UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Interactive effects of alcohol and diabetes during pregnancy on the rat fetus Lin, Yi 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1992_fall_lin-yi.pdf [ 2.45MB ]
Metadata
JSON: 831-1.0077299.json
JSON-LD: 831-1.0077299-ld.json
RDF/XML (Pretty): 831-1.0077299-rdf.xml
RDF/JSON: 831-1.0077299-rdf.json
Turtle: 831-1.0077299-turtle.txt
N-Triples: 831-1.0077299-rdf-ntriples.txt
Original Record: 831-1.0077299-source.json
Full Text
831-1.0077299-fulltext.txt
Citation
831-1.0077299.ris

Full Text

INTERACTIVE EFFECTS OF ALCOHOL AND DIABETES DURINGPREGNANCY ON THE RAT FETUSByYi LinB. Sc. (Public Health) Tianjin Medical Institute, 1983A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESSCHOOL OF FAMILY AND NUTRITIONAL SCIENCES, DIVISION OF HUMAN NUTRITIONWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAAugust 1992© Yi Lin, 1992In presenting this thesis in partial fulfilment of the requirements for an advanced degree atthe University of British Columbia, I agree that the Library shall make it freely availablefor reference and study. I further agree that permission for extensive copying of thisthesis for scholarly purposes may be granted by the head of my department or by hisor her representatives. It is understood that copying or publication of this thesis forfinancial gain shall not be allowed without my written permission.Human NutritionThe University of British Columbia2075 Wesbrook PlaceVancouver, CanadaV6T 1Z1Date:/ / 7zAbstractSubstantial literature indicates that maternal diabetes has teratogenic effects on the fetusand that consumption of alcohol during gestation is also teratogenic to the fetus. In thepresent study, Sprague-Dawley female rats were used to test whether small amounts ofalcohol administered during organogenesis has an effect on the outcome of pregnancy innondiabetic rats; to confirm the teratogenic effect of diabetes on fetuses; and to establishwhether there is an interaction between alcohol and diabetes on the outcome of pregnancy.All rats were fed Purina rat chow and tap water ad libitum. Diabetes was induced by tailvein administration of streptozotocin (60 mg/kg body weight) one week before mating.Alcohol was administered during organogenesis (days 6-11 of gestation) intragastrically.The incidence and types of external and skeletal malformations were recorded in thefetuses of all treatment groups on day 21 of gestation. In addition, the fetal developmentwas followed by assessing the ossification of the skeleton on gestational day 21 with theaid of alizarin red S and alcian blue 8GS staining.The results of this investigation support the view of previous studies that maternaldiabetes is teratogenic to the offspring in terms of significantly increased malformationrates, and retarded fetal development (smaller fetal weight, bigger placenta, and retardedskeletal ossification compared to controls).Consumption of a small amount of alcohol (2g/kg/day) during organogenesis (days6-11 of gestation) did not seem to intoxicate the dams, the body weight gain was ata normal rate, and there was a bigger fetal/placental ratio compared to controls. Interms of skeletal development, alcohol exposure seemed to enhance bone development.However, there were a few fetuses with absence of ossification centers in the hyoid bone11and the short 13th rib which were significantly different from other treatment groups.Whether this represents retarded skeletal development is not clear.In the present study, a significant interaction between maternal diabetes and alcoholadministration during organogenesis was observed. The fetal external malformation ratewas significantly increased in diabetic rats exposed to alcohol compared to diabetic ratsnot exposed to alcohol. Furthermore, more fetuses had poorly ossified thoracic vertebralcenters, poorly ossified cervical arches, and poorly ossified supraoccipital bone in diabeticrats exposed to alcohol than in diabetic rats not exposed to alcohol. These observationssuggest that a small amount of alcohol consumption during organogenesis could possiblyexacerbate the embryotoxic effects of maternal diabetes.111Table of ContentsAbstract iiList of Tables viiiList of Figures xAcknowledgement xi1 Introduction1.1 Evidence that alcohol is teratogenic1.1.1 Studies in humans1.1.2 Studies in animals1.2 Evidence that diabetes is teratogenic1.2.1 Studies in humans1.2.2 Studies in animals1.2.3 The teratogenic period of diabetic pregnancy1.2.4 The proposed mechanisms1.3 Evidence for an interaction between alcohol and other factors1.4 Evidence for an interaction between alcohol and diabetes1.5 Skeletal examination in experimental teratology studies1.6 The purpose of the study2 Methods 242.1 Preliminary experiments 2411151111• . 1314• • 15• 17• . 202123iv• 25• 25• 25252626272727282829313234342.2 Experimental methods2.2.1 Chemicals and reagents2.2.2 Animals and diets2.2.3 Induction of diabetes2.2.4 Breeding2.2.5 Administration of alcohol2.2.6 Blood alcohol assay2.2.7 Body weight record2.2.8 Blood glucose measurement . .2.2.9 Termination of gestation2.2.10 Skeletal staining2.2.11 Skeletal examination2.2.12 Food record2.2.13 Statistical analysis3 Results3.1 Calorie and water consumption3.2 Reproductive performance3.3 Body weight and weight gain of dams during3.4 Blood glucose concentrations3.5 Peak blood alcohol levels3.6 Outcome of pregnancy3.7 Fetal weight, placental weight and fetal3.8 External malformations3.9 Skeletal malformations3.10 Skeletal ossification and variantspregnancyweight/placental weight ratio383942444445495254V3.10.1 Metacarpus, metatarsus 543.10.2 Sternum 563.10.3 Vertebrae 603.10.4 Skull 663.10.5 Other skeletal variants 693.10.6 Bone areas 694 Discussion 714.1 Calorie and water consumption 714.2 Reproductive performance 724.3 Weight gain 734.4 Outcome of pregnancy 744.5 Fetal and placental weights 764.6 Malformations 784.7 The proposed mechanisms for malformations 794.8 Skeletal ossification 824.8.1 Number of ossification centers 834.8.2 Skeletal variations 844.8.3 Ossification of the skull 864.8.4 Bone areas 874.9 The effect of alcohol consumption 894.10 Summary and conclusion 89Appendices 91A Photographs 91B Chi-square tests, P values 102viC Two way ANOVA statistical effects diabetes vs alcohol, P values 104D Tests of Simple Effect, P values 106Bibliography 108viiList of Tables1.1 Principal Features of the Fetal Alcohol Syndrome Observed in 245 PeopleAffected 31.2 Associated Features of the Fetal Alcohol Syndrome Observed in 245 PeopleAffected 43.3 Average water and calorie intake per day during gestation in each treatment group (Mean ± SD) 353.4 Breeding and pregnancy rates of animals in each treatment group . . . 383.5 Body weight and weight gain of dams during gestation in each treatmentgroup (mean + SD) 403.6 Blood glucose concentration of dams at different times during the experiment, in each treatment group (mmol/L) 443.7 Reproductive variables on day 21 of gestation in each treatment group. 453.8 Fetal and placental weights on day 21 of gestation in each treatment group(mean ± SD) 463.9 Numbers of fetuses with external malformations iii each treatment group 513.10 Numbers of fetuses with skeletal malformations in each treatment group 523.11 Development of ossification centers in metacarpus, metatarsus, and sternum in each treatment group 553.12 Skeletal variants observed in fetuses of each treatment group 583.13 Mean numbers of poorly ossified centers in sternum, thoracic and lumbarvertebrae in each group (mean ± SD) 59vii’3.14 Frequencies of fetuses according to the most cephalic and most caudalvertebral ossification centers and the most caudal vertebral arches . . . 613.15 Ossification centers and foramen in skull 673.16 Ossified tissue areas (mm2)measured in the selected bones (mean ± SD) 70ixList of Figures2.1 Skeletal districts for examining the ossification centers, skeletal variantsand bone areas 303.2 Daily mean calorie intake during pregnancy 363.3 Daily mean water intake during pregnancy 373.4 Maternal body weight changes during gestational period in 4 groups . 413.5 Interaction of diabetes at alcohol on maternal weight gain during D12-21 433.6 Interaction of alcohol at diabetes on maternal weight gain during D12-21 433.7 Interactioll of diabetes at alcohol on fetal weight 483.8 Interaction of alcohol at diabetes on fetal weight 483.9 Interaction of diabetes at alcohol on f/p ratio 503.10 Interaction of alcohol at diabetes on f/p ratio 503.11 Observed irregular shapes of sternebral ossification centers 573.12 Ossification centers of cervical arches 633.13 Observed irregular shapes of vertebral ossification centers 65xAcknowledgementI wish to thank my research supervisor, Dr. Melvin Lee, for his kind help and knowledgeable guidance throughout the course of this study. I am also deeply grateful to Dr.Muriel Harris for her constructive advice in writing my thesis. My thanks are happilyextended as well to Dr. Linda McCargar, Dr. Joseph Leichter, and Dr. David Kitts forserving on my committee.I would also like to thank Mrs. Virginia Green for her advice and assistance withstatistical analysis of the data, Mr. Alistair Blachford for his instruction in the techniqueof bone area measurement. My thanks also go to Mr. Zbysek Masin for his technicalsupport in the animal room.xiChapter 1Introduction1.1 Evidence that alcohol is teratogenicThe potential for damage to the offspring from prenatal alcohol exposure is now wellestablished. A link between alcohol and birth defects had been strongly suspected sincethe 1700s, with early scientific evidence gradually being accumulated during the 19thand early 20th centuries (Warner and Rosett, 1975).In 1973, Jones et al. (l973a; 1973b) rediscovered an effect of alcohol on morphogenesis. In their two original articles, 11 cases were reported and these 11 children, allraised in a fetal environment provided by an alcoholic mother, had a similar pattern ofcraniofacial, limb and cardiovascular defects with prenatal-onset growth deficiency anddevelopmental delay. They named this condition “fetal alcohol syndrome”. The characterization of “fetal alcohol syndrome” by Jones et al. stimulated a great deal of researchboth in humans and laboratory animals.1.1.1 Studies in humansWithin a few years after the original reports by Jones et al., several hundred case reports(Clarren and Smith, 1978) were published in the medical literature of many countries.A great deal of evidence supports the view that exposure to ethanol during gestation isassociated with a variety of negative outcomes including perinatal death, compromisedgrowth, and behavioral deficits. Fetal alcohol syndrome, the most severe end of the1Chapter 1. Introduction 2spectrum of effects of alcohol on the fetus, has been reported around the world with afrequency of one to three per 1,000 births based on local reports from various U.S. andEuropean studies. Abel and Sokol, basing their estimate on 20 studies from Australia,Europe, and North America, found a worldwide incidence of 1.9 cases of FAS per 1,000live births (Warren and Bast, 1988). It has been calculated that approximately 5% ofall congenital anomalies may be attributable to prenatal alcohol exposure (Sokol et al.,1986).However, not all affected infants present the full clinical picture at birth. Alongthe rest of the continuum toward normal are persons with every subcombination of fetal-alcohol-syndrome anomalies. Each anomaly can independently vary in severity and gradeinto the normal range (Clarren and Smith, 1978). The abnormalities most typicallyassociated with alcohol teratogenicity can be grouped into four categories: central nervoussystem dysfunctions; growth deficiencies; a characteristic cluster of facial abnormalities;and variable major and minor malformations. Tables 1.1 & 1.2 show the frequency ofspecific malformations within each category (Clarren and Smith, 1978).The variability of phenotype probably results from variations in dose exposure andgestational timing offset by the genetic background of the individual fetus. Nearly allpatients recognized as having the full fetal-alcohol-syndrome phenotype have been born toheavy daily alcohol users or relatively frequent intermittent heavy alcohol users (Clarrenand Smith, 1978).The critical amount of maternal alcohol intake needed during pregnancy to produceadverse fetal effects has not been definitely established. However, a definite risk has beenestablished with an oral maternal intake of 9Oml (3oz) absolute alcohol (6 average-sizeddrinks) per day during pregnancy. This consumption level was derived from animal studies by noting the minimum blood alcohol concentration (73mg/lOOmi in mice) responsiblefor a teratological effect. The amount of alcohol consumption necessary to achieve thatChapter 1. Introduction 3Table 1.1: Principal Features of the Fetal Alcohol Syndrome Observed in 245 PeopleAffectedFEATURE MANIFESTATIONCentral-nervous-systemdysfunction:Intellectual Mild to moderate mental retardation*Neurologic Microcephaly*Poor co-ordination, hypotoniafBehavioral Irritability in infancy*Hyperactivity in childhoodfGrowth deficiency:Prenatal <2 SD for length & weight*Postnatal <2 SD for length &‘ weight*Disproportionately diminishedadipose tissuefFacial characteristics:Eyes Short palpebral fissures*Nose Short, upturnedHypoplastic philtrum*Maxilla HypoplasticfMouth Thinned upper vermilion*Retrognathia in infancy*Micrognathia or relative prognathiain adolescencef* Feature seen in> 80% of patients.f Feature seen in> 50% of patients.Source: Clarren, S.K., and Smith, D.W. The fetal alcohol syndrome.N. Engi. J. Med. 298: 1063-1067, 1978.Chapter 1. Introduction 4Table 1.2: Associated Features of the Fetal Alcohol Syndrome Observed in 245 PeopleAffectedAREA FREQUENT* OCCASTONAL$Eyes Ptosis, strabismus, Myopia, clinical microphepicanthal folds thalmia, blepharophimosisEars Posterior rotation Poorly formed conchaMouth Prominent lateral Cleft lip or cleft palate,palatine ridges small teeth with faulty enamelCardiac Murmurs, especially Ventricular septal defect,in early childhood, great-vessel anomalies,usually atrial tetralogy of Fallotseptal defectRenogenital Labial hypoplasia Hypospadias, small rotatedkidneys, hydronephrosisCutaneous Hemangiomas Hirsutism in infancySkeletal Aberrant palmar Limited joint movements,creases, pectus especially fingers & elbows,excavatum nail hypoplasia, especially 5th,polydactyly, radioulnar synostosispectus carinatum, bifid xiphoid,Klippel-Feil anomaly, scoliosisMuscular Hernias of diaphragm, umbilicusor groin, diastasis recti* Reported in between 26 & 50% of patients.f Reported in between 1 & 25% of patients.Source: Clarren, S.K., and Smith, D.W. The fetal alcohol syndrome.N. Engi. J. Med. 298: 1063-1067, 1978.Chapter 1. Introduction 5blood alcohol concentration in a 130 lb. woman for a period of 6-8 hours was calculatedto be 90m1 of absolute alcohol (Maykut, 1979). Although confirmatory evidence is notyet available, it has been suggested that a continuum of fetal alcohol effects ranging fromthe normal state to abnormalities in growth and intellectual performance to congenitalmalformations occurs with increasing intake of alcohol (Maykut, 1979).The period during pregnancy when alcohol is imbibed may be another determiningfactor in producing morphologic changes. It has been suggested that the first 3-4k monthsof gestation may be the critical period for producing birth defects since rapid organcell differentiation (e.g., brain, heart, eyes) occurs at this time. Intrauterine growthretardation may be exacerbated by alcohol consumption in later pregnancy (Maykut,1979).In addition, the effects of maternal alcohol exposure on the fetus may be relatedto the peak blood alcohol concentration, which varies from individual to individual. Ithas been noticed that there are racial differences in the rate of alcohol metabolism andalso, alcohol metabolism may vary with age. Thus the variability of observed symptomsis probably due to differences in dose exposure at various gestational periods offset bygenetic background of the particular fetus (Maykut, 1979).1.1.2 Studies in animalsAnimal studies and the development of appropriate animal models have provided a wealthof information about the effects of gestational alcohol exposure (Driscoll et al., 1990).Animal models are particularly important because of the opportunity to control for factors seldom accounted for in human studies (including control of timing and dose ofalcohol exposure, control of nutrition through pair-fed controls, control of the postnatalenvironment through cross-fostering, and consideration of individual differences throughcross-strain comparisons). Extensive animal studies have now demonstrated the specificChapter 1. Introduction 6teratogenic properties of ethyl alcohol in several species of animals, with many of the abnormalities being similar to those found in man (Clarren and Smith, 1978). The mouse(Chernoff, 1975; Randall et al., 1977), guinea pig (Papara-Nicholson and Telford, 1957),beagle dog (Ellis and Pick, 1976), rat (Sandor and Amels, 1971; Tze and Lee, 1975), andpigtail macaque (Elton and Wilson, 1977) are among the many species that have beenstudied as models for the teratogenicity of alcohol.Decreased birth weight and litter size in offspring of alcohol-fed mice and rats havebeen found by many investigators (Chernoff, 1975; Martin et al., 1977b; Tze and Lee,1975), although not by all (Sandor and Amels, 1971; Weathersbee and Lodge, 1978).Many morphological effects are also found, but these are not consistent across species.In the mouse, illvestigators have produced eye defects, cardiac and neural abnormalities,digit anomalies, and cardiovascular, urogenital, and head malformations by administeringlarge doses of alcohol summarized by Streissguth, et al. (1980). In the rat, microcephalyand a shrivelled appearance have been noted by Tze and Lee (1975). However, apparentmalformations were not observed in ethanol treated rats by many other researchers (Abeland Dintcheff, 1978a; Abel, 1978b; Abel, 1979a; Lee and Leichter, 1983; Testar et al.,1986).Methodological differences among experiments of the time period during gestationwhen alcohol is administered, the dose, the route of administration, and the strain ofanimal used may contribute to the difficulties encountered in reproducing experimentalresults (Streissguth et al., 1980). The three commonly employed methods to administeralcohol are: placing alcohol in drinking water, oral intubation and liquid diet.Ideally, potential teratogens should be administered to animals by the same route thatexposure occurs in humans. For alcohol, the most obvious choice is the oral route. Thesimplest approach would be to place alcohol in the animal’s drinking water. However,this method has several disadvantages. First, it produces only low blood alcohol levels,Chapter 1. Introduction 7and it is the blood alcohol level, not the daily dose of alcohol administered, that appearsto be important in producing fetal alcohol effects, since studies in different strains of micehave shown that prenatal death, malformations, and fetal weights were directly relatedto maternal blood alcohol levels (Chernoff, 1980). Second, animals usually reduce theirfluid intake when the only fluid available is alcohol and, as a consequence, their foodintake is also reduced, introducing suboptimal nutrition as a possible confounding factor(Rilely and Meyer, 1984).To overcome these major defects, Testar et al. (1986) used a modified procedurefor oral administration of alcohol. In their study, increasing amounts of ethanol wereintroduced into drinking water during the premating period and 25% ethanol in drinkingwater was offered during gestation in rats. Total energy intake in the alcohol groupdid not differ from that of controls, either before or during gestation, and this balancewas produced by the progressive increment in alcohol derived calories and the correlateddecrease in calories derived from food. During pregnancy, ethanol provided more than30% of the total calories ingested by the rats. At the 21st day of gestation, the maternalblood alcohol concentration was 147 ± l8mg/dl.In the liquid diet method, alcohol is mixed with a liquid diet and the mixture is provided as the animal’s only source of calories. Presumably this method is nontraumatic tothe animal, and is easy to administer. Theoretically, nutritional intake can be controlledand high blood alcohol levels can be achieved and maintained. Typically, in the rat, a dietproviding 35% of the animal’s daily caloric intake in the form of alcohol is administeredduring pregnancy. Blood alcohol levels (BAL) of over lOOmg/dl are achieved during asubstantial portion of the dark cycle when the rats consume the majority of their dailyintake (Driscoll et al., 1990).Disadvantages are that animals in the same group do not all consume the same amountof alcohol and their temporal patterns of intake also differ. Both factors influence bloodChapter 1. Introduction 8alcohol levels and are not under the experimenter’s control. Furthermore, diets that differin nutritional composition might produce different blood alcohol levels even though theycontain the same concentration and percent ethanol derived calories and the animalsconsume about the same daily dose of ethanol. For example, in a study by Wiener etal. (1981) two diets differing in protein content, but containing the same concentrationof ethanol, were consumed in fairly similar amounts by pregnant rats. However, theanimals on the diet with higher levels of protein achieved blood alcohol levels only abouthalf as high as those on the diet lower in protein. Thus, even though animals are self-administering the same daily dose of ethanol, this does not necessarily ensure equal bloodalcohol levels when the nutritional composition of the diet differs. This problem was alsoencountered by Vavrousek-Jakuba et al. (1991). Since blood alcohol level appears to becritical in producing fetal alcohol effects, interpreting data in terms of the daily dose ofethanol consumed may lead to erroneous conclusion.For oral intubation, the alcohol is mixed with a vehicle and administered directly tothe stomach via a tube. The primary advantage of the intubation procedure is that theresearcher has control over the dose of alcohol and time of administration. Each animalreceives the same dose, at the same time, high blood alcohol levels can be obtained, butare transient. In rats, the doses administered are generally 4 to 8g/kg/day. Followingintubation, peak BALs are often well over 200mg/di and rats are obviously intoxicated(Driscoll et al., 1990).Disadvantages of this method are that intubation is probably stressful and involvesdaily handling of the dam, possibly confounding the results. Also, the concentration ofalcohol and volume administered must be considered because of the potentially irritatingproperties of high concentrations of alcohol and the effects of volume on absorption (Rilelyand Meyer, 1984).Again, due to differences in metabolism among species, the BAL, rather than theChapter 1. Introduction 9absolute dose of alcohol, provides a more meaningful estimate of compound bioavailability(Driscoll et al., 1990; Kumar, 1982).The following studies reviewed were based on different methodologies. In Lee andLeichter’s study (1983), Sprague-Dawley rats received 10% ethanol (v/v) in drinkingwater for one week, 20% ethanol in drinking water for another 4 weeks before matingand 30% ethanol in drinking water throughout gestation. The mean serum alcohol levelof these rats was 61 ± 23mg%. They found the litter size and birth weight were lower inthe rats of alcohol group than in the rats of both pair-fed and control groups.In the study by Testar et al. (1986), Wistar rats were given alcohol in the drinkingfluid. Before mating, the rats were given 10% ethanol in drinking fluid for one week, 15%for the second week, 20% for the third, and 25% for the fourth week. After mating, theywere given 25% ethanol throughout gestation. The ethanol calories comprised over 30%of the total energy intake during pregnancy. Ethanol intake during pregnancy was morethan lOg/kg/day. At the 21st day of gestation, rats on ethanol had blood alcohol levelsof 147±l8mg/di. There was a reduction in litter size and fetal weight in the ethanolgroup, but this was also found in pair-fed group. However, the fetal body length wasreduced only in the ethanol group.In studies using a liquid diet as the source of alcohol, ethanol provided 36% of totalcalories in almost every study. In the study by Weinberg et al. (1990), Sprague-Dawleyrats were given liquid diet throughout gestation, the blood alcohol level reached 145 tol9omg/dl. They found a significantly lowered fetal weight in alcohol-exposed fetusesthan in either pair-fed or ad libitum control fetuses.In Abel’s three consecutive studies (Abel and Dintcheff, 1978a; Abel, 1978b; Abel,1979a) alcohol was given by stomach tube throughout gestation. In these studies, fourdifferent doses were used (1.0, 2.0, 4.0, and 6.Og/kg/day of ethanol). The peak bloodalcohol levels produced by 2.0, 4.0, and 6.Og/kg doses were 8Omg/lOOml, 150 ± 15 andChapter 1. Introduction 10267 ± l9mg/lOOmi, respectively. The lower doses of 1.0 and 2.Og/kg, resulted in a reduction in litter size and litter weights in both ethanol treated and pair-fed control groups,but the difference between the ethanol treated and pair-fed control groups was not significant. With the higher doses of 4.0 and 6.Og/kg, litter weight, but not litter size wasreduced in ethanol-treated groups. Also, a dose-related reduction in litter weight at birthwas observed.In summary, most studies found a reduced litter weight and/or litter size in ethanoltreated rats, but no apparent gross malformations were observed.With respect to the findings in placental weight of alcohol treated animals, many researchers found a reduced fetal weight in ethanol treated rats, but an increased placentalweight (Aufrere and Lebourhis, 1987; Weinberg et al., 1990). Others have reported variations in placental mass, but their observations are not all similar. In alcoholic women,an increase in placental mass was noticed by Kaminski et al. (Aufrere and Lebourhis,1987) and in both mice and rats, a similar observation has been reported (Fisher et al.,1985; Gordon et al., 1985; Henderson et aL, 1981; Jones et al., 1981; Leichter and Lee,1984). However, in some cases, no change was reported (Abel, 1979b; Chernoff, 1977;Nelson et al., 1985), whereas in others a decrease in placental weight occurred (Aufrereand Lebourhis, 1987; Greizerstein and Aldrich, 1983; Kennedy, 1984). These discrepancies can be explained by differences in the experimental conditions used, includingBAL, duration of intoxication and stage of pregnancy. Indeed, if intoxication is severeand takes place during the last two-thirds of pregnancy, a decrease in placental weight isobserved (Aufrere and Lebourhis, 1987; Greizerstein and Aldrich, 1983; Kennedy, 1984).Conversely, if intoxication takes place during the first one-third of pregnancy, placentalweight is increased (Fisher et al., 1985; Gordon et al., 1985; Henderson et al., 1981; Joneset al., 1981).Chapter 1. Introduction 111.2 Evidence that diabetes is teratogenicThe first six decades of the insulin era clearly established that most, if not all, of theperinatal complications of pregnancies caused by diabetes are linked to faulty regulationof maternal metabolism and that they can be diminished by rigorous diabetes management (Freinkel, 1980). However, careful retrospective analyses disclosed that the 3- to 6-fold increase in the incidence of birth defects had not been attenuated to any meaningfulextent (Freinkel, 1988).1.2.1 Studies in humansThe association between diabetes mellitus in women and congenital malformations intheir offspring has been suspected since the nineteenth century. Tn 1885, LeCorché reported hydrocephalus in two infants of diabetic mothers. The prognosis for pregnantdiabetic women prior to the discovery of insulin was poor, and few women deliveredsuccessfully (Mills, 1982).It was not until better control of hyperglycemia, close monitoring in the last monthsof pregnancy, and early delivery for fetal distress were instituted that salvage rates indiabetic pregnancies improved substantially. At this point, the full impact of congenitalmalformations was appreciated (Mills, 1982).The definitive clinical observations on the increased incidence of congenital fetalmalformations in diabetic pregnancy were not made until 1964 by Moisted-Pedersen,Tygstrup and Pedersen (1964). These observations were later confirmed in many centers, both retrospectively (Glasgow et al., 1979; Kucera, 1971; Malins, 1979; Pedersen,1977; Soler et al., 1976) and prospectively (Cluing and Myrianthopoules, 1975). Moststudies have shown that there is unquestionably an increased incidence of major congenital abnormalities in the infants of diabetic mothers (1988).Chapter 1. Introduction 12Epidemiological data indicate that the risk of congenital malformations in diabeticpregnancies is three to four times higher than that in the nondiabetic population (Bairdand Aerts, 1987; Eriksson, 1984a; Kitzmiller et al., 1978; Kucera, 1971; Mills, 1982;Moisted-Pedersen, 1980; Pedersen, 1979; Simpson et al., 1983; Soler et al., 1976), andthe malformations affect all fetal organ systems (Kucera, 1971).Diabetes in pregnancy is associated with a number of changes in embryo-fetal development such as altered growth rate and maturation of several organ systems and anincreased rate of congenital malformation (Eriksson et al., 1989a).Most studies show a generalized increase in malformations involving multiple organsystems. These include cardiovascular, genitourinary, musculoskeletal, and other malformations (Mills, 1982). Nevertheless, no single study has contained enough cases todetermine whether or not the risk for each specific defect is significantly increased indiabetic pregnancies compared to the nondiabetic population.Kucera (1971) tried to estimate relative risks using data from available literature forthe 20 year period from 1945 through 1965. He reviewed data on 7,101 fetuses of diabeticwomen, of which 340 fetuses showed anomalies (4.79%). This series was then comparedwith a normal “control” group obtained from WHO data with a total of 431,764 fetuses.This control group had 7,124 malformations (0.65%). He found a higher incidence ofanomalies in the offspring of diabetic women. Spinal anomalies, situs viscerum inversus, gross skeletal anomalies, pseudohermaphroditism, urological anomalies, and heartanomalies are all significantly more frequent in fetuses of diabetic pregnancies. Onesyndrome which seems particularly strongly associated with diabetes is the caudal regression syndrome. Caudal regression is a condition in which agenesis or hypoplasia ofthe femorae occurs in conjunction with agenesis of the lower vertebrae. It is very rare inthe general population (Mills, 1982). Kucera (1971) found 9 cases with this syndrome in7,101 infants of diabetic mothers.Chapter 3. Results 37....20013-/15001001 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20gestational day— group 1 + group 2 group 3 —s-- groupFigure 3.3: Daily mean water intake during pregnancyChapter 3. Results 38Table 3.4: Breeding and pregnancy rates of animals in each treatment groupGroup 1 2 3 4No of rats initially 9 10 18 13No of rats that bred 9 10 16 13Breeding rate (%) 100 93.5No of rats with live fetuses 9 8 111) 8CPregnancy rate (%) 100 90 68.8 61.5No of rats excluded from study 0 2 7 5Group 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcoholBreeding rate and pregnancy rate were not significantly differentamong groupsExcluded rats:a Group 2: 1 not pregnant and 1 had one resorption but nolive fetus on day 21bGroup 3: 2 not bred, 5 not pregnant on day 21cGroup 4: 5 not pregnant on day 21interactions were observed, p<O.Ol).3.2 Reproductive performanceThere was no apparent difference between diabetic and nondiabetic rats in the proportionof animals with positive breeding (number of vaginal plugs). However, not all rats whichhad plugs subsequently produced litters. There was a trend that more diabetic ratswith plugs failed to produce litters than was the case for nondiabetic animals, althoughthe difference among groups was not statistically significant. Only rats containing livefetuses at autopsy were included in this study and subjected to subsequent examinations(table 3.4).Chapter 3. Results 393.3 Body weight and weight gain of dams during pregnancyThe average body weights of dams on days 1, 6, 12, and 21 of gestation, as well asthe average body weight gains from day 1 to day 5, 6 to day 11, and 12 to day 20 ofgestation for each group are shown in table 3.5 and Figure 3.4. The mean body weightsof the 4 experimental groups were not statistically different on day 1 of gestation. Duringgestation, diabetes, but not alcohol exposure, played the main role in the differences inbody weight. Diabetic rats, whether consuming alcohol or water, had significantly lowerbody weights on days 6, 12, and 21 of gestation compared to nondiabetic rats (p<O.Ol).On day 21, diabetic rats weighed about 90g less than nondiabetic rats. No interactionbetween diabetes and alcohol exposure on body weight was observed on days 6, 12, or21 of gestation.From day 1 to day 5 of gestation, when alcohol was not administered, diabetic rats(groups 3 & 4) gained less weight than nondiabetic rats (groups 1 & 2, p<O.O5). Fromday 6 to day 11 of gestation, when rats in groups 2 and 4 consumed alcohol, diabetic ratsgained significantly less weight than nondiabetic rats (p<O.OOl), but alcohol administration did not have a significant effect on body weight gain. It should be noted that therewere huge variations of weight gains from day 6 to day 11. in groups 3 and 4 (table 3.5).In fact, 3 out of 11 rats in group 3 and 3 out of 8 rats in group 4 lost weight instead ofgaining weight during this period of gestation. The reason for losing weight in these ratsis not clear. It cannot be attributed to the administration of alcohol, because some of thegroup 3 rats (no alcohol administered) also lost weight during this period. Interestingly,from day 12 to day 20 of gestation, in addition to the diabetic effect on body weight gain(p<O.OO1), there was also a significant interaction between diabetes and alcohol (2-wayANOVA, p<O.O5). Alcohol alone did not have a significant effect on body weight gain.Because the overall 2-way ANOVA showed a significant interaction between diabetesChapter 3. Results 40Table 3.5: Body weight and weight gain of dams during gestation in each treatmentgroup (mean ± SD)Group 1 2 3 4No of animals 9 8 11 8Body wt on D1(g) 253.6 ± 11.8 256.0 + 8.8 246.6 ± 12.4 249.7 ± 12.3Body wt on D6(g)* 275.1 ± 9.3 279.2 ± 6.5 264.1 + 16.9 260.9 ± 16.5Body wt on D12(g)* 305.7 ± 13.3 293.7 + 7.1 271.5 + 26.0 266.2 ± 16.2Body wt on D21(g)* 382.3 ± 28.5 387.9 ± 18.0 309.2 ± 38.2 284.9 ± 25.0Wt gain D1-5(g)* 21.5 ± 5.9 23.2 ± 5.6 17.5 ± 12.1 11.2 ± 10.2Wt gain D6-11(g)* 30.6 ± 7.8 14.5 ± 8.1 5.1 ± 19.4 5.2 ± 13.5Wt gain D12-20(g)*f 76.7 ± 20.0 94.2 ± 17J 37.7 ± 23•6b 18.7 ± 195bGroup 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcohol* Diabetic significantly different from non-diabetic, p<O.O5t Significant interaction between diabetes and alcohol, p<O.O5Values not sharing a common superscript in each row are significantly differentat p<O.O5Chapter 3. Results 41400-+360-360--340320 //300-280-FA11 6 12 21Days of Gestation• Group 1 + Group 2 ---E--- Group 3 ----- Group 4Figure 3.4: Maternal body weight changes during gestational period in 4 groupsChapter 3. Results 42and alcohol on body weight gain from day 12 to day 20 of gestation, the Tests of SimpleEffect were employed to identify the source of the difference. In rats not exposed toalcohol (groups 1 & 3), diabetic rats gained less weight during this time than did non-diabetic rats (p<O.0Ol). Similarly, in alcohol exposed rats (groups 2 & 4), diabetic ratsgained less weight than did nondiabetic rats (p.<O.OOl). On the other hand, nondiabeticrats, whether exposed to alcohol or not, had similar body weight gains. In diabetic rats,although there was a trend for alcohol exposed animals to gain less weight than animalsnot exposed to alcohol, the difference was not statistically significant.This pattern of interaction between diabetes and alcohol on body weight gain fromday 12 to 20 is shown in Figures 3.5 and 3.6. Figure 3.5 indicates that the difference inbody weight gain from day 12 to 20 between diabetic and nondiabetic rats was greater inalcohol exposed group than in non-alcohol exposed group (p<0.O5). Figure 3.6 shows atrend that the direction of alcohol effect on weight gain from day 12 to 20 in nondiabeticrats is opposite to that in diabetic rats. This means when rats are not diabetic, alcoholexposure tends to facilitate weight gain compared to controls and contrarily, whereaswhen rats are diabetic, alcohol exposure tends to result in less weight gain compared todiabetic rats not exposed to alcohol. However, the differences of weight gains from day12 to 20 between alcohol and water exposed groups, either in nondiabetic rats or diabeticrats, were not significant.3.4 Blood glucose concentrationsIn groups 1 and 2 (non-diabetic), blood glucose concentrations on day 21 of gestation didnot differ from those on day 1, confirming the continued nondiabetic state throughoutgestation. In groups 3 and 4, blood glucose concentrations by one week after STZ administration were significantly higher than those before STZ administration, which wereChapter 3. Results 43120100water aicoho4groupI —a— nondiabetic —.— diabeticFigure 3.5: Interaction of diabetes at alcohol on maternal weight gain during D12-21I100nondia diabeticgroup—B— water—.-- alcohol 1Figure 3.6: Interaction of alcohol at diabetes on maternal weight gain during D12-21(Reference: Kimmel, H.D. Simple and factorial experiments. In: Experimental principlesand design in psychology. New York. The Ronald Press Co. pp 153-177, 1985)Chapter 3. Results 44Table 3.6: Blood glucose concentration of dams at different times during the experiment,in each treatment group (mmol/L)Rat Group 1 Group 2 Group 3 group 4No Dl D21 Dl D21 Bf A D2 D7 D12 D21 Bf A D2 D7 D12 D211 3.9 4.8 4.7 2.9 3.6 18.4 16.9 23.6 23.3 21.1 4.6 21.3 26.6 >27 16.0 >272 3.7 5.9 3.6 4.1 4.9 15.2 17.9 21.8 18.1 21.9 4.7 >27 >27 16.9 18.0 16.43 3.8 3.8 4.8 7.6 2.8 18.3 22.5 16.7 19.3 26.6 4.3 22.0 18.4 22.3 17.5 18.84 4.8 3.5 3.4 4.1 5.7 >27 24.8 19.2 16.9 >27 4.9 25.0 >27 23.1 21.0 17.35 2.8 2.4 2.9 3.9 4.3 21.9 19.9 22.3 24.3 15.0 5.1 >27 >27 >27 18.7 24.06 4.3 7.8 2.9 3.7 4.7 23.1 17.8 21.2 18.8 22.0 5.0 16.6 25.0 17.9 17.3 19.97 6.8 4.0 3.7 3.7 2.9 19.0 26.4 18.5 15.7 22.5 4.3 25.2 18.6 24.3 20.1 25.58 3.7 6.2 4.5 3.3 4.4 16.7 25.2 17.4 23.1 22.5 3.7 >27 >27 25.2 21.0 21.39 6.8 4.4 5.5 15.8 26.1 17.7 18.0 22.010 3.8 18.8 19.6 20.0 18.4 >2711 7.2 23.4 -- 22.0 21.7Group 1: Nondiabetic on waterGroup 2: Nondiabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcoholf Blood glucose level before STZ injection1 Blood glucose level one week after STZ injectionnot different from those of nondiabetic rats. After STZ administration, blood glucoseconcentrations of all rats in groups 3 and 4 were above l5mmol/L and remained abovethis level throughout gestation, confirming the continued diabetic state (table 3.6).3.5 Peak blood alcohol levelsOn day 11 of gestation (the last day of alcohol administration to groups 2 and 4), themean peak blood alcohol concentration was 73.3 ± 13.6mg/di.3.6 Outcome of pregnancyTable 3.7 shows the outcome of pregnancy by treatment group. The mean number ofimplantation sites per dam and mean number of live fetuses per litter did not differ between diabetic and nondiabetic rats or between alcohol exposed and non-alcohol exposedChapter 3. Results 45Table 3.7: Reproductive variables on day 21 of gestation in each treatment groupGroupNo of rats examinedTotal No of implantation sitesNo of implantation sites/litterTotal No of live fetusesNo of live fetuses/litterTotal No of resorption sites% resorptionsMean litter arcsineNo of resorptions/litterNo of litters with resorptions% litters with resorptionsGroup 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcohol* mean ± SDNo means or rates were significantly different among groupsrats. The number of resorptions expressed either as the mean number of resorptions perlitter or as the percentage of resorptions in each group was not significantly differentamong the four treatment groups (both chi-square test and 2-way ANOVA). There wereno differences among treatment groups with respect to number of litters with one or moreresorptions.3.7 Fetal weight, placental weight and fetal weight/placental weight ratioThe mean fetal weight, mean placental weight and the mean ratio of fetal weight/placentalweight (f/p ratio) for each treatment group are shown in table 3.8. Photo 1 (see AppendixA) shows the relative sizes of two fetuses from a control dam (right) and a diabetic-alcoholdam (left) respectively.1 2 3 49 8 11 8128 117 131 11114.2 ± 4.3* 14.6 ± 3.8 11.9 ± 4.3 13.9 ± 3.8118 108 112 10013.1 ± 4.2 13.5 ± 3.8 10.2 ± 3.8 12.5 ± 3.59 7 16 117.0 6.0 12.2 9.915.9 ± 7.7 14.8 ± 7.6 21.1 ± 8.9 17.9 ± 9.51.0±1.1 0.9±1.0 1.5±1.7 1.4±1.55 4 8 555.6 50.0 72.7 62.5Chapter 3. Results 46Table 3.8: Fetal and placental weights on day 21 of gestation in each treatment group(mean ± SD)Group 1 2 3 4No of animals 9 8 11 8Mean fetal weight/litter (g)*f 3.38 ± 0.63 3.79 ± 0.30a 2.56 ±0•49b 2.27 ± 0.41bMean placental weight/litter (g)* 0.50 ± 0.07 0.45 ± 0.04 0.64 ± 0.18 0.54 ± 0.10Mean f/p ratio/litter*f 6.83 + 1.52a 8.42 ± 0.94b 4.14 ± 0.89c 4.29 ± 0.60cGroup 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcoholf/p ratio: ratio of fetal weight/placental weight* Diabetic significantly different from non-diabetic, p<O.05Alcohol significantly different from water, p<O.05t Significant interaction between diabetes and alcohol, p<0.O5Values not sharing a common superscript in each row are significantly different at p<O.OSThe mean fetal weight was significantly less (p<O.OO1) in litters of diabetic rats compared to nondiabetic animals, but a significant difference was not seen between alcoholand water exposed groups. However, there was a significant interaction between diabetesand alcohol on fetal weight (2-way ANOVA, p<0.O5).Because the overall two-way ANOVA indicated a significant interaction between diabetes and alcohol exposure on fetal weight, the Tests of Simple Effect were carried out.In the non-alcohol exposed group, fetuses of diabetic dams weighed significantly less thandid fetuses of nondiabetic dams (p<O.OO1). Similarly, in alcohol exposed groups, diabeteshad a very strong effect on fetal weight (p<O.OO1). As a result, fetuses of diabetic damswere lighter than fetuses of nondiabetic dams. However, in both diabetic and nondiabetic groups, the differences in fetal weight with or without alcohol treatment were notsignificant.Chapter 3. Results 47The pattern of interaction between diabetes and alcohol on fetal weight is demonstrated in Figures 3.7 and 3.8. Figure 3.7 indicates that the difference in fetal weightbetween diabetic and noridiabetic rats was greater in alcohol exposed rats than in rats notexposed to alcohol (p<O.O5). Figure 3.8 shows the directions of alcohol’s effect on fetalweight in diabetic and nondiabetic rats. However, the difference of fetal weight betweenalcohol and water exposed rats was not significant either in diabetic or nondiabetic ratsas demonstrated in Tests of Simple Effect.Placental weights were significantly different (p<O.Ol) when diabetic and nondiabeticrats were compared, with diabetic rats tending to have bigger placentas than nondiabeticrats. However, a significant difference in placental weight between alcohol and waterexposed groups was not found.In terms of the ratio of fetal weight/placental weight (f/p ratio), 2-way ANOVAshowed that both diabetes and alcohol had very strong effects (p<O.OO1 and p<O.O5respectively) on the f/p ratio and there was also an interaction between diabetes andalcohol (p<O.O5).The subsequent Tests of Simple Effect gave the following results for comparisonsbetween different pairs of groups. In non-alcohol exposed rats, the f/p ratio of diabeticrats was significantly lower than was that of nondiabetic rats (p<O.OO1). Similarly, inalcohol exposed rats, diabetic rats had lower f/p ratio compared to nondiabetic rats(p<O.OOl). In nondiabetic rats, alcohol had a significant effect on f/p ratio (p<O.Ol),but the direction is interesting. Alcohol exposed rats had a higher f/p ratio than didnon-alcohol exposed rats. In contrast, alcohol had no effect on f/p ratio in diabetic rats.As a result, all diabetic rats had a similar f/p ratio.The pattern of interaction between diabetes and alcohol on f/p ratio is shown in Figures 3.9 and 3.10. Figure 3.9 shows again that the difference in f/p ratio between diabeticand nondiabetic rats was greater in alcohol exposed rats than in non-alcohol exposed ratsChapter 3. Results 4854.Swater alc,holgroupI —s— nondiabetic -.- diabetic 1Figure 3.7: Interaction of diabetes at alcohol on fetal weight45.4....nordia diaeticgroup—s-- water -—•-• alcohol 1Figure 3.8: Interaction of alcohol at diabetes on fetal weightChapter 3. Results 49(p<O.O5). Figure 3.10 demonstrates that in nondiabetic rats , alcohol exposure raisedthe f/p ratio, however, in diabetic rats, alcohol exposure did not show any effect on f/pratio.3.8 External malformationsA wide variety of external malformations were found in diabetic rats, but none werefound in nondiabetic rats regardless of whether the animals were exposed to alcohol orwater (table 3.9). The malformations observed include: a variety of facial defects, characterized by smaller lower jaw, no lower jaw, no mouth, tongue protruding, otocephaly(see Appendix A, photo 2), and distorted face; missing tail (see Appendix A, photo 3);umbilical hernia or gastroschisis (see Appendix A, photo 4); severe edema; and ectrodactyly. Some affected fetuses had more than one malformation, such as severe edematogether with absence of the lower jaw.Although both groups of diabetic animals (those on water and those given alcohol)had malformed fetuses, those on alcohol had a greater percentage of fetuses with all typesof external malformations compared to those on water (chi-square test, p<O.05), with 5malformed fetuses out of 112 (4.5%) occurring in group 3, and 12 out of 100 (12%) ingroup 4.In terms of the number of litters with malformed fetuses, the difference betweengroups 3 and 4 was not significant, although there was a trend for diabetic animals onalcohol to have more affected litters (27.3% in group 3 vs 75% in group 4). The means ofthe litter frequencies of external malformations (transformed to Freeman-Tukey arcsines)were not different between groups 3 and 4 (t-test, p>O.O5).Chapter 3. Results 50100.5.4.3.water aJCOhO1groupnondjabetjc —*- diabeticFigure 3.9: Interaction of diabetes at alcohol on f/p ratio109nondia diabeticgroup—s-- water -.— alcoholFigure 3.10: Interaction of alcohol at diabetes on f/p ratioChapter 3. Results 51Table 3.9: Numbers of fetuses with external malformations in each treatment groupGroup 1 2 3 4No of litters examined 9 8 11 8No of fetuses examined 118 108 112 100No of fetuses withfacial defects 0 0 3 8absent tail 0 0 1 1umbilical hernia/gastroschisis 0 0 0 2edema 0 0 1 2ectrodactyly 0 0 1 1No of fetuses with external malformations 0 0 5 12Malformed fetuses as % of live fetuses 0 0 4.5Mean litter arcsine of malformed fetuses - - 13.28 ± 9.27* 17.27 ± 8.79No of litters with malformed fetuses 0/9 0/8 3/11 6/8% of litters with malformed fetuses 0 0 27.3 75Group 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcoholA single fetus may have more than one malformationsValues not sharing a common superscript in each row are significantly differentat p<O.O5* mean ± SDChapter 3. Results 52Table 3.10: Numbers of fetuses with skeletal malformations in each treatment groupGroup 1 2 3 4No of litters examined 9 8 11 8No of fetuses examined* 103 84 101 79No of fetuses withmalformed sternum 0 0 7 2malformed ribs 0 0 1 2malformed vertebrae 0 0 3 3malformed skull 0 0 0 1malformed limb 0 0 0 1No of fetuses with skeletal malformations 0 0 11 9% of malformed fetuses 0 0 10.9 11.4No of litters with malformed fetuses 0/9 0/8 5/11 4/8% of litters with malformed fetuses 0 0 45.5 50Group 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcohol* Fetuses with external malformations and fetuses which were dried outduring fixation in alcohol were excluded from skeletal examination3.9 Skeletal malformationsFetuses with external malformations were excluded from skeletal examination. All fetuseswithout gross malformations were stained with alizarin red S and alcian blue 8GS andskeletal morphology was examined.Skeletal malformations were not seen in fetuses of nondiabetic dams, whether onwater or alcohol. However, a variety of skeletal malformations were found in diabeticanimals on water and in diabetic animals on alcohol (table 3.10).Malformed sternum was found relatively frequently among fetuses with skeletal malformations. In the 9 fetuses with malformed sternums this was expressed as cleft sternum(see Appendix A, photo 5).Chapter 3. Results 53Three fetuses showed malformed ribs. One fetus, from a diabetic dam on alcohol,had the 1st left rib bifurcated and fused with the 7th vertebrae. Another fetus, from thesame group, had fusion of the 7th and 8th ribs. The third fetus, from a diabetic animalon water, had an extra right rib fused distally with the 1st rib.In the 6 fetuses with vertebral malformations, the defects occurred at different levels,i.e. cervical, thoracic or lumbar vertebrae. Two fetuses from the diabetic-alcohol grouphad abnormal cervical vertebrae. In one, there were 5 cervical arches on the left with the4th arch in a vertical position, and 6 arches on the right; while in the other fetus, twoextra pieces of cartilage were seen on top of the 5th and 6th cervical arches and crossedeach other. The other fetus, from the diabetic-water group had the 5th left cervical archbifurcated (see Appendix A, photo 5). In 2 fetuses with lumbar vertebral malformations,one, from diabetic-alcohol group, had an extra lumbar vertebrae, while the other, fromthe diabetic-water group, had more severe malformation. In this case the 1st and 2ndlumbar vertebral right arches fused, and the 1st and 2nd lumbar vertebral bodies wereossified unilaterally and the 3rd and 4th vertebral bodies ossified unilaterally and fusedtogether. Only one fetus, which was from the diabetic-water group, had thoracic vertebralmalformation. This fetus had hemivertebrae, the 1st thoracic right and the 2nd thoracicvertebrae asymmetrically displaced and the 1st right rib missing.In addition, one fetus from the diabetic-alcohol group had a malformed arm joint,while another from the same group had a missing zygomatic process and zygoma bonein the skull (see Appendix A, photo 6).Chapter 3. Results 543.10 Skeletal ossification and variantsFour entire fetal skeletons, one from each treatment group, are shown in photo 7 (seeAppendix A). A total of 367 fetuses were examined for skeletal ossification. The ossification centers in five districts including metacarpus, metatarsus, sternum, vertebrae andskull were counted in each fetus and the irregular shapes of the ossification centers werealso recorded.3.10.1 Metacarpus, metatarsusNo ossification centers were observed in the anterior and posterior phalanges of fetusesin any group.Table 3.11 shows the frequency of ossification centers in the fetuses of each group, according to the number of ossification centers, as well as the average number of ossificationcenters. For the metacarpus (both left and right sides), all fetuses in groups 1 and 2 hadeither 3 or 4 ossification centers, while most fetuses in groups 3 and 4 had a maximumof 3 ossification centers. Some fetuses in groups 3 and 4 had no metacarpal ossificationcenters, while others had only 1 or 2. Only a few showed 4 ossification centers. As aresult, the average number of ossification centers in the metacarpus (both sides) weresignificantly less in the diabetic groups than in the nondiabetic groups (p<O.OOl). However, there was no significant difference in the average number of ossification centers inmetacarpus when fetuses of alcohol exposed rats are compared to those of water exposedanimals.For the metatarsus, almost all fetuses in groups 1 and 2 had 4 ossification centers onboth sides. In groups 3 and 4, although large percentage of fetuses had 4 centers, quite afew had 3 centers and some did not show any ossification centers. The average numberof ossification centers in the metatarsus was significantly less in groups 3 and 4 comparedChapter 3. Results 55Table 3.11: Development of ossification centers in metacarpus, metatarsus, and sternumin each treatment groupSkeletal No. of ossification Group 1 Group 2 Group 3 Group 4district centraMetacarpus(L) 0 6 11 0 32 5 63 65 28 78 694 38 56 12Ave.* 3.43 ± 0.43f 3.68 + 0.39 2.91 ± 0.58 2.80 ± 0.39Metacarpus(R) 0 5 01 0 32 6 73 64 26 79 694 39 58 11Ave.* 3.46 ± 0.46 3.74 + 0.35 2.94 ± 0.57 2.83 ± 0.35Metatarsus(L) 0 7 81 0 02 0 13 1 28 304 97 84 66 405 5Ave.* 4.04 ± 0.14 4.00 ± 0.00 3.49 ± 0.83 3.16 ± 1.03Metatarsus(R) 0 7 61 0 02 0 33 1 25 314 97 84 69 495 5Ave.* 4.04 ± 0.14 4.00 ± 0.00 3.53 ± 0.79 3.20 ± 0.96Sternum 0 9 91 3 22 7 23 2 12 154 39 7 14 265 18 17 31 146 44 60 20 9Ave.* 5.12 ± 0.76 5.67 + 0.40 3.92 + 1.85 3.64 ± 1.69Total fetuses 103 84 101 79examinedGroup 1: Non-diabetic on water mean ± SDGroup 2: Non-diabetic on alcohol * Diabetic significantly differentGroup 3: Diabetic on water from non-diabetic, p<O.O5Group 4: Diabetic on alcoholChapter 3. Results 56with groups 1 and 2 (p<O.Ol). Alcohol exposure alone did not have a significant effecton the average number of metatarsal ossification centers.3.10.2 SternumPhoto 8 (see Appendix A) shows the normal sternebral ossification centers in a fetusfrom a control litter. In groups 1 and 2, almost all fetuses had 4 to 6 ossification centers.However, in groups 3 and 4, fetuses had numbers of ossification centers varying fromnone to 6. There were a large number of fetuses with less than 4 ossification centers inthe sternum (table 3.11).The average number of ossification centers in the sternum was smaller in fetuses fromdiabetic rats than in fetuses from nondiabetic rats (p<O.O5), irrespective of whether alcohol treated or not. However, the ossified sternebral centers were not all of a normal shape.The irregular shapes of sternebral ossification centers that were observed are shown inFigure 3.11 . These include dumbbell-shaped, asymmetrically dumbbell-shaped, asymmetrical and asymmetrically dumbbell-shaped, simple asymmetrical, cleaved (bipartite),asymmetrically cleaved , unilaterally ossified. The lack of apposition refers to those asymmetrical ossification centers in cases 4, 5 and 9 in Figure 3.11 . The irregular shapes areconsidered skeletal variants rather than malformations (table 3.12).The overall chi-square test showed that the difference in the frequency of dumbbell-shaped, cleaved, and unilaterally ossified centers (these shapes are referred as “poorlyossified centers” in the table) among groups was significant (p<O.OO1). The subsequentpaired comparisons indicated that groups 1 and 2 did not differ from each other withrespect to the percentage of fetuses with poorly ossified sternebral centers (24.3% ingroup 1 and 20.2% in group 2). Groups 3 and 4 were significantly different from groups1 and 2 (p<O.001), with a larger proportion of fetuses with poorly ossified centers ingroups 3 and 4 (74.2% in group 3 and 82.3% in group 4), compared to groups 1 and 2.Chapter 3. Resultsppp-pcleaved (bipartite)asymmetrically cleavedunilaterally ossifiednot ossified57I I normaldumbbell-shapedasymmetrically dumbbell-shapedasymmetrical and asymmetrically dumbbell-shapedasymmetricalasymmetrically shapedFigure 3.11: Observed irregular shapes of sternebral ossification centersChapter 3. Results 58Table 3.12: Skeletal variants observed in fetuses of each treatment groupGroup 1 2 3 4No of fetuses examined 103 84 101 79An extra rib* oa 1a 34b 211(0) (1.2) (33.7) (26.6)fThe 13th rib shortj oa 6b oa oa(0) (7.1) (0) (0)Incomplete ossification of 4a 71b 68Csupraoccipital bone (3.9) (2.4) (70.3) (86.1)Lack of foramen 5a 5a 14a 3a(4.9) (6.0) (13.9) (3.8)Poorly ossified 25 7a 75b 65bsternebral centers (24.3) (20.2) (74.2) (82.3)Lack of apposition 13a 12a 33b 34bin sternebral centers (12.6) (14.3) (32.7) (43.0)Poorly ossified 7a 63Cthoracic centers (6.8) (4.8) (59.4) (79.7)Poorly ossified oa 19blumbar centers (0.97) (0) (18.8) (24.0)Poorly ossified 0 0 44 49”cervical arches -- (43.6) (62.0)Group 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcoholPoorly ossified centers include those dumbbell-shaped, bipartite, and unilateralossified centers* Including cervical and lumbar extra ribs, and full or rudimentary sizeIncluding short or rudimentary sizesf PercentageValues not sharing a common superscript in each row are significantly differentChapter 3. Results 59Table 3.13: Mean numbers of poorly ossified centers in sternum, thoracic and lumbarvertebrae in each group (mean + SD)1 2 3 4GroupSternum*Thoracic*Lumbar*Group 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcoholMeans calculated among those fetuses with one or more poorlyossified centers* Diabetic significantly different from non-diabetic, p<O.O51.28 ± 0.54 1.06 ± 0.24 2.01 ± 1.07 2.17 ± 0.931.43 ± 1.13 1.00 ± 0 3.52 ± 2.03 3.81 ± 2.190 0 1.42 ± 0.77 2.00 ± 1.33The difference between groups 3 and 4 was not significant.It is noteworthy that many fetuses had more than one poorly ossified center. Theaverage numbers of poorly ossified centers per fetus in groups 1, 2, 3, and 4 were 1.28,1.06, 2.01, and 2.17 respectively (table 3.13). The calculation of the average numberof poorly ossified centers per fetus of each group did not include those fetuses with nopoorly ossified centers. A significant difference in the average numbers of poorly ossifiedsternebral centers per fetus was only found between diabetic and nondiabetic groups(p.<O.OO1), but not between alcohol and water treated groups. Not only were there morefetuses with poorly ossified sternebral centers, but there were also more poorly ossifiedcenters per fetus in diabetic than in nondiabetic litters.The number of fetuses showing sternebral ossification centers with a lack of appositionwas significantly different among all groups (p.<O.OO1). Similarly, significant differenceswere found between groups 1 and 3 (p=O.OO1), groups 1 and 4 (p<O.OO1) and groups 2and 4 (p<O.OO1), but not between groups 1 and 2 or groups 3 and 4. The percentages offetuses showing ossification centers with a lack of apposition were 12.6%, 14.3%, 32.7%Chapter 3. Results 60and 43.0% in group 1, group 2, group 3, and group 4 respectively (table 3.12).3.10.3 VertebraeIn rats, during development in utero, the ossification centers of the first cervical to the2nd thoracic vertebral arches appear early, then ossification proceeds caudally. For thevertebral bodies, the earlier ossification centers are the fourth thoracic to sixth lumbar,then ossification extends both cephalically and caudally (Strong, 1926). Photo 9 (seeAppendix A) shows the normal vertebral ossification centers and cervical arches in afetus from a control litter.The frequencies of fetuses by position of the most caudal vertebral arches in all 4groups are shown in table 3.14. The results for left and right sides were similar. Theoverall difference among groups was significant (p<O.OO1)The subsequent paired comparisons indicated that vertebral arches proceeded furthercaudally in fetuses of group 2 than in those of group 1 (p<O.OO1). In the control group,about 30% of fetuses had the vertebral arches ending at Ca2, about 40% of fetuses atCal, and the rest at a higher position. In contrast, over 50% of fetuses in group 2 hadthe vertebral arches ending at Ca2, with the rest at Cal.On the other hand, in groups 3 and 4, vertebral arches proceeded less caudally thanin group 1 (P<0.00l). About 50% of fetuses in group 3 and 60% in group 4 had thevertebral arches ending at the level from S4 to 51, while only 30% of fetuses in group 1had the vertebral arches ending at the level from S4 to S2, with more than 80% at S4.Again, the difference between groups 3 and 4 was not significant.In both groups 3 and 4, poorly ossified cervical arches were found frequently, but thiswas not the case in groups 1 and 2. These poorly ossified cervical arches showed either ascleaved ossification centers or unilaterally ossified centers (see Figure 3.12). When thishappened, it occurred more frequently in the 3rd to 7th cervical arches. Although thisChapter 3. Results 61Table 3.14: Frequencies of fetuses according to the most cephalic and most caudal vertebral ossification centers and the most caudal vertebral archesGroup 1 2 3 4The most cephalic vertebral centrumHigher than C7- 12C7 23 41 - -Ti 73 31 35 22T2 7 - 38 33T3- 13 15T4- 8 6Lower than T4- 6 3The most caudal vertebral centrumCa6 4 5Ca5 11 24 1Ca4 28 40 26 8Ca3 45 15 32 32Ca2 13 - 8 iiCal 2 - 8 3Higher than Cal- 25 25The most caudal vertebral arch Left (Right)Ca2 33(33) 48(48) 8(4) 3(2)Cal 40(39) 35(36) 46(47) 28(27)S4 25(25) 1(0) 13(16) 14(17)S3 2(4) - 11(9) 9(8)S2 3(2) - 15(18) 15(15)Si- 7(6) 9(9)L6 of lower than L6 - 1(1) 1(1)Group 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcoholChapter 3. Results 62phenomenon existed in both groups 3 and 4, it happened significantly more frequentlyin group 4 (62%) than in group 3 (43.6%) (p<O.O5).Table 3.14 also shows the frequencies of fetuses according to the most cephalic andmost caudal vertebral ossification centers. The frequencies of fetuses with respect to thepositions of the most cephalic vertebral bodies were significantly different among groups(p<O.OO1). The paired comparisons indicated that group 2 was significantly differentfrom group 1 (p <0.001). The majority of fetuses from group 2 had the ossified vertebralbodies ending at the 7th cervical vertebrae or higher, and the remainder had it at the 1stthoracic vertebrae, with no fetuses showing the ossification centers lower than this. Incomparison, only a few fetuses from group 1 had the ossified vertebral bodies ending atC7 and most of fetuses had them at Ti, with some fetuses at T2. Therefore, ossificationof the vertebral centra has proceeded further cephalically in group 2 fetuses (nondiabeticalcohol) than in group i (the control).In contrast, fetuses of both groups 3 and 4 had vertebral centra which extended lesscephalically than did fetuses from controls (p<O.OOl). In controls, over 90% of fetuseshad ossified vertebral centers ending at Ti, with only a few below that level, but in groups3 and 4, 1/3 or less had ossified vertebral centers ending at Ti, with the remainder belowthat level. However, groups 3 and 4 did not differ significantly from each other.Similarly, the frequencies of fetuses according to the positions of the most caudalvertebral centers were significantly different among groups (p<O.OO1). The paired comparisons indicated again that positions of caudal end vertebral centers proceeded furtherin group 2 than in group 1 (p.<0.OOi). In groups 3 and 4 the caudal end of vertebralcenter had proceeded less than in group 1 (p.<O.OO1). The difference between groups 3and 4 was not significant.In summary, ossification of the vertebral centra extended further, both cephalicallyand caudally, in fetuses of nondiabetic alcohol-exposed rats compared to other treatmentChapter 3. Results 63o 0 o 0rc ‘normal •j cleavedossification centers • •ossification centersof cervical arches . —.a•s.. •...•-o e.0.0_b•7000_b(b).Qe0b.,,ocx70 0t:i000 0 9o 00 :[ unilaterallyo o•g a ossified centersaOe•Oe o”0000o(c)(a)Figure 3.12: Ossification centers of cervical arches(a) normal ossification centers(b)(c) poor ossification centersChapter 3. Results 64groups. However, all fetuses from diabetic rats had vertebral centra which were extendedless, both cephalically and caudally, compared to fetuses of controls, although group 3(diabetic-water) did not differ from group 4 (diabetic-alcohol).The most commonly encountered irregular shapes of vertebral ossification centra areshown in Figure 3.13 . These include asymmetrical, dumbbell-shaped, asymmetricallydumbbell-shaped, asymmetrical and asymmetrically dumbbell-shaped, cleaved (bipartite), cleaved and asymmetrically ossified, unilaterally ossified, and unossified. Theseirregular shaped ossification centers are referred as “poorly ossified centers” in the text.With respect to unossified vertebral centers, only those in the range between the mostcephalic and the most caudal ossified centers are classified as unossified, and these areconsidered “poorly ossified”. Photo 10 (see Appendix A) shows a fetus, from a diabeticrat exposed to alcohol, with poorly ossified vertebral ossification centers.The proportions of fetuses with poorly ossified thoracic vertebral centers were significantly different among groups (p.<O.OOl). The paired comparisons indicated that thedifference between groups 1 and 2 was not significant. However, both groups 3 and 4were significantly different from controls (p<O.OOl). In groups 3 and 4, 59.4% fetuses and79.7% fetuses respectively had poorly ossified thoracic centra, while in group 1 a muchsmaller percentage of fetuses had poorly ossified thoracic centra (only 6.8%). It shouldbe noted that, the difference between groups 3 and 4 was significant (p<O.OOS), withgroup 4 showing 20.3% more fetuses had poorly ossified thoracic centra than in group 3(table 3.12).The average number of poorly ossified thoracic vertebral centra was 1.43 in the 7fetuses with poorly ossified vertebral centra in group 1; 1.00 in the 4 fetuses of group 2;3.35 in the 60 fetuses of group 3; and 3.81 in the 63 fetuses of group 4. The average numberof poorly ossified thoracic vertebral centers per fetus was significantly different betweendiabetic and nondiabetic groups (p<O.OO1), but not significantly different between alcoholChapter 3. Results 65normalasymmetricalasymmetricallY dumbbefl5haPeddumbbell-shaped—“ )C%%asymmetrical andasymmetrically dumbbell-shapedcleaved(bipartite)asymmetrically ossifiedC31 D%unilaterally ossifiednot ossifiedFigure 3.13: Observed irregular shapes of vertebral ossification centersChapter 3. Results 66and water treated groups (table 3.14). Thus, not only did more fetuses have poorlyossified thoracic vertebral centra in diabetic groups, but also they had a larger numberof poorly ossified centers in each fetus.The percentages of fetuses with poorly ossified lumbar vertebral centra were significantly different among groups (p<O.OO1) There was only one fetus with a single poorlyossified lumbar vertebral center in group 1 and none in group 2. However, both groups3 and 4 had relatively large percentages of fetuses with poorly ossified lumbar vertebralcenters (18.8% in group 3 and 24.0% in group 4). The difference between groups 3 and4 was not significant.The average numbers of poorly ossified lumbar vertebral centra in each fetus in groups3 and 4 were 1.42 and 2.00 respectively, which were not significantly different (table 3.14).3.10.4 SkullMost bones in the skull were ossified on day 21 of gestation in all experimental groups.This included the mandible, premaxilla, vomer, zygomatic process of the maxilla, alisphenoid, squamosal, medial pterygoid plate, nasal, frontal, parietal, exoccipital and incisive,and the posterior palatine foramen was completely enclosed by ossifying tissue.Some additional bones were ossified in all but a few fetuses. These were zygoma (notossified in 2 fetuses from group 4 and one fetus from group 3); basisphenoid (not ossifiedin one fetus from group 3); interparietal bolle (not ossified in one fetus from group 3);and the tympanic annulus (ossified on the right side, but not on the left side in one fetusfrom group 4).The bones which were frequently unossified or incompletely ossified in the skull werethe hyoid bone, presphenoid bone, and the supraoccipital bone. The basisphenoid foramen was frequently incompletely enclosed by ossifying tissue (table 3.15).Chapter 3. Results 67Table 3.15: Ossification centers and foramen in skullGroup 1 2 3 4No of fetuses examined 103 84 101 79Hyoid bone 101(2)z 70(14)b 92()ab 71()abPresphenoid bone 102(1) 84(0)a 86(15)b 69(9)bBasisphenoid foramen 86(12)a 74(5)a 53(34)b 35(40)bSupraoccipital bone 103(0)a 84(0) 100(1)a 75(3)aGroup 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcoholFigures indicate numbers of fetuses in which ossification centers were presentor (not) or foramen were completely enclosed by ossifying tissue or (not)Values not sharing a common superscript in each row are significantly differentat p<O.O05The frequencies of the presence of ossification centers in the hyoid bone were significantly different among groups (p<O.Ol). There was a larger percentage of fetuses failingto show ossification centers in the hyoid bone in group 2 than in group 1 (p.<O.OO1). Thiswas one of the two parameters affected only by alcohol exposure (16% of fetuses in group2 did not have ossification centers in the hyoid bone, only 2% in group 1).There was no difference in the frequencies of ossification centers in the hyoid bonebetween any other pairs of groups, although there was a trend that diabetic groups tendedto have more fetuses lacking ossification centers in the hyoid bone. About 9% of fetusesin both diabetic groups did not have ossification centers in the hyoid bone.The frequencies of presence of ossification centers in presphenoid bone were significantly different among groups (p <0.001). Groups 1 and 2 did not differ from each other.15% of the fetuses in group 3 and 11% of those in group 4 lacked ossification centers inpresphenoid bone, which were significantly different from group 1 (p<O.Ol). However,the difference between groups 3 and 4 was not significant.Chapter 3. Results 68The figures in table 3.15 concerning the basisphenoid foramen show whether theforamen was completely enclosed or riot (in parentheses). The frequencies of fetuses withforamen completely enclosed by ossifying tissue were significantly different among groups(p<O.OO1). Groups 1 and 2 did not differ from each other. The percentages of fetuseswith completely enclosed basisphenoid foramen in groups 1 and 2 were 83.5% and 88.1%respectively. Both groups 3 and 4 were significantly different from controls (p<O.OO1)with 52.5% of fetuses in group 3 and 44.3% of those in group 4 having completely enclosedbasisphenoid foramen. Again the difference between groups 3 and 4 was not significant.The frequencies of the presence or absence of ossification center in supraoccipital bonewere not different among groups. However, some fetuses did have ossification center insupraoccipital bone which were not complete. These fetuses either showed a dumbbellshaped ossification center or cleaved ossification centers in supraoccipital bone. Thepercentages of fetuses with incompletely ossified supraoccipital bone were significantlydifferent among groups (p<O.OO1) Groups 1 and 2 did not differ from each other. Only3.9% of fetuses in group 1 and 2.4% of fetuses in group 2 had incomplete ossificationof supraoccipital bone. Both groups 3 and 4 were significantly different from controls(p<O.OOl). There were 70.3% and 86.1% of fetuses in groups 3 and 4 respectively hadincompletely ossified supraoccipital bone. The difference between groups 3 and 4 wasalso significant (p<O.0l25). Group 4 had a higher percentage of fetuses with incompletelyossified supraoccipital bone compared to group 3.In addition, some fetuses did not have the foramen structure at all. This was considered a variant (table 3.12). Differences among groups were observed (p<O.05), but whencomparisons were made between treatment pairs, the differences were not significant.This is probably because the sample size in each group was too small.Chapter 3. Results 693.10.5 Other skeletal variantsThe frequencies of skeletal variants are shown in table 3.13. Most of these have alreadybeen described. In addition, one of the commonest skeletal variants was the occurrence ofa 14th extra rib, which either appeared at cervical or lumbar in position, and was eitherrudimentary or full size (see Appendix A, photo 10). The presence of a 14th rib wasmuch more frequent in the two diabetic groups (p<O.OOl) but was not different betweengroups 3 and 4. The other phenomenon found in the nondiabetic-alcohol group (group2) was a short or rudimentary 13th rib. This was not found in other groups (p<O.OO1).3.10.6 Bone areasAlthough ossification centers were present in all groups, the extent of ossification was notnecessarily the same among groups. Therefore, the areas of ossified bone were measuredin the humerus, ulna, radius, femur, fibula, and tibia.The areas of ossified tissue in each of these bones were significantly smaller in diabeticgroups compared to those in nondiabetic groups (p<O.OO1), but did not differ betweenalcohol and water exposed groups (table 3.16).Chapter 3. Results 70Table 3.16: Ossified tissue areas (mm2)measured in the selected bones (mean ± SD)GroupHumerus*Ulna*Radius*Femur*Fibular*Tibia*13.32 ± 0.541.22 ± 0.272.14 ± 0.461.71± 0.410.92 ± 0.211.91 ± 0.4723.30 ± 0.261.23 ± 0.152.23 ± 0.271.83 ± 0.370.90 ± 0.132.02 ± 0.3532.62 ± 0.391.00 ± 0.151.64 ± 0.331.32 ± 0.350.66 ± 0.091.62 ± 0.3242.45 ± 0.460.91 ± 0.191.40 ± 0.401.08 ± 0.310.58 ± 0.221.34 ± 0.38Group 1: Non-diabetic on waterGroup 2: Non-diabetic on alcoholGroup 3: Diabetic on waterGroup 4: Diabetic on alcohol* Diabetic significantly different from non-diabetic, p<0.05Chapter 4DiscussionThe major objectives of this study were: (1) to test whether a small amount of alcoholadministered during organogenesis has an effect on the outcome of pregnancy in nondiabetic rats; (2) to confirm the teratogenic effect of maternal diabetes on the fetus; (3) toestablish whether there is an interaction between alcohol and diabetes on the outcomeof pregnancy.4.1 Calorie and water consumptionIn the present study, the average daily calorie and water consumption were not significantly different between groups 1 and 2 during days 1-5 of gestation, when group 2 wasnot administered alcohol; days 6-11, when group 2 was given alcohol; and days 12-20,when group 2 was finished alcohol administration. These results demonstrated that theadministration of small amount of alcohol (2g/kg/day) during organogenesis (days 6-11) to nondiabetic rats (group 2) did not affect calorie and water intake. These resultsare different from those of previous studies (Abel and Dintcheff, 1978a; Tze and Lee,1975; Vavrousek-Jakuba et al., 1991), in which higher doses of alcohol were administeredchronically, and food and water consumption were depressed in alcohol consuming animals compared to controls. The results in the present study were also different fromthose of Abel’s study (1978b). In that study, the daily amount of alcohol administeredwas the same as that of ours, but chronically, and food and water consumption weredecreased.71Chapter 4. Discussion 72The average daily calorie intake and water intake of diabetic rats (groups 3 and 4)were much higher than those of nondiabetic rats (groups 1 and 2) in the present study,which is consistent with those of previous studies (Eriksson, 1984a; Giavini et al., 1986).In diabetic rats, alcohol administration significantly reduced the average daily calorieintake during the period of alcohol administration, and average daily water intake duringthe period of alcohol administration as well as afterwards. Nevertheless, the calorie andwater intake by alcohol exposed diabetic rats were still significantly higher than those ofcontrols.4.2 Reproductive performanceThere was no significant difference between diabetic and nondiabetic rats in the proportion of rats apparently breeding, as determined by the presence of vaginal plugs (100%in nondiabetic rats and 93.5% in diabetic rats). This is consistent with the study byGiavini et al. (1986). However, not all rats which had vaginal plugs produced litters.Of the animals with plugs, 100% of those in group 1 (controls); 90% of those in group2 (non-diabetic, alcohol); 68% of those in group 3 (diabetic, water); and 61.5% of thosein group 4 (diabetic, alcohol) produced litters on day 21. That these differences didnot attain statistical significance may be due to the small numbers of animals in eachgroup. This is similar to the findings of previous studies (Eriksson et al., 1 989a; Erikssonet al., 1982), in which it was found that more diabetic rats failed to implant despite asperm-positive vaginal smear than was the case in nondiabetic controls.The reason for this is not certain. However, it is clear that diabetes is not a disease with simply abnormal carbohydrate metabolism. It is characterized by multiplemetabolic abnormalities, including abnormal protein and lipid metabolism. The development of the fertilized egg, embryo formation and fetal development are processesChapter 4. Discussion 73consisting of a high rate of cell division, requiring active protein synthesis. Altered protein synthesis in diabetes obviously will affect these processes. If the fertilized egg cannotdevelop properly, it possibly cannot implant later on. Therefore this may account for theobservation that diabetic rats often did not produce litters despite vaginal plugs. Evenin those dams with successful implantation, there may be an interference with the laterprocesses of development, resulting in resorptions or fetal malformations.4.3 Weight gainIt has been observed that although overall maternal weight gain during pregnancy may benormal, weight gain during organogenesis, or during a portion of organogenesis, may beadversely affected by experimental treatments. For this reason Black and Marks (1986)recommended that weight gain in rats be reported separately for days 6-10 of gestation(where the weight of any embryos is insignificant) and during the treatment period, inaddition to the entire gestational period.In the present study, the body weight gains were recorded separately for 3 periodsof the gestation (days 1-5, 6-11, 12-20). Diabetic rats gained less weight during all 3periods of gestation and, as a result, weighed significantly less than did the controlsthroughout gestation. These results corroborate previous studies in which diabetic ratsgained less weight than controls during gestation (Eriksson et al., 1989a; Eriksson etal., 1989b; Eriksson and Jansson, 1984b; Sybuiski and Maughan, 1971; Urir-Hare et al.,1989). However, the study by Giavini et al. (1986) was an exception, in that they foundthat diabetic rats had a similar weight gain during gestation and their body weights werenot significantly different from controls at term. The reason for the differences in theresults from different studies is not known.On the other hand, alcohol administration to nondiabetic rats via stomach tubeChapter 4. Discussion 74during organogenesis (days 6-11 of gestation) at the level of 2g/kg did not affect bodyweight gains during the period of administration or afterwards. At term, alcohol exposednondiabetic rats had similar body weights to those of controls. This result differs fromprevious studies (Testar et al., 1986; Tze and Lee, 1975; Vavrousek-Jakuba et al., 1991),in which depressed weight gain was found during pregnancy in rats following treatmentwith higher doses of ethanol. Furthermore, in Abel’s study (1978b), the amount of alcoholadministered was the same as in the present study, and he too found a depressed weightgain in rats exposed to alcohol during gestation. In that case, the pair-fed animals alsohad a depressed weight gain. Abel suggested that the lower weight gain of the ethanol-treated animals was probably the result of ethanol’s depressant effect on food intakerather than on the assimilation of food. Unlike Abel’s study, in the present study alcoholwas administered only for a short period during gestation (day 6-11 of gestation). Foodand water consumption were not affected by alcohol administration; therefore, calorie andwater consumption patterns were similar to those of controls. As a result, the weightgain pattern did not significantly differ from that of the controls.In the present study, a significant interaction between diabetes and alcohol exposureon body weight gain from day 12 to day 20 of gestation was found. The difference inweight gain during this period between diabetic and nondiabetic rats was greater inalcohol exposed rats than in non-alcohol exposed rats. This difference was caused by atrend toward greater weight gain in alcohol exposed rats than in controls, while weightgain was less in the diabetic-alcohol group than in the diabetic-water group.4.4 Outcome of pregnancyDifferences in the number of implantation sites per dam, litter size, mean number ofresorptions per litter and percentage of resorptions in each group, as well as in theChapter 4. Discussion 75number of litters with one or more resorptions were not statistically significant amonggroups.Our study is consistent with that of Eriksson et al. (1982) in terms of the implantationsites, in which the implantation sites per dam were as numerous in diabetic dams as inthe controls.With respect to the incidence of resorptions, our study is consistent with the studyby Giavini et al. (1986). In that study the incidence of resorptions in diabetic CD ratswas not different from that of the controls. However, many other studies had reportedmuch higher frequencies of resorptions in diabetic rats than in the controls (Eriksson etal., 1989a; Eriksson et al., 1982; Eriksson, 1984a; Eriksson et al., 1989b; Urir-Hare et al.,1989).As for litter size, findings have not been consistent among studies. Some found smallerlitter size in diabetic rats (Eriksson et al., 1989a; Urir-Hare et al., 1989), while othersfound that the mean litter size of the control and diabetic rats did not differ from eachother (Heinze and Vetter, 1987; Ornoy et al., 1984).In the studies on alcohol exposed rats, some researchers reported no decrease in littersize (Abel, 1978b; Kahns, 1968; Pilstrom and Kiessling, 1967), while others found adecrease in litter size when rats chronically consumed large amounts of alcohol (Testar etal., 1986; Tze and Lee, 1975). Chronic consumption of lower doses (1-2g/kg) of alcoholthroughout gestation, also has resulted in reduced litter size (Abel, 1978b).The inconsistent findings among studies were probably due to differences in experimental conditions, such as the use of different strains of animals, or the differences in thetiming of breeding after the rats becoming diabetic. The different amounts of alcoholadministered and the different periods of alcohol administration may have been especiallyimportant in producing different results in experiments on alcohol treated animals.Chapter 4. Discussion 764.5 Fetal and placental weightsIn the present study, diabetic rats had lighter mean fetal weights but larger placentascompared to nondiabetic rats. These are consistent with the findings of previous studies(Eriksson et al., 1989a; Eriksson et al., 1982; Eriksson et al., 1989b; Eriksson and Jansson,1984b; Giavini et al., 1986; Urir-Hare et al., 1989) in which reduced fetal weights andheavier placentas were observed in diabetic rats.It has been suggested that the increase in placental weight is a long-term compensatory mechanism, aiming to secure a sufficient nutrient supply to the fetuses, since adrastic reduction in placental blood flow to the diabetic fetuses had been reported byEriksson et al. (1984b).Similarly, the combination of decreased fetal and increased placental weights hasalso been reported in the fetuses of alcohol-treated rats (Gallo and Weinberg, 1986;Gordon et al., 1985; Jones et al., 1981; Weinberg et al., 1990; Weinberg, 1985; Wieneret al., 1981). It has been suggested that placental hyperplasia and hypertrophy may beadaptive mechanisms brought into play to maintain normal placental function in alcohol-consuming females (Gallo and Weinberg, 1986; Weinberg, 1985). In those studies, ratswere all chronically exposed to high doses of alcohol. However, in our study, a low doseof alcohol was administered and the exposure time was only from days 6-11 of gestation.We did not find a decrease in fetal weight, nor an increase in placental weight in thenondiabetic, alcohol exposed group.It is noteworthy that a significant interaction between diabetes and alcohol on fetalweight was found in the present study. The difference in fetal weight between diabeticand nondiabetic rats was greater when rats were given alcohol than when given water.This difference was caused by the trend that alcohol-exposed nondiabetic rats (group 2)had bigger fetuses than controls (group 1) and alcohol-exposed diabetic rats (group 4)Chapter 4. Discussion 77had smaller fetuses than diabetic rats not exposed to alcohol (group 3).In addition, the ratio of fetal weight/placental weight was calculated in this study.This ratio reflects the relationship of a fetus to its corresponding placenta. For example, two fetuses with different birth weights may have the same placental weight. If thecomparison were simply done between the two placentas without considering the corresponding fetuses, we would have not seen any difference. However, when fetal weight isconsidered, the smaller fetus would have a relatively larger placenta, compared to thebigger fetus. Therefore, the smaller ratio represent a relatively bigger placenta for thesize of fetus.In the present study, whether exposed to alcohol or not, diabetic rats had lower f/pratio compared to nondiabetic rats. Interestingly, in nondiabetic rats, this low dose ofalcohol consumption raised the f/p ratio compared to controls. However, in diabetic ratsthe same level of alcohol consumption did not have an effect on f/p ratio.Similar observations in rats treated with low dose of alcohol have not been reported.Low doses of alcohol during short periods of pregnancy have not been used in previousstudies.It is difficult to explain the significant difference in f/p ratios between groups 1 and 2.As neither fetal weight nor placental weight differed significantly between groups 1 and2, the difference in ratios can not be ascribed to “better” placental development or toretarded fetal development. Fetal weight tended to be greater in group 2 and placentalweight tended to be less in group 2, compared to group 1. It may be that the ratio is amore sensitive parameter than either of its components.Chapter 4. Discussion 784.6 MalformationsA wide variety of external malformations were found in fetuses of diabetic rats, butnone were found in those of nondiabetic rats regardless of whether the animals wereexposed to alcohol or not. The malformations observed included: a variety of facialdefects, characterized by smaller lower jaw, no lower jaw, no mouth, tongue exposure,otocephaly, and distorted face; missing tail; umbilical hernia/gastroschisis; severe edema;and ectrodactyly. Malformed fetuses from diabetic dams have frequently been reportedin rats (Eriksson et al., 1989a; Eriksson, 1984a; Giavini et al., 1986; Giavini et al., 1990;Urir-Hare et al., 1989). The malformations found in those studies included subcutaneousedema, micrognathia, hepatomphalocoele, exencephaly, absence of the tail (Eriksson etal., 1989a; Eriksson, 1984a; Giavini et al., 1986). The results from those studies togetherwith those of the present study illdicate that a variety of external malformations occurin experimental diabetic animals and there was no specific malformation consistentlyobserved across studies. This is in agreement with the observations in other studies,summarized by Eriksson et al. (1982).Similarly, a variety of skeletal malformations were observed in both diabetic groupsin the present study. These included: cleft sternum, malformed ribs (bifurcated or fusedribs), vertebral malformations (arch or center), malformed arm joint, missing bone inskull, etc. A wide spectrum of skeletal malformations was also found in other studieswith diabetic rats (Baker et al., 1981; Eriksson et al., 1989a; Giavini et al., 1986; Giaviniet al., 1990; Urir-Hare et al., 1989), including fused ribs, malformed vertebrae, scoliosis,absence of the tail together with caudal vertebrae (i.e. sacral dysgenesis), a failure ofneural tube fusion, and split sternum.The lack of consistent malformations observed in different studies may be causedby different timing of inducing diabetes in the animals. The different strains used inChapter 4. Discussion 79studies may be another possible factor which caused the observations of a wide varietyof malformations. There may be genetic differences in susceptibility, and vulnerabilitymay be different at different time points during organogenesis.Unlike studies in mice, malformations were not found frequently in most studies ofalcohol treated rats (Abel and Dintcheff, 1978a; Abel, 1978b; Abel, 1979a; Lee andLeichter, 1983; Testar et al., 1986), especially in study of low dose alcohol consumption(Abel, 1978b), except for the study by Tze and Lee (1975).In the present study, all malformed fetuses were from diabetic dams. However, thefrequency of external malformations was significantly higher when diabetic rats exposedto alcohol were compared to diabetic rats not exposed to alcohol. This indicates that lowdose of alcohol administered during organogenesis may act synergistically with maternaldiabetes to produce external malformations in the fetuses. However, the mechanism isnot known at present.4.7 The proposed mechanisms for malformationsThe search for a teratogenic agent in diabetes is complicated by the fact that diabetesis not simply a disorder of carbohydrate regulation. Diabetes is responsible for a lossof normal homeostasis not only of carbohydrate but of fat and protein metabolism aswell. Vascular complications may lead to additional metabolic changes such as hypoxia orimpaired renal clearance of toxins. In short, there are multiple factors in the disorderedmilieu of the pregnant diabetic which could be teratogenic (Mills, 1982).Although the precise etiology of malformations in diabetic pregnancy is obscure, maternal hyperglycemia is one suspected etiology. In some in vitro rodent embryo culturesystems, it has been confirmed that severe malformations can be caused merely by increasing the glucose concentration of the culture medium (12 to 15 mg/ml, which is 8 toChapter 4. Discussion 8010 times the level of glucose present in normal rat serum (Cockroft and Coppola, 1977),or 5 or 8 mg/ml, which are very close to the level of glucose present in severe diabeticrat serum (Sadler, 1980)).The study by Eriksson et al. (1989b), demonstrated the importance of well regulatedglucose homeostasis during pregnancy. It was shown in this study that diabetic animalswith continuous insulin treatment gained weight at normal rate, and the fetuses showedfew resorptions and no malformations. When insulin treatment was interrupted, thepregnant animals lost weight, and congenital malformations were found. The markedclustering of malformations were found in fetuses of diabetic rats with insulin-interruptionbetween days 6 and 10 of gestation.However, abnormal glucose metabolism may not explain all malformations. Erikssonet al. (1989b) observed that rats that produced malformed offspring were not markedlydifferent from those that did not give birth to malformed offspring. The maternal weightlosses during the period of interrupted insulin therapy were similar, and resorption rateswere also similar in the litters with and without malformed offspring. Serum levels ofglucose, cholesterol, urea, and creatine did not differ between the pregnant rats. On theother hand, the marked elevation of /3-hydroxybutyrate and triglycerides in the serumof animals with malformed offspring may indicate that the teratogenic environment alsoincludes a severe disturbance of maternal lipid metabolism in addition to glucose dysregulation (Eriksson et al., 1989b).Trace metal imbalances is another possible etiology of congenital malformations indiabetic pregnancy. One trace metal which may be important is zinc. Although the etiological relationship between congenital malformations in diabetic pregnancy and fetalzinc deficiency is so far obscure, there is ample evidence to indicate that zinc deficiencyin itself may cause a number of disturbances in the feto-maternal development. A varietyof external as well as skeletal malformations have been demonstrated in the fetuses ofChapter 4. Discussion 81zinc deficient rats (Da Cunha Ferreira et al., 1989; Hurley and Swenerton, 1966; Hurley, 1981). The external malformations found in fetuses of zinc deficient rats include:cleft palate; exencephaly; meningoencephalocele; meningocele; hydrocephaly; microphthalmia; micrognathia; umbilical hernia; gastroschisis; syndactyly; shortening of limbs;limb dysplasia; oligodactyly; short, angulated, wavy, and curly tails; heart abnormalities;lung abnomalities; urogenital abnormalities; spina bifida, etc. Skeletal malformations infetuses of zinc deficient rats occurred in almost all skeletal districts, such as skull, spine,tail, ribs, sternum, and limbs. Most of these malformations described above have alsobeen found in fetuses of diabetic rats.It should be noted that a zinc deficiency has been demonstrated in diabetic rats(Eriksson, 1984a). It has been suggested that zinc deficiency is an important causeof growth retardation and congenital malformations also in the offspring of diabeticmothers. On the other hand, in the study by Eriksson (1984a), there were no differencesin trace metal concentrations between normal and malformed fetuses. This finding,however, might imply that fetal trace metal disturbances do not always become manifestas malformations and that zinc deficiency in particular may play a permissive role in theetiology of malformations in the offspring.Fetal hyperinsulinism, which is secondary to the hyperglycemia, was considered to bethe causative factor, since Laadauer and Duraswami (Giavini et al., 1986) had shown thathigh concentrations of insulin provoked vertebral anomalies in chick embryos (Giavini etal., 1986). In addition, the low molecular weight somatomedin inhibitors have proved tobe teratogenic (Giavini et al., 1986).Genetic factors might also play a role in the genesis of malformations. In a substrainof Spraque-Dawley rats that is prone to fetal malformations, there is the interactionbetween genetic factors and the metabolic derangement of diabetes that is necessary forthe induction of skeletal malformation (Eriksson, 1988), but the nature of the geneticChapter 4. Discussion 82predisposition is not yet clear. It is, however, noteworthy that diabetes exacerbateschanges in proteoglycan metabolism in cartilage that are more severe in this substrainthan in other Spraque-Dawley substrains (Eriksson et al., 1986).With respect to the mechanism of teratogenic effect of alcohol, it is still unclear.Ethanol appears to be the principal agent responsible for fetal alcohol syndrome, however,a growing body of evidence suggests that acetaldehyde, with its inhibiting effect on DNAsynthesis, placental amino acid transport, and morphologic alterations in the CNS ofthe developing mammalian embryo, could explain most of the abnormalities seen in fetalalcohol syndrome. The varying effects seen in the offspring of chronic alcoholic mothersingesting a similar amount of alcohol may be due to differences in the ability to metabolizealcohol. Deficiencies of zinc, which is involved in a variety of metalloenzymes includingalcohol dehydrogenase (ADH) and protein synthesis, may also be contributory (Kumar,1982).In the present study, diabetes combined with alcohol consumption during pregnancyincreased the frequency of external malformations compared to the solely diabetic pregnancy (almost tripled the malformation frequency, 12 vs 4.5%). It seems that the twofactors may synergistically act together. However, the mechanism whereby they acttogether is not understood at present.4.8 Skeletal ossificationThe toxic effects of a drug on the fetal development may manifest itself not only in termsof malformations but also in terms of retarded ossifications. It has been demonstratedthat the stage of fetal skeletal development can provide an additional, reliable index ofretarded fetal development, in addition to fetal weight (Aliverti et al., 1979). Therefore,evaluation of the stage of ossification attained by the fetus is important in studies onChapter 4. Discussion 83teratogenesis.Aliverti et al. (1979) have suggested that certain ‘districts’ of the skeleton provideparticularly sensitive indices of the stage of fetal ossification. These are: metacarpus,metatarsus, sternum, cervical and caudal vertebrae, and anterior and posterior proximalphalanges. All of these districts of the skeleton were examined in the present study. Inaddition, the thoracic, lumbar, and sacral vertebrae, and skull were also examined.4.8.1 Number of ossification centersThe ossification centers were coullted in the districts mentioned above except the proximalphalanges, which did not yet show ossification centers either in control or other treatmentgroups. As in my study, the studies of Lee and Leichter (1983) and Eriksson et al. (1989a)showed that the anterior and posterior proximal phalanges of control and treated ratswere not ossified.Studies have consistently shown that fetuses from diabetic mothers have fewer ossification centers in most locations (Eriksson et al., 1989a; Eriksson et al., 1982; Urir-Hare etal., 1989). It has also been reported that fetuses from rats consuming chronic high dosesof alcohol have fewer ossification centers compared to those from control and pair-fedrats (Lee and Leichter, 1983).In the present study, fetuses from diabetic rats had fewer ossification centers thanthose of nondiabetic rats in the districts of metacarpus, metatarsus and sternum. Somefetuses of diabetic rats had no ossification centers in these districts. Fetuses of diabeticdams had vertebral ossification centers which extended less cephalically and caudallyand had ossified vertebral arches which extended less caudally, compared to the fetusesof nondiabetic dams. These results indicated that fetuses of diabetic dams had retardedskeletal development relative to those of nondiabetic rats, which is consistent with previous studies (Eriksson et al., 1989a; Eriksson et al., 1982; Urir-Hare et al., 1989). ItChapter 4. Discussion 84is noteworthy that the fetuses of diabetic dams weighed significantly less than those ofnondiabetic dams. This means that the development of the whole fetus was retarded,and the retarded skeletal development does not seem to be specific.Alcohol exposure did not have an effect on the average number of ossification centersin the metacarpus, metatarsus, and sternum. Interestingly, in nondiabetic rats, fetuses ofalcohol-exposed rats had vertebral ossification centers which extended more cephalicallyand caudally and had ossified vertebral arches which extended further caudally comparedto those of non-alcohol exposed rats. However, the same effect of alcohol exposure wasnot found in diabetic rats. This indicates that small amounts of alcohol may enhancefetal development in fetuses of nondiabetic rats, but not in those of diabetic rats.4.8.2 Skeletal variationsAbnormally shaped skeletal elements, though a spontaneous phenomenon, can be increased in frequency in rat and mouse fetuses following maternal administration of drugs,such as thalidomide, salicylate, retinoic acid, diphenylhydantoin, and etc (Fritz and Hess,1970).In the present study, the irregular shapes observed in the sternum includes dumbbell-shaped, asymmetrically dumbbell-shaped, asymmetrical and asymmetrically dumbbell-shaped, simple asymmetrical, cleaved (bipartite), asymmetrical and cleaved, unilaterally ossified. Those observed in vertebrae includes: asymmetrical, dumbbell-shaped,asymmetrically dumbbell-shaped, asymmetrical and asymmetrically dumbbell-shaped,cleaved, cleaved and asymmetrically ossified, unilaterally ossified, and unossifled. Whetherthe bipartite ossified centers occur as part of the usual sequence of the development isnot clear. It has been noticed that in the rat and mouse, some “bipartite” and dislocatedsternebra may remain permanently altered (Fritz and Hess, 1970). Findings of this sortare usually listed under the heading of skeletal anomalies and can be shown to vary inChapter 4. Discussion 85frequency depending on the stock or strain of animal concerned. However, it has beenfound that dumbbell-shaped thoracic vertebral centers or bipartite sternebral centers canbe completely ossified after term (Fritz and Hess, 1970). Therefore, it is difficult to tellat the time of examination whether an irregularly shaped center will become “normal”or not later on. This caused trouble for categorizing the findings.In the present study, these irregular shapes were considered to have no significant biologic effect on animal health or on body conformity and represented only slight deviationsfrom normal (Fujinaga et al., 1989), as these may happen in control group. Therefore,they were categorized as developmental variants.The dumbbell-shaped, cleaved and unilaterally ossified centers were pooled togetherand referred as “poorly ossified centers” in this study. It has been found that not onlythere were more fetuses with poorly ossified sternebral centers, thoracic and lumbarvertebral centers but there were also more poorly ossified centers per fetus in thesedistricts in diabetic than in nondiabetic litters. Also, more fetuses of diabetic dams hadsternebral ossification centers with lack of apposition than fetuses of nondiabetic dams.Alcohol exposure itself did not have an effect on the shape of sternebral ossificationcenters. It should be noted, however, that diabetic rats exposed to alcohol had significantly more fetuses with poorly ossified thoracic centers than diabetic rats not exposedto alcohol. Similarly, more fetuses of diabetic rats had poorly ossified cervical archescompared to those of nondiabetic rats. Diabetic rats exposed to alcohol had more fetuses with poorly ossified cervical arches than diabetic rats not exposed to alcohol. Theseresults indicate that the administration of small amounts of alcohol during organogenesismay exacerbate the toxic effect of maternal diabetes on the formation of thoracic centers and cervical arches, although alcohol itself did not show this effect. However, howmaternal diabetes and alcohol administration act together is not clear at this time. Ithas been observed that the fetuses with more bone retardation are those very small ones.Chapter 4. Discussion 86And there is a trend that the average fetal weight in diabetic-alcohol group was smaller(table 3.8). More skeletal retardation could be secondary to more retardation overall.The presence of the 14th extra rib was much more frequent in the two diabetic groups.Only one fetus in group 2 had a 14th rib and none in the control group. The occurrenceof an accessory rib structure is not considered to be a malformation, but it is thoughtthat an increase in their incidence during teratogenicity tests, if not already associatedwith increased malformations, indicates that the embryotoxic range of dosage is beingapproached (Wilson, 1973). In the present study, the occurrence of the 14th rib significantly increased in fetuses of diabetic rats, while malformations were also increased infetuses of diabetic rats. Therefore, the presence of the 14th rib probably indicates theembryotoxic effect of maternal diabetes.The other variation observed in this study was the short or rudimentary 13th rib,which specifically occurred only in the alcohol treated group (group 2). Whether this indicates retarded fetal development is not established. If so, it is contrary to the findingsthat fetuses of nondiabetic alcohol-exposed rats had vertebral ossification centers extended more cephalically and caudally and had ossified vertebral arches extended morecaudally compared to those of non-alcohol exposed rats, which suggested that smallamounts of alcohol may enhance fetal development.4.8.3 Ossification of the skullIn a study by Fritz et al. (1970), it was found that the bones of the skull were completelyossified on day 21, with the exception of the siipraoccipital bone, which was dumbbellshaped in a few fetuses.Our findings in control rats are consistent with those of Fritz et al. In our controls,only 2 fetuses had unossified hyoid bones, one fetus had unossified presphenoid bone and4 fetuses had dumbbell-shaped supraoccipital bone. The rest of bones in the skull wereChapter 4. Discussion 87completely ossified in all fetuses of controls.However, in the diabetic groups significantly more fetuses had a total absence of ossification in presphenoid bone compared to nondiabetic groups. There was no differencebetween the alcohol exposed group and the water exposed group. The most impressivefinding is the ossification of supraoccipital bone. In both diabetic groups, except for a fewfetuses with a complete lack ossification of supraoccipital bone, most fetuses had supraoccipital centers which were either dumbbell-shaped or bipartite. This higher frequency ofincompletely ossified supraoccipital bone in diabetic rats compared to nondiabetic ratsobviously demonstrated a retarded skull development in the fetuses of diabetic dams.It should be noted that when diabetic rats were exposed to alcohol, the frequency ofincompletely ossified supraoccipital bone in the fetuses was significantly higher than thatin diabetic rats not exposed to alcohol.Significantly more fetuses in the nondiabetic, alcohol exposed group (group 2) failedto show ossification centers in hyoid bone, compared to controls. Similar results were alsoreported by Lee and Leichter (1983). In that study, they reported that most fetuses inalcohol exposed group did not show ossification centers in hyoid bone, but did in controls.However, unossified centers were also found in other bones of the skull. In that study,the amount of alcohol administered was high, and administration was chronic comparedto that of the present study. At the low dose of alcohol administration, why the hyoidbone was affected but not other bones is difficult to understand.4.8.4 Bone areasThe number of ossification centers can only tell whether the bone has ossified or not, butnot the extent of the ossification. Therefore, the ossified tissue areas were measured inhumerus, nina, radius, femur, fibula and tibia of the fetuses. Fetuses of diabetic groups(groups 3 4) had significantly smaller areas for all these bones, compared to nondiabeticChapter 4. Discussion 88groups. This provided further evidence that the fetuses of diabetic dams had retardedskeletal development.To summarize the findings in the skeletal examination, we have demonstrated that:(1) the fetuses of diabetic dams had a retarded skeletal development as follows: Thefetuses of diabetic dams had a smaller number of ossification centers in metacarpus,metatarsus, and sternum, and had vertebral arches and centers which extended lesscephalically and/or caudally. Not only were there more fetuses with poorly ossifiedsternebral, thoracic vertebral and lumbar vertebral centers but there were also morepoorly ossified centers per fetus in these districts. More fetuses had sternebral ossification centers with lack of apposition. More fetuses had poorly ossified cervical arches.For the skull, more fetuses had an absence of presphenoid bone, incompletely enclosedforamen, and poorly ossified supraoccipital bone. The 14th extra rib occurred frequentlyin fetuses of diabetic dams compared to those of nondiabetic dams. Fetuses of diabeticdams had smaller bone areas in humerus, ulna, radius, femur, fibula and tibia; (2) alcoholexposure alone seemed to enhance skeletal development with vertebral ossification centers extended more cephalically and caudally and ossified vertebral arches extended morecaudally. On the other hand, significantly more fetuses had absence of ossification centersin the hyoid bone and a 13th short rib, possibly indicating retarded skeletal ossification;(3) interaction between diabetes and alcohol on several parameters was observed. Morefetuses had poorly ossified thoracic vertebral centers, poorly ossified cervical arches, andpoorly ossified supraoccipital bone in alcohol exposed diabetic rats than in nonalcoholexposed diabetic rats. This possibly indicates that small amount of alcohol administration during organogenesis could exacerbate the toxic effect of maternal diabetes on thefetuses.Chapter 4. Discussion 894.9 The effect of alcohol consumptionIn the present study, alcohol was administered at the level of 2g/kg/day, which is approximately equivalent to 4oz absolute alcohol (8 average-sized drinks per day) consumedby a 60kg woman. The peak blood alcohol level was 73.3mg/dl which was comparableto a level at which some adverse effects of alcohol on the offsprings of both human andanimals were observed.However, the alcohol administration alone did not have any adverse effects on thefetuses of group 2 rats. This may be caused by the transient blood alcohol level andshort period of alcohol exposure in this experiment.Although alcohol consumption alone did not show any visible effects in this study,we did observe an interaction between alcohol and diabetes. However, we can not tellwhether the observed interaction was the direct or indirect effect of alcohol, since thealcohol exposed diabetic rats significantly reduced their calorie and water consumptioncompared to the non-alcohol exposed diabetic rats. Without pair-fed controls, it ishard to say whether the undernutrition in alcohol exposed diabetic rats interacted withdiabetes, which caused the more severe adverse effect in this group.4.10 Summary and conclusionTo summarize, the results of this investigation support the view of previous studiesthat maternal diabetes is teratogenic to the offspring in terms of significantly increasedmalformation rate, retarded fetal development (small fetal weight, big placenta, andretarded skeletal ossification).Consumption of small amount of alcohol (2g/kg) during organogenesis (days 6-11 ofgestation) did not seem to intoxicate the dams. Body weight gain was at a normal rate,and there was a larger f/p ratio compared to controls. In terms of skeletal development,Chapter 4. Discussion 90alcohol exposure seemed to enhance bone development, with the exception of a few fetuseswith absence of ossification centers in the hyoid bone and the short 13th rib, possiblyindicating retarded skeletal development.A significant interaction between maternal diabetes and alcohol administration duringorganogenesis was observed. Fetal external malformation rate was significantly increasedin alcohol exposed diabetic rats compared to nonalcohol exposed diabetic rats. Furthermore, more fetuses had poorly ossified thoracic vertebral centers, poorly ossified cervicalarches, and poorly ossified supraoccipital bone in alcohol exposed diabetic rats than innonalcohol exposed diabetic rats. These observations suggested that a small amountof alcohol consumption during organogenesis can exacerbate the embryotoxic effects ofmaternal diabetes.Appendix APhotographs91Appendix A. Photographs 92Photo 1: Fetuses on day 21 of gestation, fixed in 95% alcohol. The right fetus, which isfrom a control dam, is about 2 times bigger than the left one, which is from a diabeticdam exposed to alcohol.Appendix A. PhotographsPhoto 2: A fetus from a diabetic rat exposed to akohol, showing otocephaly.93Appendix A. Photographs 94Photo 3: A fetus from a diabetic rat exposed to alcohol, showingan absence of tail.Appendix A. PhotographsPhoto 4: A fetus from a diabetic rat exposed to alcohol, showing a gastroschisis.95Appendix A. Photographs 96Photo 5: A fetal skeleton from a diabetic rat, showed the 5th left cervical arch bifurcatedand a cleft sternum.Appendix A. Photographs 97Photo 6: Ventral view of the fetal skulls on day 21 of gestation. Left skull is from a fetusof control dam; right one is from a fetus of diabetic dam exposed to alcohol. Except forthe smaller size, the right skull is less ossified and has a missing zygomatic process andzygoma bone.Appendix A. Photographs 98Photo 7: Fetal skeletons on day 21 of gestation stained with alizarin red S and alcian blue8GS. From left to right, control (group 1); non-diabetic on alcohol (group 2); diabeticon water (group 3); diabetic on alcohol (group 4) . In addition to the smaller size of thefetuses from diabetic dams, the skeletons are less ossified (more blue color).4Appendix A. Photographs 99Photo 8: A fetal skeleton from a control litter (skull removed). The focus is the normalsternebral ossification centers.Appendix A. Photographs 100Photo 9: A fetal skeleton from a control litter (skull removed). The focus is the normalvertebral ossification centers and cervical arches.Appendix A. Photographs 101Photo 10: A fetal skeleton from a diabetic rat exposed to alcohol, showed a rudimentaryextra rib on the left, and a short extra rib on the right. These extra ribs appeared atlumbar position. In addition, this fetus showed a few poorly ossified vertebral ossificationcenters.—m—Appendix BChi-square tests, P values102Appendix B. Chi-square tests, P values 103Overall lvs2 lvs3 lvs4 2vs3 2vs4 3vs4Breeding rate NSPregnant rate NS% resorptions NS% litters with resorptions NS% external malformed fetuses - - - - - - <0.05% litters with external - - - - - - NSmalformed fetuses% malformed fetuses - - - - - - NS(external & skeletal)% litters with malformed - - - - NSfetuses (external & skeletal)The most cephalic vertebral <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 NScenterThe most caudal vertebral <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 NScenterThe most caudal vertebral <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 NSarch (Left)The most caudal vertebral <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 NSarch (Right)Hyoid <0.01 <0.001 NS NS NS NS NSPresphenoid <0.001 NS <0.001 0.002 <0.001 0.001 NSForamen <0.001 NS <0.001 <0.001 <0.001 <0.001 NSSupraoccipital NS - - - - - -Extra rib <0.001 - - - - - NSThe 13th short rib <0.001 - - - - - -Incomplete ossification of <0.001 NS <0.001 <0.001 - - 0.0118supraoccipital boneLack of foramen <0.05 NS NS NS NS NS NSPoorly ossffied sternebral <0.001 NS <0.001 <0.001 <0.001 <0.001 NScentersLack of apposition in <0.001 NS 0.001 <0.001 0.0040 <0.001 NSsternebral centersPoorly ossified thoracic <0.001 NS <0.001 <0.001 <0.001 <0.001 0.0039centersPoorly ossified lumbar <0.001 NS - - - - NScentersPoorly ossified cervical <0.001 - - - - - 0.0133archesAppendix CTwo way ANOVA statistical effects diabetes vs alcohol, P values104Appendix C. Two way ANOVA statistical effects diabetes vs alcohol, P values 105Diabetic Alcohol InteractionWater D1-5 ave. <0.001 NS NSWater DO-li ave. <0.001 <0.01 <0.01Water D12-20 ave. <0.001 <0.05 <0.05Calorie D1-5 ave. <0.001 NS NSCalorie DO-il ave. <0.001 <0.01 <0.01Calorie D12-20 ave. <0.001 NS NSBodywtDl NS NS NSBody wt D6 <0.01 NS NSBody wt D12 <0.001 NS NSBody wt D21 <0.001 NS NSWt gain 1)1-5 <0.05 NS NSWt gain DO-il <0.001 NS NSWt gain 1)12-20 <0.001 NS <0.05Implantation NS NS NSResorption NS NS NSLive fetuses NS NS NSFetal wt <0.001 NS <0.05Placeutal wt <0.01 NS NSf/p ratio <0.001 <0.05 <0.05Metacarpus (L) <0.001 NS NSMetacarpus (R) <0.001 NS NSMetatarsus (L) <0.01 NS NSMetatarsus (R) <0.01 NS NSSternum <0.01 NS NSBone areasHumerus <0.001 NS NSUlna <0.001 NS NSRadius <0.001 NS NSFemur <0.001 NS NSFibular <0.001 NS NSTibia <0.01 NS NSMean no. of poorly ossified <0.001 NS NSsternebral centersMean no. of poorly ossified <0.001 NS NSthoracic centersAppendix DTests of Simple Effect, P values106Appendix D. Tests of Simple Effect, P values 107Dia at water Dia at aic Aic at non-dia Aic at dia(1 vs 3) (2 vs 4) (1 vs 2) (3 vs 4)Water D6-11 ave. <0.001 <0.001 NS <0.001Water D12-20 ave. <0.001 <0.001 NS <0.01Calorie D6-11 ave. <0.001 <0.001 NS <0.001Wt gain. D12-20 <0.001 <0.001 NS NSFetal wt <0.001 <0.001 NS NSf/p ratio <0.001 <0.001 <0.01 NSBibliography[1] Abel, E.L., and Dintcheff, B.A. (1978a) Effects of prenatal alcohol exposure ongrowth and development in rats. J. Pharmacol. Exp. Ther. 20?’: 916-921.[2] Abel, E.L. (1978b) Effects of ethanol on pregnant rats and their offspring. Psychopharmacology 57: 5-11.[3] Abel, E.L. (1979a) Prenatal effects of alcohol on adult learning in rats. Pharmacol.Biochem. Behav. 10: 239-243.[4] Abel, E.L. (1979b) Effects of ethanol exposure during different gestation weeks ofpregnancy on maternal weight gain and intrauterine growth retardation in the rat.Neurobehav. Toxicol. 1: 145-151.[5] Aliverti, V., Bonanomi, L., Giavini, E., Leone, V.G., and Mariani, L. (1979) Theextent of fetal ossification as an index of delayed development in teratogenic studieson the rat. Teratology 20: 237-242.[6] Ariyuki, F., Higaki, K., and Yasuda, M. (1980) A study of fetal growth retardationin teratological tests: An examination of the relationship between body weight andossification of coccygeal vertebrae in mouse and rat fetuses. Teratology 22: 43-49.[7] Aufrere, G., and Lebourhis, B. (1987) Effect of alcohol intoxication during pregnancy on fetal and placental weight: Experimental studies. Alcohol and Alcoholism22: 401-407.[8] Baird, J.D., and Aerts, L. (1987) Research priorities in diabetic pregnancy today:The role of animal models. Biol. Neonate 51: 119-127.[9] Baker, L., Egler, J.M., Klein, S.H.,and Goldman, A.S. (1981) Meticulous controlof diabetes during organogenesis prevents congenital lumbosacral defects in rats.Diabetes 30: 955-959.[10] Black, D.L., and Marks, T.A. (1986) Inconsistent use of terminology in animaldevelopmental toxicology studies: a discussion. Teratology 33: 333-338.[11] Brownscheidle, C.M., Wootten, V., Mathieu, M.H., Davis, D.L., and Hofman, l.A.(1983) The effects of maternal diabetes on fetal maturation and neonatal health.Metabolism 32 (Suppl. 1): 148-155.108Bibliography 109[12] Chernoff, G.F. (1975) A mouse model of the fetal alcohol syndrome. Teratology 11:14A.[13] Chernoff, G.F. (1977) The fetal alcohol syndrome in mice: An animal model. Teratology 15: 223-280.[14] Chernoff, G.F. (1980) The fetal alcohol syndrome in mice: maternal variables.Teratology 22: 71-75.[15] Clarren, S.K., and Smith, D.W. (1978) The fetal alcohol syndrome. N. Engi. J.Med. 298: 1063-1067.[16] Cluing, C.S., and Myrianthopoules, W.C. (1975) Factors affecting risks of congenital malformations. Report from the collaborative perinal project. In: Birth defectsoriginal article series (ed Bergsma D). The National Foundation-March of Dimes,Vol. 11, Symposia Specialists, Miami, pp. 23-38.[17] Cockroft, D.L., and Coppola, P.T. (1977) Teratogenic effects of excess glucose onheadfold rat embryos in culture. Teratology 16: li,(1-146.[18] Da Cunha Ferreira, R.M.C., Monreal Marquiegui, I., and Villa Elizaga, I. (1989)Teratogenicity of zinc deficiency in the rat: study of the fetal skeleton. Teratology39: 181-194.[19] Driscoll, C.D., Streissguth, A.P., and Riley, E.P. (1990) Prenatal alcohol exposure:Comparability of effects in humans and animal models. Neurotoxicol. Teratol. 12:231-297.[20] Ellis, F.W., and Pick, J.R. (1976) Beagle model of the fetal alcohol syndrome.Pharmacologist 18: 190.[21] Elton, R.H., and Wilson, M.E. (1977) Changes in ethanol consumption by pigtailedmacaques. J. Stud. Aic. 38: 2181-2183.[22] Eriksson, U., Dahistrom, E., Larsson, K.S., and Hellerstrom, C. (1982) Increasedincidence of congenital malformations in the offspring of diabetic rats and theirprevention by maternal insulin therapy. Diabetes 30: 1-6.[23] Eriksson, U.J. (1984a) Diabetes in pregnancy: Retarded fetal growth, congenitalmalformations and feto-maternal concentrations of zinc, copper and manganese inthe rat. J. Nutr. 114: 477-484.[24] Eriksson, U.J., and Jansson, L. (1984b) Diabetes in pregnancy: Decreased placentalblood flow and disturbed fetal development in the rat. Pediatric Research 18: 795-738.Bibliography 110[25] Eriksson, U.J., Unger, E., and Kjellen, L. (1986) Decreased levels of high molecular weight cartilage proteoglycans in diabetic rats of a malformation-prone strain.Diabetologia 29: 534A.[26] Eriksson, U.J. (1988) Importance of genetic predisposition and maternal environment for the occurrence of congenital malformations in offspring of diabetic rats.Teratology 37: 365-374.[27] Eriksson, U.J., Bone, A.J., Turnbull, D.M., and Baird, J.D. (1989a) Timed interruption of insulin therapy in diabetic BB/E rat pregnancy: effect on maternalmetabolism and fetal outcome. Acta Endocrinol. 120: 800-810.[28] Eriksson, R.S.M., Thunberg, L., and Eriksson, U.J. (1989b) Effects of interruptedinsulin treatment on fetal outcome of pregnant diabetic rats. Diabetes 98: 764-772.[29] Fisher, S.E., Atkinson, M., Burnap, J.K., Jacobson, S., Sehgal, P.K., Scott, W. andVan Thiel, D.H. (1982) Ethanol-associated selective fetal malnutrition: A contributing factor in the fetal alcohol syndrome. Alcohol.: Gun. Exp. Res. 6: 197-201.[30] Fisher, S.E., Inselman, L.S., Duffy, L., Atkinson, M., Spencer, H., and Chang,B. (1985) Ethanol and fetal nutrition: effect of chronic ethanol exposure on ratplacental growth and membrane associated folic acid receptor binding activity. J.Pediatr. Gastroenterol. Nutr. 4: 645-649.[31] Freinkel, N. (1980) The Banting lecture 1980: Of pregnancy and progeny. Diabetes29: 1 023-1035.[32] Freinkel, N. (1988) Diabetic embryopathy and fuel-mediated organ teratogenesis:Lessons from animal models. Horm. Metabol. Res. 20: 469-4 ‘75.[33] Fritz, H., and Hess, R. (1970) Ossification of the rat and mouse skeleton in theperinatal period. Teratology 3: 391-338.[34] Fujinaga, M., Baden, J.M., and Mazze, R.I. (1989) Susceptible period of nitrousoxide teratogenicity in sprague-dawley rats. Teratology 40: 439-444.[35] Gallo, P.V., and Weinberg, J. (1986) Organ growth and cellular development inethanol-exposed rats. Alcohol 3: 261-267.[36] Giavini, E., Broccia, M.L., Prati, M., Roversi, G.D., and Vismara, C. (1986) Effectsof streptozotocin-induced diabetes on fetal development of the rat. Teratology 34:81-88.Bibliography 111[37] Giavini, E., Prati, M., and Roversi, G. (1990) Congenital malformations in offspringof diabetic rats: Experimental study on the influence of the diet composition andmagnesium intake. Biol. Neonate 57: 207-217.[38] Glasgow, A.C.A., Harley, J.M.G., and Montgomery, D.A.D. (1979) Congenital malformations in infants of diabetic mothers. Ulster. Med. J. 48: 109-117.[39] Goldman, A.S., Baker, L., Piddington, R., Marx, B., Herold, R., and Egler, J.(1985) Hyperglycemia-induced teratogenesis is mediated by a functional deficiencyof arachidonic acid. Proc. Natl. Acad. Sci. USA 82: 8227-8231.[40] Gordon, B.H.J., Streeter, M.L., Rosso, P., and Winick, M. (1985) Prenatal alcoholexposure: abnormalities in placental growth and fetal amino acid uptake in the rat.Biol. Neonate 47: 113-119.[41] Greizerstein, H.B. and Aldrich, L.K. (1983) Ethanol and diazepan effects on intrauterine growth of the rat. Develop. Pharmacol Ther. 6: 409-418.[42] Hadden, D.R. (1986) Diabetes in pregnancy 1985. Diabetologia 29: 1-9.[43] Heinze, E., and Vetter, U. (1987) Skeletal growth of fetuses from streptozotocindiabetic rat mothers: in vivo and in vitro studies. Diabetologia 30: 100-103.[44] Henderson, G.I., Turnter, D., Patwardhan, R.V., Lumeng, L., Hoyumpa, A.M., andSchenker, 5. (1981) Inhibition of placental valine uptake after acute and chronicmaternal ethanol consumption. J. Pharmacol. Exp. Ther. 216: 465-4 72.[45] Henderson, G.I., Patwardhan, R.V., McLeroy, S., and Schenker, S. (1982) Inhibitionof placental amino acid uptake in rats following acute and chronic ethanol exposure.Alcohol.: Clin. Exp. Res. 6: 495-505.[46] Hill, E.P., and Longo, L.D. (1980) Dynamics of maternal-fetal nutrient transfer.Fed. Proc. 39: 239-244.[47] Horii, K., Watanabe, G., and Ingalls, T.H. (1966) Experimental diabetes in pregnantmice: prevention of congenital malformations in offspring by insulin. Diabetes 15:194-204.[48] Hurley, L.S., and Swenerton, H. (1966) Congenital malformations resulting fromzinc deficiency in rats. Proc. Soc. Exp. Biol. Med. 123: 692-696.[49] Hurley, L.S. (1981) Teratogenic aspects of manganese, zinc, and copper nutrition.Physiol. Rev. 61: 249-295.Bibliography 112[50] Jones, K.L., and Smith, D.W. (1973a) Recognition of the fetal alcohol syndrome inearly infancy. Lancet 2: 999-1001.[51] Jones, K.L., Smith, D.W., Ulleland, C.N., and Streissguth, A.P.S. (1973b) Patternof Malformation in offspring of chronic alcoholic mothers. Lancet 1: 1267-1271.[52] Jones, P.J.H., Leichter, J., and Lee, M. (1981) Placental blood flow in rats fedalcohol before and during gestation. Life Sci. 29: 1153-1159.[53] Kahns, A.J. (1968) Effect of ethanol exposure during embryogenesis and neonatalperiod on the incidence of Lepatoma in C3H male mice. Growth 32: 311-316.[54] Kennedy, L.A. (1984) Changes in the term mouse placenta associated with maternalalcohol consumption and fetal growth deficits. Am. J. Obstet. Gynecol. 149: 518-522.[55] Keppen, L.D., Pysher, T., and Rennert, O.M. (1985) Zinc deficiency acts as aco-teratogen with alcohol in fetal alcohol syndrome. Pediatric Res. 19: 944-947.[56] Kimmel, H.D. (1985) Simple and factorial experiments. In: Experimental principlesand design in psychology. New York: The Ronald Press Co. pp. 153-1 77.[57] Kitzmiller, J.I., Cloherty, J.P., Younger, M.D., Tabatabaii, A., Rothchild, S.B.,Sosenko, I., Epstein, M.F., Singh, S., and Neff, R.K. (1978) Diabetic pregnancyand perinatal morbidity. Am. J. Obstet. Gynecol. 131: 560-580.[58] Kucera, J. (1971) Rate and type of congenital anomalies among offspring of diabeticwomen. J. Reprod. Med. 7: 61-70.[59] Kumar, S.P. (1982) Fetal Alcohol Syndrome. Mechanisms of Teratogenesis. AnnalClin. Lab. Sci. 12: 254-257.[60] Lee, M., and Leichter, J. (1983) Skeletal development in fetuses of rats consumingalcohol during gestation. Growth 47: 254-262.[61] Leichter, J., and Lee, M. (1979) Effect of maternal ethanol administration on physical growth of the offspring in rats. Growth 43: 288-297.[62] Leichter, J., and Lee, M. (1984) Does dehydration contribute to retarded fetalgrowth in rats exposed to alcohol during gestation? Life Sci. 35: 2105-2111.[63] Majewski, F., Nothjunge, J., and Bierich, J.R. (1979) Alcohol embryopathy anddiabetic fetopathy in the same newborn. Helv. Paediat. Acta. 34: 135-139.Bibliography 113[64] Maims, J.M. (1979) Fetal abnormalities related to carbohydrate metabolism: theepidemiological approach. In: Sutherland HW, Stowers JM (eds): Carbohydratemetabolism in pregnancy and the newborn 1978. Springer- Verlag, Berlin, Heidelberg,New York, pp. 229-246.[65] Marquis, S.M., Leichter, J, Lee, M. (1984) Plasma amino acids and glucose levelsin the rat fetus and dam after chronic maternal alcohol consumption. Biol. Neonate46: 36-43.[66] Martin, J.M., Martin, D.C., Lund, C.A., and Streissguth, A.P. (1977a) Maternalalcohol ingestion and cigarette smoking and their effects on newborn conditioning.Alcohol.: Clin. Exp. Res. 1: 243-247.[67] Martin, J.C., Martin, D.C., Sigman, G., and Radow, B. (1977b) Offspring survival,development, and operant performance following maternal ethanol consumption.Dcv. Psychobiol. 10: 435-446.[68] Maykut, M.O. (1979) Consequences of prenatal maternal alcohol exposure includingthe fetal alcohol syndrome. Frog. Neuro-Psychopharmacol. 3: 465-481.[69] Mills, J.L., Baker, L., and Goldman, A.S. (1979) Malformations in infants of diabetic mothers occur before the seventh gestational week: implications for treatment.Diabetes 28: 292-293.[70] Mills, J.L. (1982) Malformations in infants of diabetic mothers. Teratology 25: 385-394.[71] Molsted-Pedersen, L. Tygstrup, I. and Pedersen, J. (1964) Congenital malformations in newborn infants of diabetic women. Lancet 1: 1124-1126.[72] Moisted-Pedersen, L. (1980) Congenital malformations in the offspring of diabeticwomen. In: Proc. Congr. mt. Diabetes Fed., 10th, Vienna, Austria. Amsterdam,Expcerpta Med., pp. 758-62.[73] Nelson, B.K., Brightwell, W.S., MacKenzie, D.R., Khan, A., Burg, J.R., Weigel,W.W., and Goad, P. T. (1985) Teratological assessment of methanol and ethanolat high inhalation levels in rats. Fundamental and Appi. Toxicol. 5: 727-736.[74] Ornoy, A., Menu, B., Zusman, I., Granat, M., Barash, V., and Shafrir, E. (1984)Placental and skeletal changes in fetuses of streptozotocin-diabetic rats. In: Lessonsfrom animal diabetes. E. Shafrir and A.E. Renold, eds. John Libbey, London, pp.775-781.[75] Papara-Nicholson, D., and Telford, I.R. (1957) Effects of alcohol on reproductionand fetal development in the guinea pig. Anat. Rec. 127: 438-4 39.Bibliography 114[76] Pedersen, J. (1977) Congenital malformations. In: The pregnant diabetic and hernewborn. 2nd Edition. Munkgaard, Copenhagen, pp. 191-196.[77] Pedersen, J. (1979) Congenital malformations in newborns of diabetic mothers. In:Sutherland HW, Stowers JM (eds): Carbohydrate metabolism in pregnancy and thenewborn 1978. Berlin, Springer, pp. 264-276.[78] Pedersen, J., and Moisted-pedesen, L. (1981) Early fetal growth delay detected byultrasound marks increased risk of congenital malformation in diabetic pregnancy.Brit. Med. J. 283: 269-271.[79] Pedersen, J.F., and Molsted-pedesen, L. (1982) Early growth delay predisposes thefetus in diabetic pregnancy to congenital malformation. Lancet 1: 737.[80] Pilstrom, L., and Kiessling, K.H. (1967) Effects of ethanol on the growth and onthe liver and brain mitochondrial functions of the offspring of rats. Acta Pharmacol.Toxicol. 25: 225-232.[81] Prager, R., Abramovici, A., Liban, E., and Laron, Z. (1974) Histopathologicalchanges in the placenta of streptozotocin induced diabetic rats. Diabetologia 10:89-91.[82] Randall, C.L., and Anton, R.F. (1984) Aspirin reduces alcohol-induced prenatalmortality and malformations in mice. Alcohol.: Gun. Exp. Res. 8: 513-515.[83] Randall, C.L., Taylor, W.J., and Walker, D.W. (1977) Ethanol-induced malformations in mice. Alcohol.: Gun. Exp. Res. 1: 219-224.[84] Randall, C.L., and Taylor, W.J. (1979) Prenatal ethanol exposure in mice: Teratogenic effects. Teratology 19: 305-312.[85] Rilely, E.P., and Meyer, L.S. (1984) Considerations for the design, implementation,and interpretation of animal models of fetal alcohol effects. Neurobehav. Toxicol.Teratol. 6: 97-101.[86] Sadler, T.W. (1980) Effects of maternal diabetes on early embryogenesis II.Hyperglycemia-induced encephaly. Teratology 21: 349-356.[87] Sadler, T.W., Hunter III, E.S., Wynn, R.E., and Phillips, L.S. (1989) Evidence formultifactorial origin of diabetes-induced embryopathies. Diabetes 38: 70-74.[88] Sokol, R.J., Ager, J., and Martier, S. (1986) Significant determinants of susceptibility to alcohol teratogenicity. Annals New York Academy of Sciences 477: 87-102.Bibliography 115[89] Sandor, S., and Amels, D. (1971) The action of ethanol on the prenatal developmentof albino rats. Rev. Roum. Embryol. 8: 105-118.[90] Simpson, J.L., Elias, S., Martin, A.O., Palmer M.S., Ogata, E.S., and Qadvany, Q.A.(1983) Diabetes in pregnancy, Northwestern University series (1977-1981). Am. J.Obstet. Gynecol. 146: 263-268.[91] Streissguth, A.P., Landesman-Dwyer, S., Martin, J.C., and Smith, D.W. (1980)Teratogenic effects of alcohol in humans and laboratory animals. Science 209: 353-361.[92] Snyder, A.K., Singh, S.P., and Pullen, G.L. (1986) Ethanol-induced intrauterinegrowth retardation: Correlation with placental glucose transfer. Alcohol.: Clin Exp.Res. 10: 167-170.[93] Soler, N.G., Walsh, C.H., and Maims, J.M. (1976) Congenital malformations ininfants of diabetic mothers. Quart. J. Med. 45: 303-313.[94] Streissguth, A.P., Landesman-Dwyer, S., Martin, J.C., and Smith, D.W. (1980)Teratogenic effects of alcohol in humans and laboratory animals. Science 209: 353-361.[95] Strong, R.M. (1926) The Grder, time, and rate of ossification of the albino rat (musnorvegicus albinus) skeleton. Am. J. Anat. 36:313-355.[96] Styrud, J., Dahlstrim, V.E., and Eriksson, U.J. (1986) Induction of skeletal malformations in the offspring of rats fed a zinc-deficient diet. Upsala J. Med. Sci. 91:29-36.[97] Sybuiski, S., and Maughan, G.B. (1971) Use of streptozotocin as diabetic agent inpregnant rats. Endocrinology 89: 1537-1540.[98] Testar, X., Lopez, D., Llobera, M., and Herrera, E. (1986) Ethanol administrationin the drinking fluid to pregnant rats as a model for the fetal alcohol syndrome.Pharmacol. Biochem. Behav. Diabetes 24: 625-630.[99] Tom, C., Juriloff, D.M., and Harris, M.J. (1991) Studies of the effect of retinoic acidon anterior neural tube closure in mice genetically liable to exencephaly. Teratology43: 27-4 0.[100] Tze, W.J., and Lee, M. (1975) Adverse effects of maternal alcohol consumptionon pregnancy and fetal growth in rats. Nature 257: 4 79-480.Bibliography 116[101] Urir-Hare, J.Y., Stern, J.S., Reaven, G.M., and Keen C.L. (1985) The effects ofmaternal diabetes on trace element status and fetal development in the rat. Diabetes34: 1031-1040.[102] Uriu-hare, J.Y., Stern, J.S., and Keen, C.L. (1989) Influence of maternal dietary Znintake on expression of diabetes-induced teratogenicity in rats. Diabetes 38: 1282-1290.[103] Vavrousek-Jakuba, E.M., Baker, R.A., and Shoemaker, W.J. (1991) Effect ofethanol on maternal and offspring characteristics: comparison of three liquid dietformulations fed during gestation. Alcohol.: Clin. Exp. Res. 15: 129-135.[104] Warner, R.H., and Rosett, H.L. (1975) The effects of drinking on offspring: Anhistorical survey of the American and British literature. J. Stud. Alcohol 36: 1395-1420.[105] Warren, K.R., and Bast, R.J. (1988) Alcohol-related birth defects: an update.Public Health Reports 108: 638-642.[106] Weathersbee, P.S., and Lodge, J.R. (1978) A review of ethanol’s effects on thereproductive process. J. Reprod. Med. 21: 68-78.[107] Weinberg, J., D’alquen, G., and Bezio, S. (1990) Interactive effects of ethanolintake and maternal nutritional status on skeletal development of fetal rats. Alcohol7: 383-388.[108] Weinberg, J. (1985) Effects of ethanol and maternal nutritional status on fetaldevelopment. Alcohol.: Clin. Exp. Res. 9: 49-55.[109] Wiener, S.G., Shoemaker, W.J., Koda, L.V., and Bloom, F.E. (1981) Interaction of ethanol and nutrition during gestation: influence on maternal and offspringdevelopment in the rat. J. Pharmacol. Exp. Ther. 216: 572-579.[110] Wilson, J.G. (1973) Environment and birth defects. Academic Press, New York.Chap. 9, pp. 173-193.[111] (1988) Congenital abnormalities in infants of diabetic mothers. Lancet June 11,1313-1315.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0077299/manifest

Comment

Related Items