6 Riou G, Favre M, Jeannel D, Bourhis J, Le Doussal V, Orth G. Association ..... alkaline phosphatase as an antenatal screening test for Down's syndrome.
data support the theory that human papillomavirus plays an important part in the many steps to cervical cancer,4" but this is not the same as saying that poor personal hygiene is the major, if not the only, cause3 6 of this cancer. Surely a more straightforward message would be to say that men should use a condom during sexual intercourse. This would probably help to prevent not only cervical cancer, because undeniably there is an increased risk for those who are sexually active, but also the sexual transmission of other viruses that promote tumours, including HIV,7 as well as the purely sexually transmitted diseases.8 It may or may not be cost effective to screen all adult women for cervical cancer, but we should at least be given the opportunity to assess the relative merits of various defined interventions rather than have the waters muddied by meaningless and potentially damaging asides. WILLIAM FOULKES
London E9 7NG 1 Howorth PJN. The health of the nation. BMJ 1991;303:184. (20 July.) 2 Department of Health. The health of the nation: a consultative documentforhealth in England. London: HMSO, 1991:3-15. 3 Peto R. Introduction: geographic patterns and trends. In: Peto R, zur Hausen H, eds. Viral aetiolog of cervtcal cancer. Cold Spring Harbor, New York: CSH Laboratories, 1986:167-71. (Banbury report 21.) 4 Zur Hausen H. Papilloma viruses and human genital cancer. In: Grundmann E, ed. Cancer of the uterine cervix. Stuttgart: Gustav Fischer Verlag, 1985. 5 Mitchell H, Drake M, Medley G. Prospective evaluation of risk of cervical cancer after cytological evidence of human papillomavirus infection. Lancet 1986;i:573-5. 6 Riou G, Favre M, Jeannel D, Bourhis J, Le Doussal V, Orth G. Association between poor prognosis in early-stage invasive cervical carcinomas and non-detection of HPV DNA. Lancet 1990;335: 1171-4. 7 Plummer FA, Simonsen JN, Cameron DW, Ndinya-Achola JO, Kreiss JK, Gakinya MN, et al. Cofactors in male-female sexual transmission of human immunodeficiency virus type I.J Infect Dis 1991;163:233-9. 8 Upchurch DM, Brady WE, Reichart CA, Hook EW. Behavioural contributions to acquisition of gonorrhoea in patients attending an inner city sexually transmitted disease clinic. _J Infect Dis
1990;161:938-41.
Availability of cadaver organs for transplantation SIR,-The comment in "This week in BM7" (4 May) on the paper by Dr M A M Salih and colleagues' that "Elective ventilation for the purposes of organ donation raises ethical questions that require wide debate" encourages us to draw attention to the serious defects in the present system for obtaining organs for transplantation as well as in the new proposal. Briefly, these defects are that the clinical brain stem tests in routine use are not exhaustive (not challenging the medullary respiratory centre with a hypoxic stimulus, for example) and cannot ensure the permanent absence of all brain stem function2'4; that their satisfaction does not ensure that the brain has been destroyed5 or that higher brain function has irreversibly stopped6; and that those from whom consent for removal of organs is requested may not have this idiosyncratic notion of death explained to them. It is still a mistakenly held belief that life support is withdrawn before surgery to remove organs. To these defects it is now suggested that we add the deliberate prolongation of dying of some patients to benefit not those patients but third parties. However deserving the third parties may be, such treatment is in breach of the Hippocratic principle and the Declaration of Geneva. Using artificial ventilation to prolong the dying of adults is already allowed by the code of practice for obtaining donor organs8 and has also been proposed or used for children and anencephalic babies. For example, one of us (DJH) was asked to ventilate a young girl dying of a brain tumour solely so that she should not die before arrange-
312
ments could be made for her organs to be removed. Reports from the US and Canada suggest that there is real doubt in the minds of theatre staff over whether beating heart donors on ventilators are still patients or truly dead when operations to remove organs begin,9' and there is reason to believe that the same anxieties exist in the United Kingdom, including among anaesthetists who both anaesthetise and paralyse organ donors." It is ironic that animals are protected from such abuse by the Animals (Scientific Procedures) Act 1986; section 1, subsection 4, states that "An animal is living until the permanent cessation of circulation or the destruction of its brain."'2 Is it not time that we accorded to patients at least the same security that we insist on for animals and use the same criteria of permanent cessation of circulation or destruction of the brain to establish that death has, indeed, occurred before embarking on surgery to remove vital organs? To accept the suggestions of Dr Salih and colleagues would take us even further from such principles. DAVID J HILL Anaesthetic Department, Addenbrooke's Hospital, Cambridge CB2 2QQ
results on screening with confidence intervals including low risks declined it. If a confidence interval was cited it should take account of all sources of random error, including fluctuations within a person over time2 and errors of assessing gestational age3 as well as assay error. Such a confidence interval would be very wide, and the consequent reduction in efficiency is likely to be large. The fact that the estimated risk is imprecise should not detract from the value of biochemical screening for Down's syndrome. Screening on the basis of maternal age gives reproducible results but has little efficiency: biochemical screening's imprecision is more than compensated for by its greater efficiency, yielding a materially greater detection rate for the same rate of amniocentesis. HOWARD S CUCKLE Department of Obstetrics and Gynaecology, St James's University Hospital, Leeds LS9 7TF NICHOLAS WALD JAMES DENSEM Department of Environmental and Preventive Mledicine, Medical College of St Bartholomew's Hospital, London EC1M 6BQ
DAVID WAINWRIGHT EVANS
Cambridge CB3 9LN G AUSTIN GRESHAM
University of Cambridge,
Cambridge 1 Salih MAM, Harvey I, Frankel S, Coupe DJ, Webb M, Cripps HA. Potential availability of cadaver organs for transplantation. BMJ 1991;302:1053-5. (4 May.) 2 Wetzl RC, Setzer N, Stiff JL, Roberts MC. Hemodynamic responses in brain dead organ donor patients. Anesth Analg 1985;64: 125-8. 3 Shiogai R, Takeuchi K, Ogashiwa M, Hara M, Kadowaki C, Maeda T, et al. Brain stem auditory evoked potentials monitoring in the diagnosis of brain death-clinical applications and reliability. NeurosurgRev 1989;12(suppl 1):328-39. 4 Sinclair ME, Suter PM. Lower oesophageal contractility as an indicator of brain death in paralysed and mechanically ventilated patients with head injury. BMJ 1987;294:935-6. 5 Byrne PA, O'Reilly S, Quay PM. Brain death-an opposing viewpoint. 7AMA 1979;242:1985-90. 6 Grigg MM, Kelly MA, Celesia GG, Ghobrial MW, Ross ER. Electroencephalographic activity after brain death. Arch Neurol 1987;44:948-54. 7 Denison KM. On the ethics of transplant surgery. Theologia Cambrensis 1991;3(2):5-8. 8 Working Party on Behalf of the Health Departments of Great Britain and Northern Ireland. Cadaveric organs for transplantation. A code of practice. London: DoH, 1983: para 26. 9 Youngner SJ, Landefeld CS, Coulton CJ, Juknialis BW, Leary M. "Brain death" and organ retrieval. A cross-sectional survey of knowledge and concepts among health professionals. JAMA 1989;261:2205-10. 10 Gelb AW, Robertson KM. Anaesthetic management of the brain dead for organ donation. Can J7 Anaesth 1990;37: 806-12. 11 Timmins AC, Hinds CJ. Management of the multiple-organ donor. Current Opinion in Anaesthesiology 1991;4:287-92. 12 BMA. Rights and responsibilities of doctors. London: BMA,
1988:92.
Estimating risk of Down's syndrome SIR,-Messrs K Spencer and P Carpenter are concerned about the effect of assay error in biochemical screening on the estimated risk of Down's syndrome and suggest that the risk should be reported together with a confidence interval.' In our view, this would be confusing. In screening, a categorical decision on whether to take further action needs to be made from a continuous variable, in this case the risk of Down's syndrome based on the maternal age and biochemical profile. A line has to be drawn somewhere, and the use of confidence intervals would blur this, probably leading to a loss of efficiency of screening-that is, a lower detection rate for a given rate of amniocentesis. This would occur if some women with a truly low risk who had negative results on screening with confidence intervals encompassing high risks were offered amniocentesis and some women with a truly high risk who had positive
I Spencer K, Carpenter P. Estimating risk of Down's syndrome. BMJ7 1991;302:1536-7. (22 June.) 2 Cuckle HS, Wald NJ, Nanchahal K, Densem J. Repeat maternial serum alpha-fetoprotein testing in antenatal screening programmes for Down's syndrome. Br j Obstet Gynaecol
1989;%:52-60. 3 Holding S. Estimations of gestational age and screening for
Down's syndrome. BMJ 1991;302:965. (20 April.)
Prenatal screening for Down's syndrome SIR,-The article by Mr Trevor A Sheldon and Dr John Simpson on the cost effectiveness of triple screening for Down's syndrome' raises some interesting points but fails to answer the fundamental question, which is, What are the marginal costs and marginal benefits of changing from screening by maternal age to triple screening? It can be derived from figures in the article that screening by maternal age detected 3 9 fetuses with Down's syndrome a year. This is a detection rate of 30 9%. By using their table III, the total cost for this screening service was 3 9 x£17 000 (cost per case detected)=£66 300. According to example 2 in the box in their article, if we assume a sensitivity for triple screening of 60%, uptake of screening of 80%, and uptake of amniocentesis of 75%, a triple screening programme will detect 4 6 fetuses with Down's syndrome at a cost of £134970. The marginal cost is therefore £68 670 for a marginal benefit of 0-7 extra fetuses with Down's syndrome detected. This is equivalent to a marginal cost per case of £98 100 for the extra case detected each 17 months as a result of introducing triple screening. Though this figure is still comparable with the £90 000 that it costs society to look after a child with Down's syndrome, it is surely this figure of £98 100 to detect an additional case rather than the average cost of £29 341 that should be considered when making decisions on whether to introduce triple screening for Down's syndrome. TIM DAVIES OAK SOE KHA East Dorset Health Authority, Royal Victoria Hospital, Bournemouth BH1 4HX 1 Sheldon TA, Simpson J. Appraisal of a new scheme for prenatal screening for Down's syndrome. BMJ 1991;302:1133-6. (11 May.)
SIR,-Dr Jennifer G Wishart is concerned about screening for Down's syndrome by measuring maternal blood neutrophil alkaline phosphatase
BMJ VOLUME 303
3 AUGUST 1991
activity' because the study that forms its scientific basis was retrospective and included mainly older women2 and because the test can be subject to large differences between observers. The fact that the original study was retrospective would have biased the results if knowledge of the outcome of pregnancy influenced the person measuring the neutrophil alkaline phosphatase activity, which is in part subjective, or if the trauma of amniocentesis increased the activity. This is unlikely to have happened as the tests were done without knowledge of whether the blood film was from a woman whose fetus had Down's syndrome and the average activity in women who had an amniocentesis was similar to that in other women. The distribution of maternal age in the study was indeed atypical of that in the general population, but this does not limit the applicability of the results because, in both pregnancies in which the fetus had Down's syndrome and pregnancies in which it did not, no association between maternal age and the activity was found. Measurement of neutrophil alkaline phosphatase activity is semiquantitative so that there may be differences between observers. In the screening service offered by this department we have sought to minimise this by double reading blood films during a training period. As a result the scores obtained by our two readers are comparable, with a 7% coefficient of variance. The NHS provides screening of maternal serum for Down's syndrome in some hospitals, including our own, and we hope that this will eventually be extended throughout the country. We will be using much of the funds raised through our private screening service to research ways of simplifying the neutrophil alkaline phosphatase test so that it can be made less labour intensive and, hence, more suitable for mass screening in the NHS. With the support of the Medical Research Council we will also be studying the effect of different approaches to counselling pregnant women on anxiety associated with screening. HOWARD S CUCKLE RICHARD J LILFORD JAMES G THORNTON Academic Department of Obstetrics and Gynaecology,
St James's University Hospital, Leeds LS9 7TF I Wishart JG. Prenatal screening for Down's syndrome. BMJ 1991;302:54-5. (6 July.) 2 Cuckle HS, Wald NJ, Goodburn SP, Sneddon J, Amess JAL, Dunn SC. Measurement of activity of urea resistant neutrophil alkaline phosphatase as an antenatal screening test for Down's syndrome. BMJ 1990; 301:1024-6.
Ethics, commerce, and kidneys SIR,-A rational decision, whether it be to sell a kidney or to sell a house, depends on more than a fair price or a free will. ' It demands that the person concerned has full information regarding all the possible consequences of his or her choice and the ability to process the information rationally. Conveying medical information to patients for the purpose of obtaining informed consent is widely accepted as being difficult when there is a clear benefit to the patient as well as an entailed risk. If a person sells his or her kidney he or she can expect no physical benefit, only a loss, with consequences that may be at best simply painful and at worst life threatening. The vendor relies on accurate information from someone who may have a vested interest in the welfare of the recipient and who views the donor only as a means to a greater good-that of the recipient. Because a person's ability to appreciate the consequences of an action depends on his or her competence we would not allow a child or mentally subnormal person to sell a kidney whatever the price. What then of an unsophisticated person
BMJ
VOLUME
303
3
AUGUST
1991
selling a kidney without apparently appreciating the difference between a hotel and a hospital? It is too simplistic to equate exploitation only with market forces; and it is a mistake to ignore the fact that vulnerability, as measured in terms of competence, is not always easy to assess. G M SAYERS Directorate for Medicine for the Elderly, Hackney Hospital, London E9 6BE 1 Wight JP. Ethics, commerce, and kidneys. BMJ 1991;303:110. (13 July.)
SIR,-I agree with Dr J P Wight that we need to know why obtaining organs for transplantation through payment is not permitted, but I disagree that the reason is practical rather than ethical.' I think that on certain fundamental questions his argument is flawed and that the conclusions he draws are potentially dangerous. Believing that he can exert free will in matters such as selling his house, he assumes that free will is not compromised by financial concerns in other matters. He further argues that allowing the sale of organs increases personal freedom by increasing freedom of action. In my view these assumptions are incorrect. Free will cannot be assumed to exist in all situations. Even if lines of action are available to us our choices may be limited by various constraints, and financial means are particularly powerful. Simply giving a fair price does not protect people from exploitation. Poverty is becoming less well tolerated by society, and welfare support is increasingly difficult to obtain. It is not difficult to imagine an unemployed parent with debts being advised by the social security office of various methods of raising money, which might include the sale of an organ. A choice between hunger or the break up of your family and a kidney gives little room for freedom. Similarly, our health service may become run on a system of credit that we can choose to use in various ways. The catch phrase is patient choice. There is little freedom in choosing between an operation for cataract and a colonoscopy if they are both needed. The needs of treatment might demand that we sell a kidney to obtain enough credit. It is only a short step to consider using euthanasia to obtain organs. One of the dangers of allowing euthanasia even without payment is that pressure may influence people's decision on the matter. I might choose euthanasia to relieve not only my own suffering but the financial burden on my family. The benefits of selling an organ could influence this decision. It is a sinister reminder that in Nazi Germany euthanasia was first introduced by medical researchers for the mentally ill and those with learning disabilities.2 I think that the question is not whether we should be free to sell parts of our body but whether it is appropriate to place financial value on human life. Our profession must guard against undue simplification as a solution to real ethical conflicts. T D SIMPSON London N8 9RT I Wight JP. Ethics, commerce, and kidneys. BMJ 1991;303:110. (13 July.) 2 Birley JLT. Psychiatrists and citizens. Br J Psychiatry 1991; 159:1-6.
ability of the five sampling methods to diagnose cervical intraepithelial neoplasia (p values being between 0-2 and 0 97). A reader of Dr Wolfendale's editorial might be left with the incorrect impression that "the Cervex sampler improved the yield of... smears suggestive of cervical intraepithelial neoplasia grade III when compared with the modified Ayre spatula." In fact, the paper provides no evidence in favour of this conclusion. The quoted paper has several problems. The most serious is in the analysis of data relating a diagnosis of cervical intraepithelial neoplasia to the five methods of cervical sampling (its table III). Boon et al obtain p
