W. Z. Billewicz

THE ASSESSMENT OF FETAL GROWTH

THE importance of low birthweight i n determining perinatal mortality, and as a possible cause of higher morbidity and impaired development during childhood, has been widely recognized. Two main categories of low birthweight babies ---those whose fetal growth rate has been normal but whose gestational age is short; and the “light for dates” baby, whose prenatal growth has probably been impaired-have been recognized to have different prognoses (e.g. Van den Berg and Yerushalmy, 1966). The assessment of fetal growth in the individual case involves the use of standards which describe the distribution of birthweights at given gestational ages. Neligan (1 965) and Gruenwald (1966) have reviewed most of the published work and have produced standards of their own. Most of the available standards are unsatisfactory in one way or another, either because the sample on which they were based was atypical, or because inaccuracies in the data gave rise to features which are likely to be “unbiological”. For example, both Neligan and Gruenwald found that the distributions of birthweights at early gestational ages showed evidence of bimodality, which they attributed to a proportion of cases in which the gestational ages had been underestimated. Both authors suggested, and Gruenwald applied, a correction which was based on the assumption that birthweights are symmetrically distributed around the principal mode of each bimodal distribution. Other investigators have used other devices to limit improbabilities in their data. Thus, Lubchenco, Hansman, Dressler and Boyd (1963) simply discarded infants whose birthweights were “not compatible” with their gestational ages; and McKeown and Gibson (1951) rejected the lowest and highest 1 per cent of values when birthweights were distributed by gestation within each pound (453 g.) of weight. This report is based on an analysis of 52,004 legitimate, single births which took place in the City of Aberdeen during the years 1948--64. The material is of exceptional value for the following reasons. Clinical records of all the births in a well-defined, socially variegated but ethnically rather homogeneous urban population were available. About 95 per cent of the inhabitants of Aberdeen were born in Scotland, the great majority in the northeast region. During the years under review, more than 80 per cent of Aberdeen births took place in one teaching maternity hospital, which was also responsible for the antenatal care of its booked patients. Most of the information on menstrual dates was collected by senior members of the obstetrical staff, many of whom were using the clinical records for research purposes. Finally, the heights and weights of nearly all the mothers were recorded, so that the influence of maternal size on birthweight could be investigated.

SOURCES OF VARIATION IN MENSTRUAL BLOOD LOSS

Menstrual blood losses were measured in 348 women aged 17 to 45 years in a Northumbrian mining village. Menstrual loss was found to be related to parity and to the birthweight of previous children. It is possibly associated with height, and may be partly controlled by uterine size and blood flow.

The weight of the placenta in relation to birthweight.

MATERIAL AND METHODS The analysis of birthweights was based on the records of 52,004 legitimate, single births in the City of Aberdeen during the years 1948-64. Cases of uncertain gestational age, macerated fetuses and perinatal deaths from malformations were excluded. For present purposes, we have also excluded domiciliary births, for which placental weights were seldom available, and fourth or later pregnancies, a high proportion of which were domiciliary births. Analyses involving certain clinical data could not be undertaken on births before 1958. Placentas were weighed routinely by the delivery room staff, without special preparation. Weights (originally recorded in pounds and ounces) therefore include attached cord and membranes and any contained blood or adherent blood clot. Data from the case records were summarized on Hollerith punch cards. Since birthweights were grouped in half-pound (227 8.) and placental weights in 2 oz. (57 g.) categories, it was impossible to obtain from the cards exact placental weight/birthweight ratios in individual cases and only the ratios of means are reported. The results described below do not suggest that it would be profitable to go back to the original case records in order to obtain more detailed information. Gestational ages are expressed in completed weeks, e.g. “39 weeks” includes all deliveries from 273 to 279 days, inclusive.

A SERIAL STUDY OF CHANGES OCCURRING IN THE ORAL GLUCOSE TOLERANCE TEST DURING PREGNANCY

Nineteen healthy pregnant women with no family history of diabetes took part in a serial study in which they had a standard 50 g. oral glucose tolerance test (OGTT) at 10, 20, 30 and 38 weeks gestation and again 10 to 12 weeks after delivery. While every test result remained within “normal” limits as defined by any of the standard criteria, subtle and progressive changes in the shape of the glucose response curve were shown to occur throughout pregnancy. These changes in shape were evaluated by means of the H index recently described by Billewicz et al. (1973); in 14 patients the values attained were characteristic of a “suspect” or “abnormal” response in non–pregnant subjects. Nevertheless since all of these women remained clinically normal throughout, and it is unlikely that 14 out of 19 will become diabetic later in life, it seems unreasonable that these changes per se should be regarded as evidence of impending pathology.

THE EPIDEMIOLOGY OF OEDEMA DURING PREGNANCY

THE interpretation of oedema during pregnancy is difficult, especially when it is not accompanied by other evidence of pre-eclampsia. It is doubtful if many obstetricians would support the extreme view of Cosgrove (1950): “. . . any degree of edema, even that manifested only by excessive weight gain is essentially abnormal, unphysiological and dangerous, necessitating alert attention and active therapy.” According to Greenhill (1965): “. . . mild edema unassociated with other symptoms . . . is of no special significance”, while Kellar (1959) considered that: “Oedema of the ankle region is very common in normal pregnancy and it is very difficult to say when this has passed the physiological limit.” In Kellar’s view therefore, the hydration of pregnancy may sometimes become manifest as oedema without being unphysiological. Our own interest in this subject was aroused during a study of the gain in total body water in women with normal pregnancies (Hytten, Thomson, and Taggart, 1966). None had hypertension or proteinuria, but a high proportion (46 out of 93) exhibited some oedema, usually of minor degree, but occasionally quite severe. It was not clear whether this should be regarded as unusual; we could find no published study of the incidence of oedema in normal and abnormal pregnancies. A study of the records of the Aberdeen Maternity Hospital was accordingly undertaken.

The energy cost of human lactation.

A total of 55 diet surveys were made on 49 women in Scotland within about 2 months of parturition some of whom were fully breast-feeding and some of whom were fully bottle-feeding their babies in an attempt to measure the effect of lactation on energy exchanges in the mother. 23 women were breast-feeding and 32 were bottle-feeding their babies; the lactating mothers took 591 kcal per day more than those who were not lactating. Both groups of mothers were losing weight and the estimated contribution of such losses to the total energy supply was added to and the amounts expended on basal metabolism deducted from the dietary energy intakes. The activity of each group was fairly similar; hence it was concluded that the average amount of energy available to support lactation was 618 kcal daily. The average energy value of the milk produced was estimated from the weights of the babies to be 597 kcal daily. From these data it was concluded that the energy exchanges in human lactation have an efficiency of 90% or more with the lower limit being 80%. Therefore an additional supply of 600 kcal in a daily diet should support lactation and 500 kcal is offered as the official recommended allowance.

Birth weights and placental weights in pre-eclampsia.

IT is commonly held that pre-eclampsia and essential hypertension tend to be associated with depression of foetal growth, presumably as a result of “placental insufficiency”. For example, Morris, Osborn and Wright (1955) state that in severe early pre-eclampsia the development of the foetus “is always retarded and it remains grossly undersized”; and that “the placenta is almost always very small, infarcted, seemingly underdeveloped, and of poor function”. We have not traced any reports confirming such impressions statistically. Clinical impressions may be misleading since so many cases of pre-eclampsia and hypertension are delivered prematurely and the “normal” range of birth weights at any stage of gestation varies so widely. METHOD The present analysis is based upon 7,168 single births to booked married city primiparae delivered in the Aberdeen Maternity Hospital during the years 1948-55; these hospital cases comprise nearly 90 per cent of all first births in the city. It is not easy to prepare objective statistics relating to pre-eclampsia, owing to uncertainties of diagnosis, especially in mild cases. For present purposes, criteria based on those of Nelson (1955) have been used. Preeclampsia is defined as a condition in which the diastolic blood pressure rises, after the 26th week of pregnancy, to 90 or more on two or more occasions separated by at least a day, or shows a definite or progressive pattern if the rise occurs in labour. Severity is assessed in accordance with the presence or absence of albuminuria (catheter specimen of urine). In this series, three grades of severity have been distinguished : (a) “severe”, when albuminuria rose to 2 g./litre (Esbach) or more; (b) “moderate”, when albuminuria was present but never rose to 2 g./litre; (c) “mild”, when there was no definite albuminuria. Oedema is not considered in the diagnosis or the grading. With these criteria, experience has shown that independent assessors obtain virtually identical results when analyzing the same hospital records, there being little room for the exercise of “judgment”. While their validity might in some instances be argued, there is no doubt that, generally speaking, the assessment of pre-eclampsia on such lines gives results which seldom conflict with clinical assessments made on more general grounds. The following cases were excluded from the analysis : (1) Macerated stillbirths. (2) Cases (971) in which the duration of gestation was uncertain because the patient was unsure of the date of her last menstrual period, or there was an obvious discrepancy in the first half of pregnancy between the stated date and such phenomena as the growth of the uterine fundus and the start of movements. An unusual birth weight was not, by itself, a reason for rejecting an otherwise acceptable date of last menstruation. Birth and placental weights in cases of preeclampsia were plotted on charts on which lines showing the average (median) and upper and lower extremes for all births at each stage of gestation had already been drawn. If the weights

Birthweights in consecutive pregnancies.

An analysis was made of birthweights in consecutive pregnancies among 6702 married women who were mostly followed up for ten years or more after the first viable birth. To facilitate comparisons between pregnancies, birthweights were standardized for sex and to 40 weeks gestation.

HAEMATOLOGICAL CHARACTERISTICS AND MENSTRUAL BLOOD LOSSES

In women of reproductive age in a Northumbrian mining village, there was a weak inverse relationship between the amount of blood lost at menstruation and various haematological indices. In those who had not recently taken iron medicines this relationship was more pronounced: the greater the loss, the lower the indices. Against expectation, however, this negative association was as strong at low as at high levels of menstrual loss. It is concluded that, within the range experienced in this community, the association between menstrual losses and haematological indices is not likely to be due to iron depletion.

Height, weight and food intake in man

Certain correlations between food intake and social status, height, weight and gain of weight during pregnancy were referred to in a previous series of papers (Thomson, 1958,1959a, b). In this paper, the same data, derived from a study of pregnant women, are used to investigate further the relationships of height and weight to diet. The most interesting conclusion is that, contrary to popular belief and the implication of physiological teaching (e.g. F.A.O. : Second Committee on Calorie Requirements, 1957), heavy people do not as a rule eat much more than light people. A review of the literature suggests that this conclusion is valid also for men and for non-pregnant women.