PLATE XI
BRONCHOPNEUMONIA.
(Age 51 ) Spreading or confluent consolidation. Necrotic, lobular consolidation in upper lobe. Accession 12623, Army Medical Museum. Colored photograph.
294
The consolidation in the early stages of the involvement of the bronchus showed the greatest width of the extension into the parenchyma near the hilus, gradually diminishing if only the direct trunk was involved. The more usual picture, however, at death was an involvement most dense near the hilus which formed
FIG. 84.- Bronchiole involved in an inflammatory reaction of the surrounding tissue, which extends to the submucosa. There is an early purulent bronchiolitis. Organisms which are morphologically pneumococci are present in the exudate in the alveoli surrounding this bronchiole. The inflammation in the alveoli is fibrinous in character. The reaction about the bronchiole is an interstitial type not distinguishable from that produced by the streptococcus. Accession number 1385, Army Medical Museum. Negative number 45277. Hematoxylin and eosin stain; X 130
the apex of a cone, the cone being the area supplied by the main bronchus and most or all of its branches. There was considerable irregularity in the character of consolidation of the areas surrounding different branches of the bronchus, yet in all, the older or more dense portions of the consolidation appeared nearer the larger trunks or visible bronchi.
295
As a whole the consolidations were less uniform, less dense, and more crepitant toward the periphery of the lung and the most marked congestions were found at the periphery of such spreading areas. In this locality also one was more apt to find hemorrhages, recognizable as such on gross examination. The lower lobes were somewhat more frequently affected than the upper and sometimes most of a lower lobe would be so involved. In such cases, however, there was usually a layer of air-containing tissue beneath the pleura.
As previously stated, this type of reaction was more often associated with pneumococci and when members of this group were responsible, the lesions were apt to show considerable fibrin and a relatively dry exudate. (Fig. 85.) Streptococcus viridans produced less fibrin, as was also true of meningococcus, an organism occasionally found in this type of lesion. It also appeared sometimes to be produced byStreptococcus hemolyticus, and in such cases the appearance was that of a spreading bronchopneumonia along all the bronchi near the hilus and extending out a relatively short distance into the parenchyma. This consolidation was hemorrhagic in character and, on section, so full of blood that it appeared like a relatively fresh clot. In such cases death probably occurred before extension to the peripheral portions of the lung took place. The most common lesions in the periphery, however, in the streptococcus cases were lobular consolidations which were usually less hemorrhagic than the central areas.
Microscopically, spreading bronchopneumonia presented the most varied picture, including practically every type of acute inflammatory reaction in the lung which has ever been described. Even in a single section representing a surface area of two or three square centimeters purulent, hemorrhagic, fibrinous, and serous reactions were not infrequently seen.
The usual picture in cases due to the pneumococcus in which fibrin formation was a factor were first, a purulent bronchitis and desquamation of the mucous membrane, the lumen being filled with pus and desquamated epithelial cells. The submucosa was edematous and infiltrated with leucocytes, most of them being of the polymorphonuclear variety. The hyaline basement layer varied in its appearance, being swollen early, disappearing during the later stages. The submucous glands of the larger bronchi varied greatly in the degree to which they were affected. In some instances they were the site of a fibrinous inflammatory reaction, particularly those nearest to the mucosa. As a whole, the changes in these structures were not marked, capillary vessels were prominent, and the reticulum stain revealed an apparent excess of them, though such excess was observed in the fulminant cases heretofore described, the inflammatory reaction of which, in many instances at least, preceded the bronchopneumonia now being considered. Pus and inflammatory edema infiltrated the bronchial wall and extended out into the alveoli accompanied by more or less fibrin formation. In the alveoli the vessels became engorged and a serous exudate was poured out, accompanied by a few cells, polymorphonuclears and lymphocytes. This exudate escaped in small quantities through the epithelium and later, raised the epithelium of the alveoli and caused its desquamation. Strands of fibrin then appeared in the exudate, the leucocytes gradually increased as did fibrin until the typical picture of the fibrinous reaction of croupous pneumonia was attained. Adjacent alveoli were involved as the process extended, yet it was rare to have even several adjacent microscopical fields show a uniform
296
picture, the portions distal to the bronchus usually presenting an earlier stage in the inflammatory process.
The variation in the amount of fibrin in the exudate was very striking. Tissue surrounding one bronchus might contain considerable quantities of it,
FIG. 85.- Postinfluenzal pneumonia. Pneumococcus, Type II, found in blood cultures, sputum, and cultures from the lung. Duration of pneumonic symptoms four days. The lesions are confluent lobular pneumonia, which is most advanced in the right lung, with necrosis near the hilum. There are a few peribronchiolar lesions in both lungs, not masked by the lobular consolidation. Acute hemorrhagic tracheitis and bronchitis was present, with hemorrhagic serofibrinous pleuritis. The amount of fluid in the pleural cavities was small. At necropsy, the lungs were moist and dripped blood with the exception of the lower portion of the right upper lobe which was drier. Foci of consolidation in the left lung were nodular in character and dusky red in color. The consolidations become paler and more distinct as the result of fixation. The peribronchial and mediastinal lymph nodes were enlarged, contained an increased amount of fluid, and showed necrotic areas about the periphery. Accession number 2551, Army Medical Museum. Negative number 30652.
and alveoli filled with fibrin surrounding this structure were contiguous with areas, the site of a purulent, serous or hemorrhagic reaction in which fibrin formation was scarcely visible. This may be explained by assuming either that the organism in one instance produced fibrin, in the other did not, or what
297
is distinctly more probable, that the two sets of tissue were affected by two distinct secondary invaders, the one affecting one bronchus, the other, a different one. It is true, however, that alveoli arising from the same terminal bronchiole not infrequently showed different types of inflammatory reaction and even though more than one organism was found in the exudate as examined in the sections, it seems unlikely that different organisms were the cause of these variations in response within so small a unit.
SPREADING BRONCHOPNEUMONIA (INTERSTITIAL)
In a certain number of the cases of this spreading type of bronchopneumonia the infection appeared to extend along the peribronchial and perivascular tissues as a phlegmonous inflammatory reaction following the lymphatic trunks. So far as can be determined from the material examined, there were always peripheral lobular lesions, sometimes of minor degree, which were drained by the peribronchial lymphatics of bronchi not affected by the inflammation of which the peripheral lesions were part. Lymphatics surrounding one bronchus receive tributaries from others and lymph flow is undoubtedly greater when, as was constant in these cases, the flow was retarded or blocked by inflammation in the lymphatics of bronchi primarily involved. This type of lesion was more frequently due toStreptococcus hemolyticus, butStreptococcus viridans, or at least organisms so identified, were occasionally the etiological factors, as were also Group IV pneumococci.
The reaction apparently occurred as a definite entity entirely apart from one almost indistinguishable from it produced, as has been described above, by direct extension from the neighboring bronchi. It was seen to greater or lesser degree about varying numbers of bronchi of all lungs in which the interstitial reaction formed a prominent part in the pulmonary inflammation. The tissues surrounding the bronchus and its surrounding areolar tissue were involved by the spreading of the inflammation to them by continuity, while at the same time the phlegmon involved the entire bronchial wall to and finally including the mucous membrane, producing first a sharp catarrhal reaction followed by desquamation of the cells and finally by a purulent bronchitis in which were large numbers of the organism which presumably caused the inflammation. (Figs. 86, 87, 88, 89 and 90.)
The reaction in the alveoli resembled that seen in the spreading pneumonias originating within the bronchus, except that as pneumococci were relatively infrequently the cause, fibrin production was less and frequently not apparent at all. Hemorrhage in smaller or larger amounts was usual and the bronchi and larger vessels were often surrounded by definite hemorrhagic zones, while hemorrhagic extravasations might consolidate a large portion of the central part of a lung or appear scattered here and there, usually relatively near larger bronchi where the lymphatic channels were larger and more tissue was present of the type apparently favoring this reaction, namely, the areolar tissue surrrounding bronchi and blood vessels.
Abscesses formed in the peribronchial tissues and not infrequently the termination of the purulent inflammation of the alveoli was a complete destruction of the alveolar walls and abscess formation, a reaction relatively common with
303
the streptococcus and rare with pneumococcus infection. This peribronchitis frequently was associated with pneumonic areas representing sections of lung supplied by relatively small bronchi or bronchioles and consisting of one or more anatomical lobules. These were usually relatively uniformly involved, the inflammation being at the same stage throughout. This stage was usually somewhat earlier than that reached by the consolidation in the central portion of the lung, and the general picture suggested strongly that these were pneumonia resulting from the inhalation of purulent material in bronchi, the terminal portion of which up to that time had been relatively slightly affected.
The phlegmon of the wall, in its later stages, was accompanied by considerable apparent increase in the number of blood vessels, fibroblasts increased, the wall of the bronchus was thickened by fibrous tissue, the contraction of which distorted the lumen. and in a certain proportion of the cases which died after many weeks, usually from empyema, rather extensive bronchiectases were found in relatively large bronchi. Not infrequently these communicated with abscesses in the parenchyma, the latter being of varying size from microscopical structures to large ones which had been recognized by their physical signs and X-ray appearances during life.
ACUTE BRONCHITIS AND PNEUMONITIS WITHOUT CONSOLIDATION
A few cases of an intense bronchitis were followed by an infiltration of the alveolar walls with leucocytes, slight inflammatory edema and very little exudate into the alveoli, death apparently occurring before such exudation took place. The lungs of such cases collapsed to a considerable degree on opening the chest, but were dark red in color and appeared more voluminous than normal lungs. On section, the bronchi showed a thin, serofibrinous exudate, the walls being dry, while the parenchyma was a somewhat dusky red, very slightly denser than normal, and showed an increased blood content but with little serum or froth escaping with the blood. Microscopically, a serofibrinous bronchitis was found with desquamation of the columnar cells. The ducts leading to the bronchial glands were usually plugged with serofibrinous exudate containing relatively few leucocytes and some blood. The submucosa showed leucocytic infiltration and edema and the glands usually showed a considerable degree of necrosis and fibrinous exudate. Small hemorrhages were present about the smaller vessels in the lower submucosa. Smaller bronchi did not show marked involvement of epithelium but the submucosa was infiltrated with leucocytes particularly about the capillaries. The alveolar walls showed engorgement of the vessels, with the escape of polymorphonuclear leucocytes into the walls, and occasional ones in the alveoli. In a few cases in which the tissues were well preserved pneumococcus-like forms occasionally were found in the alveolar walls and more rarely in the alveoli. They were seen occasionally also in the blood of the smaller vessels. In the cases in the Army Medical Museum, cultures of blood and lung showed pneumococcus Type II. Cases, apparently of this type, were described as having occurred in the Civil War and were then diagnosed simply as acute bronchitis. The most advanced process was in the bronchi and the distribution and character of the lesion suggest that the organisms found their way into the blood stream through the bronchial
304
veins and were distributed to the lungs by the pulmonary circulation. The organisms were present in the blood as well as in the lung, but the greatest intensity of the reaction appears to have occurred in the pulmonary tissues. The infection was apparently so overwhelming that death occurred before exudation could take place, and this would appear to be the logical result of a blood-stream dissemination. (Figs. 91, 92, 93, and 94.)
SECONDARY LOBULAR PNEUMONIA
This lesion was probably the most frequent type of secondary pneumonia though it was practically never the only type present. As stated previously, isolated anatomical lobules and groups of lobules frequently were involved in cases, the majority of the lesions in which were of other varieties. (See frontis-piece.) Usually it was present in lungs, the seat of extensive interstitial inflammation, though often affecting lobes and parts of lobes not markedly involved in the interstitial process. When it affected the greater part of a lobe, difficulty was experienced in distinguishing it microscopically from the earlier stages of lobar pneumonia, but microscopical examination showed differences in the character of the reaction in different and often adjacent lobules. (Fig. 95, 96, 97, 98, 99.)
Pneumococci and streptococci were the organisms most frequently found in the cultures of the lesions. The types of the pneumococci varied in the different camps but organisms of Group IV predominated in all. The incidence of hemolytic streptococci varied and tended to increase toward the latter part of the epidemic waves of acute respiratory disease, including the influenza pandemic of 1918. At some camps, however, these organisms were abundant early in the pandemic and their proportion in such places did not vary markedly during its course.
This type or distribution of consolidation varied markedly in the character of the exudative process. This was due to two factors, the bacterium causing it and the relative duration. The sequence of events in the progress of the anatomical involvement, however, appeared to be the same. The infection appears to have extended along air passages and involved with greater or lesser rapidity and completeness an entire anatomical lobular sector of the lung. In some cases the cellulitis and lymphangitis of the perilobular tissue and the interlobar septa were undoubtedly the lesions from which infection spread to the lobules between and about them, as the densest parts of the lobular consolidations often were nearest these structures, the central part of the lobule or that part farthest from the affected septa being the least involved. (Figs. 100 and 101.)
Every type of appearance, gross and microscopic, was seen, varying from lobule to lobule, and, except for those lobular consolidations of fibrinous character, usually due to pneumococci, the microscopical reactions varied in the same lobules. In the lesions due toStreptococcus hemolyticus there was no regular sequence of events.
In the lobular pneumonias, the upper respiratory tract varied in the pathological picture presented according to the organism responsible for the pathological conditions. Thus the most intense reactions in the upper respiratory tract were found in those cases in which culture from the nasopharynx and
305
FIG. 91.- Section of wall of the main bronchus, showing desquamation of epithelium and fibrinous and hemorrhagic inflammation of the submucosa. The ducts from submucosal glands are filled with a fibrinous exudate. Pneumococcus, Type II, isolated from the lung and blood. Accession number 16380, Army Medical Museum. Negative number 45581. Hematoxylin and eosin stain; X 25
316
sputum showed hemolytic streptococci. The changes in the trachea, in the cases due to pneumococci, showed little more than the reaction found in those acute, fulminant cases which have been described previously, and there was usually some epithelium left on the surface. In the streptococcus cases the columnar cells, for the most part, were thrown off, the submucosa infiltrated with leucocytes and the lymphatics distended with polymorphonuclear cells. Except for the depths of the crypts the mucous membrane appears to have lost both the columnar and basal cells. Hemorrhages in the submucosa about the capillaries in the deeper portions were usual, the glands of the submucosa showing hypersecretion and usually some degenerative changes. (Figs. 102 and 103.) In the bronchi the same type of inflammatory reaction was seen in the mucosa and in addition the inflammation extended out between the cartilaginous rings into the surrounding lymphatics. (Fig. 104.) In the smaller bronchi and bronchioles in these streptococcus infections a thin exudate was thrown out accompanied by desquamation of the columnar cells and a moderate degree of proliferation of the basal layer, though in all of the cases in which tissues were well fixed small Gram-negative bacteria were likewise present in these locations, thus making it possible that the proliferative changes were not due to the streptococcus. In the thin, fibrinopurulent exudate the organisms were not particularly numerous and usually were found in the lymph spaces beneath the hyaline basement membrane as well as mingled with the cells of the mucosa, and contained in the serous exudate which sometimes raised the mucus membrane from the underlying reticulum. (Figs. 105, 106, and 107.) In the smaller bronchioles in the midst of lobular consolidations it is difficult or impossible to tell whether the inflammatory reaction extended along the bronchioles and spread out from them or extended to them from adjacent inflamed alveoli.
In the alveoli of lobular consolidations apparently caused by pneumococci, the changes approached those seen in typical lobar pneumonia, and in microscopical sections, small areas may be indistinguishable from the microscopical picture in the lobar type of distribution. Quite frequently, however, there was much less uniformity even in a lobule than is found in an entire lobe of lobar pneumonia. Usually there was less fibrin formation and more variation in amount and kind of exudate even in adjacent fields of a microscopical section. In the pneumococcus cases, however, leucocytic infiltration of the alveolar walls was relatively rare but was usual in the streptococcus infections. (Figs. 108, 109, and 110.)
The infiltration of the alveolar walls in the streptococcus pneumonias varied markedly in the amount of inflammatory edema, cellular infiltration and hemorrhage present. Leucocytic infiltration of the alveolar walls appeared to be characteristic of the process and usually was accompanied in the early stages by a serous exudate in the alveoli. (Fig. 111.) In some instances the leucocytic exudate of the alveolar walls appears to have been superimposed upon the typical alveolar reaction seen in the fulminant cases and associated with small Gram-negative bacteria, and in such, if the tissue was well fixed, these bacteria usually were found. (Fig. 112.) The amount of hemorrhage in the alveolar walls and in the alveoli varied in amount and distribution, sometimes being abundant in the walls with little escape in the alveoli; at other times the hemorrhage appeared to be practically entirely within the alveoli.
317
(Fig. 113.) In cases of the most fulminant type the leucocytic exudate frequently showed a preponderance of mononuclear cells most of which were lymphocytes. (Figs. 114, 115, 116, 117, and 118.) In the more rapidly fatal streptococcus cases hemorrhage was usually abundant, and in many the patients did not survive a sufficient time to allow a marked cellular infiltration to occur.
FIG. 102.- Section of bronchial wall including group or mucous glands in a streptococcus bronchitis. A number of glands show advanced degenerated changes of the cells while the mucosa of the bronchus is completely desquamated. Accession number 16657, Army Medical Museum. Negative number 45226. Hematoxylin and eosin stain: X 1510
In most cases where considerable portions of the lung were affected by lobular pneumonia there was found a spreading bronchopneumonia of the central portions of the lung where it was impossible to make out lobular areas, and the presumption is that the primary reaction was that of bronchitis and peribronchitis. (Fig. 119.)
341
If the cases survived a sufficient time the massive consolidations tended to become more uniform, particularly when they reached the purulent stage. (Fig. 120.) These massive consolidations usually showed necrosis in the central portions or, if located near the periphery of the lung, the necrosis appeared to start in the subpleural lobules. The density decreased toward the periphery in the earlier stages but began to return toward the normal in parts nearer the hilus of the lung, presumably as the result of the resumption of circulation. (Figs. 121, 122, and 123.) The continued extension of the process, thus maintaining the block of the lymphatic circulation, played an important part in the production of the necrosis and solution of the tissues. The necrosis occurred relatively early in lungs affected by Streptococcus hemolyticus and the rupture of such areas into the pleura was frequently the cause of empyema. Occasionally these ruptured into the bronchi and a pulmonary abcess formed which in some cases, as indicated by clinical and roentgenoscopical evidence, were healed, leaving a definite scar. Microscopically there appeared in the early stages of this abeess formation dense collections of polymorphonuclear leucocytes in the relatively loosely consolidated lobules. (Fig. 124.) A reticulum stain shows that in these areas the reticulum of the alveolar walls has been destroyed. (Fig. 125.) This finding is quite different from that of lobar pneumonia where similar stains show that, in spite of the almost uniform consolidation, fibers of the reticulum are still present outlining the air spaces.
In cases in which individual, isolated lobules were involved in the necrotic process the solution of tissue appeared to start in areas of denser consolidation instead of uniformly involving the entire lobule. This is explained by the variations in the circulation and lymphatic drainage still remaining. (Fig. 126.) Sometimes considerable areas of the lung showed scattered rather than confluent areas of necrosis with abcess formation, but the most frequent picture was that of a massive consolidation with necrosis of its central portion or that portion of the consolidation nearest the pleura.
Where scattered lobules were involved the tissue between was usually atelectatic and if the patient survived a sufficient time, the tissue between the abscesses became organized and a dense fibrous matrix filled with abscesses resulted. This maintained a focus of infection for an indefinite length of time and as it usually communicated with an empyema cavity the whole mass occasionally was involved in an organizing process extending from the parietal pleura irregularly into the pulmonary parenchyma. (Fig. 127.) In this dense tissue, even after what appeared to be complete organization, one found in the microscopical sections collections of mononuclear leucocytes with occasional polymorphonuclears, and it usually was possible to find Gram-positive cocci by means of bacterial stains. Ordinarily granulation tissue projected into the abscess cavities from walls of a greater or lesser density, but the contraction of the fibrous tissue about the veins and smaller vessels was responsible for necrosis and breaking down of the granulations just as is true of other healing abscesses. (Fig. 128.)
345
INTERSTITIAL PULMONARY LYMPHANGITIS
Inflammatory reactions in interstitial tissues of the lung have been described from time to time, and indeed the term interstitial pneumonia has been used not infrequently, though more often as descriptive of chronic or proliferative types of reaction than of acute. It is doubtful if an acute interstitial inflammation ever occurs as a primary condition. In the material in the Army Medical Museum collections it was always associated with pneumonia, usually of the lobular variety. The designation in fact can be justified only in cases in which the interstitial reaction so preponderates among the factors leading to the death of the individual, that the pneumonic consolidation accompanying it appears insignificant.
There is extreme variation in the rapidity of the process in the amount of tissue involved and in the resistance of the host. The various stages found were dependent on the length of time the patient survived following the onset of the disease.
Following the onset of measles or influenza by relatively few days, the acute or fulminant hemolytic streptococcus infections lead to death in three or four days. Two types of such cases were found.
In one type there was an acute streptococcus infection of the upper respiratory passages, with intense involvement of the tonsils and nasal pharynx by an acute inflammation characterized by reddening, moderate swelling, and a pale, filmy exudate over the surface of the mucous membrane. It extended into the larynx where there was considerable swelling of the vocal cords; hoarseness was a prominent symptom, and not infrequently considerable respiratory difficulty resulted from the laryngeal edema. The acute inflammatory reaction extended down the trachea and into the bronchi, producing a seropurulent exudate. The inflammation then spread to the alveoli producing a diffuse hemorrhagic pneumonitis. This lesion followed both measles and influenza cases, and in the examination of some of these in the specimens at the Army Medical Museum, the purulent exudate in the bronchi was found to contain large numbers of Gram-negative bacteria as well as numerous streptococci, the latter organism decreasing in numbers toward the periphery of the lung. (Figs. 129, 130.)
The large amount of purulent material in the bronchi may not have been wholly due to the streptococcus. The purulent inflammation may have preceded it as a result of the action of either the measles or influenza virus, or of both. This possibility should be considered, because the reaction of the streptococcus elsewhere is not the production of large amounts of pus until later in the disease, the primary reaction being serous or hemorrhagic. (Figs. 131 and 132.) The frankly purulent exudate in the bronchi was associated with serous and hemorrhagic lesions in the lung and serofibrinous reactions in the pleura.
In the other type the organisms in the bronchi gained access to the media-astinal nodes causing a phlegmon in and about these nodes, extended directly to the pleura at the hilus and produced an intense pleuritis with rapid accumulation of fluid. This process was accompanied by an overwhelming toxemia and death resulted sometimes with no evidence of involvement of the parenchyma of the lung, though bronchitis always was found. The macroscopical appearance in such cases was striking. On opening the chest, large quantities
346
of fluid were found in the pleural cavities, frequently bilateral, which clinically had accumulated with great rapidity, so that during life aspirations had been necessary much more frequently than usual. The rapid accumulation of the fluid in the early hours or days of the disease, being so unusual, caused some cases to be considered pneumonia, and some of them came to necropsy without the presence of fluid having been ascertained. The lungs were found collapsed and dark in color, the mediastinal tissues were edematous, and the lymph nodes were enlarged, moist, and usually presented small hemorrhagic spots. On section, the compressed portions of the lungs were relatively dry and deep purplish red in color. Near the hilus the lung was moist, contained considerable fluid, and peribronchial thickening was usually evident. Here also there were frequently some hemorrhages and usually small areas of consolidation, though it was evident that the pleural and not the pulmonary involvement was responsible for the symptoms and death of the individual. (Fig. 133.)
In cases which were just as fulminant in character from the clinical standpoint, pleural involvement was slight or absent, but instead, the peribronchial and perivascular lymphatics containing areolar tissue were involved in a spreading phlegmon which extended to the interstitial tissues of both lungs. The bronchi always were involved in an inflammatory reaction, and it is probable that the interstitial reaction was secondary. Serial roentgenograms show the thickening beginning near the hilus and extending out into the periphery of the lung. This appearance is probably due more to the accumulation of exudate piling up more quickly in the numerous large lymphatics near the hilus draining from large peripheral areas while those in the periphery required a longer time to reach that degree of thickening necessary to produce shadows on the plate. The extension of the shadow outward, however, was exceedingly rapid, requiring a matter of hours only to produce definite lines reaching to the periphery.
Pleural symptoms frequently had their first appearance at about the time the roentgenograms showed that these lines had reached the vicinity of the pleura, though extension to this membrane from the vicinity of the hilus occurred at various stages of spread of the interstitial reaction in the lung. The pleura became involved and gave clinical signs before any lung was involved, as previously stated, and at any time during the spread of the process, or it might remain uninvolved until the interstitial raction reached the pleura by way of the interlobular septa. In some instances, generalized involvement did not occur until interlobar empyema ruptured and discharged into the cavity. Sometimes several localized empyemiata were found walled off by fibrinous exudate without general involvement of the entire pleura. In the case predominantly interstitial the lung was slightly heavier than normal. The pleural surface rarely was uninvolved; usually there were patches of fibrinous inflammation overlying consolidated areas. The entire surface sometimes showed a thin layer of fibrin adherent, the cavity being dry. Usually irregular consolidations could be felt near the pleural surface, but rarely no definite consolidation of nodular character was evident. On section, in the rapidly fatal cases not accompanied by marked pleural involvement, the lung presented a moist, dripping surface from which frothy, bloody serum flowed. (P1. I.) From the bronchi purulent secretion often streaked with blood could
352
be expressed. Larger and smaller areas of frank hemorrhage with clotting were to be found scattered throughout the lung but most evident and largest in the vicinity of the main bronchial trunks. Other than these there was, in the typical case, no definite consolidation, though the tissue seemed somewhat stiffer than normal and under a hand lens the walls of the pulmonary structures appeared unduly prominent.
Microscopically the changes in the parenchyma were very distinctive. Purulent inflammation in the bronchi was present and this pus contained varying numbers of streptococci, the numbers being distinctly greater in proportion in the exudate in the larger trunks, decreasing toward the periphery, and being practically absent in many terminal bronchioles, where in well fixed tissues Gram-negative bacteria were found. In the fulminant cases the degree of destruction of the bronchial mucosa was never very great. The most marked changes were in the peribronchial areolar tissue and consisted of a phlegmonous inflammatory reaction characterized by inflammatory edema, a deposit of relatively small amounts of fibrin and an infiltration of leucocytes, many being of the mononuclear type. Cells of the character of large lymphocytes were frequent but there were considerable numbers of somewhat larger mononuclear cells, most of which had slightly basophilic cytoplasm; a certain proportion appeared eosinophilic, though rarely were granular cells present in numbers except polymorphonuclear neutrophiles. The lymph vessels were dilated with serum containing leucocytes, some fibrin, and streptococci (figs. 134, 135, 136), apparent clots being formed, in many instances, which took a deep eosin stain, and while they were for the most part granular, hyaline areas occurred in them. Capillary blood vessels were engorged and some of them contained masses of cells which appeared laked (conglutination thrombi) or contained deep eosin staining material without evidence of morphological elements (hyaline thrombi). Whether or not these thrombi were present during the life of the individual is open to question. Appearances similar in every way are certainly more frequent in poorly fixed tissue, but they are noted more frequently in descriptions of the various types of pulmonary inflammations of the World War than in those of the respiratory affections ordinarily met with. Interstitial changes involved equally the areolar tissue around the bronchi and blood vessels, though they were more noticeable about the tissue surrounding the pulmonary artery and bronchus than about the vein; in fact, the structures about the vein were rarely markedly involved. (Figs. 137, 138.)
In addition to the involvement of the peribronchial and perivascular structures the phlegmon extended to those bands of connective tissue which pass into the lobe, finally separating it into anatomical lobules. (Figs. 139, 140.) These septa were involved, the distribution of the involvement varying markedly in different cases; they were most characteristic, as seen in section, when the involvement was of those interlobular septa just beneath the pleura, particularly in the latter stages when frank pus was formed.
There is a valid question as to whether the lymphangitis--for such is the character of the reaction in the peribronchial and perivascular tissues and interlobar septa--precedes the changes in the parenchyma or alveoli and terminal bronchioles or is the result of drainage from these structures. It is generally
PLATE XII
STREPTOCOCCUS LYMPHANGITIS.
Hemorrhagic perivascular and peribronchial lesions, Streptococcus hemolyticus. Thrombosis of vessel. Accession 3026. Army Medical Museum. H. & E. stain. Autochrome.
358
accepted that the bronchi and bronchioles are involved first, by the inflammation of the initial infection by the measles or influenza virus. The infection with the streptococcus spreads along the surface of the upper air passages just as does the infection seen in so many cases of clinical influenza. Indeed the wave of exudative inflammation characteristic ofStreptococcus hemolyticus involvement of the pharynx and larynx was observed as a succeeding one to that of the initial influenzal or measles process. That theStreptococcus hemolyticus extended to the mediastinum directly from the hilus lymph nodes seems indubitable, but it is difficult to explain the extension of the process outward along the peribronchial and perivascular lymphatics against the normal or usual flow in these vessels. The small size of streptococcus lesions of the hand which give rise to marked lymphangitis extending up the arm and the rapidity with which such lymphangitis extends is well known. It seems probable that the interstitial lymphangitis is secondary to peripheral involvement, at least of the bronchi rather than that it extends outward and involves the parenchyma secondarily. If it is viewed in this light, it appears erroneous or misleading to coin the words "interstitial pneumonia" for the process. The correct term is interstitial lymphangitis and cellulitis or phlegnion secondary to bronchitis, bronchiolitis and lobular pneumonia of streptococcus origin; in other words it is secondary to the spread of this organism along the air passages. Infections by this organism in the early days of the preceding measles or influenza, sometimes produced extremely extensive pulmonary involvement which, however, might well be the result of its combination with the other virus, no division being possible between the lesions produced by the one and the other.
In cases that lived a longer time the changes in the interstitial tissues were recognizable on gross examination. The peribronchial and perivascular tissues became paler as the result of increase in the leucocytic infiltration and were more distended. The same is true of the septal tissues, so that there formed a mosaic-like pattern dividing the anatomical lobules. At first reddish, these became paler and the lymph vessels became visible as tortuous beaded strix distended with fluid or clotted purulent exudate. (Pl. II; figs. 141, 142, 143, 144, 145.)
The purulent inflammation extended out into the parenchyma and abscesses were formed in the tissue. Extension to the pleura caused generalized or local enipyemia. Often extensions to the interlobar pleura caused interlobar pus pockets which, as the result of fibrinous adhesions at the junction of visceral anti parietal pleura, were retained in the interlobar clefts and were not evacuated when the pleura was drained by operative procedures.
Some of these interlobar pus pockets destroyed the adjacent parenchyma, forming abscesses extending into the lung. (Fig. 146.) The thickening along the bronchi and blood vessels made these structures appear as thick-walled tubes which, on gross examination, were differentiated one from the other with difficulty, unless section revealed the contents. On cross section they were even more difficult to differentiate. The cut ends protruded above the contracted parenchyma. (Figs. 147, 148.) Early there was a surrounding zone of hemorrhage but later the exudate became purulent so that the thickened walls were pale and opaque resembling cross sections of old water pipes encrusted
PLATE XIII
ZENKER'S DEGENERATION OCCURRING IN INFLUENZA.
Rectus abdominis muscle. Accession 6951, Army Medical Museum. Colored photograph
PLATE XIV
ZENKER’S DEGENERATION OCCURRING IN INFLUENZA.
Accession 6951, Army Medical Museum. Weigert hematoxylin picro-eosin stain. Autochrome.
368
with calcareous deposit. Abscesses also formed along the bronchial tree and occasionally ruptured into the bronchi. Some abscesses destroyed sections of the bronchial wall and thus communicated by large openings into the bronchial lumen, producing in effect irregular or saccular bronchiectases. In other cases the phlegmonous inflammation, by weakening the entire circumference of a section of bronchial wall through destruction of its tissues, aided in the production of more even dilatations, the more common variety of bronchiectasis.
Abscesses in the periphery of the lung caused pleural involvement and empyema by extension, sometimes as the result of gross rupture. Abscesses nearer the hilus extended and sometimes produced large pulmonary abscesses before rupture into a bronchus resulted in drainage and relief of tension. The inflammation occasionally extended to the walls of vessels usually of pulmonary veins causing thrombosis which rarely completely blocked the larger vessels. Bacteriemia from which streptococci were cultured was found in cases where this was observed. (Figs. 149, 150.)
The abscesses compressed bronchi and produced atelectatic areas distally to the point of compression, where organizing processes quickly supervened. These atelectatic areas were sometimes the seat of inflammation at the time of the shutting off of the air and sometimes relatively unaffected. In the former case the exudate became organized and in the latter the epithelium of the alveoli became columnar and often piled up in masses of cells of squamous type imitating foci of epithelioma. In some of the cases of interstitial lymphangitis of considerable duration, organization of the exudate within the septa produced a very striking picture. The cause of death in these cases was usually empyema, the lung having recovered to a greater or lesser degree. It is probable that in many of the recovered cases patches of this organizing process were responsible for some of the opaque areas shown in the roentgenograms. The cross sections of such lungs show varying amounts of increased thickness of the walls of the bronchi and of the perivascular tissues. The most striking lesion is the thickening of the interlobular septa which stand out as pale areas surroundinig the irregular lobular divisions of the lung. If the process involved a larger part of the lobules or a whole lobe, abscesses always were found in the interstitial tissue and also in the parenchyma, and practically constantly areas of recent lobular pneumonia representing a recent extension of the process were evident. (Figs. 151 and 152.) In some instances considerable organization was seen relatively soon after the apparent onset of pneumonia, and it was impossible to state the exact age of the processes on the macroscopic appearance, histological examination being necessary to demonstrate the relative amounts of fibrous tissue and phlegmonous infiltration present.
Histological examination showed a gradual increase of the reticulum throughout the interlobular septa. This reticulum was gradually replaced by collagenous fibers and the relatively large number of vessels of capillary size present in the early stages were replaced by fewer vessels of larger caliber. The varying stages of the process could frequently be studied by taking multiple sections from the same lung. (Figs. 153, 154, 155.)
Streptococcus lymphangitis, in addition to forming a prominent part in some cases, was seen in all types of pneumonia in camps where Streptococcus
377
hemolyticus was prevalent. In spreading bronchopneunionia due to other organisms interstitial inflammation was seen affecting other parts of the same lung. Often it was found along the bronchi serving the posterior vertebral portion of the lower lobe, more rarely the upper lobe. The mediastinal tissues were infected by streptococci with subsequent empyema in cases the pull- monary lesions of which were apparently due to other organisms, principally pneumococci.
In empyema with lobar pneumonia the streptococcus was found in the exudate about as frequently as was the pneumococcus. In such cases it was not always possible to trace its path but sometimes the presence of a mediastinitis or interstitial lymphangitis in other lobes revealed its route.
STAPHYLOCOCCUS AUREUS PNEUMONIAa
Staphylococci were frequently found in cultures from the lung of pneumonia following influenza, but for the most part they were not believed to produce the pneumonic lesions. Pneumonia due to staphylococcus aureus was described at Camp Jackson and was occasionally reported from other camps. The lesions due to a septicemia or pyemia in which this organism was the etiologic agent were reported occasionally, but these can scarcely be considered pneumonia. The number of such cases as compared with the total cases of pneumonia was relatively few, but at Camp Jackson the clinical type was considered to be a distinct one. It is doubtful, however, if one could with surety diagnose Staphylococcus aureus pneumonia during life.
In the typical cases, the patients were extremely prostrated almost from the onset of their infection. After being ill from three to four days with influenza, their condition became critical. They exhibited an unusual type of cyanosis. The cherry-red, indigo-blue color was indeed very striking, though not pathognomonic of the infection.
The onset was rarely accompanied by the chill and localized pain of a typical lobar pneumonia though the course of the disease was extremely rapid. The facies, the anxious expression, and deep cyanosis suggested a grave prognosis from the onset, at a period when physical signs of pulmonary involvement were but scanty. Epistaxis was common. The fever on the whole was high, ranging between 104º and 106º F., with frequent remissions to 101º F. The pulse rate was usually low but poor, and in many cases almost imperceptible. Respirations usually ranged between 24 and 36 and in some cases rose to 50 or 60 without much obvious discomfort. These patients rarely had the painful and labored breathing seen in pneumococcus infections.
Of the patients studied at Camp Jackson, 12 died between the first and fifth day, and 73 between the sixth and tenth day of the disease.
The most characteristic feature of this type of pneumonia is the sputum. When typical, the sputum is friable, purulent material of a dirty salmon-pink resembling anchovy sauce or the contents of an overripe furuncle. Occasionally the sputum is hemorrhagic, at first suggesting an acute pulmonary hemorrhage;
a The following statements are based on a study by H. T. Chickering and James H. Park: Staphylococcus Aureus Pneumonia.Journal American Medical Association, 1919, lxxii, 617, and protocols and specimens in the collections of the Army Medical Museum.
378
but on close inspection its purulent character can be distinguished. Smears of such sputum stained according to Gram's method, showed many grouped Gram-positive cocci and pus cells. When such sputum was obtained the prognosis was grave; typical sputum, however, was not always procured; and frequently cases that ultimately resulted fatally produced only greenish yellow purulent material which on culture yielded the staphylococcus as the predominant organism.
If sputum containing the staphylococcus was streaked on glucose-free blood-agar plates, the colonies appeared as opaque, round shiny disks about 1 mm. in diameter, surrounded by a wide zone of hemolysis after 18 hours' incubation. On removal from the incubator, the colonies sometimes appeared white; but if the plates were allowed to remain a few hours in the daylight or sunlight, the yellow pigment appeared.
In the routine examination of hundreds of specimens of sputum,Staphylococcus aureus was encountered frequently from the throats of normal persons and from patients having a definite pneumococcus infection, but not usually as the predominating organism. It is impossible to say what proportion of men having pneumonic lesions and showing Staphylococcus aureus in the sputum were harboring the organisms in the alveoli of the lungs. A considerable number of individuals presenting these findings recovered, but it is doubtful that many of them actually had a staphylococcus infection of the lungs when one considers the pathology of the disease. In many cases, in addition to the predominatingStaphylococcus aureus in the sputum, there were recovered various types of pneumococcus, B. influenzae, streptococcus, and other organisms. The presence of these bacteria in association with the staphylococcus in the lung was confirmed in many instances in the post-mortem examinations. Bacteriemia was an unfavorable prognostic sign.
Few cases showed clinical evidence of accumulation of fluid in the chest or pericardium. When present, the fluid was slightly cloudy-amber in color, sometimes with a tinge of blood, when discovered early in the disease. Later, frank greenish-yellow pus sometimes was obtained. Direct smears of the fluid stained by Gram's method showed many pus cells and grouped Gram-positive cocci. Many of the bacteria were phagocyted. Cultures of the fluid on plain or blood agar showed the typical colonies of Staphylococcus aureus. The culture plates were incubated for 72 hours before discarding; otherwise the small, colorless, dewdrop-like colonies of B. influenzae, which are prone to develop about the large staphylococcus colony, might have escaped notice after 18 hours' incubation, when they were pin-point in size and could be seen only with a hand lens.
If, in exploring the chest, no fluid is obtained, the needle penetrating solid or semiconsolidated lung tissue, the needle may be washed out by aspirating a little plain bouillon broth culture medium into the syringe and cultures made with the blood-tinged fluid.Staphylococcus aureus was frequently recovered bv this method.
Leucocyte determinations were made in a considerable number of cases ofStaphylococcus aureus pneumonia. Both the patients dying of pureStaphylococcus aureus infection and those having a mixed infection with B. influenzae,
379
pneumococcus or hemolytic streptococcus, in many instances showed leucopenia. The majority were about normal; a few showed moderate leucocytosis. Many of those having an increased number of white blood corpuscles had localized complications, as empyema or pericarditis. Brief mention was made above of the lack of definite localized shadows in the lungs on fluoroscopic examination early in the disease. In many instances the diffuse mottling of the whole lung suggested miliary tuberculosis. In other instances, small multiple areas of consolidation were demonstrated. Occasionally whole lobes were involved, there being present a confluent bronchopneumonia or a lobar type of lesion caused by pneumococci with a Staphylococcus aureus infection superimposed.
Complications were rare, due largely to the high immediate mortality. Only three patients had discoverable empyema. Other complications were few in number. Seven patients had acute otitis media. One patient had a staphylococcus meningitis. Nine patients had signs of meningeal irritation, but examination of the spinal fluids proved them to be normal. One patient developedStaphylococcus aureus pericarditis. The pericardium, both parietal and visceral layers, were covered with a thick, shaggy coat of fibrin. A small amount of purulent fluid was found in the pericardial sac.
Subcutaneous emphysema, a complication rarely seen in primary lobar pneumonia, was encountered six times in the total series of about 1,400 cases of pneumonia. In two of these casesStaphylococcus aureus was isolated from the lung. The emphysema was very extensive, involving the head, neck, and trunk. In one case the eyelids were completely closed. It apparently em- barrassed the respiration but little.
One patient had a complicating suppurative parotitis from which the staphylococcus was cultivated. Several patients had multiple furuncles.
At the necropsy examinations made during this epidemic in 14 cases it was possible to cultivateStaphylococcus aureus directly from the lung tissue, either as the sole organism or in association with other microorganisms.
From this material an unusual opportunity was presented to study the lesions associated with the presence ofStaphylococcus aureus in the lung tissue. The lungs on gross inspection were not usually so voluminous as those seen in the hemolytic streptococcus infections, nor did they have the same shotty feeling on palpation. Usually the dependent portions of the lungs, the lower lobes, or very frequently the posterior portions of all the lobes, being the dependent parts with the body in the horizontal position, were involved.
Sometimes there was a pleural exudate. A fibrinous exudate was much less commonly found than in lobar pneumonia. The surface of the affected portions of the lung was deep purplish-blue. Careful examination of the pleura usually revealed, in addition to showers of petechiae, small yellowish-white spots, pinhead in size, situated just beneath the pleura, which proved to be minute abscesses. The cut surface of the lung was either intensely hemorrhagic, resembling the cut surface of the spleen if the course of disease had been of brief duration, or had the appearance of a confluent bronchopneumonia. However, small bronchioles standing up above the cut surface and surrounded by a zone of dark red infiltrated tissue, as in the bronchopneumonias following
380
measles, were not seen. On the contrary, innumerable small abscesses, ranging in size from 1 to 10 mm. in diameter, were usual. In early processes, these minute white spots were not broken down. In those of longer duration, many small abscesses became confluent, having softened and become filled with thick, greenish-yellow pus. Nine of the 14 cases examined post mortem revealed these small abscesses. Of the cases not showing abscess formation, three were very early, the patients dying on the fourth, sixth and seventh days from the onset of the primary epidemic infection. The sections of these cases showed intense congestion, with rupture of the alveolar walls and exudation of serum and red blood corpuscles into the alveoli. Another case, in which the patient died on the eleventh day without abscess formation, was complicated by sup- purative pericarditis. In the last case there was a mixed infection with B. influenzae, in which process Staphylococcus aureus apparently played a minor role. Direct smears of the abscesses showed pus cells and many grouped Gram-positive cocci. The majority of the abscesses were situated at the periphery of the lung near the pleura, though the whole lung was sometimes involved. The propensity of Staphylococcus aureus to form multiple small abscesses in the lung appeared to be characteristic of this organism. The gross appearances suggest that the abscesses resulted from a pyemic process, but microscopical examination revealed the fact that the abscesses had originated within the bronchial lumen, the resulting lesions resembling those of inhalation pneumonia. Extensions along peripheral lymphatics to the pleura gave rise to the minute foci seen beneath the surface of this membrane.
The inflammation extended along the bronchi, producing a purulent bronchitis with destruction of the mucosa and rapid extension through the walls of the bronchi and bronchioles, destroying these walls and extending into the surrounding alveoli. It thus formed abscesses, the center of which was the bronchus and the periphery those surrounding alveoli not dissolved by the lytic action. (Figs. 156, 157, 158, 159, 160, 161.) The alveoli surrounding these bronchial abscesses were involved in an inflammation of greater or lesser extent. Where the abscesses were surrounded by a relatively narrow zone of inflammation there was an exudate of serum, relatively few leucocytes and few bacteria, essentially a reaction surrounding a focal abscess in the lung and dependent more on the reaction secondary to the toxins of the abscess than on extension of the bacteria into the tissues.
In other cases the inflammation extended outward from the bronchus for considerable distances, apparently following along the air passages, causing a seropurulent reaction over wide areas of the pulmonary parenchyma. This reaction was a spreading bronchopneumonia, but was characterized by the occurrence of the focal abscesses about the bronchi which supplied or traversed the consolidated area. The organism, true to its characteristics, produced lysis of the tissue and abscess formation with a minimum of phlegmonous reaction about the lesions, but in addition spread along air passages and sometimes caused widespread pneumonic consolidation. Staphylococci were relatively abundant in the pus of the abscess but were few in the alveolar walls and exudate of the pneumonic areas. Lymphangitis, so prominent a part of the picture in streptococcus lesions, was not noticeable.
387
A post-mortem examination of one of the two cases presenting signs of subcutaneous emphysema was made. It was impossible to discover the pathway of the escape of air from the lungs. The mediastinal tissue was filled with bubbles of air. Along the lappet of the inferior edge of the right upper lobe, the visceral pleura was distended with large bubbles of air varying in size front 1 to 10 mm. in diameter.
On account of the crowded condition of the morgue at Camp Jackson it became impossible to hold many post-mortem examinations. Consequently an attempt was made to learn more of the bacterial infection of the lungs of the fatal cases by exploring the lungs post mortem with an ordinary chest exploring needle and syringe. The syringes and needles were sterilized by boiling and the skin was sterilized by the application of tincture of iodin. The needle was thrust into the lung tissue and aspiration continued until about 1 c.c. of bloody fluid was obtained. This fluid was then cultivated on glucose-free blood-agar plates and slants and in plain broth, and in many cases direct smears of the lung juice were made. These smears checked closely with the cultures. The organisms recovered in this way were found as well in the sputa, blood cultures, and empyema fluids.
The occurrence ofStaphylococcus aureus in the lungs in the fatal cases of pneumonia at Camp Jackson was striking; 49 percent of the 312 cases cultivated showing this organism present alone, 92 cases; in association with the influenza bacillus, 22 cases; with the pneumococcus, 29 cases; with various streptococci, 12 cases. In many instances, death was predicted from the appearance of the sputum alone, and the post-mortem cultures verified the information obtained from the sputum examination.
The lesions were somewhat similar to some of those produced by streptococcus in other camps, though the breaking down into abscesses appeared to occur with greater rapidity. It differed from the abscesses due to streptococci in showing little or no evidence of lymphangitis in the lymphatics draining the affected area. It was associated usually with other organisms which undoubtedly were responsible in part for the pulmonary lesions.
REFERENCES
(1) MacCullum, W. G.: Pneumonia in Army Camps. Monograph of the Rockefeller Institute for Medical Research, No. 10, April 16, 1919.
(2) Miller, William Snow: Studies on the Normal and Pathological Histology of the Lung. The American Review of Tuberculosis, 1925, Vol. XI, 1.
