THE PATHOLOGY OF TRIGEMINAL NEURALGIA, ILLUSTRATED BY THE MICROSCOPIC INATION OF TWO GASSERIAN GANGLIA.
Theremoval of the Gasserian ganglion for persistent trigeminal neuralgia made the study of the pathology of this disease possible. The operation was based upon the assump tion, either that the ganglion was the focus of the disease, or that by its removal the central sensory paths for pain would be permanently interrupted. Grounds for this assumption were found both clinically and pathologically.
Previous to this time there were only two sources from which such a study could be approached. First, from the ex amination of portions of the peripheral branches of the fifth nerve, which were removed by the so-called palliative opera tions; and, secondly, from hypotheses and conjectures based upon the pathology of neuralgias elsewhere. The first of these offers only a limited field for study, and the sources of error are considerable, so much so, in fact, that the recorded observations are now no longer to be relied upon. The futility of this method of study is in the main due to two things, first, pathological changes found in the resected nerves might well be the effects of the mechanical force of the operation. Sec ond, the absence of positive findings merely pointed to the fact that the diseased condition was probably farther up the course of the nerve, or in the ganglion itself. A pathology based merely upon hypothesis or analogy is, of course, at this day not worthy of serious consideration.
If we believe, then, in the purely objective method of study, it is evident that most of the early work on the pathol ogy of trifacial neuralgia can have merely an historic interest. The theories of Dana in regard to the role of the blood-vessels ; of Oppenheim, that of an inflammatory degenerative process in the vessels ; of Mills, "a form of degeneration of the cells of the ganglion, in which the disease gradually invades all portions of the body, and, as one cell after another drops out, the instability of the ganglion is increased, and with this the tendency to violent sensory discharges ;" ° and of others, fail to satisfy the demand which modern neuropathology insists upon, and must be set aside as inadequate.
The first serious attempt, then, to investigate the pathol ogy of trigeminal neuralgia began about the time of the ap pearance of Krause's monograph, " Die Neuralgie des Trigemi nus," in 1896. In this work there is an account of the microscopic study of four Gasserian ganglia removed by him, and studied by Sanger.
Although in all some 200 operations for the removal of the ganglia have been done since then, a careful examination has been made in only a comparatively small proportion of them. In fact, at the time of the symposium on affections of the fifth nerve held in Philadelphia at the College of Physi cians, April 20, 1900, Spiller reported that no authentic ex amination by Nissl method of the ganglion had as yet been recorded, and until such examinations have been made in con siderable numbers, no very definite statement of the pathology is possible.
It was for this reason that the study of these two ganglia was undertaken, and, as it was possible to use the Nissl stain in both of them, it was hoped that the findings might be of some interest.
Before going into the detailed study of the ganglia and their anatomy, it might be well to suggest certain rules, in the absence of which pathological findings in the ganglia are open to criticism. First, a Gasserian ganglion upon the peripheral branches of which surgical operations have previously been made in the way of nerve stretching, resection, etc., is unfit for pathological study, or, rather, any conclusions drawn from the findings must be for the most part invalidated, for the reason that the mechanical effects of the operation may cause an ascending neuritis, which might produce changes in the ganglion itself, in the cells, or in the periganglionic tissue. Second, a Gasserian ganglion which is removed by morcella tion, or is much torn or cut, cannot be regarded as a favorable object for pathological study. Third, no conclusions in regard to the condition of the nerve-cells are justified, unless they have been studied by the Nissl method or its various modifi cations. Of course, the ganglion must be examined in a good state of preservation.
The first of the ganglia given to me by Dr. Bartlett ful filled most of these conditions, except that a peripheral opera tion had been done some ten years before; the second one may be regarded as an ideal specimen for study.
A few words in regard to the microscopic anatomy of the ganglion Gasserii and the nerves arising from it may make clearer the description of the pathological appearances which follow.
The ganglion in its structure and function is analogous to the sensory ganglion of the posterior nerve-roots of the cord. It is composed of numerous nerve-cells arranged in rows and layers. These cells are sensory in type and structure, and are unipolar. The cells are round, having a nucleus and a nucleolus, which occupy the centre of the cell with the Nissl bodies arranged concentrically around them. The whole gan glion is surrounded and enclosed by connective tissue. This connective tissue runs in between and around each cell, form ing a supporting membrane to the cell and for the blood vessels. A capillary network finely meshed surrounds the nerve-cell and the nerve-fibres arising from The nerve-cells are unipolar, that is, they possess but one process. This is the axis cylinder, which shortly after leaving the cell divides into two branches, one going to the periphery and the other to the brain. The myelinized fibres of the periph eral branches form the three main branches of the nerve, and those of the central branches form the great sensory root, which goes to the pons and plunges into it. As soon as it reaches the pons the individual branches again divide into an ascending and descending portion, analogous to the posterior spinal nerves. The ascending root of finer calibre than the other loses itself quickly in the cells of the trigeniinal nucleus, or in the substantia gelatinosa, while the descending root forms the well-known spinal trigeminal root, which has been followed as far as the crossing in the It can be seen from this brief description that the tri geminus has an immense distribution, both in the medulla and in the cord ; and the degeneration involving this immense area, which must necessarily follow in every removal of the ganglion, should be taken into account when the question of operation is considered.
I wish to call your attention to the central sensory root, as this has been found diseased in a few cases, and has given rise to the question of central neuralgia.
The two ganglia studied in this paper were given to me in P;) per cent. formol solution. They were, as can be seen from the photographs, cleanly removed, showing practically no mutilation, and in every way in excellent condition for microscopic study. After being placed in alcohol for a few days they were embedded in celloidin, and cut in the follow ing fashion. No. i at right angles to the long axis, and No. 2 parallel to it and transversely. I wish to say that the second method is far better, as•each section gives a view of the whole ganglion body, and if portions of the nerve-roots are sepa rately mounted, cross sections can easily be made.
Complete serial sections of both ganglia were made, and numerous sections from each level were examined. The fol lowing stains were used : Nissl-methylene blue method, thio nin, toluidin blue, neutral red, for the nerve-cells; Van Giesen, Haemalaun, Carmin and Weigert's hwmatoxylin with a pre vious treatment in chromic acid for the nerve-cells and inter stitial structure. Various other methods were tried, but on account of the formalin hardening were mainly unsuccess ful.
Before recounting the findings in these ganglia, a short analysis of the literature on the subject may be of interest. I have been able to collect the reports of twenty ganglia exam ined after their removal for trigeminal neuralgia. They have, with the exception of Krause's cases, been reported in the last three years, and chiefly by American authors. Of these twenty reports comparatively few will stand the test of strict neuro pathologic criticism. By far the most notable contribution to the literature on this subject is by Spiller, of Philadelphia, who has studied ten cases. The best discussion of trigeminal neuralgia from all points of view is found in the series of papers read before the College of Physicians at Philadelphia at the symposium upon the fifth nerve in its neurological and surgical relations, April 20, 1901. Every worker in this field of medicine must acknowledge his indebtedness to the work done there. To take up in any great detail these twenty reports would require more space than is at my disposal, so I shall merely attempt a brief account of the findings as reported.
In Spiller's first series of seven cases, six showed patho logical changes. It is to be noted that no Nissl stain was used, as the specimens were hardened in Mailer's fluid. The find ings were in the main as follows : Medullary sheaths much swollen, likewise the axis cylinders, fibres atrophied, empty nerve-sheaths, atrophied ganglion cells, cells faintly stained, and in one case sclerosed blood-vessels. In one case a decided increase of connective tissue, a distinct In two other cases reported by Spiller degeneration by Marchi method in the medullary sheaths, concentric bodies of brain sand within the ganglion, tumefaction of the axis cylinders in portions of the sensory root ; in the second case the ganglion cells were Although the cells were found affected, there is not evidence enough to show that they are the seat of a pri mary morbid process. In Spiller's series is included also the case of endothelioma. involving the ganglion operated on by Keen." In four cases operated upon by Krause and studied by Sanger, the ganglion cells were found diseased in the way of pigmentation, vacuolization, and atrophic appearances. No change was found in the blood-vessel, and no interstitial changes. Krause believes that the changes are first paren chymatous in nature, and then the myelinized nerves are involved." 1 Leser found widened lymph spaces with an in crease of wandering cells. Swollen appearance of ganglion cells similar to those described by Krause." L. F. Barker reports two cases both examined by the Nissl method, scarcely any normal-looking cells, chromophilic cells, yellow pigment, cells swollen, in some, absence of tigroid substance, or Nissl bodies. Some of the nuclei swollen, no typi cal Marchi degeneration, pigmentary changes in cells, no vas cular changes, no interstitial hypertrophy. The same concen tric bodies as Spiller describes." Two cases reported by T. Crawford Renton showed absolutely no abnormality." Coehlo's case examined by Pestana by the Nissl stain,—cell changes of every degree of severity from the slightest modification of the chromatic sub stance to the complete destruction of the cell and the substitu tion for it of connective tissue. Infiltration of black pigment in the cells. In places, following the total destruction of the cell, its capsule became filled with round cells of new forma tion. 7 It is difficult to summarize these findings on account of their variety and contradiction. It is much more difficult to determine just what weight they should have in any final ex position of the pathology of the disease under consideration. Without reference to any one or series of cases, the patho logical findings as a whole might be divided into groups, as follows. In a few cases the ganglion was found to be abso lutely normal. Of this group not much can be said except that a certain doubt must always be felt in any report of an absolutely normal condition of a ganglion so pathological in its manifestations as the ganglion in a case of persistent tri facial neuralgia of long duration, for it is assumed that only in such cases was an operation performed. A second group shows evidence of neuritis and degeneration, so pronounced that there can be no question that this must be considered in any explanation of the pathology of this disease which may be attempted. The third group, composed of Spiller's one case and the second one of mine, shows an interstitial inflam mation. A fourth group, composing the largest number of reports, shows nerve-cell changes of various degrees of sever ity. In all of them, however, the changes are less those of a primary cell affection than a secondary one. It must be ad mitted that these cell changes of themselves would form an insufficient basis for the pathology of the disease. The pres ence of brain sand, or concretions, was noted in three cases, and in the second of my series. In all probability these cells have no vital importance. In a very few cases changes in the blood-vessels were noted. Some consideration of these find ings will be taken up in the discussion of the changes found in the two ganglia which are the subject of this paper. The pathological details of the first ganglion being less definite and less reliable, for the reasons previously given, than the second, only a brief description of them will be given.
The examination of each ganglion may be divided into three parts; first, the study of the nerve-cell ; second, of the interstitial tissue, including the blood-vessels ; third, of the nerve-fibres, including what was taken to be the sensory root in the second case.
Ganglion No. x. In the Nissl preparations the nerve-cells took the stain most unequally, some being stained so densely (chromophilic cells) that no study of their finer anatomy could be made, while others, again, were so slightly stained that they gave the appearance of cell atrophy. The cells were extremely irregu lar in form, comparatively few showing the full rounded appear ance of a normal sensory cell. In many places empty capsular spaces were seen, which formerly contained nerve-cells. Upon closer examination some of these spaces showed what was evi dently the remains of degenerated nerve-cells, collections of pig ment, scattered tigroid substance, and small, round bodies very much like an extruded nucleolus. The cells vary much in regard to size. As a general rule, the very small cell took the stain very intensely and showed the greatest abnormality. It is interesting to note in this connection that Head and Campbell, in their recent study of the spinal ganglia in herpes zoster, found that the smaller cells were first and most profoundly affected, and they believe that these cells probably subserve the function of pain." The analogy between the spinal and Gasserian ganglia has been pre viously alluded to, and in this respect this finding is an interesting corroboration. Some of the cells were much retracted and shrunken, leaving about them large clear spaces generally devoid of any structure at all. The cells which took the Nissl stain in a typical manner showed for the most part no abnormality, the nucleus and nucleolus being in the centre and the Nissl bodies concentrically arranged. In quite a large proportion of cells, evidence of chromatolysis, slight, as a rule, and of the central variety, was found. Pigmentation of the cells was common, the deposited pigment being found, as is usually the in the periphery of the cell. Some few cells showed the typical appear ance of a nerve-cell pathologically affected in an extreme degree, with migration of the nucleus towards the periphery, vacuoliza tion, etc. They were too few to be of great significance. The interstitial substance was to all appearance normal, no sclerosis being observed. In fact, many cases gave the impression of an interstitial atrophy rather than the reverse ; the blood-vessels were normal. The nerve-fibres stained by the Van Giesen method were mostly normal in appearance.
Any interpretation of the findings in Ganglia No. I must take into account the early peripheral operation and the haemorrhage which accompanied its removal, free blood being found in great abundance scattered throughout the ganglion.
Ganglion No. 2. This was a much more favorable specimen for study than the first, on account of the absence of a peripheral operation, the absence of profuse hxmorrhage, and on account of the method of sectioning. Very beautiful preparations were ob tained by toluidin blue, with a counter-staining of erythrosin and by Nissl. The nerve-cells in the thionin, Nissl, and toluidin blue stains show the same irregularity in stain reaction as in the first case. This difference seemed to follow no rule, a deeply stained cell being found next to one very faintly stained. In no sense could there be any grouping of cells based upon their reaction to the stains in point of color. In the main, while the cells could not be said to be absolutely normal, the irregular and atrophied ap pearances in Ganglion No. I were not observed. The empty spaces found in the first case were absent, they being, as a rule, well filled out, allowing for the retraction due to the hardening fluids. Practically, no nuclear changes were found, either in regard to position, form, or staining quality ; some cells showed the usual central or peripheral chromotolysis. The pigment deposit in the cells is very curious, and merits a brief description. Instead of being in the periphery, as in the first it was here found cen trally, chiefly in the neighborhood of the nucleus. The pigment is composed of fine granules, differing somewhat in size, showing a greenish tint in the Nissl preparation and black in the other stains. The pigment was frequently found arranged around the nucleus in the form of a crescent or ring. The form and arrange ment of the pigment recalled the pigment deposit in the red blood corpuscles in malaria. The concentric bodies or brain sand described by Spiller and Barker were found in a few sections towards the centre of the ganglion.
The study of the interstitial substance in this ganglion showed some very interesting facts, and gives a fairly definite explanation of the pathology in the second case. The interstitial substance was greatly in excess of the normal, as shown by the tremendous increase of the nuclei. This nuclear increase was due to a pro liferation of two different kinds of cell elements, one having a slightly elongated nucleus characteristic of a connective-tissue cell, the other smaller and circular. This latter was really a cell pro liferation, showing in many places the typical appearance of a small, round cell infiltration. In places this cell increase extended into the periganglionic spaces, which were quite filled with small, round cells. Bands of connective tissue anomalous in distribution were to be seen. In the Van Giesen and strictly nuclear stains this increase in interstitial cell elements could be very clearly seen.
The nerve-fibres and nerve-roots as well as what was taken to be the sensory root, and the nerve branches, which were cut separately in cross sections studied chiefly in Van Giesen prepara tions, showed, beyond a tendency to stain rather faintly and irregularly, no certain evidence of degeneration or atrophy. As it was impossible to use the Marchi stain in formol hardened prepa rations, the question of degeneration must be regarded as still unsettled. In a few sections treated first with chromic acid and then stained with Weigert's hmmatoxylin method, appearances in the nerve-fibre and sheath similar to those described by previous investigators were seen, but as these specimens were faulty no reliance can be placed upon them. The blood-vessels were found to be normal.
The findings of these two ganglia might be summed up as follows : In both of them the nerve-cells must be regarded as pathologically altered, but in neither to such a degree or extent as to consider them primarily affected. The definite changes which are found in acute or long-standing paren chymatous nerve-cell affections are certainly missed in these . two cases. The more intense cell changes of the first case lose their importance, in a measure at least, on account of the previous peripheral operation on this ganglion. Pathological the cells certainly are, but secondarily affected.
Neuritis and atrophy are probably present in both cases in spite of the failure of the Van Giesen stain to definitely prove their existence. The concentric bodies found in the second case are similar to those described by Spiller and Bar ker, who designate them by this name, or brain sand, evidently not considering them amyloid bodies at all. Redlich has de scribed such bodies as corpora amylacea, or amyloid bodies, and says that they stain with the nuclear stains, and only the larger ones show a concentric structure around a central nu cleus. They are usually found in the spinal cord around the points of exit of the posterior They occupy analogous positions in this specimen. These bodies, then, in spite of Spiller's and Barker's hesitancy to call them so, are probably corpora amylacea. Their pathological significance is very slight, as their number is so small.
In attempting to resolve the pathological findings of these two ganglia into some general basis for a pathology of tri geminal neuralgia, the same difficulty is present as in every disease where symptoms are the expression of various causes. From these two ganglia, as well as from the reports of the 45 ganglia before alluded to, it is quite evident that trigeminal neuralgia is not a definite disease, but merely the symptoms of various processes affecting the fifth nerve anywhere in its course from the ganglion to its peripheral termination. It is extremely probable that no disease of the nerve-cells per se exists as a primary parenchymatous affection. In the present state of our knowledge we are justified in assuming two main divisions of trigeminal neuralgic affections. First, and the more common, is a neuritis beginning in the terminal divisions of the fifth nerve and having a tendency to ascend to the ganglion. Second, an interstitial inflammation, chronic and progressive, of the ganglion body itself. Of the two sped mens studied, I should venture to place No. r in the first, and No. 2 in the second category. A third division is possible, and there have been two cases reported, that is, a central neu ralgia or neuritis affecting the sensory root as it leaves the ganglion on its way to .the pons.
Concerning the etiology of these divisions, our knowledge is too vague to do more than speculate. That trauma and new growths may have some importance has been amply substan tiated by the case of Spiller, where an endothelioma was found affecting the ganglia, and by the report of cases where a direct trauma in the region of the ganglion was followed by the neu ralgic manifestations, possibly in Case No. 2. The two other causes which play an important role are in all probability toxic and mechanical. Just how these act is at present doubtful. In the second the interstitial hypertrophy might very well exercise mechanical pressure on the ganglion cells, pro ducing in this way the attacks of pain similar to the root-pain of spinal affections. The nerve-cells show the effect of this pressure by their various states of functional fatigue, so to speak, which is shown by their varying reaction to the Nissl stain. However, this is pure hypothesis, which was denounced in the beginning of this paper as being unscientific.
A few words from a neuropathologic point of view in regard to the utility of the operation may not be amiss. It is evident that this operation has a definite basis from the patho logical stand-point. Wherever the process is located, the re moval of the Gasserian ganglion must be the final means of relief. It is simply a question in what cases operation is justi fiable. This is a matter beyond the scope of the present paper. In regard to the so-called central neuralgias, I fail to see what influence this can have upon the utility of the operation. The sensory root, if diseased, can only degenerate to the terminal ends of the neurons involved, and following the law of sec ondary degeneration the process must stop. Pathological processes in the brain itself other than those due to pressure, and those affecting the meninges, cause no symptoms of pain, and, as the Gasserian ganglia contain the cells of nutrition of the sensory root, their removal is equivalent to placing this portion of the nerve outside of the realm of active symptoma tology.
There remain, then, two general lines along which the study of neuralgia of the trigeminus should proceed. First, such an improvement of the operative technique as to render the operation comparatively safe, as suggested by Keen; and, second, such an improvement in our clinical knowledge that it will be possible to tell what portion of the trigeminus is affected. If it is the ganglion, then operation for its removal will be indicated. If in the peripheral branches, then periph eral operation is advisable. Such a knowledge can only be gained by a careful study of each ganglion removed in con nection with the symptoms present in each case.