group of patients described in our original paper. Vickers in his review (5) accepts that none of the studies he reviews is methodologically perfect. None of the trials of acupuncture for hyperemesis done under anaesthesia showed any effect. Jewell and Young in the Cochrane review of interventions for nausea and vomiting in early pregnancy conclude that the results from trials of P6 acupressure are equivocal (10). In conclusion, we accept that there may be a role for acupuncture in hyperemesis gravidarum, but feel that the accusation of disproportionate, uncomfortable and unsuitable therapeutic practice are unfair, and that the severity of illness in the patients we described warranted urgent treatment in the form of enteral nutrition, which is preferable to parenteral nutrition for the reasons we described. Self-propelling nasojejunal tubes are the method of choice for delivering this nutrition where nasogastric feeding fails. Callum B. Pearce Hamish D. Duncan Department of Gastroenterology Queen Alexandra Hospital Portsmouth PO6 3LY, UK e-mail: callum@pearcey.screaming.net References 1. Pearce C B, Collett J, Goggin P M, Duncan H D. Enteral nutrition by nasojejunal tube in hyperemesis gravidarum. Clin Nutr 2001; 20: 461–464 2. Ernst E. The role of complementary and alternative medicine. Br Med J 2000; 321: 1133–1135 3. Knight B, Mudge C, Openshaw S, White A, Hart A. Effect of acupuncture on nausea of pregnancy: a randomized controlled trial. Obstet Gynecol 2001; 97(2): 184–188 4. Carlsson C P, Axemo P, Bodin A, Carstensen H, et al. Manual acupuncture reduces hyperemesis gravidarum: a placebo- controlled, randomized, single-blind, crossover study. J Pain Symptom Manage 2000; 20(4): 273–279 5. Vickers A J. Can acupuncture have specific effects on health? A systematic review of acupuncture antiemesis trials. J R Soc Med 1996; 89: 303–311. 6. Lee A, Done M L. The use of nonpharmacologic techniques to prevent postoperative nausea and vomiting: a meta-analysis. Anesth Analg 1999; 88: 1362 7. Dundee J W, McMillan C. Positive evidence for P6 acupuncture antiemesis. Postgrad Med J 1991; 67(787): 417–422 8. Dundee J W, Ghaly R G, Fitzpatrick KT, Abram WP, Lynch G A. Acupuncture prophylaxis of cancer chemotherapy-induced sickness. J R Soc Med 1989; 82(5): 268–271 9. Dundee J W, Sourial F B, Ghaly R G, Bell P F. P6 acupressure reduces morning sickness. J R Soc Med 1988; 81(8): 456–457 10. Jewell D, Young G. Interventions for nausea and vomiting in early pregnancy (cochrane review). Cochrane Lib 2001; (4) doi:10.1054/clnu.2001.0547, available online at http://www.idealibrary.com on Age and sex influence on appetite Dear Sir, Abnormalities of eating behaviour leading to anorexia and reduced food intake frequently characterise the development and progression of a number of diseases. Unfortunately, they represent an underestimated issue in patients’ clinical management and only recently their impact on morbidity, mortality and quality of life has been acknowledged. The paper by Bannerman et al. is therefore welcome and underscores the growing interest of researchers and clinicians in anorexia (1). Also, it is a valuable contribution to establish the ‘anorexia aware- ness’, hopefully involving an audience larger than the usual psychiatric and physiological communities. As pointed out by the authors themselves, the paper has some limitations. In particular, the authors state twice in the manuscript that ‘there is no evidence that appetite parameters are directly related to age and sex in adults’. In our opinion, this emphatic statement is not convincing. As far as gender is concerned, we believe that a large body of evidence exists indicating that sex hormones influence some hypothalamic areas involved in the control of eating behaviour. As a consequence, it is well established that the menstrual cycle is a modulator of appetite (2). Changes in eating behaviour have been described in healthy women, but appetite is profoundly influenced also in oligomenorrheic patients and in healthy women with menstrual distress. More recently, some gender differences have also been detected in the jejunal migrating motor complex (3). Finally, animal data suggest that gender may also influence the mode of appearance of anorexia (4). Ageing has a profound impact on appetite, leading to a decline in food intake. This decline in food intake has multiple causes, the most relevant being depression and loneliness. However, an increase in the activity of the peripheral satiation system has been demonstrated to significantly contribute to the anorexia of ageing. Healthy older persons are less hungry and more full before, and become more rapidly satiated after eating a standard meal than younger persons (for review see ref. 5). The pathogenesis of the anorexia of ageing is yet to be fully elucidated, but cytokines and hormones appear to be involved. We acknowledge the likelihood that the gender and age differences existing between the groups studied by Bannerman et al. did not significantly influence the observed results. We are also convinced that the unbalanced matching of patients and controls does not diminish the clinical relevance of the paper. However, we feel that the ‘anorexia awareness’ could be seriously biased by the authors’ too emphatic statement on the 186 CORRESPONDENCE