National Academies Press: OpenBook
« Previous: ANALYSIS OF THE SEX RATIO DATA
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 99
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 100
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 101
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 102
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 103
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 104
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 105
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 106
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 107
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 108
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 109
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 110
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 111
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 112
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 113
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 114
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 115
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 116
Suggested Citation:"ANALYSIS OF THE MALFORMATION DATA." National Research Council. 1956. Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki. Washington, DC: The National Academies Press. doi: 10.17226/18776.
×
Page 117

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Chapter Vlll ANALYSIS OF THE MALFORMATION DATA DATA on major malformations were collected at the examination shortly following birth, and again at the 9-months examination. For the reasons advanced in Section 2.6, it was deemed advisable to consider the results of these ex- aminations separately. Moreover, to permit pool- ing the information from the two examinations, if such pooling was warranted, all infants found on the "at-birth" examination to have one or more major malformations were excluded from consideration in the analysis of the 9-months data. 8.1 The trait. — In any consideration of the relation of parental irradiation history to the occurrence of major congenital malformations in the offspring, it is of course first necessary to define major congenital malformation. Since there is no absolute dividing line between "ma- jor" and "minor" malformation, an attempt to formulate an a priori definition encounters cer- tain difficulties. In actual practice, the examining physicians were required to record all abnormali- ties, down to and including such variations as minor haemangiomas (excluding the very com- mon occipital-nuchal type) and auricular pits. When the final report on each infant was coded, a decision was made as to the presence of major malformation. Any one type of malformation was consistently coded as either major or minor. Table 8.1 is a listing of the malformations which in the course of this investigation were coded as major. A discussion of the relative fre- quencies of the various malformations, and the various combinations in which they occurred, will be presented elsewhere. For our present purposes, then, we shall define major malforma- tion simply as any condition found on that list. Obviously there are many other major congenital malformations which do not appear on that list, simply because they either did not occur in this series, or, if they did occur, were not recognized at birth. This listing deals solely with conditions recognizable at birth by the usual physical examination. A later chapter will deal with autopsy findings. 8.2 Reliability of diagnosis. — A question which arose repeatedly during the course of this study — and which will undoubtedly occur to the reader — concerned the completeness with which the significant malformations oc- curring in the populations under study were recognized and reported. This question has already been commented on briefly (Sec. 2.6). It should at the outset be made quite clear that no study organized as this one was can be expected to yield an accounting of all signifi- cant congenital defects present in the newborn population. Thus, the diagnosis of mental de- fect, or severe visual or hearing loss, is notori- ously unreliable during the first several months of life. From the standpoint of the present study, while for obvious reasons one should strive for as high a level of diagnostic accuracy as possible, the primary danger to be guarded against is that different diagnostic standards obtain for the various segments of the popula- tion under study. There is no reason to believe that diagnostic standards were not uniform with respect to infants in one exposure subclass as compared with infants in another. We have al- ready commented briefly on some of the evi- dence suggesting that no bias has been intro- duced by differences in the degree to which various groups cooperated in the study (Sec. 5.7). It is important to reiterate that each infant recorded in this study, with the exception of a certain number of stillbirths or neonatal deaths, as recorded in Section 2.5, was examined by a Japanese physician. It was recognized from the outset that both the loyalties and training of the midwife were such that she might at times fail to record the obvious. This suspicion was well justified by instances in which a child with 99

100 Chapter VIII Genetic Effects of Atomic Bombs harelip has been reported as normal by the at- tending midwife. Beginning in early 1950, the mothers of all living infants found to have either a major or a minor malformation in the course of the home visit were invited to bring marily occasioned by the disposal of the body of an infant alleged to have a major defect before examination by an ABCC physician, every mal- formation recognized at birth was verified be- fore the defect was accepted. Prior to early TABLE 8.1 AN ALPHABETICAL LISTING OF THOSE MALFORMATIONS OBSERVED IN THIS STUDY WHICH OCCURRING ALONE OR IN COMBINATION WITH ONE ANOTHER WERE GRADED AS MAJOR CONGENITAL DEFECT Achondroplasia Albinism Amputation, congenital Anencephaly Anonychia Anophthalmos Arthrogryposis Atresia ani or rectum Atresia of the external nares B lepharophimosis Brachydactyly Cataracts Central nervous system defect, severe Claw hand Cleft palate Club foot Coloboma iiMis Corneal defect Cryptophthalmia Cystic hygroma Diaphragm, defect of Diastasis recti, severe Dislocations, multiple, congenital Dysplasia of acetabulum, congenital Ectodermal defect, congenital Ehlers-Danlos syndrome Encephalocele Exostosis, severe, of bone External ear, major malformation of Facial cleft, oblique Flexion deformity of fingers and toes Foetus amorphous Foetus papyraceous Funnel chest Gastroschisis Harelip Heart disease, congenital, cyanotic Hemimelus Hydrocephalus their children to an ABCC "Verification Clinic" at some convenient time shortly after the diag- nosis. Transportation to and from the clinic was supplied by the ABCC. There the child was seen by another Japanese physician and an American physician. Thus, with exceptions pri- Hypertelorism Hypospadias, marked Ichthyosis congenita Inguinal hernia (females) Intestinal obstruction, ? congenital defect Lymphangioma Macroglossia Major bone, absence of Malformation, complex and ill-defined Microcephaly Micropenis Microphthalmus Microtia Mongolism Neurofibromatosis (von Recklinghausen's dis- ease) Nystagmus, congenital Oligodactyly Omphalocele Polydactyly Polyostotic fibrous dysplasia Polyoty Ptosis of eyelids Pupillary membrane, persistent Radio-ulnar synostosis Recto-vaginal fistula Rhinencephaly Situs inversus viscerum Spina binda with or without meningocele or myelomeningocele Status Bonnevie-Ullrich Sympodia Syndactyly Synorchism Teratoma Thyroglossal duct, persistent Tumor, type undetermined 1951, major malformations were usually veri- fied by a second home visit on the part of either one of the more experienced Japanese physicians or an American pediatrician. An analysis of the results of these home visits and Verification Clinic indicate a high degree of accuracy of

Analysis of the Malformation Data 101 diagnosis for those defects recognized as present. The crucial question is, how many defects which should have been diagnosed at birth went our control material with roughly comparable material collected by other investigators. The most extensive and reliable Japanese series to come to our attention is that compiled by Dr. TABLE 8.2 THE TYPES AND FREQUENCY OF VARIOUS MAJOR CONGENITAL MALFORMATIONS OBSERVED AT THE TOKYO RED CROSS MATERNITY HOSPITAL DURING THE YEARS 1922 THROUGH 1940 (The over-all frequency is 456 malformations among 49,645 births [0.92%]. The table is based on data made available by Drs. S. Mitani and V. Kuji.) No. of Description of abnormality cases Acrania or hemicrania (anencephaly) .... 32 Hydrocephalus 14 Congenital ascites 2 Spina bifida with club foot 3 Sacral teratoma 3 Umbilical cord hernia (omphalocele) .... 9 Spina bifida 8 Club foot (T. varus, T. equino-varus) . . 63 Harelip with cleft palate 53 Polydactyly (fingers) 38 Cleft palate 30 Club foot (T. valgus) 18 Hemangioma or lymphangioma 16 Congenital teeth 13 Malformation of ear lobe 13 Micromelus 11 Atresia ani et vaginalis (including imper- forate anus) 10 Syndactyly of toes 10 Microphthalmia (and others) 7 Luxation of knee joint 6 Polydactyly (toes) 6 Microcephaly 4 Pseudohermaphroditism 4 Hydrocephalus, cleft palate, gastroschisis 1 Atresia ani, umbilical hernia 1 Atresia ani, malformation of external au- ditory canal 1 Atresia ani, cleft palate (mild degree?).. 1 Atresia ani, sympodia 1 Atresia ani, cleft palate 1 Cleft palate, oblique facial cleft 1 Cleft palate, malformation of fingers 1 Cleft palate, malformation of ear 1 Harelip, pseudohermaphroditism 1 Harelip, oblique facial cleft, absent nose, syndactyly (fingers) 1 Harelip, polydactyly (fingers), Syndactyly (toes) 1 Fissured palate, club foot 1 Fissured palate, microphthalmia 1 Club hand, club foot, small thumb, miss- ing right labium 1 Elbow contraction, club foot, luxation of knee joint 1 unrecognized? There are three general ap- proaches to the evaluation of this question which we shall now consider. (A) The first approach involves a compari- son of the frequency of major malformation in No. of Description of abnormality cases Inguinal hernia, micromelus, hydrocepha- lus 1 Gigantism of feet and arms 1 (fin- Polydactyly (toes), harelip Polydactyly (fingers), syndactyly gers) 1 Syndactyly (fingers), harelip 1 Torticollis, abdominal hernia 1 Torticollis, club foot, facial asymmetry... 1 Monophthalmia, polydactyly (toes) 1 Monophthalmia, polydactyly (fingers) ... 1 Hydrocele 4 Cleft hand (lobster claw) 4 Polydactyly (toes) with syndactyly (toes) 4 Ankylo-glossum (tongue-tie) 3 Club foot (T. calcaneus) 3 Abnormal elbow joint 3 Malformation of external auditory canal.. 3 Amelus (arm or arms missing) 3 Inguinal and scrotal hernia 2 Meningocele (cranium) 2 Club hand 2 Brachydactyly (2 fingers) 2 Contraction of the knee joint 2 Polydactyly (toes) with syndactyly (fin- gers) 2 Malformation of fingers 2 Absence of nose 1 Club foot (T. equinus).. Parasitic teratoma of jaw. Tumor of "waist" Bifid uvula Hypospadias or epispadias Cleft sternum (lower portion) 1 Phimosis (extreme) 1 White hair (partial albinism) 1 Tumor (small) of vagina 1 Anencephaly, polydactyly (fingers) . Club foot, syndactyly (toes) Club foot, malformation of fingers.. Malformation of face Monster (amorphous) Polydactyly (fingers and toes) Missing fingers, Syndactyly (toes) . .. V. Kuji (personal communication) at the Tokyo Red Cross Maternity Hospital. The findings are shown in Table 8.2. The over-all frequency of malformed individuals in this series (0.92 per cent) and in the children of the non-irradiated

102 Chapter VIII Genetic Effects of Atomic Bombs parents included in this study (0.85 per cent) is quite comparable. The agreement is all the more striking in view of two facts: (1) certain of the "malformations" on the Kuji list, such as congenital teeth, hydrocoele, and partial albinism, would be considered minor in our terminology, and (2) a hospital series in Japan cannot be considered random, but probably in- cludes a disproportionate representation of the complications of pregnancy. In view of the tendency of gross malformations to be associated with such complications as hydramnios (Prindle, Ingalls, and Kirkwood, 1955), this should bias the Red Cross Hospital series in an upward di- rection. There is no evidence from this compari- son that any considerable number of congenital defects diagnosable at birth are being over- looked in Hiroshima and Nagasaki. (B) A second approach to the question of missed diagnoses of significant defect consists of checking the ABCC findings against those of another medical group in Hiroshima or Nagasaki. Such a spot check was carried out for the last three months of 1951 in Nagasaki. Out of twelve major "malformations" seen at the two largest hospitals in the city (the Nagasaki Medical School Hospital and the Mitsubishi Hospital), five were already known to the ABCC. The remaining seven were the follow- ing: congenital dislocation of the hip — 3 cases; congenital heart disease— 2 cases; idiocy with funnel chest — 1 case; hydrocephalus — 1 case. In view of the frequent delayed appear- ance of congenital dislocation of the hip and hydrocephalus, these cases are not thought to constitute "missed" diagnoses. As noted earlier (Sec. 2.6) congenital heart disease was not in- cluded in the "at-birth" series of the ABCC. Only the funnel chest perhaps should have been detected in the ABCC series. Again, then, there is no evidence that any substantial proportion of major defect diagnosable at birth by the techniques of physical examination was missed. (C) A third approach to the question of accuracy of diagnosis comes from a comparison of the "at-birth" with the "9-months" series. In Section 2.6 we have recorded the results of a "spot check" in Hiroshima in 1952 concerning the clerical and medical problems which can arise in a program concerned with congenital malformations such as this, and the steps taken to meet some of these problems. Every effort was made to insure that these errors did not recur in the remainder of the study. It might be noted, however, that, as can be seen in Sec- tion 2.6, most of the errors were ones which would overestimate the number of those diag- noses which were made at birth and not subse- quently verified at nine months. We may now compare the final figures for major malforma- tions present in those infants examined at age 9 months with the findings recorded for these same infants at birth. Such a comparison, repro- duced in Table 8.3, provides the best means available for reaching an opinion concerning the confidence to be placed in the "at-birth" data on congenital malformation. Perusal of the table brings out a number of points of interest. Out of a total of 563 infants with one or more major defects diagnosed either at the examination shortly after birth or at the ex- amination at an age of approximately 9 months, or on both occasions, only 112 were clearly verifications at age 9 months of an "at-birth" diagnosis. However, there were altogether 130 infants with one or more major defects in whom the diagnosis could not be verified at 9 months because of the death of the child in question; it seems likely that the great majority of these children did indeed suffer from a major mal- formation. The 2 cases of defect at birth which were not verified at age 9 months were both cases of polydactyly which we assume to have undergone surgery. Of the total number of de- fective infants detected by these two approaches, then, the figures taken at face value indicate that only 43 per cent of them were observed at birth. However, it is clear that in many of the 319 infants with one or more defects not diagnosed shortly following birth, the defects are of such a nature that their certain detection at that time would have been difficult if not impossible. This is especially true of the two diagnoses which contribute most to this list (dysplasia of the acetabulum — 133 cases; and inguinal hernia in females— 56 cases; total — 189 cases, 33.6 per cent of the grand total of 563 infants or 59.3 per cent of the additional infants diagnosed at 9 months). The relatively high frequency of acetabular dysplasia may re- flect a particular interest in this condition on the part of several of the American pediatri- cians associated with the project. We have at- tempted to indicate in the table by a single asterisk those diagnoses which should in all probability have been made at the time of the

Analysis of the Malformation Data 103 TABLE 8.3 A COMPARISON, FOR HIROSHIMA AND NAGASAKI, OF THE FINDINGS AS REGARDS MAJOR CONGENITAL MALFORMATION IN INFANTS EXAMINED AT APPROXIMATELY AGE 9 MONTHS, AND IN THOSE SAME INFANTS WHEN EXAMINED SHORTLY AFTER BIRTH (The figures are obtained from 14,768 infants in Hiroshima and 12,324 infants in Nagasaki. Infants born of consanguineous marriages have not been excluded from this comparison. Since no diagnoses of congenital heart disease were accepted on the "at-birth" examination, this diagnosis does not appear in the comparison. There were 50 such diagnoses made in Hiroshima in the "9-months" examination and 55 in Nagasaki. Ex- planation of asterisks in text.) Hire- Comparison shima Nagasaki Total 1. Total number of infants in whom a diagnosis of one or more major malformations was made on the "at-birth" and the "9-months" examination 57 55 112 2. Total number of infants in whom a diagnosis of one or more major malformations was made on the "at-birth" examination, but who could not be re-examined at age 9 months because of the death of the child 74 56 130 3. Total number of infants in whom a diagnosis of one or more major malformations was made in the "at-birth" examination, but in whom no major defect was detected at age 9 months.. . 2 0 2 4. Total number of infants in whom no diagnosis of major mal- formation was made at birth, but in whom one or more major defects were detected in the "9-months" examination * 165 154 319 Achondroplasia * 1 2 3 Anterior synechiae 0 1 1 Branchial cyst 0 1 1 Cataracts** 1 0 1 Club foot** 224 Coloboma of eyelids * 1 1 2 Coloboma of iris * 2 0 2 Cyst, thyroglossal 1 0 1 Complex and ill-defined condition * 0 1 1 Cryptorchism 0 2 2 Deformity of bone 0 1 1 Dextrocardia * 0 1 1 Dysplasia of acetabulum 76 57 133 Ectodermal defect, congenital 0 1 1 Funnel chest ** 8 5 13 Hammer toe 0 1 1 Harelip and cleft palate * 1 2 3 Hernia, inguinal, in females 35 21 56 Hydrocephalus 4 1 5 Hypertelorism ** 0 1 1 Hypoplasia of bone 0 1 1 Ichthyosis (congenital type)** 3 0 3 Laryngeal stridor, congenital 0 1 1 Lymphangioma 0 1 1 Maldevelopment of central nervous system 0 1 1 Mental defect, severe 2 18 20 Microcephaly 4 2 6 Micrognathia * 0 1 1 Micropenis * 1 0 1 Microphthalmus * 1 0 1 Microtia * 0 2 2 Mongolism** 8 7 15 Nystagmus, congenital 0 6 6 Polydactyly * 1 0 1 Ptosis 1 2 3

104 Chapter VIII Genetic Effects of Atomic Bombs TABLE 83—Continued Hiro- Comparison shima Nagasaki Total Situs inversus viscerum 1 3 4 Spastic infantile paralysis 4 0 4 Spina bifida * 1 0 1 Strabismus 5 3 8 Syndactyly * 1 1 2 Teeth, malformation of 0 1 1 Tumor, cutaneous 0 1 1 Wilm's tumor 0 1 1 Xeroderma pigmentosum 0 1 1 165 154 319 563 *A defective infant, regardless of number of major defects, is entered under Item 4 in the table only once. In case two or more major defects were detected for the first time at the 9-months examination, only the most severe of the defects is listed under Item 4. "at-birth" examination, and by a double aster- isk those diagnoses which as a rule are more difficult to make in newborn infants but which might well have been made at that time. (How- ever, each infant was entered into Table 8.3 but once, and the numbers entered opposite specific diagnoses under item 4 refer to the number of infants in whom that diagnosis was considered the most severe one). There are 35 diagnoses in the first category, and 22 in the second. All of the remaining defects may either not have been diagnosable at birth or are of such a nature that uncertainty might surround them. In Section 2.6 the possibility was raised that such relatively common diagnoses as acetabular dysplasia and pilonidal sinus "unbalance" the 9-months data. These are both findings which border on being normal variations, in their less extreme forms. Ideally, in an analysis of this type one would introduce a weighting constant to take into account the genetic component in the etiology of the various malformations under study, but in view of the obvious impossibility of this, there seemed no alternative to giving each diagnosis, including the common ones, equal weight. Despite the many observations which have been published on congenital malformation in man, it is a surprising fact that to our knowl- edge there exists only one series comparable to this one, in terms of re-examination at some later date of a series of infants examined shortly after birth. Mclntosh et al. (1954), in a study of 5,739 products of conception weighing over 500 grams who were followed, unless death intervened, until 12 months of age, and some even longer, found that only 43.2 per cent of all malformations detected in the course of the study were observable at birth. Although for a number of reasons, to be dealt with elsewhere, these figures are not directly comparable to our own, the general agreement between the two series is striking. Table 8.3 also serves another purpose, in that it permits a partial comparison of the diagnoses being made in Hiroshima and Nagasaki. With one possible exception, the findings in the two cities appear as uniform as might be hoped for under these circumstances. It will be noted that severe mental defect is listed 18 times for Naga- saki, but only twice for Hiroshima. It seems quite likely that this reflects a difference in ap- proach in the two cities, rather than a real difference. 8.3 The genetic argument for radiation-in- duced changes. — Within the spectrum of mu- tations induced by the irradiation of experi- mental animals, the most dramatic, but least common, group of mutations are the so-called "visibles." These mutations produce, as the name implies, visible alterations in the pheno- type. Although direct evidence is lacking, it seems probable that in man a certain fraction of these mutations would lead to gross physical departures from the norm readily detectable at birth. Accordingly, one measure of genetic damage consequent to atomic bombing would entail determining whether differences in the frequency of congenital malformation arose

Analysis of the Malformation Data. 105 consistent with parental exposure. On an a priori basis, we might expect that the frequency of malformation would increase with increasing parental exposure. The extent to which the frequency of mal- formation would be altered by parental exposure would, naturally, be a function of the com- ponent in the etiology of such malformations attributable to genetic factors. While this com- ponent cannot be exactly specified, that it is not inappreciable is indicated by (1) the number of specific malformations known to be inherited, and (2) the finding that the total malformation rate within the Japanese data increases with increasing parental relatedness (Schull, un- published). 8.4 Concomitant variation influencing the in- dicator. — Among the numerous agents for which there exists some evidence regarding a role in the etiology of congenital malformations in man, and which may prove important in the analysis of these data, are such diverse factors as maternal age, parity, nutrition, and viral in- fections in the first trimester of pregnancy. No really complete review of the literature on the etiology of human malformations has been published. Certain aspects of the etiology have been treated by Warkany (1947), Gruenwald (1947), and Landtman (1948). Within the data here presented, there appear to exist only two sources of concomitant varia- tion which require comment. These are: (a) known differences between exposure categories in maternal age and parity, and (b) possible nutritional differences between the exposure categories. Of these concomitant variables maternal age and parity are undoubtedly the most important. An extensive literature exists describing the effects of these variables on pregnancy outcome. In the main, however, this literature is con- cerned with the effect of maternal age and/or parity on specific malformations rather than on the total malformation rate. Two exceptions are the work of Carter and McCarthy (1951) and Coffey and Jessop (1955), who find a positive association between maternal age and total malformation rate. The relationship of maternal age and parity to malformation rate in the present data is sum- marized in Tables 8.4 and 8.5 and Figures 8.1 and 8.2. Table 8.4 presents the effect of age for specific parities. The data have been analyzed according to the method of Krooth (1955). In the column headed "A.D." have been entered by parity class, the absolute differences between the proportion of all normal infants born at a specific age interval and the proportion of all malformed infants born at the same age interval. A x2 test has been applied to the significance of the findings for each parity class, with the usual convention of indicating significance at the 5 per cent level with one asterisk, and at the 1 per cent level with two asterisks. For each parity, an index of absolute differ- ence (IAD) has been calculated as suggested by Krooth (1955), namely, where ^4 corresponds to the individual entries in the column headed "A.D." For each parity the proportion of the total number of births which occurred at this parity (pT) has been calculated. The IAD for maternal age, all parities considered, is simply the weighted mean of the IAD'S f°r the individual parities. The same ma- terial is presented graphically in Figure 8.1. In the entire data, there is a highly significant age effect, although only two of the individual parity classes show effects at the level of significance. It is noteworthy that by inspection, age effects seem to be limited to the four highest parity classes, suggesting an age-parity interaction. A much more detailed analysis of these data will be presented elsewhere. Table 8.5 presents a similar analysis with respect to parity effects. There is a highly signifi- cant relationship between parity and the fre- quency of malformation in the data as a whole, consistent in all six age classes, although reach- ing the level of significance in just one of these six classes. The findings are shown graphically in Figure 8.2. These results, like the age effects, will be discussed in detail elsewhere. Suffice it to say for the present that the effects of age and parity appear to follow different patterns, the age effect, as noted previously, emerging strongly at the higher ages, the parity effects being more uniformly distributed. Carter and McCarthy (1951) were inclined to attribute the positive association between ma- ternal age and total malformation rate almost entirely to the known association of mongolism with increasing maternal age. This interpreta- tion would not be valid for the Japanese data, since there are included only three children with

106 Chapter VIII Genetic Effects of Atomic Bombs TABLE 8.4 THE EFFECT OF MATERNAL AGE AT FIXED PARITY ON THE FREQUENCY OF MALFORMED INFANTS (Detailed explanation in text.) Total Normal Malformed births births infants Parity 1 < 21 1,929 1,913 16 21-25 8,697 8,636 61 26-30 3,168 3,151 17 31-35 621 617 4 36-10 1801 41 + 19/ ;" Total 14,614 14,515 99 X24 = 1.771 IAD = 0.0510 Parity 2 <21 424 421 3 21-25 6,392 6,334 58 26-30 6,156 6,101 55 31-35 1,431 1,420 11 36-10 357| 41+ SO/ Total 14,810 14,681 129 X24 = 1.117 IAD = 0.0290 Parity 3 <21 37 36 1 21-25 2,016 1,999 17 26-30 5,869 5,810 59 31-35 2,422 2,396 26 36-40 53l\ 525 6 41+ 67 J 66 Total 10,942 10,832 110 X1, = 0.618 IAD = 0.0300 Parity 4 21-25 487 J 484 26-30 2,872 2,853 19 31-35 2,537 2,523 14 36-40 761 752 9 41+ 87 85 2 Total 6,749 6,697 52 X24 = 11.091* IAU = 0.1676 Parity 5 21-25 77\ — —\ 26-30 1,005J 1,069 13J 31-35 1,783 1,758 25 36-40 847 839 8 41 + 102 98 4 Total 3,814 3,764 50 X2. = 6.466 IAD = 0.0869 Pet cent malformed 0.83 0.70 0.54 0.64 0.50 pT = 0.2664 0.71 0.91 0.89 0.77 0.49 pT = 0.2700 2.70 0.84 1.01 1.07 1.17 = 0.1995 1.63 0.66 0.55 1.18 2.30 pT = 0.1230 1.20 1.40 0.94 3.92 pT = 0.0695 AD .02982 .02119 .04538 .00211 .00354 .10204 .00541 .01817 .01079 .01145 .01209 .05791 .00577 .02999 .00001 .01517 .00609 .00300 .06003 .08157 .05963 .10750 .06080 .02577 .33527 .02400 .03295 .06290 .05396 .17381

Analysis of the Malformation Data 107 TABLE 8.4—Continued Total births Normal births Malformed infants Per cent malformed Parity 6 21-25 in - --i 26-30 241/ 248 4/ l-*» 31-35 989 979 10 1.01 36-40 794 780 14 1.76 41 + 147 142 5 3.40 Total 2,182 2,149 33 — X23 = 5.535 Uu = 0.1526 pT = 0.0398 Parities 7 + 21-25 4-1 - -"I 26-30 52 / 54 2 / 31-35 455 449 6 1.32 36-40 935 928 7 0.75 41+ 293 283 10 3.41 Total 1,739 1,714 25 — X28 = 13.046** IAD = 0.2834 p,. = 0.0317 All parities <21 2,395 2,375 20 0.84 21-25 17,684 17,540 144 0.81 26-30 19,363 19,194 169 0.87 31-35 10,238 10,142 96 0.94 36-40 4,405 4,358 47 1.07 41 + 765 743 22 2.88 Total 54,850 54,352 498 X20 = 36.398** I'AD (for mother's age)= 2pTIAD = 0.0691 AD .00587 .15231 .06145 .08548 .30511 .04849 .02196 .26142 .23489 .56676 .00354 .03355 .01378 .00617 .01420 .03051 4.0 3.5 3.0 MALFORMATION (PER CENT) 2.5 2.0- 1.* 1-0 0.5- 15 20 25 30 MOTHER'S 33 40 AGE 4ft FIGURE 8.1 — The distribution of the frequency of infants with major malformation by maternal age for specified parities.

108 Chapter VIII Genetic Effects of Atomic Bombs 4.0T 3.5 3.0- 2.5- MALFORMATION (PER CENT) 2.0- • 1.5- • 1.0- 0.5- 1234567 PARITY FIGURE 8.2 — The distribution of the frequency of grossly malformed infants by parity for specified maternal ages. TABLE 8.5 THE EFFECT OF MATERNAL PARITY AT FIXED AGE ON THE FREQUENCY OF MALFORMED INFANTS (Detailed explanation in text.) Mother's age: < 21 1 Total births Normal births Malformed infants Per cent malformed AD 1,929 1,913 16 0.82 .00547 2 424 421 3 0.71 .02726 3 . 371 41 >• 2.38 .03274 4 5/ ' Total 2,395 2,375 20 .06547 X22 = 1.296 IAD = 0.0327 8,636 1^ = 0.0437 0.70 Mother's age: 21-25 1 . 8,697 61 .06875 2 6,392 6,334 58 0.91 .04166 3 2,016 1,999 17 0.84 .00409 4 487~l 81 5 77 "f 571 r 1.38 .02301 6 I 74- 4 Total 17,684 17,540 144 — .13751 X2. = 4.387 IAD = 0.0688 1^ = 0.3224

Analysis of the Malformation Data 109 TABLE 8.5—Continued Total births Normal births Malformed infants Per cent malformed AD Mother's age: 26-30 1 3,168 3,151 17 0.54 .06358 2 '. 6,156 6,101 55 0.89 .00758 3 5,869 5,810 59 1.01 .04641 4 2,872 2,853 19 0.66 .03621 5 1,005 992 13 1.29 .02524 6 241 237 4 1.66 .01132 7+ 52 50 2 3.85 .00923 Total 19,363 19,194 169 .19957 X28 = 15.935* IID = 0.0998 PT = 0.3530 Mother's age: 31-35 l 621 617 4 0.64 .01917 2 1,431 1,420 11 0.77 .02543 3 2,422 2,396 26 1.07 .03458 4 2,537 2,523 14 0.55 .10294 5 1,783 1,758 25 1.40 .08707 6 989 979 10 1.01 .00763 74. 455 449 6 1.32 .01823 Total 10,238 10,142 96 — .29505 X2a = 10.472 UD = 0.1475 PT = O-186? Mother's age: 36-40 1 180 179 1 0.56 .01979 2 357 355 2 0.56 .03891 3 531 525 6 1.13 .00719 4 761 752 9 1.18 .01893 5 . 847 839 8 0.94 .02231 6 794 780 14 1.76 .11889 7-f- 935 928 7 0.75 .06400 Total 4,405 4,358 47 — -29002 X28 = 6.095 I AD = 0.14 50 PT = 0.0803 Mother's age: 41 + 1 19] —] 2 SO}- 135 — }• 0.74 .13625 3 67 J lJ 4 87 85 2 2.30 .02349 5 ... 102 98 4 3.92 .04992 6' 147 142 5 3.40 .03615 7 _j." 293 283 10 3.41 .07366 Total 765 743 22 .31947 X24 = 3.184 IAD = 0.1597 ^ = 0.0139 All ages: l 14,614 14,515 99 0.68 .06826 2 14,810 14,681 129 0.87 .01107 3' 10,942 10,832 110 1.01 .02159 4 6,749 6,697 52 0.77 .01880 5 3,814 3,764 50 1.31 .03115 6 2,182 2,149 33 1.51 .02673 7 + 1,739 1,714 25 1.44 .01867 Total 54,850 54,352 498 — X26 = 32.594** I"AD (parity)= 0.1002

110 Chapter VIII Genetic Effects of Atomic Bombs diagnoses of mongolism, and of these three, one was born at a maternal age of 25 and the other two at 34 and 40 years of age. The diag- nosis of mongolism, while not particularly un- common at 9 months of age (1 out of 1,580 children), was singularly infrequent in the at- birth data. This is not surprising in view of the not-uncommon failure to diagnose mongolism at birth in Caucasian infants, and the added diagnostic difficulties among Japanese. In view of the greater average age and parity of the mother as parental exposure increases, a spurious irradiation effect could arise if the data were unadjusted for this fact. It might also be pointed out, however, that if no differences are demonstrable without age-parity adjustments, it is unlikely that differences will be detected following adjustment because of the direction of the bias. It was felt that while some adjustment for age-parity effects was indicated, any attempt to subdivide the present data in such a manner as to take into consideration both age and parity effects would jeopardize the validity of the statistical tests employed. Somewhat arbitrarily the decision was reached to take into considera- tion age but not parity effects. Accordingly, for this analysis the terminations have been classified not only with respect to sex, city, and parental exposure, but also maternal age. Five age categories were recognized, namely, 15-20, 21-35, 36-38, 39-41, and 42 and over. Since the selection of the age intervals is purely arbi- trary, the intervals were so selected as to be of the shortest span consistent with significant numbers in those areas where maternal age appears to be most important. It must be borne in mind, however, that the "age effects" which emerge from this analysis are also in part "parity effects." Some bias may be introduced as a consequence of ignoring parity but this bias would again appear to favor a higher rate of malformation among the exposed population. The necessity for accounting for possible differences between exposure cells in maternal nutrition is more difficult to appraise. Warkany's (1947) extensive studies on the effect of severe dietary stresses on pregnancy outcome in gravid rats and mice would lead to the conclusion that diet is of considerable importance. In these animals, the degree of dietary stress needed to induce malformation is, however, far more severe than could have obtained even in the war torn cities of Japan. Moreover, neither Smith (1947) nor Antonov (1947), in their studies of the populations of Rotterdam and Leningrad following severe dietary stress during World War II, could demonstrate an increase in the frequency of congenital malformations. What- ever dietary stresses may have existed in the populations of Hiroshima and Nagasaki during the earlier stages of this study were very prob- ably not so severe as those which obtained in Rotterdam and Leningrad. With respect to possible differences in nutritional status between exposure categories, detailed dietary histories on random samples of the population are not available. However, in a marginal economy such as existed in post-war Japan, there is undoubt- edly a high correlation between economic status and nutritional level; we have already seen (Sec. 5.3) that in neither city is there a demon- strable relationship between economic level and exposure history. It will be assumed in this analysis that there are no important differences in maternal nutrition with respect to the various exposure subclasses. Lastly, by way of completing a discussion of the concomitants, it might be mentioned that during the course of this study no exceptional epidemics of German measles or other viral infections were known to have occurred in Hiroshima or Nagasaki. 8.5 The "at-birth" data. — In Tables 8.6 and 8.7 are presented the "at-birth" frequencies of major malformation by parental exposure, sex of infant, and city, with and without classi- fication of the termination by maternal age. Inspection of these tables reveals no striking, general differences in the frequency of infants with at least one major malformation by sex, city or parental exposure. The results of the analysis of these two tables are given in Tables 8.8 and 8.9. From Table 8.8 (a and b) we note no demonstrable differences when age is ig- nored, among the cities, sexes, or parental ex- posures. It is worth noting at this point that the frequency of malformation in the unexposed groups in Hiroshima and Nagasaki is very simi- lar to the frequency of malformation in Kure where during the years control studies were in effect there 57 malformed infants were reported in 7,544 registered terminations (0.76%). In Table 8.9, with the data further parti- tioned to take into account age differences among the parents, the only "main effect"

Analysis of the Malformation Data 111 TABLE 8.6 FREQUENCY OF MALFORMED INFANTS BY PARENTAL EXPO- SURE, SEX OF INFANT, AND CITY (Unrelated parents) Male infants Hiroshima Mothers 1 2 3 4-5 Total fn 9,005 2,832 1,114 571 13,522 Km 73 30 4 6 113 IP .0081 .0106 .0036 .0105 .0084 f" 772 941 220 109 2,042 2<m n 14 — 1 26 IP .0142 .0149 — .0092 .0127 Z f° 313 197 268 51 832 I' )<m 4 2 1 2 9 1 IP .0128 .0102 .0037 .0370 .0108 fn 209 122 80 56 467 4-5^ m IP .0048 1 .0082 1 — — .0043 2 — — f n 10,299 4,092 1,682 790 16,863 TotaN m U .0086 89 .0115 47 .0030 5 .0114 9 .0089 150 Male infants Nagasaki Mothers 1 2 3 4-5 Total fn 7,608 4,849 360 279 13,096 l<m 68 39 4 3 114 U .0089 .0080 .0111 .0108 .0087 f" 1,120 2,112 134 56 3,422 2^m 7 23 2 — 32 U .0063 .0109 .0149 — .0094 Q f" 129 140 51 18 338 U h * 3-^m 4 — 1 — 5 i IP .0310 — .0196 — .0148 fn 75 103 14 15 207 4-5 ^m IP 2 — — — 2 .0267 — — — .0097 f" 8,932 7,204 559 368 17,063 Total < m U .0091 81 .0086 62 .0125 7 .0082 3 .0090 153

112 Chapter VIII Genetic Effects of Atomic Bombs TABLE 8.6— Continued Female infants Hiroshima Mothers 1 2 3 4-5 Total r fn 8,289 2,536 1,071 546 12,442 l<m IP .0105 87 .0083 21 .0093 10 .0128 7 .0100 125 fn 728 909 204 93 1,934 2-^m [p .0082 6 .0132 12 .0098 2 .0215 2 .0114 22 w 3-^m fn 283 188 253 56 780 4* 3 2 4 1 10 .c • IP .0106 .0106 .0158 .0179 .0128 £ fn 186 122 77 61 446 4-5 4m U .0161 3 .0082 1 .0130 1 .0164 1 .0135 6 fn 9,486 3,755 1,605 756 15,602 Total -I m IP .0104 99 .0096 36 .0106 17 .0146 11 .0104 163 Female infants Nagasaki Mothers 1 2 3 4-5 Total f 1° 1-jm 7,002 66 .0094 4,467 387 280 3 .0107 12,136 105 .0087 IP 31 .0069 5 .0129 fn 1,050 2,032 145 60 3,287 2\m 4 13 3 — 20 IP .0038 .0064 .0207 — .0061 G 3^m f" 114 133 43 17 307 £. 1 — — — 1 21 IP .0088 — — — .0033 fn 64 75 21 13 173 4-5^ m U — .0267 2 — — .0116 2 — — — fn 8,230 6,707 596 370 15,903 1 IP Total J m .0086 71 .0069 46 .0134 8 .0081 3 .0080 128

j 8 I NTAL % a P oa o, .1 III 3.2 X £13 * *J a^. g ssO !•*- irs ^ K% r-^o ^^'N s« O ^osO Cs Os— O 00 *fs 1 — Tr tN O irs — sO — fst «\ 2 |S?§ -*r4._i rsi00so 00 — 00 so I I Os| 1 £3 £~£ •*- 2 " • I BI I p .-" ° 9 *" p £j q M q q q ^ r r • so *sso *— vs rj f*s^s* ^ c- <" i o o •^ q q q ® sOOOs £"2 •- ^ r*- i*s*o * Os "3 • ^K ?Vs5ds O -iH — ' rii o oo Os O so I-e O 00 -- in — — "s g - q c..; *.j •„.) oo l l «r-— so^n oo l I *n 1 I M | I sO g ^O •\ in I I P* I l q *M I S-S 211 -11 "Ml "I I S~£ 211 211 -II 55- i-\ ~ c1 ^soso vr — cr; 0D I I P —O \ n — ri — *— c£ M so q ^ q q q I ^_j •"I I £•"£ -II "II * I I 12 2"g 8 S'-S eN g * q q rsOs *-* \ I Os I Os tA | | — 1 "M s^a 21 i =11 ~ii q 21 I 5NS 31 I =1 I «! I •" 8 '*i-*r- Mosp rsiinrsi inr4p r-- —o •^ 'N so^en ^* x* o os M r* r|1*— M— •-•— *« q ^ q q q q fs I I osso r i T I I t — T O i i 0Oll O — fN sO I I *n Os— |l ^ q q fN r-t — 04 O S — ^* — — "* n i S SI I *| I IES2 vs 00 00 — *H ^ «<\ CM q q q .-* fN — — mM sOO— — oor- •-renoo tN^r^ O so r*--fl-Os oo *^ trs ri go 00 ^ 00 O^•-*OsO -^'N — -J ss? 'N — c cBo. c c c. c£o. c c ex. c&o, aSdi BC J^ c So. GO •-* _1_ 'n *V ' ! 2 « en— oo rs i — a I- •n N« S''^ ~ | | s 1 1 o 1 1 0D sO — 'fs 00 q * q " q s I o — *^ O 1 1 r^ — *s ri tN 1 1 R2S o ^ q - q s 3 |-§ B. § Os— 00 fsf ** 00 00 — O r- i r^.ce oo M^sO q & q fO ? *5 «sOg "2 'N tN Os so — Q rsi — - 00 ^ §"i « §S-E i SI 1 r; 1 1 §"S rn ^ - l si l -II ml 1 - II SI 1 l 1 SI 1 "M 1 eN j J - 1 1 SI 1 „ , —0 0-0 s I '* a 21 1 40 1 I 00 tN *N CS A I =~S 21 1 o 1 1 r^ 1 1 tO -T q fN I r- w~*n *r 1 1 1 oo O sj | I T *- r~ •* rsi 00 1 1 S^5 q q rs| § 1 1 SI 1 "1 1 "1 l 1 1 1 SI 1 ~ 1 511 211 »-0 in 1 1 x-s 2. q » 1 811 SM — I I r^ 1 1 SI 1 fN tN = 1 00 so 'N i 1 21 I 0D 1 1 5"8 — 'N q " S •s.sO 0O'WOs 00'-J"* ^ Os fN00 9*9 r^t^oo *r P £ 8 §1 S 1 *fs f* r*. O q so- ^ « 1 °* 1 1 rj ' "1 1 « l l 21 I 1 00 1 1 — 1 1 s 1 ^OvsOs 0s I 1 1 » so •n 1 I SI 1 - 1 1 OsirsOs r- r- g ' q •n* *R S2S |*| o^ i^ ?S£ sor-e-• «sOO r - S ^ 8 ^ 8 sO Os sO 5 ? q q § I S2S SI 1 EI i Q o — - — 0 rj - ° q M q Si o 3-o °-S i"2 — CO SSo ° N" °' - ° °. o s" ° I Osr ,g K^S SI 1 2| I « 1 1 S^B 'N q q v-* — q a T 1 S»9 S 1 1 S 1 1 " 1 1 ?°~§ •3 o q .S oos ^ » S°£ *- s0Om a o • o S 1^ O 3 q q oC c: 1 a 1 «— « £2? 5"B S fN s=a c£a, c£o. c H r- c€o, cSa o *-s oo — 4. 4 I 3 a H o. « IN tO 113

114 Chapter VIII Genetic Effects of Atomic Bombs emerging at the level of significance is the al- ready discussed effect of maternal age on the frequency of congenital malformation. Among the first order interactions none is significant save mother's exposure-mother's age. However, since the effect of mother's exposure is not inde- pendent of age no generalizations can be made with respect to the effect of maternal irradiation TABLE 8.8 CHI-SQUARE ANALYSIS OF THE FRE- QUENCY OF CONGENITALLY MALFORMED INFANTS BY SEX, CITY, AND PARENTAL EXPOSURE (Unrelated parents) a. All exposure cells (4X4) Source DF x' P Total . 63 63.077 .50-.60 Interactions, first order CS . 1 1.039 .30-.50 CM . 3 9.427 .02-.05 CF . 3 2.054 .50-.70 SM . 3 5.210 .10-.20 SF . 3 3.057 .30-.50 MF . 9 6.107 .70-.80 Main effects City (C) . 1 2.269 .10-.20 Sex (S) . 1 0.169 .50-.70 Mothers (M) . 3 1.907 .50-.70 Fathers (F) . 3 1.232 .70-.80 b. Excluding parents in Category 1 (3X3) Source DF x' P Total . 35 28.806 .80-.90 Interactions, first order CS . 1 1.027 .30-.50 CM . 2 6.411 .02-.05 CF . 2 3.282 .10-.20 SM . 2 3.052 .20-.30 SF . 2 2.364 .30-.50 MF . 4 4.080 .30-.50 Main effects City (C) . 1 2.406 .10-.20 Sex (S) . 1 0.015 .90-.95 Mothers (M) . 2 0.222 .80-.90 Fathers (F) . 2 0.480 .80-.90 based on the main effects test alone. In order to interpret the effect of maternal exposure, we must consider separately the maternal age classi- fications. Table 8.10 presents the distribution of the frequency of malformation by mother's age and exposure. For purposes of comparison and its bearing on a possible somatic maternal effect, a comparable table (Table 8.13) is given for father's exposure. In Table 8.11 are given the results of analyzing the effect of mother's exposure at each of the five different maternal age levels. For these five age levels mother's exposure seems to exert an effect only in the 15-20 year age interval. From inspection of Table 8.10 we note that the frequency of mal- TABLE 8.9 CHI-SQUARE ANALYSIS OF THE FRE- QUENCY OF CONGENITALLY MALFORMED INFANTS BY CITY, MATERNAL AGE, AND PARENTAL EXPOSURE (Unrelated parents) All exposure cells Source Total DF 156 x1 P Interactions, first order CM 164.205 .50-.60 3 9.657 .02-.05 CF 3 2.053 .50-.70 CA 4 1.914 .70-.80 AM 12 23.786 .02-.05 AF 12 7.907 .70-.80 MF 9 6.107 .70- .80 Main effects Aee (A) . 4 18.057 .001-.01 City (C)' Age intervals 15-20 1 0.438 .50-.70 21-35 1 2.720 .05-. 10 36-38 1 0.032 .80-.90 39-41 1 0.062 .80-.90 42+ . . .. I 1.863 .10-.20 Sum 5 5.115 .30-.50 Mothers (M)'-» Age intervals 15-20 3 14.581 .001-.01 21-35 3 4.463 .20-.30 36-38 3 2.277 .50-.70 39-41 3 3.111 .30-.50 42 + 3 1.462 .50-.70 Fathers (F)1 Age intervals 1 5-20 15 25.894 .02-.05 3 0.411 .90-.95 21-35 3 2.619 .30-.50 36-38 3 3.555 .30-.50 39-41 3 1.303 .70-.80 42 4- 3 1.400 .70-.80 Sum 15 9.288 .80-.90 • Adjusted for age (see Sec. 6.5). "See Table 8.11. formation tends to increase with increasing ma- ternal exposure with an eight-fold difference between mother's "1" and mother's "4-5," or a two-fold difference between mother's "2" and mother's "4—5." This difference ceases to be significant, however, when consideration is

Analysis of the Malformation Data 115 limited to only those cells in which both parents are exposed. Moreover, it is not significant if only mothers who were unexposed are excluded from consideration (X2 = 1.062; DF = 2). The significance of the difference seems to be largely a function of the inordinately low frequency of malformation observed among the unexposed mothers in the 15-20-year age interval. In amination necessitated two departures from the usual analytical approach. Firstly, parental ex- posure categories 3, 4, and 5 were pooled. Sec- ondly, all extraneous variation was ignored. TABLE 8.12 THE DISTRIBUTION BY MATERNAL EXPOSURE AND PARITY OF MALFORMED INFANTS BORN TO MOTHERS OF AGES 15-20 Mothers TABLE 8.10 THE DISTRIBUTION OF FREQUENCY OF MALFORMED INFANTS CLASSIFIED BY MOTHER'S AGE AND EXPOSURE ONLY Parity 1234-5 Total fn 1,204 805 168 77 2,254 Km 3 13 1 3 20 [p .0025 .0161 .0060 .0390 .0089 Mothers fn 266 185 50 24 525 2-1 m 1 3 2 — 6 [p .0038 .0162 .0400 — .0114 Age 123 4-5 Total fn 1,492 1,013 221 106 2,832 15-20-^ m 5 16 3 3 27 [p .0034 .0158 .0136 .0283 .0095 fn 22 23 3 5 53 3-4^ m 1 — — — 1 [p .0455 — — — .0189 fn 32,256 17,917 3,685 1,933 55,791 21-35-{m 294 138 30 22 484 [p .0091 .0077 .0081 .0114 .0087 fn 1,492 1,013 221 106 2,832 TotaHm 5 16 3 3 27 [p .0034 .0158 .0136 .0283 .0095 {n 2,007 1,569 324 129 4,029 m 23 14 1 1 39 p .0115 .0089 .0031 .0078 .0097 fn 901 920 152 87 2,060 39-4l-|m 10 17 2 — 29 [p .0111 .0185 .0132 — .0141 TABLE 8.13 THE DISTRIBUTION OF FREQUENCY OF MALFORMED INFANTS CLASSIFIED AT BIRTH BY MOTHER'S AGE AND FATHER'S EXPOSURE ONLY fn 291 339 60 29 719 42+^m 8 6 1 — 15 [p .0275 .0177 .0167 — .0209 Fathers fn 36,947 21,758 4,442 2,284 65,431 TotaNm 340 191 37 26 594 [p .0092 .0088 .0083 .0114 .0091 Age 123 4-5 Total fn 2,233 458 104 37 2,832 15-20-jm 22 4 1 — 27 (p .0099 .0087 .0096 — .0095 TABLE 8.11 CHI-SQUARE ANALYSIS OF THE EFFECT OF MOTHER'S EXPOSURE ON THE FREQUENCY OF MALFORMED INFANTS AT EACH OF THE FIVE DIFFERENT AGE LEVELS fn 44,764 8,201 1,753 1,073 55,791 2 1-35 -{m 379 74 21 10 484 IP .0085 .0090 .0120 .0093 .0087 fn 2,655 1,074 210 90 4,029 36-38^ m 29 7 1 2 39 [p .0109 .0065 .0048 .0222 .0097 Age group DF x' P 15-20 3 14.581 .001-.01 fn 1,203 672 122 63 2,060 39-41-1 m 18 10 1 — 29 [p .0150 .0149 .0082 — .0141 21-35 3 4.463 .20-.30 36-38 3 2.276 .50-.70 fn 341 280 68 30 719 42 -Hm 9 5 1 — 15 [p .0264 .0179 .0147 — .0209 39-41 2* 2.421 .20-.30 42+ 2* 1.196 .50-.70 • Groups 3, 4, and 5 have been pooled. fn 51,196 10,685 2,257 1,293 65,431 Total^m 457 100 25 12 594 [p .0089 .0094 .0111 .0093 .0091 Table 8.12 is presented the distribution of mal- formed infants among mothers in the 15-20- year age interval by parity and maternal ex- posure. We note that unexposed mothers (in These considerations must be borne in mind in appraising Tables 8.14 and 8.15.1 In Table 8.14 or second pregnancies experienced an extremely low malformation rate. An explanation for this extremely low rate is not immediately apparent. 8.6 The "9-months" data. — The sparcity of the material available from the 9-months ex- 1 The discrepancy in terms of total number of in- fants recorded in Tables 8.3 and 8.13 is due to the fact that in the former any child seen in the "9- months follow-up" is scored (these infants will vary in age from 7j to 17 months), whereas in the latter only those infants whose ages were 8-10 months in-

116 Chapter VIII Genetic Effects of Atomic Bombs TABLE 8.14 THE FREQUENCY OF MALFORMED INFANTS AMONG ALL INFANTS RE-EXAMINED AT 9-MoNTHS OF AGE, BY CITY AND PARENTAL EXPOSURE (Malformed infants are defined here as being those infants found to have one or more major defects at 9 months of age not diagnosed at birth.) Hiroshima Mothers 1 2 3-5 Total fn iJm 5,424 116 .0214 1,718 25 1,016 13 .0128 8,158 154 .0189 IP .0146 1° 2-1 m 458 5 ''01 12 .0239 180 1,139 19 .0167 | IP .0109 2 .0111 1• f" 303 178 250 3 .0120 731 £ 3-5^ m 5 .0165 2 10 .0137 IP .0112 f" Total <m 6,185 126 .0204 2,397 39 1,446 18 .0124 10,028 183 .0182 IP .0163 Nagasaki Mothers 1 2 3-5 Total f" 4,016 2,474 371 6,861 iJm 97 45 5 147 IP .0242 .0182 .0135 .0214 fn 563 990 115 1,668 2^m 16 22 4 42 B IP .0284 .0222 .0348 .0252 o> — * fn 116 142 61 319 01 £ 3-5 ^m 1 2 3 6 [P .0086 .0141 .0492 .0188 fn 4,695 3,606 547 8,848 TotaK m 114 69 12 195 IP .0243 .0191 .0219 .0220 are presented the frequencies of malformed infants, diagnosed at 9 months but not so diag- nosed at birth, distributed by city and parental exposure. The analysis of these data is presented in Table 8.15. Inspection of the analysis reveals no significant differences between cities or father's exposure categories, but an effect of mother's exposure significant at the 5 per cent level. This effect, however, is contrary to what one might expect from radiation, since there are clusive are scored. The purpose of the more rigid definition in the latter table was to minimize another source of extraneous variation, namely, age of infant. TABLE 8.15 CHI-SQUARE ANALYSIS OF THE FRE- QUENCY OF MALFORMED INFANTS AT 9 MONTHS OF AGE BY CITY AND PARENTAL EXPOSURE (Unrelated parents) Source Total Interactions, first order CM CF MF DF 17 Main effects City (C) 1 Mothers (M) 2 Fathers (F) 2 22.543 1.062 1.011 3.942 3.441 6.070 1.643 .10-.20 .50-.70 .50-.70 .30-.50 .05-.10 .02-.05 .30-.50 TABLE 8.16 THE DISTRIBUTION BY MATERNAL EXPOSURE OF THE SEVEN MOST COMMON MAJOR CONGENITAL MALFORMATIONS IN THE JAPANESE, EXCLUSIVE OF CONGENITAL HEART DISEASE (The incidences of the malformations are given in terms of number of malformations per 1,000 births.) Mothers 1 2 3 4-5 Type of malformation No./ 1,000 births No./ 1,000 births No./ 1,000 births No./l.OOO births No. 76 No. 17 No. No, 0.70 0.78 4 0.90 2 0.88 Cleft palate single ^7 0.87 7 0.32 3 0.68 Club foot 52 1.41 28 1.29 5 1.13 3 1.31 Harelip with cleft palate. . . Harelip without cleft palate. Polydactyly 55 1.49 1.00 1.00 34 18 28 1.56 0.83 1.29 3 5 0.68 1.13 0.68 6 2.63 37 V7 3 2 2 0.88 0.88 Syndactyly 71 0.62 10 0.46 3 0.68 Sum of malformations... Total Births 262 7.09 142 6.53 21,758 26 5.85 4,442 15 6.57 2,284 36,947 Total 49 42 88 98 62 70 36 445 65,431

Analysis of the Malformation Data 117 relatively more malformations among the chil- dren of category 1 mothers, and this despite the fact that age-parity effects, which should introduce a bias in the direction of hypothesis, have not been allowed for. 8.7 Analysis by specific malformation type. — The genetic basis of most of the common types of malformations is poorly understood. Accordingly, the remote possibility had to be considered that the frequency of children with certain kinds of malformations was affected in a rather specific way by parental radiation. Despite the improbability of the existence of such an effect in the absence of an effect of radiation exposure in the material as a whole, an analysis of this possibility has been under- taken. The results, with respect to the seven most common malformation types, are given in Table 8.16. When the data are partitioned in this manner there are relatively few malforma- tions of any specific type. For this reason, the classification according to radiation history is by mother's exposure only. Inspection of Table 8.16 reveals no noteworthy trends. 8.8 Summary. — Analysis of the frequency of malformed infants by city and parental ex- posure reveals no significant, consistent effect of parental exposure. The small departures from expectation which are observed cannot be said to be in the direction of genetic hypothesis.

Next: ANALYSIS OF THE STILLBIRTH DATA »
Effect of Exposure to the Atomic Bombs on Pregnancy Termination in Hiroshima and Nagasaki Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!