Chapter V (1)

INTRACRANIAL INJURIES: EXTRA- AND SUB-DURAL HÆMORRHAGES. INJURY TO THE BRAIN

HÆMORRHAGES

I. Extra-dural hæmorrhage.

Extra-dural hæmorrhage may be arterial or venous in origin. In the former case the blood is derived almost exclusively from the middle meningeal artery. When of venous origin the source is usually to be found in a laceration of one of the sinuses of the brain. When blood is poured out from a torn sinus it tends to occupy, and be widely diffused in, the subdural space. A venous extra-dural extravasation is relatively of infrequent occurrence, and the pressure exercised on the brain seldom suffices to permit of the development of local or general symptoms of cerebral compression. On these grounds extra-dural hæmorrhage may be regarded as almost necessarily of arterial origin, and as derived from a torn middle meningeal artery.

Middle meningeal hæmorrhage.

_General considerations._ The middle meningeal artery enters the middle fossa of the skull through the foramen spinosum, and divides, after a short intracranial course, into two main terminal divisions, anterior and posterior. The _anterior_ branch passes forwards towards the anterior inferior angle of the parietal bone, then changing direction and turning upwards and backwards towards the vertex of the skull. The _posterior_ branch passes horizontally backwards--grooving the squamous portion of the temporal bone--towards the posterior inferior angle of the parietal bone.

Throughout their intracranial course the main trunk and its terminal branches are embedded in the outer wall of the dura mater, except in the spheno-parietal region, where the anterior branch of the artery usually occupies a channel in the bone--one between ¹⁄₂ and 1 inch in extent.

The anterior branch overlies the pre-Rolandic motor area, whilst the posterior division is related to the temporo-sphenoidal and lower parietal regions--‘silent’ areas of the brain.

Middle meningeal hæmorrhage occurs most commonly in connexion with a fissured or comminuted fracture of the temporal region--the result of direct violence. A fracture, however, is not necessarily present, the hæmorrhage may take place on the side opposite to that at which the injury was received (laceration by contre-coup), and both vessels may be involved. Crisp English narrates three cases in which bilateral extravasation resulted. These more rare examples of middle meningeal hæmorrhage are more or less confined to patients suffering from marked arterial degeneration.

Middle meningeal hæmorrhage, uncomplicated by brain injury, is of infrequent occurrence, the associated injuries to bone and brain confusing the diagnosis. Moreover, the special pressure effects are dependent not only on the absence of serious brain lesion, but also on the non-existence of a safety-valve, such, for instance, as is afforded by a comminuted fracture of the bones entering into the formation of the temporal fossa, or of the roof of the middle and external ears. In the former case, blood will force its way into the temporal region, there forming a temporal hæmatoma--one that may pulsate--whilst, in the latter case, the blood escapes freely from the external auditory meatus. It should be noted, however, that pressure applied to the temporal hæmatoma may lead to the development of irritative or paralytic symptoms confined to the muscles of the opposite side of the body, whilst the restriction of aural bleeding, by means of plugs inserted into the ear, will lead to the early development of compression symptoms.

The amount of blood extravasated varies--according to the calibre of the vessel involved--from a drachm to several ounces. The largest coagulum that has come under my observation was 4¹⁄₂ inches in the long diameter and 1 inch in thickness. Krönlein narrates a case in which the clot weighed 9 ounces.

In shape the clot is elliptical; in consistency it is either fluid or jelly-like. During the early stages of its formation it can be readily removed. Later on, it adheres to the dura mater, and, when removed, leaves that membrane rough and discoloured.

Some uncertainty exists as to the relative frequency with which the trunk and the two terminal branches are exposed to injury. There can be no doubt that the majority of cases in which typical clinical symptoms are present are such as evidence injury to the anterior terminal division; this is due to the anatomical relation of the clot to the motor cortex. Injury to the main trunk is of the rarest occurrence, for the foramen spinosum lies immediately anterior to the petro-sphenoidal suture, the course pursued by typical middle fossa fractures. I have seen one case only, and have read the accounts of two others, in which the foramen spinosum was directly implicated.

The attachment of the dura mater to the sides and base of the skull exercises a most important influence on the direction in which the blood spreads. Firmly adherent to the lesser wing of the sphenoid in front and to the summit of the petrous bone behind, the membrane intervening between these two regions is but loosely attached to, and readily stripped away from, the floor of the middle fossa. As the clot increases in size it exerts considerable mechanical pressure and tends to separate still further the dura from the bone. The anterior and posterior limitations compel the blood to extend first in the outward direction and then upwards towards the vertex of the skull.

Krönlein divides middle meningeal extravasations, according to their regional distribution, into three main groups:--

Temporo-parietal (the most common variety).

Parieto-occipital (rare).

Parieto-frontal (very rare).

Middle meningeal hæmorrhages yielding typical clinical symptoms are of infrequent occurrence, yet such extravasations are very commonly present in severe lesions of the skull. This is evidenced by the fact that middle meningeal hæmorrhage--of a greater or lesser degree--was found in 20 per cent. of all cases of fractured skull, and in 45 per cent. of those in which the middle fossa was involved. My experience is confirmed by Dwight and Nichols. For example, Dwight, in 149 autopsies, notes that middle meningeal hæmorrhage was existent in 49 cases (29 per cent.). He maintains further that the artery is injured in nearly every case in which the fracture, extending from vault to base or vice versa, involves the middle fossa of the skull. Nichols reports 11 cases in 32 autopsies (34 per cent.).

=Symptomatology.= As the direct result of the blow the patient is ‘concussed’, remaining in that condition for a variable period of time, a matter of seconds, minutes, or hours, according to the nature of the associated damage to the bone and brain. In the most typical cases, the patient, on regaining consciousness, should recover--again for a variable period of time--complete control over mind and body, returning to work, walking home or visiting hospital or doctor. More usually, however, he remains slightly dazed, though recognizing his surroundings and capable of answering questions. In any case--unless the hæmorrhage is accompanied by grave cerebral lesion--there should be some attempt at recovery, some return to consciousness. The importance of this ‘lucid interval’ cannot be over-estimated. It should be noted, however, that this interval of consciousness is not of itself absolutely diagnostic of middle meningeal hæmorrhage--it ought to be associated with definite localizing symptoms of brain compression (see below). I have encountered several cases in which this lucid interval was present, and in which operative procedures were carried out in the anticipation of finding an extra-dural extravasation, only to find a subdural hæmorrhage. Such subdural hæmorrhages are not infrequently associated with a similar return to the conscious state, but are rarely--if ever--accompanied by the early development of symptoms of localized brain compression. In the event of doubt as to whether the surgeon has to deal with an extra or subdural extravasation lumbar puncture will probably clear up the diagnosis.

The duration of this ‘lucid’ interval is exceedingly variable. It is usually a matter of minutes or hours, though both König and Wiesman narrate cases in which eight days elapsed before the onset of definite localized compression symptoms. I also have recollections of a case in which the patient walked some miles across the moor to see his doctor, and finding him away from home, walked home again, then becoming unconscious and dying shortly afterwards.

All depends on the calibre of the vessel injured and on the existence of a safety-valve, by means of which some of the extravasated blood can escape externally (see p. 97).

No rule can be laid down as to the duration of this ‘lucid’ interval, but all authorities are agreed as to its existence, even though it may be of exceeding short duration. Its importance, from a clinical point of view, cannot be overestimated. It was present in 43 out of 63 cases reported by Jacobson, well marked in 32 (50 per cent.), less marked in 11 (18 per cent.). It was observed in 60 per cent. of cases that came under my care.

During this period the blood is occupied in stripping away the dura from the bone--the larger the vessel the easier the task and the shorter the ‘lucid’ interval.

The further progress of the case may be considered according to the _general and local effects_ of the hæmorrhage.

=General effects.= As soon as the clot becomes of sufficient size as to exert general pressure on the brain, symptoms of compression become evident, the patient first entering on the stage of ‘slow cerebration’. He is dull and apathetic, showing marked delay in responding to questions. As the pressure increases, he gradually passes into a state of unconsciousness.

The pulse becomes full and slow, the blood-pressure rises progressively (to 200 or more mms. of Hg.), respiration is laboured and stertorous, and the temperature raised (rising progressively). The skin is hot, the face turgid and cyanosed. If the pressure is unrelieved, stimulation of the medullary centres gives place to exhaustion, the pulse becoming irregular and rapid, blood-pressure falls rapidly, and respiration is of irregular rhythm and finally Cheyne-Stokes in character. Finally, the patient becomes completely comatose, the pupils are widely dilated, and the cornea insensitive. Both urine and fæces may be passed (through paralysis of the sphincters), the extremities exhibit flaccid paralysis and the reflexes are abolished. Death occurs primarily through paralysis of the respiratory centre.

=Local effects.= In many cases considerable help to diagnosis may be obtained by external examination--bruising and ecchymosis, or definite hæmatomata, in the temporal or parietal regions. In some cases the hæmatoma may pulsate, and in others the application of pressure to the hæmatoma will lead to the development of fits involving the face and extremities on the contra-lateral side. Copious hæmorrhage from the ear should always arouse suspicion as to the probability of an extra-dural hæmorrhage (see p. 97).

The _anterior_ branch of the artery, as it passes upwards towards the vertex, overlies the pre-Rolandic or motor area. The pressure exercised on that region of the brain leads--according to the site of extravasation and the degree of pressure exercised--to the development of twitchings, convulsions, spasticity, or paralysis of the muscles of the face, upper and lower extremities on the contra-lateral side of the body.

The late Professor von Bergmann[26] pointed out that the arm area is most commonly affected, and that the leg area is never alone implicated. Wiesmann[27] states ‘that isolated paralysis of the leg area is never seen, but only paresis when the arm is paralysed, or both may be similarly affected at once; convulsions may precede the affection’.

In discussing the frequency of local compression effects, Wiesmann points out that some abnormal condition of the opposite extremities was only absent in 16 out of a total of 257 cases. In 37 cases reported by Jacobson,[28] hemiplegia, of a greater or lesser extent, was present in 19 (50 per cent.). The same authority explains that ‘with regard to the onset of convulsions, if there be restlessness, spasmodic twitchings, or movements of the limbs, it is only too probable that in addition to middle meningeal hæmorrhage, contusion or laceration of the brain-substance will be found at more places than one’.

In cases that have come under my observation twitchings of the muscles of the face and upper extremities were observed in 37 per cent. of cases, and paralysis in the same number. Some alteration in the condition of the muscles on the contra-lateral side was present therefore in 74 per cent. of cases.

If the hæmorrhage be situated on the left side of the brain, both motor and sensory cortical speech areas may be involved, especially the former.

With respect to _reflexes_, it has previously been stated that the reflexes are abolished. This is true with respect to the later stages of fully-developed compression. In the earlier state the reflexes on the contra-lateral side may be increased, the difference between the two sides affording some help in the differential diagnosis.

=Pupillary changes= are variable. According to Jonathan Hutchinson, the pupil on the affected side becomes dilated, the blood extending inwards towards the cavernous sinus and exercising direct compression of the third nerve. The frequency with which a dilated pupil occurs in cases of middle meningeal hæmorrhage, and the causation thereof, are, however, matters of dispute. Phelps writes as follows: ‘Reference to cases cited show that the so-called Hutchinson pupil may be observed as well in hæmorrhages occurring in other situations than the middle fossa, and in some cases of cerebral lesion without injury at all. There is no reason to doubt that it is the result of cerebral contusion, but in what relation they stand to specific cerebral injuries is undetermined.’

Walton states that ‘doubtless various factors play an important part in producing the dilated pupil, but if disorders of any simple mechanism are to be credited with the production of the Hutchinson pupil or other pupillary changes, the only lesion worthy of the place is disturbance, irritative or paralytic, of the intracranial fibres of the cilio-spinal tract’.

The late Professor von Bergmann, from 70 cases, reports as follows:--

In 39 cases both pupils were markedly dilated (56 per cent.).

In 7 cases both pupils were markedly contracted (10 per cent.).

In 20 cases the pupil on the same side was dilated (30 per cent.).

In 4 cases the pupil on the opposite side was dilated (6 per cent.).

Parsons, in answering some queries in 1903, writes:--

‘I should anticipate from cortical irritation bilateral pupillary
dilatation, perhaps more marked on the opposite side: from pressure
the opposite effects, so far as mere cortical paralysis goes, but this
would be more liable to be vitiated by vascular changes--the complex
condition seen in coma. It is very difficult to elicit pupillary
constriction by cortical stimuli--very easy to elicit dilatation, but
always bilateral though often more marked on the opposite side.’

In cases examined by me the following conditions were found:--

Pupils equal 46 per cent. of cases
Pupils constricted on the affected side 36 per cent. of cases
Pupils dilated on the affected side 18 per cent. of cases

From my experience, therefore, and from that of the authorities quoted above, it would appear that alterations in the size of the pupil are of very doubtful significance in the diagnosis of middle meningeal hæmorrhage, and that the ‘Hutchinson pupil’ is but rarely observed.

The results obtained by ophthalmoscopic examination are of much greater value. Such investigations should be carried out as a routine procedure. In the earlier stages of the hæmorrhage some œdema of the disk, together with dilatation of retinal veins and diminution in the calibre of the arteries, will be observed on the affected side. This affords a most valuable aid in the differential diagnosis as to the side on which the hæmorrhage is situated--not always such a simple matter as appears at first sight.

=Treatment.= Operation is indicated in all cases of middle meningeal hæmorrhage if the condition of the patient be compatible with such treatment. The earlier the operation is carried out the better.

=Operation.= _For Hæmorrhage from the anterior branch._ After the usual preparatory treatment, the ‘site of election’ for trephining is marked out on the scalp by taking a point which lies 2 inches posterior to the external angular frontal process, and the same distance above the zygoma. A bradawl may be introduced so as to indent the external table of the skull, thus aiding the subsequent accurate application of the trephine.

The incision commences immediately posterior to the external angular frontal process, curves upwards and backwards along the temporal crest, and terminates in front of the tragus of the ear. The flap, comprising skin, temporal fascia and muscle, and pericranium, is turned to its zygomatic base and all bleeding-points secured.

The subsequent procedures vary according to the condition of the bone:--

_If there be no fracture of the bone_, the pin of the trephine is applied to the indented spot and the disk of bone removed. After the removal of the disk the clot will be exposed; but, in order to obtain full exposure, the gap must be enlarged with the craniectomy forceps, mainly in the upward and downward directions.

_If a fissured fracture be found_, the trephine is applied in close relation to the intersection of the line of the artery and the line of the fracture, the gap being enlarged in the required direction.

_When the fracture is comminuted_, the fragments of bone are either elevated or removed until sufficient room has been obtained.

The greater part of the clot can be removed with a Volkmann spoon, or with an ordinary teaspoon, and the region flushed out with hot saline solution.

_To secure the bleeding-points._ When the artery is injured beyond the limits of the gap, the bone must be nibbled away till the bleeding-point is exposed. In some cases it may be necessary to remove a considerable portion of the bone entering into the formation of the temporal fossa. In the meantime, hæmorrhage may be controlled by plugging towards the base of the skull with strips of gauze. Sufficient room having been obtained, the gauze is withdrawn, when smart sponging should enable the operator to determine the source of the hæmorrhage. At this stage of the operation, a head-lamp is useful. After the exposure of the bleeding-points, the following measures can be adopted whereby to arrest the bleeding:--

(_a_) The application of a ligature or of silver ‘clips’ (see Fig. 6) on either side of the bleeding-point.

(_b_) The vessel may be underrun on either side of the bleeding-point.

(_c_) The vessel may be seized with artery forceps and twisted.

(_d_) The hæmorrhage may be arrested by means of gauze plugs.

(_e_) The foramen spinosum may be occluded with bone or wooden pegs.

(_f_) The common or external carotid arteries may be ligatured.

Every effort should be made to control the hæmorrhage by means of ligature, silver clip, underrunning, or torsion. Gauze plugs may usually be relied on to stop the bleeding, but they possess the great disadvantage of exercising pressure on the cortex and preventing expansion of the brain.

Occlusion of the foramen spinosum and ligature of the carotid arteries complicate and prolong the operation. The foramen spinosum is occluded in the following manner: the dura mater is quickly stripped up with the fingers away from the base of the skull, the soft parts gently but firmly retracted and the foramen identified. Its occlusion may be effected by means of a sterilized wooden match or bone peg, by a small piece of bone derived from those osseous fragments which have been removed during the operation, or with the aid of a piece of catgut; in each case the occluding medium should be well driven into the foramen.

It is open to doubt whether ligature of the carotid arteries is ever necessary, though several cases have been recorded in which such a procedure was carried out. The external carotid should be tied in preference to the common vessel, further interference with the blood-supply of the brain increasing necessarily the risk of cerebral œdema, &c.

The hæmorrhage having been arrested by one or other of these methods, the cavity is thoroughly washed out with hot saline solution and drained, a small rubber drainage-tube being brought out through the scalp-flap. The tube should be removed after twenty-four to thirty-six hours, according to the circumstances of the case.

_For hæmorrhage from the posterior branch._ The ‘site of election’ for exposure of the posterior branch lies at the intersection of the two following lines:--

A line drawn backwards from the upper border of the orbit parallel to Reid’s base-line.

A line drawn vertically upwards from the posterior border of the mastoid process.

The trephine is applied, the disk removed, and the gap enlarged with the craniectomy forceps in the forward and backward directions. The hæmatoma is evacuated and the bleeding arrested after the methods enumerated above.

=Results.= The ultimate result of operation for middle meningeal hæmorrhage depends on two factors--the time at which operation is carried out, and the question of brain lesion. There is every reason to believe that, when treated early and when uncomplicated by brain injury, the prognosis, both immediate and remote, is wholly satisfactory.

Wiesman[29] collected 257 cases, of which 110 were submitted to operation, with a mortality of 27 per cent. Of the cases in which no operation was performed 88 per cent. died.

Duchaine[30] reports 27 cases in which operation was carried out, with death in 6 cases only, a mortality of 22 per cent.

Von Bergmann collected 110 cases treated by operation, with 74 recoveries and 36 deaths, a mortality of 32·27 per cent.

Blake only lost 3 cases out of a total of 42 submitted to operation.

Even when every allowance is made for the fact that operative measures were adopted in the more hopeful cases, the above statistics show conclusively that early operation affords great hope of cure.

Before passing on to =subdural hæmorrhage=, it will be convenient to discuss briefly the question of injury to the =internal carotid artery=.

This vessel, during its forward passage in the outer wall of the cavernous sinus, from the posterior to the anterior clinoid processes, crosses the line of the typical middle fossa fracture (see p. 84). The artery is loosely embedded in the sinus, it is protected by the third, fourth, ophthalmic division of the fifth and sixth nerves, and it is situated some distance above the level of the base of the skull. In spite of this degree of protection from injury, the artery may be lacerated in this part of its intracranial passage. The basic fracture is necessarily of a very severe nature. Two cases have come under my own care: one in which both arteries were torn across, the fracture passing from one middle fossa to the other, with such wide separation of the anterior and posterior segments of the skull that the two were freely movable on one another (Fig. 53), whilst in the second case, the fracture, originating in the left frontal region, passed backwards across the anterior fossa, comminuting the base in the region of the cavernous sinus, lacerating that sinus and tearing the artery (Fig. 54).

In both cases death was almost instantaneous, blood pouring from the nose and mouth.

In some few cases the vascular lesion is confined to a minute tear in the coats of the vessel, a fistulous communication being formed between the artery and the cavernous sinus--arterio-venous aneurysm. This condition is discussed on p. 204.

II. Subdural hæmorrhage.

Hæmorrhage into the subdural space may be (_a_) diffuse, or (_b_) localized.

(_A_) =Diffuse subdural hæmorrhage.= The blood may be derived from one of the great sinuses of the brain or from superficial cerebral vessels (laceration of the brain).

The _superior longitudinal sinus_ may be torn by the in-driving of fragments of bone in a comminuted depressed fracture of the vertex, or by wide separation along the line of the sagittal suture. Blood is diffused throughout the subdural space, on one or on both sides of the falx cerebri, but always tending to gravitate towards the lower limits of the supra-tentorial space.

In the event of the wound being compound, air may enter into the sinus. Insomuch, however, as the sinus pressure is, under normal conditions, positive, this complication is of rare occurrence, unless the patient is in a state of profound shock, from loss of blood, or suffers from urgent dyspnœa.

Allusion should also be made to the condition described by Stromeyer, as ‘sinus pericranii’, where a subpericranial hæmatoma communicates with a venous sinus. Such hæmatomata may pulsate, are more or less reducible on pressure, and are increased on straining or on the application of pressure to the internal jugular vein.

The _lateral sinus_ may be laid open in any part of its course, more especially at the angle of junction between the lateral and sigmoid sinuses, in close relation to the occipito-mastoid suture, a region not infrequently involved in basic fractures. The blood effused will occupy the supra- or infra-tentorial spaces according to the situation of the rent in the sinus-wall.

The _cavernous sinus_ is frequently involved in anterior and middle fossa fractures. The blood effused usually escapes into the nose and mouth.

=Symptoms.= The rapid diffusion of blood throughout the subdural space, and the usual coexistence of extensive injury to bone and brain, seldom permit of the development of such definite symptoms as might accurately define the localization of the hæmorrhage. The most important feature in the diagnosis lies in the fact that compression symptoms are early in onset and progressive in nature. The ‘lucid’ interval, so prominent a feature in middle meningeal extravasations, is either absent or of such short duration as to be difficult of recognition. The rapid development of symptoms pointing to brain compression, and the nature of the injury itself should enable the surgeon, in the majority of cases, to come to the conclusion that the patient is suffering from diffuse subdural hæmorrhage. In all cases of doubt, lumbar puncture should be carried out, the presence of blood-corpuscles in the cerebro-spinal fluid confirming the diagnosis.

=Indications for operation.= Whether one considers the case from the point of view of the more immediate prognosis or from the more remote aspect of the case, operation is indicated, if the condition of the patient is compatible with such treatment. In considering the advisability of adopting an active form of treatment, it must be taken into account that, if the patient recovers without operation, the presence of an extensive subdural blood-clot must necessarily interfere, as an after-result, with the functions of the cortex, and that the removal thereof affords considerable hope of permanent cure.

=Operation.= Operative measures may be considered as follows:--

The treatment of the fracture and of the dural rent.

The treatment of the subdural extravasation.

_The treatment of the fracture and of the dural rent._ A scalp-flap is framed, suited to the occasion and formed with due regard to existent scalp laceration. The fracture being fully exposed, depressed or comminuted fragments of bone are elevated or removed. Complete exposure of the sinus-wall, both in front and behind the site of laceration, is required, and for this purpose the craniectomy forceps may be called into requisition. In the event of copious bleeding during these procedures, strips of gauze should be inserted on either side of the sinus rent between the dura and the bone, thus compressing the sinus and stopping the bleeding.

The sinus rent may be treated after one or other of the following methods:--

1. It may be sewn up with mattress sutures of fine catgut, by which means it may be possible to stop the hæmorrhage without interfering with the sinus blood-stream.

2. The open mouth of the sinus may be puckered up by means of a purse-string suture.

3. The needle may be passed right round the sinus on either side of the rent, piercing the falx cerebri in the case of the superior longitudinal sinus and the tentorium cerebelli in wounds of the lateral sinus. The two ligatures are then tightened up and the bleeding controlled.

4. The gauze plugs (see above) may be allowed to remain for twenty-four hours or more, and then carefully withdrawn.

_The treatment of the subdural hæmorrhage._ Though the complete evacuation of the clot is seldom practicable, much may be done to remedy the condition. The tenseness of the dura mater, the absence of pulsation, and the peculiar plum-colour imparted to that membrane by the presence of underlying blood-clot point to the nature of the trouble.

The dura mater is incised and the presenting clot removed, as far as circumstances permit, with the aid of a spoon and irrigation (hot saline solution at a temperature between 110° and 115° Fahrenheit). Free drainage must be supplied, and it is necessary, therefore, that a counter incision should be made at the most dependent part of the clot. A suitable region may present itself if the bone be extensively comminuted. Under other circumstances, a narrow channel may be cut in the bone, prolonged in the downward direction, till the lower limits of the clot have been reached (see Fig. 56).

A rubber drainage-tube may be inserted in the manner depicted in the same figure. Except at the point of emergence of the tube, the dura mater is sewn up (with fine interrupted catgut sutures).

(_B_) =Localized subdural hæmorrhage (subdural hæmatocele).= We are greatly indebted to Bowen[31] for our knowledge of subdural hæmatomata. That subdural hæmorrhage might take the form of a localized collection was a recognized fact, but the condition was but little understood, the typical clinical symptoms were not recognized, and surgeons hesitated to adopt surgical remedies. All these facts impeded advance in both diagnosis and treatment. Subdural hæmatoceles, even at this date, are often regarded as rarities, but, in the light of present knowledge, it would appear that wider recognition of the hæmorrhage in its clinical aspect will show that such ideas are erroneous.

In the preceding section it was shown that subdural extravasation resulting from sinus-injury tends to become diffused throughout the subdural space. With respect to localized hæmorrhages our data are by no means complete, but all available evidence tends to show that subdural hæmatoceles are dependent on laceration of the pia-arachnoid vessels--a condition practically synonymous with superficial laceration and contusion of the cortex. The blood, derived from small cerebral veins and minute cerebral arteries, exercises primarily but slight pressure effect. Its force is expended on compressing and emptying the subjacent and surrounding cerebral vessels, producing, in other words, a condition of _local cerebral anæmia_. This pressure is insufficient to lead to the development of general compression, but suffices to produce certain rather indefinite symptoms. We have, therefore, before us a very different picture to that depicted in cases of middle meningeal hæmorrhage. In such extra-dural hæmorrhages some definite period of time must elapse before the dura mater is sufficiently stripped from the bone to allow of the formation of a clot of size sufficing to exercise both local and general compression effects. Pressure effects then become very manifest.

On the other hand, in localized subdural extravasations there is an immediate effect, for the brain is contused or lacerated. On recovering from the immediate effects of the injury--concussion of a greater or lesser degree of intensity--the patient does not regain complete mental and bodily convalescence. He suffers from symptoms suggestive of brain irritation--headaches, photophobia, mental irritability, insomnia, loss of appetite, the pulse accelerated, and the temperature slightly raised. Later on, more definite symptoms arise, but, intervening between the day of the accident and the time at which these more definite localizing symptoms develop, there is an unmistakable _latent_ period--a period to be sharply differentiated from the previously mentioned _lucid_ interval (see p. 139).

This ‘latent’ period lasts for a variable period of time--seldom less than a week or more than three months. At the end of that time the picture changes, and the patient evidences symptoms obviously referable to local brain compression. From a clinical point of view it is fortunate that subdural hæmatoceles tend to involve the fronto-parietal region with the consequent development of motor symptoms, especially paresis or paralysis of the contra-lateral side. Hence the name sometimes applied to the condition--traumatic late apoplexy. The speech areas are implicated if the injury be situated on the left side. Paralysis limited to the lower extremity is exceedingly suggestive, such a palsy occurring only with the greatest rarity in extra-dural hæmorrhages (from the middle meningeal artery).

If the pressure be unrelieved by operation the patient passes from the excitatory to the paralytic stage of brain compression, gradually falling into a condition of coma, the pulse increasing in frequency and the respiration becoming more and more embarrassed.

In the study of this condition, the following are the points to which special attention should be paid:--

1. The comparative want of severity with respect to the injury received.

2. The absence of any ‘lucid’ interval, such as is present, for example, in middle meningeal hæmorrhage.

3. The presence of a definite ‘latent’ period.

4. The late development of symptoms pointing to general and local brain compression.

Finally, it may be stated in general that the more indefinite and the more deferred the pressure symptoms, the greater the probability that we have to deal with a localized subdural hæmorrhage. Lumbar puncture may materially aid in the differential diagnosis between extra-dural and subdural hæmorrhages.

The following case affords a typical example of subdural hæmatocele:--

An elderly man fell down, striking the right side of the head against
the pavement. He lost his senses for a few minutes and was then
brought to the hospital, dazed and complaining of headache (note the
absence of any _lucid_ interval). He was sent home in a cab and was
assisted to bed by his wife. He kept to his bed during the next week,
complaining of constant headache, and was exceedingly irritable (this
is the ‘latent’ period). Towards the end of the third week his wife
noticed that he seldom used the left upper and lower extremities. This
paresis increased steadily, the face was drawn to the right side, and
his condition became so serious that the wife brought him again to the
hospital (this is the ‘manifest’ period).

When seen by me he was only partly conscious, the mental condition
varying, however, during the day. He was exceedingly irritable,
muttering to himself in a low tone, words being more or less
unintelligible. The left upper extremity was paralysed, the face drawn
to the right side, and the left lower extremity was but little used.
The pupils were equal, the disks normal. The temperature in the left
axilla was two degrees lower than on the opposite side.

The hæmatocele was exposed, cleared out, and the cavity drained.
Recovery was rapid and complete. Two years later all was well.

=Operation.= The lesion is usually of so gross a nature that little difficulty will be experienced in determining the site for trephining. The protective gauze and scalp-tourniquet are applied as usual (see p. 14), and a bradawl introduced through the scalp so as to indent the external table and allow of the subsequent accurate application of the trephine. A scalp-flap, suited to the occasion, is framed, the skull trephined, and the disk of bone elevated and removed. The appearance of the dura mater now allows the operator to verify his diagnosis--the membrane is non-pulsatile, it bulges markedly outwards and presents a blue-purple colour. The bone is then nibbled away in the downward direction towards the lower limit of the clot, the scalp incision being prolonged according to requirements.

In the region of the trephine-hole the dura is incised in a crucial manner, and the four flaps held aside by catgut sutures passed through the apex of each flap. A blunt director is introduced beneath the dura, passing towards the lower limit of the clot, and the membrane slit up to within a short distance of the lower margin of the gap. All meningeal vessels that cross the line proposed for dural section must first be underrun on either side of that line. Retraction of the dura will now allow of adequate exposure of the underlying hæmatoma. Its removal can be carried out with the aid of a spoon, and by means of gentle irrigation (hot saline at a temperature between 110° and 115° Fahrenheit). It is usually impossible to remove the whole of the coagulum, but the greater portion can be got rid of in the manner described.

The dura mater is approximated above by the cross-union of the four apical sutures, whilst the downward prolongation is also sutured except at the most dependent point, where a rubber drainage-tube is inserted beneath the dura mater and brought out through the scalp-flap (see Fig. 56). This tube should be anchored to the dura mater with a single fine catgut suture.

The tube should be allowed to remain in position for at least thirty-six hours. In this the surgeon must be guided by the amount of discharge and the general progress of the case.

=Results.= The most complete results obtained by operative treatment are supplied by Bowen, who classifies a series of 72 cases as follows:--

Class A. Subdural hæmorrhage, apparently without other serious injury to bone or brain.

36 cases. 22 recovered after operation.
14 died--11 without and 3 after operation.

Class B. With serious brain-injury.

36 cases. 6 recovered after operation.
30 died--10 without and 20 after operation.

In other words, every case died in which no operative measures were adopted, whilst after operation 28 recovered and 23 died. My own experience is limited to 4 cases, 3 of which recovered.

=Pia-arachnoid hæmorrhage.= From evidence obtained at post-mortem examinations, it may be concluded that the great majority of cases of severe head-injury--especially those in which ‘irritation’ of the brain is a prominent feature--are associated with pia-arachnoid hæmorrhage. This is evidenced by the presence of a film of blood and serum over certain regions of the brain, not confined to the region subjacent to the skull area primarily involved, but also present over the opposite pole of the brain (injury by contre-coup).

Subpial hæmorrhage differs from the subarachnoid variety in that the blood follows the invaginations of the pia mater, lining, therefore, both major and minor cerebral sulci. Subpial and subarachnoid hæmorrhages are, however, generally co-existent, and the term--pia-arachnoid hæmorrhage--aptly describes the conditions commonly found.

The =symptoms= resulting from such blood extravasations are such as are described in the sections dealing with concussion and irritation of the brain (see p. 166).

INJURY TO THE BRAIN

Injury to the brain must always be regarded as by far the most important of the many complications associated with fracture of the skull. A fracture is not necessarily accompanied by brain-injury, and laceration of the brain may exist without a fracture. There can, however, be no question that the more severe cases of head-injury are almost invariably associated with some degree of brain-injury--varying from contusion to extensive laceration, cortical or central.

No part of the brain is exempt from injury, but two regions, the antero-inferior aspect of the frontal lobe and the antero-external aspect of the temporo-sphenoidal lobe, show a special liability to contusion and laceration.

The brain may be injured immediately subjacent to the site at which the blow is delivered (direct injury), or at the pole directly opposite that at which the blow was applied (indirect injury).

Direct injury results from the in-driving of comminuted and depressed fragments of bone, from the passage of a foreign body, and from the direct transmission and diffusion of forces through the subjacent cerebral substance.

The indirect forms of injury are less readily explained. Their frequency cannot be disputed. I drew attention to the subject in my Hunterian Lectures in 1904, there stating that indirect injury to the brain was more common than the direct form. My opinion is confirmed by Phelps,[32] who states that injury by indirect violence is of almost constant occurrence.

Three theories have been advanced in explanation:--

1. ‘Their production has been ascribed to the change in form that is suffered by a skull in virtue of its elasticity when subjected to violence which causes distortion of the brain to the point of bursting.’

2. ‘In the displacement of the cerebro-spinal fluid by the consecutive depression and bulging that is believed to follow a blow on the head; a momentary vacuum is formed at either axis of force, and the vessels of the brain and meninges rupture from lack of support.’

Neither of these theories are tenable, insomuch as they are based on an acceptance of the ‘bursting and compression’ theories (see p. 74)--theories which, in my opinion, it is impossible to accept.

3. The ‘contre-coup’ theory. According to this theory the injury is due to sudden and violent displacement of the brain against the opposing osseous barrier. This theory has been opposed by many--notably by Helferich--on the ground that the brain is said to completely fill the cranial cavity, and that ‘shaking’ from side to side is impossible.

This theory affords, however, the most satisfactory explanation of contralateral laceration. The following points may be advanced in its favour:--

(_a_) The frontal and temporo-sphenoidal lobes are more liable to contusion and laceration than any other parts of the brain, both regions possessing certain anatomical relations accounting for their ready injury. The frontal pole is related to the angle of junction between the horizontal and vertical plates of the frontal bone, the temporo-sphenoidal lobe to the cul-de-sac, at the anterior part of the middle fossa, that is overhung by the wings of the sphenoid bone.

(_b_) The lacerated area is almost invariably situated at that part of the brain which lies opposite to the region struck in the line of the transmitted force. Furthermore, brain lesions are most prevalent when the accident results from a fall from a height, a class of accident in which ‘shaking’ of the brain is most likely to occur, and least common when the basal fracture is dependent on a bilateral compression force.

(_c_) The statements that ‘shaking’ cannot take place, and that the soft cerebral substance cannot be driven forcibly and violently against the opposing bony barrier, so as to be locally bruised and torn, are incorrect.

That such violent displacement of the brain can and does occur is proved by the following case:--

A prize-fighter was knocked out by a blow on the right side of the
head, dying a few hours later. The post-mortem revealed no fracture of
vault or base, but a linear laceration of the brain on the mesial
aspect of the right hemisphere which exactly corresponded to the free
margin of the falx cerebri. In this case, therefore, conclusive
evidence was obtained that the brain had been driven from right to
left against a firm and resisting barrier, corresponding laceration
resulting.

The lacerated area is usually cone-shaped, the base corresponding to the surface of the brain, the apex facing towards the lateral ventricles. In the most severe cases the anterior or lateral horns of the lateral ventricles may be involved at the apex of the cone, and, in the event of further cerebral softening, the horns of the ventricles may, as it were, bulge outwards, thus assisting in the formation of those traumatic cephaloceles that communicate between the lateral ventricle and the surface of the brain.

On the other hand, if the degree of laceration be slight and the injury compatible with life, organization and partial resolution takes place, with the formation, in nearly all cases, of adhesions between the cortex and the membranes, of false membranes, arachnoid cysts, and of blood cysts. In the more serious cases cerebral fibrosis results, in direct proportion to the degree of cerebral destruction, with corresponding degeneration, both of the cerebral cortex and of the tracts that evolve therefrom.

=Symptoms.= Certain regions of the brain may, when injured, give rise to localizing symptoms. More generally, however, all localizing features are obscured by an overpowering element--the presence of concussion, irritation, or compression of the brain.

Still, in some cases, the localizing features are sufficiently marked, at some period of the case, to allow of a more definite diagnosis. The symptoms, therefore, must be considered under the two heads, (A) _localizing_, and (B) _general_.

A. Localizing symptoms.

_Localizing to the frontal lobes._ Phelps,[33] from a series of 130 cases, arrived at the following conclusions--that, ‘in nearly every case in which consciousness was retained or regained, and the mental faculties not perverted by general delirium, laceration of the left frontal lobe was attended by default in intellectual control, and that the lesion was usually of the anterior region and implicated its inferior surface.’ ‘Subcortical disintegration was characterized by abrogation of mental power, and superficial laceration by aberration in its manifestations.’ And secondly, that, ‘in nearly every instance in which laceration was confined to the right side, the mental faculties remained unaffected, except as they were obscured by delirium and stupor occasioned by coincident general lesion.’

Although the above statements have not altogether been confirmed by cases that have come under my observation, yet, coming from such a source, they must be accepted unless refuted.

Bianchi states that, after extirpation of the frontal lobe of the ape, the whole intellectual life of the animal is changed. Perception, attention, inhibition, and in particular memory and association are enfeebled. The psychical tone is lowered; the ape lacks initiative, courage, and spirit.

Laceration of the under aspect of the lobe may, by direct involvement of the olfactory region, interfere with the faculties of smell and taste.

_Localizing to those parts that minister to the function of speech._ The condition of the patient usually prevents one from obtaining early information as to the power of speech. Isolated lesions of the posterior part of the third left frontal convolution result in _motor aphasia_, a condition characterized by the inability on the part of the patient to speak, in spite of the fact that comprehension is unimpaired. The brain-injury is, however, seldom so defined as to be characterized by loss of speech without other phenomena. In very close relation to the area responsible for speech is the _writing_ centre, situated at the posterior end of the middle frontal gyrus, near the centre for the movements of the hands and fingers. Injury to this centre results in _agraphia_ and _alexia_.

_Word-deafness_ results from lesions implicating the upper temporo-sphenoidal lobe, and _word-blindness_ from injuries to the supramarginal and angular gyri (see Fig. 57).

Phelps pointed out that the lesion must be an extensive one if the symptoms included under the term _sensory aphasia_ (word-blindness, word-deafness) are to be evolved. No localized injury to any small area suffices, and the cortex must be deeply involved. Lacerations of the brain are usually of so gross a nature that further differentiation is seldom to be obtained. Here it might be noted that the cortical centres responsible for speech and its accessories, word-seeing, word-writing, and possibly word-hearing, are all situated on the left side in normal right-handed individuals. The corresponding areas on the right side may be regarded as ‘silent’ areas, and may be utilized, if desired, for decompression purposes. No operation, with that purpose in view, should be carried out, however, unless the surgeon is assured that the patient is right-handed in his actions. A case has recently come to my knowledge where a right-parietal decompression operation was carried out with disastrous results.

_Localizing to the temporo-sphenoidal lobe._ The relation of sensory aphasia to lesions of the temporo-sphenoidal lobe has already been mentioned. The anterior poles of this portion of the brain, a region frequently involved in head injuries, may lead to the development of impaired smell and taste, especially if the lesion includes the uncinate lobe and be situated on the left side. In other respects, the temporo-sphenoidal lobe may be regarded as a ‘silent’ area of the brain.

_Localizing to the pre- and post-central convolutions._ Injury to the pre-central or motor area usually leads to the development of definite symptoms--twitchings, convulsions, or paralysis of the face and extremities on the contra-lateral side. In the earlier stages reflexes are abolished. Later on, as the result of degenerative changes in the pyramidal tracts, spasticity, contractures, and rigidity, with increased reflexes will be observed in the affected limbs. On the other hand, the muscles do not show any reaction of degeneration. Babinski’s sign is generally present.

Injury to the post-central convolutions might be expected to lead to various alterations in tactile and muscle sense, in stereognosis, and in sense of pain and temperature, but the general condition of the patient seldom permits accurate demonstration. Such sensory disturbances are more frequently observed as late results of head-injury.

_Localizing of the occipital lobes._ Laceration of the occipital lobes may lead to homonymous hemianopia, for which defect it is probable that the degree of laceration must be considerable, involving mainly the mesial aspect of the occipital lobe (see Fig. 69). Schäfer has shown that the greater portion of the occipital lobe may be removed in monkeys without producing loss of vision, and it is only when the lesion involves the parieto-occipital fissure and passes into the occipito-temporal convolutions that loss of vision is permanent.

Slighter degrees of occipital injury may lead to subjective symptoms, such as flashes of light, colour changes, &c.

_Localizing to the cerebellum._ Lesions of the cerebellum rarely permit of the development of such localizing symptoms as are observed in cerebellar tumours. Inco-ordination of movement, ataxia, and other symptoms observed in cerebellar tumour formation are, from the general condition of the patient, incapable of demonstration in cerebellar laceration. It is necessary, however, to draw attention to the significance of yawning and gaping. This symptom has been observed in several recent cases, and, as far as my observation goes, is only present in cerebellar lesions.

In the consideration of these localizing symptoms, it must be borne in mind that brain lacerations are of an exceedingly gross nature, and that there may be--as the late Professor von Bergmann[34] pointed out--‘notwithstanding the severest brain symptoms, not the slightest discoverable anatomical alteration in the brain. On the other hand, notwithstanding the absence of all brain symptoms, extensive and striking destruction of the brain substance.’

B. General symptoms.

Concussion or cerebral shock.

_Its pathology._ Concussion used to be regarded as implying a ‘molecular disturbance’ of the brain--a definition so unsatisfactory that it may be relegated to a desired obscurity. At the present day, though some difference of opinion exists as to whether the condition is or is not necessarily dependent in its development on associated cortical or subcortical lesions, macroscopical or microscopical, concussion is regarded as implying a condition of _acute cerebral anæmia_ through sudden inhibition or paralysis of the vaso-motor centre.

With respect to the existence of visible lesions of the brain, Kocher maintains that such lesions--hæmorrhages and the like--are necessarily present, proposing that the term ‘brain concussion’ should be eliminated in favour of ‘brain contusion’. On the other hand, the late Professor von Bergmann pointed out that such a course was not possible, instancing the many cases of concussion where contusion could not possibly have existed.

There cannot, in my opinion, be any question as to which is the correct view. It is unreasonable to ask one to believe that the milder cases of concussion, cases in which the patient merely suffers from loss of consciousness for a few minutes, from which he rapidly recovers and is discharged from treatment within a few days, and from which he suffers in the future from no remote ill effects whatsoever, it is unreasonable to ask one to believe that such cases have experienced so severe a lesion as brain contusion or laceration. Both these conditions would require a long period of convalescence, and would tend to leave in their train very serious after-results.

It is obvious, therefore, that there exists a degree of concussion from which the patient can quickly and entirely recover. It is also clear that in many cases the patient passes through a long and serious illness, whilst, in some cases, a fatal termination may ensue--and that too within a few hours. It is consequently clear that we have to deal with varying grades of concussion, some of which are of so mild a nature as to be clearly unassociated with definite cerebral lesion, whilst others are of so serious a nature that death may take place within a few hours, and in such cases the gravest cerebral lesions may be found at the autopsy.

In other words, it would appear that concussion is more or less dependent in its severity on the degree of associated cerebral change.

It is, however, to the milder type of case that we should turn in order to determine what is exactly meant by concussion. In such cases the patient merely suffers from a mild degree of _cerebral shock_, a condition closely resembling shock in general. ‘Shock’ is defined as a ‘condition resulting from fall in blood-pressure, due to inhibition or paralysis of the vaso-motor centre’. The afferent impulses that pass up the various sensory paths inhibit or paralyse the vaso-motor centre, a centre that has for its main function a tonic constricting influence on the peripheral arteries. Hence, these vessels dilate atonically, the blood collects in the great splanchnic area, and, insomuch as the cerebral arteries possess no separate vaso-motor apparatus, they passively follow the changes in the general circulation. The brain is consequently in a state of arterial anæmia, the arteries relatively emptied, the veins passively engorged.

Clinically, concussion or cerebral shock differs from ordinary shock in one manifestation only--sudden loss of consciousness. This development is readily explained on the ground that concussion results from a violent blow applied in the near neighbourhood of the vaso-motor centre, with consequent _sudden_ vaso-motor depression and _acute_ anæmia of the cerebral cortex. This leads to immediate great fall of blood-pressure, loss of consciousness, and lack of muscular control.