892 asthma (1184, 1004-1-351), and a lower incidence of arthritis (0-745, 0 594-0 932) and myasthenia gravis (0267, 0-099-0-722). Such results are not compatible with Geschwind, Behan, and Galaburda’s theory.z However, in the event of further replication they undoubtedly need explanation. One possibility is that left-handers and right-handers differ in their HLA haplotypes,6 thereby resulting in different patterns of disease. Academic Department of Psychiatry, St Mary’s Hospital Medical School, Imperial College of Science, Technology and Medicine, London W2 1PG, UK I. C. MCMANUS Department of Psychology, University of Waterloo, Waterloo, Ontario, Canada M. P. BRYDEN M. B. BULMAN-FLEMING 1. Meyers S, Janowitz HD. Left-handedness and inflammatory bowel disease. J Clin Gastroenterol 1985; 71: 33-35. 2. McManus IC, Bryden MP. Geschwind’s theory of cerebral lateralization: developing a formal causal model. Psychol Bull 1991; 110: 237-53. 3. Searleman A, Fugagli AK. Suspected autoimmune disorders and left-handedness: evidence from individuals with diabetes, Crohn’s disease and ulcerative colitis. Neuropsychologia 1987; 25: 367-74. 4. McKeever WF, Rich DA. Left handedness and immune disorders. Cortex 1990; 26: 33-40. 5. Geschwind N, Behan P. Left-handedness: association with immune disease, migraine and development learning disorder. Proc Natl Acad Sci USA 1982; 79: 5097-100. 6. Yeo RA, Gangestad SW. Developmental origins of variation in human hand preference. Genetica (in press). Adverse effects from traditional Chinese medicine SIR,-Dr Tomlinson and colleagues (Feb 6, p 370) point out that there were more cases of aconite poisoning in Hong Kong than we reported (Nov 21, p 1254). Our study population comprised 7 cases of herb-induced aconite poisoning that were prospectively documented over 4 months, and 10 others retrospectively identified. Recruitment required establishment beyond reasonable doubt that aconites from aconitum rootstocks were the only plausible cause for intoxication. We are aware that the problem of herb-induced aconite poisoning is likely to be great in Chinese communities worldwide. In China alone there were over 600 reported cases in the past 30 years,l which prompted the introduction in 1984 of legal regulations on aconites and other potent herbs. The true incidence of aconite poisoning in Hong Kong might be higher; indeed, since our report there have been another 5 well-documented cases. However, identification of such cases requires clinicians to be aware of the dangers of aconites. In none of our retrospectively identified cases was the diagnosis of aconite intoxication suspected while the patient was in hospital, despite the availability in every case of herbal prescription forms that clearly included aconitum rootstocks. However, we caution against overdiagnosis in the absence of sufficient evidence, and agree with Dr But (March 6, p 637) that proper identification of herbs is essential. The current lack of legal control over herbal practice in Hong Kong dates back to the Nanking Treaty of 1842, in which Chinese medicine was protected from control as part of "Chinese customs and usages". In 1989, a government-appointed working party on Chinese medicine was established to review its status and to identify areas for improvement. Since the report of a mini-epidemic of life-threatening aconite poisoning by our group in October, 1991, local education measures have been instituted to alert the medical community and the public.2 Initiatives were also taken by the Chinese Medicinal Material Research Centre to inform and work with the herbal industry, and associations of practitioners of traditional Chinese medicine, to find measures to minimise further incidents of aconite poisoning. An interim report of the working party3 in January, 1992, identified the following issues as high priority: education in the proper use of traditional Chinese medicine, the drawing up of a "potent herbs" list to facilitate control, and the introduction of registration and regulation of practitioners of traditional Chinese medicine. These suggestions might be applicable in many other parts of the world. The report by Professor Vanherweghem and colleagues (Feb 13, p 387) of nephrotoxicity associated with a slimming regimen that included Western medicines and herbal ingredients from Chinese and European plants, confirms the need for appropriate use of medicines. The prescription of herbal medicines requires as much vigilance and caution as that of Western medicines. In the cases reported by Vanherweghem and colleagues, the herbs were not used according to established principles and cannot be accepted as Chinese medicine. As pointed out by Professor Atherton and colleagues (March 6, p 637), the cases reported in Belgium raise the additional issue of the risk of adverse interactions between herbs and modem medicine. In China, there is much research into the possibility of integration of traditional Chinese and modem medicine involving careful experimental and clinical design, and systematic observation and monitoring of efficacies, interactions, and side-effects. The Belgian cocktail is not an integration of traditional and modern medicine, but a confusion of the two. Department of Medicine, University of Hong Kong; Queen Mary Hospital, Department of Biology and the Chinese Medicinal Material Research Centre; and Department of Anaesthesia and Intensive Care Chinese University of Hong Kong, Hong Kong YAU-TING TAI PAUL PUI-HAY BUT KARL YOUNG CHU-PAK LAU 1. Wu YB. Aconite poisoning: review of experience in China over the past 30 years. Jiangsu Zhong Yi (Jiangsu J Chin Med) 1988; 12: 39-42. 2. Tai YT, But PP-H, Young K, Lau CP. Aconite poisoning induced by the Chinese medicinal herbs Chuan Wu and Caowu: clinical presentation, toxicology and management. Urgent newsletter of the Hospital Services Department and the Department of Health, Government of Hong Kong. Hong Kong: Government Press, 1991, Oct 29. 3. Working Party on Chinese Medicine. Interim report. Hong Kong: Government Press, 1992. SIR,-Professor Vanherweghem and colleagues report rapidly progressive interstitial renal fibrosis in 9 young women who had followed a slimming regimen including Chinese herbs. We studied 6 women with renal failure secondary to this regimen (2 of whom are included in the report by Vanherweghem and colleagues). We describe the unusual presentation of 1 of these women with bilateral ureterohydronephrosis secondary to extensive periureteral fibrosis. This 28-year-old woman had two uneventful pregnancies in 1988 and 1990. She was referred to us in October, 1991, during the 12th week of her third pregnancy because she had sterile microscopic haematuria and leucocyturia of recent onset. Ultrasonography disclosed major bilateral ureterohydronephrosis extending to the pelvis. Cystoscopy and magnetic resonance imaging were unhelpful. Her serum creatinine rose from 101 to 176 umol/L at delivery, while her haemoglobin fell from 9-6 to 6-9 g/dL, necessitating transfusion. After delivery of a normal boy in April, 1992, her ureterohydronephrosis did not regress and her renal function further deteriorated. Bilateral retrograde ureteropyelography showed a severe narrowing of both distal ureters, the cause of which remained elusive despite intravenous pyelography, magnetic resonance imaging, and laparoscopy. Despite the successful insertion of ureteral catheters, her renal function continued to deteriorate. We searched, therefore, for a cause of primary renal disease and learned in August, 1992, that the patient had followed the slimming regimen of clinic X described by Vanherweghem and colleagues. After her first pregnancy, from December, 1988, to June, 1989, she took the apparently harmless formula 1 preparation, and after her second pregnancy, from July 1990, to August, 1991, she took the apparently toxic formula 2. Haemodialysis was started in January, 1993, and the ureteral catheters were removed. As part of her pretransplantation programme, a left nephroureterectomy was performed. Macroscopically, major thickening of the pelvis and ureter and a homogeneously shrunken end-stage kidney were observed. Microscopically, intense sclerosis of the renal cortex without primary glomerular or vascular lesions was observed. In addition, the pelvis and the ureter were embedded in a 0 to 1 cmthicksheath of dense fibrotic tissue. This observation suggests that the Chinese herbs incriminated in the reports of Vanherweghem and colleagues induce a generalised fibrotic process. Not only do these herbs cause renal sclerosis, but they may also stimulate extensive fibrosis around both ureters. As in