59
These include height, health behaviours, social circumstances, and the work environment. Unfortunately, explanations for the grade differences in health reported in the Whitehall cohorts continue to elude us for two reasons. First, the grade differences in behavioural, social, and work-related risk factors in Whitehall II were clearly correlated with each other and with some of the established risk factors. The researchers made no attempt to disentangle the complex interrelations that undoubtedly exist between these many variables.
explanations for these differentials.
Consequently it
was not
possible
to
examine the
independent
contribution of any particular factor, or combination of factors, to the health of civil servants. Second, whereas Whitehall I studied the relation between grade and mortality prospectively over ten years,’ Whitehall II was a cross-sectional investigation of health, grade, and associated risks. Although cross-sectional studies can describe the association between prevalent disease and potential risk factors, they cannot establish causality.2 Most people would agree with Marmot et al that attention should be paid to the social environment, job design, and consequences of income inequality. Whitehall II, however, has so far not been able to describe the direct, independent effect of these factors on the health of civil servants. Department of Public Health and Primary Care, Royal Free Hospital School of Medicine, London NW3 2PF, UK
Cardiovascular Research, University Laboratory of Physiology, Oxford 0X1 3PT, UK
Citations of
JONATHAN ELFORD
1. Marmot
MG, Shipley MJ, Rose G. Inequalities in death: specific explanations of a general pattern. Lancet 1984; i: 1003-06. 2. Hennekens CH, Buring JE. Epidemiology in medicine. Boston: Little Brown, 1987.
SiR,—The Whitehall II study of civil servants is depressing for the picture it presents of health inequalities, but for those interested in trying to do something about them it can be viewed optimistically as an opportunity to identify more precisely some of the possible causal mechanisms. Marmot et al state that "An analysis of the relation between speed of promotion and initial health status will be published elsewhere". It is not clear whether this refers to how quickly people have reached their present employment grade in relation to their current health status, or how quickly people will progress from now on, in relation to health status. With respect to causality, the second approach is likely to be more useful than the first. In the event that this second approach is the intention, I suggest that it might be useful to identify, within any particular grade, those who (a) actively seek promotion, and are successful; (b) actively seek promotion, and are not (initially, at least) successful; and (c) do not seek promotion. It is not only health status that might discriminate between these groups, but also the whole range of other influences that Marmot and his colleagues seek as possible explanations of the observed cross-sectional differences between grades (biological, behavioural, and psychological characteristics). A longitudinal follow-up of these subjects could be very enlightening. Department of Social Medicine, Medical School, Birmingham B15 2TJ, UK
some of which were incorporated in a revised version. We were aware that we were close to the work so when the revised paper arrived we sent it out of the country to referees 3 and 4. They reported, together, that the method should be independently tested but did not think it had been well done here. Wilmshurst was sent the report and we offered to look at a third try. A protracted correspondence followed, but no rewrite was submitted. This sort of thing, though rare, will happen on any journal. This is not really an instance of negative findings not being published but had much more do to with trying to change a paper from a polemic to a reasoned critique. The subject was and is controversial and all editors are happy to publish controversy; it brightens up the pages and gets people thinking. We would have been glad to carry the Wilmshurst reply and sorry that we never got to the point where we had a reasoned and reasonable bit of science to publish. It is a shame that he became so involved with accusations of bias that he apparently could not see what points were being made by our referees. It is probably true that negative result papers do not get accepted. Editors should watch out for this; we try to.
improvement,
T. MARSHALL
P. SLEIGHT D. BERGEL
published work
SiR,—Your April 27 editorial does not consider another possible for the fall in the citation impact and the rise in uncitedness of published articles from the UK during the past decade, despite a rise in the number of publications-unwillingness on the part of our American colleagues to cite papers published in non-American journals. This view is shared by several nonAmerican clinical scientists to whom I have spoken and also by some American clinicians (mostly non-faculty members). I wonder if Science Watch will be able to review its analysis to confirm or dispel the above possibility.
explanation
Department of Clinical Haematology, George Hospital, Kogarah, Sydney,
St
NSW 2217, Australia
A. MANOHARAN
Cyclosporin and autoimmune disease SIR,-As chairman of the Second Congress on Immunointervention in Autoimmune Diseases (CIAD ’91) recently held in Paris, I would like to let your readers know that your conference report (June 1, p 1338) covers only a small part of what was discussed over the four days of that meeting and focuses on cyclosporin trials. Three major conclusions could be drawn from the many presentations from scientists and clinicians working with cyclosporin. The efficacy of the drug in several diseases has been well established by randomised controlled trials comparing cyclosporin with placebo or standard therapy-eg, in severe psoriasis,t,2 uveitis,33 rheumatoid arthritis4 and nephrotic
syndromes Although side-effects, especially nephrotoxicity, are a Publication bias SIR,-We well remember the
many letters about the paper
submitted by Dr Wihnshurst (June 8, p 1419) late in 1987 and are happy to break cover. Wilmshurst’s paper was sent to us four months after the appearance of a study testing in man a method based on animal studies in which one of us (D. B.) had been involved. The author of that paper had been trained in Oxford and was interested in the method because he had been taught by its originator. Wilmshurst had looked at it for us and did not like it. We took a second opinion and published because we thought the idea needed looking at. The ensuing submission from Wilmshurst and his colleague did contain some interesting new data. Referee 1 would have liked to see it published but felt that many changes were needed. D. B. agreed, saying that the data were worth publishing "if they can do it properly" (ie, write it better). (The bit about the typewriter is clever.) Both gave detailed suggestions for
concern, the risk is very low when the recommended
daily dose of less than 5 mg/kg is used. No opportunistic infections are seen. Only exceptional cases of lymphoma (less than 1 per 1000) are observed, and they are usually reversible when the immunosuppression is stopped. New indications, based on controlled studies and not just anecdotal evidence, are emerging (eg, Crohn’s diseaseaplastic anaemia,? primary biliary cirrhosis,8 lichen planus,9 atopic dermatitis and allergic asthma [reported at the Congress])—though more data are needed to establish the efficacy and evaluate the risk/benefit ratio. Similar comments apply to other immunosuppressive agents, including FK506 in liver
transplantation. Other agents are needed to complement, and eventually replace, cyclosporin, with the twin aim of improving efficacy and reducing side-effects (both direct toxicity and over-immunosuppression). FK506 and the cyclosporin analogues rapamycin and myocophenolic acid are good candidates but randomised trials
60
comparing them with cyclosporin both for efficacy and side-effects (notably nephrotoxicity, which is not exclusive to cyclosporin) are needed before we can say whether their therapeutic indices are indeed better than that for cyclosporin. In vitro minimum active concentrations do not by themselves defme the therapeutic index. Furthermore
agents may still carry the risk of overand will not solve a major problem encountered with all these agents in most patients-namely, disease relapse after cessation of treatment. The answer may come from new approaches fully discussed at the Paris meeting-namely, monoclonal antibodies, autoantigen peptide therapy, T cell vaccination, and so on. Immunointervention is not just a matter of selecting the best immunosuppressive drug but searching for the best strategy that allows us to slow down or stop the autoimmune process in a non-toxic, disease-specific, and long-term fashion. new
immunosuppression
Department of Clinical Immunology, Hôpital Necker,
Research Unit for Genetic Disease
J.-F. BACH
75743 Paris, France
CN, Fradin MS, Messana JM, et al. Cyclosporine for plaque-type psoriasis: results of a multidose, double-blind trial. N Engl J Med 1991; 324: 277-84. 2. Mihatsch MJ, Woolff K, eds. Risk/benefit ratio of cyclosporin A in psoriasis. Br J Dermatol 1990; 122 (suppl 36). 3. Masuda K, Nakajima A, Urayama A, Nakae K, Kogure M, Inaba G. Double-masked trial of cyclosporin versus colchicine and long-term open study of cyclosporin in Behçet’s disease. Lancet 1989; i: 1093-96. 4. Tugwell P, Bombardier C, Gent M, et al. Low-dose cyclosporin versus placebo in patients with rheumatoid arthritis. Lancet 1990; 335: 1051-55. 5. Ponticelli C, ed. Cyclosporin A in nephroric syndrome. Clin Nephrol 1991; 35 (suppl 1). 6. Brynskov J, Freund L, Norby Rasmussen S, et al. A placebo-controlled, double-blind, randomized trial of cyclosporine therapy in active chronic Crohn’s disease. N Engl J Med 1989; 321: 845-50. 7. Frickhofen N, Kaltwasser JP, Schrezenmeier H, et al. Treatment of aplastic anemia with antilymphocyte globulin and methylprednisolone with or without cyclosporine. N Engl J Med 1991; 324: 1297-304. 8. Weisner RH, Ludwig J, Lindor KD, et al. A controlled trial of cyclosporine in the treatment of primary biliary cirrhosis. N Engl Med 1990; 322: 1419-24. J 9. Eisen D, Ellis CN, Duell EA, Griffiths CEM. Voorhees JJ. Effect of topical cyclosporine rinse on oral lichen planus. N Engl J Med 1990; 323: 290-94. 1. Ellis
Assessment of the mitochondrial chain
respiratory
SiR,—Trounce et all have suggested mitochondrial respiratory failure as a possible cause of ageing in man. However, the frequent scatter of control data and their overlap with those of patients limits the ability of clinicians to reach a reliable diagnosis of respiratory chain deficiencies (figure, left).2,3However, we have recently observed that a constant ratio of respiratory enzyme activities is a consistent feature of oxidative phosphorylation, irrespective of the tissue tested. This finding may be because a balanced proportion of each enzyme complex is required for optimum function of the respiratory chain. Our observation prompted us to present enzyme activities as ratios rather than absolute values. 1 0
—
COX
(nmo!/nmn
per mg
E
COX/SCCR
protein)
Mitochondrial enzyme activities ratios.
= patients; 0 =controls
We studied 7 patients (A-G) and 34 controls for cytochrome c oxidase (COX) activity, succinate-cytochrome c reductase (SCCR) activity, and COX/SCCR activity ratios in mitochondria isolated from skeletal muscle. While normal COX activities were scattered over almost two orders of magnitude, the COX/SCCR activity ratios showed a normal distribution with a narrow range of values (mean [SD] 29 [0’6]; range 2-4). Under these conditions, patients whose absolute COX values were in the low normal range could be unambiguously diagnosed as COX deficient (figure, right). Although this approach does not hold for generalised enzyme defects, it does improve the ability to recognise and quantify localised deficiencies in the mitochondrial respiratory chain.
as
ratio
both absolute values and
in Children, INSERM U-12, Hôpital des Enfants-Malades, 75743 Paris Cedex 15, France
PIERRE RUSTIN DOMINIQUE CHRETIEN THOMAS BOURGERON ANNE WUCHER
JEAN-MARIE SAUDUBRAY AGNES ROTIG ARNOLD MUNNICH
I, Byrne E, Marzuki S. Decline in skeletal muscle mitochondrial respiratory chain function: possible factor in ageing. Lancet 1989; ii: 637-39. 2. Moraes CT, Shanske S, Tritschler HJ, et al. mtDNA depletion with variable tissue expression: a novel genetic abnormality in mitochondrial diseases. Am J Hum Genet 1991; 48: 492-501. 3. Zheng X, Shoffner JM, Voljavec AS, Wallace DC. Evaluation of procedures for assaying oxidative phosphorylation enzyme activities in mitochondrial myopathy muscle biopsies. Biochem Biophys Acta 1990; 1019: 1-10. 1. Trounce
Inhaled corticosteroids, bone formation, and osteocalcin SIR,-Inhaled corticosteroids are now a part of the treatment of asthma. Low dosages of beclomethasone dipropionate (BDP), devoid of effects upon the hypothalamic-pituitary-adrenal axis, can control asthma.l,2 However, there is concern now that higher dosages are in use. A finding of lowered serum osteocalcin in individuals inhaling BDP 2000 ug daily3 suggests that high-dose therapy might be associated with adverse effects upon bone. As part of a larger study evaluating the systemic effects of inhaled BDP we measured osteocalcin across treatments using doses of BDP within the currently recommended therapeutic range (400-2000 Ilg daily). We studied 16 healthy adult volunteers, using a randomised double-blind, placebo-controlled crossover design to evaluate the systemic effects of 100, 200, 350, and 500 g BDP administered by standard metered-dose inhaler every 6 h for 10 days. Wash-out intervals of 18 days separated each treatment period. Plasma for osteocalcin measurement was collected on a fasting specimen taken within 30 min of 0900 h 4 days before starting treatment and repeated at the beginning of the fifth and ninth days of each treatment period. Plasma octeocalcin was measured by an in-house radioimmunoassay and each individual’s samples were assayed in the same batch. The difference from pre-treatment concentration at both days of testing during treatment was analysed by a three-way analysis of variance with factors subject, dose, and phase. The hypothesis of no-treatment effect was tested by a two sided t-test at each dose. During dosing with placebo plasma osteocalcin remained constant. During treatment with BDP there was a significant fall relative to placebo even at the lowest dose of 400 ug daily on treatment day 5. This effect was dose-dependent up to 1400 Ilg with a plateau thereafter (figure). With more prolonged treatment the plateau occurred at an even lower dose so that at day 9 of treatment plasma osteocalcin was lowered to the same extent by 800,1400, and 2000 Ilg dosages (figure). Osteoporosis is a sequel to chronic systemic glucocorticoid therapy. This occurs primarily as a result of reduced bone formation which is reflected by reduced osteocalcin values in patients on such therapy.’ The finding of reduced osteocalcin in adults inhaling BDP 2000 ug daily is not too surprising because at such high dose systemic glucocorticoid effects can be expected. We have now shown that BDP even at a dose of 400 ug per day is also associated with reduced osteocalcin values. This raises the possibility that even
These include height, health behaviours, social circumstances, and the work environment. Unfortunately, explanations for the grade differences in health reported in the Whitehall cohorts continue to elude us for two reasons. First, the grade differences in behavioural, social, and work-related risk factors in Whitehall II were clearly correlated with each other and with some of the established risk factors. The researchers made no attempt to disentangle the complex interrelations that undoubtedly exist between these many variables.
explanations for these differentials.
Consequently it
was not
possible
to
examine the
independent
contribution of any particular factor, or combination of factors, to the health of civil servants. Second, whereas Whitehall I studied the relation between grade and mortality prospectively over ten years,’ Whitehall II was a cross-sectional investigation of health, grade, and associated risks. Although cross-sectional studies can describe the association between prevalent disease and potential risk factors, they cannot establish causality.2 Most people would agree with Marmot et al that attention should be paid to the social environment, job design, and consequences of income inequality. Whitehall II, however, has so far not been able to describe the direct, independent effect of these factors on the health of civil servants. Department of Public Health and Primary Care, Royal Free Hospital School of Medicine, London NW3 2PF, UK
Cardiovascular Research, University Laboratory of Physiology, Oxford 0X1 3PT, UK
Citations of
JONATHAN ELFORD
1. Marmot
MG, Shipley MJ, Rose G. Inequalities in death: specific explanations of a general pattern. Lancet 1984; i: 1003-06. 2. Hennekens CH, Buring JE. Epidemiology in medicine. Boston: Little Brown, 1987.
SiR,—The Whitehall II study of civil servants is depressing for the picture it presents of health inequalities, but for those interested in trying to do something about them it can be viewed optimistically as an opportunity to identify more precisely some of the possible causal mechanisms. Marmot et al state that "An analysis of the relation between speed of promotion and initial health status will be published elsewhere". It is not clear whether this refers to how quickly people have reached their present employment grade in relation to their current health status, or how quickly people will progress from now on, in relation to health status. With respect to causality, the second approach is likely to be more useful than the first. In the event that this second approach is the intention, I suggest that it might be useful to identify, within any particular grade, those who (a) actively seek promotion, and are successful; (b) actively seek promotion, and are not (initially, at least) successful; and (c) do not seek promotion. It is not only health status that might discriminate between these groups, but also the whole range of other influences that Marmot and his colleagues seek as possible explanations of the observed cross-sectional differences between grades (biological, behavioural, and psychological characteristics). A longitudinal follow-up of these subjects could be very enlightening. Department of Social Medicine, Medical School, Birmingham B15 2TJ, UK
some of which were incorporated in a revised version. We were aware that we were close to the work so when the revised paper arrived we sent it out of the country to referees 3 and 4. They reported, together, that the method should be independently tested but did not think it had been well done here. Wilmshurst was sent the report and we offered to look at a third try. A protracted correspondence followed, but no rewrite was submitted. This sort of thing, though rare, will happen on any journal. This is not really an instance of negative findings not being published but had much more do to with trying to change a paper from a polemic to a reasoned critique. The subject was and is controversial and all editors are happy to publish controversy; it brightens up the pages and gets people thinking. We would have been glad to carry the Wilmshurst reply and sorry that we never got to the point where we had a reasoned and reasonable bit of science to publish. It is a shame that he became so involved with accusations of bias that he apparently could not see what points were being made by our referees. It is probably true that negative result papers do not get accepted. Editors should watch out for this; we try to.
improvement,
T. MARSHALL
P. SLEIGHT D. BERGEL
published work
SiR,—Your April 27 editorial does not consider another possible for the fall in the citation impact and the rise in uncitedness of published articles from the UK during the past decade, despite a rise in the number of publications-unwillingness on the part of our American colleagues to cite papers published in non-American journals. This view is shared by several nonAmerican clinical scientists to whom I have spoken and also by some American clinicians (mostly non-faculty members). I wonder if Science Watch will be able to review its analysis to confirm or dispel the above possibility.
explanation
Department of Clinical Haematology, George Hospital, Kogarah, Sydney,
St
NSW 2217, Australia
A. MANOHARAN
Cyclosporin and autoimmune disease SIR,-As chairman of the Second Congress on Immunointervention in Autoimmune Diseases (CIAD ’91) recently held in Paris, I would like to let your readers know that your conference report (June 1, p 1338) covers only a small part of what was discussed over the four days of that meeting and focuses on cyclosporin trials. Three major conclusions could be drawn from the many presentations from scientists and clinicians working with cyclosporin. The efficacy of the drug in several diseases has been well established by randomised controlled trials comparing cyclosporin with placebo or standard therapy-eg, in severe psoriasis,t,2 uveitis,33 rheumatoid arthritis4 and nephrotic
syndromes Although side-effects, especially nephrotoxicity, are a Publication bias SIR,-We well remember the
many letters about the paper
submitted by Dr Wihnshurst (June 8, p 1419) late in 1987 and are happy to break cover. Wilmshurst’s paper was sent to us four months after the appearance of a study testing in man a method based on animal studies in which one of us (D. B.) had been involved. The author of that paper had been trained in Oxford and was interested in the method because he had been taught by its originator. Wilmshurst had looked at it for us and did not like it. We took a second opinion and published because we thought the idea needed looking at. The ensuing submission from Wilmshurst and his colleague did contain some interesting new data. Referee 1 would have liked to see it published but felt that many changes were needed. D. B. agreed, saying that the data were worth publishing "if they can do it properly" (ie, write it better). (The bit about the typewriter is clever.) Both gave detailed suggestions for
concern, the risk is very low when the recommended
daily dose of less than 5 mg/kg is used. No opportunistic infections are seen. Only exceptional cases of lymphoma (less than 1 per 1000) are observed, and they are usually reversible when the immunosuppression is stopped. New indications, based on controlled studies and not just anecdotal evidence, are emerging (eg, Crohn’s diseaseaplastic anaemia,? primary biliary cirrhosis,8 lichen planus,9 atopic dermatitis and allergic asthma [reported at the Congress])—though more data are needed to establish the efficacy and evaluate the risk/benefit ratio. Similar comments apply to other immunosuppressive agents, including FK506 in liver
transplantation. Other agents are needed to complement, and eventually replace, cyclosporin, with the twin aim of improving efficacy and reducing side-effects (both direct toxicity and over-immunosuppression). FK506 and the cyclosporin analogues rapamycin and myocophenolic acid are good candidates but randomised trials
60
comparing them with cyclosporin both for efficacy and side-effects (notably nephrotoxicity, which is not exclusive to cyclosporin) are needed before we can say whether their therapeutic indices are indeed better than that for cyclosporin. In vitro minimum active concentrations do not by themselves defme the therapeutic index. Furthermore
agents may still carry the risk of overand will not solve a major problem encountered with all these agents in most patients-namely, disease relapse after cessation of treatment. The answer may come from new approaches fully discussed at the Paris meeting-namely, monoclonal antibodies, autoantigen peptide therapy, T cell vaccination, and so on. Immunointervention is not just a matter of selecting the best immunosuppressive drug but searching for the best strategy that allows us to slow down or stop the autoimmune process in a non-toxic, disease-specific, and long-term fashion. new
immunosuppression
Department of Clinical Immunology, Hôpital Necker,
Research Unit for Genetic Disease
J.-F. BACH
75743 Paris, France
CN, Fradin MS, Messana JM, et al. Cyclosporine for plaque-type psoriasis: results of a multidose, double-blind trial. N Engl J Med 1991; 324: 277-84. 2. Mihatsch MJ, Woolff K, eds. Risk/benefit ratio of cyclosporin A in psoriasis. Br J Dermatol 1990; 122 (suppl 36). 3. Masuda K, Nakajima A, Urayama A, Nakae K, Kogure M, Inaba G. Double-masked trial of cyclosporin versus colchicine and long-term open study of cyclosporin in Behçet’s disease. Lancet 1989; i: 1093-96. 4. Tugwell P, Bombardier C, Gent M, et al. Low-dose cyclosporin versus placebo in patients with rheumatoid arthritis. Lancet 1990; 335: 1051-55. 5. Ponticelli C, ed. Cyclosporin A in nephroric syndrome. Clin Nephrol 1991; 35 (suppl 1). 6. Brynskov J, Freund L, Norby Rasmussen S, et al. A placebo-controlled, double-blind, randomized trial of cyclosporine therapy in active chronic Crohn’s disease. N Engl J Med 1989; 321: 845-50. 7. Frickhofen N, Kaltwasser JP, Schrezenmeier H, et al. Treatment of aplastic anemia with antilymphocyte globulin and methylprednisolone with or without cyclosporine. N Engl J Med 1991; 324: 1297-304. 8. Weisner RH, Ludwig J, Lindor KD, et al. A controlled trial of cyclosporine in the treatment of primary biliary cirrhosis. N Engl Med 1990; 322: 1419-24. J 9. Eisen D, Ellis CN, Duell EA, Griffiths CEM. Voorhees JJ. Effect of topical cyclosporine rinse on oral lichen planus. N Engl J Med 1990; 323: 290-94. 1. Ellis
Assessment of the mitochondrial chain
respiratory
SiR,—Trounce et all have suggested mitochondrial respiratory failure as a possible cause of ageing in man. However, the frequent scatter of control data and their overlap with those of patients limits the ability of clinicians to reach a reliable diagnosis of respiratory chain deficiencies (figure, left).2,3However, we have recently observed that a constant ratio of respiratory enzyme activities is a consistent feature of oxidative phosphorylation, irrespective of the tissue tested. This finding may be because a balanced proportion of each enzyme complex is required for optimum function of the respiratory chain. Our observation prompted us to present enzyme activities as ratios rather than absolute values. 1 0
—
COX
(nmo!/nmn
per mg
E
COX/SCCR
protein)
Mitochondrial enzyme activities ratios.
= patients; 0 =controls
We studied 7 patients (A-G) and 34 controls for cytochrome c oxidase (COX) activity, succinate-cytochrome c reductase (SCCR) activity, and COX/SCCR activity ratios in mitochondria isolated from skeletal muscle. While normal COX activities were scattered over almost two orders of magnitude, the COX/SCCR activity ratios showed a normal distribution with a narrow range of values (mean [SD] 29 [0’6]; range 2-4). Under these conditions, patients whose absolute COX values were in the low normal range could be unambiguously diagnosed as COX deficient (figure, right). Although this approach does not hold for generalised enzyme defects, it does improve the ability to recognise and quantify localised deficiencies in the mitochondrial respiratory chain.
as
ratio
both absolute values and
in Children, INSERM U-12, Hôpital des Enfants-Malades, 75743 Paris Cedex 15, France
PIERRE RUSTIN DOMINIQUE CHRETIEN THOMAS BOURGERON ANNE WUCHER
JEAN-MARIE SAUDUBRAY AGNES ROTIG ARNOLD MUNNICH
I, Byrne E, Marzuki S. Decline in skeletal muscle mitochondrial respiratory chain function: possible factor in ageing. Lancet 1989; ii: 637-39. 2. Moraes CT, Shanske S, Tritschler HJ, et al. mtDNA depletion with variable tissue expression: a novel genetic abnormality in mitochondrial diseases. Am J Hum Genet 1991; 48: 492-501. 3. Zheng X, Shoffner JM, Voljavec AS, Wallace DC. Evaluation of procedures for assaying oxidative phosphorylation enzyme activities in mitochondrial myopathy muscle biopsies. Biochem Biophys Acta 1990; 1019: 1-10. 1. Trounce
Inhaled corticosteroids, bone formation, and osteocalcin SIR,-Inhaled corticosteroids are now a part of the treatment of asthma. Low dosages of beclomethasone dipropionate (BDP), devoid of effects upon the hypothalamic-pituitary-adrenal axis, can control asthma.l,2 However, there is concern now that higher dosages are in use. A finding of lowered serum osteocalcin in individuals inhaling BDP 2000 ug daily3 suggests that high-dose therapy might be associated with adverse effects upon bone. As part of a larger study evaluating the systemic effects of inhaled BDP we measured osteocalcin across treatments using doses of BDP within the currently recommended therapeutic range (400-2000 Ilg daily). We studied 16 healthy adult volunteers, using a randomised double-blind, placebo-controlled crossover design to evaluate the systemic effects of 100, 200, 350, and 500 g BDP administered by standard metered-dose inhaler every 6 h for 10 days. Wash-out intervals of 18 days separated each treatment period. Plasma for osteocalcin measurement was collected on a fasting specimen taken within 30 min of 0900 h 4 days before starting treatment and repeated at the beginning of the fifth and ninth days of each treatment period. Plasma octeocalcin was measured by an in-house radioimmunoassay and each individual’s samples were assayed in the same batch. The difference from pre-treatment concentration at both days of testing during treatment was analysed by a three-way analysis of variance with factors subject, dose, and phase. The hypothesis of no-treatment effect was tested by a two sided t-test at each dose. During dosing with placebo plasma osteocalcin remained constant. During treatment with BDP there was a significant fall relative to placebo even at the lowest dose of 400 ug daily on treatment day 5. This effect was dose-dependent up to 1400 Ilg with a plateau thereafter (figure). With more prolonged treatment the plateau occurred at an even lower dose so that at day 9 of treatment plasma osteocalcin was lowered to the same extent by 800,1400, and 2000 Ilg dosages (figure). Osteoporosis is a sequel to chronic systemic glucocorticoid therapy. This occurs primarily as a result of reduced bone formation which is reflected by reduced osteocalcin values in patients on such therapy.’ The finding of reduced osteocalcin in adults inhaling BDP 2000 ug daily is not too surprising because at such high dose systemic glucocorticoid effects can be expected. We have now shown that BDP even at a dose of 400 ug per day is also associated with reduced osteocalcin values. This raises the possibility that even
