Acta Obstet Gynecol Scand 55: 53-57, 1W6

GROWTH OF THE FETAL SKULL, WITH SPECIAL REFERENCE TO WEIGHT-FOR-DATES OF THE NEWBORN CHILD Per Bergsjs, Trygve Bakke and Tor Bjerkedal From the Department of Obstetrics and Gynecology (Head: Professor Oddmund Koller), and the Institute of Hygiene and Social Medicine (Head: Professor Tor Bjerkedal), University of Bergen School of Medicine, Bergen. Norway

Abstract. Two growth curves of the fetal biparietal diameter are presented. One is prospective, based on the last menstrual period, while the other is retrospective, calculated from the date of birth. The prospective curve shows a slight decline in average values following the week 42 of gestation. This is thought not to reflect a true shortening of the biparietal diameter in the single fetus, since serial measurements in single fetuses show growth up to the time of delivery. Individual growth curves, grouped according to the weight-for-dates of the newborn child, show a trend towards longer biparietal diameters in the larger fetuses. Serial measurements, preferably starting before week 30, appear to be a good help in predicting birth weight, but growth rates without regard to the actual measurement values were of little prognostic value in the present material.

Several curves of the normal growth of the fetal skull, measured by pulsed ultrasound, have been presented (4,5, 6, 8). These are similar in shape, but they show parallel shifts along the metric axis. These differences are probably due to different apparatus calibration. Single measurements have been valuable in the assessment of fetal age (5). Serial measurements have given more accurate information about fetal prognosis than urinary estriol assays (7). A previous report from our department showed a correlation between the fetal biparietal diameter and the weight-for-dates of the newborn child in a small series of pregnancies complicated by pre-eclampsia (1). We have now extended our observations and will present normal curves for the biparietal diameter, based both on prospective and retrospective pregnancy dating. Furthermore, in this new series we shall report the results of all individual cases of normal and pre-eclamptic pregnancies, in which two or more measurements were performed, clas-

sifed according to the weight-for-dates of the newborn child. MATERIAL AND METHODS Normal pregnancies. These were defined as single pregnancies in women with no pregnancy complications, and live children without malformations. These form the basis of the normal curves. The material consists of 165 women with normal pregnancies. The prospective normal curve. This was based on a definite dating of the last menstrual period, in women with regular periods, who did not take contraceptive pills shortly before the actual pregnancy, irrespective of whether the pregnancy was terminated by induction or spontaneously. Altogether 13 1 pregnancies fufilled these criteria. The retrospective normal curve. This was constructed retrospectively from the time of delivery, in women who went into labour spontaneously. Altogether 54 pngnancies fulfilled these criteria. Classgcation of the fetuses. The fetuses were classified

1

60

6 'a

1 ,

26

28

JC

----1'

34

IL

, 38

,

,

4:

4.'

.

44 We-5

Fig. 1 . The smoothed normal prospective curve. The actual figures are given in Table I. Acta Obstet Gynccol Scand 55 (1976)

54

P . B e r g s j ~et al.

Table I. Weekly average values of the biparietal diameter in cases of normal pregnancy, based on definite dating of the last menstrual period

Table 11. Weekly average values of the biparietal diameter measured retrospectively from the time of spontaneous birth ~~

Gestational age (weeks)

No. of observations

Mean (mm)

S.D:

24.5 25.5 26.5 27.5 28.5 29.5 30.5 31.5 32.5 33.5 34.5 35.5 36.5 37.5 38.5 39.5 40.5 41.5 42.5 43.5

3 1 5 13 13 10 21 20 19 26 25 36 32 46 45 50 29 17 6 4

70.7 63.0 65.6 71.6 71.3 73.0 75.7 77.4 81.1 82.7 85.6 87.5 88.6 91.8 91.0 92.3 93.0 92.9 92.2 90.0

5.5 0.0 5.3 5.4 6.2 3.1 4.7 6.2 5.5 4.5 4.8 3.0 4. I 8.1 3.5 3.4 3.8 3.2 2.6 2.9

retrospectively in degrees of maturity, according to standard tables of birth weight and gestational age (3). These tables were based on 125485 births in Norway 1967 and 1968. ( a ) large for dates, birth weight above 90th percentile (b) medium-sized, upper range, birth weight between 50th-90th percentiles, (c) medium-sized, lower range, birth weight between 10th-50th percentiles, (d) small for dates, birth weight below 10th percentile. The material comprises 22 children in group a , 66 children in group b, 88 children in group c, and 16 children in group d. Measurements. The biparietal diameter was measured by pulsed ultrasound, using a diagnostic A- and B-scan unit made by Kretz-Technik (series 4100 MG). Before

Blparietal diameter

(mm)

100,

Fig. 2 . The retrospective normal curve. 'The actual figures are given in Table 11. Acta Obstet Gynecol Scand 55 (1976)

Week 0.5

1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5

No. of observations

Mean (mm)

S.D.

39 29 20 14 16 6 11

92.3 90.9 89.8 89.9 87.0 86.8 83.0 75.0 80.0 75.3 73.8 78.0 72.6 66.0

4.7 3.0 3.2 4.2 5.6 3.1 3.1 4.5 4.0 4.8 5.1 2.8 7.6 0.0

68.0

0.0

5

3 7 5 2

-

-

each measurement the apparatus was calibrated with a plexiglass cylinder, which was constructed to give a distance equivalent to 2 cm in tissue. The velocity of the ultrasound beam was in the order of 1570 mlsec. The measurements were performed alternately by 3 examiners. In a test series of duplicate measurements the standard deviation of the observer differences was 1.52 mm(2). Statistical analysis. The authors received help in the recording and programming of the material for electronic data processing from the EDB-section of the Medical Faculty, University of Bergen, and are especially grateful to programmer Alf M. Aksland.

RESULTS The normal curves. T h e prospective normal curve is presented in Fig. 1 and Table I. The number of observations after week 42 is small, but there is an indication of a decline in average values a t this time. T h e retrospective normal curve is shown in Fig. 2 and Table 11. This curve is less smooth than t h e prospective curve. The cause f o r this is probably the smaller number of observations. T h e retrospective curve has a gradual increase u p t o the time of spontaneous delivery. When delivery (time 0) in the retrospective curve is set equal t o 40 weeks i n the prospective curve, t h e two curves are very similar. Individual growth curves. The individual growth curves, subgrouped according t o the weight of the newborn child, are shown in Figs. 3-6. Cases with

Growth of the fetal skull Birth weimt

Blrth weight 10 - 50 percelltlle

,90 percentile TI

mm

.

55

o Preeclamptlc matlent5

o Preeciampt8c MtlentS

Normal pregnancies

90

-

90

80

-

80

70 -

'0

d

35

30

4 0 weeks

30

35

40 Week5

Fig. 3. Serial measurements of the biparietal diameter in fetuses who were large for dates at birth (above 90th percentile).

Fig. 5 . Serial measurements of the biparietal diameter in fetuses who were medium sized, lower range, at birth (between 10th-50th percentiles).

maternal pre-eclampsia can be identified by open circles in the figures. A comparison between the figures give the impression that there is a relationship between the biparietal diameter and the birth weight of the newborn child. However, this appears to be a statistical association rather than a clear-cut prognostic index in the single case. This will be further discussed below. Women with pre-eclampsia had children in all birth weight groups, but their relative numbers increased with decreasing birth weights of the children. With a few exceptions the fetal skulls of pre-eclamptic mothers showed similar patterns to those of mothers without pre-eclampsia. The correlation coefficient between the biparietal diameter taken shortly before birth, and the weight

and length of the newborn child are shown in Table 111.

DISCUSSION The shape of the prospective curve of biparietal diameter growth of the present investigation is similar to that presented by Willocks et al. (8), with the exception of the slight decline following week 41. The differences in actual measurements between different normal curves is most likely to be due to calibration differences, and not to reflect differences of fetal skulls. The dip following week 41 is a phenomenon which is also seen in normal curves of birth weight (3). The present material is made up of serial measurements of the same individual, and

B l r l h weqht 53 -90percentile

Birth welght < 10 percentile

rnm

I

o Preeclamptic

mm

patients

'

I

3G

.

.

.

.

I

35

.

.

.

.

o Preeclamptic patients

N 3 r m l pregnancies

I

40

Fig. 4 . Serial measurements of the biparietal diameter in fetuses who were medium sized, upper range, at birth (between 50th-90th percentiles).

35

4 0 weeks

Fig. 6 . Serial measurements of the biparietal diameter in fetuses who were small for dates at birth (below loth percentile). Acra Obsrer Gynecol Scand 55 (1976)

56

P . Bergsj0 et al.

Table 111. Coefficients of correlation (r) between the biparietal diameter measured during the last week before birth, and birth weight and length Biparietal diameter observa- versus birth weight (r) tions No. of

Condition

Normal pregnancies 58 Pre-eclamptic pregnancies 15

Biparietal diameter versus birth length (r)

0.58

0.72

0.63

0.37

single measurements in other cases. The single measurements were more frequent towards the end of pregnancy, at which time several women were referred to the obstetrical department for evaluation of hypermaturity. The shape of the curve is therefore thought not to reflect a true shortening of the biparietal diameter in individual fetuses. It is more likely to reflect a biological phenomenon, which means either a preponderance of small fetal skulls among those going past term, or an incorrect pregnancy dating on the part of the mothers. That the individual fetus does not show a shortening of the biparietal diameter is demonstrated in Figs. 3-6, in which the trend of gradual growth up to the time of birth is evident. The few instances of shortening of the biparietal diameters with time are thought to represent either observer’s error (2) or pathological conditions in cases of pre-eclampsia. The retrospective curve, using spontaneous labour as the point of reference, eliminates the possible error of incorrect dating, but may include some cases of premature labour. Individual growth curves, grouped according to the weight-for-dates of the newborn child, should indicate whether it is possible to predict the size of child at birth. Willocks et al. (8) could predict the birth of a dysmature or a non-dysmature child with a fair degree of accuracy, using a growth rate of 0.17 cm per week as the dividing line. We have not calculated the weekly growth indices, but a glance at Figs. 3-6 will show that such predictions would be of little value in these cases. A comparison of the large (Fig. 3) and the small (Fig. 6) babies shows that the former, as a rule, had a slower growth rate from about week 30 onwards than the latter. On the other hand the large babies had large biparietal diameters when first measured, in contrast to the small ones. Thus, the general trend was a conActa Obstet Gynecol S c a d 55 (1976)

vergence of the curves towards a normal average with a smaller variation at term. The two groups of medium sized babies show considerable overlapping of the curves, with a slight trend for the babies in the upper range to have larger biparietal diameters. The curves for fetuses of mothers with preeclampsia show the same growth trends as those of mothers without disease, with a few exceptions of arrested growth, not thought to be due to measurement errors. In a previous communication Bergs9 & Brodtkorb (1) found a more marked correlation between the biparietal diameter and pregnancy outcome than that of the present material. The former included many women with severe pre-eclampsia, who had severely growth retarded children, while the majority of the patients of the present material had mild pre-eclampsia. The correlation coefficients (Table 111) support the impression that there is no strong association between the biparietal diameter measurements and the weight-for-dates of the newborn child. The nutritional status of the fetus affects the body to a greater degree than it affects the head. On the other hand, judged from the curves, it seems that the time of maximum growth rate of the fetal head may have prognostic significance. Fetuses destined to be large babies appear to have large biparietal diameters before week 30. Therefore a measurement of the biparietal diameter between week 20 and 30, in cases of known gestational age, may prove to be the best prognostic index of the size of the newborn child. REFERENCES 1. Bergsje, P. & Brodtkorb, C.:

Ultrasonic fetal cephalometry in pre-eclampsia. Acta Obstet Gynecol Scand 52: 249, 1973. 2. Bergs@, P., Bakke, T. & Salamonsen, L.: Observer error in ultrasonic fetal cephalometry. Acta Obstet Gynecol Scand 54: 41, 1975. 3. Bjerkedal, T., Bakketeig, L. & Lehmann, E. H.: Percentiles of birth weights of single, live births at different gestation periods. Acta Pgdiat Scand 62: 449, 1973. 4. Brodtkorb, C. & Bergsje, P.: Bestemmelse av svan-

gerskapslengden ved hjelp av ultralyd. Nord Med 86: 1569, 1971. 5. Campbell, S.: The prediction of fetal maturity by ultrasonic measurement of the biparietal diameter. J . Obstet Gynaecol Br Comm 76: 603, 1%9. 6. Joppila, P., YMstalo, P. & Pystynen, P.: Fetal head

growth measured by ultrasound in the last few weeks of pregnancy in normal, toxaemic and diabetic women. Acta Obstet Gynecol Scand49: 367, 1970.

Growth of thefetal skull 7. Robinson, H. P., Chatfield, W. R., Logan, R. W. & Hall, F.: Intrauterine fetal growth retardation: A com-

Submittedforpublication Jan. 12, 1975

parison of human placental lactogen urinary oestriol and biparietal diameter measurements. Ann Clin Biochem 11: IS, 1974. 8. Willocks, J., Donald; I., Campbell, S. & Dunsmore, I. R.: Intrauterine growth assessed by ultrasonic foetal cephalometry. J Obstet Gynaecol Br Comm 78:804,

Per Bergsje Department of Obstetrics and Gynecology Akershus Central Hospital N-1474 Nordbyhagen Norway

57

1971.

1

Acta Obstet Gynecol Scand 55 (1976)