Health care providers, be they directly managed units, self governing trusts, or private hospitals, will also be affected by fundholding. The main benefit to them is that they can secure more business by targeting fundholding practices. On the other hand, they may decide to concentrate on meeting the needs of the main purchasing authorities. The new family health services authorities may also think that the interest in the fundholding general practitioner gives them an opportunity to have a wider influence in the provision of health care. On the other hand, much of their time and effort will be directed towards those practices that are not fundholding. The final party in the assessment of the benefits and risks of fundholding is the community at large. Perhaps only 10% of the population will initially be registered with fundholding practices. If greater consumer awareness on the part of general practitioners and their patients has an appreciable impact the biggest potential benefit of fundholding is that it will improve the quality of primary and secondary care for all the population. The biggest risk is that fundholding will increase the inequalities in health care provision if some practices can deliver a better standard of service than others. This is particularly so since the minimum practice size for fundholding excludes the many singlehanded and small practices that are more common in inner city areas. As there have been few opportunities to assess the benefits and risks of fundholding before its implementation, it will be important to monitor both efficacy and ''adverse reactions" after its implementation. Some research is planned7 and more
is needed. Everyone, including the government, should realise that the inappropriate application of managerial innovations can do as much, if not more, harm to patients as an untested drug, device, or surgical procedure. MICHAEL DRUMMOND
Professor of Economics, Centre of Economics, University of York, York YO 1 5DD BERNARD CRUMP
Acting Director of Public Health, Central Birmingham Health Authority, Birmingham B 15 2TZ
ROGER HAWKES General Practitioner, Tamworth, Staffordshire B79 7JN
MICHAEL MARCHMENT Unit General Manager (Community and Mental Health), South East Staffordshire Health Authority, Burton on Trent DE14 2ED 1 Bevan G, Maynard A, Holland W, Mays N. Reform'ing UK health care to improve health: the case for research and experiment. York: Centre for Health Economics, University of York, 1989. 2 Drummond M, Maynard A. Efficiency in the NHS: lessons from abroad. Health Policy 1988;9:5974. 3 Mechanic D. Consumer choice among health insurance options. Health Affairs 1989;8:138-48. 4 Vladeck B, Goodwin E, Myers L, Sinisi M. Consumers and hospital use: the HCFA "death list." Health Affairs 1988;7:122-5. 5 Hillman AL. Financial incentives for physicians in HMOs: is there a conflict of interest? N EnglJ Med 1987;317:1743-8. 6 Hillman AL, Pauly MV, Kertein JJ. How do financial incentives affect physicians' clinical decisions and the financial performance of health maintenance organisations? N EnglJ Med 1989;321:8692. 7 Judge K. Monitoring and evaluating Working for Patients. BMJ 1989;299:1385-7.
Children with head injuries Every district needs a paediatrician to oversee services for them Young people are characterised by inexperience, impulsivity, curiosity, bravado, and high levels of physical activity, all of which make them especially vulnerable to accidents.' Accident prevention programmes are more likely to succeed in changing the environment than the nature of childhood. All children hit their heads hard at some time or other, and one in 15 will have had a head injury that "caused parents serious worry."2 Children are at risk at all ages,3 but boys are injured twice as often as girls.4 Fortunately, youth also conveys some protection: children under the age of 10 are many times more likely to survive a given head injury than are the elderly.5 Fewer than 5% die at the site of the accident, and if they reach hospital alive more than 95% survive.6 What long term problems do these survivors have? In the 1970s neurologists believed that "plasticity" (the ability of the brain to adapt and to recuperate7) protected the child against the sequelae observed in adults, and that this explained the occasional "miraculous" recovery of a toddler who fell on to his or her head from the window of a tower block and the virtual absence in children of the postconcussion syndrome.8 The intelligence quotient after head injury is usually normal,9-12 and few children have the sort of detailed neuropsychological assessment that can show specific brain damage. Full neurological examination of a 2 year old can be very difficult, and children do not always present their symptoms in a conventional way. All these factors have combined to make doctors seriously underestimate the extent of the residual problems. Those studies that have looked specifically at children have shown that 20% of the survivors of severe injury have motor sequelae, and 10% have severe learning BMJ
VOLUME 301
8 DECEMBER 1990
difficulties persisting beyond nine months. 13 They have difficulties with short term memory, concentration, and the processing of new learning"'7; language disorders"8'21; visuospatial problems22; changes in personality23 24; and behavioural or emotional disorders.25 Although improvement may continue for at least five years, the most rapid recovery takes place in the first few months.26 Last year the Paediatric Neurology Association drew attention to the lack of consistency in the long term management of these children.27 Parents who have witnessed the intensive input of skills and technology into the lifesaving process may become frustrated and embittered by the almost complete absence of effective rehabilitation. All too often within a few weeks of injury a child develops severe secondary damage from contractures, abnormal motor patterns, and behavioural disturbances-all of which are preventable. Such a child will end up in a special school, but most who are less severely injured return to mainstream schooling (often "put back a year") with little acknowledgment of their specific learning disability. This situation could easily be improved. Every child with head injuries, wherever admitted, should be under the care of a paediatrician; a paediatric rehabilitation team should be concerned in the management of severely injured children and their families from the time of admission28; and the few children (one every couple of months in an average district) who need intensive inpatient rehabilitation need to be included in the bed allocation. Removing a child to a remote specialist centre is rarely indicated and may perpetuate the delay in obtaining services. Regions might find it more 1289
efficient to develop a single paediatric rehabilitation resource team to train and advise on head injury. The report of the working party of the Medical Disability Society on the management of traumatic brain injury virtually ignored services for children.29 That omission is further evidence of the need for separate advocacy -a paediatrician in every district to develop and provide services for children with head injuries. MARION CROUCHMAN
Consultant Paediatrician, Sheldon Centre, King's College Hospital, London SE5 7RN I Manheimer D, l\ellinger GD. Personality characteristics of the child accident repeater. Child Dez, 1967;38:491-5 13. Rune V. Acute head injuries in children. Acta PaediatrScand 1970;suppl 209. Department of Health and Social Security. Hospital inpatient enquirv. London: HMSO, 1987. Craft AW, Shaw DA, Cartilidge NE. Head injuries in children. BMJ 1972;iv:200-3. Conroy C, Kraus JF. Survival after brain injury. Neuroepidemtiology 1988;7:3-22. Kraus JF, Fife D, Conroy C. Pediatric brain injuries: the nature, clinical course, and earlI outcomes in a defined United States population. Pediatrics 1987;79:501-7. 7 St James-Roberts J. Neurological plasticity, recovery from brain insult and child development. In: Reese HW, Lipsitt LP, eds. Advances in child dezvelopzent and behavior. Vol 14. New York: Academic Press, 1979:255-319. 8 Dillon H, Leopold RL. Children and the post-concussion syndrome. JAMA 1961;175:86-92. 9 Brink JD, Garrett AL, Hale WR, Woo-Sam J, Nickel XL. Recovery of motor and intellectual function in children sustaining severe head injuries. Dev Med Child Neurol 1970;12:565-71. 10 Mandleberg IA. Cognitise recovery after severe head injury: 3 WAIS verbal and performance lQs
2 3 4 5 6
as a function ot' post traumatic amnesia duration and time from injury. J Neurol Neurosurg Psychiatry 1976;39: 1001-7. 11 Chadwick 0, Rutter M, Brown G. A prospective study of children with head injuries. 2 Cognitive sequclae. Psychol/.Med 1981;11:49-61. 12 I)cncker SJ, Lofsing B. A psychometric study of identical twins discordant for closed head in jury. Acta Psvchiatrica et Neurologica Scandinavica 1958;33:(suppl 122). 13. Mahoney WCJ, D'Souza BN, Haller JA, Rogers MC, Epstein MH, Freeman JM. Long term outcome of children with sesere head. trauma and prolonged coma. Pediatrics 1983;71:756-62. 14 Chadwick 0, Rutter M, Shaffer D. A prospective study of children with head injuries. 4:'Specific cognitise deficits. Neuropsychologs' 1981;3:101-20. 15 Levin HS, Benton AL, Grossman RG. Neurobehavioural consequences of closed head injury. Oxford: Oxford University Press, 1982. 16 Brooks DN. Long and short term memory in head-injured patients. Cortex 1975;11:329-40. 17 Bawden HN, Knights RAM, Winogron HW. Speeded performance following head injury in children. J Clin Exp .Veuropsvchol 1985;7:39-54. 18 Levin HS, Grossman RG, Kelly PJ. Aphasic disorder in patients with closed head injury. Neurol Neurosurg Psychiatry 1976;39: 1062-70. 19 Sarno MT. The nature of verbal impairment after closed head injury. 7 Nertr Ment Dis 1976;168:685-92. 20 Levin HS, Grossman RG, Sarwar M, Meyers CA. Linguistic recovery after closed head injury. Brain Lang 1981;12:360-74. 21 Ewing-Cobbs L, Levin HS, Eisenberg HM. Language functions following closed-head injury in children and adolescents. Journal of Clinical and Experimental Psychology 1987;9:575-92. 22 Brooks DN. Memory and head injury.J Nerv Ment Dis 1972;155:350-5. 23 Brooks DN, MicKinlav WW. Personality and behavioural changes after sesere blunt head injury -a relative's view. J Neurol Neurosurg Psychiatry 1983;46:336-44. 24 Rosen CD, Gerring JP. Head trauma: educational reintegration. London: Taylor and Francis, 1961. 25 Brown G, Chadwick 0, Shaffer D, Rutter M, Traub M. A prospective study of children with head injuries in adulthood: 3. Psychiatric sequelae. PsvcholMed 1981;11:63-78. 26 Klonoff H, Low MD, Clark C. Head injuries in children: a prospective 5 year follow up. J Neurol A'eurosurgPsychiatry 1977;40:1211-9. 27 Robinson OR, AMellor DH. Needs of children with neurosurgical problems. BM3r 1989;298:154. 28 Crouchman MR. Head injury: how the community paediatrician can help. Arch Dis Child 1990;65:1286-7. 29 M\edical Disability Society. Report of working party on the management of traumatic bratn injury. London: Development Trust for the Young Disabled, 1988.
New directions in primary biliary cirrhosis Bile acids look promising Primary biliary cirrhosis has a poor reputation for various reasons. Not least of the problems is that patients do not like being diagnosed as having any sort of cirrhosis because of its implied allegation of alcohol abuse. In fact the condition is not a true cirrhosis at all in the early stages, but suggestions to rename it non-suppurative destructive cholangitis or disappearing bile duct disease, or even (my personal favourite) autoimmune hepatic triaditis have not caught on. The condition is often thought to be rare and invariably unpleasant whereas in reality it is common and often asymptomatic.' 5 The only known cure is liver transplantation, which is a reasonable option in those patients with liver failure indicated by pronounced jaundice.67 Because primary biliary cirrhosis may be a chronic progressive disabling disease a wide variety of treatments has been assessed. The lack of a proved specific drug cure means that management has often been supportive and aimed at alleviating symptoms. Many drugs have been used in an attempt to suppress the inflammation in the liver, minimise symptoms, and prolong survival. Glucocorticoids were the first tested.' Initial enthusiasm was tempered by the fact that osteoporotic bone disease worsened during treatment. Recent interest has been shown in using prednisolone, perhaps with adjuvant drugs to reduce the hazards of bone softening, but such treatment is problematical.9 The chelating drug penicillamine was tested because of the observation that copper accumulates in the liver of patients with primary biliary cirrhosis and that patients might benefit if this could be cleared. Though there was initial evidence of improved survival,'0 toxicity led to this line of treatment being abandoned." 12 Similarly, the opioid antagonist nalmefene has been difficult to use because of side effects.'3 Oestradiol treatment in women has been tested, but results were variable.'4 Immune suppression with azathioprine showed some 1290
promise and might lengthen survival, though again the evidence is equivocal.'5 16 Chlorambucil has also been used.'7 Cyclosporin may be too toxic for use in large doses on its own, causing hypertension and renal impairment, but combined with prednisolone has shown some apparent benefit."8-2 Methotrexate has been proposed despite the fact that large doses are known to cause cirrhosis and also acute hepatic failure.2223 The paradoxical approach of using the immune stimulant levamisole did not give encouraging results.24 Colchicine as an inhibitor of fibrosis seems a harmless line of treatment which might improve survival,2528 but again long term effects are uncertain and side effects such as blood dyscrasia and peripheral neuropathy are not unknown.29 We are therefore faced with a disease that is nasty and treatment that can be nastier. Against this setting a serendipitous discovery opened up a new possibility for treatment. Although coexistent liver disease was said to be a specific contraindication to dissolving gall stones with bile acids, some patients were so treated, and in 1981 Leuschner et al reported biochemical improvements.30 At the time there was only mild interest, but in 1987 an uncontrolled study of 15 French patients with primary biliary cirrhosis showed striking reductions in abnormal serum liver function values and good symptom control after treatment with ursodeoxycholic acid.3' The philosophy of bile acid treatment is that the hepatocyte affected by autoimmune processes is further damaged because of cholestatic accumulation of endogenous bile acids, such as chenodeoxycholic acid and cholic acid.3'34 Partial replacement of normal human bile acids by the hydrophilic and theoretically less harmful ursodeoxycholic acid may be expected to reduce liver inflammation and pruritus.35 There may also be direct effects on the immune system, indicated by reduction in abnormal HLA class I expression on hepatocytes.36 In addition to these effects there are improvements in abnormal hepatocyte membrane fluidity and some reduction
BMJ
VOLUME
301
8 DECEMBER 1990
is needed. Everyone, including the government, should realise that the inappropriate application of managerial innovations can do as much, if not more, harm to patients as an untested drug, device, or surgical procedure. MICHAEL DRUMMOND
Professor of Economics, Centre of Economics, University of York, York YO 1 5DD BERNARD CRUMP
Acting Director of Public Health, Central Birmingham Health Authority, Birmingham B 15 2TZ
ROGER HAWKES General Practitioner, Tamworth, Staffordshire B79 7JN
MICHAEL MARCHMENT Unit General Manager (Community and Mental Health), South East Staffordshire Health Authority, Burton on Trent DE14 2ED 1 Bevan G, Maynard A, Holland W, Mays N. Reform'ing UK health care to improve health: the case for research and experiment. York: Centre for Health Economics, University of York, 1989. 2 Drummond M, Maynard A. Efficiency in the NHS: lessons from abroad. Health Policy 1988;9:5974. 3 Mechanic D. Consumer choice among health insurance options. Health Affairs 1989;8:138-48. 4 Vladeck B, Goodwin E, Myers L, Sinisi M. Consumers and hospital use: the HCFA "death list." Health Affairs 1988;7:122-5. 5 Hillman AL. Financial incentives for physicians in HMOs: is there a conflict of interest? N EnglJ Med 1987;317:1743-8. 6 Hillman AL, Pauly MV, Kertein JJ. How do financial incentives affect physicians' clinical decisions and the financial performance of health maintenance organisations? N EnglJ Med 1989;321:8692. 7 Judge K. Monitoring and evaluating Working for Patients. BMJ 1989;299:1385-7.
Children with head injuries Every district needs a paediatrician to oversee services for them Young people are characterised by inexperience, impulsivity, curiosity, bravado, and high levels of physical activity, all of which make them especially vulnerable to accidents.' Accident prevention programmes are more likely to succeed in changing the environment than the nature of childhood. All children hit their heads hard at some time or other, and one in 15 will have had a head injury that "caused parents serious worry."2 Children are at risk at all ages,3 but boys are injured twice as often as girls.4 Fortunately, youth also conveys some protection: children under the age of 10 are many times more likely to survive a given head injury than are the elderly.5 Fewer than 5% die at the site of the accident, and if they reach hospital alive more than 95% survive.6 What long term problems do these survivors have? In the 1970s neurologists believed that "plasticity" (the ability of the brain to adapt and to recuperate7) protected the child against the sequelae observed in adults, and that this explained the occasional "miraculous" recovery of a toddler who fell on to his or her head from the window of a tower block and the virtual absence in children of the postconcussion syndrome.8 The intelligence quotient after head injury is usually normal,9-12 and few children have the sort of detailed neuropsychological assessment that can show specific brain damage. Full neurological examination of a 2 year old can be very difficult, and children do not always present their symptoms in a conventional way. All these factors have combined to make doctors seriously underestimate the extent of the residual problems. Those studies that have looked specifically at children have shown that 20% of the survivors of severe injury have motor sequelae, and 10% have severe learning BMJ
VOLUME 301
8 DECEMBER 1990
difficulties persisting beyond nine months. 13 They have difficulties with short term memory, concentration, and the processing of new learning"'7; language disorders"8'21; visuospatial problems22; changes in personality23 24; and behavioural or emotional disorders.25 Although improvement may continue for at least five years, the most rapid recovery takes place in the first few months.26 Last year the Paediatric Neurology Association drew attention to the lack of consistency in the long term management of these children.27 Parents who have witnessed the intensive input of skills and technology into the lifesaving process may become frustrated and embittered by the almost complete absence of effective rehabilitation. All too often within a few weeks of injury a child develops severe secondary damage from contractures, abnormal motor patterns, and behavioural disturbances-all of which are preventable. Such a child will end up in a special school, but most who are less severely injured return to mainstream schooling (often "put back a year") with little acknowledgment of their specific learning disability. This situation could easily be improved. Every child with head injuries, wherever admitted, should be under the care of a paediatrician; a paediatric rehabilitation team should be concerned in the management of severely injured children and their families from the time of admission28; and the few children (one every couple of months in an average district) who need intensive inpatient rehabilitation need to be included in the bed allocation. Removing a child to a remote specialist centre is rarely indicated and may perpetuate the delay in obtaining services. Regions might find it more 1289
efficient to develop a single paediatric rehabilitation resource team to train and advise on head injury. The report of the working party of the Medical Disability Society on the management of traumatic brain injury virtually ignored services for children.29 That omission is further evidence of the need for separate advocacy -a paediatrician in every district to develop and provide services for children with head injuries. MARION CROUCHMAN
Consultant Paediatrician, Sheldon Centre, King's College Hospital, London SE5 7RN I Manheimer D, l\ellinger GD. Personality characteristics of the child accident repeater. Child Dez, 1967;38:491-5 13. Rune V. Acute head injuries in children. Acta PaediatrScand 1970;suppl 209. Department of Health and Social Security. Hospital inpatient enquirv. London: HMSO, 1987. Craft AW, Shaw DA, Cartilidge NE. Head injuries in children. BMJ 1972;iv:200-3. Conroy C, Kraus JF. Survival after brain injury. Neuroepidemtiology 1988;7:3-22. Kraus JF, Fife D, Conroy C. Pediatric brain injuries: the nature, clinical course, and earlI outcomes in a defined United States population. Pediatrics 1987;79:501-7. 7 St James-Roberts J. Neurological plasticity, recovery from brain insult and child development. In: Reese HW, Lipsitt LP, eds. Advances in child dezvelopzent and behavior. Vol 14. New York: Academic Press, 1979:255-319. 8 Dillon H, Leopold RL. Children and the post-concussion syndrome. JAMA 1961;175:86-92. 9 Brink JD, Garrett AL, Hale WR, Woo-Sam J, Nickel XL. Recovery of motor and intellectual function in children sustaining severe head injuries. Dev Med Child Neurol 1970;12:565-71. 10 Mandleberg IA. Cognitise recovery after severe head injury: 3 WAIS verbal and performance lQs
2 3 4 5 6
as a function ot' post traumatic amnesia duration and time from injury. J Neurol Neurosurg Psychiatry 1976;39: 1001-7. 11 Chadwick 0, Rutter M, Brown G. A prospective study of children with head injuries. 2 Cognitive sequclae. Psychol/.Med 1981;11:49-61. 12 I)cncker SJ, Lofsing B. A psychometric study of identical twins discordant for closed head in jury. Acta Psvchiatrica et Neurologica Scandinavica 1958;33:(suppl 122). 13. Mahoney WCJ, D'Souza BN, Haller JA, Rogers MC, Epstein MH, Freeman JM. Long term outcome of children with sesere head. trauma and prolonged coma. Pediatrics 1983;71:756-62. 14 Chadwick 0, Rutter M, Shaffer D. A prospective study of children with head injuries. 4:'Specific cognitise deficits. Neuropsychologs' 1981;3:101-20. 15 Levin HS, Benton AL, Grossman RG. Neurobehavioural consequences of closed head injury. Oxford: Oxford University Press, 1982. 16 Brooks DN. Long and short term memory in head-injured patients. Cortex 1975;11:329-40. 17 Bawden HN, Knights RAM, Winogron HW. Speeded performance following head injury in children. J Clin Exp .Veuropsvchol 1985;7:39-54. 18 Levin HS, Grossman RG, Kelly PJ. Aphasic disorder in patients with closed head injury. Neurol Neurosurg Psychiatry 1976;39: 1062-70. 19 Sarno MT. The nature of verbal impairment after closed head injury. 7 Nertr Ment Dis 1976;168:685-92. 20 Levin HS, Grossman RG, Sarwar M, Meyers CA. Linguistic recovery after closed head injury. Brain Lang 1981;12:360-74. 21 Ewing-Cobbs L, Levin HS, Eisenberg HM. Language functions following closed-head injury in children and adolescents. Journal of Clinical and Experimental Psychology 1987;9:575-92. 22 Brooks DN. Memory and head injury.J Nerv Ment Dis 1972;155:350-5. 23 Brooks DN, MicKinlav WW. Personality and behavioural changes after sesere blunt head injury -a relative's view. J Neurol Neurosurg Psychiatry 1983;46:336-44. 24 Rosen CD, Gerring JP. Head trauma: educational reintegration. London: Taylor and Francis, 1961. 25 Brown G, Chadwick 0, Shaffer D, Rutter M, Traub M. A prospective study of children with head injuries in adulthood: 3. Psychiatric sequelae. PsvcholMed 1981;11:63-78. 26 Klonoff H, Low MD, Clark C. Head injuries in children: a prospective 5 year follow up. J Neurol A'eurosurgPsychiatry 1977;40:1211-9. 27 Robinson OR, AMellor DH. Needs of children with neurosurgical problems. BM3r 1989;298:154. 28 Crouchman MR. Head injury: how the community paediatrician can help. Arch Dis Child 1990;65:1286-7. 29 M\edical Disability Society. Report of working party on the management of traumatic bratn injury. London: Development Trust for the Young Disabled, 1988.
New directions in primary biliary cirrhosis Bile acids look promising Primary biliary cirrhosis has a poor reputation for various reasons. Not least of the problems is that patients do not like being diagnosed as having any sort of cirrhosis because of its implied allegation of alcohol abuse. In fact the condition is not a true cirrhosis at all in the early stages, but suggestions to rename it non-suppurative destructive cholangitis or disappearing bile duct disease, or even (my personal favourite) autoimmune hepatic triaditis have not caught on. The condition is often thought to be rare and invariably unpleasant whereas in reality it is common and often asymptomatic.' 5 The only known cure is liver transplantation, which is a reasonable option in those patients with liver failure indicated by pronounced jaundice.67 Because primary biliary cirrhosis may be a chronic progressive disabling disease a wide variety of treatments has been assessed. The lack of a proved specific drug cure means that management has often been supportive and aimed at alleviating symptoms. Many drugs have been used in an attempt to suppress the inflammation in the liver, minimise symptoms, and prolong survival. Glucocorticoids were the first tested.' Initial enthusiasm was tempered by the fact that osteoporotic bone disease worsened during treatment. Recent interest has been shown in using prednisolone, perhaps with adjuvant drugs to reduce the hazards of bone softening, but such treatment is problematical.9 The chelating drug penicillamine was tested because of the observation that copper accumulates in the liver of patients with primary biliary cirrhosis and that patients might benefit if this could be cleared. Though there was initial evidence of improved survival,'0 toxicity led to this line of treatment being abandoned." 12 Similarly, the opioid antagonist nalmefene has been difficult to use because of side effects.'3 Oestradiol treatment in women has been tested, but results were variable.'4 Immune suppression with azathioprine showed some 1290
promise and might lengthen survival, though again the evidence is equivocal.'5 16 Chlorambucil has also been used.'7 Cyclosporin may be too toxic for use in large doses on its own, causing hypertension and renal impairment, but combined with prednisolone has shown some apparent benefit."8-2 Methotrexate has been proposed despite the fact that large doses are known to cause cirrhosis and also acute hepatic failure.2223 The paradoxical approach of using the immune stimulant levamisole did not give encouraging results.24 Colchicine as an inhibitor of fibrosis seems a harmless line of treatment which might improve survival,2528 but again long term effects are uncertain and side effects such as blood dyscrasia and peripheral neuropathy are not unknown.29 We are therefore faced with a disease that is nasty and treatment that can be nastier. Against this setting a serendipitous discovery opened up a new possibility for treatment. Although coexistent liver disease was said to be a specific contraindication to dissolving gall stones with bile acids, some patients were so treated, and in 1981 Leuschner et al reported biochemical improvements.30 At the time there was only mild interest, but in 1987 an uncontrolled study of 15 French patients with primary biliary cirrhosis showed striking reductions in abnormal serum liver function values and good symptom control after treatment with ursodeoxycholic acid.3' The philosophy of bile acid treatment is that the hepatocyte affected by autoimmune processes is further damaged because of cholestatic accumulation of endogenous bile acids, such as chenodeoxycholic acid and cholic acid.3'34 Partial replacement of normal human bile acids by the hydrophilic and theoretically less harmful ursodeoxycholic acid may be expected to reduce liver inflammation and pruritus.35 There may also be direct effects on the immune system, indicated by reduction in abnormal HLA class I expression on hepatocytes.36 In addition to these effects there are improvements in abnormal hepatocyte membrane fluidity and some reduction
BMJ
VOLUME
301
8 DECEMBER 1990
