Document BykJ7zdLGmMVZmXrw1oxXQrBL

Vol. 324 No. 3 CORRESPONDENCE 195 evidence whatever of an aggregate decrease in health care costs for the elderly as a result of research. The evidence, I believe, points in exactly the opposite direction. The elderly have a powerful moral claim on health care, but not an unlimited claim. A primary goal of the health care system should be to help everyone have the chance to avoid a premature death and become an elderly person. That is a reasonable good. By contrast, an unlimited effort to extend the life of the elderly, regardless of cost and regardless of the burdens it can impose on the young, is not. Briarcliff Manor. NY 10510 Daniel Callahan, Ph.D. The Hastings Center 1 de Lissovoy G. Medicare and heart transplants: will lightning strike twice? Health Aff (Millwood) 1988; 7{4):61-72. 2. Health Care Financing Administration. Special report: findings from the Na tional Kidney Dialysis and Kidney Transplantation Study. Baltimore: Depart ment of Health and Human Sep/ices, 1987. 3. Latta V. Keene RE. Leading inpatient surgical procedures for aged Medicare beneficiaries. Health Care Financ Rev 1989; 11:99-110. 4. Hosking MP. Warner MA, LobdelJ CM, Offord KP, Melton LJ 111. Out comes of surgery in patients 90 vears of aee and older. JAMA 1989; 261:1909-15. 5. Office of Technology Assessment Task Force. Life-sustaining technologies and the elderly. Philadelphia: Science Information Resource Center, 1988: 400-18. The above letters were referred to the authors of the article and editorial in question, who offer the following replies: To the Editor: In his book Setting Limits' Daniel Callahan states that "severely ill, mentally alert patients who have lived out a full life span should receive care that is limited, as a rule, to general nursing care." He explicitly indicates that not only intensive care and advanced life support, but also general medical care such as antibiotic treatment, should be withheld from such patients. In the first point in his letter, Callahan appears to recommend only the limitation of "expensive high-technology medical care," whereas in his second point he returns to the broader prohibition in his book by recommending the limitation of "life-extending" care for the elder ly. The use of penicillin is potentially life-extending for an older patient. I disagree with Callahan's contention that prohibiting the treatment of elderly patients with antibiotics or other general medi cal care is not a change in present policy. As Callahan himself discusses in his book, the problem for those who wish to reduce health care costs by limiting care for the aged is that high expenditures for the elderly are due largely to the cost of regular medical care for chronic and severe acute illnesses. Only a small proportion of the high cost of medical care in the last year of elderly patients' lives, to which Dr. Hadorn refers, is for high-cost intensive care.2-3 Contrary to Dr. Hadom's statement, not only Cal lahan but also other thoughtful analysts have suggested limiting routine care for the elderly. Indeed, they must eliminate more than intensive care if they wish to reduce national health costs substan tially by rationing health care to the elderly. I do not share Dr. Callahan's pessimism about the economic effect of medical research. I see no reason why the major chronic degenerative diseases, such as osteoarthritis, osteoporosis, Parkin son's disease, and Alzheimer's disease, cannot be controlled or pre vented. Decreases in dementia due to Alzheimer's and in hip frac tures caused by osteoporosis will reduce national health care expenditures more than the ethically distasteful alternative of with holding beneficial care from the elderly. Beyond economics, success in such research will relieve suffering and improve the quality of life, goals on which we can all agree, however we feel about the "risk" chat life may be extended as well. In his editorial in the issue in which my article appeared, Dr. Reiman stated that I "take no stand on rationing in general." Al though my article focused on age as a criterion, I did state that "I am not persuaded that explicit rationing is necessary in the United States at this time." Dr. Reiman himself, in two more recent editori als,4^ has cogently indicated why we should rationalize the U.S. health care system before we consider rationing beneficial care to anyone. Boston, MA 02118 Norman G. Levcnsky, M.D. Boston University Medical Center 1. Callahan D. Setting limits: medical goals in an aging society. New York: Simon and Schuster, 1987. 2. Lubitz J. Pnhoda R. The use and costs of Medicare services in the last 2 years of life. Health Care Financ Rev 1984; 5(3).117-31. 3 Scitovsky AA. "The high cost of dying": what do the data show? Milbank Mem Fund Q Health Soc 1984 ; 62:591-608. 4. Reiman AS. The trouble with rationing. N Engl J Med 1990; 323:91 1-3. 5. Idem. Reforming the health care system. N Engl J Med 1990; 323:991-2. Editor's reply: As I have argued in two recent editorials,1-21 do not believe that rationing of beneficial services will be necessary in the United States if we develop a more rational and efficient health care system. Without substantial reform of our present system, rationing would be difficult to implement and, aside from its ethical problems, would probably not save any money in the aggregate. Dr. Stoll's conviction that rationing is inevitable reflects the view from the United Kingdom, where there are centralized budgetary controls that limit total expenditures on health care to barely more than half the U.S. level. But here in the United States we have an open-ended health care system, based largely on fee-for-service and private practice; even if patients and physicians were prepared to accept it, I do not see how rationing could work very well under these conditions. Dr. Harvey is quite right about the unequal effects of the pro posed Oregon plan to ration health care to the poor. In my view, the proposal's chief virtue lies in the raising of public consciousness. Proponents have yet to come up with an acceptable and workable plan, which says something about the difficulties of explicitly ra tioning services in this country at this time. Arnold S. Relman, M.D 1. Reiman AS. The trouble with rationing. N Engl J Med 1990; 323:911-3 2. Idem. Reforming the health care system. N Engl J Med 1990; 323:991-2. <' ASBESTOS-RELATED DISEASES To the Editor: Aware that more than 100,000 Americans will die of asbestos-related disease,1-2 the public is understandably suspicious when we conclude that an occupational or environmental exposure is not the cause of a particular ailment. In assessing the importance of potentially hazardous exposures, full and painstaking disclosure is necessary. It was with much disappointment, then, that we read the Journal review article on asbestos-related diseases by Mossman and Gee.3 Mossman and Gee have emasculated much of the evidence that is at odds with their own position. In fact, the authors so emphatically articulate a legal position on nearly every issue chat one questions just what purpose their article is to serve; contrary to their stated purpose, we find little effort "to elucidate the mechanisms of asbes tos-related disease." Journal readers seeking to balance their perspective are encour aged to look elsewhere for information on the relative pathogenicity of chrysotile asbestos4,5 (and volume 14 of the American Journal of Industrial Medicine); the genetic and epigenetic changes involved in asbestos-induced lung cancer6; the linkage between exposure to as bestos and cancer of the gastrointestinal tract,7 larynx,8-9 and kid ney10-1 2; pathological criteria13-14 and radiographic sensitivity15 in diagnosing asbestosis; and the likelihood that smoking habits are confounding any of the identified relations between asbestos and disease.16,17 Mossman and Gee, in discussing the radiographic progression of patients with asbestosis, refer only to an abstract18 that reports on persons exposed to asbestos who did not necessarily have asbestosis. This is peculiar given the availability of published studies address- LAM 019157 DPMC-12730 SH-647 196 THE NEW ENGLAND JOURNAL OF MEDICINE Jan. 17. 1991 ing this issue.10'21 In a similar vein, the authors support a thesis with the findings of an epidemiologic study,22 and then go on to report that the period of follow-up in the study was too short to demon strate anything. Of importance to clinicians is the error the authors perpetuate in claiming that benign asbestos-related pleural effu sions usually occur in the first 20 years after the initial exposure to asbestos. Hillerdal,23 in the only population-based study of this condition, reports a mean latency of 30 years and the frequent finding of bloody pleural fluid. In a field that is rife with scientific controversy, medicolegal com bat, and intense public concern, a review article should identify controversial issues, objectively describe opposing positions, and cite relevant evidence. The article by Mossman and Gee fails to meet that standard. We believe that readers of the Journal should be aware of the large community of physicians and scientists who strongly disagree with many of the authors' conclusions. This letter, unanimously endorsed by all member clinics of the Association of Occupational and Environmental Clinics, evidences the depth of that sentiment. (The Association of Occupational and Environmen tal Clinics is an organization of 29 occupational health clinics based at or formally affiliated with schools of medicine or public health. The programs embody a deeply held commitment to preventive medicine and encompass a full range of clinical services, as well as research, physician training, and community education.) Pawtucket, RI 02860 David G. Kern, M.D. Memorial Hospital of Rhode Island Boston, MA 02138 David C. Christian!, M.D. Karl T. Kelsey, M.D. Howard Hu, M.D. Harvard School of Public Health Philadelphia, PA 19104 Howard Frumkin, M.D. University of Pennsylvania Denver, CO 80206 Kathleen Kreiss, M.D. Cecile Rose, M.D. Lee S. Newman, M.D. Joseph Jarvis, M.D. National Jewish Center for Immunology and Respiratory Medicine Ann Arbor, MI 48109 David Garabrant, M.D. Thomas G. Robins, M.D. Alfred Franzblau, M.D. University of Michigan School of Public Health Piscataw-ay, NJ 08854 Howard M. Kipen, M.D. Michael Gochfeld, M.D. University of Medicine and Dentistry of New Jersey Seattle, WA 98104 Linda Rosenstock, M.D. Scott Barnhart, M.D. Harborview Medical Center and Representatives of the 22 Other Association of Occupational and Environmental Clinics Member Clinics 1. Nicholson WJ. Perkel G, Selikoff IJ. Occupational exposure io asbestos: population at nsk and projected mortality -- 1980-2030. Am J Ind Med 1982; 3:259-311. 2 Walker AM. Loughhn jE. Friedlander ER, Rothman KJ, Dreyer NA. Pro jections of asbestos-related disease 1980-2009. J Occup Med 1983; 25:40925. 3. Mossman BT. Gee JBL. Asbestos-related diseases. N Engl J Med 1989; 320'1721-30. 4. Churg A. Green FHY. eds. Pathology of occupational lung disease. New York: Igaku-Shoin. (9S8 223-4, 253. 279-82. 5 Cullen MR. Controversies in asbestos-related lung cancer. Occup Med 1987. 2.259-72. 6. Barrett JC, ed. Mechanisms of environmental carcinogenesis. Vol. 1. Role of genetic and epigenetic changes. Boca Raton. Fla.: CRC Press, 1987 7. Frumkin H. Berlin J. Asbestos exposure and gastrointestinal malignancy, review and meta-analysis. Am J Ind Med 1988; 14:79-95 [Erratum, Am J Ind Med 1988; 14:493.] 8. Doll R, Peto J. Other asbestos-related neoplasms. In: Amman K, Aisner J. eds. Asbestos-related malignancy. Orlando. Fla.: Grune &. Stratton. 1987:81-96. 9. Smith AH. Handley MA. Wood R. Epidemiological evidence indicates asbestos causes laryngeal cancer. J Occup Med 1990: 32.499-507. 10. Smith AH, Sheam VI, Wood R. Asbestos and kidney cancer the evidence supports a causal association. Am J Ind Med 1989; 16:159-66. 11. Maclure M, Poole C. Asbestos and kidney cancer Am J Ind Med 1990; 17.647-8 12. Enterline PE, Henderson V. Asbestos and kidney cancer. Am J Ind Med 1990; 17.645-6. 13. Wamock ML. Wolery G. Asbestos bodies or fibers and the diagnosis of asbestosis. Environ Res 1987; 44:29-44. 14. Roggli VL. Pratt PC. Number of asbestos bodies on iron-stained tissue sections in relation to asbestos body counts in lung (issue digests. Hum Pathol 1983; 14:355-61. 15. Kipen HM. Lihs R, Suzuki Y, Valciukas JA. Selikoff IJ. Pulmonary fibro sis in asbestos insulation workers with lung cancer: a radiological and histopathological evaluation. Br J Ind Med 1987; 44:96-100. 16. Axelson O. Confounding from smoking in occupational epidemiology. 8r J Ind Med 1989; 46:505-7. 17. Blair A. Steenland K, eds. Smoking and occupation in epidemiologic stud ies. Am J Ind Med 1988; 13:3-190. 18. Gaensler EA, Jederlimc PJ, McLoud TC. Progression of asbestosis. Chest 1987:91:305 abstract. 19. Suoranta H, Huuskonen MS, Zitting A, Juntunen J. Radiographic progres sion of asbestosis. Am J Ind Med 1982; 3:67-74. 20. Beny G. Mortality of workers certified by pneumoconiosis medical panels as having asbestosis. Br J Ind Med 1981: 38.130-7. 21. Cookson W. De Klerk N, Musk AW, Glancy JJ, Armstrong B, Hobbs M. The natural history of asbestosis in former crocidolite workers of Wirtenoom Gorge. Am Rev Respir Dis 1986; 133:994-8. 22. Hodgson JT. Jones RD. Mortality of asbestos workers in England and Wales 1971-81. Br J Ind Med 1986; 43:158-64. 23. Hillerdal G. Non-mahgnant asbestos pleural disease. Thorax 1981; 36:66975. The above letter was referred to the authors of the article in question, who offer the following reply: To the Editor: Kern and associates criticize our account of laryn geal cancer even though we cited opposing opinions, including one mentioned in their letter. Worse still, while citing one new paper supporting their views on asbestos and laryngeal cancer,1 they ig nore two other recent contrary opinions2"3' and the rejection of this association by the British Advisory Council.4 *Furthermore, the pa per they cite1 uses enhanced standard mortality ratios for lung can cer as a surrogate for measures of exposure to asbestos -- a ques tionable approach. Are we biased when we cite the position of the American Thoracic Society on benign asbestos-related disorders together with "a response thereto"? In criticizing our account of the potential progression c f asbestosis, they ignore an ent.-f paragraph in our review describing the classic clinical course. They also criti cize us on the issue of radiologic sensitivity in diagnosing asbestosis, citing one of the signatories' papers.1 We quoted this paper. We also discussed the issue in a paragraph indicating that pathologic proc esses can precede radiologic manifestations. Surely the issue in 1991 is the role of high-resolution CT scanning in the diagnosis of lung fibrosis. They also criticize our comments on the confounding effects of smoking on standard mortality ratios for lung and laryngeal cancer,, citing Axelson,6 who in fact estimates confounding errors from smoking in standard mortality ratios of up to 1.6 -- close to our earlier suggested value of about 2.0 in a paper coauthored by an ex president of the Association of Occupational and Environmental Clinics.7 An Austrian study of cement workers41 also indicates that "deaths from lung cancer in asbestos workers were higher (SN1R [standard mortality ratio], 1.7} but after adjustment for age and sex specific smoking habits, this was not significant (SMR, 1.04)." They use outdated risk estimates of deplorable mortality among asbestos workers and ignore the seven more recent estimates of risk discussed LAM 019158 DPMC-12761 Vol. 324 No. 3 CORRESPONDENCE 197 in our review. However, the most serious omission is their failure to cite three recent international scientific symposia, all of which re port mesothelioma to be due largely to amphiboles rather than chrysotile.9'11 As recently as September 1990, an editorial espoused the same views.12 Further accounts of the importance of amphiboles can be found in studies of occupational and paraoccupational cases of mesothelioma.13 This view was also espoused by Dr. Churg, whose textbook they cite.'4 Before publication of our article, some sections on the mecha nisms of asbestos-related disease were deleted because of space limitations. We therefore refer readers to our recent article15 and a book16 on current research developments in this field. It is unfortunate that labor unions and the plaintiff bar so often depend on the advice of members of the Association of Occupation al and Environmental Clinics, which we believe is frequently neither correct nor contemporary. Society deserves better. Burlington, VT 05405 New Haven, CT 06510 Brooke T. Mossman, Ph.D. University of Vermont J. Bernard L. Gee, M.D. Yale University School of Medicine 1. Smith AH, Handley MA. Wood R. Epidemiological evidence indicates asbestos causes laryngeal cancer. J Occup Med 1990; 32:499-507. 2. Edelman DA. Laryngeal cancer and occupational exposure to asbestos. Int Arch Occup Environ Health 1989; 61:223-7. 3. Liddell FDK. Laryngeal cancer and asbestos. Br J Ind Med 1990: 47:28991. 4. Industrial Injuries Advisory Council. Report on cancer of the larynx. Lon don: Her Majesty's Stationery Office. 1989. 5. Kipen HM, Lilis R. Suzuki Y, Valciukas JA, Selikoff U. Pulmonary fibro sis in asbestos insulation workers with lung cancer, a radiological and histo pathological evaluation. Br J lnd Med 1987; 44:96-100. 6. Axelson O. Confounding from smoking in occupational epidemiology. Br J lnd Med 1989; 46:505-7. 7. Welch L, Hertz J, Cullen MR. Gee JBL. Current pulmonology. In: Sim mons DA, ed. Occupational lung disease. Chicago: Year Book Medical, 1985:155. 8. Neuberger M. Kundi M. Individual asbestos exposure: smoking and mortal ity -- a cohort study in the asbestos cement industry. Br J lnd Med 1990; 47:615-20. 9. Doll R. Mineral fibres in the nonoccupational environment: concluding remarks. In: Bignon J, Peto J, Saracci R, eds. Non-occupational exposure to mineral fibres. Lyon. France: International Agency for Research on Cancer, 1989. (1ARC scientific publications no. 90.) 10. Spengler JD, Ozkaynak H, McCarthy JF, Lee H. Summary of symposium on health aspects of exposure to asbestos in buildings. Cambridge, Mass.: Energy and Environmental Policy Center. Harvard University. 1989. 11. World Health Organization report on occupational exposure limit for asbes tos, Oxford, United Kingdom, April 10-11, 1989. Geneva: World Health Organization. 1989. 12. Gibbs AR. Role of asbestos and other fibres in the development of diffuse malignant mesothelioma. Thorax 1990; 45:649-54. 13. Gibbs AR, Griffiths DM. Pooley FC, Jones JSP. Comparison of fibre types and size distributions in lung tissues of paraoccupational and occupational cases of malignant mesothelioma. Br J Occup Med 1990; 4:621-6. 14. Churg A, Green FHY, eds. Pathology of occupational lung disease. Tokyo: [gaku-Shoin, 1988. 15. Mossman BT, Bignon J, Com M. Seaton A, Gee JBL. Asbestos: scientific developments and implications for public policy. Science 1990; 247:294301. 16. Mossman BT, Begin RO. eds. Effects of mineral dusts on cells. Vol. 30 of NATO ASI senes H: cell biology. Berlin, Germany: Springer-Verlag, 1989. LACK. OF EFFICACY OF HYDERGINE IN ALZHEIMER'S DISEASE To the Editor: Thompson et al. (Aug. 16 issue)1 suggest that Hvdergine (Sandoz brand of ergoloid mesylates) is ineffective in treat ing Alzheimer's disease and may in fact "cause cognitive dysfunc tion, perhaps through a direct toxic effect or by accelerating the progression of Alzheimer's disease." Such speculation seems unsup ported by the data presented and may cause unwarranted concern among patients, their families, and their physicians. Hydergine has been studied in dozens of clinical trials during the past 20 years,2'7 and although evidence of efficacy is equivocal, no suggestion of adverse cognitive effects has emerged. The specula tion about such effects by Thompson et al. is not supported by their findings. The authors report that the patients treated with Hyder gine for 24 weeks had greater deterioration on a standard test of psychomotor speed, the Wechsler Adult Intelligence Scale (WAIS) Digit Symbol Substitution Task, than did the patients given pla cebo. One must question whether the test was properly adminis tered. The mean base-line scores for both groups would place the study subjects high among normal subjects of the same age, sug gesting the test was improperly administered or has little relevance in Alzheimer's disease. Beyond that, the scores of the drug-treated group actually improved between weeks 12 and 24 of the study, making it hard to understand how Hydergine could have toxic ef fects or speed the deterioration due to Alzheimer's disease. The results of the only test of cognitive function, the Wechsler Memory Scale, showed no change with Hydergine treatment. Analysis of the data is open to question. In view of the base-line difference between the drug and placebo groups on the WAIS Digit Symbol Substitution Task and the different results reported for this test and for the Geriatric Evaluation by Relatives Rating Instru ment at 12 and 24 weeks, the appropriate analysis of treatment effects would seem to be a repeated-measures analysis of covar iance, with base-line performance entered as a covariate. The most disturbing aspect of this paper is that a study of few patients for a short time with questionably analyzed data and mys terious base-line scores on the only cognitive test to show a change resulted in headlines in the popular press and unsupported specula tion about the toxic effects of a widely prescribed drug. If Hyder gine is useless or toxic, we need to know about it. but this study has not answered those questions. Philadelphia, PA 19129 Joel D. Posner, M.D. Lisa Landsberc, M.Ed. Medical College of Pennsylvania 1. Thompson TL II, Filley CM, Mitchell WD, Culig KM, LoVerde M, Byyny RL. Lack of efficacy of Hydergine in patients with Alzheimer's disease. N Engl J Med 1990; 323:445-8. 2. Rouy JM, Douillon AM, Compan AM. Wolmark Y Ergoloid mesylates (`Hydergine') in the treatment of mental deterioration in the elderly: a 6-momh double-blind, placebo-controlled trial. Curr Med Res Opin 1989; 11:380-9. 3. Thienhaus OJ, Wheeler BG, Simon S, Zelman FP. Hartford JT. A controlled double-blind study of high-dose dihydroergotoxine mesylate (Hydergine) in mild dementia. J Am Geriatr Soc 1987; 35:219-23. 4. Chierichetti S. Cucinotta D, Santini V. Implications of long-term study of Hydergine in elderly patients with chronic senile cerebral insufficiency. Biomed Issues 1985; 1:377-93. 5. La2zari R. Franzese A, Chienchetu S. et al. Multicenter double-blind pla cebo-controlled long-term clinical trial of Hydergine in chronic senile cere bral insufficiency: an interim report. Ageing 1983; 23:347. 6. Loew DM, Weil C. Hydergine in senile mental impairment. Gerontology 1982; 28:54-74. 7. McDonald RJ. Hydergine: a review of 26 clinical studies. Pharmakopsychiatr Neuropsychopharmakol 1979. 12.407-22. The above letter was referred to the authors of the article in question, who offer the following reply: To the Editor: We welcome the opportunity to respond to Dr. Posner and Ms. Landsberg. First, the writers take issue with the speculation that Hydergine mav have a detrimental effect in Alz heimer's disease. They base their criticism on the fact that the base line scores on the WAIS Digit Symbol Substitution Task were quite high for patients with Alzheimer's disease, and that the treated group had a lower mean score at 12 than at 24 weeks. The scores were indeed high, and in reviewing our methods, we have found that we inadvertently allowed the patients 3 minutes to complete the task instead of the standard 90 seconds. Although this variation does preclude comparison with other studies using the WAIS Digit Sym bol Substitution Task, the internal validity of our study is fully LAM 019159 DPMC-12732 198 THE NEW ENGLAND JOURNAL OF MEDICINE Jan. 17,1991 intact, and the detrimental effect of the drug is still evident. With regard to the lower mean score at 12 than at 24 weeks, this difference {-9.96 vs. -8.85) was not significant. The data certainly support our speculation about a possible harmful effect; unexpected results are often critically important in science, and we would be remiss in ignoring them. Readers will note that we drew no conclusion re garding this issue. Next, on the question of whether a repeated-measures analysis of variance would have been a more appropriate statistical test, we find no compelling reason to agree. We think that the test we used -- a one-way analysis of covariance -- is appropriate and simple and that the more complicated repeated-measures analysis is un necessary. Finally, why the dissatisfaction with the appearance of reports of our study in the popular media? Our study was not, as compared with most earlier studies of Hydergine, a small one with a "few patients," nor was it conducted "for a short time." We appreciate the careful detection of what may have been confusing the Digit Symbol Substitution Task base-line scores, but still find the more important change in scores valid. The central conclusion, that there was no hint of a beneficial effect of Hydergine in these patients with Alzheimer's disease, remains unchallenged. Information of this sort is understandably sought by a nation increasingly preoccupied with the problems of Alzheimer's disease and other dementias. We did not solicit media attention, but when approached, we thought it was appropriate to describe our findings in a reasonable manner. It is our belief that we did so. Denver, CO 80262 Christopher M. Filley, M.D. Wayne D. Mitchell, Ph.D. University of Colorado School of Medicine Philadelphia, PA 19107 Troy L. Thompson II, M.D. Jefferson Medical College Denver, CO 80262 Kathleen M. Culig, B.S.N., A.N.P. Richard L. Byyny, M.D. Mary LoVerde. M.S.N., A.N.P. University of Colorado School of Medicine TRANSPLACENTAL PASSAGE OF INSULIN To the Editor: Menon et al. (Aug. 2 issue)1 make the interesting observation that animal insulin, mainly in the form of insulin-anti body complexes, can be detected in the cord serum of some infants born to mothers with insulin-dependent diabetes mellitus treated with animal insulin. The authors conclude that the higher rate of macrosomia in infants of such mothers is partly due to the stimula tory effect of the transplacentally transferred insulin.' We disagree with this conclusion. If the animal insulin transferred from the mother to the fetus contributed to fetal macrosomia, one would expect a lower rate of macrosomia in the 17 infants of the series studied by Menon et al. in whom no animal insulin could be detected in the cord serum. In this group, however, there was an even higher rate of macrosomia (in 9 of 17 vs. 12 of 28 infants). We do not disagree that transferred animal insulin could be a growth factor in the fetus. Its effect might be balanced, however, through the inhibition of insulin secretion in the fetus, which supposedly would occur if fetal blood glucose levels were lowered by transferred insulin. What could cause the significant correlation between transferred animal insulin and the macrosomia rate? Potentially, transferred maternal antibodies could neutralize fetal insulin, causing fetal hyperglycemia and thus beta-cell hyperplasia. The results of meas urements of C peptide in cord serum do not support this possibility, however.2 The macrosomia rate of 47 percent in the population studied by the authors is higher than the rate of 10 to 25 percent in recently published studies and in our own population with insulin-depend ent diabetes mellitusJ ; and unpublished data). One co^. - hypothe size that the higher rate of macrosomia in the patients of Menon et al. was due to somewhat poorer glycemic control in some patients treated in the early 1980s, when, for example, blood glucose moni toring at home was not widespread. Poor glycemic control might then coincide with higher doses of animal insulin, because the use of highly immunogenic insulins was more common. It would be inter esting if the authors could stratify their patients according to the year in which they were pregnant. Their method of assessing glycemic control -- i.e., measurement of glycosylated hemoglobin with the minicolumn assay -- may not be well suited to detecting small differences in glycemia within the nearly normoglycemic range.4 Thus, although the authors found no difference in glycemic control as determined by the percentage of glycosylated hemoglobin, there could have been higher blood glu cose levels in the mothers of the infants with macrosomia. We agree with the authors that immunogenic insulins should not be used in pregnant women. The influence of insulin antibodies on insulin pharmacokinetics and their possible role in prolonging fetal hypoglycemia are reasons enough not to use them.5 The use of pure insulins will not reduce rates of macrosomia in diabetic women, however, unless metabolic control as close as possible to that of pregnant nondiabetic women is maintained throughout pregnancy.3 4000 Dusseldorf, Germany Renate Kimmerle, M.D. Ernst A. Chantelau, M.D. Abteilung fur Ernahrung und StofTwechsei Heinrich-Heine-Universitat 1. Menon RK, Cohen RM, Sperling MA, Cutfield WS, Mimouni F, Khoury JC. Transplacental passage of insulin in pregnant women with insulin-dependent diabetes mellitus: its role in fetal macrosomia. N Engl J Med 1990; 323:30915. 2. Mylvaganam R. Stowers JM. Steel JM, Wallace J, MacHendry JC, Wnght AD. Insulin immunogenicity in pregnancy, maternal and fetal studies. Diabetologia 1983; 24:19-25. 3. Gabbe SG, Quilligan EJ. General obstetric management of the diabetic preg nancy. Clin Obstet Gynecol 1981; 24:91-105. 4. Kaplan LA, Cline D. Gartside P, Burstein S, Sperling M, Stem EA. Hemo globin Al in hemolysates from healthy and insulin-dependent diabetic chil dren. as determined wuh a temperature-controlled mini-column assay Clin Chem 1982: 28:13-8. 5. Knip M. Lautala P, Leppaluoto J, Akerblom HK, Kouvalainen K. Relation of entroinsular hormones at birth to macrosomia and neonatal hypoglycemia in infants of diabetic mothers. J Pediatr 1983; 103:603-11. To the Editor: In their report on exogenous insulin found in the cord serum of neonates born to mothers with insulin-dependent diabetes mellitus, Menon et al. do not discuss one very interesting finding. In the infants without exogenous insulin in cord serum, the mean level of endogenous insulin was only 16 percent of that in the infants with exogenous insulin in cord serum. Furthermore, the level of endogenous insulin in the infants without exogenous insulin was only 39 percent of that in the subgroup of infants without macroso mia who had detectable exogenous insulin. Since birth weight corre lated with levels of endogenous insulin in ail the infants, it is possi ble that the remarkable increase in endogenous insulin levels contributed to the promotion of growth in addition to the exogenous insulin. Indeed, the mean excess of exogenous insulin in cord serum in the group with macrosomia, as compared with the group without macrosomia, was only 711 pmol per liter (1113 -- 402), whereas the corresponding excess of endogenous insulin was 1818 pmol per liter (2726 - 908). We conjecture that the increase in the level of endogenous fetal insulin associated with maternal anti-insulin antibody, more than transplacentally transferred exogenous insulin, underlies the fetal macrosomia. Tel Hashomer. Israel 52621 Izhar Ben-Shlomo, M.D. Jehoshla Dor. M.D. Sheba Medical Center Safed, Israel 13100 Shifra Zohar, M.D. Rebecca SiefT Hospital Tel Hashomer. Israel 52621 Shlomo Mashiach, M.D. Sheba Medical Center DPMC-127? LAM 019160