinson's disease and London WIN 8AA, UK 24-hour at least once in ulcerative colitis 24 patients. The other four patients had typ- References ical - attacks. 50 HLA- Sir: Individuals who develop Parkinson's A, B and C, 12 HLA-DR and 2 HLA-DQ HA. The Dorn study of smoking and are about twice as likely to have been I Kahn serologically defined antigens were studied disease mortality among US veterans. Report on habitual non-smokers when compared with eigh. and one half years of observations. In in every patient as previously described.4 a control population.`1 Smoking also Epidemiological Approaches to the Study of Typing for HLA-DW2 and its subsets was seems to exert considerable protective effects Cancer and other Chronic Diseases. Mono- also performed as reported. 5 Results were against ulcerative colitis.46 It has been graph No. 19. Washington DC, National compared with a control population of Cau- suggested that both these illnesses may be Cancer Institute. US Govt. Printing Office casian blood donors. Chi-square test was associated with inflexible, morose, inward- 1966: 1-125. used with a correction of probability values looking personalities.7" One of us was 2 Kessler II, Diamond KL. Epidemiological according to the number of HLA antigens struck by what was possibly an un- studies of Parkinson's disease. 1. Smoking studied. The level of significance was chosen and Parkinson's disease. Am J Epidemiol < recognised association between these two 197 1;94: 16-25. at p 0 05. conditions, five patients with both illnesses 3 Godwin-Austen RB, Lee PN, Marmot MG, Increases in the proportions of cases with coming to light over a three year period. Stern GM. Smoking and Parkinson's disease. HLA-DQWI (100%), DR2 (92 9%), DW2 With the help of a letter expressing interest J Neurol Neurosurg Psychiatry' 1982;45: (92-9%) and B7 (50%) antigens were the in this link which was published in the UK 577-81. only significant observed differences when Parkinson's Disease Society Newsletter and 4 Jick H, Walker AM. Cigarette smoking and narcoleptic patients were compared with further postal questionnaires and corre- ulcerative colitis. N Engl J Med 1983; controls (table). The 26 DR2 positive nar- spondence with the patients' general prac- 308:261-3. coleptic patients were all DR2 long/ diseases 5 Logan RFA, Edmond M, Somerville KW, titioners, 20 more people with both Langman MJS. Smoking and ulcerative col- DQWI/DW2. were found. The mean age of all 25 patients itis. Br Med J 1984;288:751-3. Two patients were DR2/DW2 negative. Protected by copyright. (14 men, 11 women) was 62 years (range 6 Vessey M, Jewell D, Smith A, Yeates D, The first one was a Caucasian 54 year old 43-79); six were in social class 1, eight in McPherson K. Chronic inflammatory bowel female phone operator with A2 A30/B7 social class 2, six in social class 3, four in disease, cigarette smoking and use of oral BW62/CW3/DRWI3 DRW14/DQWI anti- social class 4 and one in social class 5. In 19 contraceptives: findings in a large cohort of women of childbearing age. Br Med gens. Typical sleep attacks began at age 15 the ulcerative colitis preceded the Parkin- study years, up to 5 times per day. They were cor- son's disease, sometimes by many years. J 1986;292:1 101-3. 7 Bellini M, Tansella M. Obsessional scores and rected initially with 75 mg desipramine. The Twenty of the patients had never smoked subjective general psychiatric complaints of drug was stopped owing to dizziness and tobacco, three had given up 25, 21 and 10 patients with duodenal ulcer or ulcerative replaced by 100 mg clomipramine with good years ago respectively; one was a very oc- colitis. Psychol Med 1976;6:461-7. efficacy and tolerance. Excessive daytime casional pipe smoker and the other patient 8 Todes CJ, Lees AJ. The pre-marked person- was another salient feature. had smoked ten cigarettes a day all his life. ality of patients with Parkinson's disease. J Nocturnal sleep was self-estimated as excel- The prevalence of non-smoking for an Neurol Neurosurg Psychiatry 1985;48: lent and, usually, 10 hours long. However, age-sex-social class matched population in 97-100. there was a transient fatigue on awakening. the United Kingdom would be about 50%.9 9 Office of Population Censuses and Surveys. General Household Survey: cigarette smoking No actual cataplexy attacks, Two of the ex-smokers in the study had 1972-1982. Govt. Statistical Service. GHC and hypnagogic hallucinations were experi- stopped smoking at least ten years before 83/2 July 1983. enced. A 24-hour polygraphic recording was the onset of their Parkinson's disease or made in February 1986 after a 2 weeks ces- a ulcerative colitis; another smoked pipe sation of drugs. It showed four sleep epi- http://jnnp.bmj.com/ very occasionally. Eighty per cent of the sodes with direct REM sleep. Total sleep patients had never smoked tobacco and only Accepted 6 September 1986 duration was 12 h 30 min with a relative three (12%) were smoking tobacco at all at excess of REM sleep (43%). Otherwise, the time of onset of one or other disease. In since puberty, there were typical symptoms these the quantity smoked was fairly small and signs of dystrophia myotonica (Stei- (15 cigarettes/day, 10 cigarettes/day and '/2 nert's disease) with myotonic discharges in oz (14 g) tobacco/week). Although derived the EMG. At the last examination in August from a highly selected cohort, these figures HLA-DR2 negative narcolepsy 1986, our patient was still minimally affected would be in keeping with the reported nega- by the muscle disorder in her daily activity. on September 24, 2021 by guest. tive association between smoking and both Sir: Association between narcolepsy and She had no respiratory insufficiency. A ulcerative colitis and Parkinson's disease. HLA-DR2 antigen is the strongest so far brother had both narcolepsy and Steinert's Further prospective epidemiological studies described between an HLA antigen and a disease. He died from myocardial infarction should examine the possible link between disease.`3 Among 28 narcoleptic patients, at age 44. Two sisters were operated on for these two diseases further and in particular we found two HLA-DR2 negative cases. We bilateral cataracts in their fourth decade. A the tantalising notion of a shared distinctive present their case-reports and discuss the brother and a sister were unaffected. We pre-morbid personality. implications of these data. were aware of one case of muscle disorder, K BIHARI We consulted the files of the patients two cases of early cataracts and one case of AJ LEES suffering from narcolepsy and referred to early baldness in the paternal lineage. 635 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.50.5.635-a on 1 May 1987. Downloaded from

636 Letters Table HLA groups with significantly different proportions ofnarcoleptic and control 7 Leygonie-Goldenberg F, Perrier M, Duizabo cases Ph, et al. Troubles de la vigilance, du som- meil et de la fonction respiratoire dans la HLA Narcolepsy* Controlst Chi-2 p (corrected) maladie de Steinert. Rev Neurol (Paris) 1977; 133:255-70. B7 14(50%) 14(11.3%) 225 «00005 8 Guilleminault C. Editorial Narcolepsy 1985. DR2 26 (92 9%) 46 (23%) 55-9 <«0 0005 Sleep 1986;9:99- 101. DQWI 28 (100%) 141 (70-5%) 11 0 <0 002 9 Langdon N, Lock C, Welsh K, et al. Immune DW2 26 (92 9%) 38 (19%) 65 9 «0 0005 factors in narcolepsy. Sleep 1986;9:143-8. *total number of cases = 28. 10 Neely SE, Rosenberg RS, Spire JP, Antel J, ttotal number of cases = 124 for HLA-A, -B and -C; 200 for HLA-DR,-DQW and -DW2 antigens. Arnason BGW. HLA antigens in narcolepsy. Neurology 1986;36 (Suppl 1):299 (abstract). 11 Matsuki K, Juji T, Tokunaga K, Naohara T, Satake M, Honda Y. Human histo- compatibility leukocyte antigen (HLA) hap- estimated from the The second DR2 negative narcoleptic questions need elucidation: why the nar- lotype frequencies data on HLA class I, II, and III antigens in 11I patient was a 35 year old lorry driver of colepsy gene a susceptibility is in such tight japanese narcoleptics. J Clin Invest Reunion Island origin. He was born and linkage desequilibrium with the DR2 and 1 985;76:2078-83. lived there until his moving to Lyon at age DQW1 alleles? What are the mechanisms 12 Font MP, Gebuhrer L, Betuel H, Freidel C, 29. Cataplexy attacks occurred 2 years later, linking the gene to the disease? Dausset J, Cohen D. HLA DR2, DR5, and soon followed by sleep episodes and severe DRW6 associated DW subtypes correlate nocturnal dyssomnia. Polysomnography with HLA-DR,B and DQJ Restriction Frag- CHRISTIAN CONFAVREUX,* ment Length Polymorphisms. Proc Natl showed short REM sleep latency. The LUCETTE GEBUHRER,t patient was unaware of a similar case in his Acad Sci USA 1986;83:3361-5. HERVE BETUEL,t 13 Inoko Ando Matsuki family. He was HLA A36 A26/B7 H, A, Tsuji K, K, Juji T, CATHERINE FREIDEL,t Honda Y. HLA-DQ,B chain DNA restriction B35/CW4/DRW 1I DRW13/DQWI HELENE BASTUJI,* fragments can differenciate between healthy DQW3. The presence of the A36 antigen GILBERT AIMARD,* and narcoleptic individuals with HLA-DR2. indicated his negroid ancestry which was MICHEL DEVIC,* Immunogenetics, 1986;23:126-8. Protected by copyright. otherwise visible. MICHEL JOUVET* In this study, we were able to confirm the Hopital Neurologique,* Accepted 13 September 1986 extraordinary association between nar- and Centre de Transfusion Sanguine, colepsy and HLA-DR2 antigen. However, Laboratoire d'histocompatibilitW,t Lyon, two patients were DR2 negative. One of France them was negroid. The other patient was remarkable as narcolepsy was combined with dystrophia myotonica, an exceptional Correspondence to C Confavreux, H6pital Pituitary apoplexy following isosorbide association.6 7 Neurologique, 59 boulevard Pinel, 69003, administration These data have several implications. Lyon, France. Firstly, they confirm that narcoleptic The "Association pour la Recherche sur la Sir: Pituitary apoplexy usually arises spon- patients, especially negroids, may be HLA Sclerose En Plaques" support is acknowl- taneously in previously unsuspected pitui- DR2 negative.8 It follows that the edged. tary macroadenomas. A variety of possible absence of the HLA-DR2 antigen is not precipitating factors have been reviewed by sufficient to reject the diagnosis. Above all, Bernstein et al who described the first of the gene coding for DR2 antigen is not per three reported cases of pituitary apoplexy se responsible for narcolepsy. All of our References following a pituitary stimulation test.' patients were HLA-DQW Ipositive, as were Apart from thyrotrophin-releasing http://jnnp.bmj.com/ the 156 patients DQ typed in the litera- 1 Langdon N, Welsh KI, Van Dam M, Vaughan hormonel 2 and bromocriptine3 no other ture.1 3 11 Studies on the correlations RW, Parkes D. Genetic markers in nar- vasoactive drugs have been implicated in the between HLA class II specificities and DNA colepsy. Lancet 1984;2:1178-80. aetiology of this condition. We describe a Restriction Fragments Length Poly- 2 Juji T, Satake M, Honda Y, Doi Y. HLA anti- patient who developed major apoplexy fol- morphism (RFLP) defined with gens in Japanese patients with narcolepsy. lowing oral isosorbide HLA-DQ,B Tissue Antigen 1984;24:316-9. mononitrate. cDNA have shown can probes that DQW1 3 Billiard M, Seignalet J. Extraordinary associ- A 54 year old man had an anterolateral be divided in at least three types,' 2 the same ation between HLA-DR2 and narcolepsy. myocardial infarction in August 1984 and being present in all narcoleptic patients.5 13 Lancet 1985;1:226-7. was commenced on atenolol 100 mg daily. The gene coding for DQWI may be the pri- 4 Betuel H, Gebuhrer L, Lambert J, Freidel AC, He underwent coronary angiography in on September 24, 2021 by guest. mary association with narcolepsy and the Farre A. A possible new HLA-DR allele. March 1985 and was advised to commence causal factor for the disease. Another possi- Hum Immunol 1983;8:227-37. isosorbide mononitrate (Elantan 20, Sanol bility is that it is simply closer to the hypo- 5 Marcadet A, Gebuhrer L, Betuel H, et al. DNA Schwarz) 20 mg twice daily in addition to thetical narcolepsy susceptibility gene than polymorphism related to HLA-DR2 DW2 in the atenolol. One to two hours after patients with narcolepsy. Immunogenetics taking the gene coding for DR2. Further work, 1985;22:679-83. the first tablet he developed a severe vertical using RFLP studies with new restriction 6 Coccagna G, Mantovani M, Parchi C, Mironi and left sided which lasted three enzymes or DNA sequencing, is necessary to F, Lugaresi E. Alveolar hypoventilation and hours and resulted in discontinuation of the test these hypotheses and to localise pre- in myotonic dystrophy. J Neu- isosorbide. One month later he attempted to cisely the susceptibility gene. Two more rol Neurosurg Psychiatry 1975;38:977-84. restart the drug but 90 minutes after taking