Staphylococcus, Micrococcus, and Other Catalase-Positive Cocci


EPIDEMIOLOGY AND TRANSMISSION 
Staphylococcus and Related Genera

The major habitats of most staphylococcal species are the skin and mucous membranes of

mammals and birds. S. aureus subsp. aureus is considered to be the most important human

pathogen among staphylococci, followed by S. epidermidis, S. haemolyticus, and S.

saprophyticus subsp. saprophyticus. S. auricularis, S. capitissubsp. capitis and urealyticus, S.

caprae, S.cohnii subsp. cohnii and urealyticus, S.

hominis subsp. hominis andnovobiosepticus, S. lugdunensis, S. pasteuri, S. pettenkoferi, S.

saccharolyticus, S. schleiferi subsp. schleiferi, S. simiae, S. simulans, S. warneri, and S.

xylosus are also encountered in human specimens (164). These species are found mainly as

part of the resident microbiota. For references concerning species description, see “List of

Prokaryotic Names with Standing in Nomenclature” (http://www.bacterio.cict.fr/).

S. aureus and Other Coagulase-Positive Species

As determined in longitudinal studies, three types of S. aureus nasal carriers have been

historically distinguished: persistent carriers (10 to 35%, carrying one strain over time),

intermittent carriers (20 to 75%, carrying different strains), and noncarriers (5 to 50%)

(203). Since intermittent carriers and noncarriers share similar S. aureus nasal elimination

kinetics, it was recently proposed that there are only two types of nasal carriers: persistent

carriers and others (189). From the vestibulum nasi, S. aureus can be transferred to skin

and other body areas. Health care workers have a high S. aureus nasal carriage rate (50% to

90%) as do patients with insulin-dependent diabetes mellitus, patients receiving long-term

hemodialysis, and users of illegal intravenous drugs (37). The vaginal carriage rate in adult

premenopausal women is about 10% (119). The intertriginous skin folds, the axillae, and the

perineum are also found to be regularly colonized. Of particular interest, nasal colonization

plays a crucial role as a source of invasive infections (107, 195, 202).

The population of S. aureus presents a highly clonal structure dominated by approximately a

dozen major clonal complexes comprising hundreds of clonal lineages or sequence types

(58). The S. aureus populations are also divided into four distinct groups based

on agr (accessory gene regulator system) allelic variation (135).

Within health care facilities, S. aureus strains are transmitted from patient to patient

primarily via hand carriage of medical personnel. This is of utmost importance for the

transmission of methicillin-resistant S. aureus (MRSA) strains (MRSA strains are further

discussed in “A ntimicrobial Susceptibilities” below). The role of the external environment is

less important, except for certain areas such as intensive care units and burn units.

Colonized or infected health care workers may act as a reservoir (2). MRSA strains

circulating in livestock and found in meat production differ from companion animal strains.

For companion animals, MRSA acquisition is primarily a humanosis, in contrast to the newly

emerging livestock-associated MRSA and methicillin-susceptibleS. aureus (MSSA) strains

such as ST398, presenting a genuine zoonotic risk (130).

Staphylococcus aureus subsp. anaerobius has been recovered from subcutaneous abscesses

of sheep (45). So far, it has not been isolated from human clinical specimens. S.

intermedius, the recently delineated species S. pseudintermedius, and two clusters of S.

delphini are closely related coagulase-positive species commonly recovered from carnivores,

other mammals (e.g., horses and cats), and birds. Since it is very difficult to differentiate

among these species, it is very likely that previously reported S. intermedius isolates with

variant biotypes might have included S. delphini and, in particular, S.

pseudintermedius (163). S. pseudintermediusmight be the predominant coagulasepositive

Staphylococcus species recovered from normal and infected canine skin (48).

Further coagulase-positive animal-adapted staphylococci are S.

schleiferi subsp. coagulans, S. felis, and S. lutrae.

Coagulase-Negative Staphylococci

In humans, S. epidermidis is the most frequently recovered staphylococcal species colonizing

the body surface, where it is particularly prevalent on moist areas such as the axillae,

inguinal and perineal area, anterior nares, and toe webs. S. auricularis is part of the healthy

human external auditory canal microbiota colonizing exclusively this region; S. capitis is

found surrounding the sebaceous glands on the forehead and scalp following puberty; S.

haemolyticus and S. hominis are preferentially isolated from axillae and pubic areas high in

apocrine glands; and S. saprophyticus subsp. saprophyticus is frequently colonizing the

rectum and the genitourinary tract of young women (106, 159, 164). S. lugdunensis is

frequently found on the lower extremities and on the groin (28). However, these species may

be found occasionally on other body sites.

S. sciuri and S. xylosus are commensals of the skin and the mucous membranes of many

animals and, occasionally, of humans. Both species are also found in food; S.

xylosus represents one of the major starter cultures used for meat fermentation. S.

sciuri subsp. carnaticus is recovered mainly from bovine hosts, and subsp. rodentium is

found mainly in rodents. S. kloosii, S. equorum subsp. equorum, and S. gallinarum are found

on several mammals and food products. S. chromogenes, S. hyicus, and S. lentus are

common residents of cloven-hoofed animals and, in addition, may be isolated from their food

products. S. vitulinus (syn. S. pulvereri) is found preferentially on horses and whales. S.

arlettae is found on mammals and birds; S. nepalensis has been isolated from Himalayan

goats; S. muscae is described on flies. S. carnosus subsp. carnosus and utilis, S. condimenti,

S. equorum subsp. linens, S. fleurettii, S. piscifermentans, and S. succinus subsp. casei have

been associated with fermented food and dairy products;

and S. succinus subsp. succinus was isolated from an amber fragment. However, the

complete and/or true natural habitat for many of these species is still unclear.

Macrococcus, Salinicoccus,

Jeotgalicoccus, and Nosocomiicoccus

The Macrococcus genus comprises four hoofed-animal-adapted species including M.

caseolyticus, first described as S. caseolyticus (102). This species is found on food such as

sausages and meat products. The

halotolerant/halophilic Jeotgalicoccus and Salinicoccus species are recovered from fermented

seafood and salted fish or found in saline and desert soil or salt mines. Thus far, the only

reported recovery ofNosocomiicoccus ampullae has been isolation from surfaces of bottles of

saline solution used in wound cleansing (4).

Micrococcaceae and Dermacoccaceae

The skin of humans and other mammals is the primary habitat for

most Micrococcaceae and Dermacoccaceaeisolated from clinical specimens. Cutaneous

populations of micrococci are carried by most people (ca. 96%) with M. luteus as the most

frequent species followed by K. varians (104). R. mucilaginosa is probably a normal

inhabitant of the mouth and upper respiratory tract. Animal and dairy products may be

considered secondary sources of micrococci. Many of the recently discovered members of

the Micrococcaceae and Dermacoccaceae are associated with different environmental

habitats.

Alloiococcus

Recently, a high incidence of A. otitis in the outer ear canal of healthy persons was

demonstrated, suggesting that alloiococci are part of the normal bacterial microbiota (182).

CLINICAL SIGNIFICANCE Back to top

Staphylococcus

Many staphylococcal species are classical opportunists colonizing skin and mucous

membranes but may become pathogenic in a species- and strain-dependent manner

following breaks in the cutaneous epithelial barrier through trauma or medical interventions.

The recovery of a staphylococcal isolate always requires assessment of clinical significance to

determine whether it is a contaminant, colonizer, or pathogen.

S. aureus is the clinically most important species, capable of causing a wide range of human

and animal diseases. S. aureus possesses an extensive, often redundant and overlapping

arsenal of virulence factors, such as adhesins, enzymes, and toxins, and has various

strategies to evade the host immune response. In addition, the pathogen has become

resistant to many of the therapeutic agents available. National Nosocomial Infection

Surveillance and National Healthcare Safety Network data indicate that S. aureus is the most

common cause of nosocomial pneumonia and skin and soft tissue infections. S. aureus is

second only to CoNS as a cause of primary bacteremia in hospitals (84, 205).

Disease entities caused by S. aureus can be broadly divided into toxin-mediated diseases

and suppurative infections comprising skin and soft tissue infections (SSTIs), organic and

systemic infections, and foreign-body-related infections (FBRIs). The spectrum of SSTIs

ranges from superficial (impetigo, folliculitis, furuncles/carbuncles, hydradenitis suppurativa,

pyoderma, and wound infections) to deep entities (abscesses, mastitis, cellulitis, and

pyomyositis) to life-threatening necrotizing fasciitis and myositis. SSTIs are the most

frequent infections associated with community-acquired MRSA (CA-MRSA) with a single clone

(pulsed-field gel electrophoresis [PFGE] type USA 300; multilocus sequence typing [MLST]

type ST-8) being most prevalent (99). Infection of deep sites may involve any body

compartments and organ systems resulting in empyemas, osteomyelitis, arthritis,

endocarditis, pneumonia, otitis media, sinusitis, mastoiditis, and parotitis. Any localizedS.

aureus infection can become invasive and lead to bacteremia. Systemic infections comprise

primary and secondary bacteremia, meningitis, and endocarditis. Bacteremia may be

complicated by metastatic foci (e.g., vertebral osteomyelitis). Congenital or acquired defects

in host defense and the presence of foreign bodies may predispose patients to serious

infections.

Besides an acute aggressive course, S. aureus may also cause chronic, persistent, and

relapsing infections often due to a phenotypic subpopulation designated small-colony

variants (SCVs) (147). SCVs of S. aureus and other staphylococcal species (e.g., S.

epidermidis and S. lugdunensis) have been isolated from patients with chronic osteomyelitis,

abscesses, and FBRIs as well as from cystic fibrosis patients with chronic airway infection

(96, 167, 196).

Classical toxin-mediated diseases due to S. aureus include the staphylococcal toxic shock

syndrome (TSS), staphylococcal food poisoning, and staphylococcal scalded skin syndrome

(SSSS). TSS is associated with colonization by or infection with an isolate of S. aureus that is

positive for TSS toxin-1 (TSST-1) or, less frequently, for other members of the

staphylococcal pyrogenic toxin superantigen (PTSAg) family (primarily staphylococcal

enterotoxin B or C). TSS is diagnosed on clinical grounds characterized by high fever, rapidonset

hypotension, a diffuse erythematous rash that becomes desquamating 1 to 2 weeks

after onset, and involvement of three or more organ systems. After its initial description in

children, it was associated with menstruating women who were using highly absorbent

tampons. While the incidence of menstrual TSS decreased due to changes of the tampons’

absorbency and chemical composition, the frequency of nonmenstrual TSS entities has

remained constant (78). Although commonly no source of infection is confirmed,

nonmenstrual TSS is usually associated with focal postoperative wound or soft tissue

infections.

Staphylococcal food poisoning is caused by consumption of food contaminated with one or

more preformed, relatively heat-stable enterotoxins. Nausea, vomiting, abdominal cramps,

and diarrhea occur 2 to 6 hours after food ingestion. Symptoms usually subside 8 to 12

hours later.

SSSS is a type of bullous exfoliative dermatitis caused by exfoliative (epidermolytic) toxins

(ETA and ETB) (112). The syndrome is typically found in neonates and young children. In

addition to severe exfoliation affecting up to 90% or more of the entire body surface

(“Ritter’s disease”), a localized form (pemphigus neonatorum) with a few blisters is known.

Diagnosis is made on the basis of clinical features, including Nikolsky’s sign, in which the skin

wrinkles on gentle pressure.

The other coagulase-positive or -variable staphylococci are members of the skin microbiota

of various animal species and occasionally cause infections in their hosts. The members of

the S. intermedius/pseudintermedius/delphini cluster are the most common etiologic agents

of the canine pyoderma.S. hyicus is predominantly associated with the exudative epidermitis

(greasy pig syndrome) in pigs, S. schleiferisubsp.coagulans is found in dogs suffering from

external otitis, and S. aureus subsp. anaerobius is the etiological agent of abscess disease, a

specific lymphadenitis of sheep and goats. In humans, S.

intermedius/S. pseudintermedius appears to be occasionally responsible for canine-inflicted

wound infections, FBRIs, food poisoning, and invasive infections in immunocompromised

patients (19, 180). Only a few reports are known for human infections due to S.

schleiferi subsp. coagulans and S. hyicus.

Since 1980, CoNS have been increasingly recognized as nosocomial pathogens, especially S.

epidermidis. CoNS cause nosocomial infections in patients with predisposing factors such as

immunodeficiency and/or indwelling or implanted foreign polymer bodies

(84, 156, 194, 198). CoNS are less often implicated as the cause of infections of natural

tissue. CoNS are the most common cause of nosocomial bloodstream infection typically

associated with central and peripheral intravascular catheters (205). Most important in the

pathogenesis of FBRIs is the ability of CoNS to colonize the surface of the device by the

formation of a thick, multilayered biofilm (142). S. epidermidis is the predominant cause of

infections associated with prosthetic vascular grafts, prosthetic orthopedic devices, and

cerebrospinal fluid shunts. CoNS are frequently isolated causative agents of prosthetic-valve

endocarditis; rarely they are involved in infections of (previously damaged) native valves

(ca. 5%) (113). A right-sided native valve endocarditis is observed in intravenous drug

abusers. Virtually any other surgically inserted materials and devices may become infected

by CoNS. They account for 45 to 75% of all late-onset bloodstream infections in preterm and

low-birth-weight neonates in neonatal intensive care units (43).

In addition to prosthetic-valve endocarditis, unusually fulminant cases of native-valve

endocarditis may be caused by S. lugdunensis, characterized by an aggressive clinical course

with high mortality (8). Thus, patients with S. lugdunensis bacteremia should be carefully

examined for signs of endocarditis. Besides other invasive infections, this organism is also a

common pathogen involved in FBRIs (167). S. lugdunensis infections resemble those caused

by S. aureus rather than those caused by other CoNS.

Based on special urotropic and ecologic characteristics, S.

saprophyticus subsp. saprophyticus is a well-documented causative agent of acute,

commonly recurrent, urinary tract infections in young, otherwise healthy, sexually active

women and, less frequently, in young men or boys. This pathogen is the second most

common (after Escherichia coli) cause of uncomplicated cystitis among young women. While

colony counts of ≥100,000 CFU/ml in two or more cultures of midstream urine usually

indicate significant bacteriuria, lower counts may be significant in the symptomatic patient.

Infections due to the recently described subspecies S. saprophyticus subsp. bovis have not

been reported.

Since human infections due to Macrococcus, Jeotgalicoccus,

Nosocomiicoccus, and Salinicoccus have not been described, these genera are not discussed

further in this chapter.

Micrococcaceae and Dermacoccaceae

While “micrococci” are generally acknowledged as harmless saprophytes, they can also act

as opportunistic pathogens. Micrococcus, Kocuria, and Kytococcus species have been found

to cause infections such as endocarditis, pneumonia, and sepsis or FBRIs predominantly in

immunocompromised patients (3, 25, 166). Recovery of the more recently described

micrococcal species associated primarily with the environment must be assessed for clinical

significance as reported for K. rhizophila (23).

R. mucilaginosa has been implicated in cases of bacteremia, endocarditis, endophthalmitis,

intravascular catheter-related and central nervous system infections, pneumonia, peritonitis,

septicemia, and cervical necrotizing fasciitis (75, 76, 116).

Since human infections due to Acaricomes, Citricoccus, Luteipulveratus, Nesterenkonia,

Renibacterium, Sinomonas, Yimella, and Zhihengliuella have not been reported, these genera

are not discussed further in this chapter.

Alloiococcus

A. otitis has been associated with infections of the middle ear (114). While

immunostimulatory capacity suggests that A. otitis has pathogenic potential (80), other

studies revealed that A. otitis may be a commensal rather than a cause of otitis media (182).

COLLECTION, TRANSPORT, AND STORAGE Back to top

The general principles of collection, transport, and storage of specimens as given in chapters

9 and chapter 16 of this Manual are applicable to the microorganisms listed in this chapter.

No special methods or precautions are usually required for these organisms because they are

easily obtained from clinical material of most infection sites and are relatively resistant to

drying and to moderate temperature changes. While some staphylococcal species may

require anaerobic conditions or CO2 supplementation for satisfactory growth, they survive

transport and limited storage in air.

DIRECT EXAMINATION Back to top

The direct microscopic examination of normally sterile fluids such as cerebrospinal fluid and

joint aspirates may be helpful. Direct examination of nonsterile fluids may also be useful, if

the presence of inflammatory cells versus epithelial cells is taken into consideration. As the

result of direct microscopic examination, only a presumptive report of “gram-positive cocci

resembling staphylococci” should be made. Cells of microorganisms discussed here are

gram-positive, nonmotile, non-spore-forming cocci that are arranged mostly in pairs and

tetrads but occur also singly, in irregular (grape-like) clusters or in short chains (three or

four cells). However, within the Micrococcaceae and Dermacoccaceae, some species exhibit

rod-shaped cells and have been shown to be motile.

Rapid PCR-based approaches have been introduced for detection of MRSA directly from

surveillance swabs. The multiple-locus approach, detecting the mecA gene and additionally

an S. aureus-specific target (see “Species Identification by Nucleic Acid-Based and

Spectroscopic Approaches” below), may be influenced by the coexistence of MSSA and MRCoNS

in the patient’s physiological microbiota and thus may lead to false-positive MRSA

findings (21). Nevertheless, the fundamental advantage of this approach is the direct

detection of the methicillin resistance-encoding mecA gene. Tests applying this principle,

such as hyplex Staphylo Resist (plus) (BAG Health Care, Lich, Germany) and StaphPlex Panel

(Qiagen), are commercially available.

The alternative single-locus amplification strategy overcomes the MSSA/MR-CoNS

coexistence drawback by using oligonucleotide primers binding on the staphylococcal

cassette chromosome (SCCmec) right extremity and on the neighboring orfX region of the S.

aureus chromosome, amplifying both a taxonomic marker and a resistance marker in one

step (89). This principle is the basis for several rapid test systems (e.g., BD GeneOhm MRSA

assay; BD, Franklin Lakes, NJ; GenoType MRSA Direct and Geno-Quick MRSA; Hain

Lifescience; Xpert MRSA; Cepheid, Sunnyvale, CA; and LightCycler MRSA Advanced Test;

Roche, Basel, Switzerland). However, the use of the surrogate marker SCCmec region

instead of mecA as target may lead to false-positive or false-negative results, e.g., due to

the exchange of the mecA gene by other genes, partial excision of the cassette, and

variability of the cassette primer binding sites (53). Overall, the amplification-based rapid

MRSA screening assays are characterized by very good negative predictive values

(approximately 97 to 99%) and are hampered by moderate positive predictive values

(approximately 65 to 95%). To date, MRSA cultures remain essential for confirming the

molecular results, for typing purposes, and for determination of the complete susceptibility

profile.

ISOLATION PROCEDURES Back to top

Considering the widespread distribution of staphylococci and “micrococci” on the skin and

mucous membranes, careful procedures should be used to isolate organisms from the

presumed focus of infection without collecting surrounding microbiota. The basic procedures

for culture and isolation described in chapter 16 of this Manual should be followed.

The primary culture plate used for the isolation of staphylococci from clinical specimens is

Columbia blood agar containing 5% defibrinated sheep blood (see also chapter 17).

Abundant growth of most staphylococcal species occurs within 18 to 24 h. The simultaneous

use of an enrichment broth (e.g., dextrose broth) streaked after 24 and 48 h on Columbia

blood agar may enhance the recovery rate of S. aureus and other staphylococci.

The use of selective agars for S. aureus such as mannitol salt agar, egg yolk-tellurite

pyruvate containing Baird-Parker medium, Columbia colistin-nalidixic acid agar, lipase-saltmannitol

agar (Remel, Lenexa, KS), and phenylethyl alcohol agar may be appropriate for

specimens from heavily contaminated sources such as feces. It is mandatory to confirm

putative S. aureus isolates recovered on these media.

Novel selective agars using chromogenic enzyme substrates specifically for S. aureus, such

as CHROMagar Staph aureus (CHROMagar, Paris, France), BBL CHROMagar Staph aureus

(BD Diagnostics, Sparks, MD), and S. aureus ID (bioMerieux, La Balme Les Grottes, France),

have been launched on the market with chromogen-dependent coloration of the colonies. In

particular for screening purposes, the chromogenic agars have been proven to be suitably

sensitive and specific, allowing a presumptive but not final identification (141). Chromogenic

agars designed for MRSA detection are discussed in “Antimicrobial Susceptibilities” below.

The diagnosis of catheter-related bloodstream infections by CoNS and other organisms

remains a major challenge. One of the most frequently studied diagnostic techniques is

represented by the semiquantitative roll-plate catheter culture. Here, the distal segment of

the central venous catheter is cut and rolled across the surface of a Columbia blood agar

plate at least four times followed by overnight incubation. A colony count of 15 CFU/ml or

more may indicate catheter colonization (118). Examination of paired quantitative blood

cultures drawn simultaneously from the catheter and a peripheral vein enhanced by the

analysis of differential time to positivity represents an example of an approach that does not

require catheter removal (148) (see also “Evaluation, Interpretation, and Reporting of

Results” below).

Cultivation of “micrococci” and R. mucilaginosa should be performed as described for

staphylococci on Columbia blood agar at 35°C to 37°C under aerobic conditions. However,

abundant growth of Micrococcaceae andDermacoccaceae needs consistent incubation times

of 36 to 48 h.

Because of the slow growth rate, it is difficult to isolate A. otitis by conventional nonselective

culture methods (29). Blood agar plates with 6% NaCl were shown to be useful.

IDENTIFICATION Back to top

The basic criteria distinguishing catalase-positive gram-positive cocci and their relatives

among themselves and from other microbial taxa are given in “Description of the Genera”

above and in Table 1. Misidentification is likely to occur if automated test system results are

accepted without critical review by skilled lab personnel. In specialized settings, species can

be identified by chemotaxonomic procedures and molecular methods.

Staphylococcus and Related Genera

Staphylococcus species can be identified phenotypically on the basis of a variety of

conventional characteristics (Tables 1 to 3). The most clinically significant species in human

and veterinary medicine can be identified on the basis of several key characteristics (Table

3). The application of the extensive scheme originally published by Kloos and Schleifer in

1975 (105) has been mostly replaced by the use of commercial identification systems.

Colony Morphology

Most staphylococcal colonies are 1 to 3 mm in diameter within 24 h and 3 to 8 mm in

diameter after 72 h of incubation in air at 34 to 37°C. Exceptions are S.

aureus subsp. anaerobius, S. saccharolyticus, S. auricularis, S. equorum, S. vitulinus, and S.

lentus, which grow more slowly and usually require 24 to 36 h for detectable colony

development. Further incubation of agar plates for a period of up to 48 to 72 hours

(optimally followed by 2-day incubation at room temperature) enhances morphologic

differences.

On routine blood agar, the typical S. aureus colony is pigmented (cream yellow to orange),

smooth, entire, slightly raised, and hemolytic (Fig. 1). Mucoid colonies due to highly

encapsulated strains are rarely encountered. A number of isolates of S. aureus as well as

some CoNS species (e.g., S. haemolyticus and S. lugdunensis) may have a hazy or distinct

zone of beta-hemolysis around the colonies ranging from weak to strong. SCVs of S.

aureus or other staphylococcal species are characterized by pinpoint colonies (1/10 the size

of the wild type), mostly nonpigmented and nonhemolytic after 24 to 72 hours of incubation

(Fig. 1) (147,192). They are often mixed with colonies displaying the normal phenotype,

thus giving the appearance of a mixed culture. Upon subculture they may remain stable or

revert to the wild type. Depending on their auxotrophy, normal growth may be restored if

the isolate is grown in the presence of hemin, menadione, or thymidine and/or

CO2 supplementation (147).



Coagulase Production

A widely used criterion for the identification of S. aureus in the clinical laboratory is the

clotting of plasma proven by two different tests: (i) detection of the extracellular free

coagulase by the tube test due to staphylococcal coagulase that converts fibrinogen to fibrin

and (ii) detection of the cell wall bound “coagulase” (i.e., the clumping factor) by the slide

agglutination test (see below).

The tube coagulase test is performed by transferring a large, well-isolated colony from a

noninhibitory agar into 0.5 ml of reconstituted rabbit plasma. It is crucial to incubate the

tube at 37°C for 4 h and to observe the tube for clot formation by slowly tilting the tube 90°

from the vertical. Any degree of clotting represents a positive test. Aflocculent or fibrous

precipitate is not a true clot and should be recorded as negative. If no clot is formed by 4 h,

the tube should be read again after 18 h of incubation. False-negative results may occur for

some strains producing staphylokinase, which may lyse the clot after formation (usually after

prolonged incubation). Inaccurate results may occur if nonsterile plasma is used or the

colony tested is not pure.

Particularly in veterinary microbiological laboratories, the other coagulase-positive or -

variable species (Table 1) must not be disregarded. The detection of free coagulase in

staphylococci obtained from human specimens is usually equated with the species

identification of S. aureus. However, for animal-inflicted wounds, additional testing should be

performed to provide identification beyond “coagulase-positive staphylococci.”

Agglutination Assays

The classical slide agglutination test detects bacterial aggregation of S. aureus and other

clumping factor-positive staphylococcal species in the presence of rabbit plasma through the

action of clumping factor A, a cell wall-associated adhesin for fibrinogen. The detection of

this factor expressed by non-aureus species (Table 1) may require the application of human

plasma. While this test is very quick to perform (less than 1 minute), several limitations

concerning sensitivity and specificity (e.g., masking of the factor by capsular polysaccharides

and autoagglutination by picking colonies from media with high salt concentrations) have to

be noted.

To overcome the disadvantages of the slide coagulase test (low sensitivity) and of the tube

coagulase test (long incubation time), rapid latex and hemagglutination assays allowing

presumptive identification of S. aureus have been developed. Besides detecting protein A

and clumping factor A, recent third-generation assays include monoclonal antibodies

recognizing the clinically most prevalent capsular polysaccharide serotypes 5 and 8 or other

structures (e.g., Pastorex Staph-Plus [Bio-Rad Laboratories, Hercules, CA]; Slidex Staph Plus

[bioMerieux, Marcy l’Etoile, France]; and Staphaurex Plus and Staphytect Plus [Oxoid,

Cambridge, United Kingdom]). The higher sensitivity (>98 to 100%) of the third-generation

tests has reduced their specificity (72 to 99%). False-positive reactions occur with some

CoNS strains (S. haemolyticus and S. hominis)possessing type 8 capsular polysaccharide or

owning a cell wall hemagglutinin (S. saprophyticus). When an isolate is suspected to be S.

aureus, negative slide tests should be confirmed by the tube coagulase test.

Identification of Species by Susceptibility Tests

Novobiocin resistance is intrinsic to S. saprophyticus and other CoNS species (Table 2) but is

uncommon in the other clinically important CoNS species of the “S. epidermidis group.” A

disk diffusion test for estimating novobiocin susceptibility can be performed using a 5-μg

novobiocin disk on Mueller-Hinton agar or tryptic soy sheep blood agar. With an inoculum

suspension equivalent in turbidity to a 0.5 McFarland opacity standard and incubation at 35

to 37°C overnight or up to 24 h, novobiocin resistance is indicated by an inhibition zone

diameter of ≤16 mm with any of these media.

Polymyxin B resistance is usually observed for isolates of S. aureus, S. epidermidis, S.

hyicus, S. chromogenes,and to a lesser extent, for some strains of S. lugdunensis. A disk

diffusion test may be performed using a 300-U polymyxin B disk on tryptic soy sheep blood

agar with the same test conditions as those described for novobiocin. Polymyxin B resistance

is indicated by an inhibition zone diameter of <10 mm.

Identification of Species by Biochemical Procedures

For speed, standardization, cost reduction, and convenience, the classical tests for

fermentation, oxidation, degradation, and hydrolysis of various substrates (details are

available in chapter 17) have been incorporated into commercial manual and automated

biochemical test systems (see below and chapter 3). They are often complemented by

simultaneously performed antimicrobial susceptibility testing, an advanced expert system,

and an interface to the laboratory informatics software. Results with conventional tests

(Tables 1 and 3) may be slightly different from those obtained with rapid biochemical test

systems due to the use of other, more sensitive indicators. Commercial identification

systems identify the staphylococci (and some other aerobic gram-positive cocci) of clinical

importance with an accuracy of 70 to >90%. For some systems, reliability depends on

additional testing as suggested by the manufacturer. Uncommon strains or phenotypic

variants (e.g., SCVs) may have altered patterns of biochemical reactions requiring molecular

testing for identification.

The API Staph (bioMerieux) strip represents an overnight method for manual identification of

primarily staphylococci for health care and product safety applications. Necessitating the

same incubation time and also fashioned in the strip format, the ID32 Staph (bioMerieux)

may be read manually as well as automatically by the bioMerieux ATB system. The

databases of both systems comprise more than 20 staphylococcal species with clinical

significance, some “micrococcal” species, and R. mucilaginosa. A rapid version allowing 2-h

identification of S. aureus, S. epidermidis, and S. saprophyticus is provided by the same

manufacturer (Rapidec Staph). The Vitek 2 (bioMerieux) system is a fully automated

platform that performs bacterial identification and antibiotic susceptibility testing. The VITEK

2 gram-positive identification card encompasses a total of more than 100 species, including

26 staphylococcal species, a small spectrum of “micrococci,” and R.

mucilaginosa;identification of CoNS usually requires 10 h.

The MicroScan product Pos ID family (Siemens Health-care Diagnostics, Deerfield, IL)

includes “Conventional” (overnight identification time), “Rapid” (2.5-h identification time),

and “Synergies plus” (2- to 2.5-h identification time, with key antimicrobial results in as little

as 4.5 hours) panels in a microtiter format. Identification of 19 clinical staphylococcal

species, some “micrococcal” species, and R. mucilaginosa is available with either manual or

automated processing on the autoSCAN-4 and WalkAway systems.

The BD BBL Crystal identification systems’ Rapid Gram-Positive ID kit (BD Diagnostic

Systems, Sparks, MD) is a three-row panel that may be read manually or with the BBL

Crystal AutoReader requiring a 4-hour incubation period. Besides 14 staphylococcal species,

the database encompasses a total of 88 taxa including M. caseolyticus, R. mucilaginosa, and

several “micrococci.” The BBL Crystal Gram-Positive ID system represents the overnight

incubation (18-h) version with an extended taxa profile of 121 gram-positive organisms. A

further enhanced taxa profile (covering about 200 taxa) is covered by the automated

nephelometry-based BD Phoenix Automated Microbiology System using one combination

panel with the identification substrates on one side and the antimicrobial agents on the other

side of the panel. For staphylococci, about 10 to 15.5 hours are required for complete results

(identification and antimicrobial susceptibility testing) (56).

The gram-positive aerobic bacteria database (339 taxa) of the Biolog Systems family (Biolog,

Hayward, CA) comprises 34 staphylococcal and 4 macrococcal (sub-)species and many

members of the Micrococcaceae andDermacoccaceae families not found in other commercial

phenotype-based systems. The system’s redox chemistry based on the utilization of a wide

variety of carbon sources is used to generate a “metabolic fingerprint” providing results in 4

hours or less and is available with different automation levels.

The Sherlock Microbial Identification System (MIDI, Newark, DE) represents an identification

system that automates microbial identification by combining cellular fatty acid analysis with

computerized high-resolution gas chromatography. In addition to a multitude of other

species, 30 staphylococcal species, M. caseolyticus,and several “ micrococci” are listed in the

fatty acid-based database. A new sample preparation method (Instant FAME) allows for rapid

identification from pure cultures in less than 15 minutes.

Most systems are fairly successful in differentiating S. aureus, S. epidermidis, and S.

saprophyticus, while the accurate identification of less common species is more variable

(152). Systems may fail in distinguishing commonly encountered staphylococcal species, in

particular if phenotypic variants or isolates recovered from livestock and food are tested

(18, 210). A verification of the identification result by a second, independent approach is

recommended for isolates with presumptive identification as S. aureus, particularly for

oxacillin-resistant strains. Additional tests should be performed on clinically significant

isolates with questionable identification results that impact patient management.

Species Identification by Nucleic Acid-Based and Spectroscopic

Approaches

Extraction of staphylococcal nucleic acids may be challenging due to the gram-positive

nature requiring special conditions for lysis of the cell wall. For this purpose, lysostaphin,

lysozyme, proteinase K, and achromopeptidase have been described.

Besides PCR fingerprinting techniques based on DNA sequence polymorphisms, speciesspecific

variable regions of universal genes or genes unique for S. aureus or other

staphylococcal species may serve as targets for identification and differentiation of

staphylococcal isolates. Assays based on the specific amplification of fragments of the

universal 16S and 23S rRNA genes and of their spacer sequences have been published

(18,41, 55, 92, 124, 173, 178). The 16S rRNA gene is also used as target for the in situ

detection and identification of S. aureus and S. epidermidis (111). Other universal DNA

targets shown to be useful for identification of staphylococci include the elongation factor

gene (tuf), the gyrase gene (gyrA), the manganese-dependent superoxide dismutase

gene (sodA), the glyceraldehyde-3-phosphate dehydrogenase-encoding gene (gap),and a

60-kDa heat shock protein (HSP60/GroE) (73, 74, 120, 144, 211).

Sequencing of selected universal phylogenetic marker genes represents the ultimate

approach to identify known and not yet described staphylococcal species. For 16S rRNA gene

sequencing, the regions between bp 70 to 300, 420 to 500, 1,000 to 1050, and 1,250

to1,290 (corresponding to nucleotides of the Escherichia coli16S rRNA gene) are useful to

determine sequence differences among the staphylococcal species. For differentiation of

staphylococcal subspecies, sequencing of partial rpoB gene sequences seems to be superior

to partial 16S rRNA gene sequencing (123). The recognized limitations of currently available

public sequence databases apply to sequencing of staphylococcal isolates (18).

The most popular and well-studied specific target for S. aureus identification is the nuc gene

that encodes thermostable nuclease (thermonuclease or TNase) (31, 204). PCR methods

targeting the nuc gene are highly specific for S. aureus. A specific PCR for the S.

intermedius nuc gene has also been described (24). Further specific targets used for

identification of S. aureus include the genes encoding clumping

factor (clfA), coagulase(coa), manganese-dependent superoxide dismutase (sodM, absent in

CoNS), the factors essential for the expression of methicillin resistance (femA and B), and for

MRSA only, the staphylococcal insertion element 431(108, 122, 188, 191, 211).

Misidentification using the fem factors may occur due to fem-negative S. aureusstrains and

CoNS with a gene structurally related to femA.

For molecular identification of S. aureus and some other staphylococcal species isolated from

culture, several commercial tests (e.g., GenoType Staphylococcus [Hain Lifescience, Nehren,

Germany]; StaphPlex Panel [Qiagen, Germantown, MD]; AccuProbe [Gen-Probe, San Diego,

CA]; and S. aureus Evigene [AdvanDx, Woburn, MA]) are available. Some assays also detect

resistance genes (see below) and/or toxin genes. The GeneXpert (Cepheid) and BD

GeneOhm (BD Diagnostics) instruments offer assays to detect MSSA and MRSA directly from

positive blood cultures. The RiboPrinter microbial characterization system (DuPont Qualicon,

Wilmington, DE) utilizes ribotype pattern analysis for the differentiation of staphylococcal

species with patterns in the database.

The S. aureus PNA FISH (bioMerieux) is a qualitative nucleic acid hybridization assay

targeting rRNA sequences based on peptide nucleic acid (PNA) fluorescence in situ

hybridization (FISH). The assay is intended for rapid identification of S. aureus in a smear

prepared from a positive blood culture. The S. aureus/CNS PNA FISH is expanded to identify

simultaneously several CoNS species that commonly cause bacteremia (83). Diagnostic DNA

oligonucleotide microarrays that identify the genus Staphylococcus, clinically important

staphylococcal species, other pathogens, drug resistance genes, and toxin genes have been

designed and tested on clinical isolates (66, 129).

Alternative high-throughput approaches involving mass spectral analysis of surfaceassociated

proteins of intact staphylococcal cells are represented by the matrix-assisted laser

desorption ionization–time of flight and related methods (32, 150, 206). Here, the quality of

the database and the standardization of variable parameters are crucial to achieve

reproducible results. Nondestructive techniques such as Fourier transform infrared (FT-IR)

and Raman spectroscopy are currently being developed as alternative methods for the rapid

identification of staphylococci (6). This approach also allows discrimination between the SCV

and the normal S. aureus phenotype (20).

Diagnosis of Toxin-Mediated Staphylococcal Syndromes

The diagnosis of TSS and SSSS is based on clinical signs supplemented by serologic tests

and the detection of the toxin production by staphylococcal isolates (S. aureus, rarely other

species). While the skin manifestations are mostly culture negative, isolates are usually

recovered from the suspected site of infection. Blood cultures are positive in fewer than 5%

of cases of staphylococcal TSS. In patients with TSS, protective antibodies against causative

PTSAgs are absent or present at very low levels. However, serocon-version after onset of the

disease and during convalescence may be observed. The same phenomenon holds true for

exfoliatin antibodies in patients with SSSS. Currently available immunoassays for antibody

detection are for research use only. Beyond the recognition of the characteristic rapid onset

and the clinical signs, staphylococcal food poisoning is difficult to verify because the

incriminated food source may not contain cultivable staphylococcal cells and requires

detection of staphylococcal enterotoxin.

Traditional immunological procedures may be used to measure the toxin in culture

supernatants of isolated strains, in contaminated food extracts, or in patient specimens. Kits

for the detection of strains producing TSST-1 (TSTRPLA [Oxoid, Cambridge, United

Kingdom]; TST-RPLA “Seiken” [Denka Seiken, Tokyo, Japan]; and TSST-1 Evigene

[AdvanDx, Woburn, MA]), staphylococcal enterotoxins (Rida Screen set A, B, C, D, E [RBiopharm,

Darmstadt, Germany]; SET-RPLA “Seiken” [Denka Seiken]; SET-RPLA Kit toxin

detection kit [Oxoid]), and ETA/ ETB (EXT-RPLA “Seiken” [Denka Seiken]) are offered. A

quantitative real-time immuno-PCR was recently described for the detection of small

amounts of enterotoxins (61). An adaptation of a flow cytometry-assisted multiplex

immunoassay (Bio-Plex system; Bio-Rad, Hercules, CA) for the detection of ETA and ETB has

been recently reported (94). Since phenotypic methods may be hampered by low sensitivity

and specificity (cross-reactivity between PTSAgs) and in vitro expression of PTSAgs might be

negatively influenced by various factors, detection of PTSAg/ET-encoding sequences by

nucleic acid-based methods has come into favor (17,22, 128, 209).

Due to renewed interest in Panton-Valentine leukocidin (PVL), PCR procedures targeting

leukocidal (synergohymenotropic) pore-forming toxins produced by S. aureus were

established (115). The detection of PVL-encoding genes is included in the GenoType

Staphylococcus and MRSA test systems (Hain Lifescience) and offered by the PVL EVIGENE

kit (AdvanDx, Woburn, MA).

Micrococcaceae and Dermacoccaceae

Micrococci and staphylococci might be easily confused with one another on the basis of

similar cellular morphologies, Gram stain appearance, and positive catalase activities. The

exact species affiliation of “micrococci” may be frequently misjudged. The frequent

pigmentation of micrococcal colonies with high convex profile leads to their presumptive

identification as members of the Micrococcaceae and Dermacoccaceae families. Colonies

of M. luteus, K. varians, and the kytococcal species are characterized by yellowish tints. K.

kristinaeand D. nishinomiyaensis appear orange-like, and K. rosea shows pink to red

colonies. Some species (e.g., M. lylae) or strains of the usually pigmented species are nonpigmented.

On routine blood agar, R. mucilaginosacolonies are mucoid or sticky, transparent

to white, and nonhemolytic and in the majority of cases adhere to the agar (differentiation

from streptococci). This organism may be distinguished from other similar organisms by its

weak catalase reaction and its inability to grow in the presence of 5% NaCl.

In the clinical laboratory, “micrococcal” species can be preliminarily distinguished from

staphylococci by their resistance to furazolidone (100 μg/disk; resistance is indicated by a

≤9-mm zone diameter) and lysostaphin (200 μg/disk; resistance, no zone) and susceptibility

to bacitracin (0.04 U/disk; susceptibility, ≥10-mm zone diameter) in contrast to members of

the Staphylococcus genus, which show inverse susceptibility patterns (zone diameters:

furazolidone, ≥15mm; lysostaphin, 10 to 16 mm; bacitracin, no zone) (Table 1). Details are

available in previous editions of this Manual. In contrast to most staphylococci, micrococci

are positive by the modified oxidase test (59). Regarded as a reference method to

distinguish “micrococcal” species from staphylococci, the fermentation of glucose in a

manner similar to the oxidation-fermentation test for nonfermenters requires a specific

oxidation-fermentation medium and prolonged incubation. In contrast to staphylococci,

“micrococci” are characterized by the lack of acid production from glucose under anaerobic

conditions.

Key features for differentiation of species reported to occur in human specimens are given

in Table 4. Data concerning the applicability of manual and automated identification systems

for members of Micrococcaceae andDermacoccaceae are given in the

respective Staphylococcus section; however, their use is limited to a small spectrum of the

clinical “micrococcal” species. In cases of doubt and extraordinary clinical relevance, the use

of sequencing-based approaches is recommended for definite species recognition.



Alloiococcus

After 48 h of incubation at 37°C, alloiococci form small alpha-hemolytic colonies on blood

agar. Colonies formed on brain heart infusion agar with 5% rabbit blood are small, moist,

and slightly yellow at 72 h, and the blood is partially hemolyzed. Growth occurs in the

presence of 6.5% NaCl and on bile esculin agar. No growth occurs at 10°C or 45°C. A.

otitis can be distinguished from similar organisms by its positive catalase and negative

oxidase activities, its obligate aerobic nature, and its inability to produce acid from glucose or

other carbohydrates (Table 1). Arginine dihydrolase is not produced. Pyrrolidonyl

arylamidase, leucine aminopeptidase, and P-galactose are produced. Starch and esculin are

not hydrolyzed; hippurate is mostly hydrolyzed (29). A. otitis is included in the database of

the API Strep gallery (bioMerieux).

As the bacterium is quite inert biochemically, molecular methods are often necessary to

confirm the identification. For the molecular verification of suspected colonies, an A. otitisspecific

PCR assay has been described (1). A PCR assay for direct detection of this

microorganism in clinical specimens has been reported as part of a multiplex approach

targeting pathogens that cause otitis media with effusion (82).

TYPING SYSTEMS Back to top

Traditional phenotyping techniques such as phage typing, capsule serotyping, antibiotic

susceptibility pattern analysis, and other biotyping methods have been replaced by molecular

band-based and sequence-based typing methods (179). The reference method for defining

the core genetic population structure of S. aureus is MLST. In a standardized manner, the

allelic polymorphism of seven housekeeping genes is indexed. A web-based database is

available (http://saureus.mlst.net/). MLST has limited discriminatory power and low

throughput capacity in the context of MRSA outbreak investigation and surveillance. In the

case of common ancestry but assumed distinct epidemiological origin of MRSA isolates,

subtyping of SCCmec may provide additional information (155).

For local MRSA outbreak investigation as well as for long-term MRSA surveillance, SmaI

macrorestriction pattern analysis by PFGE represents a highly discriminatory “gold standard”

tool with detailed performance and interpretation guidelines (133, 183). However, PFGE is

technically demanding with low throughput, limited portability, problems with inter-center

reproducibility, and different national nomenclatures. The emerging livestock-associated

ST398 isolates are mostly nontypeable by standard PFGE protocols (88). Attribution of PFGE

clusters to genetic lineages may be problematic (179).

The number of polymorphisms and the sequence of tandem repeat elements of the

hypervariable X region of the S. aureus protein A (spa) gene are the basis of a single-locus

sequence typing approach that has become one of the primary genotyping methods for

MRSA surveillance (65, 170). Beside total reproducibility, other advantages include low costs,

high throughput, a standardized nomenclature, and complete portability of data transferable

into an international database (http://spaserver.ridom.de/) curated by SeqNet.org

(http://www.seqnet.org/) (81). For particular genetic lineages, misclassification may occur,

necessitating the use of additional tests for reliable inference.

Other molecular typing techniques used for typing of staphylococci include several other

band-based molecular fingerprinting approaches, ribotyping, and more recently, multiplelocus

variable-number tandem-repeat analysis and microarray-based approaches

(110, 160, 185). Modern approaches based on phenotype structures applied for

staphylococci include the matrix-assisted laser desorption ionization– time of flight/mass

spectrometry and whole-cell fingerprinting techniques, such as the FT-IR spectroscopy (7).

While many typing systems have been developed and evaluated for S. aureus, fewer

applications are available for CoNS. These include antibiotic resistance analysis, phage

typing, slime production, and some modern genotyping procedures (PFGE and ribotyping)

(69, 190).

For “micrococcal” species, phage typing and PFGE approaches applying phage sets and

restriction enzymes, respectively, different from those used for staphylococci (132) have

been described. A restriction fragment length polymorphism approach has been described

for Alloiococcus (29).

SEROLOGIC TESTS Back to top

Because serological testing for antistaphylococcal antibodies lacks specificity and predictive

accuracy, it plays no role for the diagnosis of most staphylococcal diseases. The one

exception is the determination of protective antibodies in the case of toxin-mediated

syndromes such as TSS and SSSS (see above). For the other microorganisms discussed in

this chapter, detection of antibodies is not clinically useful.

ANTIMICROBIAL SUSCEPTIBILITIES Back to top

Staphylococcus and Related Genera

Genetic Basis and Prevalence of Antimicrobial Susceptibilities

Methicillin resistance mediated by the mecA gene has the greatest impact on patient

management by excluding all traditional β-lactam antibiotics from the antibiotic

armamentarium. Since the early 1980s, the prevalence of health care-associated MRSA (HAMRSA)

has increased in many regions. In some areas of the United States, the prevalence of

MRSA is >50% (50, 101). In Europe, with the low-prevalence exception of The Netherlands

and the Scandinavian countries, approximately 20% or more of S. aureus isolates are

methicillin resistant (177). In Asia, Australia, and Africa, high MRSA rates (approximately 20

to 80%) have also been noted (158, 169, 212). The worldwide burden of infections caused

by MRSA is increasing due to the advent of CA-MRSA in the past decade and, most recently,

by livestock-associated S. aureus infections (85, 109). A number of studies and reviews

describing antimicrobial and biocide susceptibilities of clinically important staphylococcal

species have been published (51, 64, 87). Networks have also been established that provide

online or published resistance data for staphylococci: the National Nosocomial Infections

Surveillance System (http://www.cdc.gov/ncidod/dhqp/), the European Antimicrobial

Resistance Surveillance System (http://www.rivm.nl/earss/), and the International

Nosocomial Infection Control Consortium (http://www.inicc.org/).

The mecA gene is acquired by S. aureus and other staphylococcal species on a foreign,

mobile DNA element (SCCmec) and encodes an additional penicillin-binding protein (PBP),

PBP2a. Seven major variants of SCCmec, types I to VII, and many subtypes have been

recognized (27, 44, 90, 91, 136). SCCmec types I, II, III, and VI are predominantly

associated with HA-MRSA (47).

Multidrug resistance is regularly observed in HAMRSA and usually includes resistance to

aminoglycosides, fosfomycin, fusidic acid, glycopeptides, ketolides, lincosamides, macrolides,

quinolones, rifampin, tetracyclines, and trimethoprim- sulfamethoxazole (51, 62, 87). There

are rare reports of S. aureus and CoNS isolates resistant to new agents such as linezolid,

daptomycin, and tigecycline (54, 143, 171).

In contrast to HA-MRSA, CA-MRSA strains are more susceptible to non-β-lactam antibiotic

classes and harbor different SCCmec (mostly types IVa, V, and VII) (13, 27, 44, 91). CAMRSA

isolates are more likely to carry PVL, a pore-forming toxin with potent cytolytic and

inflammatory activities (72). Whereas in Europe an array of diverse CA-MRSA clones (mainly

the “European” ST-80, spa type t044) have been reported, the PFGE type USA300 (ST-

8, spa types t008 and t024) predominates in the United States followed by USA400

(47, 101). Populations at increased risk for CA-MRSA infections include children in day care

centers, athletes, military recruits, jailed inmates, and men who have sex with men

(52, 97, 138). CA-MRSA causes superficial skin and soft tissue infections and can also be

associated with necrotizing fasciitis and myositis, necrotizing pneumonia, and other severe

entities (115). Infections by PVL-positive S. aureus in young, otherwise healthy children

following a respiratory viral infection (most frequently influenza) can be a devastating

disease (72). However, some of the distinctions between HA-MRSA and CA-MRSA strains

may disappear since CA-MRSA is now becoming endemic in hospitals and acquiring additional

resistances (52, 162, 168).

Since 1996, vancomycin-intermediate S. aureus (VISA) isolates (MIC of 4 to 8 μg/ml) (39)

and their putative precursors, termed heterogeneous VISA (hVISA) strains, have been

identified first in Japan followed by detection in the United States and other regions

(33, 86, 154). Reduced susceptibility is thought to be caused by cell wall alterations resulting

in reorganization and thickening (30, 42). In 2002, the first of multiple vancomycinresistant

S. aureus (VRSA) strains containing the vanA gene were reported in the United

States (34). S. aureus isolates currently defined as vancomycin-resistant exhibit MICs of ≥16

μg/ml (39). Since most VISAs are also resistant to teicoplanin, the acronym GISA

(glycopeptide-intermediate S. aureus) is preferred by some authors. A few clinical isolates

that are resistant to teicoplanin but fully susceptible to vancomycin have been reported (57).

More than 90% of all of nosocomial CoNS (mainly S. epidermidis and S. haemolyticus) are

resistant to penicillin due to β-lactamase production, and approximately 60 to 80% are

resistant to methicillin and other agents (51,93, 121). Rare acquisition of mecA-encoded

methicillin-resistance in S. intermedius/S. pseudintermedius has been reported (15).

Determination of Antimicrobial Susceptibilities

Antimicrobial susceptibility testing of staphylococci may be performed conventionally by

Clinical and Laboratory Standards Institute (CLSI) reference methods (38) or commercial

systems as described in chapter 67 to 70 of this Manual. Direct detection of MRSA is

discussed in “Direct Examination” above. For SCVs, no approved method has been developed

to determine the susceptibility (146, 192).

Detection of MRSA represents the most important task in determining the antimicrobial

susceptibilities of staphylococci. To distinguish MRSA from MSSA, traditional methods may

have reduced sensitivity and specificity due to heteroresistance and borderline oxacillinresistant

S. aureus characterized by β-lactamase hyperproduction. In the case of

heterogeneous PBP2a expression, only a small fraction of the bacterial cell population

(10−8 to 10−4) expresses the resistance phenotype under in vitro test conditions. The

resistant subpopulation may be overlooked because it usually grows more slowly than the

susceptible population. The successful detection of heteroresistant strains is favored by

cooler incubation temperatures (30 to 35°C), the presence of NaCl (2 to 4%), and prolonged

incubation time (up to 48 h). Cefoxitin (30 μg) disk diffusion tests correlate better with the

presence of mecA than do the oxacillin disks used previously. The cefoxitin disk diffusion

method may be used for testing S. aureus and S. lugdunensis (resistance is indicated by a

zone diameter of ≤21 mm) as well as other CoNS (resistance, ≤24 mm) for mecA-mediated

resistance (60, 172). For broth microdilution testing, oxacillin or cefoxitin may be used to

detect mecA- mediated resistance in S. aureusand S. lugdunensis (resistance: cefoxitin, MIC

≥ 8 μg/ml; oxacillin, MIC ≥ 4 μg/ml). For other CoNS isolates, the presence of mecA is

predicted by applying lower oxacillin MIC breakpoints (resistance: oxacillin, MIC ≥ 0.5

μg/ml). Cefoxitin has also improved the detection of MRSA by automated susceptibility

testing systems (157). The slow growth rate of SCVs prevents the use of disk diffusion and

automated methods to determine the susceptibility of these strains (147).

An alternative method for detection of methicillin resistance is the use of anti-PBP2a

monoclonal antibodies available as a latex agglutination assay that may be performed on

isolated colonies from a pure culture (MRSA Screen; Denka Seiken, Tokyo, Japan) (134). If

the latex test is used for SCVs, then the number of colonies must be increased 100-fold

(100).

For heterogeneous methicillin-resistant staphylococci, strains displaying growth-impaired

phenotype (SCVs) or borderline oxacillin-resistant S. aureus strains, the methicillin

resistance of cultivated staphylococcal isolates may be determined by detection of

the mecA gene (131). For the genetic verification of MRSA isolates, the occurrence of both

the mecA gene and a species-specific marker (see above) has to be proven. The use of pure

colony material is a vital premise for this approach.

For MRSA screening to detect colonization (preferentially nasal, also pharyngeal), selective

media (e.g., mannitol salt agar) supplemented with oxacillin are widely used. Inclusion of a

broth enrichment step prior to plating enhances sensitivity but delays results. Media

containing chromogenic enzyme substrates (e.g., MRSA ID [bioMerieux]; BBL CHROMagar

MRSA [BD]; and Brilliance MRSA agar [Oxoid]) have better specificity; however, confirmation

of MRSA with a coagulase test is recommended for some products (36, 49, 140).

Detection of VISA is unreliable and probably underreported by routine susceptibility testing

methods, including automated methods (9). According to current CLSI recommendations, S.

aureus isolates with vancomycin MICs of 4 to 8 μg/ml are classified as VISA (39). These

vancomycin MIC breakpoints were lowered for S. aureus (the intermediate category

remained defined as 8 to 16 μg/ml for CoNS) in order to better detect hVISA strains that are

potentially associated with vancomycin clinical failure (35, 184). Detection of hVISA requires

a “modified population analysis profile–area under the curve” method or a macroEtest

method (201, 207). Recently, FT-IR spectroscopy has been successfully applied for rapid and

accurate identification of VISA and hVISA among isolates of MRSA (5). VRSA strains (MIC,

≥16 μg/ml) are reliably detected by the broth microdilution reference method, most FDAcleared

automated systems, or a brain heart infusion vancomycin (6 μg/ml) agar screen

plate (39). In 2009, the European Union Committee on Antimicrobial Susceptibility Testing

reduced the glycopeptide MIC breakpoints (resistance: vancomycin, >2 μg/ml for S.

aureus and CoNS; teicoplanin, >2 μg/ml for S. aureus and >4 μg/ml for CoNS) to avoid

reporting VISA isolates as “intermediate”

(http://www.srga.org/eucastwt/MICTAB/MICglycopeptides_v2.html).

Additional information on the determination of antimicrobial susceptibilities is contained

in chapters 67 to 70 of this Manual. There are no CLSI methods for susceptibility testing of

“micrococci” and alloiococci.

Treatment

For additional and annually updated information regarding treatment, the reader should

consult the current edition of The Sanford Guide to Antimicrobial Therapy (71) or other

guidelines. Effective treatment of focal infections such as empyema and abscesses requires

incision and drainage. Penicillin G is the most effective compound for the treatment of the

uncommon penicillin-susceptible S. aureus strain. In general, penicillin-resistant, oxacillinsusceptible

staphylococcal strains should be treated with penicillinase-stable penicillins, β-

lactam/β-lactamase inhibitor combinations, and cephalosporins (151). For patients with

penicillin allergy or chronic renal failure, clindamycin or vancomycin may be an option in the

case of MSSA. However, the use of vancomycin, known to be poorly bactericidal against

staphylococci, is not recommended for severe infections due to MSSA as it is inferior to β-

lactams in terms of mortality and bacteriological outcome.

Strains that are oxacillin- or cefoxitin-resistant (MRSA) should be considered resistant to all

β-lactams including penicillins, carbapenems, and cephalosporins (except for the new “fifthgeneration”

cephalosporins [ceftobiprole and ceftaroline] with anti-MRSA activity) (77).

Vancomycin and the newer agents such as linezolid, daptomycin, dalbavancin, and

tigecycline are suitable options for empiric therapy of MRSA infections (note the different

spectra and the approved indications of these compounds) (63, 161, 208). Occasionally,

trimethoprim-sulfamethoxazole (SXT) may be helpful, but it should be used with caution

(145). In particular in terms of “collateral damage,” quinolones should be avoided for the

therapy of staphylococcal infections (139).

While there are only a few, partly uncorroborated studies and case reports available

supporting combination therapy for treatment of severe staphylococcal infections,

aminoglycosides, rifampin, fosfomycin, co-trimoxazole, and fusidic acid in combination with

glycopeptides and β-lactams have been recommended (70). A careful risk-benefit

assessment concerning drug-drug interactions and side effects should be performed. Since

resistance towards rifampin, fusidic acid, and fosfomycin develops rapidly, these compounds

must not be administered alone.

Since the majority of clinically recovered CoNS strains are methicillin resistant, most

infections by CoNS require treatment with vancomycin or, where appropriate, the new

agents described above. Replacement of these by β-lactamase-resistant penicillins is

advisable for methicillin-susceptible isolates. When used simultaneously, antibiotics with cell

wall activity (β-lactams and vancomycin) combined with rifampin were shown to act

synergistically; however, this combination is not recommended for catheter-related

bloodstream infections (125). FBRIs remain a therapeutic challenge and frequently require

removal of the device (125, 198).

Prospective studies on the most appropriate treatment for patients infected with

staphylococcal SCVs are unavailable. A reduced susceptibility to aminoglycosides can be

expected (16). Resistance to trimethoprim-sulfamethoxazole is observed in thymidineauxotrophic

SCVs (96). Due to the fact that SCVs may persist intracellularly, a combined

treatment regimen of rifampin (intracellular activity) with either β-lactam antibiotics or

vancomycin (for methicillin-resistant SCVs) may be effective.

Micrococcaceae and Dermacoccaceae

Systematic data on susceptibilities of the two

families Micrococcaceae and Dermacoccaceae are rare, and the real species affiliation in

older reports is often unclear. Members of the genera Micrococcus and Kocuria appear to be

susceptible to β-lactams, macrolides, tetracycline, linezolid, rifampin, and the glycopeptides;

however, clinical isolates resistant to these agents have been reported (25, 95, 197). While

most kytococcal isolates have been susceptible to carbapenems, gentamicin, ciprofloxacin,

tetracycline, rifampin, and glycopeptides, kytococci are usually resistant to penicillin G,

cephalosporins, and oxacillin (not mecA- based) (25, 95, 174). An antibiotic regimen that

has been suggested for the treatment of infection by members of both families is a

combination of vancomycin with rifampin and gentamicin (199). R. mucilaginosa appears to

be variable in its antimicrobial susceptibility (197, 200). The observation that R.

mucilaginosa exhibits poor to no growth on Mueller-Hinton agar makes susceptibility testing

a challenge for clinical laboratories. A supplementation with 5% sheep blood and incubation

in 6% CO2 may enhance susceptibility testing (197, 200).

Alloiococcus

A. otitis has been reported as susceptible to ampicillin, cefotaxime, tetracycline, and

vancomycin but resistant to macrolides, azithromycin, and co-trimoxazole (10, 29, 46). In a

ddition to gentamicin-resistant isolates, some A. otitis isolates with intermediate levels of

resistance to β-lactams have been reported, although they are β-lactamase negative.

EVALUATION, INTERPRETATION, AND REPORTING OF

RESULTS Back to top

Distinguishing contaminants and colonizers from staphylococcal and “micrococcal” isolates

causing infection continues to be an important challenge for laboratorians and clinicians. It is

imperative to have an appreciation of the quality of the specimen under consideration.

Clinical features and the results of other investigations should be taken into account during

the interpretative process. There is no replacement for good communication between

laboratory staff and primary physicians.

Cultivation and identification of the causative pathogen to the species level represent the

“gold standard” for the diagnosis of staphylococcal infections. The most critical step when

interpreting a culture that is suspect for staphylococci is to distinguish between S.

aureus and other species. Due to the importance of the report of “S. aureus” and, in

particular, of “MRSa” for prognosis, therapy, hospital hygiene, and infection control, any

uncertainty regarding the species identification or the susceptibility to oxacillin (cefoxitin)

should be investigated via a second independent method. Considering costs and rapidity,

applicable routine proceedings may comprise the use of respective latex agglutination assays

combined with the use of automated systems for identification and susceptibility testing. In

case of doubt, further tests, preferentially nucleic acid-based approaches, should be applied.

However, the need for additional procedures must be weighed against delay in informing the

physician concerning a preliminary S. aureus identification as the putative causative agent.

Species-level identification of CoNS associated with infection should be considered, though it

is still a matter for debate because species-level identification of CoNS rarely leads to an

intervention or to changes in therapy. To rule out S. saprophyticus for urine isolates, the

novobiocin test could be used. Isolates from deep-tissue infections and blood cultures of

patients with suspected endocarditis should be differentiated to the species level, since the

identification of S. lugdunensis raises the index of suspicion for aggressive disease. To rule

out S. lugdunensis for invasive and sterile-site isolates, its positivity for pyrrolidonase and

ornithine decarboxylase could be considered (Tables 2 and 3). Other CoNS have also been

increasingly described as causative pathogens of severe infection. Ribotyping and/or PFGE

should be used when determination of genetic relatedness is required. Contaminants and

colonizing CoNS do not require susceptibility testing or identification. Several reports suggest

that cultures of colonially indistinguishable CoNS may contain multiple different strains (67).

Extending the incubation period of the initial cultures to 72 h may enable different colony

types to be more readily identified.

Unless there is strong evidence to the contrary, isolation of S. aureus from a sterile-site

culture (such as aspirated pus, blood, or cerebrospinal fluid) should be considered clinically

significant. Contamination of high-quality samples by S. aureus is rare, and further samples

should be taken if there is clinical doubt. Interpreting the isolation of S. aureus from

specimens contaminated with elements of a normal microbiota requires consideration of

setting, clinical features, and recent interventions. Quantitative culture may be helpful, for

example, when interpreting a bronchoalveolar lavage or urine sample. In persons with S.

aureus endocarditis, bacteremia is continuous and associated with higher loads, while in

cases with transient bacteremia (e.g., manipulations with mucous membrane trauma)

bacteremia is associated with lower loads and typically has a short duration. Because

quantitative blood cultures are not routinely performed, the time between incubation onset

and growth detection (defined as the time to positivity) may provide additional information in

continuous blood culture-monitoring systems. Rapid growth of S. aureus (within 14 h after

the initiation of incubation) has been associated with a high likelihood of endovascular

infection, delayed clearance, and complications (98). Isolation of S. aureus from surgical

wounds and other sites such as ulcers may represent infection or colonization, and the

clinician ’s response to the culture report should be guided by a bedside assessment of signs

and symptoms of infection. Colonization alone is an insufficient reason to treat, unless the

patient is colonized by MRSA and decolonization is undertaken as part of a specific infection

control policy.

Compared to S. aureus, interpreting the significance of cultures that are positive for CoNS is

more challenging. CoNS are an important cause of nosocomial bloodstream infections, but

they are also the most common contaminants of blood cultures (79). It is obvious that

samples taken from colonized sites will contain members of these species. Blood samples or

biopsy specimens taken without careful skin cleaning and disinfection will become

contaminated with the microbiota of the skin or mucous membranes. However, even careful

attention to collection techniques will not prevent all episodes of contamination.

Interpretation of blood cultures positive for CoNS requires knowledge of the presence of

prosthetic material in the intravascular compartment, risk factors for true CoNS sepsis such

as prematurity or the presence of an impaired immune system, and clinical features of

sepsis. Factors helpful in distinguishing between true positive and contaminated cultures

taken from a patient with clinical features of infection include (i) isolation of a strain in pure

culture from the infected site or body fluid and (ii) the repeated isolation of the same strain

or combination of strains over the course of the infection (103, 186). An algorithm to reduce

misclassification of nosocomial bloodstream infections due to CoNS was defined as at least

two blood cultures positive for CoNS within 5 days or one positive blood culture plus clinical

evidence of infection (26).

The presence of the same CoNS strain on an intravenous catheter tip and in a blood culture

is supportive evidence for intravenous catheter-associated bacteremia. Measurement of

differential time to positivity between blood cultures drawn through the central venous

catheter and those drawn from the peripheral vein was reported as highly sensitive and

specific for the in situ diagnosis of catheter-related bloodstream infection in patients with

short- and long-term catheters; however, other studies did not show that this method was of

major diagnostic value (149, 153, 165). Isolation of CoNS from peritoneal dialysis fluid or

cerebrospinal fluid taken from ventricular shunts in a patient suspected of having infection is

usually significant. While most contaminated clinical specimens produce mixed cultures of

different strains and/or species, some infections may be attributable to more than one strain

or species (67).

For patients with complicated cases, who probably have a true CoNS infection, it is advisable

to develop a sampling strategy. This is particularly pertinent when dealing with patients with

low-grade infection associated with implanted prosthetic material such as a joint replacement

or vascular graft. Samples should include those from each anatomical layer or region, and

fresh instruments should be used to gather deep-site samples (11).

Susceptibility testing, and in particular the detection of methicillin resistance, should always

be carried out following the identification of S. aureus. For surveillance cultures to detect

MRSA colonization, susceptibility testing beyond the determination of methicillin resistance

may only be needed for patients undergoing decolonization to predict the success of

mupirocin therapy. All CoNS associated with true infection require susceptibility testing.

The considerations of clinical significance discussed for CoNS are also appropriate for

members of theMicrococcaceae and Dermacoccaceae families; however, the criteria used for

distinguishing etiologically relevant isolates from contaminants and colonizers, respectively,

should be applied much more strictly. Since species of the Kytococcus genus are resistant to

some β-lactams, they should be (if clinically relevant) carefully distinguished from other

“micrococci” that are usually susceptible.

No comments:

Post a Comment