Chapter LX: Part 60

{868} Murchison[15] classified the first case of this disease involving the intestines, liver, mesentery, and heart. The same observer the following year described a case in which the glands of the neck, mediastinum, axillæ, and spleen were involved. Wunderlich has recorded a case of malignant mediastinal disease which commenced in the glands of the neck; but the cervical glands may be enormously enlarged without implication of the bronchial.

[Footnote 15: _Path. Soc. Trans._, London, vols. xx. and xxi., together with a summary of the literature of the subject.]

The general disease dates back to the time of Hodgkin, Bright, and Wilkes, and was then known as anæmia lymphatica. It has been specifically described by Virchow,[16] Cornil, and Ranvier as independent of leukæmia, and was designated lymphadénie. It was noticed by Trousseau under the title of adénie, and Ogle and numerous clinical observers since have also recorded cases.

[Footnote 16: _Die Krankhaften Geschwülste_, Band ii. p. 376.]

As a mediastinal growth the characteristics of lympho-sarcoma can be made more vivid by the reproduction of one of the first cases recorded of this disease. On removing the sternum and cartilages they were found adherent to a mass occupying the anterior mediastinum. The morbid growth reached backward to the trachea, surrounding it with a thickness posteriorly of a quarter of an inch; it extended downward to the bifurcation of the trachea, and, involving the superior prolongation of the pericardium, invaded and greatly thickened the parietal part of that membrane, covering the heart at {869} its upper half. The diseased structure reached upward to the root of the neck, involving the anterior mediastinal glands, and surrounded the trachea by a thin layer as high as the thyroid cartilage. Some of the glands on each side of the neck were affected as high as the angle of the jaw. Laterally, the morbid growth extended on each side to the line of junction of the cartilages with their ribs, displacing the anterior margin of the lung. The pulmonary pleura was involved and thickened at the line of contact, and the right lung at the upper part of its anterior margin was invaded from the pleura by white, fibrous-looking branched bands. At the lower part of the anterior lobe the lung was also invaded from the pericardium. Some of the glands at the root of the lung were involved by extension, but they were not generally affected, nor was the lung invaded except to the limited extent above mentioned. The heart and pericardium were free from disease.[17]

[Footnote 17: Powell, _Path. Trans._, vol. xxi., London.]

The malignant growths of the mediastinal region implicate the surrounding structures so rapidly that it is, as a rule, quite impossible to determine, even after death, the starting-point of the disease; and while lymphadenoma can originate in the same tissues as the other forms of sarcoma already alluded to, yet it in most cases probably originates in the lymphatics of the anterior or posterior mediastinum.

In reference to the location of this form of morbid growth, we find, on consulting a series of cases reported by Fenwick, Eve, Payne, Peacock, Powell, Murchison, Bennett, Dickinson, that the region for principal development seems to be the anterior mediastinal space, although in several instances the posterior mediastinal region was also involved.

The characteristics of the growth of lymphadenoma are the involvement by continuity of all adjacent tissues, thus affording a contrast to secondary sarcomata. The glands of the neck are sometimes invaded, but are unaffected in a considerable proportion of cases. The lungs may be involved slowly, the growth following the lymphatic paths along the bronchial or vascular sheaths. The malignancy of lympho-sarcoma is unquestionable, but as a local growth it is less so than when the process is general; it is less malignant than cancer or certain forms of sarcoma.

Carcinoma of the Mediastinum.--Primary carcinoma of the mediastinum, as separated from the foregoing groups, is relatively rare; even as a secondary growth the same is true, unless it directly penetrates the chest-wall from a cancerous breast. The cancerous growths present a special peculiarity in the fact that they incorporate all the tissues with which they come in contact, and are followed by contraction. Carcinoma often originates in the lymph-tissue at the root of the lung, and may form a mass which may involve the bronchial glands, lower part of the trachea, the right and left bronchi, and surround the aorta and oesophagus. Scirrhous cancer frequently originates in the tissues at the root of the lung surrounding the bronchi and vessels, compressing them, and extending by branching rays through the lung-substance toward the periphery, following the course of the large bronchi, the lymph or arterial vessels. Carcinomatous formation may also originate in the follicles of the mucous glands of the bronchial tubes, and the mucous membrane of the same is frequently ulcerated by extension of the morbid process. The mucous membrane of the bronchi may be covered with villous-like formations springing from the surrounding growth.[18] Obstruction of the bronchial lumen by carcinomatous growth may prevent the expectoration of the bronchial secretions, and dilatation of the bronchial tubes may be consecutive. These dilated tubes may become filled with pus from associated bronchitis or forms of catarrhal pneumonia.

[Footnote 18: See cases by Bennett and Williams, _Lond. Path. Trans._, vols. xix. and xxiv.; also Burrows, _Med.-Chir. Trans._, vol. xxvii.]

{870} The special pathological characteristics of cancerous growths are that they exist most frequently in the posterior mediastinum, and therefore exert special pressure on the respiratory passages. Again, they are subject to contraction, by which the various pulmonary structures are fused together. Hard, nodulated, cervical glands usually appear in the supra-clavicular spaces, affording special contrast in this respect with the pure sarcomata. Since, in general, the same tissues may be affected as in lympho-sarcoma or other processes affecting the bronchial glands, a positive diagnosis can usually only be made by a microscopic study of the growth. Only one lung is usually implicated, while the sarcomata spread by extension in all directions and may involve both lungs.

The effect upon the lungs of mediastinal pressure on the bronchial tubes may be very serious. Collapse of the bronchial tubes and oedema of the lungs may ensue, or subacute catarrhal inflammation with consolidation--a process which has been described by Fuchs as a form of pneumonia under the title of apneumatosis. The affected tissues not uncommonly break down by necrotic disintegration, which may lead to the formation of cavities sometimes erroneously described as resulting from softening of cancerous nodules.

Pleural effusions are prominent in the clinical history of malignant intra-thoracic disease, and especially in mediastinal processes. These effusions are consequent on pressure on the intra-thoracic circulation, or may be traceable to inflammation, either developed by irritation of the contiguous morbid process or extension of the same upon the serous membrane. Purulent pleural collections have been noted in certain cases, and they may be hemorrhagic. In 31 cases in which the character of the effusion was mentioned, 6 only were tinged with blood. This characteristic is therefore simply of relative importance. Pericardial effusion is also possible from causes similar to those operating upon the pleural tissues. Pressure may occasion dilatation or thrombosis in the vena cava. The vessels of the neck suffer, either directly from pressure inducing dilatation, or by being converted into rigid tubes, allowing of no adaptation to the amount of blood passing through them. There may be corresponding collateral swelling of the azygos or hemi-azygos veins, and at the same time collateral circulation is established between the jugular and the subclavian on the one side and the azygos and hemi-azygos on the other through the superior intercostal veins. The external thoracic veins may, in some cases, become enlarged, and infrequently compression of the inferior cava may occasion effusion into the abdominal cavity and cause oedema of the lower extremities. Morbid growths have occasionally invaded the spinal canal and excited sufficient pressure to occasion paralysis.[19]

[Footnote 19: Bennett, _loc. cit._]

There are certain forms of mediastinal and pulmonary tumors very seldom met with; for example, fibromata and osteomata,[20] the latter sometimes occurring as an exostoses springing from the posterior surface of the sternum. Dermoid cysts of this region, as in the lungs, are also most unique. Mohr records the case of a woman æt. twenty-eight who had spat up hair since her sixteenth year. In the left lung was found a cyst which communicated with the bronchus. Inside of it was found several rounded knobs, here and there pedunculated, varying in size from a nut to a hen's egg, consisting of fibrous tissue provided with sebaceous and sweat-glands, and from which sprang numerous long hairs. The remaining contents consisted of fat and balls of hair. Teeth, bone, and cartilage can sometimes be recognized in these cysts.[21]

[Footnote 20: _Die Krankhaften Geschwülste_, ii. p. 102; Förster, _loc. cit._, p. 105; Wagner, _Arch. für Physiol. Heilk._, 1859, p. 411; Luschka, _Virchow's Arch._, Bd. x. p. 500; Förster, _Ibid._, Bd. xiii. p. 105; Didardier, _L'Union méd._, 1867, No. 83.]

[Footnote 21: _Nederland Weekblat. vor Geneesk._, 1851, p. 44.]

{871} Enchondroma may occur in the mediastinum or lungs; it is rare as a primary process, but is more often found as secondary to enchondromata of the bones.[22]

[Footnote 22: Lebert, _Physiol. Pathol._, ii. p. 213; also, Förster, _Virchow's Arch._, xiii. p. 106.]

ETIOLOGY.--Predisposing and Exciting Causes.--The etiology of morbid growths in the mediastinum, as elsewhere, is subject to debate and conjecture. The most practical query relates to location--viz. that sarcomatous growths originate in the anterior mediastinum, and carcinoma more frequently in the posterior.

The trade of shoemaker was followed by several subjects of sarcoma observed by the writer. These men were accustomed to press the last against the sternum. With a pure family history free from taint of malignant disease the etiology of sarcomata may be more readily linked with some cause of irritation than is the etiology of cancerous tumors. This irritation may be a blow or other direct injury or some local irritation, as antecedent inflammatory process in the lungs, bronchial mucous membrane, or pleura. Intemperance, insufficient food, and over-exercise have been noticed in rare instances as antecedents. In reference to lympho-sarcoma, preceding causal irritation may have existed, but in two-thirds of the cases the etiology is obscure. Hereditary transmission has not been distinctly proven in regard to any of the forms of sarcomata. The etiology of cancerous tumors is still more vague, though possibly the previously-named conditions may have preceded the growth. Louis, speaking generally upon intra-thoracic cancer, places it fourth in the scale of comparative frequency of organs affected--viz. uterus, stomach, liver, and lungs. The history of the removal of a morbid growth may attest the secondary character of some growths apparently primary.

In the question of age and sex the autopsies at Kronstadt already referred to show that in 158 malignant growths 127 were carcinomatous; 81 occurred in men of an average age of fifty-three, and 46 in women of an average age of fifty-six. So we may conclude that carcinomatous growths occur after the middle period of life. In 31 cases of sarcomatous tumors, 20 occurred in men of an average age of thirty-eight, and 11 in women of an average age of forty-eight. Powell gives 24.8 as the mean age for the occurrence of mediastinal growths in general. As a rule, a mediastinal tumor recognized at a relatively early period of life, before the thirtieth year, is most likely to be one of the forms of sarcomata. Sarcomatous tumors, however, sometimes occur in the aged; for example, in a woman æt. seventy-six (Laboriou[23]) and in a woman over sixty reported by Wilson.[24] The question of liability through sex is somewhat uncertain, but while these growths may occur in either, a slight preponderance exists in favor of the male sex, especially if the growth be a lympho-sarcoma.

[Footnote 23: _Virchow's Arch._, _loc. cit._]

[Footnote 24: _Trans. Path. Soc. Philada._, Jan., 1884.]

SYMPTOMS.--In studying the semeiotic characteristics of mediastinal growths an accurate history of the case is a prerequisite of paramount importance to a correct understanding of the essential features of the malady. It should be borne in mind that no single fact determined by the methods of physical diagnosis has special pathological significance, but simply indicates certain definite physical conditions in the region under examination. The purely objective physical signs are so closely intertwined with the general symptoms of morbid process that any study is partial which does not recognize this combination. Both physical signs and general symptoms must be in turn considered in connection with a thoughtful analysis of the processes of morbid anatomy, because symptomatology is the study of the expression of pathological changes. The general nutrition of patients suffering from primary sarcoma or lympho-sarcoma is often good in the early stages of the disease unless the oesophagus is pressed upon or implicated, and at the last {872} patients may even die in a well-nourished condition. Indeed, the special import of the peculiar respiratory disturbance with pain seems set at naught by the appearance of fair health. In some cases of sarcoma or lympho-sarcoma, however, emaciation is progressive, though slower than in cancerous growths.

When cancer itself is primary, the ordinary characteristic train of disturbances of nutrition, with cachexia, follows, and then emaciation is rapid and decided. The loss of nutrition with anæmia is more marked in secondary sarcoma, and in cases of secondary cancer cachexia is the rule.

In reviewing the clinical history attention is specially directed to the development of the mediastinal growth by the gradual increment of subjective sensations of shortness of breath, with a sense of discomfort or tightness in the chest, with or without radiating pains. The respiratory phenomena present great diversity, yet the neurotic character of the dyspnoea is characteristic. Rest or change of posture may remove all oppression, or on the least exertion dyspnoea may be at once manifested. With limited physical signs there may be great distress of breathing or orthopnoea, while in many cases with unquestioned evidence of tumor there may be only a little quickening of respiration. As a rule, tumors of the anterior mediastinum are less characterized by dyspnoea than those involving the posterior space. The dyspnoea depends upon the size and seat of the tumor, and increases day by day with its growth; but in certain cases the tumor is so placed that pressure on the trachea, bronchus, or direct pneumogastric irritation may induce severe paroxysmal attacks of dyspnoea, with laryngeal symptoms resembling the condition so common in aneurismal tumors. Pressure symptoms, traceable to irritation of the pneumogastrics, are, however, as a rule, less marked than in aneurism. Pressure on the trachea without implication of the laryngeal nerves can occasion many of the symptoms usually assigned to the latter cause.[25] Even when the tenth nerves have been surrounded or involved by the growth, special symptoms may be absent, although in other cases serious phenomena follow, such as vomiting or other gastric disturbance, or even inability to swallow; sometimes palpitation, angina, irregular action, or tendency to faintness may follow implication of the cardiac plexus.

[Footnote 25: Bristow, _St. Thomas's Hosp. Rep._, vol. lxxi.: "Influence of Pressure on Trachea without Implication of the Recurrent Laryngeal Nerves."]

The symptom of pain is usually far less than in cases of aneurism, since it is only in rare instances that the chest-walls become eroded by the outward pressure of the tumor, as so frequently occurs in aneurism. Moreover, the growth more readily adapts itself to the contour of the chest, and tends to envelop rather than compress organs or nerves. From the time that pressure commences, either on the trachea, bronchi, or intra-thoracic nerves, cough is more or less constant. It may, however, be due to pulmonary changes occasioned by the pressure or actual involvement of the lung by the growth. Cough is an earlier symptom when the growth is situated in the posterior mediastinum than when it is located anteriorly. It is usually laryngeal and ringing in timbre, and may occur paroxysmally, as in pertussis. It is ineffectual, dry, or attended with only scanty mucous or frothy expectoration. The sputa may be tinged with blood, or profuse hæmoptysis is a possible symptom. A microscopic examination of the sputa in a case of intra-thoracic tumor is always important, because portions of the morbid growth may be found, or by perforation of the trachea or oesophagus the pus from a mediastinal abscess may be mixed with the sputa. Mediastinal tumors are not, as a rule, characterized by febrile symptoms. Inflammatory complications of the lungs or pleura may account for the exceptional thermometric variations. Cases have been reported by Bennett and Church in which there was persistent elevation of temperature, with daily fluctuations and rapidity of pulse and respiration. In one instance of lympho-sarcoma the paroxysms of fever corresponded with the periods of {873} growth in the enlarged glands, but in this case the lymphatics of the general system were also implicated. From the fact that the growths are strictly mesial, dysphagia is a far more common and persistent symptom than in aneurism, especially in growths of the posterior mediastinum. When the growth is situated in the anterior mediastinum the dysphagia is less frequent; but it must be borne in mind that prolongations of the tumor may occasion lateral oesophageal pressure, or narrowing of the lumen of the oesophagus can occur from pressure upon the trachea by the growth. Exceptionally, dysphagia may be due to implication of the oesophagus in the new growth. (See Cancer of the Lung.) Neural influences may increase the dysphagia, in which case it is doubtless a reflex phenomenon and is associated with hiccough or vomiting. The passage of a bougie can be readily effected in such cases.

In reference to the foregoing pressure symptoms one fact deserves recognition--viz. that in aneurismal tumors the pressure symptoms are subject to variations in intensity due to changes in the intra-aneurismal tension, while in morbid growths in the mediastinum the pressure symptoms exhibit a progressive tendency, advance upon the same lines, and are more constant than in aneurism. With this principle in mind, the additional pressure symptoms in doubtful cases of mediastinal growth must all be considered; for instance, in some histories recorded by Rossbach the pupils could be dilated by firm pressure on the tumor above the clavicle. The pulses in the brachials or radials may be unequal, and variations of rhythm, volume, and rate may be noted as evidences of pressure, which may occasion thrombosis by retarding the circulation in the innominate, subclavian, or azygos vein. The blood may reach the heart by the collateral circulation elsewhere described or by the dilated mammary superior and inferior epigastrics and the inferior vena cava. Pressure may therefore give rise to cyanosis, oedema of the upper or lower portions of the body, with enlargement of the superficial veins, or dropsy may be traceable to hydræmia.

PHYSICAL SIGNS.--When mediastinal tumors are of small size, physical signs may afford no help in making a diagnosis, and they will always vary according to the location of the growth.

Inspection may reveal venous repletion of the veins of the face and neck, with distension of the superficial veins of the chest; the latter symptom is more frequently obvious than in aneurism. In the case of sarcoma represented by Fig. 53 the foreign growth was so limited to the mesial line as not to involve the vessels or create pressure symptoms upon them. If the anterior mediastinum is implicated, there may be circumscribed alterations in the contour of the chest. Prominence of the upper piece of the sternum and of the sternal attachment of one or more ribs may be recognized. The sternum itself may appear thickened upon palpation of the notch. One side of the chest may be larger than the other above the nipple-line; the affected side, however, may be smaller, since vicarious respiratory function may create distension. The usual changes in the contour of the chest-walls will indicate pleural effusions. (See Cancer of the Lungs.) Since tumors of the anterior mediastinum overlie the aorta, transmitted pulsation may be detected in rare instances; this pulsation can be differentiated from aneurismal vibrations by the absence of the sense of expansile pulsation characteristic of dilated aorta or aneurism, but it sometimes closely resembles that yielded by an aneurismal sac thickly lined by coagulum.

Lympho-sarcoma and cancer are often accompanied by painless, movable glandular enlargements, recognizable by palpation in the supra-clavicular spaces; but the absence of the glandular implication in sarcomata is conspicuous; swelling occasionally manifests itself in the suprasternal notch. Tumors of the posterior mediastinum must attain considerable size before {874} they can be recognizable by the foregoing methods. Growths in the latter space are those especially liable to complication by pleural effusion in one or both sides. Mediastinal growths may occasion collapse of the lung, or cirrhotic processes or pleural adhesions may diminish the circumferential measurements of the chest. The heart may be displaced backward, downward, to the left or to the right side; and since in aneurism, uncombined with valvular disease, little cardiac displacement occurs, this sign is of importance. Rarely, as in Cotton's case in Brompton Hospital, the heart may be fixed in situ by the extension of the growth on both sides of it.

Percussion.--It has been observed that a very small tumor may, from its particular site, at a very early stage give rise to symptoms both of pressure and disordered innervation of great severity, whilst another may attain considerable magnitude before the patient experiences any distress or any decided evidence of pressure is manifested. It is equally true that percussion and auscultation may be most valuable, or, on the other hand, indefinite. The degree of dulness occasioned by a morbid growth in the anterior mediastinum is dependent on its size, large growths yielding flatness; but when the tumors are small the osteal resonance of the sternum is simply hardened. Respiratory percussion is available if the growth lies anteriorly. The full, clear resonance of full-held inspiration contrasts with the increased dulness developed when the lungs are stripped from the mediastinal space by forced expiration. The boundaries of the growth on either side of the sternum may be defined by percussion, and it is possible that the greater part of one side may be occupied by the new formation. The adjacent tissues are involved by direct invasion, or indirectly by extension along the bronchus from behind forward, thus involving the middle tier of the lung. Mediastinal tumors therefore in their mode of growth yield a contrast with pleural effusions, because the latter usually advance steadily from below upward. When the growth is located in the posterior mediastinum, percussion should be practised after the manner recommended by Mussey to facilitate recognition of enlarged bronchial glands. (See Pulmonary Syphilis.) The apices, humeral, scapular, basic, or marginal regions often yield a tympanitic type of resonance, since they are often in a condition of vesicular emphysema. The pericardial sac may be distended with effusion or implicated in the growth, and an area of pyramidal dulness with the base above may be recognizable.

Auscultation.--In growths situated anteriorly, in the mesial line, one of the most forcible lessons may be impressed by the distance and obscurity of the second sounds of the heart over the aortic and pulmonary artery, cartilages, or the upper piece of the sternum. The cardiac sounds may be transmitted downward, and can be heard distinctly in some abnormal position. Even a murmur can occur due to compression of the aorta, or pericardial friction. The respiratory murmur will probably seem feeble and distant over one or both apices, and whistling near the trachea.

If the posterior mediastinal space be involved, the respiratory murmur may represent some type of bronchial breathing, or if the lumen of the trachea or one of the bronchi be decidedly lessened, the respiratory murmur may be whistling, feeble, or suppressed over the affected side. Over the other bronchus the respiratory murmur may be more high-pitched than in health, and slightly exaggerated. The rhythm is often jerky and paroxysmal; the paroxysms are more or less constant, but are liable at times to increase. Auscultation should be especially practised over the roots of the lungs or in the neighborhood of the second dorsal vertebra. Frequently it can be demonstrated, both by auscultation and percussion, that there is diminished air-supply to one or other of the lungs, while the respiratory murmur is not sufficiently changed for classification. The respiration may acquire a stridulous or sibilant character, most marked on expiration, but {875} less often than in aneurism, because there is a greater tendency to occlusion of the bronchi. It should always be remembered that the lung undergoes very various and opposite changes as the result of pressure on the bronchi, interrupting the entrance and egress of air from the lobules, and the physical signs of emphysema, infarction, congestion, or consolidation may exist in one or the other side. The ordinary methods of physical examination indicate the existence of pleural effusions, but large growths extending from the mediastinum or originating in the lung may so closely simulate such effusions that a positive diagnosis can be arrived at only by paracentesis.

When tumors exist in the form of very small nodules as diffused sarcomata, no changes in the character of the respiration may be noted. Friction râles and pleuro-pericardial frictions may be heard in some cases. Distension of the bronchial tubes from pressure may occasion the dilatation of the distal bronchial passages and pulmonary lobules with retained muco-purulent secretions. The cross-sections of the bronchi have been described as multiple abscesses. Areas of collapse or slow inflammation with softening of the secondary inflammatory product can follow. The bronchial pressure may prevent the sufficient transit of air through the bronchi to create râles, or moist râles indicative of tracheo-bronchitis or oedema may abound.

The study of the vocal resonance and fremitus presents nothing novel, but corresponds with the generally-understood principles.

DURATION.--It is very difficult to determine accurately the duration of malignant diseases of the mediastinum, since for a long time the patient may be quite free from any local subjective symptom, even though a growth may have attained to a considerable size. Moreover, intra-thoracic malignant disease, especially in the non-cancerous varieties and if the digestive tract be normal, may be unattended by any of those symptoms commonly associated with malignant process, such as a peculiar tint of skin, progressive and great emaciation, or the aspect of suffering. Sarcomatous tumors usually grow rapidly, as in a case related by Jaccoud, in which death occurred within eight days after admission to the hospital. Prior to this time the patient had suffered from no objective symptoms whatever, although when admitted there was physical evidence of a large growth extending from the clavicle to the nipple.[26] West records a fatal case at two and a half months; Horstman, one in which the disease originated on the right of the sternum, as evidenced by a very small area of dulness; the entire right side of the thorax was invaded within five weeks.[27] Berevidge reports a case of sudden death from hæmoptysis in a man aged sixty-four years, who up to that time had appeared healthy, and only a few days before had complained of a slight cough and a feeling of oppression in the chest. At the autopsy two cancerous masses the size of a hazelnut were found, one of which overlaid a bronchus which was ulcerated to a considerable extent. The bronchi were filled with blood. Virchow mentions a case the duration of which was only two months. Walsh, speaking of malignant growths in general, assigns three and a half months as the minimum duration of these cases.

[Footnote 26: _Leçon de Clin. méd._, p. 636, Paris, 1867.]

[Footnote 27: _Trans. Path. Soc. London_, 1883.]

Undoubtedly, the duration will depend on the freedom from pressure upon the oesophagus, or from interference with digestion due to pneumogastric irritation, or from malignant processes in the stomach or intestines. Pain, and consequent loss of sleep, will also accelerate the termination of any case. Lebert assumes an average duration of thirteen months, and Walsh states the maximum duration in intra-thoracic malignant processes at twenty-seven months. The soft secondary malignant sarcomata or carcinomata grow more quickly, and have a relatively shorter course, than the harder forms of the same species. Lymphadenoma may persist a long time, and appear for a while to be stationary and unattended by any serious impairment of the general {876} health, but the cases are exceptional. The persistence of fibrous, fatty, or cystic tumors depends chiefly on the mechanical inconvenience occasioned by them. All forms of malignant intra-thoracic disease, however, are steadily progressive to a fatal termination. Death commonly arises from the gradual increase in seriousness of the pressure symptoms. Inability to lie down, harassing cough, want of sleep, all tend to induce fatigue which may prove fatal. Deficient aëration of the blood may occasion stupor, or sudden simultaneous pleural and pericardial effusion or general pulmonary oedema may terminate the scene. In exceptional instances death has resulted from laryngeal spasm or from acute hypertrophy of the thyroid gland with tracheal occlusion. In a remarkable case reported by Bennett paroxysmal dyspnoea had been the only symptom of intra-thoracic disease for a few months, when suddenly a severe seizure occurred which persisted uninterruptedly for three days, till weakness and exhaustion terminated in death by asphyxia. In this case the thyroid gland was found enlarged to the size of a double fist, but the enlargement was mainly below the sternum and along the sides of the trachea, which was literally surrounded by the greatly-enlarged and firm lateral lobes of the thyroid, so as to be completely flattened laterally. The structure of the thyroid appeared healthy, but very firm, and the enlargement was due solely to hypertrophy, and not to cystic or other disease, nor was there any exophthalmos.[28] Death is possible from sudden asthmatic attack, or, more rarely of all, by hæmoptysis.

[Footnote 28: See "Cancerous and Other Intra-thoracic Growths," Bennett, _The Lumleian Lect._, 1872, p. 169.]

PROGNOSIS.--The prognosis is invariably unfavorable, and must continue so unless the more recent attempts for removal of primary growths in the anterior thoracic regions yield grounds for a more hopeful outlook. We may also hope that some remedy may influence or control the development of lymphoma. Considerable relief may be obtained by rest, suitable feeding, careful regulation of the digestive system, and such hygienic measures as may seem most available.

DIAGNOSIS.--From Aneurism.--When we consider that in the diagnosis of aneurism of the aorta every sign and symptom has in turn been found fallacious in the ever-varying conditions under which aneurisms appear, and that one is forced to say that aneurism has no pathognomonic signs or symptoms, the difficulties in the way of the diagnosis of intra-thoracic morbid growths may be recognized. Moreover, the diversity in the peculiarities of each case, the multifarious character of the pressure symptoms and physical signs, and the absence of a precise order of phenomena peculiar to tumors in this situation, may render a positive diagnosis in the early stages very difficult.

Aneurism in the absence of unequivocal signs of its existence may be excluded on the following grounds: the absence of conditions which predispose to disease of the coats of the arteries--_i.e._ syphilis, alcoholism, Bright's disease, rheumatism, laborious avocations, violent exercise. Aneurism may occur at any age, but it is rare before the age of thirty years, and most prevalent between the ages of forty and fifty years. Aneurism is also less frequent in the female sex. The distal pressure symptoms of aneurism are more variable than in other morbid growths of the mediastinum, and especially dysphagia is less constant. Great emaciation without intense pain is adverse to the diagnosis of aneurism, while severe pain with occasional exacerbations is favorable to this diagnosis. However, instances of morbid growths are recorded in which intercosto-humeral neuralgia was an initial symptom.

"An extensive area of dulness must in aneurism mean a large sac, and with such a large tumor we should almost invariably get marked expansive pulsation. Again, aneurismal sacs, before they produce extensive dulness in {877} any portion of the parietes of the chest, point, as it were, in some particular direction, becoming distinctly prominent and producing an eccentric motion around them in consequence of the thoracic parietes being absorbed or yielding at the point of greatest pressure" (Graves). Hæmoptysis may occur not only from aneurismal leakage, but from the effects of pressure of morbid growths upon a bronchus or the invasion of the same by the malignant process. Blood-spitting cannot therefore be regarded as an important differential symptom. Unless valvular disease be associated with aneurism, the displacement of the heart is less frequent in aneurism than in morbid growths.

From Abscess.--The etiological relation in this process is traumatic, or mediastinal abscess occurs in connection with caries or fracture or after an operation in the neighborhood of the throat or neck, or of suppurative disease elsewhere in the thorax, as abscess of the lung or empyema. The pain in cases of abscess is deep-seated, constant, slowly increasing, rather than the paroxysmal pain of aneurism or solid tumor. The febrile movement may afford decided aid in the diagnosis, but it is also true that high temperature may mark the progress of lymphadenomata, as in Bennett and Sutton's case, in which from Jan. 11th to Feb. 28th the thermometrical wave vibrated between 103.5° maximum, with a pulse of 148 per minute, to 100.5° minimum, with a pulse of 108. In this remarkable case sweating was also a prominent feature; and a somewhat similar example has been recorded by Murchison. In corresponding circumstances the existence of secondary processes in the lungs or elsewhere, with enlarged glands in the neck, may prevent error. In mediastinal abscess there will probably be a tendency to point, with the appearance of a fluctuating, circumscribed, superficial tumor at the sternal border or adjacent to this bone. There may also be tenderness on pressure associated with the pain, and an oedematous condition of the tissues of that portion of the sternal region covering the tumor, although this symptom sometimes attends malignant new formations. Pulsation may accompany abscess, but will be of the transmitted variety. In suspicious cases the sternal bone can be drilled and an exploratory needle introduced into the tumor.

The general diagnosis of mediastinal tumor can be more easily made upon the basis of regional invasion. But in any suspicious case an elaborate and thorough clinical history is an essential prerequisite. In proportion as one completes the natural history of a case of obscure intra-thoracic disease the more likely one is to approach by exclusion a correct interpretation of the existing physical signs and symptoms.

Growths in the Anterior Mediastinum.--Tumors located in the anterior mediastinal space overlie the heart and aorta, and consequently the heart-sounds, especially the second, may be indistinct or muffled; or the second sounds may be audible in some new situation, owing to displacement of the heart. The sternal region may be distinctly prominent or bulged, and at the notch the bone may appear thickened. The resonance in the interscapular regions remains unimpaired, but over the sternum percussion should yield a very dull sound if the growth be large, but when a comparatively small tumor exists the sternal resonance will be hardened and high-pitched. An additional explanation of this modification exists in instances where the growth is not adherent to the sternum and the bone is arched over the tumor.

The respiration may be whistling or stridulous if the stethoscope is placed over the trachea, and over one or other apex anteriorly the respiratory murmur may be feeble or blowing, in proportion to the volume of air which is permitted to enter the chest. Posteriorly, the respiratory murmur may be unaffected at first, although as the growth advances evidence of pressure on the bronchial tubes may be detected over the interscapular region. The superficial veins of the chest may be enlarged, especially those below the level of the upper segment of the sternum. Dysphagia is usually slight in {878} proportion to the other pressure symptoms or entirely absent. It may be simply a symptom of irritation of the intra-thoracic nerves or due to enlargement of the glands of the mediastinum.

Mediastinal growths usually develop in the middle line; they spread in all directions, especially laterally, but avoid at first the roots of the lungs. Pressure is rather exercised upon the parts in the mesial line. They reach a large size and grow with great rapidity, producing symptoms rather as a consequence of their size than by virtue of contractile properties.

From Pericarditis.--A possible pericarditis may be mistaken for a tumor of the anterior mediastinum. The diagnosis of pericarditis must be sustained by evidence showing the dependence of this process upon rheumatism, syphilis, nephritis, or propagated inflammation. The distension of the pericardial sac due to pericarditis exhibits a definite outline. The dulness of a tumor is irregular, with a tier of dulness upon a higher level than in effusion. The absence of various pressure signs is marked in pericarditis, while disturbance of the heart's rhythm is more frequent. Kussmaul states that there are two signs characteristic of chronic pericardial inflammation with thickening and adhesion--viz.: a "complete or almost complete failure of the radial pulse during inspiration, and simultaneously visible swelling of the great veins of the neck, instead of the collapse that usually takes place during this portion of the expiratory act. Adhesion of the great vessels to the sternum, either directly or through the medium of the pericardium, is supposed to account for these phenomena."

Febrile movement is usually present in pericarditis, and, while a possible temporary feature in new growths, is not persistent unless complicated by inflammation in the pulmonary tissues. Finally, the progress of the case will often decide the question.

Growths in the Posterior Mediastinum.--In growths located in the posterior mediastinum one or the other bronchus is one of the earliest structures implicated by the pressure, because in these cases the chief mass of tumor is found at the root of the lung. Secondary lesions in the lungs directly traceable to pressure are frequent, but unilateral, although secondary cancer from malignant lesions elsewhere than in the lungs may be bilateral. Pressure symptoms as a class occur early, are grave, constant, and progressive. Percussion according to directions of Guéneau de Mussy may be made available. Abolition or great impairment of breath and voice sounds over one or other posterior aspect of the chest is the rule, since these tumors are prone to contraction. Sometimes the respiratory murmur is whistling or blowing if the bronchial pressure is less decided. Progressive emaciation and cachexia are commonly present, not only from the inherent tendencies of the disease, but also depending upon the disturbance of the functions of many important organs which have been encroached upon by the tumor. The exclusion of a malignant disease of the oesophagus is very difficult. The passage of a bougie might determine the seat of obstruction, and thus assist in the diagnosis, but great caution must be observed lest penetration of the softened tissues occur. (See Cancer of Oesophagus.)

From Pleural Effusion.--The greatest difficulty may be experienced in deciding between uncomplicated pleurisy and effusion complicated by morbid growth.

Aside from the history of the case and state of nutrition, paracentesis may aid the diagnosis, since, if the fluid is turbid, highly albuminous, with a large proportion of coagulable fibrin, it is an evidence of inflammatory origin; but if it is clear, limpid, and on standing gives but a delicate veil of pseudo-fibrin, it indicates a passive or mechanical cause. Hemorrhagic exudation is only of relative importance. The recognition of pleural friction râles over parts flat on percussion will be an evidence of tumor. Hæmoptysis in this {879} association would negative the idea of simple effusion. The presence of signs of pressure on central parts is indicative of tumor (Walsh), but Powell has recorded an instance of simple pleural effusion accompanied by husky voice and laryngeal cough; and also an instance in which, from a similar cause, there was increased size, tortuosity, and throbbing of the radial and brachial arteries on the affected side without oedema of the limb, yet probably attributable to obstruction of the return circulation.

Enlarged glands in the neck, or enlarged veins with evidence of thrombosis of the descending vena cava, would indicate tumor. Dulness from a tumor itself might resemble sacculated effusion, yet there might be retraction in place of distension of the chest, and particularly characteristic dulness in the mediastinal region as compared with the circumferential regions, or peripheral patches of resonance may be suggestive and lead to critical revision of the symptoms.

From Chronic Pneumonia.--Mediastinal growth invading the lung from its root has been mistaken for chronic pneumonia. Walsh lays stress on the following signs as distinguishing tumor: 1. A tendency to increase instead of diminution of bulk of the affected side. 2. Implication of the mediastinum, with dyspnoea out of proportion to the extent of consolidation. 3. Different characters of respiration in the two diseases. To these may be added pressure symptoms in general in cases of tumor, with displacement of the heart toward the side unaffected by the pulmonary process. Hæmoptysis is very often a concomitant of bronchial pressure, but occurs so frequently in basic pneumonia, especially in the syphilitic, that it is devoid of importance except from the standpoint of relative investigation. With reference to symptoms of bronchial irritation without assignable cause, we should always do well to remember the observation of Stokes, that they may be characteristic of disseminated morbid process.

Differentiation of Malignant Growths.--The younger the patient the more probable the existence of lymphoma or sarcoma. The majority of primary tumors of the mediastinum are lymphomatous, and when the growths originate in the anterior space they are almost certainly lympho-sarcoma or sarcoma. Widespread enlargement of the lymphatic glands, with or without enlargement of the spleen, indicates a lymphadenoma.

Finally, primary lympho-sarcoma or sarcoma tends to spread by extension of the process by continuity of structure, although secondary forms of the process present lesions distributed through the lungs.

The evidence in favor of sarcoma may be drawn from exclusion of the other forms of morbid process, from the rapidity of the growth, and from the history of previous operative interference for the removal of foreign growth, especially if the previous disease were sarcomatous.

Carcinomata may be suspected in cases in which there has been an hereditary predisposition to carcinomatous disease or the previous or concomitant existence of cancerous disease in the mammæ or elsewhere, particularly if the period of life is relatively advanced. The development of the tumor may be more slow than other forms of growth, and is associated with tendency to progressive emaciation in the absence of evidences of direct pressure on the oesophagus and the existence of cachexia. Carcinomatous disease is more commonly coincident with the presence of hard, nodular, immovable masses in the neck.

Cystic tumors present signs of fluctuation. Syphilitic gummata must be diagnosticated by exclusion and the existence of the syphilitic history. The possibility of substernal thickening due to syphilis, with reflex disturbances, particularly oesophageal spasm, must be borne in mind.

Those rare forms of disease due to hyperplasia or caseous deposit in the thoracic glands, independent of pulmonary disease, must be recognized by {880} exclusion. The fact must be remembered that with great enlargement of glands in the neck and elsewhere the bronchial glands may remain constantly unaffected.

TREATMENT.--From the inaccessible location of these growths but little assistance can be rendered by surgery. The progress of this branch of science has of recent years included resection or excision of the sternum or some of the ribs for the removal of growths involving the mediastinum or pleura. Küster[29] has successfully made partial resections of the sternum for the removal of mediastinal tumors, and the entire bone has been excised by König[30] in a case of sarcoma. The pericardial and both pleural cavities were opened in the course of the dissection; the wound became gangrenous, and the heart was afterward surrounded with pus: notwithstanding this, the wound slowly healed and the patient ultimately recovered. In cases treated by this method pleural adhesions usually prevent double pneumothorax; portions of the ribs have been resected with the sternum, and have been succeeded by unilateral pneumothorax, and recovery has ensued. (See Fig. 55.)

[Footnote 29: _Berliner klinische Wochenschrift_, No. 20, 1883, pp. 127, 136, 274.]

[Footnote 30: _Centralblatt f. Chir._, No. 42, 1882.]

Paracentesis must sometimes be practised to relieve accumulation of fluid in the pleural sacs in instances in which dyspnoea is serious, and life may be prolonged by repeatedly practising this operation. Reflex laryngeal irritation, or paroxysmal dyspnoea with stridulous breathing, requires the use of inhalations or atomization of antispasmodics, and among the most useful of these are ether and chloroform. This group of neural symptoms can sometimes be markedly palliated by hypodermic use of morphia with atropia. But too often the symptoms are caused by actual pressure, and not by nerve-irritation, and this mode of treatment is futile, and therefore these measures should be employed with caution.

{881} Sleeplessness, cough, bronchial or other pulmonary complications, must be managed upon general principles. The local pains may be met by local treatment, such as mustard sinapisms or soothing lotions; even blisters may secure temporary relief. The digestive system should be carefully studied, and assimilable and appropriate food should be selected. In lymphadenoma combinations of iodine with arsenic, as in Donovan's solution, may be tried, but, unfortunately, the utmost aid from present resources consists in a palliative and expectant policy.

{882}

DISEASES OF THE BLOOD AND BLOOD-GLANDULAR SYSTEM.

BY WILLIAM OSLER, M.D.

INTRODUCTION.

The blood is a fluid tissue composed of cells floating in an albuminous plasma, and it differs from other tissues not less in the arrangement of its elements than in the activity of the changes which go on in it. It is the mart into which is poured from the alimentary canal the commodities needed in nutrition, and the elements of the body select from it the various materials which they require, giving in exchange those chemical combinations which result from the metabolism of the tissues. In spite of ceaseless changes, a uniformity of composition is one of the most striking features of the blood in health. This is maintained, as regards the constituents of the plasma, by the activity of the organs which regulate income and expenditure--the alimentary canal and liver on the one hand, and the kidneys, lungs, and skin on the other; while histological uniformity is maintained by the adenoid or cytogenous tissue throughout the body, the function of which is to replace the wornout blood-corpuscles.

The corpuscles form rather less than one-half by weight of the blood. The plasma contains about 90 per cent. of water, which holds in solution proteids in the form of serum, albumen, and the fibrin-forming factors; sugar in traces; creatin, hypoxanthin, and urea; various fatty bodies in small amount; salts, chiefly sodium; and gases. The corpuscles (red) consist of hæmoglobin (90 per cent.), proteid bodies, and traces of lecithin and cholesterin.

So far as we know at present of the function of these two portions of the blood, the plasma ministers to the general nutrition of the tissues, while the corpuscles (red) are chiefly concerned with respiratory processes, acting as the carriers of oxygen and carbonic oxide.

We shall first give a brief account of the histological characters of the blood, and of the relation of the groups of adenoid or cytogenous tissue to the corpuscles.

Two forms of corpuscles are usually described, but we can recognize four varieties of blood-corpuscles in the body: (1) red, (2) white, (3) nucleated red, and (4) the hæmatoblasts (Hayem), or blood-plates of Bizzozero.

(1) Red Corpuscles.--In each cubic millimeter of plasma there are about 5,000,000 red cells. The percentage may vary within health limits from 90 to 110. The corpuscles are circular, non-nucleated, biconcave disks, homogeneous, to ordinary inspection structureless, and consist of a colorless stroma which is possibly reticulated, and a red coloring matter, the hæmoglobin. In {883} health they are tolerably uniform in size, about 7.9 µ[1] in diameter, or in English measurement 1/3200 of an inch (Gulliver). Even in normal blood there may be slight variations in size between 6.5 µ and 8.5 µ, the average, according to Hayem, being 7.5 µ.

[Footnote 1: µ is used to signify a micro-millimeter or 1/1000 part of a millimeter.]

(2) Colorless or white corpuscles, nucleated masses of protoplasm, with an average diameter of 10 µ, or about 1/2500 of an inch. The majority have a finely granular protoplasm, but in a few the granules are coarse and do not completely fill the clear protoplasm. The ultimate structure is reticular (Heitzman). Erhlich[2] has shown by their varying reaction to eosin that there are chemical differences among the colorless cells quite unrecognizable by other means. In healthy blood they display active amoeboid changes at ordinary temperatures. Their protoplasm does not, as is commonly stated, rapidly disintegrate, but if kept at a medium temperature retains its vitality, as shown by movements, for hours. The number per cubic millimeter is from 8 to 15 millions, and the ratio to the red is variously computed as 1 to 300 or 1 to 500.

[Footnote 2: _Frerichs find Leyden's Archiv_, Bd. i.]

(3) Nucleated red corpuscles, which occur in the blood of the foetus and the infant, gradually diminishing until at the third or fourth year they disappear. In the adult they do not occur in the blood in health, but are normal constituents of the red marrow of the short bones. They measure from 1/1500 to 1/2000 of an inch, and are of somewhat variable intensity of color, often quite as deep as the ordinary red forms. There may be two or even three nuclei, not colored, grouped together, often eccentric, and in some instances protruding from the cell.

(4) The hæmatoblasts of Hayem, the blood-plates of Bizzozero, the elementary or intermediate corpuscles, are small discoid colorless corpuscles about 3 µ in diameter, and are normal constituents of healthy blood. When the blood is withdrawn, they aggregate together into irregular clumps or masses, which have long been known as Schultze's granule-masses. It can be readily demonstrated in new-born rats or kittens, in which these masses abound, that the corpuscles composing them are isolated in the vessels, and only run together when the blood is drawn. The statement is commonly made that the granule-masses of Schultze result from the disintegration of the white corpuscle (of the red, Erhlich), but half an hour's study of the question in a new-born rat will convince any competent histologist that we have here to do with a separate blood-element.[3] It appears to have important relations with the production of fibrin.

[Footnote 3: Consult _Proceedings Royal Society_, 1874; _Centralblatt f. d. Med. Wissenschaften_; _Medical News_, 1882, 2; Bizzozero, _Virchow's Archiv_, Bd. xi.; Hayem, _Recherches sur l'Anatomie normal et pathologique du Sang_, Paris, 1878.]

Of the origin and life-history of the red corpuscles during post-embryonic life we have still much to learn. They are stated to develop--

(1) From colorless corpuscles, the lymph-cells or leucocytes. In the lymph-glands, the Malpighian bodies of the spleen, in the thymus, or the adenoid tissue of the tonsil, of the lymph-elements in the intestines and other regions, colorless cells are constantly being manufactured, and the general belief has been since Hewson's time that the red corpuscles develop in some way or other from these leucocytes. How or where has not yet been settled. It does not apparently go on in the blood, or we should surely catch, in the many observations and with the excellent powers now in use, a glimpse of the birth of one of them. Some observers (Johnstone[4]) maintain that they develop from the granular protoplasm of the adenoid reticulum by a process of budding. This may be so, but we should expect to find the lymph in the efferent {884} vessels and of the thoracic duct much more rich in red cells than is usually the case, and in specimens of healthy glands we should find young-looking elements such as he describes.

[Footnote 4: _Seguin's Archiv_, vol. vi.]

(2) From the nucleated red corpuscles. In the embryo this undoubtedly takes place, and as the weeks of development proceed, the ordinary red forms gradually predominate. In the child the red nucleated cells disappear early, and are then found only in the red marrow. So far as my observations go,[5] they apparently originate from colorless marrow-cells, which gradually become more homogeneous, and hæmoglobin develops in the protoplasm. The nucleus degenerates and disappears, when the cell has the appearance of an ordinary red disk. Rindfleisch thinks that the nucleus of the nucleated red is extruded in the development. It is possible that from these nucleated red corpuscles cells may originate in another way--viz. by budding. This I have seen and sketched in the marrow-cells,[6] and Malassez has studied the same process.[7] The gemmæ are small, and sprout from the protoplasm, not the nuclei, and when they break off they resemble the microcytes which occur so abundantly in certain anæmic states. Bizzozero[8] holds that these nucleated red corpuscles are independent elements which do not develop from the colorless marrow-cells. They multiply by fission, and develop into the ordinary red forms with the disappearance of the nuclei. Several recent investigations support this view.[9]

[Footnote 5: _Centralblatt f. d. Med. Wissensch._, 1878.]

[Footnote 6: _Trans. Am. Ass. Ad. Science_, 1882.]

[Footnote 7: _Archives de Physiologie_, 1882.]

[Footnote 8: _Centralblatt f. d. Med. Wissenschaften_, Bd. xix.]

[Footnote 9: _Fortschritte der Medicin_, 1885, No. 1.]