Streptococcus

The group of nonpyogenic streptococci includes mostly alpha-hemolytic as well as

nonhemolytic and even beta-hemolytic streptococcal species from the large category of

viridans group streptococci. In a study of the genus Streptococcus based on sequence

comparisons of small-subunit (16S) rRNA genes, five species groups of viridans group

streptococci were demonstrated (59) in addition to the pyogenic group (beta-hemolytic,

large-colony formers). These nonpyogenic groups were designated the S. mitis group, the S.

anginosus group, the S. mutans group, the S. salivarius group, and the S. bovis group.

Several streptococcal species were not unequivocally assigned and remain ungrouped

(36, 70). Among alpha-hemolytic streptococci, S. pneumoniaecan be separated from other

streptococci of the viridans group through bile solubility and optochin susceptibility.

However, phenotypic characterization and taxonomic considerations place S.

pneumoniae into the S. mitis group (59). The relationship of S. pneumoniae to other species

of the S. mitis group is so close that 16S rRNA gene analysis reveals greater than 99%

identity to the nucleotide sequences of S. mitis and S. oralisand the current concept of

separate species in this group has been challenged (66). A novel, closely related species, S.

pseudopneumoniae, has recently been split off from S. pneumoniae, following DNA-DNA

hybridization studies and phenotypic characterization (1). Strains are nonencapsulated,

insoluble in bile, and optochin susceptible only when incubated in ambient air and may be

associated with chronic obstructive pulmonary disease (61).

S. mitis, S. sanguinis, S. parasanguinis, S. gordonii, S. cristatus, S. oralis, S. infantis, S.

peroris (60), S. australis(131), S. oligofermentans (118), S. massiliensis (46), S. sinensis, S.

orisratti, S. pseudopneumoniae, and S. pneumoniae are members of the S. mitis group.

These species form a group whose classification and nomenclature have been a source of

considerable confusion in the past. Among the changes that were made are the

reclassification of the original S. mitis type strain as an S. gordonii strain and the subsequent

replacement by a new S. mitis type strain (NCTC12261T) (65). Based on phenotypic

reactions (especially arginine hydrolysis and the esculin test), the S. mitis group can be

further subdivided into the S. sanguinis and the S. mitis groups (36), but based on 16S rRNA

analysis, these two groups appear to belong together (59). Since correlation of the renamed

streptococcal species with human infections is still difficult, we chose to present these

species as part of the S. mitis group until further information becomes available.

The small-colony-forming S. anginosus group consists of the three distinct species S.

anginosus, S. constellatus,and S. intermedius (128). It includes streptococcal species

previously referred to as Lancefield group F streptococci, “S. milleri” group or “S.

milleri,” but “S. milleri” has no standing taxonomically. The S. mutansgroup comprises the

species S. mutans, S. sobrinus, S. criceti, S. ratti, S. downei, and numerous species that

have only been identified from animals thus far (S. ferus, S. macacae, S. hyovaginalis, and

S. devriesei).

The human species S. salivarius, S. vestibularis, and S. thermophilus, which is found in dairy

products, belong to the S. salivarius group. The whole S. salivarius group is closely related to

the S. bovis group. Some streptococcal species that are currently part of the S.

bovis group (S. infantarius and S. alactolyticus) (103) were formerly part of the S.

salivarius group (36).

The S. bovis group has experienced extensive taxonomic changes (30, 103, 104). These

changes were made because DNA-DNA reassociation studies revealed considerable

heterogeneity among the human isolates described as S. bovis biotypes. Four DNA clusters

are currently recognized. DNA cluster I consists of animal strains of S. bovis and S.

equinus, which were shown to belong to a single species. The earlier species name S.

equinus has been formally adopted. DNA cluster II consists of S. gallolyticus, with three

subspecies: subsp.gallolyticus (formerly S. bovis biotype I), subsp. pasteurianus (formerly S.

bovis biotype II.2), and subsp.macedonicus. DNA cluster III consists of S.

infantarius (formerly S. bovis biotype II.1), with two subspecies: subsp. infantarius and

subsp. coli (formerly called S. lutetiensis). DNA cluster IV consists of S. alactolyticus.

DESCRIPTION OF THE GENUS Back to top

Bacterial species belonging to the genus Streptococcus are catalase-negative, gram-positive

cocci of less than 2 μm that tend to grow in chains in liquid media. Most species of the

genus Streptococcus have a low G+C content of DNA ranging between 34 and 46%. The cell

wall composition is typical for gram-positive bacteria and consists mainly of peptidoglycan

with glucosamine and muramic acid as amino sugars and galactosamine as a variable

component. A variety of carbohydrates, surface protein antigens, and teichoic acid are

attached to the cell wall and are, among other characteristics, responsible for intra- and

interspecies differences among streptococci. Streptococci are facultative anaerobic bacteria.

Due to a lack of heme compounds, streptococci are incapable of respiratory metabolism.

Some species of the viridans streptococcal group and S. pneumoniaerequire 5% CO2 levels

for adequate growth, and the growth of many streptococcal species is enhanced in the

presence of 5% CO2. The optimum temperature for growth of most streptococci is around

37°C, while some species like S. uberis also grow at temperatures as low as 10°C. The

complex nutritional requirements of streptococci are usually provided by the addition of

blood or serum to the growth medium. Glucose and other carbohydrates are metabolized

fermentatively with lactic acid as the major metabolic end product. The addition of glucose or

other carbohydrates to liquid medium enhances growth but lowers the pH, resulting in

growth inhibition unless the medium is highly buffered (e.g., Todd-Hewitt broth [THB]).

Leucine aminopeptidase (LAP) is produced by all streptococci and enterococci but can also be

found in lactococci, pediococci, and other catalase-negative gram-positive cocci. It helps to

distinguish these species from the LAP-negative Aerococcus species and Leuconostoc. All

streptococci are catalase negative upon exposure to 3% hydrogen peroxide with the

exception of S. didelphis, a veterinary pathogen (102). False-positive catalase reactions may

occur if bacteria are grown on blood-containing media.

EPIDEMIOLOGY AND TRANSMISSION Back to top

Streptococci can cause infections in humans and in many different animal species including

mammals and fish. Different species of streptococci are frequently found as commensal

bacteria on mucous membranes. Occasionally streptococci are present as transient skin

microbiota. Several species exhibit a high virulence potential, but even the highly pathogenic

streptococcal species are frequently found as colonizing strains. S. pneumoniae was

responsible for approximately 42,000 invasive infections in the United States in 2007,

leading to an estimated 4,500 deaths (http://www.cdc.gov/abcs), and was found as a

colonizing bacterial species in many asymptomatic carriers. The asymptomatic carriage rate

for S. pneumoniae differs considerably between children and adults. Detection rates of 30 to

70% have been reported for young children, depending on the sampling method, while

carriage rates among healthy adults are often reported to be below 5% (48, 97).

Significantly higher colonization rates for adults living in households with preschool children

suggests the occurrence of household transmission between parents and their children (53).

Due to active bacterial surveillance in the emerging infections program network (107),

reliable epidemiologic data on invasive infections due to S. pneumoniae (described

above), S. pyogenes (group A), and S. agalactiae(group B) have been obtained for a

population of almost 30 million people in the United States during 2007. National estimates

are that S. pyogenes caused 11,400 cases of invasive disease and 1,350 deaths, with the

peak of infections in people older than 65 years. Invasive infections due to S.

agalactiae were second to those due to S. pneumoniae, with an estimated 20,000 cases and

1,475 deaths in 2007. Reflecting the ongoing changes in the epidemiology of group B

streptococcal disease, the highest attack rates were observed in patients less than 1 year

and adults greater than 65 years of age. Apart from causing invasive infections, pyogenic

streptococci are frequently encountered as colonizing strains. While asymptomatic

pharyngeal colonization with S. pyogenes occurs in less than 5% of the adult population, S.

agalactiae colonization rates of the urogenital and gastrointestinal tracts can be

demonstrated in 10 to 30% of the female as well as the male population. No significant

differences are observed in the colonization rates of pregnant and nonpregnant women.

Transmission of streptococcal infections can occur by different routes. Pathogenic species

like S. pyogenes andS. pneumoniae are primarily transmitted through droplets or direct

contact. Transmission can first lead to colonization, with the potential for the development of

a subsequent infection. Transmission from mother to child is typical for neonatal invasive S.

agalactiae infections. Newborns acquire the bacteria usually during delivery, although

transmissions, shortly after birth, from the mother or health care personnel to infants have

been documented, especially in late-onset neonatal infections. Endogenous infections most

often occur by viridans group streptococci as part of the oral microbiota. The tooth decaycausing

species S. mutans is also transmitted from mother to child during early infancy,

most probably through oral secretions.

Streptococcal infections do not represent classical zoo noses, although most species have a

preferred host. While occasional animal-to-human transmissions do occur, as in the case

of Streptococcus suis, genotypic and phenotypic analyses of animal and human strains

demonstrated that most strains causing human infections were distinct from the strains

causing animal infections. For large-colony-forming group C and G streptococci, such an

analysis led to an important change in species designations (120, 122). Currently all betahemolytic

group C and L and human group G streptococci are defined as S.

dysgalactiae subsp. equisimilis, while alpha-hemolytic group C streptococcal animal isolates

are classified as S. dysgalactiae subsp. dysgalactiae and animal group G streptococcal strains

as S. canis (17). Other closely related veterinary species are S. equi subsp. equiand S.

equi subsp. zooepidemicus. The predominant reservoir for S.

dysgalactiae subsp. equisimilis strains is the human host, and transmission usually occurs

among humans.

CLINICAL SIGNIFICANCE Back to top

Streptococcus pyogenes (Group A Streptococci)

S. pyogenes colonizes the human throat and skin and has developed complex virulence

mechanisms to avoid host defenses (23, 33). The upper respiratory tract and skin lesions

serve as primary focal sites of infections and principal reservoirs of transmission. S.

pyogenes can cause superficial or deep infections due to toxin-mediated and immunologically

mediated mechanisms of disease. S. pyogenes is the most common cause of bacterial

pharyngitis and impetigo. In the past, S. pyogenes was a common cause of childbed fever or

puerperal sepsis. S. pyogenes is responsible for deep or invasive infections, especially

bacteremia, sepsis, deep soft tissue infections, such as erysipelas, cellulitis, and necrotizing

fasciitis. Less common presentations include myositis, osteomyelitis, septic arthritis,

pneumonia, meningitis, endocarditis, pericarditis, and severe neonatal infections following

intrapartum transmission. One or more erythrogenic exotoxins produced by S. pyogenes may

cause a confluent erythemathous sandpaper-like rash characteristic of scarlet fever. While

systemic toxic effects occur rarely with scarlet fever, severe clinical manifestations in

streptococcal toxic shock syndrome (STSS) may result from massive superantigen-induced

cytokine and lymphokine production. Nonsuppurative complications include poststreptococcal

glomerulonephritis (GN) and acute rheumatic fever (ARF). While either of these conditions

may follow pharyngitis, only GN is linked with skin infections due to S. pyogenes. S.

pyogenes has also been associated with pediatric autoimmune neuropsychiatric disorders

(114).

The causes of the emergence of STSS, frequently accompanied by necrotizing fasciitis, and

the resurgence of invasive S. pyogenes infections since the mid-1980s are mostly

unexplained (112). S. pyogenes remains exquisitely sensitive to penicillin. Despite the

continuous exposure and subsequent type-specific immunity, the most prevalent M types

associated with STSS continue to be M1 and M3, together accounting for approximately 50%

of invasive infections. Since identical strains have accounted for less serious infections (86),

host factors and comorbid conditions account for different diseases. The incidence of STSS

seems to be highest among young children, particularly those with varicella, and the elderly.

Other persons at risk include those with diabetes mellitus, chronic cardiac or pulmonary

diseases, HIV infection, and intravenous drug or alcohol abuse. The risk for severe invasive

infection in contacts has been estimated to be 200 times greater than for the general

population, but most contacts are asymptomatically colonized (26).

Streptococcus agalactiae (GBS)

S. agalactiae was first identified as the cause of bovine mastitis at the end of the 19th

century. Since the 1970s it has been reported as the cause of invasive neonatal infections.

Neonatal infections present as two different clinical entities: early-onset neonatal disease,

characterized by sepsis and pneumonia within the first 7 days of life; and late-onset disease

with meningitis and sepsis between day 7 and 3 months of age. The most important risk

factor for the development of invasive neonatal disease is the colonization of the maternal

urogenital or gastrointestinal tract by S. agalactiae, which is found in 10 to 30% of pregnant

women. Prevention of early-onset neonatal infections can be achieved in the majority of

cases by administration of intrapartum antibiotic prophylaxis starting at least 4 hours before

delivery. Official CDC recommendations for the prevention of neonatal S.

agalactiae infections were first issued in 1996, were revised in 2002 (105), and resulted in a

substantial decline of early-onset neonatal group B streptococcus (GBS) disease (106).

InvasiveS. agalactiae infections of adult patients may be observed as postpartum infections

or in immunocompromised adult patients with alcoholism, diabetes mellitus, cancer, or HIV

infection (41). The spectrum of infections in adult patients includes pneumonia, bacteremia,

meningitis, endocarditis, urinary tract infections, skin and soft tissue infections, and

osteomyelitis.

Streptococcus dysgalactiae subsp. equisimilis (Human Group C

and G Streptococci)

Human isolates of large-colony-forming beta-hemolytic streptococci harboring the Lancefield

group C or group G antigens belong to this novel species (120, 122). While most isolates of

this species possess either the Lancefield group C or the group G antigen, strains harboring

the Lancefield group L as well as the group A antigen (16) have been described. The clinical

spectrum of disease caused by S. dysgalactiae subsp. equisimilisresembles infections caused

by S. pyogenes (17). The responsible strains harbor genes similar to virulence factor genes

of S. pyogenes, such as emm-like genes, and can be isolated from upper respiratory tract

infections, skin infections, soft tissue infections, and invasive infections such as necrotizing

fasciitis, STSS, bacteremia, and endocarditis. However, convincing reports of scarlet fever

due to S. dysgalactiae subsp.equisimilis have so far not been published. Similar to what is

observed with S. pyogenes, cases of GN and ARF have been reported (4, 52) following S.

equi subsp. zooepidemicus (GN) and S. dysgalactiae subsp. equisimilisinfections (GN and

ARF).

Streptococcus pneumoniae

S. pneumoniae is described separately in this section due to its clinical features that

distinguish it from other species of the S. mitis group. S. pneumoniae is the most frequently

isolated respiratory pathogen in community-acquired pneumonia. In as many as 30% of

community-acquired pneumonia cases, S. pneumoniaecan be found in blood cultures of

patients. S. pneumoniae is also a major cause of meningitis, leading to high morbidity and

mortality in pediatric and adult patients. The most frequently observed infection due to S.

pneumoniae is otitis media, with an estimate of one infection for every child up to the age of

6 years in the United States. Other infections due to S. pneumoniae include sinusitis,

peritonitis, and rare cases of endocarditis. S. pneumoniae colonizes the upper respiratory

tract, especially in children, without evidence of infection. Prevention of pneumococcal

infections can be achieved by immunization with a 23-valent capsular polysaccharide vaccine

in adults or the 7-valent conjugate vaccine in children. Conjugate vaccines for children

including a larger number of serotypes have recently been released in Europe. Widespread

use of these vaccines has resulted in a reduction of invasive pneumococcal infections during

the past several years but also in changes of the serotypes responsible for invasive and

noninvasive infections (20, 62, 99).

Streptococcus mitis Group

S. mitis, S. sanguinis, S. parasanguinis, S. gordonii, S. cristatus, S. oralis, S. infantis, S.

peroris, S. australis, S. sinensis, S. orisratti, S. oligofermentans, S. massiliensis, S.

pseudopneumoniae, and S. pneumoniae are members of this group. Members of the S.

mitis group are regular commensals of the oral cavity, the gastrointestinal tract, and the

female genital tract. The S. mitis group can be found as a transient microbiota of the normal

skin and may represent contaminants when isolated from blood cultures. At the same time,

these species are the most frequently isolated bacteria in bacterial endocarditis in native

valve and, less frequently, in prosthetic valve infections. Careful evaluation of the clinical

situation is therefore crucial to correctly interpret the clinical significance of blood culture

isolates from the S. mitis group. In neutropenic patients, streptococcal species from the S.

mitis group are often responsible for life-threatening sepsis and pneumonia cases following

immunosuppression by chemotherapy (15). Treatment of these infections is further

complicated by high penicillin resistance rates.

Streptococcus anginosus Group

Species from the S. anginosus group (S. anginosus, S. constellatus, and S. intermedius) are

often harmless commensals of the oropharyngeal, urogenital, and gastrointestinal

microbiota. However, these organisms are strongly associated with abscess formation in the

brain, oropharynx, or peritoneal cavity. A subspecies of S. constellatus, S.

constellatus subsp. pharyngis, has also been described and associated with pharyngitis

(130). All species of this group are small-colony-forming bacteria (colony size, ≤0.5 mm)

that can display variable patterns of hemolysis (alpha, beta, or gamma). Since they can also

harbor the Lancefield group antigen A, C, F, or G (or none at all), it is especially important to

reliably distinguish them from large-colony-forming (>0.5 mm) beta-hemolytic streptococci

of the pyogenic group. Association of certain species with specific isolation sites has been

reported. While S. anginosus is frequently found in specimens from the urogenital or

gastrointestinal tracts, S. constellatus is commonly isolated from the respiratory tract, and S.

intermedius is most often identified in abscesses of the brain or liver.

Streptococcus salivarius Group

Streptococcal species that belong to the S. salivarius group include S. salivarius and S.

vestibularis. They have been primarily isolated from the oral cavity and blood. Another

species of this group, S. thermophilus, is found only in dairy products. S. salivarius has been

repeatedly reported as a cause of bacteremia, endocarditis, and meningitis (sometimes

iatrogenic), while S. vestibularis has not been clearly associated with human infection.

Isolation of S. salivarius from blood cultures does correlate to some extent with intestinal

neoplasia (101).

Streptococcus mutans Group

S. mutans and S. sobrinus belong to the S. mutans group. They are the most commonly

isolated species of the group that originate from human clinical specimens, usually obtained

from the oral cavity. S. criceti, S. ratti, andS. downei have occasionally been identified from

human sources, while the other streptococcal species of theS. mutans group (S. ferus, S.

macacae, S. hyovaginalis, and S. devriesei) have only been identified in animals. S.

mutans is the primary etiologic agent of dental caries, and infection is transmissible. By 18

years of age, 85% of the population have at least one caries lesion (111). Permanent

colonization with S. mutans occurs under normal living conditions in the Western world

between the second and the end of the third year of life (111). Molecular analysis of mother

and infant isolates reveals that strains are usually acquired from the mother and that the

colonization rate of infants depends on the bacterial load of the mother (19). Analyses of

streptococcal blood culture isolates show that S. mutans is the most frequently isolated

species of this group in cases of bacteremia (36).

Streptococcus bovis Group

Extensive taxonomic changes have occurred in the S. bovis group, and strains formerly

known as human S. bovis isolates are designated as different species (see “Taxonomy”

above). The group now includes S. equinus, S. gallolyticus, S. infantarius, and S.

alactolyticus. Species from this group are frequently encountered in blood cultures of

patients with bacteremia, sepsis, and endocarditis. The clinical significance of blood cultures

growing streptococci from the S. bovis group lies in the association of S.

gallolyticus subsp. gallolyticuswith gastrointestinal disorders including colon cancer and

chronic liver disease and S. gallolyticus subsp.pasteurianus with meningitis (6, 43, 69, 88).

Other Streptococci Infrequently Isolated from Human

Specimens

Streptococcal species that are primarily animal pathogens are sometimes isolated from

human hosts, in most cases from humans that are in close contact with animals. S. suis, S.

porcinus, and S. iniae belong to this category. S. suis is a swine pathogen that has

occasionally been isolated from cases of human meningitis and bacteremia. S. suis is

encapsulated and appears to be alpha-hemolytic on sheep blood agar plates, although some

strains are beta-hemolytic on horse blood agar. S. suis strains are positive for the Lancefield

group antigen R, S, or T, which helps to distinguish them from the phenotypically similar

species S. gordonii, S. sanguinis, and S. parasanguinis. Similar to S. suis, S.

porcinus (Lancefield groups E, P, U, and V) is primarily a swine pathogen. Beta- hemolytic S.

porcinus strains have rarely been isolated from human sources such as peripheral blood,

wounds, and the female genital tract (37). A recent molecular study of S. porcinus isolates

from the female genital tract, however, indicates that these isolates belong to a novel

species designated S. pseudoporcinus (8). S. porcinus can be misidentified as S.

agalactiae due to its isolation from the female genital tract, false-positive reactions with

commercially available group B antisera, and a positive CAMP test reaction.S.

porcinus and S. pseudoporcinus can be L-pyrrolidonyl-β-naphthylamide (PYR) positive and do

not hydrolyze hippurate, in contrast to S. agalactiae. S. iniae is a fish pathogen that is betahemolytic

but does not possess any Lancefield group antigens. It has been isolated from soft

tissue infections, bacteremia, endocarditis, and meningitis in people handling fish

(38, 125). S. iniae isolates resemble S. pyogenes strains due to the fact that both are PYR

positive. Beta-hemolysis of the species can be observed only around agar stabs or under

anaerobic culture conditions. Commercial identification systems do not correctly identify the

species; the failure to react with Lancefield group antisera is important to notice, since it is

rare among beta-hemolytic streptococci.

COLLECTION, TRANSPORT, AND STORAGE OF

SPECIMENS Back to top

Specimens suspected of harboring streptococci should be collected by methods outlined

elsewhere in thisManual (chapter 16). Since many streptococcal species lose viability fairly

quickly, it is best to place swabs in an appropriate moist transport medium and process

specimens rapidly. If transport time is below 1 to 2 hours, a special transport system is not

absolutely necessary. S. pyogenes can safely be transported on dry swabs; desiccation

enhances recovery from mixed cultures by suppression of the accompanying microbiota (77).

Detailed recommendations for collection and storage of swabs from pregnant women to

detect S. agalactiaecolonization have been issued by the U.S. Centers for Disease Control

and Prevention. These recommendations are summarized below under “Special Procedures

for Streptococcus agalactiae Screening.”

DIRECT EXAMINATION Back to top

Microscopy

Microscopic examination shows streptococci as gram-positive bacteria growing in chains of

varying length. S. pneumoniae organisms most often present as gram-positive diplococci

with an elongated appearance, but a reliable microscopic distinction of S. pneumoniae from

enterococci and other streptococci is not possible. In blood culture specimens, S.

pneumoniae tends to form chains of varying length, similar to other streptococci. Direct

identification of streptococci by microscopic methods is most helpful in the case of clinical

specimens from sterile body sites, such as cerebrospinal fluid (CSF). Tiny, irregular cocci in

clumps of chains seen in abscess- or peritonitis-associated aspirates are suggestive of the S.

anginosus group. Interpretation of Gram stain results from nonsterile body sites is difficult,

due to the residential microbiota that frequently includes streptococci. Thus, for example,

throat swabs should not be examined by Gram stain for diagnosis of “strep throat.”

Direct Antigen Detection of S. pyogenes from Throat

Specimens

S. pyogenes is the most common cause of acute pharyngitis and accounts for 15 to 30% of

cases of acute pharyngitis in children and 5 to 10% of cases in adults. If diagnosis can be

provided rapidly, antibiotic therapy can be initiated promptly to relieve symptoms, to avoid

sequelae, and to reduce transmission. Numerous assays for direct detection of the group Aspecific

carbohydrate antigen in throat swabs by agglutination methods or immunoassays

(enzyme, liposome, or optical), also referred to as “rapid antigen assays,” have become

commercially available during the past 2 decades. A list of FDA-cleared tests is accessible via

the Internet

(http://www.accessdata.fda.gov/scripts/cdrh/devicesatfda/index.cfm?Search_Term=866.374

0). Although these tests provide rapid results and allow early treatment decisions, the throat

culture remains the gold standard. Sensitivities of rapid antigen tests range from 58% to

96% and have never equalled that of culture (40, 119). Negative rapid antigen test results

should therefore be confirmed by culture in children and adolescents, when typical clinical

signs are present (13). The specificity, however, is generally high, even though false-positive

antigen results are seen from patients previously diagnosed and/or treated for S.

pyogenes infection (21). Moreover, the low positive predictive value of rapid group A antigen

tests in the adult population frequently results in the prescribing of unnecessary antimicrobial

therapy (89).

Antigen Detection of S. agalactiae in Urogenital Tract Samples

Several different commercially available antigen detection tests have been developed for the

identification ofS. agalactiae in samples from the urogenital tract. Independent from the

technique involved (latex agglutination, enzyme immunoassay, or optical immunoassay), all

of the currently available tests lack sufficient sensitivities to detect bacterial colonization

with S. agalactiae (115). They are not recommended by the CDC for screening of pregnant

women. Even though modified protocols with an incubation of the samples in selective broth

prior to antigen testing appear to increase assay sensitivities, the current CDC

recommendations rely on selective broth culture performed at 35 to 37 weeks of gestation

for this purpose (see “Special Procedures forStreptococcus agalactiae Screening” under

“Isolation Procedures” below).

Antigen Detection of S. pneumoniae in Urine Samples

An immunochromatographic membrane test relying on the detection of the cell wallassociated

polysaccharide that is common to all S. pneumoniae serotypes (C- polysaccharide

antigen) (Binax NOW; Binax Inc., Portland, ME) has proven helpful for the identification of S.

pneumoniae infections in adult patients, especially in patients that already received antibiotic

treatment. Compared to conventional diagnostic methods, reported sensitivities of antigen

detection in urine samples range between 50 and 80% and specificities are approximately

90% (49, 83). Due to the fact that the test is also positive in S. pneumoniae carriage without

infection, as is often observed among infants (32), it is of limited value in pediatric patients.

The test should not be used for children below the age of 6 years (32), and comprehensive

studies on schoolchildren with lower colonization rates have not been performed. It can

currently only be recommended in adults as an addition to conventional diagnostic culture

techniques for S. pneumoniae (75) and is probably most helpful for patients who received

antimicrobial treatment before cultures were obtained.

Streptococcal Antigen Detection in CSF

Commercially available antigen detection tests for the diagnosis of pathogenic

microorganisms in CSF samples include reagents for the detection of S. agalactiae and S.

pneumoniae. These tests have also been used on positive blood culture specimens. The tests

are not recommended for routine use, as the results should not be used to change decisions

about empiric therapy based on clinical and laboratory criteria (116). It has also been shown

that the sensitivity of direct antigen detection in CSF is low (<30%) and offers no advantage

over a conventional cytospun Gram stain (78). However, very promising results have been

published for the use of the S. pneumoniae urinary antigen test on CSF samples (81).

Nucleic Acid Detection Techniques

S. pyogenes

A rapid method for the detection of S. pyogenes in pharyngeal specimens is based on a

single-stranded chemiluminescent nucleic acid probe assay to identify specific rRNA

sequences (Group A Streptococcus Direct Test; Gen-Probe, Inc., San Diego, CA). This test

performed well in comparative studies with the culture technique. Sensitivity and specificity

for the probe test ranged from 89% to 95% and 98% to 100%, respectively, compared to

the results of the culture technique with a sensitivity of 98% to 99% (21, 92). These data

suggest that the probe test may be suitable as a primary test or as a backup test to negative

antigen tests, particularly for batch screening of throat cultures. Moreover, a real-time PCR

assay (LightCycler Strep A assay; Roche Diagnostics, Indianapolis, IN) has been recently

developed for the detection of S. pyogenes in throat swabs (119). Real-time PCR proved to

be more sensitive than the standard culture method (119), and it appears to be unnecessary

to perform cultures when results of the real-time PCR are negative. Real-time PCR allows the

detection of beta-hemolytic species S. pyogenes and S. dysgalactiae subsp. equisimilis.

S. agalactiae

A rapid method for the detection of S. agalactiae colonization in pregnant women at the time

of the delivery has recently been developed and evaluated (11). The test is based on the

detection of the S. agalactiae cfbgene (91) by a fluorogenic real-time PCR assay (BD

GeneOhm StrepB; Becton Dickinson, Sparks, MD), and results can be obtained in a few

hours with a reported sensitivity of 94% and specificity of 95.9% (27). The test has been

evaluated and approved by the FDA for rectal/vaginal swabs. It is commercially available and

performed well in a multicenter evaluation study (27). A novel FDA-cleared automated test

for PCR detection is the GBS GeneXpert test from Cepheid (Sunnyvale, CA), which proved to

be highly sensitive but not very specific (less than 65%) in clinical evaluation (44). Both

tests are performed directly on clinical samples. While the costs are exceedingly higher than

selective culture, a major advantage is that results may be available within a short time

frame, and vaginal colonization status can be assessed at the time of delivery. In comparison

with the gold standard of antenatal selective broth culture, as recommended by the CDC, the

tests may offer alternatives for the future. Nevertheless, it has to be kept in mind that PCR

tests performed at the time of delivery have the major problem that despite rapid

performance time, time to delivery is often too short to allow effective administration of

peripartum antibiotics, if needed.

S. pneumoniae

Several different PCR assays have been developed for the identification of S.

pneumoniae from culture isolates. Tests are based on the detection of the genes for

autolysin lytA, the pneumococcal surface antigenpsaA, and the pneumolysin gene ply.

Comparison of the ability to distinguish difficult-to-identify S. pneumoniaestrains and closely

related atypical streptococci revealed that the lytA-based PCR was the most specific method

(79). A novel and improved PCR for the detection of lytA has recently been evaluated,

confirming these results (18). While results based on the detection of psaA are also

acceptable, the different pneumolysin-targeted methods appear to be relatively nonspecific.

So far these assays are not commercially available and have to be established as “in-house”

PCRs. Nucleic acid probes for the detection of cultured isolates of S. pneumoniae are

commercially available (AccuProbe; GenProbe, San Diego, CA) (28). Detection relies on

hybridization of a specific probe to 16S rRNA sequences but fails to distinguish S.

pseudopneumoniaefrom S. pneumoniae. These tests are not routinely performed for

standard identification procedures but can aid in the identification of atypical S.

pneumoniae isolates with unusual patterns of bile solubility and optochin susceptibility.

ISOLATION PROCEDURES Back to top

General Procedures

Streptococci are usually grown on blood agar media because the assessment of the

hemolytic reaction is important for identification. Growth of streptococci is often enhanced in

the presence of an exogenous catalase source. Streptococcal species with low or absent

hydrogen peroxide production, such as S. agalactiae,can be grown on other commonly used

nonselective media without blood.

Agar media selective for gram-positive bacteria (e.g., phenylethyl alcohol-containing agar or

Columbia agar with colistin and nalidixic acid) support the growth of streptococci. The

optimal incubation temperature range for most streptococcal species lies between 35°C and

37°C. Supplemental carbon dioxide (5% CO2) or anaerobic conditions enhance the growth of

many streptococcal species since streptococci are facultative anaerobes. Although some

streptococci grow well in ambient air, incubation in 5% CO2 is recommended for the culture

of S. pneumoniae and other streptococcal species of the viridans group.

Special Procedures for Streptococcus pyogenes Throat

Cultures

A properly performed and interpreted throat culture on a 5% sheep agar with trypticase soy

base incubated in air remains the gold standard for the diagnosis of S. pyogenes acute

pharyngitis (85). The isolation of only a few colonies of S. pyogenes does not allow the

differentiation between a carrier and an acutely infected individual and may reflect

inadequate specimen collection (12). Lack of hemolysis, overgrowth, and production of toxic

bacterial metabolites by nonpathogenic organisms of the upper respiratory tract microbiota

or depletion of substrates often leads to false-negative results or delays caused by laborintensive

reisolation steps. In order to enhance S. pyogenes isolation, numerous studies

analyzed incubation conditions in anaerobic or CO2-enriched atmosphere as well as different

media selective for beta-hemolytic streptococci (63, 72, 126). Due to cost restraints and an

uncertain benefit, these additional efforts are not generally recommended. After 18 to 24 h

of incubation, culture plates should be examined for growth of beta-hemolytic colonies.

Negative cultures should be reexamined after an additional 24-h incubation period.

Presumptive identification of S. pyogenes can be achieved by susceptibility to bacitracin or

testing for PYR activity. Other beta- hemolytic streptococci are occasionally positive in one of

these tests, but not in both. Definitive diagnosis includes the demonstration of the Lancefield

group A antigen by immunoassay. Although other species may rarely possess the group A

antigen (Table 1), they lack PYR activity (36).

Special Procedures for Streptococcus agalactiae Screening

Early-onset neonatal GBS (S. agalactiae) infections can be prevented by administration of

antibiotic prophylaxis during delivery (105). An essential requirement for efficient

prophylaxis is the reliable detection of colonization with S. agalactiae in pregnant women

before delivery. Screening should be performed between weeks 35 and 37 of pregnancy. A

lower vaginal and a rectal swab (i.e., insertion of a swab through the anal sphincter) should

be obtained either with one or two different swabs and placed in appropriate transport

medium (Amies or Stuart’s medium without charcoal; see chapter 16). While culture counts

decline to some extent, viability ofS. agalactiae is preserved in transport medium kept at

room temperature or 4°C for up to 4 days. To reduce costs, vaginal and rectal swabs can be

placed in a single transport medium tube and cultured together. Swabs should be cultured in

selective broth medium for 18 to 24 hours at 35 to 37°C in ambient air or 5% CO2 and

subsequently plated on tryptic soy agar (TSA) blood agar plates or S. agalactiae selective

agar medium. Selective broth medium is commercially available (Trans-Vag broth

supplemented with 5% sheep blood [Remel Inc., Lenexa, KS] or LIM broth [BBL Microbiology

Systems, Cockeysville, MD]). Selective broth can also be prepared by supplementation of

THB with nalidixic acid (15 μg/ml) and colistin (10 μg/ml) or by supplementation of THB with

nalidixic acid (15 μg/ml) and gentamicin (8 μg/ml). TSA blood agar plates should be checked

for typical colonies (narrow zone of beta-hemolysis) of S. agalactiae after 24 and 48 hours of

incubation at 35 to 37°C. Identification of S. agalactiae is then achieved by standard

techniques as described below. Selective media relying on the detection of the orange S.

agalactiae pigment (Granada medium or StrepB Carrot broth [Hardy Diagnostics, Santa

Maria, CA] or GBS broth [Northeast Laboratory Services, Waterville, ME]) are highly specific

and sensitive (100, 123). Subculture of enrichment broth on Granada medium enhances

sensitivity and obviates the need for further identification steps due to excellent specificity,

but nonhemolytic strains cannot be detected with pigment-dependent selective media. PCRbased

detection of S. agalactiae following overnight enrichment broth increases sensitivity

and detection time in comparison to conventional selective culture (14). For the identification

of questionable cultured strains as S. agalactiae, a 16S RNA-based nucleic acid detection test

(AccuProbe; GenProbe, San Diego, CA) can be helpful. However, all of the nucleic acid-based

detection techniques are more expensive than conventional enrichment culture.

IDENTIFICATION Back to top

Description of Colonies

Colonies of streptococci usually appear gray or almost white with moist or glistening

features. Dry colonies are rarely encountered. Colony size varies among the different betahemolytic

species and helps to distinguish groups of streptococci. Beta-hemolytic

streptococci of the pyogenic group (S. pyogenes, S. agalactiae, and S.

dysgalactiae subsp. equisimilis) form colonies of >0.5 mm after 24 hours of incubation, in

contrast to the beta-hemolytic strains of the S. anginosus group (formerly called “S. milleri”

group), which present with pinpoint colonies of ≤0.5 mm after the same incubation time

(Fig. 1). Members of the S. anginosusgroup emit a distinct odor resembling butterscotch or

caramel, presumably due to the production of diacetyl by the species belonging to this

group. Among the beta-hemolytic species of the pyogenic group, S. agalactiaeproduces the

largest colonies with a relatively small zone of hemolysis. Nonhemolytic S. agalactiae strains

do occur and resemble enterococci.

Identification of Beta-Hemolytic Streptococci by Lancefield

Antigen Immunoassays

Commercially available Lancefield antigen grouping sera are primarily used for the

differentiation of beta-hemolytic streptococci. Products for rapid antigen extraction and

subsequent agglutination can be obtained from many different suppliers. The presence of the

Lancefield group B antigen in beta-hemolytic isolates from human clinical specimens

correlates with the species S. agalactiae. Similarly, the detection of the Lancefield group F

antigen in small-colony-forming streptococci from human clinical material allows a fairly

reliable identification of a strain as a member of the S. anginosus group. The presence of

Lancefield group A, C, or G antigens necessitates further testing (Table 1). Beta-hemolytic

streptococcal strains not reacting with any of the Lancefield antisera are rare and should be

further identified by phenotypic tests or nucleic acid detection techniques.

Identification of Beta-Hemolytic Streptococci by Phenotypic

Tests

A number of streptococcal identification products incorporating batteries of physiologic tests

are commercially available (see chapters 3 and 17). In general, these products perform well

with commonly isolated pathogenic streptococci but may lack accuracy for identifying

streptococci of the viridans group. For the bulk of pathogenic streptococci isolated in clinical

laboratories (e.g., S. pyogenes, S. agalactiae, and S. pneumoniae), serologic or presumptive

physiologic tests (as described below) offer an acceptable alternative to commercially

available identification systems.

PYR Test

The presence of the enzyme PYR is often tested to distinguish S. pyogenes from other betahemolytic

streptococci. Hydrolysis of L-pyrrolidonyl-β-naphthylamide by the enzyme to β-

naphthylamide produces a red color with the addition of cinnamaldehyde reagent (chapter

17). The beta-hemolytic streptococcal species S. iniae and S. porcinus can be PYR positive

but are only rarely identified in human clinical specimens, since they are primarily animalassociated

species. PYR spot tests are commercially available. It is important to

distinguish Streptococcus from Enterococcus prior to PYR testing, and strains of other related

genera may be PYR positive (including the genera Abiotrophia, Aerococcus, Enterococcus,

Gemella, and Lactococcus). However, PYR-positive beta-hemolytic enterococcal isolates

typically present with a different colonial morphology (smaller zone of beta-hemolysis and

bigger colony size) and when combined with other phenotypic characteristics (see chapter

21) may be distinguished from streptococci. To avoid false-positive reactions caused by other

PYR-positive bacterial species (for example, staphylococci), the test should be performed on

pure cultures only.

Bacitracin Susceptibility

With rare exceptions, S. pyogenes displays bacitracin susceptibility, in contrast to other

human beta-hemolytic streptococci. Together with Lancefield antigen determination, it can

be used for the identification of S. pyogenes since beta-hemolytic strains of other

streptococcal species that may contain the group A antigen are bacitracin resistant. The test

can also be used to distinguish S. pyogenes from other PYR-positive beta-hemolytic

streptococci (S. iniae and S. porcinus). A bacitracin disk (0.04 U) is applied to a sheep blood

agar plate that has been heavily inoculated with three or four colonies of a pure culture of

the strain to be tested. It is important to perform the test from a subculture on sheep blood

agar, since placement of bacitracin disks on primary plates is not sensitive enough. After

overnight incubation at 35°C in 5% CO2, any zone of inhibition around the disk is interpreted

as indicating susceptibility. Importantly, bacitracin-resistant S. pyogenes isolates have been

reported, and clusters of bacitracin-resistant strains were observed in several European

countries (74, 80, 90).

VP Test

The Voges-Proskauer (VP) test detects the formation of acetoin from glucose fermentation. It

is performed on streptococci as a modification of the classical VP reaction that is used for the

differentiation of enteric bacteria. Small-colony-forming beta-hemolytic streptococci of the S.

anginosus group that are VP positive may be distinguished from large colony-forming betahemolytic

streptococci harboring identical Lancefield antigens (A, C, or G). Streptococci of

the S. mitis group are VP negative. For the modified VP reaction described by Facklam and

Washington in the fifth edition of this Manual (40a), the culture growth of an entire agar

plate is used to inoculate 2 ml of VP broth and incubated at 35°C for 6 hours. Following the

addition of 5%α-naphthol and 40% KOH, the tube is shaken vigorously for a few seconds

and incubated at room temperature for 30 min. A positive test yields a pink-red color that

results from the reaction of diacetyl with guanidine.

BGUR Test

Detection of β-glucuronidase (BGUR) activity distinguishes S.

dysgalactiae subsp. equisimilis strains containing Lancefield group antigens C or G from

BGUR-negative, small-colony-forming streptococci of the S. anginosusgroup with the same

Lancefield group antigens. Rapid methods for the BGUR test are commercially available.

Alternatively, a rapid fluorogenic assay with methylumbelliferyl-β-D-glucuronide (MUG)-

containing MacConkey agar, often used for Escherichia coli, has been described (68).

CAMP Test

The CAMP factor reaction was first described in 1944 by Christie, Atkins, and Munch-Petersen

and refers to the synergistic lysis of erythrocytes by the beta-hemolysin of Staphylococcus

aureus and the extracellular CFB protein of S. agalactiae. The gene and its expression can be

demonstrated in the vast majority (>98%) of S. agalactiae isolates, but CAMP-negative

mutants do occur. The strain to be tested and a Staphylococcus aureusstrain (ATCC 25923)

are streaked onto a sheep blood agar plate at a 90° angle. Plates are incubated in ambient

air overnight at 36 } 1°C. A positive reaction can be detected by the presence of a triangular

zone of enhanced beta-hemolysis in the diffusion zone of the beta-hemolysin of S.

aureus and the CAMP factor (Fig. 2). CAMP factor-positive strains can also be detected by a

method using beta-lysin containing disks (Remel) or by a rapid CAMP factor spot method

(96). Despite the fact that close homologs of the CAMP factor gene are present in many S.

pyogenes strains, most beta-hemolytic streptococci other than S. agalactiae are negative in

the above-described CAMP factor test, except for the rare human isolates of S. iniae, S.

porcinus, and S. pseudoporcinus. Several gram-positive rods including corynebacteria

and Listeria monocytogenes strains may also be CAMP factor positive.

S. mutans Group

The S. mutans group includes S. mutans, S. sobrinus, S. criceti, S. ratti, S. downei, S. ferus,

S. hyovaginalis, S. devriesei, and S. macacae. S. mutans and S. sobrinus are frequently

found in human hosts, while the other species are only rarely encountered in humans or

represent animal pathogens. The species of the S. mutansgroup are characterized by the

production of extracellular polysaccharides from sucrose, which can be tested by culturing

the bacteria on sucrose-containing agar and by the ability to produce acid from a relatively

wide range of carbohydrates. S. mutans strains may present with an atypical morphology for

streptococci, forming short rods on solid media or in broth culture under acidic conditions. On

blood agar, colonies are often hard, adherent, and usually alpha-hemolytic. Under anaerobic

growth conditions, some strains are beta-hemolytic.S. sobrinus strains are mostly

nonhemolytic or occasionally alpha-hemolytic. On sucrose-containing agar, species from this

group form colonies that are rough (frosted glass appearance), heaped, and surrounded by

liquid-containing glucan.

S. salivarius Group

Streptococcal species in the S. salivarius group are S. salivarius, S. vestibularis, and S.

thermophilus. S. salivarius strains are usually non- or alpha-hemolytic on blood agar. On

sucrose-containing agar, strains form large mucoid or hard colonies due to the production of

extracellular polysaccharides. A high proportion of S. salivarius strains react with the

Lancefield group K antiserum. Species in this group may also react with the streptococcal

group D antiserum. It is unclear if these strains truly possess the group D antigen or yield a

nonspecific cross-reaction. S. vestibularis is alpha-hemolytic, and the failure of this species

to produce extracellular polysaccharides on sucrose-containing agar is helpful in

distinguishing S. vestibularis from S. salivarius strains. S. thermophilus is found in dairy

products but has not been isolated from clinical specimens.

S. bovis Group

The species belonging to the S. bovis group (S. equinus, S. gallolyticus, S.

infantarius, and S. alactolyticus) are nonenterococcal group D streptococci that are PYR

negative. Most strains grow on bile esculin agar and are unable to grow in 6.5% NaCl. On

blood agar, strains are either nonhemolytic or alpha-hemolytic. Strains of theS. bovis group

share phenotypic characteristics with S. mutans strains, such as production of glucan,

fermentation of mannitol, and growth on bile esculin agar. However, the S. bovis group does

not ferment sorbitol. S. gallolyticus subsp. gallolyticus and S. infantarius subsp. coli typically

ferment starch or glycogen and give a Lancefield group D reaction, in contrast to S.

gallolyticus subsp. pasteurianus. A detailed phenotypic characterization and emended

description of the different subspecies have recently been published (6). For the

identification of species in this group, testing of BGUR, α- and β-galactosidase, β-

mannosidase, acid production from starch, glycogen, inulin, and mannitol is helpful. As

described earlier, strains formerly known as S. bovis currently belong to several species of

the S. bovis group.

Physiologic Tests

Optochin Test

Most S. pneumoniae isolates are optochin susceptible. Before application of the optochin

disk, several colonies of a pure culture are streaked onto a sheep blood agar plate. Optochin

testing should be performed on plates that are incubated at 35 to 36°C overnight in 5%

CO2 because up to 8% of strains will not grow under ambient conditions. S.

pneumoniae isolates show zones of inhibition of ≥14 mm with a 6-mm-diameter disk

(containing 5 μg of optochin) and zones of inhibition of ≥16 mm with a 10-mm-diameter

disk. Incubation in 5% CO2 yielded increased specificity (1, 98). Optochin-resistant S.

pneumoniae strains have been reported as well as optochin-susceptible S. mitis isolates

(especially when tested under ambient conditions). Since optochin testing may miss between

4% and 11% of bile-soluble S. pneumoniae isolates (1, 98), strains displaying a smaller zone

of inhibition (9 to 13 mm for the 6-mm-diameter disk) should be subjected to additional

testing (e.g., bile solubility and genetic testing) to confirm species identification.

Application of an optochin disk onto the primary culture medium may facilitate a rapid

presumptive identification but may miss pneumococcal isolates in a mixed culture. The

optochin susceptibility test should be repeated with a pure culture in cases of mixed cultures,

or additional tests should be performed (e.g., bile solubility).

Bile Solubility Test

Bile solubility can be performed either in a test tube or by direct application of the reagent to

an agar plate. For the test tube method, a saline suspension of a pure culture is adjusted to

a McFarland standard of 0.5 to 1.0, and 0.5 ml of the suspension is added to a small tube.

The bacterial suspension is mixed with 0.5 ml of 10% sodium deoxycholate (bile) and

incubated at 35°C. A control containing 0.5 ml of bacterial suspension with 0.5 ml of saline

should be prepared for each strain tested. A positive result is characterized by clearing of the

bile suspension within 3 hours. Clearing can start as early as 5 to 15 min after inoculation

and allows the identification of a strain as S. pneumoniae. For the plate method, one drop of

10% sodium deoxycholate is placed directly onto a colony of the strain in question and

incubated at 35°C for 15 to 30 min in ambient air. It is important to keep the plates in a

horizontal position in order to prevent the reagent from washing away the colony. Colonies

of S. pneumoniae will disappear or demonstrate a flattened colony morphology, while other

viridans group streptococci will appear unchanged. In contrast to optochin susceptibility, bile

solubility testing demonstrated excellent sensitivity and specificity in a recent comprehensive

evaluation (98).

Bile Esculin Test

Bile esculin medium (available from commercial sources) in either plates or slants should be

inoculated with one to three colonies of the organism to be tested and incubated at 35°C in

ambient air for up to 48 h. Optimal results can be achieved by using media supplemented

with 4% oxgall (equivalent of 40% bile) (Remel, Lenexa, KS) and a standardized inoculum of

106 CFU (22). A definitive blackening of plated media or blackening of at least one-half of an

agar slant is considered a positive test, indicative of species belonging to the S. bovis group

or enterococci. Occasional other viridans group streptococci are positive with this test or

display weakly positive reactions that are difficult to interpret. Isolates from patients with

serious infections (e.g., endocarditis) should be more completely characterized.

Arginine Hydrolysis

Arginine hydrolysis is a key reaction for the identification of viridans group streptococci.

Discrepancies can occur among test methods (127). Two commonly used methods are

detailed here. Moeller’s decarboxylase broth containing arginine (Becton Dickinson, Sparks,

MD, and other sources) should be inoculated with the test organism, overlaid with mineral

oil, and incubated at 35 to 37°C for up to 7 days. Degradation of arginine results in elevated

pH, indicated by development of a purple color. Negative results are indicated by a yellow

color, which is due to acid accumulation from metabolism of glucose only. For the microtiter

plate method (7), three drops of the arginine-containing reagent are inoculated with 1 drop

of an overnight THB culture and incubated for 24 h at 37°C anaerobically. Production of

ammonia is detected by the appearance of an orange color following addition of 1 drop of

Nessler’s reagent.

Urea Hydrolysis

Christensen urea agar (Becton Dickinson and other sources) is inoculated and incubated

aerobically at 35°C for up to 7 days. Development of a pink color indicates a positive

reaction. An alternative format is to dispense Christensen’s medium without agar into a

microtiter tray well and, after inoculation, overlay it with mineral oil prior to incubation.

VP Test

The VP test can be performed as described above for the identification of beta-hemolytic

streptococci. A standard method for performing the VP test, requiring extended incubation, is

described in chapter 17.

Esculin Hydrolysis

Esculin agar slants (Becton Dickinson and other sources) are inoculated and incubated for up

to 1 week. A positive reaction appears as a blackening of the medium; no change in color

indicates a negative esculin hydrolysis test.

Hyaluronidase Production

Hyaluronidase activity can be detected on brain heart infusion agar plates supplemented with

2 mg/liter of sodium hyaluronate (Sigma-Aldrich, St. Louis, MO). The strains to be tested are

inoculated by stabbing into the agar, and plates are incubated anaerobically at 37°C

overnight. After the plate is flooded with 2 M acetic acid, hyaluronidase activity is indicated

by the appearance of a clear zone around the stab. A quantitative method for determining

hyaluronidase activity can be performed in microtiter trays (54).

Production of Extracellular Polysaccharide

Strains may be isolated as single colonies on sucrose-containing agar. The two most

commonly used media are (i) mitis-salivarius agar containing 0.001% (wt/vol) potassium

tellurite (Becton Dickinson) and (ii) tryptone-yeast-cystine agar (Lab M, Bury, United

Kingdom). Incubation may require up to 5 days at 37°C under anaerobic incubation

conditions.

TYPING SYSTEMS Back to top

In the majority of cases, typing of streptococci has no immediate clinical or therapeutic

consequences. It is most often performed by reference laboratories for the purposes of

epidemiologic studies and the evaluation of vaccine efficacy. Although classical antibodydependent

typing systems for capsular serotypes and surface proteins have been used for

years, molecular methods have become attractive, since they do not require special

techniques or the maintenance of rarely used reagents such as a large antibody panel.

Another advantage is the independence of DNA sequences from culture conditions and gene

expression. For the differentiation of distinct clones, pulsed-field gel electrophoresis (PFGE)

and multilocus sequence typing (MLST) systems have been established for many

streptococcal species (34).

S. pneumoniae comprises more than 90 antigenically distinct capsular serotypes that can be

detected by the Neufeld test (Quellung reaction), which is still regarded as the gold standard

for epidemiologic studies. Pure cultures of pneumococci are grown on a freshly prepared 5%

sheep blood agar plate or a 10% horse blood agar plate at 35°C to 37°C and 5% CO2 for 18

to 24 hours. A small amount of bacterial growth (less than a 10-μl loop) is resuspended in a

droplet of phosphate-buffered or physiological saline (McFarland standard of approximately

0.5). A few microliters of the saline suspension is mixed with an equal amount of specific

pneumococcal rabbit antisera on a glass slide. The specimen is subsequently evaluated for

capsular swelling (a clear area surrounding the bacterial cells) by phase-contrast microscopy

(×1,000 magnification; oil immersion) (2). The reaction is stable for approximately 30 min.

Best results are achieved when 10 to 50 bacterial cells are visible per high-power (×1,000)

microscopic field. To increase visibility of the result, it is possible to add 0.3% aqueous

methylene blue in the same amount as the antiserum to the mixture. Following the same

principle, commercially available kits (Pneumotest Statens Serum Institut, Copenhagen,

Denmark) allow rapid testing of S. pneumoniae serotypes with pooled antisera by a

checkerboard method. A rapid antigen detection test using pooled antisera coupled to latex

beads (Statens Serum Institut) has been developed (110). Due to strain discrepancies,

confirmation by Neufeld Quellung reaction is recommended. For the distinction of single

clones, PFGE (73) or MLST typing schemes (34) have been used in pneumococcal

investigations.

Ten different antibody-defined capsular polysaccharides have been described for S.

agalactiae (Ia, Ib, and II through IX). The percentage of nontypeable strains can be

minimized by optimization of capsular expression (10). In addition to antibody detection of

capsular serotypes, PCR- and DNA sequencing-based techniques allow the detection of

capsular serotypes (71, 95). Individual clones of S. agalactiae have been detected either by

MLST (57) or by PFGE (9).

Conventional typing of S. pyogenes is based upon the antigenic specificity of the surfaceexpressed

T and M proteins (56). The trypsin-resistant T protein is part of the recently

discovered pilus structures (82). The T type can be identified by agglutination with

commercially available serologic assays utilizing approximately 20 accepted anti-T sera. M

proteins are major antiphagocytic virulence factors of S. pyogenes (42). N-terminal sequence

variation in genes encoding these highly protective antigens is the basis of the S.

pyogenesprecipitation typing system. At present, 83 M serotypes are unequivocally validated

and internationally recognized to be serologically unique and are designated M1 to M93 in

the Lancefield classification (39). M serotypes that are not included are from non-S.

pyogenes organisms or correspond to an existing M serotype.

A molecular typing system is based on the nucleotide sequences encoding the amino termini

of M proteins. The emm gene sequences encode M proteins and have been correlated with

Lancefield M serotypes. This methodology allows assignment to a validated M protein gene

sequence (emm1 through emm124) and the identification of new emm sequence types and

has evolved into the gold standard of S. pyogenes typing (39). A large database of

approximately 350 emm gene sequences from strains originally used for Lancefield

serotyping and including emm sequences from beta-hemolytic groups C, G, and L

streptococci is maintained at the CDC

(http://www.cdc.gov/ncidod/biotech/infotech_hp.html). Recently, MLST has been developed

for S. pyogenes. Population genetic studies demonstrated stable associations

between emm type and MLST among isolates obtained decades apart and/or from different

continents (35).

In outbreak situations that include S. pyogenes, restriction enzyme-mediated digestion

of emm amplicons is a valuable tool for rapid identification of isolates containing

similar emm genes (5). For clusters of isolates sharing the same emm type, PFGE profiles

may be helpful for distinguishing similar strains (87).

SEROLOGIC TESTS Back to top

Determination of streptococcal antibodies is indicated for the diagnosis of poststreptococcal

disease, such as ARF or GN (108). A fourfold rise in antibody titer is regarded as definitive

proof of an antecedent streptococcal infection. Multiple variables, including site of infection,

time since the onset of infection, age, the background prevalence of streptococcal infections

(3), antimicrobial therapy, and other comorbidities, influence antibody levels. The most

widely used antibodies are anti-streptolysin O and anti-DNase B.

Antibodies against streptolysin O (ASO) reach a maximum at 3 to 6 weeks after infection.

While ASO responses following streptococcal upper respiratory tract infections are usually

elevated, pyoderma caused byS. pyogenes does not elicit a strong ASO

response. Streptococcus dysgalactiae subsp. equisimilis can also produce streptolysin O, and

thus elevated ASO titers are not specific for S. pyogenes infections.

Among the four streptococcal DNases produced, the host response is most consistent against

DNase B. Anti-DNase B titers may not reach maximum titers for 6 to 8 weeks but remain

elevated longer than ASO titers and are more reliable than ASO for the confirmation of a

preceding streptococcal skin infection. Moreover, since only 80 to 85% of patients with

rheumatic fever have elevated ASO titers, additional anti-DNase B titers may be helpful.

Due to frequent exposure to S. pyogenes, ASO and anti-DNase B titers are higher in children

in the United States from 2 to 12 years of age. Geometric mean values are 89 Todd units for

ASO and 112 Todd units for anti-DNase B, while the upper limits of normal values are 240

Todd units (ASO) and 640 Todd units (anti-DNAse B) (58). Prompt antibiotic therapy of

streptococcal infections can reduce the titer but does not abolish antibody production.

Streptococcal carriers do not experience a rise in streptococcal antibody titers.

The hemagglutination-based streptozyme test (Streptozyme; Carter-Wallace, Inc., Cranbury,

NJ) was developed to detect antibodies against multiple extracellular streptococcal products.

However, variabilities in test standardization and inconsistent specificities have been

reported (45). Assays for antibody detection against other S. pyogenes proteins

(hyaluronidase, streptokinase, and NAD glycohydrolase) are technically difficult to perform

and are not commercially available.

ANTIMICROBIAL SUSCEPTIBILITIES Back to top

Beta-Hemolytic Streptococci

Penicillin remains the drug of choice for the empirical treatment of streptococcal infections

due to S. pyogenes,because in contrast to S. pneumoniae and other alpha-hemolytic

streptococci, S. pyogenes remains uniformly susceptible to penicillin. Reports about reduced

penicillin susceptibility in strains of S. pyogenes have not been confirmed by reference

laboratories. This is, however, no longer true for S. agalactiae. Recent reports show the

emergence of diminished susceptibility to penicillin G caused by a mutation of the penicillin

binding proteins Pbp2x in isolated strains in Asia and the United States (24, 67). Due to

suspected or confirmed penicillin allergies in more than 10% of patients, macrolides are

often given as an alternative treatment. Macrolide resistance rates among isolates of S.

pyogenes and S. agalactiae have been increasing in North America as well as in Europe (29).

Resistance rates correlate with the use of macrolides in clinical practice, and geographic

differences in resistance rates are often due to differences in macrolide use. In the United

States, the rate of macrolide resistance among S. agalactiae rose from 12 to 20% from 1990

to 2000 (84) and has recently been reported as 38% (50). Beta-hemolytic streptococcal

isolates with a reduced susceptibility to glycopeptides have not been reported. Due to the

uniform susceptibility of S. pyogenes to penicillin, resistance testing for penicillins or other

beta-lactams approved for treatment of S. pyogenes and S. agalactiae is not necessary for

clinical purposes. So far the existence of rare isolates of S. agalactiae with reduced

susceptibility to penicillin has not resulted in a change of this recommendation, which may of

course change if increasing numbers of such strains are encountered. Susceptibility testing

for macrolides should be performed by using erythromycin, since resistance and

susceptibility of azithromycin, clarithromycin, and dirithromycin can be predicted by testing

erythromycin.

S. pneumoniae and Viridans Group Streptococci

In view of the development of penicillin resistance in S. pneumoniae and other alphahemolytic

streptococci, penicillin can no longer be recommended as the empirical treatment

of choice in many countries. Penicillin is considered a preferred antimicrobial agent for

only S. pneumoniae and other alpha-hemolytic streptococci with demonstrated

susceptibilities to penicillin. Penicillin resistance in S. pneumoniae is caused by altered

penicillin-binding proteins. Approximately 25% of S. pneumoniae isolates from the United

States were not fully susceptible to penicillin in 2007 (http://www.cdc.gov/abcs). But the

recent changes of S. pneumoniaebreakpoints in nonmeningeal isolates (susceptibility, ≤2

μg/ml; intermediate resistance, 4 μg/ml; resistance, ≥8 μg/ml) for penicillin in CLSI

definitions caused this value to drop to less than 10% (124). Susceptibility to penicillin can

be determined by a disk diffusion test with 1 μg of oxacillin. According to the current CLSI

guidelines, in all cases where oxacillin zone sizes (≤19 mm) indicate a reduced susceptibility

to penicillin, MIC determinations for penicillin should be performed. For susceptibility testing

of all other β-lactams in S. pneumoniae, MIC determinations are recommended. S.

pneumoniae infections should be treated according to current guidelines (76). Depending on

the clinical situation, treatment options include penicillin, extended-spectrum cephalosporins,

macrolides, fluoroquinolones, and vancomycin. In addition, more than one-third of blood

culture isolates of the viridans group collected in the late 1990s in the United States were not

susceptible to penicillin (31). Elevated percentages of penicillin-resistant strains can be found

among S. mitisand S. salivarius isolates. S. pneumoniae was uniformly susceptible to

macrolides until the late 1980s in the United States, but macrolide resistance is now evident

in about 25% of S. pneumoniae strains (117).

The increased use of fluoroquinolones to treat S. pneumoniae infections has been

accompanied by a rise in fluoroquinolone-resistant S. pneumoniae strains. Resistance occurs

in a stepwise fashion and is due to mutations in DNA topoisomerase IV or a subunit of DNA

gyrase. While the overall prevalence of fluoroquinolone resistance is below 1% according to

the CDC’s Active Bacterial Core surveillance data (http://www.cdc.gov/abcs), the increase in

resistant strains during recent years emphasizes the need for close monitoring. Clinical

failures of levofloxacin therapy due to resistance have been reported (25). Vancomycinresistant

S. pneumoniae isolates have not been described. However, the isolation of a

vancomycin-resistant S. bovis isolate has been reported (93).

EVALUATION, INTERPRETATION, AND REPORTING OF

RESULTS Back to top

Streptococci from the pyogenic group are important human pathogens. Timely identification

of these species in clinical specimens is therefore crucial to treat infections adequately and to

reduce transmission. Tonsillopharyngitis caused by S. pyogenes remains a substantial health

problem in childhood and adolescence. Diagnosis by either rapid antigen tests or

bacteriological culture minimizes unjustified antibiotic treatment of viral pharyngitis.

Serologic tests are usually applied in cases of suspected poststreptococcal sequelae. Proper

identification and reporting, however, should not be limited to S. pyogenes, since S.

dysgalactiae subsp.equisimilis (human group C and G streptococci) has been documented as

an agent of pharyngitis including cases complicated by nonsuppurative sequelae. In this

context it is important to correctly differentiate these pathogens from the small-colonyforming

beta-hemolytic species of the S. anginosus group that make up part of the

oropharyngeal microbiota. While invasive neonatal S. agalactiae infections are declining due

to improved prenatal screening and peripartal antibiotic prophylaxis, increased detection

of S. agalactiae from adult patients has been reported (109). Thorough identification and

reporting of this organism should therefore not be confined to screening swabs during

pregnancy or in newborns.

Despite the fact that S. pneumoniae is often found as a colonizer in respiratory samples, it

should always be clearly distinguished from viridans group streptococci and reported.

Cultural methods remain the mainstay in pneumococcal pneumonia and sepsis as well as

meningitis. To enable the initiation of adequate antibiotic treatment, resistance testing

should be performed for all isolates. Special care should be taken to ensure the reporting of

the correct β-lactam breakpoints for non-CSF and CSF S. pneumoniae isolates, which have

just recently been changed. If antibiotic treatment was started before microbiological

samples were obtained, the urinary antigen test or nucleic acid detection techniques may

help to properly identify the causative agent, especially in invasive infections.

The correct identification of viridans group streptococci and the distinction of strains causing

infections from isolates of the physiological microbiota remain a major challenge.

Identification to the group or species level should be confined to strains causing abscesses,

endocarditis, and serious infections in neutropenic patients. Many S. mitis isolates are no

longer penicillin susceptible, and special attention has to be paid to susceptibility testing.

Due to the association of S. gallolyticus subsp. gallolyticus with malignancies of the

gastrointestinal tract and in view of the recent taxonomic changes within the S. bovis group,

reports of novel species designations should include the information that the species belongs

to the S. bovis group (121).