Certain conclusions may be drawn from the literature on the bacteriology of the respiratory lesions associated with influenza. All reports show that a few organisms have been found more or less constantly in influenza and influenzal pneumonia: the pneumococcus group, the streptococci (hemolytic, non-hemolytic, pandemicus, etc.) and the Pfeiffer bacillus. They may occur alone, together, or with less frequently found organisms. Among the latter, the staphylococcus, the Micrococcus catarrhalis, Bacillus pneumoniÆ (Friedlander), diphtheroids, and undetermined organisms, all have been reported (2, 48, 62, 67, 68, 92).

Another feature has been the variation of the predominating organism, or organisms, in different localities, and in the same locality at different times. For example, Wolbach (162) at Camp Devens, Massachusetts, demonstrated the Pfeiffer bacillus at autopsy in twenty-three out of twenty-eight cases. In fourteen, it was in pure culture. Keegan (67) at Chelsea, Massachusetts, also found it in eighty-two per cent of the lungs at necropsy, in thirty-one per cent of which it was in pure culture. MacCallum (92), working at Camp Lee, Virginia, found the pneumococcus, Type IV, the predominating organism and rarely the Pfeiffer bacillus. “At the Johns Hopkins Hospital similar methods revealed no influenza bacilli whatever.” At Camp Dix, New Jersey, however, MacCallum found the Pfeiffer bacillus in every case. At Camp Grant, Illinois, Hirsch and McKinney (60) state that the epidemic was due to a virulent strain of pneumococcus and that the Pfeiffer bacillus played no rÔle. At the Puget Sound Navy Yard, Ely and co-workers (37) did not find the Pfeiffer bacillus; they attributed the epidemic to the hemolytic streptococcus. Goodpasture (48), working at the same hospital as Keegan, reports that the bacteria found in December, 1918, and January, 1919, were different from those found in the early months of the epidemic, inasmuch as in the latter group the hemolytic streptococcus was found in one hundred per cent of the cases and the Pfeiffer bacillus in twelve per cent. The foreign literature shows similar variations in the bacteriology.

The organisms associated with influenzal pneumonia are the so-called “mouth organisms.” They are not only found in the mouths and upper air passages of the influenza patients, but also in those of normal individuals. This points to the fact that the bacteria of the mouth have gained access to the lung, probably already injured by a primary agent, in sufficient numbers to bring about a serious inflammatory process. In this connection it is of interest to note the relatively high frequency of the mouth organisms, pneumococcus, Types III and IV, in influenzal pneumonia as compared to the less frequent mouth inhabitants, Types I and II, which are responsible for two-thirds of the cases of true lobar pneumonia (5, 45, 92, 121).

Only the eighty-two cases at the New Haven Hospital are included in the following report. Routine post-mortem cultures were taken from blood, lung, serous fluid wherever present, and exudates from the trachea and bronchi in the later cases. Blood and serous fluids were cultured into neutral infusion broth and plated on blood agar after twelve to thirty-six hours’ incubation. Lung and bronchial cultures were streaked on blood agar plates and for the last third of the series on Avery’s oleate media. Cultures were examined on each successive day and were discarded only after one week. Undoubtedly, we have failed to find B. influenzÆ in many of the earlier cases because of lack of familiarity with the organism, of variations in its morphology (33), and of unsuitable culture media. The organisms tabulated below include only those determined by cultural methods; those found by direct smears or in histopathological preparations are not considered. (See Table No. I.)

The hemolytic streptococcus has been found frequently in association with purulent pleural effusions in influenza, Thomas (146), Stone and Swift (138), Ely et al (37); and Goodpasture (48) has suggested an etiologic relationship between the type of effusion and the infecting organism in influenza, as has been brought forward for the similar post-measles empyema in the army camps. At the New Haven Hospital the Streptococcus hemolyticus was found frequently in non-fatal empyema. However, at post-mortem examination the frequency of this organism in pleural effusions of various types was no greater than that of several other organisms, but it occurred in the only two cases of frank empyema of this series. (See Table No. II.)

Various observers have emphasized the types of organism associated with different gross and microscopic manifestations of influenzal pneumonia. Pfeiffer described the peribronchial type with purulent bronchitis, from which the influenza bacillus was isolated, and the same association has been noted by MacCallum (92), Wolbach (162), Wegelin (156), Dietrich (34), and others. Opie et al (110), however, in a series from which B. influenzÆ was isolated in over eighty-five per cent of the necropsies rarely found this picture. Wolbach states that the gross anatomical picture in influenzal pneumonia is similar to that following measles, from which a hemolytic streptococcus has been isolated in a high percentage of cases. MacCallum (93) classified this type as interstitial pneumonia. It is interesting to note that interstitial pneumonia has been rare in many localities where the hemolytic streptococcus has been prevalent during the past year. Stone and Swift (138) state that “despite the prevalence of the streptococcus at necropsy, only eight instances of so-called interstitial pneumonia were found in a series of fifty-five cases,” and Goodpasture (48) failed to find a single example in a series of sixteen cases. The pneumococci Types I and II, frequently encountered in the usual forms of lobar pneumonia, have been found exceptionally in this epidemic. Type II, however, has been reported (67, 107) present in about the same proportion as in true lobar pneumonia (5). Chickering and Park (25) described a series of cases of pneumonia due to the staphylococcus characterized by multiple miliary abscesses. Necrotization and abscess formation, however, have been striking features of the pathology of this epidemic, even when the staphylococcus has not been demonstrable. Recently Wadsworth (154) demonstrated experimentally that organization in pneumonia does not result from the pneumococcus or the staphylococcus alone, but only follows when both organisms are associated. On the contrary, Blanton and Irons (12) found that “there was no difference to be made out in the nature of the process caused by the streptococcus, pneumococcus, or influenza bacillus.”

The type of the pneumonic process, as well as the invading organism, has varied widely in different localities. From the foregoing brief survey, it will be seen that before the associations suggested above can be proven, much more definite evidence must be presented.

An attempt has been made to correlate the bacteriological findings in this series with the distribution and type of pneumonic process. These are tabulated in Table III. It will be seen at a glance that no relationship is demonstrable between the type of single or associated organisms and the distribution of the pneumonia, whether it is lobar, pseudolobar, peribronchial or lobular and whether acute fulminating, necrotizing or organizing. (See Table No. III.)

The confusion which exists in associating different bacterial types with well defined and characteristic anatomical lesions is true, not only for the disease as it occurs spontaneously in man, but also for its reproduction in experimental animals. A review of the literature teaches that not all of the factors concerned in producing the different anatomical types are known.

Pertinent experimental studies were reported recently by Blake and Cecil from the Army Medical School in Washington. One cubic centimeter of a very virulent pneumococcus (M. L. D. for a mouse in forty-eight hours equals one-millionth of 1 cubic centimeter) inoculated in monkeys through the skin of the neck into the trachea just beneath the larynx has uniformly produced a lobar pneumonia, clinically as well as anatomically the nearest experimental approach to spontaneous lobar pneumonia of man. This work indicates, not only the possible importance the species may play, but also points to the rÔle of the infecting organism which, when sufficiently virulent, incites a characteristic reaction even when inoculated in a minimal quantity into the trachea as distinct from the bronchioles or lung tissue.

Pneumonia unassociated with a simultaneous or preliminary incapacitation of the mechanism of the upper respiratory tract, differs from the pulmonary lesions encountered after measles, gas poisoning, influenza, etc., and those experimentally produced by intrabronchial insufflations. In the lobar type the virulence of the infecting organism, combined perhaps with species variation,—implying as this does different capacities for reaction on the part of the host,—seems to be the fundamental principle. It remains to be determined whether other organisms of equal virulence with that of the pneumococcus are capable of producing characteristic anatomical reactions when inoculated in a similar manner.

In the other group, where there is not only a free ingress to the pulmonary parenchyma by the bacteria of the mouth, but where there is, in all probability, also a simultaneous or preliminary pulmonary damage furnishing a proper medium for relatively innocuous organisms, the lesion of response will surely depend upon other factors as much as upon the infecting organism. Among these obscure factors, the most important, unquestionably, is the extent of the damage to the lung tissue before the entry of the organisms, or simultaneously with it, into this area. The systemic capacity to compensate also must be considered.