Case 102 is an instance of such a lesion, the post-mortem examination showing clearly that the spinal canal was not encroached upon by the bullet. The cord in this instance appeared little changed macroscopically, and this fact was observed in other instances, both during operations and post mortem.
Contusion.—This condition is very closely allied to the last. In cases 101 and 103 the spinal canal was as little encroached upon as in 102, but the bullet struck the somewhat elastic neural arch in each case, and post mortem an adhesion between the cord and the enveloping dura opposite the point at which impact of the bullet was closest suggests that, in spite of the escape of the bone from fracture, it may have been momentarily depressed to a sufficient degree to contuse the cord, or the latter may have suffered a contre-coup injury. For these reasons the inclusion of the cases as instances of pure concussion is not warranted. In both Nos. 99 and 100 the neural arch had actually suffered fracture, and although the bone was not depressed or exercising pressure at the time of the autopsies, it was no doubt driven in temporarily at the moment of impact of the bullet.
At the post-mortem examinations of injuries of this nature it was common to find one to four segments of the spinal cord completely disorganised. At the end of some five weeks, the common duration of life, the structure of the cord was represented by a semi-diffluent yellowish material, the consistence of which was so deficient in firmness as to allow the partial collapse of the membranes covering the affected portion, so as to exhibit a definite narrowing when the whole was held up (see fig. 79). In such cases traces of extra- or intra-dural hæmorrhage sometimes still persisted.
Hæmorrhage.—This occurred as surface extravasation and in the form of parenchymatous hæmorrhages. I saw the former both in the extra-dural and peri-pial forms, but never in sufficient quantity to exert a degree of pressure calculated to produce symptoms of total transverse lesion. Here again, however, it is difficult to speak with confidence since the conditions which regulate the tension within the normal spinal canal are so complicated and liable to variation, that it is very difficult to estimate the effect of any given hæmorrhage discovered.
My friend Mr. R. H. Mills-Roberts described to me one fatal case under his care in the Welsh Hospital in which extra-dural hæmorrhage was so abundant as, in his opinion, to have taken a prominent part in the production of the paralytic symptoms.
Examples of both extra- and intra-dural (peri-pial) hæmorrhage are afforded by cases 99, 102, and 103; in none was it large in amount or widely distributed. The condition was probably also frequently associated in varying degree with that to be immediately described below.
Intra-medullary hæmorrhage (hæmato-myelia).—The importance of this condition is lessened in small-calibre bullet injuries by the fact already alluded to, that it is almost invariably accompanied by concussion changes. In one instance in which death took place at the end of eight days, partly as the result of concurrent injury, in a man in whom signs of total transverse lesion of the cord were present, the substance of the cord was found to be closely scattered over with hæmorrhages of various sizes and extending for a longitudinal area of some three inches.
As to the frequency with which hæmorrhage into the substance of the cord occurred, I regret to be unable to give an opinion. In the late post-mortem examinations I witnessed, a yellow discoloration of the softened cord was the only macroscopic evidence of hæmorrhage.
Hæmorrhages of this nature may, however, account for the grave paralytic symptoms in some cases of partial or total transverse lesion not due to direct compression or laceration.
The conditions of concussion, contusion, or hæmatomyelia were, I believe, responsible for at least nine-tenths of the cases in which a total transverse lesion was indicated by the symptoms. The extreme importance of realising this fact and the rarity of the production of symptoms by continuing compression both from the prognostic and the therapeutic point of view is obvious.