83

Estimating
Head Shape

Fetal

Age:

Effect

of

on BPD

Frank

P. Hadlock1 R. L. Deter2 R. J. Carpenter2 S. K. Park1

Several recent obstetrical sonographic examinations in this department demonstrated that variations in the shape of the fetal skull (e.g., dolichocephaly, brachycephaly) may adversely affect the accuracy of the biparletal diameter (BPD) measurement in estimating fetal age. In each case the cephalic index of the fetal skull (short axis/ long axis x 100) was in either the dolichocephalic or brachycephalic range based on established postnatal criteria. Consequently, normal values were determined (mean, 78.3) for the cephalic index in utero based on 31 6 obstetrical sonographic studies

performed

at 14-40

weeks.

Preliminary

experience

indicates

that

a cephalic

index

greater than I SD from the mean (<74, >83) may be associated with a significant alteration in the BPD measurement expected for a given gestational age, and that the head circumference can be used effectively as an alternative means of establishing gestational age.

The
gestational

bipanietal

accurate reported accurate

for a good

(BPD) has proven to be a reliable indicator of fetal [1 -3]. In the second trimester of pregnancy, it is to within 1 -1 .5 weeks (2 SD) [2-3], but in the third trimester the accuracy is considerably less; a BPD obtained after 28 weeks is only to within 3 weeks (2 SD), even if the image meets the criteria
(menstrual) age

diameter

BPD

[3-4].

The

observed

variation

in the third

trimester

is undoubtedly

multifactonial in etiology, related only in part to technical errors in imaging. If we assume a technically adequate BPD image, and an accurate measurement, and if we eliminate pathologic causes of variation in fetal head size (e.g., microcephaly, hydrocephaly, growth retardation), there remain two obvious reasons why women with the same last menstrual period may have fetuses with

different BPD measurements: (1 ) genetic variations in head size in fetuses of the same conceptual age and (2) differences in time of ovulation and fertilization with respect to the first day of the last menstrual period. Our recent experience has suggested that variations in the shape of the fetal
skull such as dolichocephaly and brachycephaly may also have a significant

effect
determine variations of the

on

BPD
the in the

measurements.
normal shape relation

An
between

investigation
the which fetal short

was
and

therefore
long axis

undertaken
of the fetal

to
skull

at the BPD plane,

Received
revision

in the hope

that it would
skull,

produce
may gestational

a simple
adversely age.

method
affect

for detecting
the accuracy

October
12,

21,
1981.

1980:

accepted

after

of the fetal

February

BPD

measurement

in predicting

Department of Radiology, Baylor College of Medicine, Houston, TX 77019, and Jefferson Davis Hospital, 1801 AlIen Parkway, Houston, TX

Subjects

and Methods
31 6 consecutive patients using a commercially available dynamic image

77019. Baylor

Address College

reprint

requests

to F. P. Hadlock.

2Department of Obstetrics
of Medicine,

and Gynecology,
TX 77030.

We examined

Houston,

AJR I 37:83-85, July 1981 0361 -803X/81 /1371-0083 $00.00 American Roentgen Ray Society

scanner (Toshiba Medical Systems, Carson, Cal.). The gestational age, based on the BPD measurement, was 1 4-40 weeks. The widest transverse and longitudinal dimensions of the skull at the level of the BPD [3, 4] were measured from outer margin to outer margin (fig. 1), and the cephalic index [5, 6] (short axis/long axis x 100) was calculated.

84

HADLOCK

ET

AL.

AJR:137,

July

1981

Representative
Case 1

Case Reports

A 24-year-old
sonographically
identified. The

woman, gravida 3, para 2, Ab 0, was examined at 35 menstrual weeks. A single viable fetus was
bipanietal diameter was 7.9 cm (32 weeks), which

suggested that the fetus was growth retarded or that the menstrual dates were 3 weeks in error. However, the head was noted to be
rather elongated [5-6] measurement [7]. Measurements
abdominal [9] were both
(fig. 3) and this was confirmed by a cephalic index

of 68, which is by definition dolichocephaly of the head circumference [8] (32 cm) and
at the with level the of the umbilical menstrual vein (31 .8 cm) of 35 patients history

circumference consistent

weeks
later,
Fig. 1.-Real-time sonognaphic image of fetal

amenorrhea.

The patient

delivered

skull

at level of bipanietal

weight
with

and the head circumference [8] (3,300 g), and Dubowitz

fetus.

7 weeks [8] (36 cm), length [8] (50 cm), score [10] were all consistent

spontaneously

diameter.

Cephalic

index

is A/B

x 100.

a 42 week

96.0

Case

2 woman,
menstrual of 7.6 this reason cm

A 27-year-old
sonography x
LIJ

gravida
weeks (30

5, para 4, Ab 0, was referred

to rule weeks) out was placenta consistent previa. with

for
The the

at 30 diameter

88.0

. . .

. . .

. .

3.

bipanietal

3
.

2. 2

224.
. . . 2 5. 2#{149} 3#{149}22.. 62. . .2.

menstrual
served, and

history.
for

A posterior

marginal
a repeat

placenta

previa
was

was
suggested.

ob-

_J

80.0

2
.

. .

<
.

2 . 2. #{149} 3 #{149}222 4 . #{149}22 2 . 5 22

..

. 34..3.3. 3425 534.. 3 62#{149}2 4. . 62 . . 2.3. 24 2 2 3.3

examination

(Li

2.

.3.3

.5.

23 ...4

4.2
.

72.0

The second examination was 6 weeks later (36 menstrual weeks), and the placenta was posterior with no evidence of a marginal placenta previa. The BPD on this examination was 8.2 cm (33 weeks), and this was initially believed to represent evidence of growth retardation. However, on reviewing the sonograms (fig. 4) it

640
7.0 14.0

was obvious
_J 42.0
at

that the head shape on the second

more elongated
from is the within of the which by

examination

was

considerably
ments index of of the 76,

than on the first.

first the study normal

Additional
[5-7];

measurea cephalic the head

21.0

28.0

35.0

head which

demonstrated range study represents

GESTATIONAL
Fig. 2.-Distribution of cephalic index

AGE (wks)
measurements in 31 6 fetuses

circumference
Additional phalic index

was 28 cm, which

measurements of 70,

is at the mean
second

at 30.5
demonstrated

weeks

[8].
a ce-

14-40

weeks.

definition

dolichocephaly

Results
The stnated values the BPD distribution no has significant of cephalic change index with measurements increasing of 1 0 fetuses by head shape; demongestational

[7], and the head circumference was 32.5 cm, which approximates the mean at 36 weeks [8]. At 3 weeks later the patient spontaneously delivered a 3,630 g boy, which was judged to be a term
infant by neonatal examination. The head circumference at delivery

was 35 cm, which

is appropriate

for a term infant

[8].

age (mean,
was

78.3;
resulted

SD, 4.4)
in the

(fig.
affected

2). Preliminary

use of these
in which eight of

detection

significantly

Discussion
Previous the cephalic aged investigators index have established postnatally. normal values for

these were dolichocephalic and The range of error in predicting BPD varied
finding

two were brachycephalic. the gestational age by the second trimester to 5.5

measured

In a study

of skull range

from
was

1 .5 weeks
in an

in the

nadiognaphs,
infants

Haas
4 weeks

[5] found

a mean
with

value

of 81 .7 in 52 be slightly by pnojecof length.

weeks
this

in the late third

gestational age

trimester. Our earliest 1 8 week dolichocephalic

had been established

observation of fetus whose

by an early

to 1 2 months,

an observed

true

of 73.5-90.4. higher than

tion of breadth

He noted that these values might actual values because the distortion
was greaten in his series than that

crown-rump the time of temperature.

length measurement and by documentation of ovulation by measurement of the basal body However, most of our cases (eight of 1 0) have

Jordaan
neonates

[6]

measured

the

cephalic
section;

index

directly

in 50
a mean

delivered

by cesanean

he found

in the third trimester. Interestingly, in two and cephalic index changed from normal values to abnormal values oven a 6 week period in the third trimester (see case 2 below). Our experience to date suggests that a cephalic index greaten than 1 SD from the mean (<74, >83), may be associated with a significant change in the BPD measurement expected for any given gestational age.

been observed cases the BPD

value data, tional

change

of 80.6, with a normal range of 76-85 (2 SD). Our based on a large sample oven a wide range of gestaages (1 4-40 weeks), demonstrated no significant
in the cephalic index be
and

with

gestational

age.

Our the

mean greater
from

value
distortion

of 78.3
of the

is slightly
This may
frontal

lower
related

than
in parallel

the
part
bones

values
to
resulting

observed

postnatally. the sonognaphic

occipital

beam

passing

to these

structures

AJR:137,

July

1981

ESTIMATING

FETAL

AGE

85

Fig. 3.-Case 1 . Real-time sonognaphic image shows dolichocephalic shaped fetal skull in 35 week fetus. BPD (7.9 cm) suggested gestational age of 32 weeks, while head circumference (32 cm) was appropriate for 35 weeks.

Fig. suggested weeks.

4.-Case gestational

2. Real-time
fetus. B, 6 weeks

sonographic
later. More

images.
elongated

A,

BPD

of

7.6

cm

(30
fetal

weeks)
skull;

in 30
BPD (8.2

week
cm)

normocephalic

(dolichocephalic)

age of 33 weeks:

however,

head circumference

(32.5

cm) was appropriate

for 36

rather than perpendicular. The range of normal (2 SD) in our series was rather large (70-86), which is very similar to values reported by Haas [5] in a study of 705 adults oven
the age of 21 years.

age, and encourage

REFERENCES

we hope that the others to examine

presentation of this the same problem.

data

will

Dolichocephaly

is defined

by a cephalic

index

below

75.9,
1 . Mitchell D. Accuracy gestational age. Arch

while bnachycephaly is said to index exceeds 81 [7]. Our case variations in the shape of the fetal the accuracy of the BPD faced with this situation, parameters Direct tape of fetal growth measurements

occur when the cephalic reports indicate that such skull may adversely affect age. When turn to other age. been

of pre- and postnatal

Dis Child I 979;54

assessment

of

:896-904

2. Kurtz AB, Wapner

RJ, Kurtz RJ, et al. Analysis

of bipanietal

in predicting gestational the sonographer must

3. 4.

in estimating the gestational of head circumference have

diameter as an accurate indicator of gestational age. JCU I 980;8 :319-326 Sabbagha RE, Hughey M. Standardization of sonar cephalometry and gestational age. Obstet Gynecol 1 978;52 : 402-406 Campbell 5, Thoms A. Ultrasound measurement of the fetal
head to abdomen circumference ratio in the assessment of

used for years as one postnatal index of age in neonates [8], and recently normal measurements of the fetal head circumference in uteno using sonography have been neported [1 1 , 1 2]. The accuracy of the head circumference
measurement in predicting gestational age in the third during mea-

trimester

of pregnancy

(2-3

weeks)

[8,

1 1 -1 2] is corn-

parable to the accuracy this period [3]. In our

using the BPD cases the head

measurement circumference

Br J Obstet Gynaecol 1977;84 :165-174 5. Haas LL. Roentgenological skull measurements and their diagnostic applications. AJR 1952;67 :197-209 6. Jordaan HVF. The difterential enlargement of the neurocranium in the full-term fetus. S Aft Med J 1976;50: 1978-1981 7. Friel JP, ed. Dorlands illustrated medical dictionary, 25th ed.

growth

retardation.

Philadelphia: Saunders, 1974:222, 470 8. Usher R, McLean F. Intrauterine growth of live-born

infants born between 1 969;74 : 901 -910
9. Sabbagha RE.

Caucasian
J Pediatr
In: Suspan vol 4. of

surements have been within 1 week of the true gestational age based on the menstrual history and/on postnatal evaluation [8, 1 0]. In addition, normal values for the fetal abdom-

25 and 44 weeks
in high-risk

of gestation.
obstetrics.

Ultrasound

inal ported
useful

circumference
as

(measured
of

at the
gestational fetal

level
age

of the
have

ductus
been ne1 0.

FP, ed. Current

Philadelphia:
Dubowitz gestational 10 1 1 . Campbell LM,

venosus)

a function

concepts in obstetrics Lea & Febiger, 1979:55

Dubowitz in the V. Goldberg newborn infant.

and gynecology,
C. Clinical assessment

recently
adjunct

[9];

this

measurement

may
gestational

prove
age

to be a
in such

in establishing

age

J Pediatr

1970;77

: 1-

cases. We
ence, tients

are

currently

measuring

the

BPD,

head

cincumfen-

abdominal with well

circumference, and established menstrual

cephalic dates.

index in paIn this way we

1 2.

S. Fetal head circumference against gestational age. In: Sanders R, James AE, eds. The principles and practice of ultrasonography in obstetrics and gynecology, 2d ed. New York: Appleton-Century-Cnofts, 1 980;454
H, Arabin PB, Baumann ML. Control of fetal devel-

hope to define more specifically the boundaries of the cephalic index at which the head circumference is consistently more accurate than the BPD in establishing gestational

Hoffbauer

opment Obstet

with multiple ultrasonic 1979;6: 147-156

body

measures.

Contrib

Gynec