The following list shows the bacteriology of a small group of autopsies in which the sputum was examined after onset of pneumonia:

In one instance of streptococcus pneumonia (Autopsy 274) infection with streptococci occurred subsequent to the examination of the sputum made five days before death; pneumococcus was found in the washed sputum.

With lobar pneumonia there was evidence that superimposed infection occurred more frequently during the stage of red than of gray hepatization. With the lobular consolidation of bronchopneumonia this relation has not been found. Among 27 instances of red lobular consolidation, hemolytic streptococcus has occurred 6 times, namely in 22.2 per cent; among 26 instances of red and gray consolidation, 8 times, namely, in 30.7 per cent; among 13 instances of gray consolidation, 5 times, namely, in 38.5 per cent. Infection with hemolytic streptococci is more frequent when the lesion has persisted to the stage of gray hepatization. This difference between lobar and bronchopneumonia is probably dependent in part at least upon the more severe and persistent lesions of the bronchi with bronchopneumonia.

The histology of consolidation which is definitely limited to secondary lobules or groups of lobules varies considerably. When death occurs in the early stage of the lesion, consolidated patches are deep red and somewhat edematous, so that bloody serous fluid escapes from the cut surface of the lung and red blood corpuscles are present in the alveoli in great abundance together with polynuclear leucocytes, fibrin and serum in varying quantity. It is not uncommon to find evidence that the lesion has had its origin in the bronchioles and extended from them to other parts of the lobule. Polynuclear leucocytes may be relatively abundant within and immediately about the bronchioles and alveolar ducts, whereas the intervening alveoli and infundibula are filled with red blood corpuscles among which are polynuclear leucocytes and perhaps some fibrin. It may be evident that bronchiolar pneumonia with hemorrhage into intervening alveoli is in process of transformation into a more diffuse leucocytic pneumonia, for polynuclear leucocytes are making their way from the alveolar wall into the blood-filled lumen and, as the result of the presence of blood, remain for a time close to the lining of the alveolus.

When the consolidated lobules have assumed a gray or reddish gray color, polynuclear leucocytes are more abundant and often almost homogeneously pack every alveolus within the boundaries of the lobule. In some instances there is fibrin partially obscured by the presence of leucocytes in great number.

Although fibrin is less abundant with bronchopneumonia than with lobar pneumonia, nevertheless in a considerable proportion of instances it is a very conspicuous element of the inflammatory exudate within the bronchioles, alveolar ducts and alveoli. It is unusual to find the alveolar ducts and alveoli uniformly plugged with fibrin containing leucocytes; there is a variegated distribution of exudate which has little resemblance to that of lobar pneumonia. Occasionally (Autopsies 242 and 247) polynuclear leucocytes fill the bronchioles, alveolar ducts and infundibula, whereas the surrounding tributary alveoli contain fibrin and polynuclear leucocytes in moderate number; red blood corpuscles may be present in sufficient number to give a homogeneously red color to the lobular consolidation.

In association with lobular pneumonia, fibrin within the lung tissue undergoes certain changes which outline very sharply the alveolar ducts and the other structures usually ill defined in preparations of the lung. A remarkable appearance is produced by the deposit of hyalin fibrin upon the surface of the alveolar ducts and infundibula. This lesion has been described by LeCount.

Within the alveolar tissue of the lung, spaces are seen lined by a layer of fibrin which stains homogeneously and very brightly with eosin. They are recognized as alveolar ducts by the presence of scattered bundles of smooth muscle in their wall. The layer of hyaline fibrin overlying the surface of the alveolar duct usually forms a continuous lining and covers over the orifices of the alveoli which surround the alveolar duct. These ducts are rendered still more conspicuous by the character of their contents which exhibits a sharp contrast with that of the surrounding alveoli. The alveoli duct occasionally contains a bubble of air, but more frequently it is filled with serum in which red blood corpuscles are sometimes numerous. There is within the lumen scant fibrin and very few cells, among which polynuclear leucocytes are predominant. In the surrounding alveoli on the contrary leucocytes and fibrin are abundant. A similar change is found in the infundibula very clearly defined by their conical form, which is especially well outlined below the pleura or in contact with interlobular septa. The infundibulum is outlined by hyaline fibrin which passes over the orifices of the tributary alveoli and separates the serous contents of the infundibulum from the cellular fibrinous contents of the alveoli about.

The lesion which has been described is often associated with acute bronchitis and bronchiolitis, and the alveoli immediately about the respiratory bronchioles may be filled with polynuclear leucocytes. It is very common to find large bubbles of air sharply defined within the purulent contents of the bronchiole. In some lobules the alveolar ducts, infundibula and alveoli intervening between these foci of leucocytic pneumonia are almost uniformly filled with fibrin and polynuclear leucocytes, but in other places the formation of complete layers of hyaline fibrin is in process. Bubbles of air are often seen within the alveolar ducts, and about them is an irregular layer of fibrin formed by the penetration of air into a channel previously filled with a loose network of fibrin containing serum in its meshes. The fibrin compressed against the walls of alveolar duct and infundibulum remains as a compact layer separating these structures from the alveoli which project from their walls. The bubble of air is doubtless later absorbed and replaced by serum, so that many alveolar ducts are filled with serum almost wholly free from cells, whereas alveoli outside the fibrinous membrane contain a network of fibrin with leucocytes in greater or less abundance.

In association with this fibrinous pneumonia, which has been described, hyaline thrombosis of the capillaries is not uncommon. This hyalin material within the capillaries gives reactions of fibrin, and in sections stained by the Gram-Weigert method for demonstration of fibrin, these thrombosed vessels have the appearance of capillaries irregularly injected with a blue material.

The interstitial tissue surrounding consolidated lobules is often edematous; the lymphatics are distended with serum and contain a moderate number of lymphocytes and polynuclear leucocytes.

Among the lungs which have been studied histologically, pneumococcus has been almost invariably associated with the lobular lesions which have just been described, whether hemorrhagic, leucocytic or fibrinous; the histologic changes accompanying infection of the lung with streptococcus will be described later. Pneumococcus has been cultivated from the consolidated lung and is found in section of the lung. B. influenzæ is found in cultures made from the bronchi. Table XXXVIII includes those instances in which the histology of the consolidated lung accords with the description given above.

Pneumococcus was found in all but 2 instances, and in one of these (Autopsy 336) the only culture was from the heart’s blood and in the other (Autopsy 498) cultures were unsatisfactory because proper media were not obtainable. Pneumococci of Types I, II, II atypical, III and IV are represented in the list. B. influenzæ has been found in a considerable number of instances in which cultures have been made from the lung and in every instance in which cultures have been made from the bronchi. Staphylococci are often found in the bronchi, but in most instances they do not penetrate into the lung.

Another group of cases of lobular pneumonia are important because in association with necrosis of lung tissue recognized by the microscope hemolytic streptococci have been found in the lungs. In such instances serum is abundant and polynuclear leucocytes are relatively scant though their distribution varies considerably; in some places leucocytes are fairly abundant though elsewhere almost absent, but this distribution bears no obvious relation to the bronchioles. In some instances (Autopsies 274 and 487) red blood corpuscles are numerous but in others (Autopsies 275 and 312) they are inconspicuous. The characteristic feature of the lesion is the occurrence of patches of necrosis within which the nuclei both of exudate and of alveolar walls have partially or completely disappeared. In these areas of necrosis short chains of streptococci are found in immense number whereas in living tissue they are present in moderate number. There has been a relatively inactive inflammatory reaction, great proliferation of streptococci and necrosis of invaded tissue. The bacteriology of instances of lobular pneumonia with necrosis is shown in Table XXXIX.

Lobular pneumonia, in some of these instances at least, has been caused primarily by pneumococci; necrosis has been the result of secondary invasion by streptococci. In Autopsy 275 Pneumococcus IV has been obtained from the blood, but in the presence of streptococci has presumably disappeared from the lung and bronchus. In the case represented by Autopsy 274, Pneumococcus IV has been found in the sputum five days before death at the onset of pneumonia, but at this time no hemolytic streptococci have been found. In the case represented by Autopsy 312, Pneumococcus IV, B. influenzæ and a few colonies of hemolytic streptococci have been obtained from the sputum two days after recognition of pneumonia and five days before death.

The hemorrhagic and edematous consolidation of the early pulmonary lesions of influenzal pneumonia is their most distinctive feature. Red confluent lobular pneumonia is frequently found in those who have died within the first week following the onset of influenza. The lungs are voluminous and heavy and may weigh as much as 1,500 grams; the pleura which overlies the consolidated area is blue or plum colored and usually shows scant if any evidence of pleurisy. Scattered patches of consolidation are accurately limited to lobules, but in addition there are large areas often involving the greater part of the lobes and not infrequently situated in the lowermost part of the lower lobes. This confluent consolidation may be obviously limited by lobule boundaries. The consolidated tissue is deep red and laxly consolidated; red serous fluid escapes from the cut surface. The lesion not infrequently occurs in association with hemorrhagic peribronchiolar pneumonia.

The histology of this confluent lesion has been studied in Autopsies 242, 244, 303, 336, 464, 474 and 506. The histology varies, because, in some instances, leucocytes, in other instances, fibrin, is abundant, but the presence of red blood corpuscles in large number within the alveoli gives a red color to the consolidated tissue. In these cases pneumococci, associated in the lungs or in the bronchi with B. influenzæ, have been the cause of pneumonia. In two autopsies studied histologically (Autopsies 274 and 478) there was red lobular and confluent pneumonia and the blood and lungs contain hemolytic streptococci demonstrated by cultures; microscopic examination showed the presence of a widespread necrosis of the lung tissue.

In the group of autopsies in Table XL there was red confluent lobular pneumonia. These autopsies are separated from those just cited because there was no histologic examination of the tissue.

This group of autopsies confirms the view that the red confluent lobular pneumonia is caused by pneumococci in association with B. influenzæ. Hemolytic streptococci may invade secondarily. In Autopsy 297 a few hemolytic streptococci were found in the bronchus but apparently had not entered the lungs. In the absence of histologic examination it is not possible to determine if the invasion of hemolytic streptococcus (in Autopsy 364) has caused necrosis of the pneumonic tissue.

Peribronchial Hemorrhage and Pneumonia

In a considerable number of instances, namely, in 19 autopsies, hemorrhage about the small bronchi has been recognizable upon gross examination of the lung. A conspicuous zone of hemorrhage 2 or 3 mm. in thickness surrounds small (with no cartilage) often dilated bronchi and on longitudinal section may be tracted for a considerable distance along the bronchus (Fig. 7). In many additional instances peribronchial hemorrhage has been found by microscopic examination. In some instances the peribronchial zone of hemorrhage is firmer than the tissue elsewhere and it is occasionally difficult to determine whether the lesion is hemorrhage or pneumonia. In 7 instances frank red consolidation of peribronchial tissue was recognized at autopsy; this lesion will be considered later under peribronchial pneumonia. Hemorrhage about bronchi, like other evidences of severe injury to bronchi following influenza, is more frequently found in the lowermost parts of the lungs than elsewhere. It is invariably associated with severe bronchitis; the bronchi have contained purulent fluid in 15 of 19 instances of peribronchial hemorrhage and in 10 instances the lesion has been associated with dilatation of the bronchi.

Microscopic examination furnishes further evidence of the severity of the bronchial changes which have brought about hemorrhage into the surrounding alveoli. The lumen of the bronchus contains blood and leucocytes; the epithelium is sometimes raised in places from the underlying basement membrane by blood; blood vessels of the bronchial wall are engorged, and there is hemorrhage into the tissue of the bronchus. More frequently the bronchial epithelium is completely lost and the denuded surface is often covered by a layer of fibrin intimately adherent to the inflamed mucosa. Transitions between simple hemorrhage and pneumonia are found, polynuclear leucocytes being mingled with red blood corpuscles. In several instances the alveoli in immediate contact with the bronchial wall have contained fibrin, whereas those in the surrounding zone have contained blood.

Bacteria found in the bronchi in 10 instances of peribronchial hemorrhage have been as follows:

The high incidence of B. influenzæ and the frequent association of B. influenzæ and hemolytic streptococci are noteworthy. The instance in which no organisms were found is probably due to a defect in media and should perhaps be excluded from the list.

The percentage incidence of pneumococci, hemolytic streptococci, staphylococci and B. influenzæ in the bronchus, lungs and blood of the heart is an index of the facility with which these microorganisms penetrate internal organs when the bronchi are the site of this hemorrhagic lesion.

When these figures are compared with those for all forms of bronchitis no very noteworthy differences are found; the incidence of pneumococci here is less and that of hemolytic streptococci greater. In association with the severe changes present in the bronchi, hemolytic streptococci which enter the lungs almost invariably find their way into the blood.

In 6 instances there has been frank pneumonic consolidation limited to a zone encircling small and medium-sized bronchi which have often been obviously dilated. On cross section these patches of pneumonia are circular, from 1 to 2 cm. in diameter and each contains a bronchus at its center. When the bronchus is cut longitudinally it is evident that pneumonic consolidation forms a cylindrical sheath about the tube. The consolidation varies in color from red to grayish red. In one instance (Autopsy 253) the consolidated tissue has formed a gray zone in contact with the bronchus and is red in a peripheral zone; microscopic examination shows that the alveoli about the bronchus contain fibrin, whereas those at a greater distance contain red blood corpuscles. In this instance, the associated pneumonia in another part of the lung has been somewhat anomalous and has had characters both of lobar and bronchopneumonia, for scattered in the left lung there have been patches of firm consolidation not more than 2 cm. across. The smaller of these patches are deep red, but the larger are coarsely granular and gray in the center. The patchy character of the lesion has suggested bronchopneumonia, but the coarse granulation on section and the presence of fibrinous plugs within the small bronchi have presented a close resemblance to lobar pneumonia. This autopsy is one of the few instances in which Pneumococcus II has been found, Pneumococcus II being present in blood and lungs, B. influenzæ, in lungs and bronchi. In 2 additional instances (Autopsies 374 and 392) peribronchial pneumonia, recognizable at autopsy, has been associated with consolidation having the characters of lobar pneumonia. In one instance, Autopsy 374, the right lung has contained two patches of firm, mottled red and pinkish red coarsely granular consolidation each about 6 cm. across, one situated in the upper lobe and the other in the lower lobe. Elsewhere in the lung, in definite relation to dilated bronchi, occur patches of firm, red, coarsely granular consolidation from 1 to 1.5 cm. in diameter when cut transversely. The bronchus in the center has contained purulent fluid. In the opposite lung similar consolidation has been limited to zones about dilated bronchi which contain purulent fluid. Pneumococcus IV has been obtained from the blood of the heart.

The peribronchial pneumonia which has been described occurs in association with evidence of profound injury to the bronchial wall. In 5 of 6 instances purulent bronchitis has been found at autopsy; in half of these instances bronchiectasis has been noted. The epithelium of the bronchus has been found separated from the underlying tissue by serous exudate, blood and leucocytes; epithelial cells undergo necrosis and disappear, the denuded surface being covered by fibrin. Necrosis extends a varying depth into the wall of the bronchus; blood vessels are engorged, and there is in some instances hemorrhage throughout the wall of the bronchus.

The character of the exudate in the alveoli surrounding the bronchus differs considerably in different instances. In some instances (Autopsies 374 and 392) red blood corpuscles are predominant in the alveoli in contact with the bronchial wall, whereas in a peripheral zone polynuclear leucocytes are more abundant. In other instances (Autopsies 253 and 402) alveoli next the bronchial wall contain abundant fibrin and these are surrounded by a zone in which the alveoli are filled with blood.

Peribronchial pneumonia is the result of the direct extension of the inflammatory process through the wall of the bronchus; it occurs when the epithelium of the bronchus is destroyed and the underlying tissues are injured, but may be present in a wide encircling zone even when the lesion has not penetrated the bronchial wall. The distribution of the pneumonia demonstrates very clearly that the inflammatory process does not reach the affected peribronchial alveoli by way of the bronchioles tributary to the bronchus.

The bacteriology of these instances of peribronchial pneumonia is noteworthy. (Table XLII.)

Pneumococcus has been found in every instance either in the lungs or blood. Pneumococcus II, which has been uncommon with the pneumonia following influenza at Camp Pike and has occurred only ten times in more than 200 autopsies, has been present in one-half of these cases. The constant association of the lesion with pneumococcus is particularly significant when a comparison is made between the incidence of pneumococcus with peribronchial hemorrhage, on the one hand, and peribronchial pneumonia on the other; pneumococcus has been present in less than a third of the instances of hemorrhage but in all instances of pneumonia.

In addition to the instances in which gross peribronchial consolidation has been noted at autopsy, microscopic examination has demonstrated the presence of fibrinous pneumonia surrounding bronchi in a considerable number of autopsies. In a zone encircling small bronchi (with no cartilage) alveoli are filled by plugs of dense fibrin (Fig. 20) containing in variable number polynuclear leucocytes and mononuclear cells. The width of the zone is often equal or greater than the diameter of the bronchus. Alveoli outside the zone of fibrinous inflammation may contain red blood corpuscles or serum, and desquamated epithelial cells are often abundant.

Of 21 instances of peribronchial fibrinous pneumonia 20 were associated with purulent bronchitis. Further evidence of the relation of the lesion to profound injury to the bronchi is its association with bronchiectasis in 17 instances.

Peribronchial fibrinous pneumonia, like other lesions encircling the small bronchi, bears a direct relation to the severity of microscopic changes in the bronchus. The epithelium of the bronchus is either partially or completely lost. Occasionally epithelium is raised by hemorrhage or leucocytes from the underlying tissue but more frequently it is wholly lost and the surface is covered by a layer of fibrin. In the early stages of the lesion, polynuclear leucocytes may be numerous throughout the bronchial wall, indicating that the inflammatory irritant within the lumen is affecting the entire wall and extending its influence to the surrounding pulmonary tissue. Later lymphoid and plasma cells are more abundant than polynuclear leucocytes. Coagulative necrosis and disintegration of the bronchial wall, proceeding from the inner surface outward, may extend more or less deeply, and fibrinous inflammation of adjacent alveoli is often more extensive about that segment of the bronchus which shows the greatest change. In some instances segments of the bronchial wall or even the entire wall has disappeared, so that alveoli containing fibrin form part of the wall of the cavity thus formed. When bronchiectasis has occurred, there are often fissures from the lumen through the entire wall extending into the surrounding lung tissue: here fibrinous pneumonia is particularly conspicuous, occurring in a zone about the edges of the defect. This deposition of fibrin within the alveoli adjacent to the injury doubtless has a part in limiting the distribution of bacterial infection. Nevertheless breaks in the continuity of the bronchial wall are not essential to the production of the lesion and the irritant, which is responsible for the lesion, may penetrate through the bronchial wall to surrounding alveoli and from alveoli to other alveoli immediately adjacent.

With this peribronchial pneumonia the smallest bronchi are distended with pus and their walls are infiltrated with polynuclear leucocytes, lymphoid and plasma cells. In a broad zone encircling the bronchus the alveoli are filled with plugs of fibrin. Bronchioles are similarly distended with polynuclear leucocytes; the alveoli which occur upon the wall of the bronchiole are often limited to one side of the wall and are filled with fibrin. This fibrin occasionally projects into the lumen of the bronchiole and forms a continuous layer in contact with the wall on the same side. The alveolar duct and infundibulum are distended with polynuclear leucocytes. The alveoli upon the wall of the alveolar duct and upon the proximal part of the infundibulum are filled with fibrin. The bronchus, bronchiole, alveolar duct and part of the infundibulum are thus surrounded by a continuous zone of alveoli containing fibrin. The alveoli about the distal part of the infundibulum may be filled with polynuclear leucocytes. Lung tissue between adjacent zones of fibrinous pneumonia may contain serum and desquamated epithelial cells.

Organization of peribronchial fibrin was found in 10 of the 22 autopsies in which peribronchial fibrinous pneumonia had been found. Fibroblasts have invaded the fibrin and newly formed capillaries have penetrated into it. In some instances the interalveolar septa are thickened and infiltrated with lymphoid and plasma cells, and in 7 instances there was chronic pneumonia with thickening and mononuclear infiltration of the interstitial tissue about the bronchi and blood vessels, and elsewhere. The duration of the fatal illness in 12 instances with no organization was usually from ten days to two weeks, though in 3 instances there was no organization although the respiratory disease had lasted from seventeen to nineteen days (average duration with no organization, 13.5 days). The duration of illness in 10 instances with organization of fibrin was slightly less than three weeks (average 18.9 days). These figures do not accurately represent the duration of pneumonia which usually develops after a period of several days following onset of influenza.

This group of instances of peribronchial fibrinous pneumonia has offered an opportunity to study the bacteriology of pneumonia with organization and to determine if it presents any unusual characters. The bacteriology of autopsies with peribronchial fibrinous pneumonia with no organization is shown in Table XLIII:

The bacteriology of instances of peribronchial fibrinous pneumonia with organization of the intraalveolar fibrin is shown in Table XLIV:

B. influenzæ has been present in the bronchi in every instance save one in which cultures have been made, and it is probable that in this exceptional instance cultures have remained sterile because the media employed have been defective. The incidence of B. influenzæ in the lung has been unusually high both with and without organization (66.7 per cent with no organization; 77.8 per cent with organization). Streptococci and staphylococci have been found in a considerable proportion of all instances of peribronchial fibrinous pneumonia, but there has been no notable preponderance of these microorganisms when organization has occurred. Organization has been present in instances in which pneumonia is referable to pneumococcus associated with B. influenzæ and unaccompanied by either streptococci or staphylococci (Autopsies 419 and 422). Wadsworth^[81] found no organization after inoculation of the lungs of dogs with pneumococcus or with staphylococcus alone, but produced organization when he inoculated animals with both microorganisms.

Injury to bronchi produced in part at least by B. influenzæ exposes the bronchi and lung tissue to repeated infection with a variety of microorganisms; absorption of fibrin and regeneration of alveolar epithelium are prevented, resolution fails to occur and organization of fibrin follows.

Suppurative Pneumonia With Necrosis and Abscess Formation

Three varieties of suppurative pneumonia have occurred in association with influenza.

A. Necrosis and suppuration with formation of one or several abscesses usually below the pleura and almost invariably caused by hemolytic streptococci.

B. Interstitial suppurative pneumonia caused by hemolytic streptococcus.

C. Multiple abscesses in clusters caused by staphylococci.

Suppurative pneumonia with necrosis and abscess formation will be discussed in this section. Pulmonary abscesses which occurred in 43 autopsies may be included in this group; in 4 of these autopsies abscess and interstitial suppurative pneumonia occurred in the same individual. These abscesses were much more frequently situated in the lower than in the upper lobes and more often in the right than in the left lung. In most instances there was one or several abscesses situated below the pleura of one lobe; occasionally abscesses occurred in two lobes of the same lung or in both lungs. The distribution was as follows: Abscess in only one lung occurred in right upper lobe in 6 autopsies; middle lobe, 3; lower lobe, 15; left upper lobe, 2; lower lobe, 16. Abscesses occurred in both right and left lower lobes, twice. The usual situation was at the lower and posterior part of the lower lobe at or near the basal edge, less frequently below the posterior border or upon the basal surface of the lobe. These abscesses in almost every instance were found immediately below the pleural surface, so that they appeared upon the pleura as opaque yellow spots usually surrounded by narrow zones of hemorrhage. In one instance (Autopsy 376) the abscess cavity was separated from the pleural cavity by remains of the pleura which was as thin as tissue paper and in other instances perforation had occurred (Fig. 9). In Autopsy 480 the abscess cavity which had perforated the pleura was in free communication with a bronchus of medium size.

In most instances of suppurative pneumonia there have been associated lesions of bronchopneumonia which have been peribronchiolar, hemorrhagic or lobular and have exhibited no unusual characters. The abscess or abscesses are situated within an area of pneumonic consolidation which is not limited by lobule boundaries and has not the characters of bronchopneumonic consolidation. In some instances this consolidation is limited to a zone immediately about the abscess, but often it involves the greater part of a lobe. The tissue is laxly consolidated and flabby; on section it has a dull, conspicuously cloudy appearance and is grayish red, pinkish gray or gray; it is homogeneous or very finely granular. Turbid gray fluid, which sometimes resembles thin pus, oozes from the cut surface.

Widespread necrosis of tissue is not infrequently a conspicuous feature of this pyogenic pneumonia (Fig. 8). Upon a cloudy gray background of consolidation are numerous opaque yellowish gray or yellow patches, occasionally 2 or 3 cm. across, giving a mottled character to the cut surface. Upon the pleura these necrotic patches appear as dull opaque yellow spots. They may be surrounded by a zone of hemorrhage. The opaque material is at first firm but may undergo softening, becoming semisolid and finally purulent. Necrotic patches may be scattered throughout a lobe, but fully formed abscesses are with few exceptions immediately below the pleura (Fig. 9).