share her conviction that a once daily application of a potent topical corticosteroid is the panacea for this intractable disease. Dalziel et al claim that twice daily applications are effective,' but in my experience such is the distress caused by this disorder in some women that more frequent applications are needed, certainly in the more active stage. The commonest antiseptic currently available in combined steroid-antiseptic topical remedies recommended by dermatologists is clioquinol. Clioquinol has a low sensitisation index, although the risk of sepsis in the eroded skin of the perineum after the application of potent topical corticosteroid preparations that lower resistance to infection cannot be disregarded. Furthermore, a few patients do not respond to this treatment, and the efficacy of other topical remedies is supported by enough anecdotal evidence for them to be mentioned, even ifthey are eventually discarded. Perhaps their success is a placebo effect, but for the fortunate patient who responds to them this is no less welcome. Vulvectomy remains a desperate and last resort for those patients with eroded and ulcerated pudenda in nightly distress from soreness and pruritus who fail to respond to topical steroid therapy. Although recurrence may occur in the same area in 80% of those who have this operation, patients obtain considerable relieffor many months if not years, and the recurrence is rarely as severe as the original condition. Lest male indifference be suspected may I point out that the standard treatment for affected male genitalia when this disorder does not respond to potent topical corticosteroid treatment is also surgical, including
circumcision. ALAN B SHRANK
Shrewsbury SYI 1TQ 1 Dalziel K, Millard P, Woinarowska FT. Lichen sclerosus et atrophicus treated with a potent topical steroid (dobetasol dipropionate 0-05%). BrJ Dermatol 1989;121(suppl 34):34-5.
Cell implantation in Parkinson's disease SIR,-Professor Adrian Williams's editorial discusses the disappointing outcome of cell implantation in Parkinson's disease.' Not only are implant procedures expensive, demanding on resources, and ethically contentious but at present they seem ineffective. Professor Williams comments that "Properly funded and conducted studies are needed on all new treatments, including implants." We suggest that such studies should also include electroconvulsive therapy. Though the current evidence for the efficacy of electroconvulsive therapy for motor disability in Parkinson's disease is largely anecdotal or from small uncontrolled trials, the sum of this evidence is impressive enough to warrant more controlled evaluation.2 Studies can be divided into those that used electroconvulsive therapy for patients with Parkinson's disease who had associated depression and those that used it for patients who were not depressed. Electroconvulsive therapy has been described in 51 patients with associated depression, of whom 31 (61%) had appreciable improvement in their motor features. There is a dissociation of response of motor and mood features, with the former responding more rapidly. In patients who are not depressed most studies claim a favourable response' but not all.7 Thirty three such patients are reported to have been treated, and 19 (58%) of them have had appreciable benefit. Benefits include an improvement in motor symptonss in general; a reduction in the required daily dose of levodopa; and, with regard to the onoff phenomenon, an increase in the proportion of time spent on. These improvements have lasted from a few days to 42 weeks. There are no reports of worsening of symptoms
556
of Parkinson's disease or of any other deleterious effects other than transient confusion in some patients. Electroconvulsive therapy is cheap and safe and compared with implant techniques requires little special equipment and expertise. Carefully designed, double blind clinical trials are urgently required to examine the efficacy of this treatment in Parkinson's disease. P MADELEY C A BIGGINS
J L BOYD R H S MINDHAM E G S SPOKES
University of Leeds, Leeds LS2 9LT
unlikely to be used regularly in the NHS. In addition, the BUPA schedule does not match the operation codes of the Office of Population Censuses and Surveys used in the NHS and any analysis with the BUPA schedule necessitates recoding of data. This is not to say that diagnosis related groups are a suitable alternative. They have been criticised because of the appropriateness or otherwise of American categories and weights to the British system. A more serious criticism, and one which directly affects the use of such groups in audit, is the heterogeneity of some of the diagnosis related groups.2 S L GEORGE
1 Williams A. Cell implantation in Parkinson's disease. Br Med J 1990;301:301-2. (11 August.) 2 Abrams R. ECT for Parkinson's disease. Am J Psychiatry 1989;146: 1391-3. 3 FrommGH. Observationsontheeffectsofelectroshocktreatment on patients with parkinsonism. Bulletin of Tulane University 1959;18:71-3. 4 Balldin J, Eden S, Granerus A-K, et al. Electroconvulsive therapy in Parkinson's syndrome with "on-off' phenomenon. JNeural Transm 1980;47:11-21. 5 Balldin J, Granerus A-K, Lindstedt G, Modigh K, Walinder J. Predictors for improvement after electroconvulsive therapy in parkinsonian patients with on-off symptoms. J Neural Transm 1981;52:199-21 1. 6 Andersen K, Balldin J, Gottfries CG, et al. A double blind evaluation of electroconvulsive therapy in Parkinson's disease with "on-off" phenomena. Acta Neurol Scand 1987;76:191-9. 7 Ward C, Stern GM, Pratt TRC, McKenna P. Electroconvulsive therapy in parkinsonian patients with the "on-off' syndrome. JNeural Transm 1980;49:13345.
Caseload or workload? SIR,-We agree with Messrs S M Jones and C D Collins that the surgical profession should develop an agreed measure of surgical workload, an essential precursor to surgical audit.' We believe, however, that the British United Provident Association (BUPA) schedule of procedures will be inadequate for measuring all surgical workload. We have compared the complexity of ear, nose, and throat procedures carried out over three years (1983-5) at Sheffield Children's Hospital on Sheffield residents with that of procedures carried out on non-residents. We calculated average costs per case for each group based on weights derived from the benefits payable for each category of the BUPA schedule, the diagnosis related group, and, as a check, average length of stay. The average cost per case for all ear, nose, and throat cases was £711.%. The table shows average costs for residents and non-residents of Sheffield. Average cost per case for patients undergoing ear, nose, and throat procedures at Sheffield Children's Hospital during 1983-5 calculated by three weighting systems
BUPA schedule Diagnosis related group Average length of stay
Cost for Sheffield resident (£)
Cost for nonresident (£)
713.23
708.08
663.94 669.44
858.82 844.95
The difference in average cost between Sheffield residents and non-residents was £5.15 by the BUPA method. This difference is negligible given the crudeness of the analysis and the average case cost of £711.%. The small difference also reflects the insensitivity of the BUPA schedule, which is caused by almost all patients being in either the minor or intermediate categories. The more complex classification by diagnosis related group, which gives 10 groups for ear, nose, and throat procedures in children, gave a difference of £195certainly important. When average length of stay was used to weight costs the difference was £175. Although the BUPA schedule is easily understood because of its hierarchical nature and may be suitable for surgical audit, we found that it is too insensitive to be used to set prices, at least for ear, nose, and throat surgery. For this reason it is
J E BRAZIER University of Sheffield Medical School, Sheffield S10 2RX 1 Jones SM,CollinsCD. Caseload orworkload? Scoring complexity of operative procedures as a means of analysing workload. BrMedJ 1990;301:324-5. (11 August.) 2 Sanderson S, Craig M, Winyard G, Bevan G. Using diagnosisrelated groups in the NHS. CommuniyMed 1986;8:37-46.
Psychiatric aspects of urinary incontinence SIR,-I hope that the careful study reported by Dr K R W Norton and colleagues will discourage gynaecologists and urologists from dismissing detrusor instability and the symptoms that usually accompany it-frequency, urgency or urge incontinence, a feeling of perineal pressure, and nocturia-as indicative of psychiatric disorder.' In a microbiological study of urine collected from patients undergoing urodynamic investigations we showed that 37 of 42 women with detrusor instability had evidence ofinfection with aerobic or fastidious bacteria before investigation.2 We have isolated bacteria from the urine of many women with urgency or urge incontinence.2 3 Many other women had high counts of fastidious bacteria and some had midstream urine specimens showing pyuria, a finding consistent with a diagnosis of urethral or periurethral infection.34 Evidently, urgency and urge incontinence could be caused by inflammation in and around the proximal urethra and bladder neck and irritability or instability of the bladder might result from this. The many women who suffer from this distressing complaint would be helped if clinicians and microbiologists would suspect and seek an infective cause before starting invasive investigations, which do not help and often exacerbate the symptoms.5 ROSALIND MASKELL
Public Health Laboratory, St Mary's General Hospital, Portsmouth P03 6AQ 1 Norton KRW, Bhat AV, Stanton SL. Psychiatric aspects of urinary incontinence in women attending an outpatient
urodynamic clinic. BrMedJ 1990;301:271-2. (4 August.) 2 Payne SR, Timoney AG, McKenning ST, den Hollander D, Pead LJ, Maskell RM. Microbiological look at urodynamic studies. Lancet 1988;u: 1123-6. 3 Wilkins EGL, Payne SR, Pead PJ, Moss ST, Maskell RM. Interstitial cystitis and the urethral syndrome: a possible answer. BrJ Urol 1989;64:39-44. 4 Maskeli R, Pead L, Sanderson RA. Fastidious bacteria and the urethral syndrome: a 2 year clinical and bacteriological study of 51 women. Lancet 1983;ii:1277-80. 5 Maskeli R. A new look at the diagnosis of infection of the urinary tract and its adjacent structures. I Infect 1989;19:207-17.
Perioperative deaths among children SIR,-Dr Paul M Spargo commented on the criteria for membership of the Association of Paediatric Anaesthetists of Great Britain and Ireland (not Northern Ireland as stated in his
letter).' The association was founded in 1973 to provide a
BMJ VOLUME 301
15 SEPTEMBER 1990
forum for the fairly small number of anaesthetists (now increased to over 100 home members) who spend a substantial part of their time with neonates, infants, and young children. It was recognised from the outset that there were also many anaesthetists who spend some time in paediatric anaesthesia often for children over the age of 3. In an annual scientific meeting of one and a half days it is not possible to provide a forum for the first group and an educational programme for the last. For that reason, after much deliberation, it was decided that members of the association should spend roughly halfoftheir time in paediatric anaesthesia. This is interpreted to include anaesthesia, paediatric intensive care, and a heavy on call paediatric commitment, and all these activities can be aggregated. To meet the educational need of anaesthetists with a lesser paediatric commitment the association has held two day seminars on paediatric anaesthesia in alternate years. These were initially highly successful, but the seminar for 1989 was cancelled owing to insufficient numbers. It probably indicates that the demand has been met by the welcome upsurge in seminars and similar teaching programmes in paediatric anaesthesia organised by the College of Anaesthetists, the Association of Anaesthetists, and the section of anaesthetics of the Royal Society of Medicine. Members of the Association of Paediatric Anaesthetists have made large contributions at these and other specialist meetings. The association is now considering a more flexible educational programme to replace the two day seminar. It may be a "refresher" day open to non-members preceding the annual scientific meeting. The association has also been invited to hold a session on paediatric anaesthesia at the winter meeting of the Association of Anaesthetists in January 1991, which will be for the anaesthetist who may only occasionally anaesthetise children. Postgraduate education in paediatric anaesthesia is being catered for by several organisations and is all the richer for that. WILLIAM J GLOVER
Association of Paediatric Anaesthetists, Hospital for Sick Children, Great Ormond Street, London WC1N 3JH I Spargo PM. Perioperative deaths among children. Br Med J
1990;301:343. (11 August.)
Hereditary (primary) haemochromatosis SIR,-I would like to add a caveat to Dr Niall D C Finlayson's statement that screening would probably benefit patients with hereditary haemochromatosis.' Recently I was asked for an opinion on a man who had been screened and found to have a ferritin concentration of 398 tg/l (normal 20-180 ig/l). He was seeking to extend his mortgage through extra life insurance. He volunteered the information that his father had haemochromatosis, and the insurance firm promptly doubled his premium. Tests confirmed the raised ferritin concentration, normal liver and pancreatic function, and the presence of the HLA-A3 antigen. He was otherwise healthy-a non-smoking, normotensive 40 year old marathon runner who consumed 12 units of alcohol a week. My letter to the insurance firm, stating that I would give him venesection to prevent the effects of his iron loading potential and that he was therefore a better than average risk now that the problem had been discovered, resulted in his premium being reduced to one and a half times the original amount quoted. I have advised him to approach other insurance firms armed with my letter and your editorial, hoping for a more sympathetic outcome. This raises the point yet again that screening has
BMJ VOLUME 301
15 SEPTEMBER 1990
implications for patients' lifestyles that we and they may not appreciate. C J BARTON
Royal Berkshire Hospital,
Reading RGl SAN 1 Finlayson NDC. Hereditary (primary) haemochromatosis. BrMedj 1990;301:350-1. (18-25 August.)
Liver function tests SIR,-M F Laker suggested that measuring the activity of alanine aminotransferase rather than aspartate aminotransferase may improve the specificity of investigations for liver disease,' a view adopted by many on the assumption that alanine aminotransferase is more hepatospecific than aspartate aminotransferase.2 We believe that this is not necessarily correct and that there are several cases in which aspartate aminotransferase may provide an equally reliable assessment of hepatic dysfunction. The ability of laboratory analyses to distinguish categories of hepatobiliary disease was assessed in a prospective, computer based graphical evaluation by probabilistic test analysis.3 With a cut off point of 200 U/I (reference range up to 20 U/1) aspartate aminotransferase was a powerful discriminator between viral hepatitis and other forms of hepatobiliary disease with a sensitivity of 91% and specificity of 95%. To achieve a comparable degree of discrimination a cut off point of 300 U/I was required for alanine aminotransferase (reference range up to 20 U/1). Another important case is hepatotoxicity caused by drugs such as paracetamol and isoniazid, which often results in both an appreciably raised aspartate aminotransferase activity and a raised aspartate aminotransferase: alanine aminotransferase ratio.4 This suggests that aspartate aminotransferase may be a more sensitive marker in this condition. Stability of the enzymes in serum is also important if there may be a delay between separating the sample and analysis, for instance at weekends. Aspartate aminotransferase is stable when refrigerated (4°C) or frozen (-20°C and -80'C),5 whereas alanine aminotransferase is considerably less stable and refrigeration is preferred to freezing. Although alanine aminotransferase is considered by many to be more hepatospecific, the usefulness of aspartate aminotransferase in managing hepatobiliary disorders should not be underestimated and deserves to be critically re-examined against the background of published reports. R CRAMB C M FLORKOWSKI
Queen Elizabeth Medical Centre,
Birmingham B15 2TH I Laker MF. Liver function tests. Br Med J 1990;301:250-1. (4
August.) 2 Aach RD, Szmuness W, Mosley JW, et al. Serum alanine aminotransferase of donors in relation to the risk of non-A nonB hepatitis. The transfusion-transmitted viruses study. N Engl J Med 1981;304:989-94. 3 Borsch G, Baier J, Glocke M, Nathusius W, Gerhardt W. Graphical analysis of laboratory data in the differential diagnosis of cholestasis: a computer assisted prospective study. J Clin
Chem ClinBiochemn 1988;26:509-19. 4 Himmelstein DU, Woolhandler SJ, Adler RD. Elevated SGOT/ SGPT ratio in alcoholic patients with acetaminophen hepato-
toxicity. Amj Gastroenterol 1984;79:718. 5 Niblock AE, Leung FY, Henderson AR. Serum aspartate aminotransferase storage and the effect of pyridoxal phosphate. J Lab Clin Med 1986;108:461.
Research struggles in eastern Europe SIR,-I am writing concerning Alexander Dorozynski's news item.' Getting east European members to Western meetings is expensive but it can be done. The European Spine Society is having a meeting
at the University of Zurich. The university has always refused any trade advertising but on this occasion has allowed a trade exhibition on the understanding that the money be used to bring east Europeans to the meeting. Most meetings in Britain expect to make a profit, and unfortunately this profit is used for administration and often to finance the travelling of members of the executive. In the European Spine Society nobody attends meetings without paying his or her own costs, which, of course, considerably reduces the meetings' cost. The drug and instrument companies are keen to advertise the fact that they are financing east Europeans to attend Western meetings. It also allows companies to meet the surgeons they are supporting and, with the passage of time, to have access to east European markets. MICHAEL SULLIVAN
Royal National Orthopaedic Hospital, London WI P 8AQ 1 Dorozynski A. Research struggles in eastern Europe. Br Med J 1990;301:305-6. (11 August.)
Mortality in patients with bleeding peptic ulcer SIR,-Mr K E Wheatley and colleagues clearly show the importance of a defined protocol in achieving a low mortality from bleeding peptic ulcers.' It is not so clear, however, whether low mortality can be achieved only in specialised units. Two studies from hospitals with specialist units have reported low overall mortality figures (5 5% and 4 8%) using similar protocols,2' but we have obtained comparable results in a district general hospital without such a unit.4 An identical protocol for the indications for surgery was used (again without using endoscopic stigmata) and over one year an overall mortality of 4-6% was recorded. Only one patient out of 60 with proved bleeding peptic ulcers died. We therefore suggest that although specialist units confer many logistic advantages in managing patients with gastrointestinal bleeding, they are not the most important factor in achieving a low mortality. The key is an agreed clear policy of management and active cooperation between specialties. ` D CLEMENTS D FOSTER
J STAMATAKIS W E WILKINS J S MORRIS
Princess of Wales Hospital, Bridgend CF31 I RQ I Wheatley KE, Snyman JH, Brearley S, Keighley MRB, Dykes PW. Mortality in patients with bleeding peptic ulcer when those aged 60 or over are operated on early. Br Med 7 1990;301:272. (4 August.) 2 Holman RAE, Davis M, Gough KR, Gartell P, Britton DC, Smith RB. Value of a centralised approach in the management of haematemesis and melaena: experience in a district general hospital. Gut 1990;31:504-8. 3 Sanderson JD, Taylor RFH, Pugh S, Vicar FR. Specialised gastrointestinal units for the management of upper gastrointestinal haemorrhage. Postgrad Medy 1990;66:654-6. 4 Clements D, Aslan S, Foster D, Stamatakis J, Wilkins WE, Morris JS. Acute upper gastrointestinal haemorrhage in a district general hospital. Audit of an agreed management policy. J R Coll Physicians Land (in press). 5 Madden MV, Griffith GH. Management of upper gastrointestinal bleeding in a district general hospital. J R Coll Physicians Lond
1986;20:212-5.
Correction Papal policy, poverty, and AIDS A printer's error occurred in this letter by Dr J Guillebaud (1 September, p 440). The last sentence should have read, "Can anyone explain to me why it is not wrong thus to ensure ... that millions of sperm die in the fallopian tubes without fertilising an egg, yet it is wrong to ensure, within marriage, that they die instead in a rubber condom."
557
circumcision. ALAN B SHRANK
Shrewsbury SYI 1TQ 1 Dalziel K, Millard P, Woinarowska FT. Lichen sclerosus et atrophicus treated with a potent topical steroid (dobetasol dipropionate 0-05%). BrJ Dermatol 1989;121(suppl 34):34-5.
Cell implantation in Parkinson's disease SIR,-Professor Adrian Williams's editorial discusses the disappointing outcome of cell implantation in Parkinson's disease.' Not only are implant procedures expensive, demanding on resources, and ethically contentious but at present they seem ineffective. Professor Williams comments that "Properly funded and conducted studies are needed on all new treatments, including implants." We suggest that such studies should also include electroconvulsive therapy. Though the current evidence for the efficacy of electroconvulsive therapy for motor disability in Parkinson's disease is largely anecdotal or from small uncontrolled trials, the sum of this evidence is impressive enough to warrant more controlled evaluation.2 Studies can be divided into those that used electroconvulsive therapy for patients with Parkinson's disease who had associated depression and those that used it for patients who were not depressed. Electroconvulsive therapy has been described in 51 patients with associated depression, of whom 31 (61%) had appreciable improvement in their motor features. There is a dissociation of response of motor and mood features, with the former responding more rapidly. In patients who are not depressed most studies claim a favourable response' but not all.7 Thirty three such patients are reported to have been treated, and 19 (58%) of them have had appreciable benefit. Benefits include an improvement in motor symptonss in general; a reduction in the required daily dose of levodopa; and, with regard to the onoff phenomenon, an increase in the proportion of time spent on. These improvements have lasted from a few days to 42 weeks. There are no reports of worsening of symptoms
556
of Parkinson's disease or of any other deleterious effects other than transient confusion in some patients. Electroconvulsive therapy is cheap and safe and compared with implant techniques requires little special equipment and expertise. Carefully designed, double blind clinical trials are urgently required to examine the efficacy of this treatment in Parkinson's disease. P MADELEY C A BIGGINS
J L BOYD R H S MINDHAM E G S SPOKES
University of Leeds, Leeds LS2 9LT
unlikely to be used regularly in the NHS. In addition, the BUPA schedule does not match the operation codes of the Office of Population Censuses and Surveys used in the NHS and any analysis with the BUPA schedule necessitates recoding of data. This is not to say that diagnosis related groups are a suitable alternative. They have been criticised because of the appropriateness or otherwise of American categories and weights to the British system. A more serious criticism, and one which directly affects the use of such groups in audit, is the heterogeneity of some of the diagnosis related groups.2 S L GEORGE
1 Williams A. Cell implantation in Parkinson's disease. Br Med J 1990;301:301-2. (11 August.) 2 Abrams R. ECT for Parkinson's disease. Am J Psychiatry 1989;146: 1391-3. 3 FrommGH. Observationsontheeffectsofelectroshocktreatment on patients with parkinsonism. Bulletin of Tulane University 1959;18:71-3. 4 Balldin J, Eden S, Granerus A-K, et al. Electroconvulsive therapy in Parkinson's syndrome with "on-off' phenomenon. JNeural Transm 1980;47:11-21. 5 Balldin J, Granerus A-K, Lindstedt G, Modigh K, Walinder J. Predictors for improvement after electroconvulsive therapy in parkinsonian patients with on-off symptoms. J Neural Transm 1981;52:199-21 1. 6 Andersen K, Balldin J, Gottfries CG, et al. A double blind evaluation of electroconvulsive therapy in Parkinson's disease with "on-off" phenomena. Acta Neurol Scand 1987;76:191-9. 7 Ward C, Stern GM, Pratt TRC, McKenna P. Electroconvulsive therapy in parkinsonian patients with the "on-off' syndrome. JNeural Transm 1980;49:13345.
Caseload or workload? SIR,-We agree with Messrs S M Jones and C D Collins that the surgical profession should develop an agreed measure of surgical workload, an essential precursor to surgical audit.' We believe, however, that the British United Provident Association (BUPA) schedule of procedures will be inadequate for measuring all surgical workload. We have compared the complexity of ear, nose, and throat procedures carried out over three years (1983-5) at Sheffield Children's Hospital on Sheffield residents with that of procedures carried out on non-residents. We calculated average costs per case for each group based on weights derived from the benefits payable for each category of the BUPA schedule, the diagnosis related group, and, as a check, average length of stay. The average cost per case for all ear, nose, and throat cases was £711.%. The table shows average costs for residents and non-residents of Sheffield. Average cost per case for patients undergoing ear, nose, and throat procedures at Sheffield Children's Hospital during 1983-5 calculated by three weighting systems
BUPA schedule Diagnosis related group Average length of stay
Cost for Sheffield resident (£)
Cost for nonresident (£)
713.23
708.08
663.94 669.44
858.82 844.95
The difference in average cost between Sheffield residents and non-residents was £5.15 by the BUPA method. This difference is negligible given the crudeness of the analysis and the average case cost of £711.%. The small difference also reflects the insensitivity of the BUPA schedule, which is caused by almost all patients being in either the minor or intermediate categories. The more complex classification by diagnosis related group, which gives 10 groups for ear, nose, and throat procedures in children, gave a difference of £195certainly important. When average length of stay was used to weight costs the difference was £175. Although the BUPA schedule is easily understood because of its hierarchical nature and may be suitable for surgical audit, we found that it is too insensitive to be used to set prices, at least for ear, nose, and throat surgery. For this reason it is
J E BRAZIER University of Sheffield Medical School, Sheffield S10 2RX 1 Jones SM,CollinsCD. Caseload orworkload? Scoring complexity of operative procedures as a means of analysing workload. BrMedJ 1990;301:324-5. (11 August.) 2 Sanderson S, Craig M, Winyard G, Bevan G. Using diagnosisrelated groups in the NHS. CommuniyMed 1986;8:37-46.
Psychiatric aspects of urinary incontinence SIR,-I hope that the careful study reported by Dr K R W Norton and colleagues will discourage gynaecologists and urologists from dismissing detrusor instability and the symptoms that usually accompany it-frequency, urgency or urge incontinence, a feeling of perineal pressure, and nocturia-as indicative of psychiatric disorder.' In a microbiological study of urine collected from patients undergoing urodynamic investigations we showed that 37 of 42 women with detrusor instability had evidence ofinfection with aerobic or fastidious bacteria before investigation.2 We have isolated bacteria from the urine of many women with urgency or urge incontinence.2 3 Many other women had high counts of fastidious bacteria and some had midstream urine specimens showing pyuria, a finding consistent with a diagnosis of urethral or periurethral infection.34 Evidently, urgency and urge incontinence could be caused by inflammation in and around the proximal urethra and bladder neck and irritability or instability of the bladder might result from this. The many women who suffer from this distressing complaint would be helped if clinicians and microbiologists would suspect and seek an infective cause before starting invasive investigations, which do not help and often exacerbate the symptoms.5 ROSALIND MASKELL
Public Health Laboratory, St Mary's General Hospital, Portsmouth P03 6AQ 1 Norton KRW, Bhat AV, Stanton SL. Psychiatric aspects of urinary incontinence in women attending an outpatient
urodynamic clinic. BrMedJ 1990;301:271-2. (4 August.) 2 Payne SR, Timoney AG, McKenning ST, den Hollander D, Pead LJ, Maskell RM. Microbiological look at urodynamic studies. Lancet 1988;u: 1123-6. 3 Wilkins EGL, Payne SR, Pead PJ, Moss ST, Maskell RM. Interstitial cystitis and the urethral syndrome: a possible answer. BrJ Urol 1989;64:39-44. 4 Maskeli R, Pead L, Sanderson RA. Fastidious bacteria and the urethral syndrome: a 2 year clinical and bacteriological study of 51 women. Lancet 1983;ii:1277-80. 5 Maskeli R. A new look at the diagnosis of infection of the urinary tract and its adjacent structures. I Infect 1989;19:207-17.
Perioperative deaths among children SIR,-Dr Paul M Spargo commented on the criteria for membership of the Association of Paediatric Anaesthetists of Great Britain and Ireland (not Northern Ireland as stated in his
letter).' The association was founded in 1973 to provide a
BMJ VOLUME 301
15 SEPTEMBER 1990
forum for the fairly small number of anaesthetists (now increased to over 100 home members) who spend a substantial part of their time with neonates, infants, and young children. It was recognised from the outset that there were also many anaesthetists who spend some time in paediatric anaesthesia often for children over the age of 3. In an annual scientific meeting of one and a half days it is not possible to provide a forum for the first group and an educational programme for the last. For that reason, after much deliberation, it was decided that members of the association should spend roughly halfoftheir time in paediatric anaesthesia. This is interpreted to include anaesthesia, paediatric intensive care, and a heavy on call paediatric commitment, and all these activities can be aggregated. To meet the educational need of anaesthetists with a lesser paediatric commitment the association has held two day seminars on paediatric anaesthesia in alternate years. These were initially highly successful, but the seminar for 1989 was cancelled owing to insufficient numbers. It probably indicates that the demand has been met by the welcome upsurge in seminars and similar teaching programmes in paediatric anaesthesia organised by the College of Anaesthetists, the Association of Anaesthetists, and the section of anaesthetics of the Royal Society of Medicine. Members of the Association of Paediatric Anaesthetists have made large contributions at these and other specialist meetings. The association is now considering a more flexible educational programme to replace the two day seminar. It may be a "refresher" day open to non-members preceding the annual scientific meeting. The association has also been invited to hold a session on paediatric anaesthesia at the winter meeting of the Association of Anaesthetists in January 1991, which will be for the anaesthetist who may only occasionally anaesthetise children. Postgraduate education in paediatric anaesthesia is being catered for by several organisations and is all the richer for that. WILLIAM J GLOVER
Association of Paediatric Anaesthetists, Hospital for Sick Children, Great Ormond Street, London WC1N 3JH I Spargo PM. Perioperative deaths among children. Br Med J
1990;301:343. (11 August.)
Hereditary (primary) haemochromatosis SIR,-I would like to add a caveat to Dr Niall D C Finlayson's statement that screening would probably benefit patients with hereditary haemochromatosis.' Recently I was asked for an opinion on a man who had been screened and found to have a ferritin concentration of 398 tg/l (normal 20-180 ig/l). He was seeking to extend his mortgage through extra life insurance. He volunteered the information that his father had haemochromatosis, and the insurance firm promptly doubled his premium. Tests confirmed the raised ferritin concentration, normal liver and pancreatic function, and the presence of the HLA-A3 antigen. He was otherwise healthy-a non-smoking, normotensive 40 year old marathon runner who consumed 12 units of alcohol a week. My letter to the insurance firm, stating that I would give him venesection to prevent the effects of his iron loading potential and that he was therefore a better than average risk now that the problem had been discovered, resulted in his premium being reduced to one and a half times the original amount quoted. I have advised him to approach other insurance firms armed with my letter and your editorial, hoping for a more sympathetic outcome. This raises the point yet again that screening has
BMJ VOLUME 301
15 SEPTEMBER 1990
implications for patients' lifestyles that we and they may not appreciate. C J BARTON
Royal Berkshire Hospital,
Reading RGl SAN 1 Finlayson NDC. Hereditary (primary) haemochromatosis. BrMedj 1990;301:350-1. (18-25 August.)
Liver function tests SIR,-M F Laker suggested that measuring the activity of alanine aminotransferase rather than aspartate aminotransferase may improve the specificity of investigations for liver disease,' a view adopted by many on the assumption that alanine aminotransferase is more hepatospecific than aspartate aminotransferase.2 We believe that this is not necessarily correct and that there are several cases in which aspartate aminotransferase may provide an equally reliable assessment of hepatic dysfunction. The ability of laboratory analyses to distinguish categories of hepatobiliary disease was assessed in a prospective, computer based graphical evaluation by probabilistic test analysis.3 With a cut off point of 200 U/I (reference range up to 20 U/1) aspartate aminotransferase was a powerful discriminator between viral hepatitis and other forms of hepatobiliary disease with a sensitivity of 91% and specificity of 95%. To achieve a comparable degree of discrimination a cut off point of 300 U/I was required for alanine aminotransferase (reference range up to 20 U/1). Another important case is hepatotoxicity caused by drugs such as paracetamol and isoniazid, which often results in both an appreciably raised aspartate aminotransferase activity and a raised aspartate aminotransferase: alanine aminotransferase ratio.4 This suggests that aspartate aminotransferase may be a more sensitive marker in this condition. Stability of the enzymes in serum is also important if there may be a delay between separating the sample and analysis, for instance at weekends. Aspartate aminotransferase is stable when refrigerated (4°C) or frozen (-20°C and -80'C),5 whereas alanine aminotransferase is considerably less stable and refrigeration is preferred to freezing. Although alanine aminotransferase is considered by many to be more hepatospecific, the usefulness of aspartate aminotransferase in managing hepatobiliary disorders should not be underestimated and deserves to be critically re-examined against the background of published reports. R CRAMB C M FLORKOWSKI
Queen Elizabeth Medical Centre,
Birmingham B15 2TH I Laker MF. Liver function tests. Br Med J 1990;301:250-1. (4
August.) 2 Aach RD, Szmuness W, Mosley JW, et al. Serum alanine aminotransferase of donors in relation to the risk of non-A nonB hepatitis. The transfusion-transmitted viruses study. N Engl J Med 1981;304:989-94. 3 Borsch G, Baier J, Glocke M, Nathusius W, Gerhardt W. Graphical analysis of laboratory data in the differential diagnosis of cholestasis: a computer assisted prospective study. J Clin
Chem ClinBiochemn 1988;26:509-19. 4 Himmelstein DU, Woolhandler SJ, Adler RD. Elevated SGOT/ SGPT ratio in alcoholic patients with acetaminophen hepato-
toxicity. Amj Gastroenterol 1984;79:718. 5 Niblock AE, Leung FY, Henderson AR. Serum aspartate aminotransferase storage and the effect of pyridoxal phosphate. J Lab Clin Med 1986;108:461.
Research struggles in eastern Europe SIR,-I am writing concerning Alexander Dorozynski's news item.' Getting east European members to Western meetings is expensive but it can be done. The European Spine Society is having a meeting
at the University of Zurich. The university has always refused any trade advertising but on this occasion has allowed a trade exhibition on the understanding that the money be used to bring east Europeans to the meeting. Most meetings in Britain expect to make a profit, and unfortunately this profit is used for administration and often to finance the travelling of members of the executive. In the European Spine Society nobody attends meetings without paying his or her own costs, which, of course, considerably reduces the meetings' cost. The drug and instrument companies are keen to advertise the fact that they are financing east Europeans to attend Western meetings. It also allows companies to meet the surgeons they are supporting and, with the passage of time, to have access to east European markets. MICHAEL SULLIVAN
Royal National Orthopaedic Hospital, London WI P 8AQ 1 Dorozynski A. Research struggles in eastern Europe. Br Med J 1990;301:305-6. (11 August.)
Mortality in patients with bleeding peptic ulcer SIR,-Mr K E Wheatley and colleagues clearly show the importance of a defined protocol in achieving a low mortality from bleeding peptic ulcers.' It is not so clear, however, whether low mortality can be achieved only in specialised units. Two studies from hospitals with specialist units have reported low overall mortality figures (5 5% and 4 8%) using similar protocols,2' but we have obtained comparable results in a district general hospital without such a unit.4 An identical protocol for the indications for surgery was used (again without using endoscopic stigmata) and over one year an overall mortality of 4-6% was recorded. Only one patient out of 60 with proved bleeding peptic ulcers died. We therefore suggest that although specialist units confer many logistic advantages in managing patients with gastrointestinal bleeding, they are not the most important factor in achieving a low mortality. The key is an agreed clear policy of management and active cooperation between specialties. ` D CLEMENTS D FOSTER
J STAMATAKIS W E WILKINS J S MORRIS
Princess of Wales Hospital, Bridgend CF31 I RQ I Wheatley KE, Snyman JH, Brearley S, Keighley MRB, Dykes PW. Mortality in patients with bleeding peptic ulcer when those aged 60 or over are operated on early. Br Med 7 1990;301:272. (4 August.) 2 Holman RAE, Davis M, Gough KR, Gartell P, Britton DC, Smith RB. Value of a centralised approach in the management of haematemesis and melaena: experience in a district general hospital. Gut 1990;31:504-8. 3 Sanderson JD, Taylor RFH, Pugh S, Vicar FR. Specialised gastrointestinal units for the management of upper gastrointestinal haemorrhage. Postgrad Medy 1990;66:654-6. 4 Clements D, Aslan S, Foster D, Stamatakis J, Wilkins WE, Morris JS. Acute upper gastrointestinal haemorrhage in a district general hospital. Audit of an agreed management policy. J R Coll Physicians Land (in press). 5 Madden MV, Griffith GH. Management of upper gastrointestinal bleeding in a district general hospital. J R Coll Physicians Lond
1986;20:212-5.
Correction Papal policy, poverty, and AIDS A printer's error occurred in this letter by Dr J Guillebaud (1 September, p 440). The last sentence should have read, "Can anyone explain to me why it is not wrong thus to ensure ... that millions of sperm die in the fallopian tubes without fertilising an egg, yet it is wrong to ensure, within marriage, that they die instead in a rubber condom."
557
