736 SA MEDICAL Jo R TAL 26 April 1975
An Evaluation of the Radiological Signs
of Fetal Death
M. I. SHAFF
SUMMARY The radiological sign of fetal death are in essence a
study of the processes of maceration, and the number of
Radiological investigation is warranted in the antenatal signs present depends to a large extent on the degree of
patient only if the findings are likely to influence future maceration. A radiological consultation in cases of sus-
management. The major radiological signs of fetal death pected intra-uterine death must be aimed at arriving at a
are gas in the fetus, overlapping of the cranial bones definite conclusion as to the state of the fetus. Vague
and Deuel's halo sign. Gas patterns in the portal and reports are of no assistance to an obstetrician faced with
umbilical vessels are unique and pathognomonic. Deuel's so important a problem, and only serve to undermine the
halo sign would be found more commonly if the correct confidence of the clinician in the value of X-ray investiga-
radiographic view was employed. Spalding's sign is of tion and interpretation.
no value before 20 weeks or if the fetal head is subjected To demonstrate the signs of fetal death. an under-
to pressure. Hyperflexion of the fetus more often indicates standing of radiographic technique is a important as
intra-uterine growth retardation than death in utero. the ability to interpret radiological signs. All radiographs
must be of high quality and we routinely use the prone
oblique position with the fetal back closest to the film
S. A/r. Med. J., 49, 736 (1975). holder. A tight compression band is required to displace
maternal tissues. and a 3-cm reduction in tissue thickness
enables the incident dose to be halved for a given exit
Consequent on the demonstration of the potential hazards dose.' A low k V technique is used as this provides the
of X-ray diagnosis in the pregnant mother and the greatest fetal detail. The resultant increase in radiation dose
development of isotopic and ultrasonic scanning techniques. concomitant with these settings is negligible. Additional
there was a fall-off in the number of obstetric referrals views may be required to elicit specific signs and these
to X-ray departments. In recent times. however, as the are detailed below where appropriate. If cases for radio-
values and limitations of ultrasound have become better logy have been properly selected. these additional views
defined, the use of radiology is increasing.' Ultrasonic are justified. The 3 most valuable signs of intra-uterine
diagnosis requires technical skill and suitable equipment. death are: (i) intravascular or intrafetal gas: (ii) overlap-
lts results, particularly with the grey scale, are impressive.' ping of the cranial bones (Spalding's ign): and (iii) Deuers
However. in the diagnosis of fetal abnormalities and fetal halo sign.
maturity after the 34th week there may be marked errors.
Furthermore. even in the assessment of fetal death by
Doppler ultrasound technique there is an error of 1 - 5 0 6.'
M ost obstetricians realise that radiology. isotopes and
Gas in the Fetus
ultrasound are complementary to each other. and that the
choice of the order of investigation can only be made by This is the most important and earliest sign of intra-uterine
a careful evaluation of the form of imagery best suited death, being present in 84°~ of cases within 10 days and
to solve the problem at hand. Even then the availability may be seen as early as 12 hours after death." The common
of equipment and skilled operators is a limiting factor. sites for the accumulation of gas are heart. aorta and
Radiology offers a simple. non-dangerous. readily available portal veins. Less common sites such as the ventricles of
and easily acquired method of diagnosing fetal death. the brain. joints_ pleura and retroperitoneal tissues are
fetal abnormalities and fetal maturity (after 34 weeks' encountered from time to time. The reliability of intra-
maturity). Our attitude to antenatal radiology is that if fetal gas a an indication of fetal death is beyond
the indication is established by careful clinical evaluation. question. Film must be carefully scrutinised lest intra-
a course of radiographic positions and ettings must then fetal gas go undetected. The difficulty that may be encoun-
be chosen. aiming at films best suited to solve the clinical tered is diflerentiation of maternal bowel shadows from
problems. Radiological inve. tigation is unwarranted if intmfetal gas. In this regard two points must be emphasised.
the findings are unlikely to influence further management. The first is the radiogr:.lphic positioning. A lateral film.
This article will re iew the signs of fetal death. The other using a compression band placed tightly over the maternal
aspects of fetal radiology will be subjects for further abdomen to displace the matern:.ll bowel, is of great assis-
publications. tance in difficult cases. The second is the detection of patho-
gnomonic and often neglected gas patterns in the umbilical
Department of Radiology. Groote Srhuur Hospital, Cape Town vessels or in the portal radicles. Samuel and Gunn G
\1. I. HAFF. :\f.B. B.CH._ D.:\f.R.D. de cri bed ga in the umbilical vessels in 5 out of 15 reported
Date received: 5 February 197-. cases of fetal death. More recently. Gruber' noted thi
26 April 197 s MEDJE E TYD KRIF 737
fealure in 3 conseculive cases. and in I of these it wa the palding's sign may be achieved by dependant radiography
'iole sign of felal dealh. He was further able to describe in vertex presentalions.
lhe specific paiterns of gas in the umbilical veins and
arteries. Gas in the umbilical arteries forms a double or
single helix, depending on whether one or both arteries are Deuel's Halo Sign
involved. Gas in lhe umbilical veins has a linear undulating
appearance likened to a discarded or kinked piece of In 1947 Deuel described an appearance of the fetal skull
string (Fig. I). These signs must be sought suboccipitally reminiscent of medieval paintings of Madonna and Child.
as the cord often surrounds lhe fetal neck. When seen. He called thi appearance 'Heiligenschein' (Fig. 2). It result
these signs are conclusive and establish a diagnosis which from accumulation of fluid beneath the loose subgalea
might otherwise have remained in doubt. Portal vein gas and subsequent displacement of the fat stripe lying in the
is also a common and helpful sign. and fine linear branching subcutis. This sign take al least 3 day to develop.
in the portal radicles must be searched for in all doubtful Different authors aitribute a 38 - 90 0 0 incidence to the
cases. halo sign: We feel that this wide discrepancy arises entirely
from lhe techni.que used for its detection. A small-coned
anteroposterior film with maternal compression over the
fetal head, maximum 85 kVP, and a bucky grid as
described by Ohlson is essential for its detection. to Where
the halo sign is not evident on plain film radiography it
may be inferred by amniography (Fig. 3). This technique
enables fetal soft tissues to be studied directly. The halo
sign must not be confused with perivesical fat or a caput
succidaneum. The former may be differentiated by
virtue of the fact that it diverges from the fetal cranium,
and the laiter is localised and found only in labour.
Hydrops fetalis remains the one condition in which the
halo sign may occur. the fetus still being alive (Fig. 3).
Under these circumstances, however, it is a strong progno -
licator of impending fetal death.
Fig. 1. The 'discarded string sign' due to gas in tbe
The nature and ongm of intrafelal gas is uncertain.
The most acceptable viewpoint is that it is composed mainly
of carbon dioxide liberated by reticulocytes capable of
anaerobic metabolism. That gas may go back into solution
has been well documented by other author .6
Fig. 2. Overplapping and malaJignment of the bones of
Spalding's Sign the skull \'au:t with an associated halo.
Overlapping and malalignment of the bone of the cranial
vault as described by Spalding in 1922, is both less constant SPINAL FLEXION
and less reliable than intrafetal gas. It is valueless before
lhe 20th week of gestation or when the fetal skull has Minor degrees of spinal flexion are difficult to detect and
been subjected to increased pressures. as in labour, near an attitude of extreme flexion, the 'ball sign', more often
lerm and afler membranes have ruptured. The essential indicates intra-uterine growth retardation and does nOI
component of lhis sign is malalignment of the bones of necessarily indicate fetal death (Fig. 4).11,,, Frequently an
the cranial vault consequent on brain shrinkage. In lhe illusion of hyperflexion i created by virtue of the X-ray
10th month. provided the membranes are intact and labour beam passing through the long axis of the fetus. A radio-
has not commenced, disalignment exceeding 4 mm is graph in the opposite oblique position will refute or
considered to be indicative of fetal death.' Accentuation of confirm thi impression. A comment on the attitude of
73 A MEDICAL JOUR 'AL 26 pril J97-
Fig. 3. Amniographic demonstration of thickened calvarial Fig. 4. Hyper:flexion of the fetus, the 'ball sign' in a
oft tissues in a live hydropic fetus 'inferred halo'. case of intra-uterine growth retardation.
tbe fetus as an indicator of its functional state is an 3. Barnen. E. and Morley, P. (1974): Abdominal Echography. London.
essential part of the report in antenatal radiology. The
radiologist in this way may draw attention to impending
4. ~~~~ee?I~ori~s·G. B. (1973): Radiology in Obscecrics and Ance-nacal
Procedures. London: Butterwonhs.
death, and urinary oestriols and ultrasound may be used 5. Stewart. A. M. (1961): Brit. J. Radiol., 34, 187.
6. Samuel, E. and Gunn, K. (1955): Roentgenology, 73, 974.
to establish the diagnosi. 7. Gruber, F. H. (1967); Radiology. 89, 881.
Borell U. and Femstrom, 1. (195 ): Acta radial. (Stockb.), 49, 409
REFERE'CE 9. oon~n C. D. (1974): Radiol. Clin. N. Amer. 12, 29.
10. Ohlson,' L. (1962): Acta radial. (St?ckb.) 57, 57.
1. Ste" art, J. H. (1974); Aus!. Radiol., 18. 314 .. 11. Barnen E and aim, A. (1965): Bnt. J. Radlol., 38, 338.
2. Kossoff, G., Garrett. W. J. and Radovanovlch. G. (197~): Ibid .. 18. 62. 12. Croall, 'J. ~nd Grecb, P. (1970): J. Obstet. Gynaec. Brit. Cwlth, 77, 802