BRITISH MEDICAL JOURNAL
6 MARCH 1976
The difficulty of relating the pathological lesions found in the autonomic nerves post mortem to these rather rare and often intermittent clinical syndromes is obvious. In cases of impotence, for example, various different histological changes in the nerves of the corpora cavernosa have been found at necropsy;7 but examination of the vagus, one of the longest autonomic nerves, has yielded less conclusive abnormalities. Abnormal vagal function can be shown in life1 2 8 and is almost always present in diabetic diarrhoea; yet of 12 patients with diabetic diarrhoea examined at necropsy,9-11 identifiable vagal lesions were found in only one.11 The important unmyelinated fibres have not, however, been adequately examined in such studies. The intrinsic nerve plexuses in the gut wall are probably normal;12 and the greater splanchnic nerve, which is a major sympathetic pathway, has also been reported to be normal in diabetic diarrhoea.11 In a more recent study, however, Low et al13 have shown demyelination and other striking abnormalities of this nerve in diabetics with post-mortem evidence of peripheral neuropathy and presumed autonomic neuropathy. These findings were compared with those in alcoholic neuropathy (where autonomic defects are usually absent), where the greater splanchnic nerves were all normal.14 Abnormality of the greater splanchnic nerve is of interest partly because of its possible role in relation to diabetic diarrhoea, but chiefly because failure to control the splanchnic vascular bed could play an important part in the development of postural hypertension.1 5 The central nervous system16 has rarely been examined in studies of autonomic neuropathy. Abnormalities of the paravertebral sympathetic chain have been found.10 17 The most striking of these are giant neurones, which may be seen in the ganglia and in which all stages of degeneration may be observed. It is frustrating that all the changes which have been reported in diabetic autonomic nerves are also found in patients without clinical autonomic neuropathy and more rarely in non-diabetics. No doubt technical difficulties and misinterpretation of microscopical changes account for some of these anomalies. The relation of structure and function remains elusive in the study of autonomic neuropathy, in which biochemical abnormalities and microvascular disease are known to be relevant factors.18 Recent observations on abnormalities of platelet aggregation1 9 have once again focused attention on the vascular aspects of neuropathy, which perhaps need to be examined more closely. Wheeler, T, and Watkins, P J, Bitish Medical jouirnal, 1973, 4, 584. Bennett, T, Hosking, D J, and Hampton, J R, British Medical Jrournal, 1975, 2, 585. 3 Ewing, D J, et al, Clinical Scienice anid Molecular Biology, 1974, 46, 295. 4Murray, A, et al, British Heart Journal, 1975, 37, 882. 5 British Medical_Journal, 1974, 3, 2. 6 Watkins, P J, British Medical Journal, 1973, 1, 583. 7Faerman, I, et al, Diabetes, 1974, 23, 971. 8 Lloyd-Mostyn, R H, and Watkins, P J, British Medical J7ournal, 1975, 3, 15. 9 Berge, K G, et al, Diabetes, 1956, 5, 25. 10 Hensley, G T, and Soergel, K H, Archives of Pathology, 1968, 85, 587. 1 Kristensson, K, et al, Acta Pathologica Microbiologica Scandinavica, Section A, 1971, 79, 684. 12 Whalen, G E, Soergel, K H, and Greenen, J E, Gastroenterology, 1969, 56, 1 2
1021.
13 14
Low, P A, et al, Brain, Low, P A, et al, Brain,
1975, 98, 341. 1975, 98, 357. 15 Wilkins, R W, Culbertson, J W, and Ingelfinger, F J, Jrournal of Clinical Investigationz, 1951, 30, 312. 16 Reske-Nielsen, E, and Lundbaek, K, Diabetologia, 1968, 4, 34. 17 Appenzeller, 0, and Richardson, E P, Neurology, 1966, 16, 1205. 18 Thomas, P K, and Ward, J D, in The Complications of Diabetes, eds H Keen and J Jarrett, p 151. London, Edward Arnold, 1975. 9 O'Malley, B C, et al, Lancet, 1975, 2, 1275.
545
Oral contraceptives and breast neoplasia The relation between contraceptive steroids and tumours of the breast has been studied extensively in animals, and the findings in beagle dogs have led to the withdrawal of two progestogens (chlormadinone acetate and megestrol acetate) from clinical use in Britain.1 The relevance of such experiments on animals to human health is, however, extremely uncertain, and hence epidemiological studies assume great importance. So far as benign lesions of the breast (fibroadenoma and chronic cystic disease) are concerned, the epidemiological findings seem clear-cut: both retrospective2-6 and prospective studies7 8 have shown consistently that the use of oral contraceptives is negatively associated with the appearance of benign lumps in the breast, and that this effect is most pronounced in long-term users. Taken together with the known positive association between benign lesions of the breast and breast cancer,9 10 these results have led some authorities to suggest that oral contraceptives might be more likely to protect against malignant disease of the breast than to cause it. 12
In assessing the results of epidemiological studies in which the relation between oral contraceptives and breast cancer has been examined directly it is essential to remember that most human carcinogens do not produce an overt effect until at least 10-15 years have elapsed from the time of first exposure. Hence it may be some years yet before any final conclusions can be drawn about possible carcinogenic effects of "the pill." In the meantime, however, it is reassuring that none of the published retrospective studies4 6 11 12 or prospective studies7 8 have indicated any overall relationship between oral contraceptive use and breast cancer. To this statement concerning an overall relationship it is necessary to add a qualification. In the largest of the retrospective studies Fasal and Paffenbargerf6 did identify certain subgroups of women who showed a positive association between oral contraceptive use and cancer of the breast. This study began in 1970 in the San Francisco Bay area. Women aged up to 49 years with newly diagnosed breast cancer (or with a benign lesion of the breast) were identified at one or other of 19 hospitals and matched with medical and surgical control patients for age, race, religion, and hospital. Information on oral contraceptive use and other important factors was obtained from interviews at the patients' homes. There were 452 patients with cancer, half of whom had at some time used oral contraceptives; the corresponding figure for 433 medical controls was 48% and for 439 surgical controls 440%. These differences between the cases and controls were not statistically significant. Likewise, there were no significant differences between the cases and controls with respect to the proportions currently using oral contraceptives, the mean duration of use, or the time since oral contraceptives had first been used or last used. Fasal and Paffenbarger did, however, find a statistically significant excess of women who had used oral contraceptives for 2-4 years among the cases (but not of women using the preparations for shorter or longer periods), and they also found an apparent increase in the risk of developing breast cancer among women who had used oral contraceptives for six years or more and had also previously had a biopsy for benign breast disease. The meaning of these findings is difficult to assess, because Fasal and Paffenbarger subdivided their data in so many ways that some "significant" differences
546
would have been expected to have occurred by chance. Nevertheless, other workers should be stimulated to examine their data closely to see if they can detect the same associations. In the meantime, all the available evidence suggests little reason for alarm. British MedicalJrournal, 1975, 4, 608. Vessey, M P, Doll, R, and Sutton, P M, British Medical Journal, 1972, 3, 719. 3 Sartwell, P E, Arthes, F G, and Tonascia, J A, New England journal of Medicine, 1973, 288, 551. 4 Boston Collaborative Drug Surveillance Programme, Lancet, 1973, 1, 1399. 5 Kelsey, J L, Lindfors, K K, and White, C, International Journal of Epidemiology, 1974, 3, 333. 6 Fasal, E, and Paffenbarger, R S, Journal of the National Cancer Institute, 1975, 55, 767. 7 Kay, C R, ed, Oral Contraceptives and Health. London, Pitman, 1974. 8 Ory, H, Cole, P, and MacMahon, B, New England Jrournal of Medicine, 1976, 294, 419. Davis, H H, Simons, M, and Davis, J B, Cancer, 1964, 17, 957. 10 Black, M M, et al, Cancer, 1972, 29, 338. 1 Arthes, F G, Sartwell, P E, and Lewison, E F, Cancer, 1971, 28, 1391. 12 Vessey, M P, Doll, R, and Jones, K, Lancet, 1975, 1, 941. 1 2
Specialists (British style) By long tradition the British assume that when foreigners do something differently from us their method must be inferior. This attitude ran through the meeting called last week by the General Medical Council to discuss the recommendations of the Merrison Report1: no one seriously questioned why the training of an NHS consultant was longer and different from that of a European specialist; it was simply taken for granted that "our standards are higher than theirs." The Merrison Committee had itself recommended that the GMC should call a conference to sound the views of the profession on the regulation of undergraduate and postgraduate education, so that enabling legislation could be drafted along lines broadly agreed by the royal colleges, the BMA, and other main bodies of medical opinion. After the new Medical Act became law the reconstituted, democratically elected GMC could then attend to the details of changes. There is a convincing case for specialist registration by the same body that regulates undergraduate training: it gives greater flexibility in allocating different aspects of medicine to the undergraduate and postgraduate years and makes it easier for there to be experiment (and therefore improvement) in the pattern of medical education. Nevertheless, early on in the course of the debate it became clear that some of the royal colleges and the two joint committees on higher medical and surgical training would not willingly give up their control of the length, content, and quality of postgraduate study in the major specialties. There was no objection to the reconstituted GMC's maintaining a specialist register; but the criteria for registration should, the conference thought, be agreed by the appropriate postgraduate advisory bodies. There would, indeed, have been no real problem had not 1976 been the year in which Britain and the rest of the EEC have to agree on mutual recognition of medical degrees and "formal qualifications of specialised medicine." The original six EEC countries operate a system of specialist registration
BRITISH MEDICAL JOURNAL
6 MARCH 1976
in which recognition of specialist status is given after completion of three, four, or five years of appropriate postgraduate training-the length varying with the specialty. Free movement of doctors within the EEC requires that Britain should adopt a similar system. The difficulty is caused by the differences in the training programmes approved for EEC specialist registration and those (longer) programmes required for accreditation by the British joint committees on higher medical and surgical training. After prolonged debate the conference decided that the only acceptable solution would be a dual system of registration. This would allow a doctor undertaking formal postgraduate training in Britain to be given a certificate of specialist training once he had completed the minimum number of years required by EEC regulations, but he would still have to complete any extra years required for accreditation by his joint committee. Should we not pause a moment to ask why the British have evolved a consultant system so different from that in Europeand whether the Europeans may not have something to be said in their favour ? Without doubt the main problem that has bedevilled the NHS hospital service since its inception has been its impossible staffing structure. Highly trained consultants supported by layers of junior staff make sense in teaching hospitals: they do not and cannot in district hospitals. For 25 years we have preserved a fiction that every hospital consultant should have his retinue of registrar and houseman -and we have preserved it only by importing more than half of the junior staff from abroad. That system is now grinding to a halt, but there are no clear plans for an alternative. There is a lot of talk about increasing consultant posts faster than junior staff numbers, but the changes that have been made in their proportions are tiny. Official manpower projections are still based on a hospital service consisting largely of established consultants and doctors-in-training. The structure in Europe is very different.2 There, most nonteaching hospitals have a staff largely made up of specialists with very few juniors: the specialists admit and clerk the patients, run the outpatient clinics, and generally carry out many of the duties of NHS junior staff. Given the manpower, the British system may well be ideal for the patient with an acute illness in one of the "sharpended" specialties, and it gives the consultant a demanding, satisfying role. But the system works only if half the junior staff are transients seeking no permanent career in the NHS (either in the hospital service or general practice). The time has surely come when the medical profession in Britain should face reality and plan a staffing structure for the hospital service that does not rely on doctors from overseas, that takes account of the growing proportion of women doctors, and that recognises that there is a demand for permanent, part-time hospital appointments from doctors other than GP principals. Once a rational structure is agreed it would become possible to plan a system of undergraduate and postgraduate education to provide the doctors needed. Our present system is very effective in training consultants: but it can provide only a small proportion with the opportunity to make full use of that training. Report of the committee of inquiry into the regulation of the medical profession (chairman Dr A W Merrison). London, HMSO, 1975. 2 Murray, J F, British Medical3Journal, 1976, 1, 507.
6 MARCH 1976
The difficulty of relating the pathological lesions found in the autonomic nerves post mortem to these rather rare and often intermittent clinical syndromes is obvious. In cases of impotence, for example, various different histological changes in the nerves of the corpora cavernosa have been found at necropsy;7 but examination of the vagus, one of the longest autonomic nerves, has yielded less conclusive abnormalities. Abnormal vagal function can be shown in life1 2 8 and is almost always present in diabetic diarrhoea; yet of 12 patients with diabetic diarrhoea examined at necropsy,9-11 identifiable vagal lesions were found in only one.11 The important unmyelinated fibres have not, however, been adequately examined in such studies. The intrinsic nerve plexuses in the gut wall are probably normal;12 and the greater splanchnic nerve, which is a major sympathetic pathway, has also been reported to be normal in diabetic diarrhoea.11 In a more recent study, however, Low et al13 have shown demyelination and other striking abnormalities of this nerve in diabetics with post-mortem evidence of peripheral neuropathy and presumed autonomic neuropathy. These findings were compared with those in alcoholic neuropathy (where autonomic defects are usually absent), where the greater splanchnic nerves were all normal.14 Abnormality of the greater splanchnic nerve is of interest partly because of its possible role in relation to diabetic diarrhoea, but chiefly because failure to control the splanchnic vascular bed could play an important part in the development of postural hypertension.1 5 The central nervous system16 has rarely been examined in studies of autonomic neuropathy. Abnormalities of the paravertebral sympathetic chain have been found.10 17 The most striking of these are giant neurones, which may be seen in the ganglia and in which all stages of degeneration may be observed. It is frustrating that all the changes which have been reported in diabetic autonomic nerves are also found in patients without clinical autonomic neuropathy and more rarely in non-diabetics. No doubt technical difficulties and misinterpretation of microscopical changes account for some of these anomalies. The relation of structure and function remains elusive in the study of autonomic neuropathy, in which biochemical abnormalities and microvascular disease are known to be relevant factors.18 Recent observations on abnormalities of platelet aggregation1 9 have once again focused attention on the vascular aspects of neuropathy, which perhaps need to be examined more closely. Wheeler, T, and Watkins, P J, Bitish Medical jouirnal, 1973, 4, 584. Bennett, T, Hosking, D J, and Hampton, J R, British Medical Jrournal, 1975, 2, 585. 3 Ewing, D J, et al, Clinical Scienice anid Molecular Biology, 1974, 46, 295. 4Murray, A, et al, British Heart Journal, 1975, 37, 882. 5 British Medical_Journal, 1974, 3, 2. 6 Watkins, P J, British Medical Journal, 1973, 1, 583. 7Faerman, I, et al, Diabetes, 1974, 23, 971. 8 Lloyd-Mostyn, R H, and Watkins, P J, British Medical J7ournal, 1975, 3, 15. 9 Berge, K G, et al, Diabetes, 1956, 5, 25. 10 Hensley, G T, and Soergel, K H, Archives of Pathology, 1968, 85, 587. 1 Kristensson, K, et al, Acta Pathologica Microbiologica Scandinavica, Section A, 1971, 79, 684. 12 Whalen, G E, Soergel, K H, and Greenen, J E, Gastroenterology, 1969, 56, 1 2
1021.
13 14
Low, P A, et al, Brain, Low, P A, et al, Brain,
1975, 98, 341. 1975, 98, 357. 15 Wilkins, R W, Culbertson, J W, and Ingelfinger, F J, Jrournal of Clinical Investigationz, 1951, 30, 312. 16 Reske-Nielsen, E, and Lundbaek, K, Diabetologia, 1968, 4, 34. 17 Appenzeller, 0, and Richardson, E P, Neurology, 1966, 16, 1205. 18 Thomas, P K, and Ward, J D, in The Complications of Diabetes, eds H Keen and J Jarrett, p 151. London, Edward Arnold, 1975. 9 O'Malley, B C, et al, Lancet, 1975, 2, 1275.
545
Oral contraceptives and breast neoplasia The relation between contraceptive steroids and tumours of the breast has been studied extensively in animals, and the findings in beagle dogs have led to the withdrawal of two progestogens (chlormadinone acetate and megestrol acetate) from clinical use in Britain.1 The relevance of such experiments on animals to human health is, however, extremely uncertain, and hence epidemiological studies assume great importance. So far as benign lesions of the breast (fibroadenoma and chronic cystic disease) are concerned, the epidemiological findings seem clear-cut: both retrospective2-6 and prospective studies7 8 have shown consistently that the use of oral contraceptives is negatively associated with the appearance of benign lumps in the breast, and that this effect is most pronounced in long-term users. Taken together with the known positive association between benign lesions of the breast and breast cancer,9 10 these results have led some authorities to suggest that oral contraceptives might be more likely to protect against malignant disease of the breast than to cause it. 12
In assessing the results of epidemiological studies in which the relation between oral contraceptives and breast cancer has been examined directly it is essential to remember that most human carcinogens do not produce an overt effect until at least 10-15 years have elapsed from the time of first exposure. Hence it may be some years yet before any final conclusions can be drawn about possible carcinogenic effects of "the pill." In the meantime, however, it is reassuring that none of the published retrospective studies4 6 11 12 or prospective studies7 8 have indicated any overall relationship between oral contraceptive use and breast cancer. To this statement concerning an overall relationship it is necessary to add a qualification. In the largest of the retrospective studies Fasal and Paffenbargerf6 did identify certain subgroups of women who showed a positive association between oral contraceptive use and cancer of the breast. This study began in 1970 in the San Francisco Bay area. Women aged up to 49 years with newly diagnosed breast cancer (or with a benign lesion of the breast) were identified at one or other of 19 hospitals and matched with medical and surgical control patients for age, race, religion, and hospital. Information on oral contraceptive use and other important factors was obtained from interviews at the patients' homes. There were 452 patients with cancer, half of whom had at some time used oral contraceptives; the corresponding figure for 433 medical controls was 48% and for 439 surgical controls 440%. These differences between the cases and controls were not statistically significant. Likewise, there were no significant differences between the cases and controls with respect to the proportions currently using oral contraceptives, the mean duration of use, or the time since oral contraceptives had first been used or last used. Fasal and Paffenbarger did, however, find a statistically significant excess of women who had used oral contraceptives for 2-4 years among the cases (but not of women using the preparations for shorter or longer periods), and they also found an apparent increase in the risk of developing breast cancer among women who had used oral contraceptives for six years or more and had also previously had a biopsy for benign breast disease. The meaning of these findings is difficult to assess, because Fasal and Paffenbarger subdivided their data in so many ways that some "significant" differences
546
would have been expected to have occurred by chance. Nevertheless, other workers should be stimulated to examine their data closely to see if they can detect the same associations. In the meantime, all the available evidence suggests little reason for alarm. British MedicalJrournal, 1975, 4, 608. Vessey, M P, Doll, R, and Sutton, P M, British Medical Journal, 1972, 3, 719. 3 Sartwell, P E, Arthes, F G, and Tonascia, J A, New England journal of Medicine, 1973, 288, 551. 4 Boston Collaborative Drug Surveillance Programme, Lancet, 1973, 1, 1399. 5 Kelsey, J L, Lindfors, K K, and White, C, International Journal of Epidemiology, 1974, 3, 333. 6 Fasal, E, and Paffenbarger, R S, Journal of the National Cancer Institute, 1975, 55, 767. 7 Kay, C R, ed, Oral Contraceptives and Health. London, Pitman, 1974. 8 Ory, H, Cole, P, and MacMahon, B, New England Jrournal of Medicine, 1976, 294, 419. Davis, H H, Simons, M, and Davis, J B, Cancer, 1964, 17, 957. 10 Black, M M, et al, Cancer, 1972, 29, 338. 1 Arthes, F G, Sartwell, P E, and Lewison, E F, Cancer, 1971, 28, 1391. 12 Vessey, M P, Doll, R, and Jones, K, Lancet, 1975, 1, 941. 1 2
Specialists (British style) By long tradition the British assume that when foreigners do something differently from us their method must be inferior. This attitude ran through the meeting called last week by the General Medical Council to discuss the recommendations of the Merrison Report1: no one seriously questioned why the training of an NHS consultant was longer and different from that of a European specialist; it was simply taken for granted that "our standards are higher than theirs." The Merrison Committee had itself recommended that the GMC should call a conference to sound the views of the profession on the regulation of undergraduate and postgraduate education, so that enabling legislation could be drafted along lines broadly agreed by the royal colleges, the BMA, and other main bodies of medical opinion. After the new Medical Act became law the reconstituted, democratically elected GMC could then attend to the details of changes. There is a convincing case for specialist registration by the same body that regulates undergraduate training: it gives greater flexibility in allocating different aspects of medicine to the undergraduate and postgraduate years and makes it easier for there to be experiment (and therefore improvement) in the pattern of medical education. Nevertheless, early on in the course of the debate it became clear that some of the royal colleges and the two joint committees on higher medical and surgical training would not willingly give up their control of the length, content, and quality of postgraduate study in the major specialties. There was no objection to the reconstituted GMC's maintaining a specialist register; but the criteria for registration should, the conference thought, be agreed by the appropriate postgraduate advisory bodies. There would, indeed, have been no real problem had not 1976 been the year in which Britain and the rest of the EEC have to agree on mutual recognition of medical degrees and "formal qualifications of specialised medicine." The original six EEC countries operate a system of specialist registration
BRITISH MEDICAL JOURNAL
6 MARCH 1976
in which recognition of specialist status is given after completion of three, four, or five years of appropriate postgraduate training-the length varying with the specialty. Free movement of doctors within the EEC requires that Britain should adopt a similar system. The difficulty is caused by the differences in the training programmes approved for EEC specialist registration and those (longer) programmes required for accreditation by the British joint committees on higher medical and surgical training. After prolonged debate the conference decided that the only acceptable solution would be a dual system of registration. This would allow a doctor undertaking formal postgraduate training in Britain to be given a certificate of specialist training once he had completed the minimum number of years required by EEC regulations, but he would still have to complete any extra years required for accreditation by his joint committee. Should we not pause a moment to ask why the British have evolved a consultant system so different from that in Europeand whether the Europeans may not have something to be said in their favour ? Without doubt the main problem that has bedevilled the NHS hospital service since its inception has been its impossible staffing structure. Highly trained consultants supported by layers of junior staff make sense in teaching hospitals: they do not and cannot in district hospitals. For 25 years we have preserved a fiction that every hospital consultant should have his retinue of registrar and houseman -and we have preserved it only by importing more than half of the junior staff from abroad. That system is now grinding to a halt, but there are no clear plans for an alternative. There is a lot of talk about increasing consultant posts faster than junior staff numbers, but the changes that have been made in their proportions are tiny. Official manpower projections are still based on a hospital service consisting largely of established consultants and doctors-in-training. The structure in Europe is very different.2 There, most nonteaching hospitals have a staff largely made up of specialists with very few juniors: the specialists admit and clerk the patients, run the outpatient clinics, and generally carry out many of the duties of NHS junior staff. Given the manpower, the British system may well be ideal for the patient with an acute illness in one of the "sharpended" specialties, and it gives the consultant a demanding, satisfying role. But the system works only if half the junior staff are transients seeking no permanent career in the NHS (either in the hospital service or general practice). The time has surely come when the medical profession in Britain should face reality and plan a staffing structure for the hospital service that does not rely on doctors from overseas, that takes account of the growing proportion of women doctors, and that recognises that there is a demand for permanent, part-time hospital appointments from doctors other than GP principals. Once a rational structure is agreed it would become possible to plan a system of undergraduate and postgraduate education to provide the doctors needed. Our present system is very effective in training consultants: but it can provide only a small proportion with the opportunity to make full use of that training. Report of the committee of inquiry into the regulation of the medical profession (chairman Dr A W Merrison). London, HMSO, 1975. 2 Murray, J F, British Medical3Journal, 1976, 1, 507.
