Iliac artery dissection in α1-antitrypsin deficiency

of the study and during treatment with the unified Parkinson’s disease rating scale (UPDRS), mini-mental state

beginning

and brief psychiatric rating scale (BPRS); moreover, the patients filled out, with family assistance, a hallucinosis questionnaire (scoring 0-5). The doses of risperidone were increased up to a maximum of 0-67 (0-47) mg per day (range 0-25-1-25); the mean duration of the treatment was 14-67 (7-34) weeks (range 8-24). Hallucinations improved significantly in all patients and disappeared in 3 (50%); the BPRS scores significantly improved, whereas the MMS remained unchanged (table). The basal motor symptoms measured by UPDRS did not worsen and the total daily levodopa doses did not change (before 556 [209] mg; after 622 [204]). The number of hours spent in an "on" state before and after risperidone was unchanged. Risperidone was well tolerated; only 2 patients reported mild increases in salivation and 1 other had temporary hypotension. These results suggest that risperidone may effectively and safely control hallucinations in levodopa-treated patients with Parkinson’s disease without causing a worsening of

(MMS),

extrapyramidal symptoms. G Meco, A Alessandria, V Bonifati, P Giustini Department of Neurosciences, "La Sapienza" University, 00185 Roma, Italy

1

2

3

4 5

Cummings JL. Neuropsychiatric complications of drug treatment of Parkinson’s disease. In: Huber SJ, Cummings JL, eds. Parkinson’s disease: neurobehavioural aspects. New York: Oxford University Press, 1992: 313-27. Cummings JL. Psychosis in neurologic disease: neurobiology and pathogenesis. Neuropsychiatr Neuropsychol Behav Neurol 1992; 5: 144-50. Zoldan J, Friedberg G, Goldberg-Stern H, Melamed E. Ondansetron for hallucinosis in advanced Parkinson’s disease. Lancet 1993; 341: 562-63. Friedman JH, Lannon MC. Clozapine in the treatment of psychosis in Parkinson’s disease. Neurology 1989; 39: 1219-21. Livingston MG. Risperidone. Lancet 1994; 343: 457-60.

Symptomatic junctional bradycardia due to lithium intoxication in patient with previously normal electrocardiogram SIR—Symptomatic tachyarrhythmia

and

bradyarrhythmia

induced by lithium have been reported in patients with and abnormal disease had a We a who electrocardiograms (ECG). report patient normal ECG who with previously symptomatic presented junctional bradycardia due to lithium intoxication. A 75-year-old woman collapsed in an orthopaedic hospital a week after an operation. She had no chest pain, but her ECG showed bradycardia for which she was given two doses of 600 mg of atropine intravenously and urgently transferred to our hospital. She had no history of cardiovascular disease or diabetes mellitus but had been taking phenytoin for epilepsy and lithium for depression for more than 20 years. Warfarin had been administered after surgery. On arrival,

pre-existing

conductive

conscious but withdrawn, her heart rate was 42 per blood min, pressure 90/60 mm Hg, general examination was unremarkable, and a 12-lead ECG showed junctional bradycardia with narrow complexes and no evidence of ischaemic or intraventricular conductive disorder. All drugs were discontinued. She had evidence of renal dysfunction (urea 10 mmol/L, sodium 135 mmol/L, potassium 4-2 mmol/L, and creatinine 211 )JLmol/L). Magnesium, calcium, and cardiac enzymes were normal. Her phenytoin and warfarin serum concentrations came back within the therapeutic range but lithium was 2-13 mmol/L (normal 0-8-1-2). After rehydration her condition improved and by the fourth day her ECG had returned to sinus rhythm with no evidence of ischaemic or conductive abnormalities. She has remained well and symptom-free until now (5 months after discharge). Sinus arrest and asystole have been reported in a patient with severe lithium intoxication but his recovery and followup ECG showed right bundle branch block and left axis deviation.’ Rosenquist and colleagues concluded that lithium treatment at therapeutic concentrations is unlikely to cause conductive disorders in patients with normal hearts and ECGs.2 In contrast, our case shows reversible symptomatic junctional bradycardia associated with lithium intoxication in a patient whose ECG suggests otherwise normal conduction. Electrophysiological study was not felt to be justified in view of her complete recovery. There has been no mention of phenytoin in therapeutic doses causing such an effect or potentiating lithium toxicity to the heart. she

was

Shaheer Farag, Robert D S Watson, David Honeybourne Department of Thoracic Medicine and Cardiology, Dudley Road Hospital, Birmingham B18 7QH, UK 1 2

Handler CE. Sinus arrest and asystole due to severe lithium intoxication. Int J Cardiol 1991; 30: 364-66. Rosenquist M, Bergfeldt L, Aili H, Mathè AA. Sinus node dysfunction during long term lithium treatment. Br Heart J 1993; 70: 371-75.

Ong ACM,

Iliac artery dissection in

&agr;1-antitrypsin

deficiency SIR—The role of al-antitrypsin &agr;1-AT) deficiency in the pathogenesis of arterial aneurysm or dissection is controversial. Schievink and colleagues (Feb 19, p 452) report 3 cases of ruptured intracranial aneurysm and 1 of spontaneous dissection of the cervical internal artery among 362 patients with ai-AT deficiency. Likewise, the heterozygous PiMZ phenotype is more common than predicted in patients with abdominal aortic aneurysms.’ On the other hand, Elzouki and Erikkson (April 23, p 1037), recorded no arterial abnormality in a retrospective study of 30 consecutive homozygous PiZZ patients who underwent necropsy. A shortened lifespan of homozygous PiZZ patients might explain these negative results. We report a patient in whom iliac artery dissection revealed &agr;1-AT deficiency. A 34-year-old man presented with a regressive acute arterial occlusion of the left leg which arose 48 h before 1371

admission while he was taking exercise. He had a history of moderately active ulcerative colitis since 1984 associated with antineutrophil cytoplasmic autoantibodies (ANCA) of unknown specificity. He was receiving 40 mg prednisolone daily for haemolytic anaemia diagnosed 4 months before admission. There were no family history or other risk factor of vascular disease. Physical examination showed a massive

left femoral thrill and bruit. Blood pressure, skin temperature, and colouration were normal, and tibial pulses were present. Emergency arteriography showed a dissection involving left primitive, internal, and external iliac arteries. Opacification of the aorta and its visceral branches and doppler sonography of cervical arteries were normal. ai-AT concentration was reduced (0-9 g/L). Heterozygous PiSZ phenotype was confirmed by isoelectric focusing. No symptom or laboratory findings suggested emphysema or liver disease. The patient was treated conservatively with low dose anticoagulant therapy. a l-AT deficiency is known to be associated with emphysema, liver cirrhosis, and panniculitis. The link between ai-AT deficiency and arterial injuries is more controversial. The young age of our patient and the absence of risk factor of vascular disease argue in favour of a link between iliac dissection and &agr;1-AT deficiency. An imbalance between proteolytic enzymes and inhibitors may precipitate arterial dissection or aneurysm formation. Of note, inflammatory vasculitis with ANCA directed to proteinase 3 might also be associated with ai-AT deficiency.2

Stéphane Cattan, Xavier Mariette, Françoise Labrousse, Jean-Claude Brouet Service d’Immuno-Hématologie, Hôpital Saint-Louis, 75010 Paris, France; and Laboratoire Central de Biochemie,

1

2

Hôpital Laennec, Paris

Cohen JR, Sarfati I, Ratner L, Tilson D. &agr;1-Antitrypsin phenotypes in patients with abdominal aortic aneurysms. J Surg Res 1990; 49: 319-21. Esnault VL, Testa A, Audrain M, et al. Alpha 1-antitrypsin genetic polymorphism in ANCA-positive systemic vasculitis. Kidney Int 1993; 43: 1329-32.

Why has coeliac disease increased in Swedish children? SIR—Coeliac disease has increased over the past decade in Sweden. In the 1970s, 1 case per 900 liveborns was reported but a three-fold increase to 1 in 300 has been observed since the mid-1980s throughout the country.’ Yet in some other European countries the incidence of coeliac disease in children is falling.The wider use of antibody screening to find atypical cases could contribute to the increase. However, this alone cannot explain most new cases since the disease now presents early, usually before 3 years of age, with typical symptoms of malabsorption.’ We suggest that changes in feeding practice could explain the increase. Before 1973 gluten could be introduced in the infant diet from the first weeks of life. In 1973 the introduction of gluten was postponed to 4 months and in 1983 to 6 months of age by recommendations from the National Board of Health and Welfare. In addition, in the early 1980s the amount of gluten doubled in follow-up formulas containing whole grain flour. Bottle feeding with a cereal-containing formula is traditionally practised in Sweden. Therefore, gluten is now usually introduced in a large amount with the bottle soon after the age of 6 months. The falling incidence of coeliac disease in children from other European countries has been explained as a result of a protective effect of breast feeding.3 We suggest two possible 1372

mechanisms by which breast milk can confer protection. First, milk of 97% of lactating mothers contains specific IgA to gluten; and, second, human milk exerts a T-cell suppressive effect. Milk IgA antibodies may diminish immune response to ingested foods by mechanisms such as agglutination of the antigen to immune complexes on the mucosal surface so that uptake is prevented, or by hiding of antigenic epitopes. Another possible immune-modulating property of human milk may be exerted by its T-cell-specific suppressive effect as shown by experiments on peripheral lymphocytes stimulated with phytohaemagglutinin, antiOKT3, and alloantigens.4 It has also been shown that the proliferative response of lymphocytes in response to phytohaemagglutinin is lower in breast-fed infants than in those fed cow’s milk.5Evidence from animal studies suggests that tolerance induction is facilitated by a continuous, early, and low exposure of the antigen, particularly where there is an excess of antibodies specific to the ingested antigen. In practical terms, this would imply that important foodstuffs should be introduced in the infant diet under the protection of breast feeding-ie, during the first months of life. In Sweden the median duration of breast feeding has been 5-6 months for the past 10 years. If the introduction of gluten is postponed to 6 months, the primary exposure to gluten in most children occurs without the suggested protective effect of breast milk. This change, with the increased gluten in follow-up formulas, and the abrupt manner of exposure of the offending antigen in a formula, could explain the increase in coeliac disease. An earlier more cautious exposure to a diet low in gluten, while the baby is still breast fed, might lower the incidence of coeliac disease in Swedish children. It has been argued that the current Swedish infant-feeding regimen provokes the development of coeliac disease in genetically predisposed individuals, which a Swedish and a recent Italian study may support (Jan 22, p 200). Since the clinical picture of coeliac disease is clearer in early childhood than later in life, fewer cases would be missed and treatment would be given early. However, additional cases may surface later in life. Moreover, the prevalence of coeliac disease can be questioned since according to a recent report from Estonia none of the 1461 investigated adults had the disease, yet in Estonia and Denmark, the annual incidence of coeliac disease in children has been 0-1per 1000 for the past 16 years (April 16, p 984). These figures emphasise that environmental factors are important for the development of coeliac disease. There is also an ethical dimension. Is it justified to deliberately provoke a disease in early childhood, which otherwise would not be necessary until decades later, if ever? Should we not rather prevent, as long as possible, disease onset

by appropriate feeding practices?

Per Juto, Gunnar Meeuwisse, Lucia Mincheva-Nilsson Department of Clinical Virology, University Hospital of Northern Sweden, S-901 85 Umeå, Sweden; Department of Pediatrics, Regional Hospital, Karlskrona; and Department of Immunology, University of Umeå 1

2

3

4

5

Cavell B, Stenhammar L, Ascher H, et al. Increasing incidence of coeliac disease in Sweden: results of a national study. Acta Paediatr 1992; 81: 589-92. Stevens FM, Egan-Mitchell B, Cryan E, McCarthy CF, McNicholl B. Decreasing incidence of coeliac disease. Arch Dis Child 1987; 62: 465-68. Greco L, Auricchio S, Mayer M, Grimaldi M. Case control study on nutritional risk factors in celiac disease. J Gastroenterol Nutr 1988; 7: 395-99. Mincheva-Nilsson L, Hammarström M-L, Juto P, Hammarström S. Human milk contains proteins that stimulate and suppress T lymphocyte proliferation. Clin Exp Immunol 1990; 79: 463-69. Juto P, Möller C, Engberg S, Bjorkstén B. Influence of type of feeding on lymphocyte function and development of infantile allergy. Clin

Allergy 1982;

12: 409-16.