I34

THE

BRITISH

JOURNAL

and function and that the neoplasm results not from
a single disordered cell, but from the activity of this
disordered system. As a neoplasm grows it tends
to strip the periosteum to cause reactive bone
formation.
Many bony tumours, I believe, are a combination
of neoplastic and reactive bone. Right-angle spiculation, so commonly seen in body tumours, is much
more likely to be reactive than neoplastic. There is,
in bony tumours as in neoplasms elsewhere in the
body, a wide variance in the histological picture:
not only from one growth to another but from one
part of a growth to another part. The impression
that I gained from this series is that every phase in
the development of bone is paralleled by its neoplastic
counterpart.

SUMlMARY
Twenty cases of different types of osteomata of
the skull are described and the histological pictures
portrayed in some cases. The life-history of a
tumour has been known for 10-20 years, and an
attempt is made to give some account of histopathological correlation to the various types of
neoplasms that arise primarily in the bone-forming
organs of the skull.

I would like to thank Mr. R. Ridley of
the Department of Photography, King’s College,
Newcastle upon Tyne, for the quality of the illustrations, and Miss M. Bousfield and Miss A. Cairns
for their help with the preparation of this paper. I
am also grateful for a grant from the Research Fund
of the Royal Victoria Infirmary.

OF

SURGERY

REFERENCES
ABBOTT,
K. H., and COURVILLE,
C. B. (1939)~
Bull. Los
Angeles neurol. SOC.,4, 101.
_ _ - - (1945),Zbid., 10, 19.
ANDREW,
J. (1956)~
Brit. 3.Surg., 43, 489.
ARMITAGE,
G. (IS~I),Zbid., 18, 565.
BAKER,
S. L. (I950), A Text-book of X-ray Diagnosis, ed.
S . C. Shanks and P. Kerley, 4, 55. London:
Lewis.
J. (1903), Tumours, Innocent and
BLAND-SUTTON,
Malignant:
their Clinical Characteristics and
Appropriate Treatment. 3rd ed. Chicago : W. T.
Keener & Co.
CRUVEILHIER,
J. (1856), Traitd Anat. pathol. gen., 3, 869.
Paris : Baillikre.
DOLBEAU,
H. F. (1871),Mim. Acad. M i d . Paris, 30, I ;
original contribution, Bull. Acad. M i d . Paris, 3 I,
1076.
ECHLIN,F. (I934), Arch. Surg., 28, 357.
GALEN
(1824)~
Opera, ed. C. G. Kuhn, 7, 728. Leipzig :
Cnobloch.
F. K. (I953), Helv. chzr. acta, 20, 83.
KESSEL,
KNAPP,H. (ISSI), Trans. med. SOC.St. N.Y., 21, 252.
KROGMAN,
W.M. (I940),Bull. Hist. Med., 8, 28.
MAFEY, G.(I~IO),“Contribution a l’ktude d’une variktk
particulikre d’exotose de la voiite cranienne, le spongiostkome”. Thbse de Paris, No. 87.
MOODIE,
R. L. (1926),Ann. med. Hist., 8, 394.
PAGET,
J. (1865),Lectures on Surgical Pathology, 3rd Amer.
ed., 477.
RICHET,D. D. A. (1871),Bull. Acad. M i d . Paris, 36,
564.
ROKITANSKY,
C. (1850), A Manual of Pathological
Anatomy, 3, 122. London : The Sydenham Society.
ROWBOTHAM,
G. F., and CLARKE,
P. R. R. (I955),Brit.J.
Surg., 42, I.
VIRCHOW,
R. (1865),Die Krankhaften Geschwiilste, 2, 30.
Berlin : A. Hirschwald.

INTERNAL CAROTID THROMBOSIS AND SPONTANEOUS
PNEUMOCEPHALUS AFTER HEAD INJURY*
BYR. ALEX DAWS
FROM THE DEPARTMENT OF NEUROSURGERY AT THE CARDIFF ROYAL INFIRMARY

THROMBOSIS
of the internal carotid artery was at
one time considered to be a rare occurrence.
Cerebral angiography, however, has corrected this
false impression and the condition is now frequently
demonstrated during life. The various clinical
manifestations which may follow occlusion of this
vessel and the angiographic appearances are well
established.
Internal carotid thrombosis may be due to several
causes. Elvidge and Werner (1951)~
from a study
of the literature, stated that arteriosclerosis and
thrombo-angiitis obliterans were the two most
frequent aetiological factors. Thrombosis of the
carotid vessels has been described as a result of direct
trauma of the wall of the artery by penetrating
wounds of the neck (Whalley, 1946; Caldwell and
Haddon, 1948). Braudo (1956)has reported internal
carotid thrombosis in children after injury in the
region of the soft palate. Thrombosis may also
follow non-penetrating injury to the neck (Northcroft

* From the Department of Neurological Surgery,
Morriston Hospital, Swansea.

and Morgan, 1944; Schneider and Lemmen, 1952;
and Clarke, Dickson, and Smith, 1955). Boldrey,
Maass, and Miller (1956)have discussed compression of the internal carotid artery by the lateral
process of the atlas as an aetiological factor in some
cases of the disease. Little attention appears to
have been paid to the fact that head injury may play
a direct part in precipitating thrombosis in this
artery. The late Sir Hugh Cairns (1942) reported
a case which followed a head injury, and quoted
another case from the literature. Sedzimir (1955)
has reported a further 7 cases in which the thrombosis resulted from trauma to the skull.
The importance of injury to the head in a case
of internal carotid thrombosis is almost impossible
to assess when the trauma occurs long before clinical
evidence of such a lesion becomes apparent. When,
however, thrombosis of the artery with disturbance
of neurological function follows immediately, or
within a few hours of a head injury, the relationship
can be accepted. The following report describes a
case of head injury which was complicated three hours
after the accident by internal carotid thrombosis

INTERNAL

CAROTID

THROMBOSIS

I35

and, in addition, by spontaneous pneumocephalus.
As in the case reported by Cairns (1942), and in
reported by Sedzimir (1955),a
three of the
fracture of the base of the
close
the
at autopsy*
carotid artery was
In the
reported here thrombosis had Occurred
just below the base of the skull and had involved the

AT OPERATION.-The
usual exploratory burr-holes
were made, but there was no evidence of an intracranial
Air
was
found in the subdural space
collection of blood.
on both sides. The brain was under increased tension
and the intraventricular pressure was high. The right
hemisphere, in the frontal and temporal regions, felt soft
to the exploring cannula, and necrotic brain oozed through
the ruptured pia on withdrawing the cannula.
The patient’s condition slowly deteriorated and he
died on March 17, 1956.
AuToPsY.-There was a fissure fracture of the base
of the skull in the middle fossa. The fracture line passed
from the squamous temporal bone on the left through the

A

B

FIG. 167.-A,

Anteroposterior, and 6, lateral, radiographs of the skull showing extensive pneumocephalus.

intracranial segments of the artery. Distally t h e
thrombus entered t h e right anterior and middle
cerebral arteries.

CASE REPORT
The patient, a 55-year-old man, was admitted as an
emergency on March 13, 1956. He had been struck on
the left side of the face by a plank of wood and was thrown
backwards striking his head on a wall. Consciousness
was lost for only a few seconds.
ON EXAMINATION.-oII admission to hospital he was
conscious but a little confused. There was an extensive
abrasion over the left side of his face and bruising of the
scalp in the right occipital and mastoid regions. The
mandible was fractured near to the symphysis. As a result
of this, and the associated swelling of the tissues in the
floor of the mouth, there was some respiratory obstruction
with slight cyanosis. No abnormal physical signs were
demonstrated in the central nervous system at that time.
A clear airway was established and the patient transferred
to the ward for observation.
Three hours after admission his condition deteriorated.
There was slight clouding of consciousness and he
developed a left hemiplegia with impairment of sensation
on this side. X-ray examination of the skull revealed
extensive spontaneous pneumocephalus (Fig. 167) but no
fracture was demonstrated. There was no clinical
evidence of injury to the neck and normal carotid artery
pulsation was judged to be present on both sides. There
was little further change in the patient’s condition until
the evening of March 15 when he developed signs of
raised intracranial pressure. The right pupil became
dilated, the plantar responses extensor, and he rapidly
passed into coma.

foramen spinosum, foramen lacerum, across the sella
turcica, and into the right optic canal. The line then
passed backwards through the right foramen lacerum,
foramen ovale, and along the anterior surface of the
petrous temporal bone. There was a small loose fragment
of bone in the anterior wall of the sella to the right of the
midline. In spite of a careful search the site of the dural
tear could not be located. The sphenoid sinus was full
of blood-clot.
There was extensive infarction of the right cerebral
hemisphere. The right internal carotid artery was
occluded by a thrombus commencing just below the base
of the skull. The intracranial portion was thrombosed
and the thrombus had extended into the right anterior and
middle cerebral arteries. On histological examination
atheroma was present in the internal carotid, most marked
in the region of the siphon. In that part of the artery,
just proximal to the siphon, a longitudinal tear in the
media and intima was demonstrated. Thrombosis had
occurred in this region and had extended both proximally,
occluding the last z cm. of the cervical portion, and also
distally into the intracranial portion, the middle and
anterior cerebral arteries.

DISCUSSION
T h i s case presented a n interesting problem i n
differential diagnosis. Hemiplegia following a lucid
interval after a head injury, and in t h e absence of
obvious injury t o the neck, is normally regarded as
due to intracranial haemorrhage. Such a possibility
was considered, but seemed unlikely as the hemiplegia was known t o have been of sudden onset, and
was initially complete. Moreover, there was no

136

THE

BRITISH

JOURNAL

significant disturbance of consciousness. On these
grounds thrombosis of a major artery, probably the
right internal carotid, was diagnosed. No difference
in carotid pulsation could be detected on the two
sides of the neck. Internal carotid pulsation, however, is difficult to assess and Gurdjian and Webster
(1953) did not think that differences in the degree of
pulsation on the two sides could be depended upon
for an accurate diagnosis.
Deterioration in the patient's condition forty-eight
hours after the onset of the hemiplegia, with signs of
raised intracranial pressure and deepening coma,
cast doubt on the diagnosis and the usual exploratory
burr-holes were made to exclude an intracranial
collection of blood. Autopsy revealed the presence
of internal carotid thrombosis and showed that the
subsequent rise in intracranial pressure was due to
infarction of the cerebrum.
When cerebral infarction and associated swelling
of the hemisphere occurs as an early feature in cases
of internal carotid thrombosis, particularly when there
is a history of head injury, it is impossible to exclude
an intracranial hamorrhage on clinical grounds alone,
and surgical exploration is necessary. A gross
degree of cerebral swelling, however, is not often
encountered in occlusion of the internal carotid
artery. The early direct extension of the thrombotic process to involve the vessels of the circle of
Willis was undoubtedly responsible for the
extensive ischamic change with gross disturbance
of neurological function which occurred in this
case.
Intimal tears of the internal carotid artery have
been reported by Verneuil and Greco (quoted by
Schneider and Lemmen, 1952), by Caldwell and
Haddon (1948) and Northcroft and Morgan (1944).
I n all these cases damage to the wall of the
artery followed injury to the neck. I n the case
reported here there was no evidence, either clinically
or at autopsy, of trauma to the neck. The exact
mechanics of the head injury could not be ascertained
in this patient, though it is probable that the tear
in the artery wall resulted from strains transmitted
directly to the vessel by the head injury (Sedzimir,
1955).

Traumatic pneumocephalus, although now recognized with increasing frequency, remains a relatively
rare occurrence. While there are other more
unusual causes of the condition it is most commonly
found in cases of closed head injury with fractures
involving the paranasal air sinuses. I n a survey of
308 such cases by Lewin (1954) intracranial air was
demonstrated in 24 of the patients. The occurrence
of pneumocephalus depends upon a fracture of the
wall of an air sinus and a tear of the overlying dura
mater and arachnoid, thus producing a communication between the meninges and the exterior. Increased intranasal air tension must also be present

OF

SURGERY

to force the air through the fistulous communication
into the intracranial cavity where it may become
trapped in the subdural space, subarachnoid space,
ventricles, or occasionally in the substance of the
brain. I n this patient the partial respiratory
obstruction which was present on admission must
have been largely responsible for the very early
occurrence of the pneumocephalus. I n spite of a
careful search at autopsy the site of the dural tear
could not be identified. There was a loose fragment
of bone in the fracture line in the anterior wall of the
sella turcica. As the sphenoid sinus was the only
air sinus involved by the fracture the fistula must
have been in this region. Cerebrospinal rhinorrhea was not present in spite of the gross degree of
pneumocephalus. This was surprising as profuse
rhinorrhea is usually characteristic of sphenoidal
sinus dural tears owing to their anatomical relationship to the cisterna chiasmatica, as has been pointed
out by Lewin and Cairns (1951). The absence of
rhinorrhea can possibly be explained by the fact
that at autopsy the sphenoid sinus was found to be
completely obstructed by solid blood-clot, the
organization of which might have resulted in a
delayed onset had the patient survived.
SUMMARY
A case of internal carotid artery thrombosis
and hemiplegia after a head injury is described.
Spontaneous pneumocephalus also occurred as a
complication. Aetiology and the problem of differential diagnosis are briefly discussed.

I wish to express my thanks to Mr. Norman
Whalley for help in the preparation of this paper,
and also for permission to publish the case admitted
under his care. I would also like to thank Professor
Lambert Rogers for his help and encouragement.
REFERENCES
BOLDREY,
E., MAASS,L., and MILLER,
E. (1956),J. Neurosurg., 13, 127.
BRAUDO,
M. (1956), Brit. med. J., I, 665.
H. (I942), Lisboa me'd., 19, 375.
CAIRNS,
H. W., and HADDON,F. C. (1948), Ann.
CALDWELL,
intern. Med., 28, 1132.
CLARKE,
P. R. R., DICKSON,
J., and SMITH,B. J. (1955),
Brit. 3. Surg., 43, 215.
ELVIDGE,
A. R., and WERNER,
A. ( I ~ s I ) Arch.
,
Neurol.
Psychiat., Chicago, 66, 752.
GURDJIAN,
E. S., and WEBSTER,
J. E. (1953),J. Amer. med.
Ass., 151, 541.
LEWIN,W. (1954), Brit. J . Surg., 42, I .
- - and CAIRNS,H. ( I ~ s I )Brit.
,
med.J., I , I .
NORTHCROFT,
G. B., and MORGAN,
A. D. (1944), Brit. J .
Surg., 32, 105.
R. C., and LEMMEN,
L. J. (1952)J.
SCHNEIDER,
.~
. Neurosurg., 9, 495.
SEDZIMIR,
C. B. (1955),3. Neurol. Psychiat., 18, 293.
N. (1946), Brit.J. Surg., 33,400.
WHALLEY,