The relative frequency with which different types of pneumococci produce lobar pneumonia under the conditions existing when Camp Pike was attacked by an epidemic of influenza is indicated by Table XXXII in which instances of lobar pneumonia alone and of combined lobar and bronchopneumonia are listed separately.

Pneumococcus I and II, which are found approximately in two-thirds of instances of lobar pneumonia occurring in cities, have an insignificant part in the production of these lesions. Pneumococcus IV and atypical Pneumococcus II, which are commonly found in the mouth, are the predominant cause of these lesions, and with Pneumococcus III, also an inhabitant of the mouths of normal individuals, have been the cause of two-thirds of all instances of lobar pneumonia observed in this camp.

There is no noteworthy difference in the occurrence of these types of pneumococci among instances of lobar pneumonia, on the one hand, and of combined lobar and bronchopneumonia, on the other. Different types exhibit no noteworthy differences in their ability to penetrate into lungs and blood.

Hemolytic Streptococcus with Lobar Pneumonia.—There can be no doubt that the concurrent infection with microorganisms other than pneumococcus modifies the progress of lobar pneumonia. With lobar pneumonia alone hemolytic streptococci have entered the bronchi in 30 per cent of instances and have penetrated into the lungs in 20.6 per cent; with associated lobar and bronchopneumonia the same microorganism has entered the bronchi in 34.3 per cent of instances and invaded the lung in 31.6 per cent. Hemolytic streptococci are the only microorganisms other than pneumococci which, in association with lobar pneumonia, have found their way from the lungs to the blood stream; more than one-third of all instances of lobar pneumonia in which hemolytic streptococci find entrance into the bronchi die with streptococcus septicemia.

Separation of instances of lobar pneumonia into groups on the basis of the occurrence of red or gray hepatization shows that infection with hemolytic streptococcus is more likely to occur during the early stages of the disease. The average duration of lobar pneumonia with red hepatization has been 3.7 days, with red and gray hepatization, 5.1 days, and with gray hepatization, 7.5 days. Infection with hemolytic streptococcus has occurred in association with red or gray hepatization as shown in Table XXXIII.

Notwithstanding the longer duration of the disease and consequent prolongation of exposure to infection, lobar pneumonia, which has reached the stage of gray hepatization, has shown the smallest incidence of infection with hemolytic streptococci. In the stage of gray hepatization there is diminished susceptibility to secondary infection with this microorganism.

Characteristic histologic changes have been found in the lungs of those who have died with lobar pneumonia followed by invasion of lungs and blood by hemolytic streptococci (e. g., Autopsies 273, 430), but with no evidence of suppuration found at autopsy. Within the pneumonic lung occur patches of necrosis implicating both exuded cells and alveolar walls; in some places nuclei have disappeared; elsewhere nuclear fragments are abundant. In these patches of necrosis Gram-positive streptococci in short chains occur in immense number. In some instances (e. g., Autopsies 273, 346, 479) interlobular septa are very edematous and often contain a network of fibrin; lymphatics are dilated and contain polynuclear leucocytes in abundance. Streptococci are found within these lymphatics. The histologic changes which have been described represent the earliest stages of abscess formation and interstitial suppuration, lesions almost invariably caused by hemolytic streptococci.

Relation of Lobar Pneumonia to Influenza.—Some writers have suggested that lobar pneumonia, heretofore observed during the course of epidemics of influenza, is an independent disease with no relation to influenza, both diseases being referable perhaps to similar meteorologic or other conditions. Chart 2, which shows by weeks from September 1 to October 31 the relation of deaths from lobar pneumonia (indicated by double hatch) to deaths from all forms of pneumonia, disproves this suggestion. The two curves follow parallel courses; that representing lobar pneumonia reaches a maximum approximately one week after the outbreak of influenza had reached its height. Lobar pneumonia, like other forms of pneumonia, was secondary to influenza. When a chart is plotted to represent the dates of onset of fatal cases of lobar pneumonia (indicated by single hatch in Chart 2), it becomes evident that the greatest number of these cases of pneumonia had their onset at the beginning of the influenza epidemic, approximately one week before it reached its height. Fatal lobar pneumonia developed less frequently in the latter part of the epidemic; to obtain an explanation of this relation it is necessary to chart separately cases of lobar pneumonia with secondary streptococcus infection, for we have already learned that streptococcus infection was the predominant cause of death in the early period of the influenza epidemic. Exclusion of these instances of secondary streptococcus infection makes no noteworthy change in the character of the chart. Fatal lobar pneumonia, like all forms of fatal pneumonia (p. 140), was more frequent in the first half than in the second half of the epidemic. This difference is referable either to greater virulence of the virus of influenza or to the greater susceptibility of those first selected by the disease or, as more probable, to conditions such as crowding together of patients with influenza, favoring the transmission of microorganisms which cause pneumonia.

Bronchopneumonia

For the purpose of the present study it is convenient to group together instances of bronchopneumonia which have been unaccompanied, on the one hand, by lobar pneumonia (p. 155) or, on the other hand, by suppuration, which with few exceptions is caused by hemolytic streptococci or by staphylococci. A group of cases in which lobar and bronchopneumonia have occurred in the same individual have already been considered. In many instances, bronchopneumonia is accompanied by abscess formation or by some other form of suppuration; these lesions will be discussed elsewhere.

Bronchopneumonia unaccompanied by lobar pneumonia or by suppuration occurred in 80 autopsies.

Pneumonic consolidation distributed with relation to the bronchi exhibits considerable variety, and an attempt to define a type of bronchopneumonia characteristic of influenza would be futile. Nevertheless, the bronchopneumonia of influenza has in many instances distinctive characters.

Lesions of bronchopneumonia which are frequently found in the autopsies under consideration may be conveniently designated by descriptive terms, indicative of their location in the lung tissue. These lesions, of which two or more often occur in the same lung, are:

1. Peribronchiolar consolidation with which the inflammatory exudate is limited to the alveoli in the immediate neighborhood of the bronchioles.

2. Hemorrhagic peribronchiolar consolidation in which gray patches of peribronchiolar pneumonia occur upon a deep red background produced by hemorrhage into alveoli. Pfeiffer believed that this lesion was characteristic of influenza.

3. Lobular consolidation with which consolidation is limited to lobules or groups of lobules.

4. Peribronchial pneumonia with which small bronchi are encircled by pneumonic consolidation.

Each one of these lesions will be discussed separately.

Following is a list of the bacteria which have been isolated from the consolidated lung of individuals with bronchopneumonia unaccompanied by lobar pneumonia or by suppuration:

The similarity of this list to that representing the bacteriology of bronchitis is evident; there is the same multiplicity of microorganisms and the frequent occurrence of mixed infections. B. influenzæ is much less frequently found in the lung. The relative pathogenicity of the large group of microorganisms enumerated above is better indicated by the following list which shows what microorganisms have penetrated into the blood in autopsies performed on individuals with bronchopneumonia:

Table XXXIV shows the percentage incidence of pneumococcus, hemolytic streptococcus, staphylococcus and B. influenzæ in the bronchi, lungs and blood and is inserted for comparison with the similar table (Table XXX) showing the incidence of these bacteria in lobar pneumonia.

Table XXXIV shows that pneumococci have a less important part in the production of broncho than of lobar pneumonia; with lobar pneumonia this microorganism was found in the lungs in 77.3 per cent of instances and in the blood, in 65.5 per cent, whereas with bronchopneumonia it was found in the lungs in 42.6 per cent and in the blood in 31.4 per cent. Hemolytic streptococci (in lungs and blood) and staphylococci (in lungs), on the contrary, were more common with bronchopneumonia, and doubtless have a part in the production of the lesion. Streptococcus viridans, B. coli and M. catarrhalis, which are not infrequently found in the bronchi (p. 151), occasionally enter the lungs with bronchopneumonia but are rarely found with lobar pneumonia. B. influenzæ has been found in less than 80 per cent of instances in the bronchi and in about half of the lungs, maintaining an incidence approximately the same as that with lobar pneumonia.

Table XXXV shows the types of pneumococci found in association with bronchopneumonia and is inserted for comparison with the similar table (Table XXXII) showing types of pneumococci with lobar pneumonia.

With broncho as with lobar pneumonia pneumococci commonly found in the mouth, namely, atypical II, and Types III and IV, have a more important part in production of the lesion than the so-called fixed types, I and II. Atypical Pneumococcus II has been less frequently encountered with broncho than with lobar pneumonia.

Peribronchiolar Consolidation.—In many instances of bronchopneumonia, usually in association with lobular or confluent consolidation, small firm nodules of consolidation are clustered about the bronchioles (Fig. 2). These nodular foci of consolidation are usually 1.5 to 2 mm. in diameter, being sometimes slightly smaller or slightly larger. They are usually gray and occasionally surrounded by a red halo; sometimes they are yellowish gray. They are clustered about the smallest bronchial tubes to form groups which are from 0.5 to 1 cm. across. A group of nodular foci of consolidation occupies the central part of a lobule of lung tissue. When pneumonia has been of short duration these foci are fairly soft and not sharply defined, and in many instances this form of bronchopneumonia is first recognized by microscopic examination. When the disease has lasted from ten days to two weeks, the consolidated nodules are very firm and sharply circumscribed, closely resembling tubercles. When they have assumed this character, microscopic examination shows that chronic changes indicated by new formation of interstitial tissue have occurred.

The lesion may be designated peribronchiolar consolidation. It has occurred usually in association with other types of pneumonic lesion in 61 instances, being recognized at autopsy in 18 and by microscopic examination in 43.

In association with this lesion there are almost invariably severe lesions of the bronchi. Purulent bronchitis was noted in 47 of the 61 instances, in which this nodular bronchopneumonia was found at autopsy. An index of the severity of the bronchial injury is the frequency with which bronchiectasis has occurred; dilatation of small bronchi was observed in 24 instances. In 10 instances the bronchi were encircled by conspicuous zones of hemorrhage.

In association with this peribronchiolar lesion the lung is often voluminous and fails to collapse on removal from the chest. Pressure upon the lung squeezes from the smallest bronchi, both in the neighborhood of the nodular consolidation and elsewhere, a droplet of viscid, semifluid mucopurulent material. The presence of this tenacious material throughout the small bronchi doubtless explains the failure of the lung tissue to collapse. Interstitial emphysema has been present in some of these lungs.

A red zone of hemorrhage has occasionally been observed about the foci of peribronchiolar pneumonia. A further stage in the same process is represented by hemorrhage into all of the alveoli separating these patches of consolidation. This hemorrhagic lesion, which will be described in more detail later, has been found repeatedly in the same lung with peribronchiolar pneumonia, being present in 8 among the 61 autopsies cited. Lobular bronchopneumonia accompanied the peribronchiolar lesion 27 times and lobar pneumonia accompanied it 20 times.

When an abscess caused by hemolytic streptococcus is associated with peribronchiolar pneumonia, empyema is present, but otherwise pleurisy is absent or limited to a scant fibrinous exudate.

Histologic examination demonstrates very clearly the relation of this lesion to the bronchioles (Fig. 3). These passages are filled and distended with an inflammatory exudate consisting almost entirely of polynuclear leucocytes. The respiratory bronchioles are beset with alveoli often limited to one side of the tubule and these alveoli are filled with leucocytes. The alveolar ducts, distinguishable from the bronchioles by the absence of columnar or cubical epithelium and by possession of smooth muscle, are similarly filled with leucocytes; the numerous alveoli which form the walls of the alveolar ducts are distended by an inflammatory exudate. In sections which pass through an alveolar duct and one or more of its infundibula, the further extension of the lesion may be determined (Fig. 4). The infundibulum in proximity with the alveolar duct contains polynuclear leucocytes and the same cells are seen in the alveoli which here form its wall, but the intensity of the inflammatory reaction diminishes toward the periphery, so that the distal part of the infundibulum, which is much distended and in consequence more readily definable than usual, is free from inflammatory exudate.

Occasionally there is irregularly distributed hemorrhage and perhaps some edema in the alveoli immediately adjacent to those which form the peribronchiolar focus of inflammation. In such instances small bronchi, that is, air passages, lined by columnar epithelium and devoid of tributary alveoli, may be surrounded by a zone of hemorrhage; immediately surrounding the bronchus, the wall of which shows intense inflammation, alveoli, in a zone of which the radius represents several alveoli, are filled with blood. This hemorrhagic zone is continued from the bronchus over the focus of inflammation which surrounds the bronchiole.

Another variation in the character of the lesion is doubtless referable to variation in the severity of primary bronchial injury. Alveoli immediately surrounding small bronchi are filled with dense plugs of fibrin. The alveoli which besot the walls of the bronchioles contain fibrin, but the alveolar duct and its tributary alveoli are filled with polynuclear leucocytes.

The bacteria which have been cultivated from the lung in autopsies with peribronchiolar pneumonia are as follows:

The following list which shows the bacteria found in the blood is an index to the pathogenicity of pneumococci and hemolytic streptococci:

The percentage incidence of pneumococcus, hemolytic streptococcus, staphylococcus and B. influenzæ in bronchus, lung and blood, given in Table XXXVI, is inserted to indicate with what readiness each one of these microorganisms passes from the bronchus through the lung into the circulating blood.

B. influenzæ is present in the bronchi in a very large proportion (84.8 per cent) of those in whom this type of bronchopneumonia has been found at autopsy; it is much less frequently recovered from the lungs. Staphylococci, in part S. albus and in part S. aureus, are less frequently found in the bronchi and are recovered from the lungs in a relatively small proportion of autopsies. The percentage incidence of pneumococci and streptococci in lungs and blood demonstrates the pathogenicity of these microorganisms, for whereas pneumococci and hemolytic streptococci are found in the consolidated lungs in 43.9 and 61.0 per cent of instances of the lesion respectively, they make their way into the blood in 40.3 and 36.8 per cent of instances.

Coexisting infection with pneumococci and hemolytic streptococci has been not uncommon e. g., Autopsy 275 in which both were in the blood; in 2 instances (Autopsies 333 and 378) in which pneumococci were obtained from the blood, hemolytic streptococci were found in the lungs and bronchi; in 3 instances (Autopsies 258, 273 and 445) in which hemolytic streptococci were present in the blood, pneumococci were obtained from the lungs.

In the group of autopsies under consideration, examination of the sputum was made during life and after onset of pneumonia in 11 instances. The microorganisms found in the sputum and at autopsy were as follows:

In 2 instances (Autopsies 285 and 346) among this small group of cases, pneumococci but no hemolytic streptococci were found in the sputum several days before death, whereas death occurred as the result of secondary invasion with hemolytic streptococci and no pneumococci were found at autopsy. It is probable that this sequence of events is not uncommon. B. influenzæ finds its way into the bronchi and pneumococci follow it; pneumonia limited to peribronchiolar alveoli may occur in consequence of this invasion. Later hemolytic streptococci may follow the same path and cause death with bacteremia.

Hemorrhagic Peribronchiolar Consolidation.—Peribronchiolar pneumonia accompanied by diffuse accumulation of blood within the alveoli is one of the most frequent complications of influenza. The lung tissue is laxly consolidated, and on section there is a homogeneous dull deep red background upon which are seen small gray spots (1.5 to 2 mm. in diameter) grouped in clusters about the smallest bronchi (Fig. 5). Wide areas of lung tissue are implicated and the lesion is more common in the dependent parts of the lung than elsewhere. In common with other forms of bronchopneumonia the lesion is in most instances associated with changes in the bronchi; in 55 instances of hemorrhagic bronchiolar pneumonia purulent bronchitis was found in 43 instances; it is noteworthy that purulent bronchitis often is not evident in the presence of pulmonary edema and edema is not infrequent with this pneumonic lesion.

Microscopic examination demonstrates the presence of acute bronchitis; the lumina of the small bronchi contain polynuclear leucocytes and red blood corpuscles. Accumulation of blood may separate the epithelium from the basement membrane. The mucosa immediately below the epithelium contains polynuclear leucocytes in fair abundance and the blood vessels of the bronchial wall are much engorged. Respiratory bronchioles are distended with polynuclear leucocytes and red blood corpuscles. In a zone about each bronchiole, in areas corresponding to the small gray spots seen upon the cut surface of the lung, the alveoli are filled with polynuclear leucocytes. In the lung tissue intervening between these spots of leucocytic pneumonia the alveoli are distended with red blood corpuscles.

In favorable sections it is occasionally possible to follow the bronchiole and alveolar duct, both filled with leucocytes, into an infundibulum. The proximal part of the infundibulum contains polynuclear leucocytes, whereas the distal part and its tributary alveoli are filled with serum and red blood corpuscles.

When the lesion has persisted for a short time there is evidence of beginning migration of polynuclear leucocytes from the blood vessels into the alveoli which are filled with blood. The alveolar walls contain numerous polynuclear leucocytes and leucocytes which have entered the intraalveolar blood are numerous in contact with the wall but occur in scant number in the center of the alveolar lumen.

Alveolar epithelium in contact with the blood in the lumen is usually swollen and often uniformly nucleated.

The inflammatory process is evidently transmitted from the bronchioles and to a less degree from the small bronchi to the adjacent alveoli. Polynuclear leucocytes fill the lumen of the bronchiole and the alveoli immediately adjacent; at the periphery of the focus of pneumonia, the alveoli may contain fibrin. In such instances small bronchi (lined by a continuous layer of columnar epithelial cells) may be surrounded by alveoli containing fibrin.

In sections from one part of the lung, the alveoli between the peribronchiolar foci of pneumonia may be uniformly filled with red blood corpuscles, whereas in sections from another part pneumonic foci may be surrounded by a zone of intraalveolar hemorrhage or of hemorrhage and edema outside of which some air-containing tissue occurs. There are transitions between this halo of intraalveolar hemorrhage and edema surrounding each bronchiolar focus and complete hemorrhagic infiltration of all intervening alveoli.

Large mononuclear cells are occasionally fairly numerous within the alveoli containing blood. These cells act as phagocytes ingesting red corpuscles, so that at times they are filled with corpuscles. Disintegration of red corpuscles occurs and brown pigment remains within the cell. It is not uncommon to find numerous mononuclear pigment containing cells which resemble those found with chronic passive congestion of the lungs.

Lungs, the site of hemorrhagic peribronchiolar pneumonia, may undergo chronic changes which will be described elsewhere.

The lesion which has been designated hemorrhagic peribronchiolar pneumonia is that which Pfeiffer regarded as the characteristic type of influenzal pneumonia. In the small bronchi containing pus and in lung tissue, Pfeiffer states, influenza bacilli are predominant and present in astonishing number in smear preparations. The demonstration of B. influenzæ by cultures from pneumonic lung is mentioned by him but its association with other microorganisms in such cultures is not discussed.

Microorganisms which we have isolated from the lungs of individuals with hemorrhagic peribronchiolar pneumonia are as follows:

With this type of pneumonia B. influenzæ has not been isolated in pure culture; B. influenzæ alone is recorded only once (Autopsy 435), but in this instance the culture has been so obscured by contamination that the occurrence of pneumococci or streptococci cannot be excluded; S. hemolyticus has doubtless been present in this lung, for it has been found in the heart’s blood, in the bronchus, and in the peritoneal exudate of the same individual.

The incidence of pneumococci and hemolytic streptococci in this list does not differ materially from that with peribronchiolar pneumonia unaccompanied by extensive intraalveolar hemorrhage, though hemolytic streptococci are somewhat more frequent with the hemorrhagic lesion. The following table shows the frequency with which pneumococci and hemolytic streptococci have penetrated into the blood:

Table XXXVII showing the percentage incidence of pneumococci, hemolytic streptococci, staphylococci and B. influenzæ further emphasizes the similarity between the bacteriology of peribronchiolar pneumonia (Table XXXVI) and the closely related hemorrhagic lesion:

Pneumococci have been found in the lungs (31.6 per cent) and blood (25 per cent), somewhat less frequently than with peribronchiolar pneumonia (43.9 and 40.3 per cent respectively), and hemolytic streptococci have been found in the blood more frequently (52.1 per cent) than with the latter (36.8 per cent) but otherwise the bacteriology of the two lesions corresponds closely. The low incidence of B. influenzæ in the bronchi (72 per cent) with hemorrhagic peribronchiolar pneumonia is perhaps incorrect as the result of the relatively small number of bacteriologic examinations (namely, 25), but the incidence of the same microorganism in the lung has been higher (52.6 per cent) than with nonhemorrhagic peribronchiolar lesion (43.9 per cent).

In some instances infection with hemolytic streptococci has occurred after the onset of pneumonia. The following list compares the results of bacteriologic examination of the sputum made after the onset of pneumonia with that of blood, lungs or bronchus after death:

Instances of secondary infection with hemolytic streptococcus occur in the list, namely, Autopsies 346 and 376.

From the foregoing studies of the bacteriology of peribronchiolar and hemorrhagic peribronchiolar pneumonia the following conclusions may be drawn: (a) B. influenzæ is found in most instances of these lesions in the bronchi and in about half of all instances in the lungs, but does not occur unaccompanied by other microorganisms. (b) In a considerable number of autopsies pneumococcus is the only microorganism that accompanies B. influenzæ; from the lungs it penetrates into the blood from which it is obtained in pure culture. (c) In a considerable number of instances S. hemolyticus accompanies B. influenzæ, and in some of these instances (representing a large proportion of the relatively small number of cases examined during life), examination of the sputum has demonstrated that infection has been secondary to a pneumonia with which no hemolytic streptococci have been found in the sputum.

Lobular Consolidation.—Consolidation of scattered lobules or groups of lobules has occurred in nearly all instances, namely, 71 of 80 autopsies with bronchopneumonia unaccompanied by lobar pneumonia or by suppuration. When death follows shortly after the onset of pneumonia, patches of consolidation have a dull deep red color; blood-tinged fluid escapes from the cut surface which is almost homogeneous or finely granular. The consolidated tissue seen through the pleura, which is raised above the general level, has a bluish red color. Isolated lobules or groups of lobules which have undergone consolidation may be scattered throughout the lungs, but not infrequently there is confluent consolidation of the greater part of lobes, of whole lobes or of almost an entire lung. Such lungs are very heavy and may weigh 1,400 or 1,500 grams; bloody serous fluid exudes from the cut surface. The lesion resembles the red hepatization of lobar pneumonia, but confluent patches of pneumonia are usually well defined by lobule boundaries. The tissue is soft and the granulation of lobar pneumonia is absent. In many instances the lobular or confluent areas of consolidation are reddish gray; in some instances consolidated tissue is in places red and elsewhere gray, and in a smaller group of autopsies there is gray consolidation only (Fig. 6). Red lobular consolidation is often seen in those who have died within the first four days following the onset of pneumonia, but is almost equally frequent after from five to ten days; the average duration of pneumonia in these cases was 5.5 days. Combined red and gray consolidation was more frequently found when pneumonia had lasted more than five days, the average duration of pneumonia being 7.3 days. The greater number of instances of gray consolidation were found after seven days of pneumonia, the average duration of the disease being 10.0 days. These figures are cited to show that lobular, like lobar, consolidation passes gradually from a stage of red to gray hepatization, but the change occurs more slowly and is often long delayed.

Lobular pneumonia, which occurred 71 times among 80 cases classified as bronchopneumonia, may be regarded as an almost constant lesion of the disease. It is found not only in association with other lesions of bronchopneumonia, but with lobar pneumonia of influenza as well.

The bacteriology of this lesion shows no deviation from that of the slightly larger group of bronchopneumonia (p. 163). All types of pneumococcus have been found in association with the lesion, Pneumococcus I in 2 instances, Pneumococcus II in 1 instance; atypical Pneumococcus II and Pneumococcus IV have been found much more frequently. Pneumococci have been found in more than a third of these autopsies (42.9 per cent in the lungs, 33.3 per cent in the blood); hemolytic streptococci in less than one-third (28.5 per cent in the lungs, 30.2 per cent in the blood).