CHAPTER XVI PNEUMOCOCCUS :
PNEUMONIA
PNEUMOCOCCI are Gram-positive cocci, exhibiting a marked tendency to arrangement as' lanceolate diplococci and growing poorly on ordinary media ; they usually ferment inulin, are soluble in bile and are frequent pathogens in man. Classification. Pasteur (1881) during his work on rabies described a capsulated diplococcus, which he found in the saliva of a child. Fraenkel (1886) saw a similar organism in the sputum of a pneumonia patient ; but the association of the organism with pneumonia was definitely established by Weichselbaum (1886), who also found t h a t it was normally present in the mouth. To this organism he gave the name Diplococcus pneumoniœ. Schottmüller (1903) described an organism which he termed Strept. mucosus ; this is now accepted as a pneumococcus (type I I I ) . These observations were readily confirmed, and the presence of the pneumococcus in the sputum of normal individuals and also its rôle as the setiological agent of pneumonia became definitely established. Our knowledge was further extended when some American workers, in 1917, found t h a t the pneumococcus could be sub-divided serologically into a number of distinct types, viz. types I, I I , I I I , plus a heterogeneous collection referred to as group IV. The taxonomic position of the pneumococcus is somewhat uncertain. I n the classification issued by the Society of American Bacteriologists it is included as the type species of the genus Diplococcus. This arrangement is, however, not universally accepted ; many workers consider t h a t it is a species of the genus " Streptococcus ", as it bears a close resemblance in some respects to the Strept. viridans. There are, however, several well-marked differences, e.g., bile-solubility, inulin-fermentation and agglutination, which are reasonably constant and appear to place the pneumococcus in a separate genus from the streptococci. Morphology and Staining Reactions. When freshly isolated from the tissues the pneumococcus is usually arranged in pairs, each coccus being more oval or lanceolate than spherical, with the 184
PNEUMOCOCCUS : PNEUMONIA
185
broader ends opposed ; each pair is usually enclosed in a single capsule. On prolonged cultivation on artificial media, particularly the fluid varieties, the cocci tend to become more spherical and to be arranged in short chains of three to five. The pneumococcus is non-motile, having no flagella, and does not form spores. Capsule-formation is usually well marked on fresh isolation from the tissues, particularly with type III strains ; the power to form capsules, however, becomes considerably decreased on prolonged artificial cultivation. The pneumococcus stains readily with the usual dyes and it is strongly Gram-positive. The capsules are not stained by the ordinary staining methods, and, when present, are represented by unstained haloes around each pair of cocci ; they can, however, be readily demonstrated by special methods, such as those of Muir and Hiss {q.v.). Cultural Characteristics. The pneumococcus can usually be cultivated without much difficulty. It is important to note that,
F I G . 13.—Colonial appearance of the pneumococcus (in section).
on first isolation, growth on the simpler media is frequently very poor. Enriched * media, such as serum-agar or blood-agar, are therefore usually employed. Growth may be obtained throughout a temperature range of 25°-40° C, but the optimum temperature is 37° C. The optimum reaction of the medium is ρΉ. 7-6-7-8. The organism is a facultative anaerobe, and consequently maximum growth takes place in the presence of free oxygen. In fluid media, such as Hartley's broth or serum-broth, growth in 24 hours usually occurs as a well-marked general turbidity. On further incubation the medium may become clear, owing to autolysis of the organisms. On solid media growth is frequently delicate ; the colonies are small, round and semi-transparent. On blood-agar, when freshly isolated, a more or less typical appearance is found ; the colonies are flat with a sharply raised edge, which tends to be raised above the rest of the colony. On further incubation a central raised portion frequently becomes differentiated from the remainder of the colony (Fig. 13) ; also around the colonies the a type of haemolysis is found. The colonies of the pneumococcus (type III) are generally definitely mucoid, owing to the large amount of capsular substance
186
A TEXT-BOOK
OF
BACTERIOLOGY
formed by these strains. I n all cases old cultures may give atypical appearances. Biochemical Activity. The pneumococcus exhibits little demonstrable proteolytic activity ; gelatin is not liquefied and indole is not formed. Its fermentation activities are more marked ; lactose, maltose, glucose, sucrose and inulin are fermented by most strains with the production of acid, but not gas. As growth is difficult to obtain in the usual peptone-water-sugars an enriched medium is required. The one in frequent use is Hiss's serum-water, which contains 1 part of serum to 3 parts of water, to this the sugar and indicator are added ; acid formation is indicated by a change in the colour of the indicator and coagulation of the medium. Litmus milk is acidified and frequently clotted. Resistance. The pneumococcus is readily destroyed by bactericidal agents, being killed by a temperature of 56° C. in 15-20 minutes. I t is a delicate organism and the conservation of stock-cultures is not easy. A satisfactory method of preservation in stock-culture is the inoculation of a medium composed of semi-solid agar to which fresh rabbit's blood has been added, and conservation, without incubation, in the refrigerator. Under such conditions survival may last for several months. Drying in vacuo at a low temperature gives excellent results. Bile-solubility. I t has long been recognized t h a t the addition of bile salts to a culture of the pneumococcus in a fluid medium results in lysis of the organisms. This property is now used extensively as a means of identification. The mechanism of the action has not been definitely determined, but it is thought t h a t the bile salts act by accelerating the normal autolytic process of the bacterial cell. The lytic activity of bile salts appears to be related to their hydroxyl grouping. Salts having no hydroxyl groups are inactive ; salts with hydroxyl groups, such as desoxycholic acid and apocholic acid, are very active. Bile-solubility is inhibited by heating the culture for 30 minutes at 56° C. In carrying out the test, a small amount (2-4 drops) of ox-bile or a suitable bile salt, diluted 1/10 or 1/100, is added to 5 c.c. of a broth culture of the pneumococcus, which, after standing about 10-15 minutes at 37° C , should be cleared. The reaction of the culture should not be below ^ H 6*6, as, at this point, precipitation of the bile salts takes place. Serology. The existence of different serological strains was first noted by Neufeld and Handel (1909). These preliminary observations were elaborated by Dochez, Avery and others in
PNEÜMOCOCCUS : PNEUMONIA
187
America, and Lister in South Africa. The American workers (1913 onwards) found that there were three common infecting types of pneumococcus, which they labelled I, II and III ; in addition, a heterogeneous group, labelled group IV, was described. Lister (1916), investigating pneumonia in South Africa, described three main serological types, which he styled A, B and C. It was subsequently demonstrated that type B corresponded with the American type II, and type C with type I, but type A had apparently no American counterpart. These observations were soon confirmed, and their importance in the epidemiology of pneumonia recognized. The American system of classification was subsequently adopted, and it is now widely employed ; at the present time some twenty-nine types have been identified in group IV, i.e., there are in all thirty-two distinct types. Investigating the antigenic structure of the various types of pneumococcus, Heidelberger, Avery and their co-workers (1923) made a striking discovery. They found that the pneumococcus contains two main antigenic components : (1) A nucleo-protein and (2) a complex carbohydrate ; other factors are undoubtedly present, but their significance is uncertain. It was also discovered that, with the three main types, the protein is common to all, but the carbohydrate is distinct and specific to each. The carbohydrate has the complex chemical structure of a polysaccharide and has been termed " specific soluble substance " or " S.S.S." The carbohydrate in the case of the type I pneumococcus is dextro-rotatory and contains nitrogen ; that of type II is dextro-rotatory, nitrogen-free, and on hydrolysis yields a glucose unit ; that of type III is lsevo-rotatory, nitrogen-free, and on hydrolysis yields glucose and aldobionic acid. Moreover, this substance was found to be intimately associated with the capsule of the organism and also to behave as a haptene, i.e., it is a partial antigen, reacting in vitro with the specific antibodies produced by the homologous organisms, but unable alone to stimulate their production in vivo. The smooth to rough (S —> R) variation is accompanied by a loss of the capsule and virulence. The typespecificity is thus lost and, as a result of the presence of the common nucleo-protein fraction, a reaction is obtained with antisera prepared against autolysed cultures of any pneumococcal type. The serological tests used extensively in the study of the pneumococcus are agglutination and precipitation ; in the former test the whole bacterial cell is involved, and in the latter the specific soluble substance.
188
A TEXT-BOOK
OF
BACTERIOLOGY
Toxin Production. The pneumococcus does not produce a soluble toxin. A substance similar in many ways to an endotoxin has, however, been extracted from the organism by autolytic dissolution or by alternate freezing and thawing. It has been suggested that the toxic symptoms, which are so frequently present in pneumococcal infections, may be due to absorption and diffusion of the specific soluble substance. The " S.S.S." has been found to be closely associated with the virulence of the organism, and it has been detected in the blood-stream and the urine of pneumonia patients, but proof that it is responsible entirely for the toxic symptoms has not been obtained {vide infra). The pneumococcus produces a soluble hsemolysin. It can be demonstrated in young broth cultures, but as it is readily inactivated by oxidation it is soon destroyed in aerobic cultures. On blood-agar its action is naturally most marked when grown under anaerobic conditions. The hsemolysin is not type-specific and appears to be independent of virulence. Its exact function, particularly in the pathogenesis of pneumococcal infections, is obscure. It is antigenic and an anti-hsemolysin has been demonstrated in the blood of pneumonia patients. Pathogenicity. In man the pneumococcus is normally found as a commensal, being present in the mouth and naso-pharynx. It is also responsible for many diverse conditions such as primary pneumonia, broncho-pneumonia, empyema, otitis media, meningitis, arthritis, endocarditis and primary peritonitis of children. The pathogenicity of the pneumococcus for laboratory animals varies considerably. Rats are highly resistant, while white mice and rabbits are extremely susceptible. In white mice intraperitoneal injection of a fresh virulent strain is followed by a purulent peritonitis, septicaemia and death in 24-48 hours. The normal resistance of man appears to be midway between these extremes. Pneumonia
Lobar pneumonia, in approximately 90-95 per cent, of cases, is caused by the pneumococcus. The pneumococcus is also frequently found as a commensal in the mouth, but it is interesting to note that the incidence of the various types of the organism in pneumonia is in marked contrast with the incidence of the types in the mouths of normal individuals (Tables VII and VIII). It is seen that some 50-70 per cent, of pneumonia cases have been caused by types I and II. The figures have varied in the
189
PNEUMOCOCCUS : PNEUMONIA TABLE
VII
Frequency of the Various Types of the Pneumococcus in Lobar Pneumonia Years
Country
U.S.A. U.S.A. Denmark Britain (Glasgow).
Percentage incidence of types
No. of Cases
I
1913-17 454 1920-30 4,310 1923-25 298 1930-32 1,215
33-3 32-1 28 36-2
II
29-3 23-0 28 35-5
III
13-1 11-1 10 3-6
Group IV
24-6 33-8 34 24 7
(After Cruickshank, 1933.) different countries, but in all instances type III infections have been in the minority, and the distribution of the other types has been, on the whole, similar. A study of cases occurring in a New York hospital during recent years has shown that the type I pneumococcus was consistently the most prevalent, followed by types III, VIII, V and VII. VIII Incidence of Types of the Pneumococcus in Normal Mouths (After Avery, etc.) TABLE
Type
Number
Percentage
I II III Group IV
1 22 34 64
0-8 18-2 28· 1 52-9
The type incidence in normal individuals is quite different ; here more than 50 per cent, of the strains belong to group IV and only 19 per cent, belong to types I and II (Table VIII). The relative frequency of types I and II in pneumonia and
190
A TEXT-BOOK
OF
BACTERIOLOGY
their rarity in the mouths of normal individuals indicate t h a t infection in pneumonia is not, endogenous, but is obtained from an outside source, i.e., it is exogenous. The organism is generally conveyed from healthy or convalescent carriers by droplet infection. Resistance is normally high, and only when this has been reduced by such non-specific causes as fatigue, alcoholism and under-nourishment does pneumonia usually develop. Recent observations indicate t h a t pneumonia is rarely a primary disease but is usually preceded by some infection of the upper respiratory tract, such as the common cold, laryngitis or influenza. These infections are considered to facilitate the spread of the pneumococcus from the naso-pharynx to the lungs. The infectivity of the pneumococcus for man is relatively low ; attendants of cases rarely develop pneumonia, but they frequently become carriers of the organism infecting the patient, while epidemics are rare and occur usually in crowded communities of individuals with a decreased resistance, as found among soldiers in wartime. An alternative theory of the mode of infection is t h a t members of the heterogeneous group IV already present in the throat change in some way to types I and I I , and so give rise to an endogenous type of infection. Griffith has found t h a t by injecting a mouse with a live rough culture of one type (I), together with a killed smooth culture of another type (II), he could recover a smooth living organism of the type I I . The possibility of infection resulting from such a biological change in type, however, appears less probable than the theory of exogenous infection. Pathogenesis. After the causative organisms have been implanted in the throat by droplet infection they pass to the lung, but the precise mode of spread has not been definitely determined. Three routes are probable : (a) A surface spread along the air-passages, (b) via the peritracheal and peribronchial lymphatics, and (c) by the blood-stream. Spread either directly by the air-passages or by the blood-stream is not indicated by either clinical or experimental observations which suggest spread by the lymphatics. Various stages of the disease process have been examined in monkeys after intratracheal inoculation of virulent pneumococci. The findings indicate t h a t the pneumococcus penetrates the bronchial or trachéal mucosa and passes to the root of the lung by lymphatic spread. Rapid invasion of the lung tissue occurs, resulting in consolidation and the development of clinical signs.
191
PNEUMOCOCCUS : PNEUMONIA
The Response of the Patient. Pneumonia tends to run a characteristic course which is readily recognized by the clinician. The severity of the infection is dependent largely on the type of the infecting pneumococcus and the age and general condition of the patient ; the mortality rate is highest in patients over 35, while type II and III pneumonias are more frequently fatal than type I infections. The introduction of serum-therapy has had a decided beneficial influence on type I infection, but the effect is of less significance in pneumonia due to the other types (see Table IX). TABLE
IX
Mortality Rate in Relation to the Type of Infecting Pneumococcus (After Park and his colleagues, 1928) · Without serum
With serum Type of pneumococcus
I II III Group IV
Cases
266 176 82 313
Deaths Deaths percent.
51 61 27 76
19 35 33 24
Cases
249 165 92 324
Deaths Deaths per cent.
82 74 27 83
33 45 29 26
Criteria, indicating the reaction of the patient to the infection and the probable prognosis, were admirably analysed by Cruickshank in the Milroy Lectures of 1933 under the' following * headings :— (1) Temperature. In all cases there is a sharp rise of temperature. In type I infections the level tends to be high throughout the infection ; whereas in type II cases there is a gradual fall from the initial high level. Fall by crisis is more frequent in type I than in type II infections and occurs about the sixth day. This appears to be primarily due to the antibody response, which is more marked in individuals under 35 than over this age. (2) Leucocyte Count. Infection is accompanied by a definite leucocytosis and the mortality rate appears to vary inversely
192
A TE XT-BOOK
OF
BACTERIOLOGY
with the response. When the count is relatively low, under 10,000 per cubic millimetre, the fatality rate is high. The leucocyte count thus constitutes a particularly important prognostic factor. I t may also provide a useful indication of the presence of complications. (3) Bacterœmia. Pneumococci can frequently be isolated from the blood-stream in cases of pneumonia even in the initial stages of infection, and their presence may have an important prognostic significance according to the type of the infecting organism. In the absence of serum-therapy, bacteraemia is present in about one-third of type I cases and about 40 per cent, of type II cases. Bacteraemia has a grave prognostic significance ; in type I cases, not serum-treated, the mortality rate has been found to be as high as 80 per cent., while figures, collected in America and Glasgow, indicated t h a t in type III infections the mortality rate of cases with a positive blood culture was 100 per cent. These figures, except in the case of type III infections, are greatly reduced by the early administration of serum. Repeated blood-cultures may be required as pneumococci may not be demonstrated in the blood-stream until the fourth or fifth day of the disease. (4) Toxaemia. The presence of toxaemia is detected by clinical observations, based on the appearance of delirium, prostration, incontinence, etc. I t has been suggested t h a t the generalization of the S.S.S. is concerned with the toxaemic state. I t has already been stated t h a t the specific soluble substance or S.S.S. may be present in the blood-stream and urine. I n types II and III pneumonia toxaemia is more severe and S.S.S. has been detected in the urine more frequently than in type I infections, in which the mortality rate is also lower. The amount of S.S.S. elaborated appears to have a great bearing on the course of infection. The type I pneumococcus forms little S.S.S., but is highly invasive ; in type I pneumonia the patient's response is usually brisk, and the temperature falls by crisis, little toxaemia being present. Type II pneumococcus produces more S.S.S. and gives rise to a more severe type of infection with more toxaemia. Type III organisms produce most S.S.S. and give rise to a most severe and often fatal form of infection. I t has been found t h a t the S.S.S. combines with the antibodies and so prevents the sensitization and subsequent phagocytosis of the pneumococci. I t is, however, uncertain whether it exercises a directly toxic action per se ; no convincing evidence of this has yet been obtained.
193
PNEÜMOCOCCUS : PNEUMONIA
Other Pneumococcal Infections Broncho-pneumonia. Lobular or broncho-pneumonia is a condition which tends to occur when the individual resistance has been lowered. It is thus not uncommonly found during the winter months as a sequel to such debilitating diseases as influenza, measles and whooping cough in the very young and the old. Its pathogenesis appears to be quite different from that of the lobar type, and it may be caused by a large variety of organisms, particularly pneumococci, hsemolytic streptococci and the influenza bacilli. Many observers have found that, when caused by the pneumococcus, in 75-90 per cent, of cases the causative organism is a member of the heterogeneous group IV. This suggests that the infection is endogenous ; i.e., owing to the decreased resistance of the individual the relatively non-pathogenic pneumococci, normally present in the naso-pharynx, become actively pathogenic. The infection in these cases spreads directly along the bronchial mucosa, by which the anatomical distribution of the lesions is determined. The mortality rate is high and is largely dependent on the age and physical condition of the patient ; the rate is highest in the very young and the very old. Empyema. Pneumococcal empyema is almost invariably a complication of pneumonia. It occurs most commonly in type I infections (55-85 per cent.) in which serum-therapy has not been carried out ; the striking frequency in type I infections is considered to be a manifestation of the invasiveness of this particular type of pneumococcus. Otitis Media and Meningitis. These conditions occur not infrequently in children, and may be either primary or secondary to an attack of pneumonia. Primary infection occurs directly from the mouth and members of group IV are usually the infecting organisms. Peritonitis. Primary pneumococcal peritonitis is sometimes encountered in children. Infection is thought to occur by direct spread via the Fallopian tube from the vulva. DIAGNOSIS. The material sent for examination in pneumococcal infections is usually pus or sputum. The identification not only of the organism but also of the type is commonly desired, and the following scheme of examination is recommended :— (1) Examination of smears by Gram's method and perhaps also for the presence of capsules. (2) Culture on blood-agar and in broth. If the culture is pure the broth can be used for bile-solubility and agglutination. The T.B.
0
194
A TEXT-BOOK
OF
BACTERIOLOGY
blood-agar culture will give the typical colonial appearance and a haemolysis ; if growth is mixed, pick-off separate colonies from the blood-agar plate and sub-culture. (3) Inoculation of inulin. (4) Bile-solubility test. (5) Serological tests. Agglutination is generally carried out by testing the three type sera against the organism, either by the ordinary macroscopic method using a pure culture, or by one of the recently introduced rapid methods. The introduction of serum-therapy in pneumonia has demanded the rapid typing of the causative organism. Several rapid and satisfactory slide methods have been introduced. One useful method originally described by Neufeld and modified by Armstrong does not depend upon agglutination. In this, small portions of sputum are mixed on a slide with a drop of each type serum, and each preparation is then covered with a coverglass. Microscopical examination is performed after a few minutes' interval, the cocci are swollen and present a groundglass appearance in the presence of the homologous serum. Much experience is, however, required before reliable results can be obtained. (6) Animal inoculation. Samples of pneumonic sputum after being well washed and emulsified in saline are injected intraperitoneally into mice ; after 4-6 hours a definite peritoneal exudate forms and is collected. Staining for capsules and agglutination tests, macroscopic or by the slide method, are then carried out. Precipitation tests, using the supernatant fluid, may also be employed, particularly when heavy contamination of the original specimen is present. The differentiation of the pneumococcus from Strept. viridans is sometimes difficult ; the differences usually found are given in Table X. Certain variations may be encountered, e.g., some strains of Strept. viridans may produce a general turbidity in broth and/or ferment inulin. The crucial test is bile-solubility ; all pneumococci must be bile-soluble, whereas this property is never possessed by Strept. viridans. A positive agglutination reaction is, of course, conclusive in the case of types I, II and III. Specific Prophylaxis. Active immunization by vaccines is not called for in ordinary civil practice, but it has been employed in special circumstances when the incidence of pneumonia has reached epidemic proportions. Lister (1917) carried out an
PNEUMOCOCCUS:
195
PNEUMONIA
extensive campaign at the Rand mines among the native workers, in whom alarming epidemics developed. Vaccines, prepared from the three common types, were injected intravenously and later subcutaneously in three large doses. There was subsequently a TABLE
X
The Differentiation of the Pneumococcus from Strept. viridans Test
Morphology Capsules Broth Blood-agar
Pneumococcus
Diplococci | ve
Turbidity a haemolysis and characteristic colonies.
1 ve Bile-solubility Inulin fermenta- Usually + v e tion Agglutination + v e (with type sera)
Strept.
viridans
Usually chainformation Not present Usually a granular deposit. a haemolysis ve
Usually - v e ve
marked decrease in the incidence of pneumonia due to these types among the natives. Active immunization was also employed at various American Army camps during the war with some success. The results, while difficult to assess, indicate that, following the use of vaccines, there is an increased resistance against the homologous types. Felton has recently introduced a preparation of the specific capsular polysaccharide for active immunization against types I and II pneumococci. The results, while encouraging, are not yet conclusive. Recent investigations suggest that, contrary to the accepted view, recovery from pneumonia is accompanied by an increased resistance of short duration to further infection with the homologous type ; when second attacks occur they are usually due to infection with one of the other types. The precise mechanism of the immunity response in pneumococcal infections is, however, still undetermined. Therapy. The application of specific therapy to pneumonia has been mainly due to the efforts of certain American workers during the past 15 years, particularly to the introduction by Felton of a O 2
196
A TEXT-BOOK
OF
BACTERIOLOGY
method of refining and concentrating the serum antibodies. Extensive investigations have been made and several facts have been definitely established. Type I pneumonia has responded best to serum-therapy ; in type II infection the results have been uncertain, and in type III cases the serum appears to be of little value. In group IV cases serum has also given valuable results in infection with types V, VII and VIIL These observations emphasize the necessity for the rapid identification of the type of infecting pneumococcus in cases of pneumonia. The serum must be administered early, intravenously and in adequate doses. The results are often striking, as shown by a diminution in the severity of the symptoms and a reduction in temperature, pulse rate and respiratory rate. The course of the disease is shortened, the mortality rate lowered and the incidence of complication reduced. The main action of the serum appears to be the promotion of phagocytosis. In order to save delay some workers employ a polyvalent serum, containing types I and II antibodies, at once. The type of the infecting organism is then determined, and if this proves to be either I or II the homologous serum is then administered. It is considered that some of the failures of serum-therapy can be explained by the late administration of the serum and also by the faulty typing of the infecting organism. Figures recently published in America show that in 160 cases of type I pneumonia treated during the first 24 hours of the disease the mortality was only 5 per cent. The serum dosage has varied in different countries. In America larger doses are used than in Britain, where an initial dose of 10-20,000 units is frequently given, followed by similar amounts every 8-12 hours until the temperature falls below 102° F. The total amount employed by different workers has been 100,000, 75,000 and 36,000 units. In some cases, particularly when the serum is given after the fourth day and bacteraemia is present, doses of 40-100,000 units have been given every 8 hours. The sufficiency of the serum-dosage may be estimated either by attention to the clinical condition of the patient and by specific tests such as the presence of homologous agglutinins in the serum or the aJJergic reaction described by Francis. Francis (1933) introduced a useful index for serum-therapy ; a small amount of the homologous polysaccharide is injected intradermally, if an immediate positive reaction, as evidenced by a local erythema, occurs, serum-therapy need not be continued. Other workers have controlled the administration of serum by the presence of
PNEUMOCOCCUS
: PNEUMONIA
197
homologous agglutinins in the serum ; when a marked agglutination reaction occurs serum-therapy is discontinued. Antipneumococcal serum is prepared by the intravenous injection of young killed cocci into the horse ; after a prolonged period of immunization, perhaps as long as 9 to 15 months, a potent serum is usually obtained. The antibody is concentrated, but this process is rendered difficult by the fact t h a t the antibody is associated with the water-insoluble protein. Much loss occurs during concentration and the final product is somewhat expensive. I t also gives rise not infrequently to unpleasant reactions. Standardization of the serum has been rather unsatisfactory, and it is only recently t h a t an official standard has been accepted. A unit has been defined by Felton as the amount of serum which will protect mice against one million lethal doses of a culture when serum and culture are given simultaneously. I n this country a dry standard serum, tested against Felton's serum by the above method, is employed as a basis of standardization. This serum has recently been adopted by the League of Nations Commission on Biological Standardization as the official international standard. I n the case of type I antiserum the activity contained in 0-0886 mg. of the serum is defined as one international unit ; with type I I antiserum one international unit is defined as the activity contained in 0-0894 mg. of the serum. A further development in serotherapy has been the preparation of antisera in rabbits. The antibody molecule of rabbit serum is less than t h a t of horse serum, and it is suggested t h a t it consequently penetrates more readily into tissues and through membranes. Specific therapy by means of vaccines has also been employed. The vaccines are heat-killed suspensions of young cultures of the various types of pneumococci. When administered in the early stages of the disease it is claimed t h a t there occurs within 24 hours a fall of temperature, a decrease in pulse and respiration rates, with improvement in the general condition of the patient. Chemotherapy by means of sulphapyridine (2-sulphanilylaminopyridine), often referred to as Dagenan or M. & B. 693, is now being extensively used in the treatment of lobar pneumonia ; the results have been highly satisfactory. The action of the drug is not type-specific as successful results have been reported in infections with most types ; some variation in the susceptibility of individual strains has, however, been reported.
198
A TEXT-BOOK
OF
BACTERIOLOGY
Drug-fast strains have been obtained, and these do not respond in experimental infections to the therapeutic action of the drug. Experimental work indicates t h a t the most satisfactory form of treatment is a combination of serum and the drug, but, in view of the difficulties involved in specific serum-therapy, as a general rule the drug is used alone. Recent experimental work indicates t h a t a combination of the drug with the type-specific vaccine is a particularly satisfactory form of therapy. I t was found t h a t a single dose of vaccine combined with the oral administration of sulphapyridine saved mice t h a t would have died had either method been used alone. The combination of specific therapy and chemo-therapy would appear to offer definite advantages over the use of either of these methods alone. The*drug apparently acts as a bacteriostatic and an associated increase in the specific defence mechanism of the host should accelerate the removal of the pneumococci. The development of toxic manifestations following the administration of sulphapyridine has already been discussed (p. 69). Careful observation should be made when the drug is being given, particularly if relatively large doses are being used, and a copious intake of fluids rigorously enforced.