Structural, mechanical, and material properties of fetal cranial bone

Structural, mechanical, and material properties of fetal cranial bone TIMOTHY .4nn

J.

KRIEWALL,

PH.D.

Arbor, Michigan

The structural stiffness of 20 parietal bones from 10 fetal cadavers were tested under controlled laboratory settings. The slope at the onset of the load-deflection curve dictated the stiffness of the bone. These values were correlated with the bone’s modulus of elasticity, mineral content, and bone density, as well as with anthropometric and gestational data. The stiffness exhibited as much as a tenfold increase over the last 10 weeks of gestation and was highly correlated with birth weight. Anthropometric factors correlated more with the change in stiffness than did material factors, such as changes in modulus, mineral content, and bone density. The bone exhibited a pronounced fiber orientation which significantly affected the modulus of the test specimens. The results are discussed in terms of obstetric management of preterm and term labor. (AM. J. OBSTET. GYNECOL. 143:707, 1982.)

FETAL SKULL molding has been recognized as a phenomenon of birth for centuries. The first serious examination of the interaction between the maternal pelvis and fetal head was described by a Dutch obstetrician, Hendrik van Deventer (1651-1724).’ He recognized that pehic contraction could result in prolonged and difficult labor, and that the resulting fetal skull molding could lead to brain damage. Investigations ensued in the following centuries in order to identify infants at risk for birth trauma. The advent of x-rays seemed to offer hope that clinical pelvimetry would rule out cephalopelvic disproportion (CPD). However, since recent studies indicate that x-ray pelvimetry is largely ineffective in the diagnosis of CPD,‘. ” investigators, today, still seek new methods for diagnosing and eliminating the effects of disproportion. A recent effort suggests factoring maternal shoe size into the clinical assessment of labor.’ Clearly, more needs to be understood about the phenomenon before its untoward effects can be prevented.

From the Department of Obstetrics and Gynecology and the Bioenginrering Program, The Unizvrsity of Michigan. Suppn-ted in part No. HD 11202. Received

for

Re~wd

December

Acr~~t~rl

by National

publication

March

Reprint request.x 270.lA-09/3M.

Institutes

October

of Health

Grant

7, 1981.

I, 1981. 2, 1982. Dr. Timothy J. Krieurall, Bldg. 3M Center-, St. Pnul, Minnetotn

OOOZ-9378/8~/110707+OH$OO.X0/0~

1982 The C.V.Mosby

55141. Co

Our approach at The University of Michigan has been to elucidate the directions and extent of molding during the antepartum and intrapartum periods and to define the load-bearing capacity of the fetal head. The directions and extent of molding have been described elsewhere.” The purpose of this report is to shed light on the biomechanics of the fetal skull and its capacity to withstand the forces applied to it by the soft and hard tissues of the maternal pelvis. The effects of fetal age and development on the structural properties will be examined, and the constitutive properties of the bone itself will be correlated with its structural behavior. Material

and methods

Twenty parietal bones taken from 10 fetal cadaver calvaria were assessed for their stiffness, a measure of structural properties which include both geometric and material effects. In addition, 554 specimens were cut from 16 calvaria. including the previous 10. in order to assess the bone’s modulus of elasticity, a material property which influences the bone’s stiffness but which is independent of geometric considerations. Table I lists the biographical data on the fetuses and the cause of death. The specimens came from The University of Michigan’s Teratology Unit, and prior approval of this study was granted by the Hospital’s Human LTse Committee. The cranial bones were separated from the cadavers at postmortem and refrigerated at - 10” C in normal saline-soaked totvels in air-tight containers until test707

708

Iuly 15, 19x2 Am. ,I. Ohstet. <;yxol.

Kriewall

Fig. 1. To measure stiffness of whole parietal bones, a force, F. was applied at the inner aspect of the parietal eminence, creating a deHection. D. Tile ratio of the applied force to the resultant deHection is the stiftness.

ing co~dci be performed. The stiffness of the parietal bones was measured by using the 10 calvaria of specimens 11 through 21 of Table 1. The stiffnesses of the first seven calvaria lvere not assessed. In order to measure the stiffness, the parietal bones were excised and suspended concal e up on a tripod of lo\\-friction rollers, as shown in Fig. 1. A downward force was applied at the inner surface of the parietal eminence to produce a deflection of 1.25 mm. The fierce l\as cycled four times at a strain rate of 03 mmimin in order to precondition the bone according to the method of Fung.’ The stiffness was recorded on the fifth cycle in units or grams force per unit of displacement. The bone \vas kept moist at all times, and the test5 were perf’ormed at ambient temperature. In order to assess the modulus of elasticity, specimans that were 2 mm Ivide by 23 mm long were cut from the full thickness of bone by means of a technique described elsewhere.‘, ’ Since tile specimens had to be kept wet during testing. and since the thickness varied over the length of each specimen, the specimens were tested in three-point bending with the use of an Instron Model TT-C material testing machine under a constant saline drip. The Method of Linit Loads leas used to compute the elastic modulus in bending.’

The fetal cranium does not exhibit the triple-layel structure of the adult skull with the inner diplo@ layer, but rather is a single layer with varying porosity. The bone also exhibits a distinct fiber pattern (Fig. 2). A radial pattern that originates from the bone’s center 01 ossification (e.g., from the parietal eminence) can be seen clearly in fetal skull bone.” Thus, care is required to record the direction of the fibers with respect to the long axis of the bending specimens. Intuitively, a specimen with fibers that run mostly perpendicular to the long axis of the specimen inherently will be weaker than one with fibers that run mostly parallel to it. ,4s a gross determination of general orientation of the fibers, bending specimens were classified as being either “perpendicular” or “parallel” when the majority of fibers in the specimen ran accordingly. After the bending tests, the specific gravity of most of the specimens was measured. followed by ashing for the assessment of mineral content. A total of 424 specimens were ashed; the remainder of 6.5, with a representative selection from each parietal bone, were prepared for histologic analysis. Both bulk and specific densities (the former includes interporosity water. whereas the latter does not) were measured bv means of tvater displacement from a pycnometer bottle and the wet and dry weights of the bone. The ashing procedures were standard and are described else\vhere.’

Results The measured values of stiffness. modulus. density, and ash are given in Table II. The stiffness given is the average for the two parietal bones of each calvaria. Fig. 3 compares stiffness with gestational age. Linear regression results in a correlation coefficient of r = 0.79 (p < 0.01). Fig. 4 compares stiffness with fetal weight and an even closer correlation is found (r = 0.92, p < 0.005). Fig. 5 compares the modulus of elasticity with fetal weight; a positive trend almost exists for parallel-fiber specimens (r = 0.49. p = 0.05+), whereas a stronger correlation is found with perpendicular-fiber specimens (r = 0.70. p < 0.01). However, if a correlation between stiffness and modulus is tested. no correlation is found. as shown in Fig. 6 (r = 0.40 for parallel-fiber elements, and r = -0.14 for perpendicular-fiber elements). Fig. 7 compares stiffness with bone thickness. and a significant correlation is found (r = 0.68, p < 0.05). Fig. X shows that the same correlation exists between bone thickness and fetal weight (r = 0.68. p < 0.05). The length of the anteroposterior chord of the parietal bone, another measure of fetal development, was significantly correlated with stiffness (r = 0.82, p < 0.01).

Properties

Fig. 2. The

mechanical properties of flexed because of the fibrous nature such as the parietal eminence. h, The pattern. c, The inner surface is more blood vessels pass are seen throughout artifacts caused by the knife used in

Table

I. Biographical

and anthropometric

of fetal

cranial

bone

709

fetal skull bone depend on the direction in which the bone is of the bone. a, The libers radiate from centers of ossification. voids through the exterior surface of the bone define the fiber convoluted than the outer surface. d, The voids through which the thickness of the bone. (The horizontal lines in d are cutting.)

data on the specimens

Calvarium No. *

sex

1

40

M

2 3 4 6

3X 27 25 28

F

7 II

40 40

12

40

13 14

3.1 IX 2.X00 970

t t

1.02.i

0.76 0.71 o.ti9 0.41 0.64

M

3.640

OAti

t

M

4.054

I .02

I”1

F

2,378

0.69

I12

34 30

M F

2,178 1.49 I

0.9 t 0.5 1

15 16 17

42 39 30

F F M

2,850 2.922 987

0.66 0.79 0.71

11x 124 7x

I8 20

40 33

F M

3.612 I .7x3

1.14 0.X6

I 04

21

20

F

1.400

0.7 I

96

The length near the level

M M M

6.50

of the anteroposterior chord is the longest straight-line of the parietal eminence, approximately co-linear with

*Biologic degradation tNot measured.

or experimental

procedures

rendered

unusable

Congenital heart failure Cnknown Insuflicient respirator) effort Prematurit? Brollchoplllmorlar~ dysplasia. sepsis Meconium aspiration Congenital heart disease. meconium H ydrops fetalis (nonimmune) Sepsis-plieumonitis Respiratory distress skndrome. sepsis H!poplastic left heart Congenital heart disease (:~stadenonlatoitl malformatlon in lung Diaphragmatic hernia Respiratory distress syndrome Respirator\ distress syndl-ome

t t t

X6 94

t

measure between the occipitofrontal those

specimens

the

temporal

and

lambdoidal

diameter. whose

numbers

are

missing.

sutures

710

Kriewall

3000l

r = 0.79

2400 -

p
0

a z

l

6000

t

0

2 2

1200

-

z ‘G 0-l

600 -

0

/

LA 0

0'

'

'

'

'

Gestational Fig.

lated

I

' 25

I

I

' 50

Gth

gestational

age.

0 r =

pendictllar-fiber

specimens.

e

1800

.z 5

1200

/

l The

l t

/

m

tainly

0

l

l l

0

l-‘-&L SW

2700

3600

4500

Birth Weight (g ) 4. The

stiffness is highly correlated

Fig.

\vcight

(r

= 0.8s.

that Gth

r

0.X?

llte

general

of

the

bone’s

material

properties

are

equivocal. Fig. 9 shows that no correlation is evident between stiffness and ash content. Fig. IO indicates that a significant negative correlation exists between stiffness and dry density (r = 0.69, p < 0.05). but no corrclation is evident between stiffness and bulk density. The remarkable feature of Fig. 10 is the consistency of the densities across subjects. No significant correlations were found between either the parallel- or perpendicular-tiber specimens’ moduli and the bulk or specific. densities of the bone. However, the modulus was significantly correlated with gestational age (r = 0.66, p < 0.01) for parallel-fiber specimens. The modulus was

biparietal (r =

be

to

lxocluct

developmental

siTe, can ccrof’ bone

;Iiitc~i~o1~ost~rioi-

highly

Gth

Indi\-idually,

and

birth

thick-

chord

c orrelated

p < 0.005). thickness

mea-

our

lveight

is birth

are

data corre-

= 0.68 (11 < 0.0.3) and that anrei.oposterior ,~ntl birth ivcigllt arc correlated, uitli

(1) <

0.01

). PI‘lli~.

enlar~~mcnt

vclopmcnt. effects

the.

to

bcmc

Iciigrli

I‘ =

I I

;LS 5ve

of’ factors.

and general T‘lle

of

in

chord

with hirth weight.

icngtli

boric

dtic

stiffness.

sho\cn

latecl,

parirtal

b\ a number

character-istics

rllr

time\

indicate I600

;I ~\holc

such as thickness

attcct

nc’s\

l

600

gcomc.tric

attributes.

/-

of

(ai1 1)~ inlluenccd

it

It5

t

0

sliffnca\

5urctl

t

t!z

The

respecrively).

0.92

p <0.0005

Fig.

Fig. 5. The hone‘s modulus of elasticiry does not show the high correlation with birth tveight that stiff’ness does.

i

T

.,-

4500

also sianific-anti\ correlated with ash content (r = 0.66, p < 0.005 and r = 0.65. p < 0.0.3 for parallel and per-

2400

C

2700 3600 Weight (g)

1800

Birth

3000

0

900

Age (weeks)

stiffness of whole parietal bone is linearly corre-

3. The

n

~Jorclaan!’ tliaincrrr

of of

has also

cohort, rhc

is attributable

cranitlm

sho\vn.

iol-

I:, tori-c.1atc.d

jritli csam1Ac, \vitli

to

fetal

de-

tbar

the

birth

veighr

0.51,).

Besides geometric characteristics, the inherent material properties of’ bone can affect the measured stiffness. One such parameter that is independent of the geomet2-\ This per

of

lxirameter unit

area)

the

materiai relates induced

is the the in

modulus

streah the

of‘

(defined

material

elasticitv.

as the to

the

force

applied

strain (defined as t11e amount of’ deformation per tinit 01‘ original length). For cxam1~le. a beam of oak and a beam of steel of exactly the same dimensions \vill ha\-e I astlv difftrent stit’fness hecausc of their differ-cnt moduli of tlasticitv. ‘l‘he elastic modulus of steel is on

Properties

3000

-

2400

-

of fetal

cranial

bone

711

0

a \0

1600

-

1200

-

600

-

zi E .3j

0 0 00

0’

0

’ 0,

, @I

2000

,

4000

,

,

,

6000

,

,

8000

lO,OOO

Modulus ( MPa 1 Fig. 6. No significant

correlation

between

parietal

bone

stiffness

and

its modulus

of elasticity

is

evident.

Table

II. Measured

values of stiffness,

modulus

of elasticity,

Modulus Calvanum No. 1 2 3 4 6 7 11 12 13 14 1.5 16 17 18 20 21 The measurements some of these values *Not measured.

stiffrMw (,&i/mm)

PUlllel fi bm

* * * * * * 2,080 690 WI 140 1.440 1.780 340 2,.560 230 420 of stiffness are missing.

density

were

= megapascal:

6,895MPa

= lO”pounds-forceisqinch.

SPPClJiic

* * * * * * I..56 1 ..‘,‘i 1.55 1.56 l.tiX 1.59 1.57 1.58 1 .w 1.70

to the experimental

order of 2 1,000 MPa,* and that of dry white oak is 1,200 MPa. Many factors. in turn, affect a material’s modulus of elasticity, with the material composition and its structural organization being the most important. Currey”’ has shown a correlation between mineral content, as measured by ash weight, and the elastic modulus of various bony structures. Indeed, our findings corroborate Currey’s in that the modulus was significantly cor-

(gmlml)

Bullc

1.740 X70 180 I20 140 3.300 * I.684 1,646 1 . 6%* * 1,044 1.434 I.996 1 1-938

the

*MPa

lkw.\2t~

Puprndiculnr fibers

added

and ash content

(MPn)

4.010 4.240 940 1 .wo 3,620 3.720 2.961 3.594 3.3YO 3,634 7.360 4.30 1 3,478 5.167 3,029 4,232 and

densities,

procedure

* * * * * * 2.07 2.14 2.13 2.2.5 2.09 2.12 2.14 2.04 2.15 2.36 midway

through

A.\/1 (% dns wrqhilt) 61 61 55 5’4 ii) 6T 64 63 61 62 64 62 65 63 57 .‘,?I the investigation:

thus.

related with ash content. The effect of fiber orientation on the modulus of fetal parietal bone has been shown qualitatively,’ also. Of the two. then, is it the increase in the structural characteristics of the bone, for instance, anthropometric factors, or the material characteristics, for instance, variations in the modulus, that most affects the stiffness? The stiffness is strongly correlated with birth weight (Fig. 4). In contrast, the elastic modulus is not (Fig. 5). A test of direct correlation between stiffness and modulus demonstrates that, in fact, there is none

712

Kriewall

3000

r

3000

r =0.44

2400 -2 E \ 0 -

2400

1800

NS

1800 I

0

600 t

1 0’

l I

I

I

I

I 40

I

Iq

0 I

I 80

Ash Content ( % dry weight 1 Thickness

( mm )

7. Stiffness is correlated with the geometric characteristics of the bone. such as its thickness. Fig.

1.25

1.00E z YE 0 ‘i l-

.75-

.50 -

0

r = 0.68 pco.05

.25 -

t 01

’

’ 900

’

’ 1800

’

’ 2700

’

’ 3600

’

1 4500

Birth Weight (g) Fig. 8. The thickness of the bone increases with birth weight and, in part, is responsible for the stiffening of the bone.

(see Fig. 6). In terms of structural characteristics, Fig. 7 reveals a strong correlation hetween stiffness and hone thickness, whereas Fig. 8 reveals that thickness is correlated with birth weight. Stiffness also does not seem to he correlated well with other material parameters, such as ash content or hone density, as is reflected by Figs. 9 and 10, respectively. Interestingly, although the ash content of fetal skull in this study ranged between 55% and 68% dry weight. Currey ” found that the ash content of adult cortical bone varied between 63% and 68%. Therefore, fetal

Fig. 9. Since the ash content is fairly consistent across subjects. no significant correlation is apparent bettveen stiffness and mineralization of the bone.

bone does not lack basic mineralization. In summary, anthropometrir factors Seem to account more for the increase in load-hearing capacity than do the material characteristics. What, then, affects the load-bearing capacity or stiffness of the fetal cranium? Fetal growth is certainly a factor. The measured fivefold to tenfold change o\.er the last 10 weeks of‘ gestation explains why preterm infants are at risk for neurological trauma as a result of birth. L,indgren,” among others. has demonstrated that molding is caused hy the sof’t tissues of the pelvis. Since the pressures of‘ a preterm labor are as strong as those of a term lahol-, comparable loads to the presenting part are created li)r both groups of infants. Thus. clinical findings, such as those reported hv Fortune and Kitchen,“’ wherein very lots,-birth weight infants had a high incidence of lethal, serious. and trivial malformations, can be explained structurallv. Since the loads are the same, but the stiffness is markedly louver in preterm infants, greater deformations \vill OCCIJI-\vith them than with term infants. In conclusion. the ability of the f&al head to withstand traumatic insults increases dramatically with hirth weight. The statistical difference between the a\.erage stiffness of infants who weigh less than 2,500 gm and that of those who weigh more than 2,500 gm is highly significant (p < 0.0001). Growl th factors affect this change more than do material factors. Bone mineral content and density do not change appreciahl! o\.cr the span of weights or gestational ages investigated. The fact that the modulus of elasticity in hend-

\‘olume

114

Number

6

3000

Properties

of fetal

cranial

bone

713

l

2400

-

Dry Density r = -0.69

.

0

p < 0.05 o Bulk Density r = 0.10

-&

1800-

NS

$ E=

.

0 0

1200

-

. . ol”““‘l”fil

0 ,000

1.60

i

5

.

P 4

:

r = 0.88 p-Co.005

25

2.40

Density (g/cm31 1800

Fig. 10. Again, the bulk and dry densities do not vary dramatically across subjects. Therefore. significant correlations are not expected.

ing

for

parallel-fiber

specimens

is significantly

greater

than that for perpendicular-fiber specimens (p < 0.00005) indicates that loads applied in directions other than normal might have severe consequences. This could happen with malpresentations and with the application of forceps. Finally, these data indicate that the application of forceps to “protect” the preterm infant’s head may be contraindicated. In fact, soft tissues may cause neurological trauma before the forceps can be applied. In addition, point contact between the cranium and the blades of the forceps can cause extreme deformations that may be unappreciable to the obstetrician: what he feels is the sum of all the forces applied to the forceps. For a vaginal delivery, the fetal head, especially that of the preterm infant, will fare better when intrapartum loads are distributed smoothly over its cranium in directions along normal bending lines. Of course, more than mechanical trauma can cause neurological morbidity. Premature infants are at increased risk from many factors, including incomplete

Birth

2700

3600

4500

Weight (g)

Fig. 11. To demonstrate the growth of the bone, the product of the parietal bone’s average thickness and length of the anteroposterior chord was correlated with birth weight. The correlation is highly significant.

cardiovascular changes after birth (e.g., patent ductus arteriosus), anoxia due to immature respiratory development, and intracranial hemorrhage due to an immature germinal matrix of the central nervous system. The intent of this report has been to describe only the etiologic factors of neurological sequelae that are associated with the biomechanical properties of the fetal skull. The able assistance of graduate students Donita Bylski, Jay Goldberg, and Jim Bulgrin is acknowledged. I would like to thank Alan Tsai, Ph.D., in the School of Public Health, for his assistance in gaining the ash results, Mason Barr, M.D., Co-director of the Teratology Unit, for gathering tissue specimens, and Bruce A. Work, Jr., M.D., for his counsel during the study.

REFERENCES

1. Cranfrani, T.: A Short History of Obstetrics and Gynecology, Springfield, Illinois, 1960. Charles C Thomas, Publisher. 2. Fine, E. A., Bracken, M., and Berkowitz, R. L.: An evaluation of the usefulness of x-ray pelvimetry: Comparison of the Thorns and modified Ball methods with manual pelvimetry, AM. J. OBSTET. GYNECOL. 137:15, 1980. 3. Kriewall, T. J.. and McPherson, G. K.: Effects of uterine contractility on the fetal cranium, in Milunsky, A., Friedman, E. A., and Cluck, L., editor: Advances in Perinatal Medicine, New York, 1981, vol. 1, Plenum Medical, pp. 295-356.

4.

Kennedy, size and

J. L., and Greenwald, E.: Correlation of shoe obstetric outcome: An anthropometric study, AM. J. OBSTET. GYNECOL. 140~466, 1981. 5. Kriewall, T. J., Stys, S. J., and McPherson, G. K.: Neonatal head shape after delivery: An index of molding, J. Perinat. Med. 5:260, 1977. 6. Fung, Y. C. B.: Stress-strain history relations of soft tissues in simple elongation, In Fung, Y. C. B., Perrone, M., and Anliken, A., editors: Biomechanics, Englewood Cliffs, New Jersey. 1972, Prentice-Hall. 7. McPherson, G. K., and Kriewall, T. J.: The elastic modulus of fetal cranial bone: A first step towards an under-

714

Kriewall

standing of the biomechanics of fetal head molding. .J. Biomech. 13:9, 1980. 8. Kriewall. T. J,. McPherson, G. li., and Tsai, A.: Bending properties and ash content of fetal cranial bone, J. Biomech. 14:73, 1981. 9. Jordaan, H. \‘. F.: The dif’ferential enlargement of the neurocranium in the f’ull-term fetus, S. Afi-. Med. J. 50: 197x. 1976. IO. Currcy, J. D.: Mechanical properties ot‘bonr tissues \vith grearly ditf‘ering functions, J. Biomech. 12:313, 1979.

1 1. Currey, J. D.: The mechanical consequences of variation in the mineral content of‘ bone, J. Biomech. 2: 1, 1969. 12. Lindgren, L.: Effects ot pressure gradient on the feral cranium, in Milunsky, A., Friedman. E. A., and Gluck. L., editors: Advances in Perinatal Medicine, New York, 1981, vol. I, Plenum Medical, pp. 357-482. 13. Fortune. D. IV., and Kitchen. N’. H.: Malformationa in infants 01’ \rry ION birth Meight. Med. ,J. Aust. 1:239, 1977.