Document v6j2n7jadwO8gp2Qr735z3RGZ

FILE NAME: Talc (TALC) DATE: 1997 DOC#: TALC027 DOCUMENT DESCRIPTION: Journal Letter to the Editor RE Perineal Powder Exposure and the Risk of Ovarian Cancer American Journal of Epidemiology Copyright 1997 by The Johns Hopkins University School of Hygiene and Public Health All rights reserved LETTERS TO THE EDITOR Vol. 146, No. 9 Printed in U.S.A. The material on this page was copied from the collection of the National Library of Medicine by a third party and may be protected by U S. Copyright law R: "PERINEAL POWDER EXPOSURE AND THE RISK OF OVARIAN CANCER ` Cook et al. (1) suggest that their data support the hypoth esis that talcum powder use causes ovarian cancer. We ; believe that the current biologic and epidemiologic data do ; not provide strong evidence for a causal association. It is J unknown how talcum powder could migrate to the ovaries, " or whether the chemico-physical aspects of talcum powder i can induce ovarian cancer. The cited paper by Egli and l Newton (2) showed migration of carbon black particles | from vaginal deposition to the oviducts in two of three i patients. Different results were obtained by Deboer (3), who l found carbon black translocation from the vagina to the ? uterus in only 2 of 37 surgical patients. Based on a previ| ously published paper (4), the authors state that talc-like particles are found more frequently in ovarian tumors than ? in normal tissue. However, in that study, the amount of particles ranged from 6,900-55,100 /g wet tissue in tissues i from three healthy ovaries to 6,400-24,500 /g wet tissue . from three malignant ovaries (4). Further efforts are needed to determine whether and how talc-like particles can mi grate up the reproductive tract, to confirm findings of talc particles in healthy and malignant ovarian tissue and to explore possible mechanisms of talc-induced ovarian carci nogenesis. In epidemiologic studies of weak associations, dose , response trends are especially important to demonstrate causality (5). Cook et al. (1) found no trend in the odds - ratios with increasing number of perineal applications de li spite a fivefold difference between the lowest and highest , exposure categories. Other studies either did not determine the duration and frequency of perineal dusting (6-9), or *showed no trend in risk with duration of exposure (9, 10), or - a suggestive trend (11). The lack of a trend in several studies argues against a biologic effect. The modest positive find in g s may be due to greater awareness and recall among the cases. This seems possible given the numerous questions on talc use in these epidemiologic surveys and the considerable differences in exposure percentages between the control groups. Reliability studies on genital hygienic practices are needed. R eferences 1. Cook LS, Kamb ML, Weiss NS. Perineal powder exposure and the risk o f ovarian cancer. Am J Epidemiol 1997; 145: 4 5 9 --65. 2. Egli GE, Newton M. The transport of carbon panicles in the human female reproductive tract. Fertil Steril 1961;12:151-5. 3. Deboer CH. Transport o f particulate matter through the human female genital tract. J Reprod Fertil 1972;28:295-7. 4. Henderson WJ, Joslin CA, Turnbull AC, et al. Talc and carcinoma o f the ovary and cervix. J Obstet Gynaecol Br Commonw 1971;78:266-72. 5. Wynder EL. Guidelines to the epidemiology of weak associ ations. Prev Med 1987;16:211-12. 6. Harlow BL, Weiss NS. A case-control study of borderline ovarian tumors: the influence o f perineal exposure to talc. Am J Epidemiol 1989;130:390-4. 7. Chen Y, Wu PC, Ge WJ, et al. Risk factors for epithelial ovarian cancer in Beijing, China. Int J Epidemiol 1992-21: 23--9. ' 8. Booth M, Beral V, Smith P. Risk factors for ovarian cancer- a case-control study. Br J Cancer 1989;60:592-8. 9. Whittemore AS, Wu ML, Paffenbarger RS Jr, et al. Personal and environmental characteristics related to epithelial ovarian cancer. Am J Epidemiol 1988;128:1228-40. 10. Cramer DW, Welch WR, Scully RE, et al. Ovarian cancer and talc. Cancer 1982;50:372-6. 11. Harlow BL, Cramer DW, Bell DA. Perineal exposure to talc and ovarian cancer risk. Obstet Gynecol 1992;80:19-26. Joshua E. Muscat Ernst L. Wynder American Health Foundation 320 East 43rd Street New York, NY 10017 Editor's note: In accordance with Journal policy, Dr. Cook and her coauthors were given the opportunity to reply to the above letter, but they chose not to do so. RE: "ASSESSING THE DIRECTION OF CAUSALITY IN CROSS-SECTIONAL STUDIES" : We read with some concern the paper by Flanders et al. (1), in which they question the direct causality of the rela tion between dioxin exposure shown in studies (2, 3) that were carried out among veterans of Operation Ranch Hand who were participants in the ongoing Air Force Health Study (AFHS). As Flanders et al. correctly note in their first example, the exposed (E ) and control (C) group triglyceride means are nearly identical in the AFHS. Therefore, we believe that any model relating triglyceride and dioxin in the AFHS should accommodate equal triglyceride group means. Using the same notation used by Flanders et al., let Yi, Y2, and A denote the dioxin exposure level, the mea sured exposure level, and the triglyceride level, respec tively. Without loss of generality, we have dropped the subject subscript from the notations. Then, a model to study triglycerides should accommodate the condition P c 0 0 -- /r(X), (1) where p.c(X) and p.E(X), respectively, are the overall triglyc eride means for the control and exposed groups. Flanders et al. consider two models. Their model 1 is a direct causation model to study the effect of dioxin exposure on health, and model 2 is a reverse causation model to study the effect of triglycerides on measured dioxin. In both models, they assume that the exposed group is stratified by dioxin categories: j, low; E2, medium; and 3, high; and Mio < Mu < M 12 < Mi3, (2) 786 The material on this page was copied from the collection of the National Library of Medicine by a third party and may be protected by U S Copyright law. diabetologists and in major medical centers where funduscopic examina Estim ated Relative Risk of O varian Cancer, According to Reported Use of Talc tion is done routinely and competent ly. However, in the office of the primary care physicians, where most diabetics in this country receive much of their care, annual examination of the fundi through dilated pupils regrettably is performed infrequently if at all. Given that circumstance, an abnormal tourniquet test result de mands a competent funduscopic ex amination to rule out proliferative retinopathy, often by referral to an ophthalmologist. I wish to emphasize that I am not advocating that the tourniquet test replace regular fundu scopic examination. No talc mentioned Any talc mentioned No diaphragm used Diaphragm used, no talc Diaphragm, with talc No body talc Some body talc "All over" Genital Legs only Not genital Unknown where Caaes 62 67 92 14 25 77 54 37 7 1 6 3 Controls 61 100 118 11 41 84 78 67 3 0 8 10 *On genitals, sanitary napkins, or underwear. Estimated Relative Risk 1.0 0.7 1.0 1.8 0.8 1.0 0.8 0.7 2.5 0.8 0.3 85% Confidence Interval 0.4-1.1 0.7-3.7' 0.4-1.4 0.5-1.2 0.4-1.2 0.7-10.0 0.3-2.5 0.1-1.2 If Drs Aaby and Zegarra have a cost-effective strategy to ensure ade quate annual examination of the 11 million diabetics in the United States of epithelial ovarian cancer conducted from 1974 to 1977 in the Washington, "by a physician who can recognize early proliferative diabetic retinopa- . thy," I would happily endorse it and discard the tourniquet test; until then, the tourniquet test will identify nine of every ten patients with dia betic retinopathy who need to be referred to such a physician. Many of DC, area.2The cases were 197 women with pathologically confirmed pri mary epithelial ovarian cancers treated in participating hospitals. The controls were 197 women treated at the same hospitals for conditions other than gynecologic, psychiatric, or malignant diseases or pregnancy. these patients' conditions are cur rently undiagnosed until loss of vision occurs. Decrease in capillary fragility with improved diabetic control noted in several patients was not meant to imply regression of diabetic retinopa thy. Histological study, however, may confirm that the tourniquet test does accurately reflect the progression or The controls were frequency matched to the cases on age, race, and hospital. The interviewers asked questions about reproductive and sexual histo ry, medical history, drug use, and other exposures. Questions about talc use were added to the questionnaire after the study began, so 135 cases and 171 controls were asked about talc exposure. regression of diabetic dermal micro The reported talc use among cases angiopathy. At present, the vascular and controls is given in the Table. We or platelet abnormality causing capil estimated the relative risk to talc lary fragility in diabetes is unknown. users as 0.7 (95% confidence interval I am currently involved in a study [CI]=0.4 to 1.1). The estimate was correlating the tourniquet test with fluorescein retinal angiography in unaffected by adjustment for race, age, and gravidity. Neither women I those patients who do not have identi fiable diabetic retinopathy on oph thalmoscopic examination. who used talc on their diaphragms nor those who used it as body powder seemed to be at excess risk. Women William A. Reynolds, MO who used talc as a body powder were W estern M ontana Clinic M issoula 1. Cartwright GE: Diagnostic Laboratory Hematology, ed 4. New York, Grue & Stratton Inc, 1968, p 367. 2. Stobbe H, Rirup C: Zum Nachweis von Kapillar schaden im Rahmen der Mikroangiopathie-Diagnostik beim Diabetes mellitus mittels Strauversuchs. Schweiz Med Wockenschr 1979;109:1808-1810. f' 3. Rodriguez R, Root HF: Capillary fragility and _ diabetic retinitis. N Engl J Med 1948;238:391-397. asked how they used it. Among the ten who specifically mentioned use on sanitary napkins, underwear, or the genital area, the relative risk was estimated as 2.5, but the small num ber of exposed women yielded an unreliable estimate (95% CI=0.7 to 10.0). Our data thus indicate no overall Chance, bias in selection or obser vation, or confounding may have influenced these estimates. One im portant potential bias to consider in this and Cramer's study is a differ ence between cases and controls in recollecting or reporting talcum pow der use, especially in the genital area. Talc exposure was not a major focus of this study, and few data are avail able to assess the likelihood of recall bias. Such a bias could stem from cases' heightened awareness or from the fact that controls were inter viewed in the hospital while most cases were interviewed at home. On the other hand, the questions about talc use were rather simple and unambiguous. Also, we noted that cases and controls were equally likely to report douching. Since reporting of use of douches might be subject to the same recall biases as talc use, this observation suggests that little recall bias operated. Another possible inter pretation of our findings of no appar ent effect of using talc on the dia phragm but some effect of perineal use of powder is that talc itself does not increase risk of ovarian cancer but that patients with ovarian cancer have or perceive a greater need for using body powder in the genital area, for reasons related either to the biolo gy of the disease or to life-style. We agree with Cramer and co-workers that other epidemiologic data will be useful. Patricia Hartge, M S c Robert Hoover. M D Talc and Ovarian Cancer To the Editor.--Cramer and co-work ers1 recently reported observing an * association between talc use and risk of ovarian cancer. We therefore examined data on talc use that two of us (L.M. and L.P.L.) had collected as part of a case-control interview study association between talc use and risk of ovarian cancer. Although a small group of women who specifically reported genital use of body talcum powders showed an excess relative risk, use of talc on a diaphragm, which would be the closest exposure to the ovaries, did not seem to elevate risk. N ational C a n c e r Institute Bethesda, Md Ljnoa P. U sher, M P H L arry Mc Gow an, M D G eo rg e W ashington University Medical Center W ashington, DC 1. Cramer DW, Welch WR, Scully RE, et ai: Ovarian cancer and talc. Cancer 1982;50:372-376. 2. McGowan L, Parent L, Lednar W, et al: The woman a t risk for developing ovarian cancer. Gynecol Oncol 1979;7:325-344. 1844 JAMA, Oct 14, 1983-- Voi 250, No. 14 Letters wsxmsMmm, BW M S Br in duly a expen.` provet in you youm part of Guard earthc a you'll i c h a lle N ew o tunitie inform tunitie Guard Natior The