for the purpose of diagnosis, which for the moment is relatively unimportant. Contrary to the opinion entertained by some that the character of the treatment is directly dependent upon the nature of the effusion, let it be asserted that among consumptives the question of entering the pleural cavity should be decided strictly upon the combination of symptoms and physical signs. In all such cases these may furnish sufficient data to constitute a safe working basis without recourse to that refinement of diagnosis which exalts the findings of the laboratory and the autopsy at the expense of the patient. If the clinical manifestations warrant the performance of aspiration, or assuming that the pleural cavity has already been entered, it is conceded that the future management is subject to some extent to the character of the effusion. The present contention is simply that among pulmonary invalids the precise determination of the nature of the liquid by routine exploratory puncture is entirely unnecessary as regards a future course of action, in view of the guidance and direction afforded by other means.

CHAPTER LIV

EMPYEMA

Clinical Manifestations. Before proceeding directly to the treatment of purulent pleurisies it is well to call attention briefly to certain features of symptomatology and diagnosis which are especially common. to this variety.

On account of the presence of unyielding adhesions, circumscribed empyema is more frequent among consumptives than among non-tuberculous individuals. Circumscribed purulent exudates may exist not only between the lung and the chest-wall, but also between the diaphragm and the lung or the separate lobes of the lung. Empyema may result from caries of rib or vertebra, serous pleurisy, or an intercurring pneumonia and typhoid. Allusion has been made to the fact that effusions of this character may be of gradual or of acute onset, accompanied in some cases with general systemic infection, and in others without the slightest evidence of constitutional disturbance. The younger the patient, the more likely are the symptoms to be acute in nature. Cases associated with systemic infection are characterized by chills, fever, sweats, and frequently leukocytosis. Cases exhibiting marked septic phenomena are usually of streptococcic, pneumococcic, or staphylococcic origin. Pain is comparatively unimportant, other than it may at times afford an indication of the probable site of the infection. Localized tenderness is of much greater importance than pain. Musser lays great stress upon firm and deep pressure in the interspaces in an effort to elicit a point of tenderness which may suggest the site of the empyema. Purulent effusions, however, often exist entirely devoid of any rational symptoms suggestive of their presence. Their early recognition necessitates painstaking examinations of the chest, which should be repeated at short intervals.

Among pulmonary invalids unilateral immobility of the chest-wall or a localized impairment of expansion is of much less significance than

among the non-tuberculous, because of other pathologic changes capable of producing a limited respiratory movement. The attention of the examiner, however, may be directed by this means to detailed physical investigation. Edema of the chest-wall is sometimes present, though not a constant accompaniment of empyema. In some cases the pus accumulation may rupture spontaneously into a bronchial tube and be discharged through the mouth in large quantities, or produce death from inundation of the bronchial tract. Rarely, the purulent exudate may point externally and be evacuated in this manner-the so-called 66 empyema necessitatis." Among consumptives the vocal fremitus in interlobar empyemas is vague and capable of misinterpretation, on account of the transmission of the vibrations through areas of pulmonary consolidation.

A consideration of some importance pertaining to percussion signs in such cases is the relation of the area of dulness to the interlobar septa. Musser has been particularly successful in localizing purulent empyemas by following the lines of the septa. He reports the area of dulness well below the lobar fissures in cases in which the empyema exists deeply between the lobes. Despite the absence of breath- and voice-sounds over a given area it is not especially infrequent to distinguish in these cases an increased skodaic tympany.

On account of the antecedent pathologic changes in the lung and pleura among pulmonary invalids, but little importance need be attached. to the shape or location of the pleural effusion.

An auscultatory feature of special interest in connection with the small empyemas of consumptives is the remarkable frequency with which moist bubbling râles are transmitted with perfect clearness through a considerable effusion. This phenomenon has frequently led to errors in diagnosis because of the inference that the presence of loud bubbling râles is incompatible with an effusion.

Exploratory Puncture. The possible dangers attending exploratory puncture are more practical than is usually supposed. The negative results which so often attend its employment are, to say the least, misleading, and permit of dangerous delays. This line of remark with reference to the inadvisability of paracentesis for diagnostic purposes is not to be construed as opposing aspiration with a large-sized needle whenever the symptoms and physical signs suggest the expediency of such undertaking. I would sooner resort to efficient surgical exploration of a suspicious area, provided the subjective and objective signs were sufficiently clear, than to dismiss such interference solely on the evidence of repeated negative exploratory puncture.

It is unwise, however, to accept the more or less radical position assumed by some in favor of surgical investigation, unless there is clinical evidence of an undoubted localized focus of infection. Certain it is that the chest cannot be opened and explored with the same impunity as the abdomen, on account of the possibility of inducing a dangerous pneumothorax. I will cite briefly an interesting experience which illustrates the difficulties so frequently encountered in the diagnosis and management of intrathoracic disease.

The patient was a woman of twenty-seven years who came to Colorado from Tennessee on account of suspected pulmonary tuberculosis. Her illness had been of six months' duration, following an acute onset which was characterized by severe pleuritic pain lasting nearly six weeks.

There had been much loss of strength and flesh, with constant fever and severe cough, the expectoration being purulent and amounting to about six ounces daily. Upon arrival she was markedly anemic, emaciated, and experienced moderate dyspnea upon slight exertion, the temperature ranging between 102° and 103° F., and the pulse from 120 to 130. There were almost daily chills. Upon examination the respiratory movements upon the right side were found somewhat limited. There was moderate dulness in the back from the spine of the scapula to the base; also in front from the fourth rib to the base, and in the axilla. Respiratory sounds throughout this region were enfeebled and markedly distant. There was complete absence of vocal fremitus and vocal resonance. The symptoms and signs pointed to an intrathoracic pus-collection, probably of pleural origin. Aspiration was practised in the eighth interspace in the line of the angle of the scapula with negative result. This was repeated several times during the ensuing week, until at least seven or eight punctures had been made. Despite failure to discover pus it was impossible to entertain any other diagnosis than that of an abscess within the chest. It was determined to explore the lung itself, but, owing to inability to elicit tenderness at any point, or to detect a sharply circumscribed area of flatness, it was somewhat difficult to select a site for deep exploration. The needle was inserted nearly to its full length (four inches) at a point just below and slightly within the lower angle of the scapula. Something less than an ounce of pus was withdrawn, which was found to be of streptococcic nature. One or two days subsequently rib resection was performed at this point by Dr. Charles A. Powers. Extremely firm and extensive pleural adhesions entirely obliterated the pleural cavity at the site of operation. The insertion of the needle through the deeply injected pleura into the lung resulted in the withdrawal of one-half teaspoonful to one teaspoonful of pus. A deep incision was made into the lung, and was followed by moderate finger curetage. All the lung tissue within reach of the finger was found honeycombed with very small pus-cavities. The trabeculæ were broken down as much as possible with the finger, and a single pulmonary cavity, the size of a small orange, was produced into which drainage-tubes were inserted. There was no elevation of temperature following the operation. The patient gained thirty-five pounds in weight, assumed a healthy appearance, and was sent home at the end of three and one-half months in excellent general condition, the physical signs, however, remaining practically unaltered. Advices received from her attending surgeon in Nashville, Dr. McGannon, are to the effect that her general condition remains excellent, although the abscess is discharging slightly. She has had one or two slight hemorrhages following paroxysmal cough. The site of the operation is shown in the accompanying photograph (Fig. 104), taken shortly after her return. The extent of pathologic change in the lung is seen in the skiagraph recently made. The important lesson to be learned from such an experience is to the effect that exploratory operation should be resorted to despite negative punctures, provided the symptoms and signs point strongly toward an intrathoracic pus-collection. I know this to be contrary to the teaching of many surgeons, who decline to extend operative interference in thoracic cases unless a verification of the condition is afforded by the use of the needle. It is easy to appreciate that in this case pus might not have been discovered even after numerous attempts at aspiration. In such

event the patient must have inevitably proceeded to a rapidly fatal issue. Among consumptives a decision relative to the expediency of exploratory puncture is in many cases extremely difficult. As a general rule, routine recourse to the use of the exploring needle as a means of diagnosis is productive of less satisfactory results than are obtained by reserving aspiration for those cases exhibiting positive surgical indications for its employment.

Methods of Treatment.-In former years it was the general dictum of the medical profession that the treatment of all cases of empyema should be that of surgical interference, the only difference of opinion relating to the choice of method. It may be of interest to quote brief extracts from a paper prepared by me thirteen years ago, and determine to what extent one can indorse the views then entertained.

"The important practical thought to be emphasized in this connection is the recognition of the existence of several species of bacteria in

Fig. 104. Showing site of operation in pulmonary abscess with recovery. (Compare with radiograph, Fig. 78, on p. 282.)

the exudate, endowed with varying properties and possessing marked differences in their virulence. The most benign of these characterize the empyemas of children and the metapneumonic pleurises of adults, and thereby furnish to the physician a justification for not resorting immediately, in all instances, to the more radical and mutilative measures of treatment. The therapeutic indications are conceded to be, first, prompt and thorough evacuation of the pus; second, prevention of reaccumulation by means of free and continuous drainage; third, the maintenance of asepsis; and, finally, the obliteration of the pus-secreting cavity through adequate provision for the expansion of the lung and the collapse of the chest-wall. Save in extreme cases a general tuberculous infection never contraindicates an operation from which satisfactory results are frequently obtained."

A single preliminary aspiration was advocated in children and in the metapneumonic pleurisies of adults. This was not based upon any

faith in the adequacy of aspiration to effect a cure, but rather with an aim to afford temporary relief, and at the same time to establish a definite diagnosis. The purulent nature of the effusion, particularly in adults, was thought almost invariably to demand subsequent operative measures. Free opening of the pleural cavity was strongly recommended on account of the complete exit offered to the coagula and organic débris, and the much improved facilities for a continuous discharge. Save for the employment of the single preliminary aspiration in children, this method was urged as an initial procedure in the treatment of all cases of empyema regardless of other qualifying conditions. It was insisted that if resection of rib was more frequently employed in the early stages of empyema before opportunity was afforded for the development of unfavorable conditions, there would result far less frequently the necessity for recourse to so severe a procedure as the multiple rib resection.

For several years the conclusions which were largely derived from the experience of others were conscientiously applied to appropriate cases of tuberculosis, with almost invariably unfortunate results. The essential principle of treatment was to perform pleurotomy, provided the general condition of the consumptive was not materially impaired, regardless of such vitally important considerations as fever, chills, sweats, and emaciation. If the condition of the patient in far-advanced phthisis was desperate, it was thought more merciful to permit him to die without inflicting the added torture of an operation. In the light of a considerable experience it has become apparent that the previous course was directly and radically wrong. Cases will be reported at length in order to illustrate the great responsibility assumed in advocating a radical operation for those comparatively well, and in withholding such surgical aid from others in urgent need, though apparently beyond hope. It is well to bear in mind that rib resection is necessarily followed by one of two conditions. There either takes place a considerable expansion of the previously compressed lung, which affords opportunity for renewed activity of the tuberculous process and rapid cavity formation, or there develops failure of the lung to expand, involving long-continued pus-formation and great danger of amyloid change. In the absence of such clinical indications as fever, sweats, and chills it seems exceedingly ill considered to precipitate the patient into the midst of such peril.

In 1895 a young man, a patient of Dr. F. C. Shattuck, consulted me immediately upon arrival in Colorado, exhibiting moderate tuberculous infection of the right lung. His illness had been of fourteen months' duration, the first symptoms relative to the pulmonary involvement beginning October, 1894. Cough and expectoration were moderate. There was slight daily elevation of temperature, with some acceleration of the pulse. Examination of the chest disclosed moist râles in the right lung from the apex to the third rib. After a period of several weeks a pleural effusion was recognized and thirty ounces of sterile serous exudate were withdrawn. Subsequently the fluid was removed many times at intervals of from three to six weeks. In the light of my present convictions this procedure was quite unwarranted. There had developed no increased elevation of temperature, no greater rapidity of pulse, or other constitutional disturbance suggesting its removal. Upon the other hand, the general condition