Document qaVmo9NGqzkMDR6qpoVOMbVJq

^used after 19*3- ln P*ricular, the incidence of cornu:^recorded for 1965 was more lhan double the 1963 figure. I fcad difficulty in grasping how such evidence indicates an as^iuod between removal of adjuvant and disappearance 0{ |(nwi, reactions. Even Dr Tim, whom he cites, seems far from iat*- MTflKnl of Baeteriology ind Virology, r^*yofMxadioier, gjaSaKr Mil 7v1 lyOEL VI1. PRESTOS Ste*ai l rcsjptv, my >umni4;v t> hum air iniir.*: ndiviJu*. 'f. '*** tf adtcrw :hcd thjt C; tho-* group I' children vjuim ihe -- * lA ny pjp<; my inn ns tit da I indk'llt ly diup to o; ho is uhisfc Sn,--The meriti of whooping-cough vaccination will nr. er tr settled by a scries of letters in The Lancet, but I feel neter(ieins that I must correct two of the statements made by Dr jjgtsdi and Professor Stewart in their letter of May 15 p. Jn the first place, Sako did publish clinical observations on nednated and control subjects, taking careful note of nprure: 30 cases in 159 exposed vaccinated vs. 137 cases tarth 13 deaths) in 149 exposed unvacrinated infants or young (tUren. This is detailed in the reference cited, and further &uils can be found in a later paper.1 But others did far more artfully designed studies than Sako, and obtained similar ^wtrs; the work of Silverthome, Miller, Kendrick, Singerhooks Coppolino, Bell, and others is liberally cued in the Venture and I will not take up space by detailed references hrt. One cannot, however, overlook the masterful studies by he British Medical Research Council,1 and the fact is that, with three exceptions, every properly controlled study on hooping-cough vaccine has shown it to be moderately to aghly effective- It is, of course, possible to prepare a poor vacOOC and, conceivably, a good vaedne that is irrelevant because ddifferences in serotype. But these problems have beer, ccrnr1") in the U.K., and Dr Fraser's analysis May 1, p. 969 ygott that, with a little more attention to the younger hsu, whooping-cough vaccine in Glasgow might prove sausfactory to ail. My second point concerns alum. In the first place us use has x "long been discontinued" as a component of vaccines; c tac four U.S. manufacturers still use it and it is included in me of the most widely used forms of diphthtria-tetanus-perMos vaccines. However, its use is diminishing since it is tougbt to be less effective as an adjuvant than the other salts dslummium. Yet this very fact renders Sako's results more, Wher than lets, impressive. .jBtoiikRoid, Geoffrey Edsall nc s strar-Jt idequjtf re n ihe iO lor.ftr I *!" is x, ihcts a? I hetc Jo txT iuIi) idei)u auk) rS nod *h inert** tdju'M* I. |V'I - 9. i>M? GENETIC RISKS OF VINYL CHLORIDE . ifStar--Dr Paddle (May 15, p, 1079) asks us to specify the (. --hod of data collection used in our study (April 13, p. 734, --so show a tabulation of the data before age adjustment. i - AAsilmentioned several times, the data were obtained by inierf wi ith the workers. The range for response-rates, which similar for the study and control groups, were also in. --fed. As we stated, the questions about pregnancy outcome ^ --t contained in a much larger interview questionnaire which -- the initial item of a cross-sectional health survey that int jadsd a physical examination, X-rays, and laboratory tests, jft'wht results of the survey demonstrated very few significant dif- and no consistent bias toward a higher prevalence for (indices measured in either the study or control group. The ytriew protocols were administered by six interviewers (two ive team* of three interviewers) from the Center fer Control, each with experience in field health surveys I"**0! interview took 15-30 min, and was ctrried out in a ' ****** v*n interview booth. The interviewers had participated * a formal review of the questionnaire before the study. Ir. **"Uon> a physician member of the survey team renewed r; responses immediately after the interview and, where y^pnate, sought further elaboration from the interviewee. 3H****'. J. PtSmt. 1947, JO, 29. Research Council Br. mod,J. 1959, i 994, FATEXHAt. AOt DISTRIBUTION FOR FETAL DEATHS ACCORDINO TO HUSBAND'S V.C. EXPOSURE Paternal p (yr) "Control*'* Pnmt-y exposure Pregnancies Fetal death* Pregnancies Fetal death* Btjort exposure: <20 20-24 25-29 50-34 >55 All age*, crude rate Mean paternal age at conception Ago-adjusted rate* 31 80 38 6 4 159 230yr. 2 `i-S't) 4 '$ a%) 4 fiQ`5<i) 0 11 '4-9SJ . . -6-9%) 70 44 2 (4-SV 56 7 (12SV 27 5 (18 SV 14 1 148 15 (10-lV 26-4yr. (61V After exposure: <20 20-24 25"29 30" 34 >35 All age*, crude rate Mean paternal age at conception Age-odiu*ted rate* 1 43 87 87 55 273 30-4 yr. 0 4 '9-J5J 3 7 S 0%) 10 fli-2%) 24 '6-8%) 0 22 3 (136V 48 U (22'Fkj 36 3 fS-.HU 33 6 (18-2(7; 139 23 (16-SV 302 yr. . (IS-8V * Fetal mortality-fate* for pnm*ry v,c, tipL.'e group are dina age adjusted to the paternal age distribution of the pregnancies m the control group. The table shows data for age-specific fetal mortality-rates before age adjustment, for both time periods (i.c , before and after the husband's exposure). Before the husband's exposure, the crude fetal mortalityrates are 10-1% for the study group and 6-9% for the control group. However, after direct age adjustment the study-group rate became 6-1%. This is the result of a younger paternal age for pregnancies in the control group. For example, pregnancies to- wives of men less than 30 years of age made up 93-7% (149/159) of the control-group pregnancies as compared to 72-3% (107/184) of the study-group pregnancies. Our paper indicated this difference in age between the groups for before exposure comparisons. Table i in that paper showed that the mean paternal ages were 23-0 years for the control group and 26-4 years for the study group. As shown here in the table, subsequent to husband's expo sure, the crude fetal mortality-rates were 8-8% and 16-5%, for the control and study groups, respectively. With direct age adjustment the 16-5% was reduced to 15-8%. Age adjustment for the subsequent to exposure comparisons resulted in little change in the rates because the age distributions for pregnan cies in both the study and control groups after exposure were similar. For example, pregnancies to wives of men younger than 30 years of age made up 48-0% (131/273) of the controlgroup pregnancies and 50-4% (70/139) of the study-group pregnancies. We hope this explanation of the age-adjustment procedure is now clear. It is the fact that the two standard age distribu tions derive from the control group s populations of pregnan cies, rather than of persons (whether fathers or not) that prob ably accounts for the numerical behaviour that Dr Paddle finds "misleading". In the subsequent to husband's exposure comparisons, the significant difference in fetal mortality-rates was a result of pregnancy outcome associated with husbands younger than 30 years of age. This difference was significant at p<0 001 (yM0-52, D.p.wl), If vinyl chloride (v.c.) is indeed related to this observed difference, the excessive fetal mortality at these age, could, in theory, be a reflection of placing newly hired personnel, with little or no seniority, in exposure cate gories that had relatively worse environmental exposures. In our paper of April 3 we cited eight references which have demonstrated that v.c. had elicited a positive response in mi crobial test systems and had been associated with significant excesses of chromosomal aberrations in lymphocytes of workers occupationally exposed to it. Three additional studies which alio have demonstrated a positive mutagenic response to Y_ v-' zL 2; TJ. z UCC - 110913 1290 THE LANCET, JUNE 12, 1976 v.c., have been drawn to our attention recently; in two of these tudies, the microbial test system was used,1 1 while in the third, Tradescantiii was used.1 2Th3erefore, we do not agree with Dr Paddle's statement that this is an "isolated" question. Divifion ot Surveillance, Hazard Evaluations and field Studies, National Institute for Occupational Safety and Healthy Cincinnati, Ohio 43226, L'.S.A, and Bureau of Epidemiology, Center for Disease Control and School of Public Health, University of North Carolina Peter F. Infante Joseph K. Wagoner Anthony J. McMichael Richard J. W.axweiler Henry Falk PLASMA-GLUCAGON AFTER PANCREATECTOMY Sir,--Dr Barnes and Dr Bloom (Jan. 31, p. 129 reported no circulating immunoreactive glucagon (l.R.G.) in both the basal state and after arginine stimulation in five pancreatectomised patients. Dr Gerich and his colleagues (April 17, p. 855) have made similar observations in another patient. We have evaluated i.R.g. dynamics in a totally pancreatectomised patient, with results quite different from those cited above. Bccauac of the heterogeneity of i.r.g. species measured by anti serum 30K, we essay before and after acetone extraction. The post-extraction assay measures predominantly i.r.g. of <9000 daltons (s-i.r.g.), and subtracting s-i.r.g. from total plasma-I.R.o. (unextracted) allows quantification of the larger molecular weight t.R.G. species (l-i.r.g.). A 30-year-old male, member of a multiple endocrine aden oma type i family, underwent a total pancreatectomy, hetnigastrectomy, and duodcnectomy in 1967 for removal of an in sulinoma. He has remained free of disease subsequently. Further evidence of total pancreatectomy was afforded by essentially unmeasurable basal levels of C peptide with failure to rise during an arginine infusion (by A. Rubcnstein, Univer sity of Chicago). His basal, fasting concentrations of s-i.r.g. and L-I.R.O., before an arginine infusion, 24 h after receiving his standard insulin dose (21 units lente and 10 units semilente) were 35 pg/ml and 104 pg/ml, respectively. These values are within normal limits for our laboratory. The accompany ing table depicts the plasma levels of small and large molecular weight i.r.g., glucose, and C peptide during the administration of arginine (500 mg/kg/30 min). The arginine infusion pro duced a small but definite rise in s-I.r.O. with no change in l-i.r.g. In addition, a 50 mg/dl increase in plasma-glucose con centration occurred concurrently with the increase in s-i.r.g. Several factors may be involved in the difference between our observations and those of Dr Barnes and Dr Bloom. Unger et al., using antiserum 30K, failed to demonstrate i.r.g. in the plasma of pdncrtatectomised dogs;4 s* u6b*s8eq9u1e0ntly several inves tigators*-* have observed circulating i.r.g. (30K) in pancreatcctomised dogs (and other species) when they are deprived of insulin for several days, demonstrating the sensitivity of extrapancreatic alpha cells to inhibition by insulin. Because of ethi cal considerations achieving this degree of insulin deficiency may not be appropriate in man, but our patient may have been more insulin deficient than those previously reported. Although alpha cells indistinguishable from pancreatic alpha cells have been reported in the stomach of man,* earlier work by Muller et al.18 failed to demonstrate a rise in i.r.g. (30K) with arginine infusion in two pancreatectomiscd patients. However, these patients did have significant plasma-i.R.G. 1. Garro, A. J., Guuenplan, J. B., Milvy, P. Mutation Ret. 1976, 38, SI. 2. Andrew,, A. W,, Zawistowiki, E. S>, Valentine, C. R. ibid. the press 3. Sparrow, A. Pen4na! communication. 4- Unger, R. H,, Ohneda, A., Aguilar-Parada, E., Eitcntratit. A. J din in ten, 1969, 41,810. 5- Vranic, M , Pck, S., Kawamsri, R. Ditbetu, 1974,2), 903, 6. Mauuyama, T., Foa, P. Proe. Soe. tip. Biol. Med. 1974,177, 4". l.Meihiter, K,, Harding, P., Chou, M., Maehiter, G., Stout, J,. Diamond. D , Field. J. Endocrinologt. 1973. 9*. 678. 8. Dobbs, It., Sakurai, H, Fatoone, G., Valveide, I., Baetens, D., Orci, L., Unger, R. Seunte, 1973, 387, 344. 9. Usiger, R. Linear, 1975, i, 1036. 10.Muller, W,, Brennan, M, Tan, M,, Aoki, T. Diabetes, 1974,23, 312. GLUCOSE, l.R.G., AND C PEPTIDE CONCENTRATIONS DURING ARGINini 1NPCSION Time 'min) Control 10 20 23 30 35 45 60 90 Glucose mg d! 224 23S 242 233 265 27 2 274 i'i.a.0. (pg/ml) 35 57 74 91 77 68 51 L't.KG. (pg/mJ) 104 103 101 99 98 92 99 C-pepnd* (ag/ml) 0-7 0*7 0-7 0-7 levels in the basal state. Plasma contains several different mol ecular weight species that are measured as i.r.g. (30K).1' 11 Therefore, it it possible that the antiserum used by Dr Barnes and Dr Bloom does not react with all the i.r.g, plasma com ponents measured in assays using antiserum 30K. In addition, the partial gastrectomy and duodcnectomy that is usually done in conjunction with a total pancreatectomy in man may result in at least partial removal of the source of extrapancreatic glu cagon and may vary in different patients depending upon the extent of the surgery. While our patient is potentially unique in being a member of a family with multiple endocrine neoplasia, we feel that on the basis of his results we can only conclude that; (1) in this pancreatectomiscd man, i.r.g. (30K) of both large and small molecular weight(s) circulates in the basal state, and that at least the small-molecular-weight fraction shows a rise with arginine; (2) the increase in plasma-glucose that accompanied the s-i.r.g. rise suggest that this s-i.r.g. is biologically active; and (3) the existence of an extrapancreatic source of biolog ically active glucagon in man has not been excluded. Department of Medicine, Diabetei Center, University of Washington, Seattle, Washington 98144. L'.S.A. Jerry P. Palmer Phillip L. Werner James W. Benson John W. Ensinck Sir,--We read with interest the article by Dr Bloom and Dr Barnes (Jan. 31, p. 219) on the absence of pancreatic gluca gon-like immunoreactivity (i.r.g.) in the plasma of panoreateclomised man. The data generated diametrically opposed airs on the diabetogenic role of pancreatic glucagon by Dr Gerich and his colleagues on the one hand (April 17, p. 855) and Dr Donowiti and Dr Felig on the other (April 17, p. 855). We wish to add information which may help to resolve these con tradictory views. In contrast to the findings of Bloom and Barnes, we11 found high levels of "pancreatic" I.R.G. in the plasma of a pancreatectomised patient, as did Muller et al.14 in two other patients. Although SO'c of the pancreatic i.r.g. could be accounted for by factors in plasma which interfere with the assay1* the rest appeared immunometrically as pancreatic i.r.g. and presum ably derived from an extrapancreatic source. In contrast to i.r.g. in porcine duodenal extracts,14 which closely resembles true pancreatic I.R.G., the i.r.g. in pancreatectomiscd human plasma is dissimilar in physicochemical properties (gel electro phoresis), is not stimulated by arginine, insulin hypoglycemia, or tolbutamide, or suppressed by glucose, and has no apparent role in blood-glucose regulation.'* We have previously17 referred to the various forms of gastrointestinal pancreatic i.r.g. found 11 Valverdc, I., Dobbs, it , Unger, R. Mttabolum, 1973, 24, 1021. 12. Valverde, I., Villanueva, M., Lozano, I., Marco. J. J. din. Endoee. Metab 1974,19, 1090. 13. Holha, J. I... Vinik, A. I., Child, P., Jackson, W. p U. Unpublished. M. Muller, W. A , Brennan, M. F., Tan, M. H , Aoki, T. T. Ditbetet. 1974, 23. 312. 13. Weir, G. C, Turner. R C,, Martin, D. B. Horn. net. Ret. 1973, 3, 241 16. Sasaki, H., Rubalea'a, B,, Bacieas, D., Blasquea, E., Snkant, C. B-, Orel- L. Unger. R. H J. dm. Invest. 1973, 56, 133. 17. Vinik, A. t.s Jackson. W, P, U\ Lancet, 1975, i, 694. UCC - 110914