Streptococcus
Maria Jevitz Patterson
General Concepts
Streptococcus pyogenes, other Streptococci, and Enterococcus
Clinical Manifestations
Acute Streptococcus pyogenes infections may take the form of pharyngitis, scarlet fever (rash), impetigo, cellulitis, or erysipelas. Invasive infections can result in necrotizing fasciitis, myositis and streptococcal toxic shock syndrome. Patients may also develop immune-mediated sequelae such as acute rheumatic fever and acute glomerulonephritis. S agalactiae may cause meningitis, neonatal sepsis, and pneumonia in neonates; adults may experience vaginitis, puerperal fever, urinary tract infection, skin infection, and endocarditis. Viridans streptococci can cause endocarditis, and Enterococcus is associated with urinary tract and biliary tract infections. Anaerobic streptococci participate in mixed infections of the abdomen, pelvis, brain, and lungs.
Structure
Streptococci are Gram-positive, nonmotile, nonsporeforming, catalase-negative cocci that occur in pairs or chains. Older cultures may lose their Gram-positive character. Most streptococci are facultative anaerobes, and some are obligate (strict) anaerobes. Most require enriched media (blood agar). Group A streptococci have a hyaluronic acid capsule.
Classification and Antigenic Types
Streptococci are classified on the basis of colony morphology, hemolysis, biochemical reactions, and (most definitively) serologic specificity. They are divided into three groups by the type of hemolysis on blood agar: b-hemolytic (clear, complete lysis of red cells), a hemolytic (incomplete, green hemolysis), and g hemolytic (no hemolysis). Serologic grouping is based on antigenic differences in cell wall carbohydrates (groups A to V), in cell wall pili-associated protein, and in the polysaccharide capsule in group B streptococci.
Pathogenesis
Streptococci are members of the normal flora. Virulence factors of group A streptococci include (1) M protein and lipoteichoic acid for attachment; (2) a hyaluronic acid capsule that inhibits phagocytosis; (3) other extracellular products, such as pyrogenic (erythrogenic) toxin, which causes the rash of scarlet fever; and (4) streptokinase, streptodornase (DNase B), and streptolysins. Some strains are nephritogenic. Immune-mediated sequelae do not reflect dissemination of bacteria. Nongroup A strains have no defined virulence factors.
Host Defenses
Antibody to M protein gives type-specific immunity to group A streptococci. Antibody to erythrogenic toxin prevents the rash of scarlet fever. Immune mechanisms are important in the pathogenesis of acute rheumatic fever. Maternal IgG protects the neonate against group B streptococci.
Epidemiology
Group A ß-hemolytic streptococci are spread by respiratory secretions and fomites. The incidence of both respiratory and skin infections peaks in childhood. Infection can be transmitted by asymptomatic carriers. Acute rheumatic fever was previously common among the poor; susceptibility may be partly genetic. Group B streptococci are common in the normal vaginal flora and occasionally cause invasive neonatal infection.
Diagnosis
Diagnosis is based on cultures from clinical specimens. Serologic methods can detect group A or B antigen; definitive antigen identification is by the precipitin test. Bacitracin sensitivity presumptively differentiates group A from other b-hemolytic streptococci (B, C, G); group B streptococci typically show hippurate hydrolysis; group D is differentiated from other viridans streptococci by bile solubility and optochin sensitivity. Acute glomerulonephritis and acute rheumatic fever are identified by anti-streptococcal antibody titers. In addition, acute rheumatic fever is diagnosed by clinical criteria.
Control
Prompt penicillin treatment of streptococcal pharyngitis reduces the antigenic stimulus and therefore prevents glomerulonephritis and acute rheumatic fever. Vancomycin resistance among the enterococci is an emerging microbial threat. Vaccines are under development.
Streptococcus pneumoniae
Clinical Manifestations
S pneumoniae causes pneumonia, meningitis, and sometimes occult bacteremia.
Structure
Pneumococci are lancet-shaped, catalase-negative, capsule-forming, a-hemolytic cocci or diplococci. Autolysis is enhanced by adding bile salts.
Classification and Antigenic Types
There are more than 85 antigenic types of S pneumoniae, which are determined by capsule antigens. There is no Lancefield group antigen.
Pathogenesis
S pneumoniae is a normal member of the respiratory tract flora; invasion results in pneumonia. The best defined virulence factor is the polysaccharide capsule, which protects the bacterium against phagocytosis.
Host Defenses
Protection against infection depends on a normal mucociliary barrier and intact phagocytic and T-independent immune responses. Type-specific anti-capsule antibody is protective.
Epidemiology
Pneumococcal pneumonia is most common in elderly, debilitated, or immunosuppressed individuals. The disease often sets in after a preceding viral infection damages the respiratory ciliated epithelium; incidence therefore peaks in the winter.
Diagnosis
Diagnosis is based on a sputum Gram stain and culture; blood or cerebrospinal fluid may also be cultured. Capsular antigen can be detected serologically. Pneumococci are distinguished from viridans streptococci by the quellung (capsular swelling) reaction, bile solubility, and optochin inhibition.
Control
Treatment is usually with penicillin. However, strains resistant to penicillin and multiple antibiotics are rapidly emerging. A vaccine is available.
INTRODUCTION
The genus Streptococcus, a heterogeneous group of Gram-positive bacteria, has broad significance in medicine and industry. Various streptococci are important ecologically as part of the normal microbial flora of animals and humans; some can also cause diseases that range from subacute to acute or even chronic. Among the significant human diseases attributable to streptococci are scarlet fever, rheumatic heart disease, glomerulonephritis, and pneumococcal pneumonia. Streptococci are essential in industrial and dairy processes and as indicators of pollution.
The nomenclature for streptococci, especially the nomenclature in medical use, has been based largely on serogroup identification of cell wall components rather than on species names. For several decades, interest has focused on two major species that cause severe infections: S pyogenes (group A streptococci) and S pneumoniae (pneumococci). In 1984, two members were assigned a new genus - the group D enterococcal species (which account for 98% of human enterococcal infections) became Enterococcus faecalis (the majority of human clinical isolates) and E faecium (associated with a remarkable capacity for antibiotic resistance).
In recent years, increasing attention has been given to other streptococcal species, partly because innovations in serogrouping methods have led to advances in understanding the pathogenetic and epidemiologic significance of these species. A variety of cell-associated and extracellular products are produced by streptococci, but their cause-effect relationship with pathogenesis has not been defined. Some of the other medically important streptococci are S agalactiae (group B), an etiologic agent of neonatal disease; E faecalis (group D), a major cause of endocarditis, and the viridans streptococci. Particularly for the viridans streptococci, taxonomy and nomenclature are not yet fully reliable or consistent. Important members of the viridans streptococci, normal commensals, include S mutans and S sanguis (involved in dental caries), S mitis (associated with bacteremia, meningitis, periodontal disease and pneumonia), and "S milleri" (associated with suppurative infections in children and adults). There remains persistent taxonomic confusion regarding "S milleri." These and other streptococci of medical importance are listed in Table 13-1 by serogroup designation, normal ecologic niche, and associated disease.
Clinical Manifestations
In humans, diseases associated with the streptococci occur chiefly in the respiratory tract, bloodstream, or as skin infections. Human disease is most commonly associated with Group A streptococci. Acute group A streptococcal disease is most often a respiratory infection (pharyngitis or tonsillitis) or a skin infection (pyoderma). Also medically significant are the late immunologic sequelae, not directly attributable to dissemination of bacteria, of group A infections (rheumatic fever following respiratory infection and glomerulonephritis following respiratory or skin infection) which remain a major worldwide health concern. Much effort is being directed toward clarifying the risk and mechanisms of these sequelae and identifying rheumatogenic and nephritogenic strains. S pneumoniae remains a primary cause of serious focal and systemic infections, the first most common cause of community acquired pneumonia in the United States and of fatal bacterial pneumonia in developing countries. Hemorrhagic shock in association with S pneumoniae sepsis in previously healthy children has been reported recently in the United States. Of major biologic importance is a renewed interest in safe and effective streptococcal vaccines.
Structure
Both S pyogenes and S pneumoniae are Gram-positive cocci, nonmotile, and nonsporulating; they usually require complex culture media. S pyogenes characteristically is a round-to-ovoid coccus 0.6-1.0 µm in diameter (Fig. 13-1). They divide in one plane and thus occur in pairs, or (especially in liquid media or clinical material) in chains of varying lengths. S pneumoniae appears as a 0.5-1.25 µm diplococcus, typically described as lancet-shaped but sometimes difficult to distinguish morphologically from other streptococci. Streptococcal cultures older than the logarithmic phase, which is the most active growth period of a culture, may lose their Gram-positive staining characteristics.
FIGURE 13-1 Morphology of the streptococci in comparison with staphylococci. Streptococci divide in a single plane and tend not to separate, causing chain formation. Capsules are antiphagocytic.
Unlike Staphylococcus (Chapter 12), all streptococci lack the enzyme catalase. Most are facultative anaerobes but some are obligate anaerobes. Streptococci often have a mucoid or smooth colonial morphology, and S pneumoniae colonies exhibit a central depression caused by rapid partial autolysis. As S pneumoniae colonies age, viability is lost during fermentative growth in the absence of catalase and peroxidase because of the accumulation of peroxide. Some group B and D streptococci produce pigment. Recently, nutritionally deficient streptococci (also known as wall-deficient, L form, thiol-requiring, satelliting, or pyridoxal-dependent) have been recovered from a variety of clinical sources, including blood, abscesses, and oral and urethral ulcers. These variants demonstrate bizarre pleomorphism microscopically and do not grow on routine subculture.
Classification, Antigenic Types and Extracellular Growth Products
The type of hemolytic reaction displayed on blood agar has long been used to classify the streptococci. b-Hemolysis is associated with complete lysis of red cells surrounding the colony, whereas a-hemolysis is a partial or "greening" hemolysis associated with reduction of red cell hemoglobin. Nonhemolytic colonies have been termed g-hemolytic. Hemolysis is affected by the species and age of red cells as well as by other properties of the base medium. Use of the hemolytic reaction in classification is not completely satisfactory. Some group A streptococci appear nonhemolytic; group B can manifest a-, b-, or even g-hemolysis; most S pneumoniae are a-hemolytic but can cause ß-hemolysis during anaerobic incubation. The viridans group, although linked by the property of a-hemolysis, is actually an extremely diverse group of organisms that does not usually react with Lancefield grouping sera. The taxonomy and biochemical and genetic relationships of these organisms continue to be clarified (Table 13-1).
Antigenic Types
The cell wall structure of group A streptococci is among the most studied of any bacteria (Fig. 13-2). The cell wall is composed of repeating units of N-acetylglucosamine and N-acetylmuramic acid, the standard peptidoglycan. For decades, the definitive identification of streptococci has rested on the serologic reactivity of cell wall polysaccharide antigens originally delineated by Rebecca Lancefield. Eighteen group-specific antigens were established. The group A polysaccharide is a polymer of N-acetylglucosamine and rhamnose. Some group antigens are shared by more than one species; no Lancefield group antigen has been identified for S pneumoniae or for some other a- or g-streptococci. With advances in serologic methods, other streptococci have been shown to possess several established group antigens.
FIGURE 13-2 Cell surface structure of S pyogenes and extracellular substances.
The cell wall also consists of several structural proteins (Figure 13-2). In group A streptococci, the R and T proteins may serve as epidemiologic markers, but the M proteins are clearly virulence factors associated with resistance to phagocytosis. More than 50 types of S pyogenes M proteins have been identified on the basis of antigenic specificity. Both the M proteins and lipoteichoic acid are supported externally to the cell wall on fimbriae, and the lipoteichoic acid, in particular, appears to mediate bacterial attachment to host epithelial cells. M protein, peptidoglycan, N-acetylglucosamine, and group-specific carbohydrate portions of the cell wall have antigenic epitopes similar in size and charge to those of mammalian muscle and connective tissue. Recently emerging strains of increased virulence are distinctly mucoid, rich in M protein and highly encapsulated.
The capsule of S pyogenes is composed of hyaluronic acid, which is chemically similar to that of host connective tissue and is therefore nonantigenic. In contrast, the antigenically reactive and chemically distinct capsular polysaccharide of S pneumoniae allows the single species to be separated into more than 80 serotypes. The antiphagocytic S pneumoniae capsule is the most clearly understood virulence factor of these organisms; type 3 S pneumoniae, which produces copious quantities of capsular material, are the most virulent. Unencapsulated S pneumoniae are avirulent. The polysaccharide capsule in S agalactiae allows differentiation into types Ia, Ib, Ic, II and III.
Finally, the cytoplasmic membrane of S pyogenes has antigens similar to those of human cardiac, skeletal, and smooth muscle, heart valve fibroblasts, and neuronal tissues, resulting in a molecular mimicry.
Extracellular Growth Products
The importance of the interaction of streptococcal products with mammalian blood and tissue components is becoming widely recognized. The soluble extracellular growth products or toxins of the streptococci, especially of S pyogenes (see Fig. 13-2), have been studied intensely. Streptolysin S is an oxygen-stable cytolysin; Streptolysin O is a reversibly oxygen-labile cytolysin. Both are leukotoxic, as is NADase. Hyaluronidase (spreading factor) can digest host connective tissue hyaluronic acid as well as the organism's own capsule. Streptokinases participate in fibrin lysis. Streptodornases A-D possess deoxyribonuclease activity; B and D possess ribonuclease activity as well. Protease activity similar to that in Staph aureus has been shown in strains causing soft tissue necrosis or toxic shock syndrome. This large repertoire of products may be important in the pathogenesis of S pyogenes by enhancing virulence; however, antibodies to these products appear not to protect the host even though they have diagnostic importance.
Three pyrogenic exotoxins of S pyogenes (SPEs) are recognized: types A, B, C. These toxins act as superantigens by a mechanism similar to those described for staphylococci, not requiring processing by antigen presenting cells. Rather, they stimulate T cells by binding class II MHC molecules directly and nonspecifically. With superantigens about 20% of T cells may be stimulated (vs 1/10,000 T cells stimulated by conventional antigens) resulting in massive detrimental cytokine release. When S pyogenes is lysogenized by certain bacteriophages, the SPEs A or C are produced; nonlysogenized strains are atoxic. SPE B is encoded by the bacterial chromosome. Re-emergence in the late 1980's of these exotoxin-producing strains has been associated with a toxic shock-like syndrome similar in pathogenesis and manifestation to staphylococcal toxic shock syndrome (Ch.12) and other forms of invasive disease associated with severe tissue destruction. SPE's have also been identified from non group A streptococci (groups B, C, F. G) in association with the toxic shock-like syndrome.
Virulence factors in the other streptococcal species, including the enterococci, are less well identified. In group B streptococci, carbohydrate surface antigens associated with antiphagocytosis have been identified, as has neuraminidase, which may play a role in pathogenesis. Among the viridans streptococci, production of the exopolysaccharide (glycocalyx) is associated with the ability to adhere to the cardiac valves and to form vegetations on the valve leaflets.
Pathogenesis
Streptococcus pyogenes and Streptococcus pneumoniae
Streptococci vary widely in pathogenic potential. Despite the remarkable array of cell-associated and extracellular products previously described (Fig.13- 2), no clear scheme of pathogenesis has been worked out. S pneumoniae and, to a lesser extent, S pyogenes are part of the normal human nasopharyngeal flora. Their numbers are usually limited by competition from the nasopharyngeal microbial ecosystem and by nonspecific host defense mechanisms, but failure of these mechanisms can result in disease. More often disease results from the acquisition of a new strain following alteration of the normal flora. S pyogenes causes inflammatory purulent lesions at the portal of entry, often the upper respiratory tract or the skin. Some strains of streptococci show a predilection for the respiratory tract; others, for the skin. Generally, streptococcal isolates from the pharynx and respiratory tract do not cause skin infections.
Invasion of other portions of the upper or lower respiratory tracts results in infections of the middle ear (otitis media), sinuses (sinusitis), or lungs (pneumonia). In addition, meningitis can occur by direct extension of infection from the middle ear or sinuses to the meninges or by way of bloodstream invasion from the pulmonary focus. Bacteremia can also result in infection of bones (osteomyelitis) or joints (arthritis).
S pyogenes (a group A streptococcus) is the leading cause of uncomplicated bacterial pharyngitis and tonsillitis (Fig. 13-3). Indeed, only group A streptococci are sought routinely in cases of pharyngitis, although groups B, C, and G are sometimes identified. S pyogenes infections can also result in sinusitis, otitis, mastoiditis, pneumonia with empyema, joint or bone infections, necrotizing fasciitis or myositis, and, more infrequently, in meningitis or endocarditis. S pyogenes infections of the skin can be superficial (impetigo) or deep (cellulitis). Although scarlet fever was formerly a severe complication of streptococcal infection, because of antibiotic therapy it is now little more than streptococcal pharyngitis accompanied by rash. Similarly, erysipelas, a form of cellulitis accompanied by fever and systemic toxicity, is less common today. There has, however, been an apparent recent increase in variety, severity and sequelae of S pyogenes infections. Because cases of streptococcal disease are not reported to national disease clearinghouses in the US, absolute numbers are not available. However, the recent resurgence of severe invasive infections has prompted descriptions of "flesh eating bacteria" in the news media. There has been no major change in susceptibility of S pyogenes to commonly used antibiotics but rather in the strain variations described above (antigenic types and extracellular growth products). However, a complete explanation for the decline and resurgence is not yet available.
FIGURE 13-3 Pathogenesis of S pyogenes infections.
The capsule of S pneumoniae renders it resistant to phagocytosis. The ability to evade this important host defense mechanism allows S pneumoniae to survive, multiply, and spread to various organs (Fig.13-4). The cell wall of S pneumoniae contains teichoic acid. The inflammatory response induced by Gram-positive cell walls differs from that induced by the endotoxin of Gram-negative organisms, but does include recruitment of polymorphonuclear neutrophils, changes in permeability and perfusion, cytokine release, and stimulation of platelet-activating factor. The role of other S pneumoniae moieties in virulence is less clear: protein A, pneumolysin, and peptide permeases. S pneumoniae is the leading cause of bacterial pneumonia beyond the neonatal period. Pleural effusion is the most common and empyema (pus in the pleural space) one of the most serious complications of S pneumoniae. This organism is also the most common cause of sinusitis, acute bacterial otitis media, and conjunctivitis beyond early childhood. Dissemination from a respiratory focus results in serious disease: outpatient bacteremia in children, meningitis, occasionally acute septic arthritis and bone infections in patients with sickle cell disease and, more rarely, peritonitis (especially in patients with nephrotic syndrome) or endocarditis.
FIGURE 13-4 Pathogenesis of S pneumoniae infections.
Postinfectious Sequelae
Infection with S pyogenes (but not S pneumoniae) can give rise to serious nonsuppurative sequelae: acute rheumatic fever and acute glomerulonephritis. These sequelae begin 1-3 weeks after the acute illness, a latent period consistent with an immune-mediated rather than pathogen-disseminated etiology. Whether all S pyogenes strains are rheumatogenic is still controversial; however, clearly not all are nephritogenic. These differences in pathogenic potential are not yet understood.
Acute rheumatic fever is a sequela only of pharyngeal infections, but acute glomerulonephritis can follow infections of the pharynx or the skin. Although there is no adequate explanation for the precise pathogenesis of acute rheumatic fever or for its failure to occur after streptococcal pyoderma, an abnormal or enhanced immune response seems essential. Also, persistence of the organism, due perhaps in part to the greater avidity with which the organism adheres to host pharyngeal cells, is associated with an increased likelihood of rheumatic fever. Acute glomerulonephritis results from deposition of antigen-antibody-complement complexes on the basement membrane of kidney glomeruli. The antigen may be streptococcal in origin or it may be a host tissue species with antigenic determinants similar to those of streptococcal antigen (cross-reactive epitopes for endocardium, sarcolemma, vascular smooth muscle). In the United States, the incidence of acute rheumatic fever had decreased dramatically. Although several areas reported a resurgence in cases in the late 1980's, subsequently, a slow, steady decline continued. Acute rheumatic fever can result in permanent damage to the heart valves. Less than 1% of sporadic streptococcal pharyngitis infections result in acute rheumatic fever; however, recurrences are common, and life-long antibiotic prophylaxis is recommended following a single case. The incidence of acute glomerulonephritis in the United States is more variable, perhaps due to cycling of nephritogenic strains, but appears to be decreasing; recurrences are uncommon, and prophylaxis following an initial attack is unnecessary.
Other Streptococcal Species
Lancefield Group Streptococci
Streptococcal groups B, C, and G initially were recognized as animal pathogens (see Table 13-1) and as part of the normal human flora. Recently, the pathogenic potential for humans of some of these non-group-A streptococci has been clarified. Group B streptococci, a major cause of bovine mastitis, are a leading cause of neonatal septicemia and meningitis, accounting for a significant changing clinical spectrum of diseases in both pregnant women and their infants. Mortality rates in full-term infants range from 2-8% but in pre-term infants are approximately 30%. Early-onset neonatal disease (associated with sepsis, meningitis and pneumonia at 6d life) is thought to be transmitted vertically from the mother; late-onset (from 7d to 3 mos age) meningitis is acquired horizontally, in some instances as a nosocomial infection. Group B organisms also have been associated with pneumonia in elderly patients. They are part of the normal oral and vaginal flora and have also been isolated in adult urinary tract infection, chorioamnionitis and endometritis, skin and soft tissue infection, osteomyelitis, meningitis, bacteremia without focus, and endocarditis. Infection in patients with HIV can occur at any age.
Streptococci of groups C and G are associated with mild, as well as severe human disease. None of these groups has been implicated in acute rheumatic fever or acute glomerulonephritis. Group D streptococci are important etiologic agents of urinary tract infections and infections associated with biliary tract procedures, as well as cases of disseminated infection, bacteremia, and endocarditis. Streptococcus bovis bacteremia has been recognized more often in cases of bowel disease.
Group F streptococci are associated with abscess formation and purulent disease. Group R streptococci, well-documented causes of meningitis and septicemia in pigs, also pose a serious health hazard to workers in the pork industry.
Viridans Streptococci
The biochemically and antigenically diverse group of organisms classified as viridans streptococci, as well as other non-groupable streptococci of the oral and gastrointestinal cavities and urogenital tract, include important etiologic agents of bacterial endocarditis. Dental manipulation and dental disease with the associated transient bacteremia are the most common predisposing factors in bacterial endocarditis, especially if heart valves have been damaged by previous rheumatic fever or by congenital cyanotic heart disease. S mutans and S sanguis are odontopathogens responsible for the formation of dental plaque, the dense adhesive microbial mass that colonizes teeth and is linked to caries and other human oral disease (see Ch. 99 ). S mutans is the more cariogenic of the two species, and its virulence is directly related to its ability to synthesize glucan from fermentable carbohydrates as well as to modify glucan in promoting increased adhesiveness.
Anaerobes
Like their aerobic counterparts, anaerobic streptococci are part of the normal flora, particularly of the mouth and intestinal tract; they are also part of the normal flora of the upper respiratory and genital tracts and the skin. These anaerobic organisms are linked to a wide variety of serious mixed infections of the female genital tract as well as to brain, pulmonary, and abdominal abscesses.
Host Defenses
The streptococci are part of the endogenous microbial flora of the nasopharynx. Disease may result from circumvention of the normal specific or nonspecific host defense mechanisms. More often, both S pyogenes and S pneumoniae are exogenous secondary invaders following viral disease or disturbances in the normal bacterial flora.
In the normal host, nonspecific defense mechanisms prevent organisms from penetrating beyond the superficial epithelium of the upper respiratory tract. These mechanisms include mucociliary movement and the cough, sneeze and epiglottal reflexes. The host phagocytic system is a second line of defense against pathogens.
Organisms can be opsonized by activation of the classical or alternate complement pathway or by specific immunoglobulin binding.
The capsules of both S pyogenes and S pneumoniae allow the organisms to evade opsonization. The hyaluronic acid outer surface of S pyogenes is only weakly antigenic; however, protective immunity results from the development of type-specific antibody to the M protein of the fimbriae, which protrude from the cell wall through the capsular structure. This antibody, which follows respiratory and skin infections, is persistent. Presumably, IgA in the respiratory secretions and serum IgG are the important protective antibody classes. S pyogenes is rapidly killed following phagocytosis enhanced by specific antibody. Prompt, effective antibiotic treatment of streptococcal infections may preclude development of this persistent antibody. Evidence has shown that antibody to the erythrogenic toxin involved in scarlet fever is also long lasting. This is the basis of the Dick test, an in vivo skin test, rarely used today, which measures host antitoxin. The capsular polysaccharides of S pneumoniae are highly antigenic and type-specific. Type-specific anticapsular antibodies to these T-independent antigens result in effective opsonization and host recovery. In untreated S pneumoniae infections, recovery clearly is due to opsonizing antibody. Even when adequate and appropriate antibiotic therapy is given, opsonizing antibody probably contributes significantly to recovery from pneumococcal disease. The normal host is somewhat resistant to S pneumoniae disease, but compromised hosts of several types are highly susceptible to serious infections: alcoholics, the semicomatose, very young, and very old individuals, patients who have undergone splenectomy, and patients with underlying diseases (specifically, chronic cardiac, pulmonary, or renal disease; sickle cell anemia; leukopenia; multiple myeloma; cirrhosis; and diabetes).
Cross-reactive antigens, especially of S pyogenes and various mammalian tissues, help explain the autoimmune responses that develop following some infections. The level of humoral response to infection with S pyogenes is greater in patients with rheumatic fever than in patients with uncomplicated pharyngitis. In addition, cell-mediated immunity may play a significant role in acute rheumatic fever.
Neonatal susceptibility to group B streptococci may result from immature neonatal phagocytic function, humoral immunity, or cell-mediated immunity, or from lack of passively acquired maternal antibody.
Evidence from the Rhesus monkey animal model in dental research shows that IgG may be a more important antibody class than IgA or IgM in protection against caries. Part of the reason may be that IgG is the antibody isotype most efficient at enhancing phagocytosis of S mutans. Cell-mediated immunity appears to participate in the protective host response against caries.
Epidemiology
The streptococci are widely distributed in nature and frequently form part of the normal human flora (see Table 29-1). Approximately 5-15% of humans carry S pyogenes or S agalactiae in the nasopharynx. S pneumoniae infects humans exclusively, and no reservoir is found in nature. The carrier rate of S pneumoniae in the normal human nasopharynx is 20-40%.
All ages, races, and sexes are susceptible to streptococcal disease. Because S pneumoniae is a particularly labile organism, sensitive to heat, cold, and drying, horizontal transmission requires close person-to-person contact. Infection is more likely at the extremes of life (<2 yr, > 65 yr), when host resistance is reduced, as described in the preceding section, or after the introduction of a more virulent strain. In the United States, pneumococcal disease is most prevalent during winter, coinciding with increased rates of acquisition but not necessarily of carriage. Alaskan natives have higher rates of invasive pneumococcal disease than do other American populations. The reason for this is unclear.
The incidence of respiratory disease attributed to S pyogenes peaks at about 6 years of age, and then again at 13 years of age, and is most common during late winter and early spring in temperate climates. Skin infections are more common among preschool-age children, and are most prevalent in late summer and early fall in temperate climates (when hot, humid weather prevails), and at all times in tropical climates. S pyogenes is spread by respiratory droplets or by contact with fomites used by the index individual, either patient or carrier. Skin infections often follow minor skin irritation, such as insect bites. There are occasional reports of streptococcal disease traced to rectal carriers, and of food-borne and vector-born outbreaks. In children, invasive disease with S pyogenes may follow varicella, or be associated with burns or malignancy; in adults with surgical or nonsurgical wounds or underlying medical problems, i.e., diabetes, cirrhosis, underlying peripheral vascular disease, or malignancy.
The world prevalence of the serious late sequelae of S pyogenes infections (acute rheumatic fever and acute glomerulonephritis) has shifted from temperate to tropical climates. In particular, acute rheumatic fever had ceased to be a major health concern in the US., despite no concomitant decline in group A streptococcal pharyngitis. These diseases previously affected persons with a low standard of living and limited access to medical care. Since 1985, there have been scattered outbreaks of acute rheumatic fever in some regions of the United States. Temporal and geographic clustering provides further evidence for "rheumatogenic" strains. Whether ethnic or racially determined factors affect this shift is not known.
Other streptococcal groups show striking epidemiologic features. An increasing prevalence of non-group-A as compared to group A streptococci in throats has been reported. Studies of the vaginal flora among women of child-bearing age show a S agalactiae carrier rate of 15-40%. Vertical transmission of the organisms to neonates of vaginally infected mothers ranges from 40-73%, but the incidence of neonates with disease (in contrast to colonized, healthy neonates) is low, 1-2%. S suis has been linked to meningitis among meat handlers. Isolation of S milleri or S bovis from the bloodstream should raise suspicion of immunosuppression or underlying disease visceral abscess formation or other bowel disease (including colon carcinoma).
In the United States, enterococci are the second most common nosocomial pathogenes associated with both endogenous colonization and patient-to-patient spread. A wide variety of infections results, especially urinary tract and surgical wound infections, with a marked propensity for antibiotic resistance. The widespread usage of newer cephalosporins, which have poor activity against enterococci, allows "break through" of enterococci as clinically significant isolates, the development of resistance in areas of heavy antibiotic use, and a selective advantage to these organisms.
Diagnosis
Clinical
It is not usually possible to diagnose streptococcal pharyngitis or tonsillitis on clinical grounds alone. Accurate differentiation from viral pharyngitis is difficult even for the experienced clinician, and therefore the use of bacteriologic methods is essential. However, distinguishing acute streptococcal pharyngitis from the carrier state may be difficult. When documented streptococcal pharyngitis is accompanied by an erythematous punctiform rash (Fig.13-4), the diagnosis of scarlet fever can be made. With streptococcal toxic shock syndrome, unlike staphylococcal toxic shock syndrome where the organism is elusive, there is often a focal infection or bacteremia. Criteria for diagnosis of streptococcal toxic shock syndrome include hypotension and shock, isolation of S pyogenes, as well as 2 or more of the following: ARDS, renal impairment, liver abnormality, coagulopathy, rash with desquamating soft tissue necrosis. The invasive, potentially fatal S pyogenes infections require early recognition, definitive diagnosis, and early aggressive treatment.
Rheumatic fever is a late sequela of pharyngitis and is marked by fever, polyarthritis, and carditis. A combination of clinical and laboratory criteria (Table 13-2) is used in the diagnosis of acute rheumatic fever. Since the original Jones criteria were published in 1944, these have been modified (1955), revised (1965, 1984) and updated (1992). The other late sequela, acute glomerulonephritis, is preceded by pharyngitis or pyoderma; is characterized by fever, blood in the urine (hematuria), and edema; and is sometimes accompanied by hypertension and elevated blood urea nitrogen (azotemia). Pneumococcal pneumonia is a life-threatening disease, often characterized by edema and rapid lobar consolidation.
Specimens For Direct Examination And Culture
S pyogenes is usually isolated from throat cultures. In cases of cellulitis or erysipelas thought to be caused by S pyogenes, aspirates obtained from the advancing edge of the lesion may be diagnostic. S pneumoniae is usually isolated from sputum or blood. Precise streptococcal identification is based on the Gram stain and on biochemical properties, as well as on serologic characteristics when group antigens are present. Table 13-3 shows biochemical tests that provide sensitive group-specific characteristics permitting presumptive identification of Gram-positive, catalase-negative cocci.
Identification
Hemolysis should not be used as a stringent identification criterion. Bacitracin susceptibility is a widely used screening method for presumptive identification of S pyogenes; however, some S pyogenes are resistant to bacitracin (up to 10%) and some group C and G streptococci (about 3-5%) are susceptible to bacitracin. Some of the group B streptococci also may be bacitracin sensitive, but are presumptively identified by their properties of hippurate hydrolysis and CAMP positivity. S pneumoniae can be separated from other a-hemolytic streptococci on the basis of sensitivity to surfactants, such as bile or optochin (ethylhydrocupreine hydrochloride). These agents activate autolytic enzymes in the organisms that hydrolyze peptidoglycan.
In many instances, presumptive identification is not carried further. Serologic grouping has not been performed as often as it might be because of the lack of available methods and the practical constraints of time and cost; however, only serologic methods, as listed in Table 13-4, provide definitive identification of the streptococci. The Lancefield capillary precipitation test is the classical serologic method. S pneumoniae, which lacks a demonstrable group antigen by the Lancefield test, is conventionally identified by the quellung or capsular swelling test that employs type-specific anticapsular antibody. Inspection of Gram-stained sputum remains a reliable predictor for initial antibiotic therapy in community-acquired pneumonia.
New methods for serogrouping that show sensitivity and specificity now are being explored. Organisms from throat swabs, incubated for only a few hours in broth, can be examined for the presence of S pyogenes using the direct fluorescent antibody or enzyme-linked immunosorbent technique. Additional rapid antigen detection systems for the group carbohydrate have become increasingly popular. However, the sensitivity (70-90%) of these currently available rapid tests for group A streptococcal carbohydrate does not allow exclusion of streptococcal pharyngitis without conventional throat culture (sensitivity of a single throat culture is 90-99%). A third generation assay, the optical immunoassay, is currently being evaluated. S pneumoniae can be identified rapidly by counterimmuno-electrophoresis, a modification of the gel precipitin method. The coagglutination test, described in Ch.12, is a more sensitive modification of the conventional direct bacterial agglutination test. The Fc portion of group-specific antibody binds to the protein A of dead staphylococci, leaving the Fab portion free to react with specific streptococcal antigen. The attachment of antibody to other carrier particles in suspension (for example, latex) also is used. The fact that whole streptococcal cells can be used in recently developed methods circumvents the difficulties involved in extracting components that retain appropriate antigenic reactivity. These newer serogrouping methods should make it more practical to identify not only b-hemolytic isolates from the blood or normally sterile sites, but also a-and nonhemolytic strains. It has become increasingly important to identify more of these strains to avoid simply misclassifying them as contaminants. Such information will expand our understanding of the importance of non-group-A streptococci.
Serologic Titers
Antibodies to some of the extracellular growth products of the streptococci are not protective but can be used in diagnosis. The antistreptolysin O (ASO) titer which peak 2-4 wks after acute infection and anti-NADase titers (which peaks 6-8 weeks after acute infection) are more commonly elevated after pharyngeal infections than after skin infections. In contrast, antihyaluronidase is elevated after skin infections, and anti-DNase B rises after both pharyngeal and skin infections. Titers observed during late sequelae (acute rheumatic fever and acute glomerulonephritis) reflect the site of primary infection. Although it is not as well known as the ASO test, the anti-DNase B test appears superior because high-titer antibody is detected following skin and pharyngeal infections and during the late sequelae. Those titers should be interpreted in terms of the age of the patient and geographic locale.
Although not used in diagnosis, bacteriocin production and phage typing of streptococci are employed in research and epidemiologic studies.
Control
Antibiotic Treatment
Penicillin remains the drug of choice for S pyogenes. It is safe, inexpensive, and of narrow spectrum, and there is no direct or indirect evidence of loss of efficacy. Prior to the 1990's, S pneumoniae was also uniformly sensitive to penicillin but a recent abrupt shift in the usefulness of penicillin has occurred. The group D enterococci are resistant to penicillins, including penicillinase-resistant penicillins such as methicillin, nafcillin, dicloxacillin, and oxacillin, and are becoming increasingly resistant to many other antibiotics. Group B streptococci are often resistant to tetracycline but remain sensitive to the clinically achievable blood levels of penicillin, even though they have penicillin minimal inhibitory concentrations (MIC) considerably higher than those of S pyogenes. Although the duration of penicillin therapy varies with the degree of invasiveness, streptococcal pharyngitis is generally adequately treated with 10 days of antibiotic therapy, and pneumococcal pneumonia with 7-14 days. If penicillin allergy occurs, an alternative drug for treating pharyngitis is erythromycin, although sporadic erythromycin and tetracycline resistance has been reported, leaving clindamycin or the newer macrolides as possible treatments. The most important goal of therapy in acute streptococcal pharyngitis is still to prevent rheumatic fever. However, therapy also hastens clinical recovery, avoids suppurative complications and renders the patient non-infectious for others. In addition to antibiotics, the patient with S pyogenes myositis or necrotizing fasciitis requires surgical debridement. Lifelong prophylaxis against recurrences of rheumatic fever is achieved with long-acting penicillin or erythromycin. Sulfonamides will not eradicate the streptococcus and thus are not acceptable therapy for streptococcal pharyngitis, but sulfadiazine is effective for preventing recurrent attacks of rheumatic fever. Additional prophylactic coverage before some dental and surgical procedures is necessary in the presence of rheumatic heart disease or prosthetic heart valves. Although streptococcal pharyngitis is usually a benign, self-limited disease, therapy is important to prevent rheumatic fever. There is no convincing evidence that antibiotic therapy prevents glomerulonephritis. Disconcertingly, some patients in recent outbreaks of acute rheumatic fever do not give a history of preceding pharyngitis.
Methods of treating the asymptomatic pharyngeal carrier of S pyogenes remain controversial. Recent evidence suggests that up to 20% of children and young adults are carriers, the carrier state involves no risk to the carrier or to others, and it is frequently difficult to eradicate despite the exquisite sensitivity of the organism to penicillin in vitro. A similar failure of antibiotic therapy to eradicate nasopharyngeal carriage or to prevent reinfection with S pneumoniae also occurs.
Although antibiotic resistance in S pneumoniae is common in many parts of the world, in the United States such strains previously had a geographically limited focus. Recent widespread emergence of S pneumoniae resistant to penicillin and other antibiotics has become a microbial threat in the United States as well. Even cefotaxime and ceftriaxone resistance has been documented. Isolates must be carefully screened for susceptibility by oxacillin disc testing, with definitive MIC determination by the E test (A B Biodisk NA, Piscataway, NJ), a convenient and reliable method for detection of resistance to penicillin and extended spectrum cephalosporins.
It is inappropriate to universally treat of pregnant women who are carriers of group B streptococci, or their colonized neonates, for several reasons: the high carrier rate; cost; the associated high risk of penicillin hypersensitivity; the potential increase in infections with penicillin-resistant organisms; the difficulty in altering colonization of women (even when their sexual partners were also treated); and the low risk of neonatal disease. The controversy continues despite recent recommendations for universal screening of preg nant women and selective intrapartum chemoprophylaxis for screen-positive mothers with preterm labor, premature or prolonged rupture of membranes, fever in labor, multiple births or previous infants with group B streptococcal disease.
Clearly, penicillin has reduced the severe morbidity and mortality associated with S pneumoniae. The emergence of resistance has now forced re-evaluation of empiric therapy. Clinicians must report clusters of S pneumoniae infection and be aware of local patterns of resistance. Penicillin susceptible organisms show MICs 0.06 mg/ml, intermediate strains 0.1-1.0 mg/ml and high level resistant strains 2 mg/ml. For nonmeningeal infection by intermediate strains, parenteral penicillin at high dose can probably be used since the mechanism of resistance involves alteration in penicillin binding proteins (PBP) and saturation. For meningeal infection with intermediate strains or any infection by high level resistant strains only ceftriaxone and cefotaxime retain sufficient activity. Resistance even to these extended spectrum cephalosporins was first reported for the US in 1991. At this writing only vancomycin remains uniformly effective but as discussed below, its use incurs potential for selection of vancomycin resistant enterococci (VRE) or risk of transferring vancomycin resistance from enterococci to S pneumoniae.
Currently, no single agent is reliably bactericidal against enterococci. Serious infections with group D enterococci often require a classic synergistic regime combining penicillin or ampicillin with an aminoglycoside, designed to weaken the cell wall with the ß-lactam and facilitate entry of the bacteriocidal aminoglycoside. Other ß-lactam drugs with good activity against enterococci include piperacillin and imipenem. An alternate drug of choice is vancomycin, but vancomycin-resistant strains of enterococci have been isolated. Nosocomial acquisition of these resistant organisms is of grave concern.
This antibiotic resistance among the streptococci/enterococci is an increasing problem. Studies show that in vitro exchange of resistant DNA can occur in conjugation via plasmids and transposons, or in transduction with bacteriophages. The mechanisms involved in the in vivo genetic exchange are not clearly defined. Evidence is accumulating that other streptococci may be the important donors of resistance markers. Transposon transfer is thought to be the most likely mechanism in S pneumoniae, although point mutations also occur. In the setting of heavy ß-lactam use, selective pressure is important in emergence of resistant strains. The first penicillin-resistant S pneumoniae were reported in 1967 in Australia and in 1974 in North America. In New Guinea, where the first penicillin-resistant strains were reported in 1971, one-third of S pneumoniae isolates from patients with severe pneumococcal disease were resistant by 1978. In Hungary in 1992, 69% of S pneumoniae isolates were penicillin resistant. This resistance is not ß-lactamase mediated but due to alteration in PBP which results in decreased binding of penicillin by the organism, rendering the drug less effective and requiring higher concentrations for saturation. Some strains resistant to erythromycin or tetracycline also have been reported, as well as some multiply resistant strains. In South Africa, outbreaks of infection with strains of S pneumoniae resistant to ß-lactam antibiotics (penicillins and cephalosporins) as well as to tetracycline, chloramphenicol, erythromycin, streptomycin, clindamycin, sulfonamides, and rifampin were reported in 1977. Although antibiotic resistance among S pneumoniae was infrequent in the United States, a major shift occurred from 1988 to 1990, resulting in the present situation of 15-25% of S pneumoniae intermediately or completely resistant to penicillin. Communities with "low prevalence" have 5-10% resistance. Single or multiply resistant strains are transmitted person to person, especially in settings of frequent salivary exchange, antibiotic use and hand-to-hand transmission (as in day care centers) or of crowding (corrections facilities, homeless shelters, nursing homes, military training groups). Control of the problem of emerging, antibiotic-resistant S pneumoniae is multifactorial: 1) surveillance for clusters of invasive disease, resistance and prevalent serotypes; 2) education of physicians and the public about antibiotic use (decrease unnecessary antibiotic use for obviously viral infections and decrease antibiotic prophylaxis for otitis by use of intermittent or expectant dosing or of non ß-lactam based prophylaxissulfa. Use topical treatment for impetigo, and short course therapies and narrow spectrum antibiotics); 3) adherence to infection control strategies in day care centers; 4) aggressive promotion of the current 23-valent S pneumoniae vaccine and support of efforts to design a new vaccine effective in those <2 years of age, analogous to the eminently successful Haemophilus influenzae type B vaccine (see Ch. 30) where bacterial polysaccharide is conjugated to protein to elicit a T cell-dependent response.
Among the enterococci, resistance to a wide variety of common antibiotics has emerged, with some strains resistant to all currently available antibiotics. There is no clinically proven treatment effective against enterococci multiply resistant to lactams, aminoglycosides, and vancomycin. The emergence of such organisms poses a stunning management dilemma. Resistance among the enterococci can be either intrinsic or acquired (by de novo genetic mutation or acquisition of DNA from resistant organisms). Enterococcal resistance to lactams is also mediated by altered PBP as in pneumococci, allowing cell wall synthesis even in the presence of antibiotic, or much less commonly by ß-lactamase. Resistance to aminoglycosides is mediated by decreased uptake or aminoglycoside modifying enzymes, and to vancomycin by decreased cell wall affinity for glycopeptide antibiotics. Further research into the mechanisms of resistance and new class(es) of antibiotics is essential.
A final concern about emerging resistance among the enterococci is the potential for genetic transfer of resistance genes to more virulent pathogens: Staph aureus, S pneumoniae and even Gram-negative organisms. So significant is this threat of emerging enterococcal resistance that the Centers for Disease Control and Prevention has issued a document addressing national guidelines. These include recommendations for 1) education of physicians and the public about the impact of vancomycin resistant enterococci (VRE), 2) vigilant surveillance for and detection of VRE, 3) strict enforcement of infection control strategies in hospitals, and 4) prudent vancomycin use or monotherapeutic use of extended spectrum cephalosporins. In a recent study of vancomycin use in US hospitals, use was about equally divided for treatment of a specific isolate, for prophylaxis, and for empiric coverage. The recommendations discourage vancomycin use for routine surgical prophylaxis, empiric prophylaxis in the patient with febrile neutropenia, the low birth weight infant or patients with vascular or peritoneal catheters, treatment of a single blood culture positive for coagulase-negative staphylococci, primary treatment of antibiotic-associated colitis, attempted eradication of colonization by methicillin-resistant Staph aureus (MRSA), or selective decontamination of the gastrointestinal tract.
Vaccination
As chemotherapeutic management becomes more difficult because of the threat of resistance, prevention becomes more important. With the introduction of antibiotics, previously successful pneumococcal vaccines fell into disuse. However, although prompt treatment with antibiotics has reduced the serious consequences of S pneumoniae infections (pre-antibiotic mortality rate of 30%), the disease incidence remains unchanged, and attention has been redirected to vaccines for S pneumoniae as well as for other streptococci. Pneumococcal vaccines (containing the pneumococcal polysaccharides of the most prevalent serotypes) have been licensed in several countries, including the United States. Initial use shows them to be useful and safe, but they remain under-utilized. The spectre of multidrug resistant S pneumoniae may provide a new incentive for their use. In 1983, the United States Food and Drug Administration licensed a vaccine containing 23 serotypes, representing coverage against nearly 89% of the pneumococcal isolates submitted to the CDC in the 1987-1988 National Surveillance Study. The population target of pneumococcal vaccines includes those at high risk for serious pneumococcal disease: the elderly (65 and older) and children (2 years of age and older) with sickle cell anemia, with an immunocompromised state (lymphoma, asplenia, myeloma, acquired immunodeficiency syndrome), with nephrotic syndrome, or with chronic cardiopulmonary disease. Vaccines for the other streptococci remain experimental.
Vaccine production for the streptococci presents several formidable problems. For both S pyogenes and S pneumoniae, a large number of serotypes must be included in effective vaccines since successful selection of a common epitope remains elusive. Continuing surveillance to determine prevalent serotypes is necessary to insure that the vaccine formulations remain appropriate. For S pyogenes, it is critical to determine rheumatogenic and nephritogenic strains to limit the required multivalency of the vaccines. Alternatively a newly described conserved portion of M protein is a distant goal. Toxicity has been associated with M protein preparations, but lack of immunogenicity in highly purified preparations of antigens is still a problem. With streptococcal vaccines, the potential risk of antigenic cross-reactivity with cardiac tissue and an associated increased risk of acute rheumatic fever must be appreciated.
In group B neonatal disease chemoprophylaxis does not appear as practical as vaccine control. Passive immunity in group B streptococcal neonatal infection appears protective. Polyvalent hyperimmune gamma globulin and human monoclonal IgM antibody which reacts with multiple serotypes are undergoing efficacy studies. Active immunization of pregnant women with undegraded sialic acid-containing polysaccharide group B antigens is another important aspect of control.
The streptococci are ubiquitous, and their significance in medicine is remarkable. Exciting advances are being made in diagnosis and in understanding the mechanisms of pathogenesis, as well as in control of these well-known organisms. Problems with antibiotic resistance must preclude complacency in dealing with these common pathogens.
REFERENCES
Awada A, van der Auwera P, Meunier F, et al. Streptococcal and enterococcal bacteremia in patients with cancer. Clin. Infect. Dis 1992; 15:33-48.
Bisno AL Medical progress: group A streptococcal infections and acute rheumatic fever. N. Engl. J. Med. 1991; 325:783-793.
Butler JC, Breiman RF, Campbell JF, et al. Pneumococcal polysaccharide vaccine efficacy: an evaluation of current recommendations. JAMA 1993; 270:1826-1831.
CDC. Addressing emerging infectious disease threats: a prevention strategy for the United States. Atlanta: US Dept Health & Human Serv. Pub Hlth Serv. CDC 1994. CDC. Preventing the spread of vancomycin resistance report from the Hospital Infection Control Practices Advisory Committee. Fed Reg. 1994; 59:25757-25763.
Coykendall AL. Classification and identification of the viridans streptococci. Clin Microbiol Rev 1989; 2:315-328.
Dillon HC. Post-streptococcal glomerulonephritis following pyoderma. Rev Infec Dis 1979; 1:935-943.
Denny FW, Wannamaker LW, Brink WR, et al. Prevention of rheumatic fever: treatment of the preceding streptococcal infection. JAMA 1950; 143:151-153.
Farley MM, Harvey RC, Stull T, et al. A population-based assessment of invasive disease due to group B streptococcus in nonpregnant adults. N. Engl. J. Med. 1993; 328:1807-1811.
Hoge CW, Schwartz B, Talkington DF, et al. The changing epidemiology of invasive group A streptococcal infections and the emergence of streptococcal toxic shock-like syndrome. JAMA 1993; 269:384-389.
Kaplan EL. Global assessment of rheumatic fever and rheumatic heart disease at the close of the century. Circulation 1993; 88:1964-1972.
McCracken G. Emergence of resistant Streptococcus pneumoniae: a problem in pediatrics. Pediatr. Infect. Dis. J. 1995; 14:424-428.
Moellering RC. Emergence of Enterococcus as a significant pathogen. Clin. Infect. Dis. 1992; 14:1173-1178.
Ruoff KL. Streptococcus anginosus ("Streptococcus milleri"). The unrecognized pathogen. Clin. Microbiol. Rev. 1988; 1:102-108.
Special Writing Group of the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease of the Council on Cardiovascular Disease in the Young, of the American Heart Association. Guidelines for the diagnosis of rheumatic fever. Jones criteria, 1992 update. JAMA 1992; 268:2069-2073.
Stevens DL, Tanner MH, Winship J, et al. Severe group A streptococcal infections associated with a toxic shock-like syndrome and scarlet fever toxin A. N Engl J Med 1989; 321:1-8.
Tanz RR, Poncher JR, Corydon KE, et al. Clindamycin treatment of chronic pharyngeal carriage of group A streptococci. J. Pediatr. 1991; 119:123-128
Wannamaker LW. Changes and changing concepts in the biology of group A streptococci and the epidemiology of streptococcal infections. Rev Infect Dis 1979; 1:967-973.
Wessels MR, Kasper DL. The changing spectrum of
group B streptococcal disease. N Engl J Med 1993; 328:1843-1844.
Zabriskie JB. Rheumatic fever: the interplay between host,
genetics and microbe. Circulation 1985; 7l:1077-1086.