INTRODUCTION — Shortly after the introduction of methicillin in 1959, outbreaks of methicillin-resistant Staphylococcus aureus (MRSA) infections were reported in the early 1960s [1]. Methicillin resistance in S. aureus is defined as an oxacillin minimum inhibitory concentration (MIC) 4 µg/mL. Isolates resistant to oxacillin or methicillin are also resistant to all beta-lactam agents including cephalosporins.
Methicillin resistance is mediated by the mecA gene, which encodes for an abnormal low-affinity binding protein, PBP-2a that permits the organism to grow and divide in the presence of methicillin and other beta-lactam antibiotics. (See "Mechanisms of antibiotic resistance in Staphylococcus aureus").
The evidence suggests that a single clone of MRSA accounted for most of the isolates recovered during the 1960s [2,3] and that five major MRSA clones had emerged worldwide by 2002 [4]. Dissemination of resistance was mediated by horizontal transfer of the mecA gene and related regulatory sequences [5]. Since that time, MRSA has increased in prevalence worldwide as both a nosocomial and, more recently, a community-associated pathogen.
The epidemiology and clinical manifestations of MRSA infection in adults will be reviewed here. The prevention and control and the treatment of MRSA infections in adults is discussed separately. (See "Prevention and control of methicillin-resistant Staphylococcus aureus" and see "Treatment of methicillin-resistant or vancomycin resistant Staphylococcus aureus infection in adults").
EPIDEMIOLOGY — MRSA was at first primarily a healthcare-associated (nosocomial) infection, with only small numbers of community-associated cases [4]. However, it is now an increasingly prevalent community-associated pathogen. Such patients acquire MRSA infection in the outpatient setting, and are infected at the time of hospital admission.
Community-associated (CA) MRSA isolates usually have different molecular and antimicrobial susceptibility characteristics from healthcare-associated (HA) MRSA. However, nosocomially acquired MRSA can spread to household and community contacts and be a source of community-associated infection [6]. In addition, CA-MRSA is emerging as a significant cause of nosocomial infection as illustrated by analysis of 132 cases of MRSA blood stream infection in patients collected at an urban public hospital in 2004 [7]. Molecular typing studies demonstrated that 39 of 116 isolates (34 percent) were MRSA USA300 genotype, the predominant cause of CA-MRSA infection in the area.
Genetics of methicillin resistance — Molecular typing techniques have provided insights into the epidemiology of MRSA and to differences between HA-MRSA and CA-MRSA. The genetic determinants of antibiotic resistance in MRSA are discussed in detail separately, but will be reviewed briefly here. (See "Mechanisms of antibiotic resistance in Staphylococcus aureus").
The mecA gene responsible for methicillin resistance is located on a mobile genetic element called staphylococcal cassette chromosome (SCCmec). Sequencing SCCmec from many MRSA strains has revealed that there are five SCCmec types (I-V) that vary in genetic makeup and size. Transfer of SCCmec from MRSA into well-adapted strains of methicillin-susceptible S. aureus (MSSA) has occurred on a number of occasions, resulting in new MRSA strains that spread rapidly in healthcare institutions. A majority of HA-MRSA clones have type I, II, or III SCCmec and are multidrug resistant [8].
In contrast, a majority of CA-MRSA strains have type IV SCCmec [9-11]. Because type IV SCCmec does not carry the multiple antibiotic resistance genes common among types I to III [10,11], CA-MRSA are usually resistant only to methicillin, other beta-lactam antibiotics, and erythromycin, but are usually susceptible to other classes of antibiotics, particularly trimethoprim-sulfamethoxazole and clindamycin [10,12-16]. CA-MRSA in some geographic areas have also become resistant to fluoroquinolones. Some CA-MRSA strains resulted from acquisition of type IV SCCmec by highly virulent strains of MSSA [17].
Nosocomial MRSA infection — The prevalence of HA MRSA among hospitalized patients varies dramatically in different geographic regions [18,19]. It is generally high in the United States, Japan, and southern Europe (eg, >30 percent in Spain, France, and Italy) but is very low (<1 percent) in Scandinavia [19].
The following observations illustrate the overall high rate of MRSA in nosocomial S. aureus infections in the United States:
Data from the United States National Nosocomial Infection Surveillance (NNIS) system through June 2004 revealed that MRSA accounted for a mean of 53 percent of S. aureus isolates recovered from intensive care unit (ICU) patients, 46 percent from non-ICU patients, and 31 percent of S. aureus recovered from outpatients [20].
Data from The Surveillance Network-USA (TSN), an electronic surveillance network of 300 clinical microbiology laboratories across the United States, found that in March 2005, MRSA accounted for 59.2 percent of S. aureus from non-ICU patients, 55 percent from ICU patients, and 47.9 percent from outpatients [21].
In a prospective survey of nosocomial bloodstream infections from the SCOPE (Surveillance and Control of Pathogens of Epidemiologic Importance) database in 49 hospitals in the United States, the proportion of S. aureus isolates resistant to methicillin increased from 22 percent in 1995 to 57 percent in 2001 [22].
Prospective data on the incidence of MRSA infection in patients undergoing vascular surgery was collected from January 2000 to December 2001 [23]. MRSA was the single most prevalent organism identified in clinical cultures accounting for 42 of the 73 (57.5 percent) S. aureus strains isolated.
HA MRSA is also prevalent among residents of long-term care facilities [24,25]. Colonized residents are frequently transferred between hospitals and long-term care facilities, creating an ongoing circuit of MRSA transmission in areas where these organisms are prevalent.
Community-associated MRSA infection — In the early 1990s, CA-MRSA infections were reported among Australian aboriginals and northern aboriginals in Canada who had little or no previous contact with healthcare facilities [26]. Subsequently, clusters or outbreaks of CA-MRSA infections affected children in day care centers, native American communities in several states, athletic teams, military personnel, men who have sex with men in Los Angeles and San Francisco, and prison inmates and guards in several states [8,16,27-34].
Incidence — In studies performed between 1997 and 2002, MRSA accounted for a significant proportion of community-associated S. aureus isolates (12 to 29 percent in adults and as many as 35 to 50 percent in some series in children) [12-15,35].
The rate may be increasing as illustrated in a prospective prevalence study that examined adult patients with skin and soft tissue infection that presented to one of 11 university-affiliated emergency departments in the United States (the EMERGEncy ID Net) during August 2004 [36]. S. aureus was isolated from 320 of 422 (76 percent) patients. The prevalence of MRSA was 59 percent overall; 97 percent were the USA300 strain and 98 percent contained the Panton-Valentine leukocidin toxin gene (pvl).
The annual CA-MRSA infection incidence was evaluated in a population-based surveillance study in Baltimore and Atlanta and from a hospital-laboratory-based surveillance of 12 hospitals in Minnesota [37]. The following results were noted:
The incidence was significantly higher among persons <2 years of age compared to those 2 years (relative risk 1.5).
The incidence was significantly higher among blacks compared to whites in Atlanta (age-adjusted relative risk, 2.7); in contrast, there was no effect of race in Baltimore.
The infecting MRSA strain was often resistant (73 percent) to commonly prescribed antimicrobial agents. However, treatment of patients with skin and soft-tissue infections with therapy to which the strain was resistant was not associated with adverse patient outcomes.
Overall, 23 percent of patients required hospitalization for their MRSA infection.
Outbreaks and clusters of infection — MRSA infection may occur in the setting of a community outbreak or as a small cluster of infections in closely associated persons. The following cases illustrate these settings:
The first Canadian outbreak caused by CA-MRSA USA300 strain involved 40 cases of infection in the Calgary Health Region in 2004 [38]. Almost all infections (98 percent) involved skin and soft tissues, except for a single case of fatal necrotizing hemorrhagic pneumonia.
Two family clusters of invasive CA-MRSA infection involving a mother, father and daughter; and a husband and wife, led to the death of a young mother of fulminant pneumonia and hospitalization of the father and daughter [39]. Although reports of family clusters of invasive MRSA infection are rare, the prevalence is probably much higher then reported because family members of a patient are not routinely tested for MRSA colonization [40]. Reports of family clusters have important implications for the treatment of those in close contact with index patients. (See "Prevention and control of methicillin-resistant Staphylococcus aureus" section on Community-associated).
Risk factors — Among individuals who play competitive sports, a number of risk factors for CA-MRSA infection have been identified [27,28,41]. These include:
Skin trauma (eg, "turf burns", lacerations or abrasions)
Lineman or linebacker position in football
A higher body mass index
Cosmetic body shaving
Physical contact with a person who has a draining lesion or is a carrier of MRSA
Sharing equipment that is not cleaned or laundered between users
Other studies have identified the following significant risk factors [34,38,42]:
Prison occupation
Military personnel
Gender
Comorbidities
Prior skin infection
Previous antibiotic use
Illicit drug use
Tattoo recipients
An outbreak among tattoo recipients illustrates the importance of infection-control measures [42]. Although gloves were worn by all tattooists, other infection control measures (eg, changing gloves between clients, hand hygiene, skin antisepsis, disinfection of equipment and surfaces) were not practiced.
Nonoutbreak infection — A prospective laboratory surveillance study conducted between August and November 2003, examined S. aureus isolates from 384 patients with nonoutbreak community-acquired S. aureus skin and soft tissue infections [43]. Among S. aureus isolates, 244 of 389 (63 percent) isolates were MRSA; 90 percent (159 of 175) of all MRSA isolates available for molecular typing belonged to MRSA USA300 clone. The MRSA USA300 clone (an emerging CA-MRSA clone) usually encodes the Panton-Valentine leukocidin (PVL) virulence factor and has predictable resistance to beta-lactam antibiotics and erythromycin but susceptibility to clindamycin, trimethoprim-sulfamethoxazole, and fluoroquinolones. The high prevalence of this clone in CA-MRSA nonoutbreak skin and soft-tissue infections should guide antibiotic choices for empiric treatment of this serious infection.
Comparison to HA-MRSA — CA-MRSA can usually be distinguished from HA-MRSA isolates by molecular techniques, suggesting that community-associated disease is not typically due to spread of nosocomial strains into the community [12,13,15]. Methicillin resistance is mediated by the mecA gene, which is located on a mobile genetic element called staphylococcal cassette chromosome (SCCmec). SCCmec and drug resistance profiles are typically different in CA-MRSA and HA-MRSA [8-15]. (See "Genetics of methicillin resistance" above).
In addition to differences in their methicillin-resistance cassettes, community isolates are more likely to encode the putative virulence factor, Panton-Valentine leukocidin (PVL) [44].
COLONIZATION — Risk factors associated with HA-MRSA include prolonged hospitalization (often greater than 14 days), preceding antimicrobial therapy (especially with cephalosporins or fluoroquinolones), presence in an ICU or burn unit, hemodialysis, having a surgical site infection, and proximity to a patient colonized or infected with MRSA [45-47].
Patients and HCWs may become colonized with MRSA and therefore serve as sources of transmission. The reported rate of MRSA colonization has varied from 0.2 to 7.2 percent in hospitalized patients [48-50] and from one to two percent overall in the community but only 0.2 percent when patients with healthcare contacts are excluded [51,52].
The rate of MRSA carriage in the community is increasing. In a case-control study of nasal surveillance cultures performed at hospital admission on 726 patients, 53 (7.3 percent) were positive for MRSA [53]. In multivariate analysis, risk factor for colonization included:
Antibiotic use within three months before admission (odds ratio [OR], 2.5; 95% CI 1.2-5.0)
Hospitalization during the past 12 months (OR, 4.0; 95% CI 2.0-8.2)
Diagnosis of skin or soft-tissue infection at admission (OR, 3.4; 95% CI 1.5-7.9)
HIV infection (OR, 3.3; 95% CI 1.7-6.3)
A retrospective study of 3455 HIV-infected patients receiving care at one large clinic in the United States found that the incidence of MRSA infection increased six-fold in 2003 compared to 2000 [54]. The primary sources of these infections were skin and soft tissue in 83 percent of patients. Risk factors included advanced immunosuppression (CD4 count <50 cells), high plasma HIV RNA (>100,000 copies/µL) and lack of antiretroviral therapy.
Reservoirs — There are three major reservoirs of HA MRSA: patients, healthcare workers (HCWs), and the inanimate environment. Patients clearly represent the greatest source from which transmission occurs. Colonized or infected HCWs may also serve as reservoirs of MRSA but are probably only significant when they remain colonized for prolonged periods. (See "Colonization" above).
Patients — The anterior nares is the body site most frequently colonized with MRSA [55,56]. Approximately 20 to 30 percent of individuals with nasal carriage of S. aureus will also be colonized on areas of intact skin, including the perineum, hands, umbilicus (in infants), and, less frequently, the axilla [55,57]. Other body sites that may be colonized with MRSA include surgical wounds, decubitus ulcers, sputum, throat, urine and intravascular catheter sites (show figure 1).
Gastrointestinal colonization by MRSA may be more common than generally recognized as illustrated in a prospective surveillance study of 1543 patients who had stool specimens submitted for Clostridium difficile toxin assay [58]. When the specimens were additionally screened for MRSA, colonization was identified in 151 (9.8 percent). Ninety-three (62 percent) of the 151 patients had no previous history of MRSA and 60 of the 93 patients were not on contact isolation precautions.
The duration of HA MRSA colonization varies from a few days or weeks to several years [56,59]. In one report of patients requiring readmission to a hospital, the median length of colonization was 40 months [56].
Patients with nares colonization at hospital admission appear to be at increased risk of subsequent MRSA infection with the same isolate. This issue was addressed in a review of 758 patients hospitalized in five inpatient hospital units [48]. The following findings were noted:
MRSA colonization was present in 3.4 percent at admission; colonization was acquired during hospitalization in 3.0 percent.
The rate of MRSA infection during hospitalization or up to one year thereafter was 19 and 25 percent in patients with baseline and acquired MRSA colonization, respectively. These values were significantly higher (relative risk 13 and 9.5) than in those colonized with methicillin-sensitive S. aureus or those not colonized with S. aureus (1.5 and 2.0 percent, respectively).
Antibiogram data suggested that the colonizing and infecting isolates were the same.
A similar rate of MRSA infection (29 percent) was noted in a report of 209 adults who were colonized with MRSA [60]. One-half of these infections occurred after discharge from the hospital.
HCWs — An average of 3 to 5 percent of HCWs are nasal carriers of MRSA in facilities where MRSA is a problem [61]. Most carriers remain colonized for only short periods [59] and thus, probably do not represent a significant reservoir. However, the few who remain colonized for a prolonged time can serve as an important reservoir of MRSA [62-64]. In the Netherlands, long-term studies found that HCWs are involved in a number of outbreaks, sometimes serving as the index case [65].
Personnel with dermatitis that is infected or colonized with MRSA are particularly prone to transmit the organism. HCWs who have nasal carriage with MRSA can transmit the organism more readily when they have a concomitant sinus infection or a superimposed acute viral respiratory infection (so-called "cloud adults") [63,64]. One hospital epidemic, for example, appeared to result from exposure to a respiratory therapist with chronic sinusitis due to the epidemic strain [63].
Inanimate environment — The importance of the inanimate environment as a reservoir has not been studied adequately. In one report, 35 percent of inanimate surfaces cultured from the rooms of patients with MRSA in a wound or in urine were contaminated with the organism; in comparison, only 6 percent of surfaces were contaminated when patients had MRSA in sites other than a wound or urine [66]. Personnel who touch contaminated surfaces may contaminate their gloves (or presumably their hands) with MRSA [66].
Medical equipment items, such as stethoscopes and blood pressure cuffs, also can be contaminated with MRSA [66-68]. However, the extent to which contaminated environmental surfaces contribute to transmission of MRSA to patients has not been established.
Transmission — MRSA is most commonly spread from one patient to another by HCWs whose hands or gloves have become transiently contaminated with MRSA [45]. Although HCWs can contaminate their hands by touching colonized or infected wounds or dressings [45], it is also possible that they can acquire the organism by touching dry intact areas of the patient's skin, which is commonly colonized [55,56].
Personnel with persistent MRSA nasal colonization presumably spread the organism to patients either via droplet transmission or by direct contact if they also have hand colonization [57,62,63]. Chronic nasal carriers will transmit MRSA via droplet more readily if they develop a respiratory infection [63,64]. Airborne transmission of MRSA can occur when nurses are changing dressings or bedding of patients who have large burns, wounds, or areas of dermatitis that are infected or heavily colonized with MRSA [59].
Tourniquets, stethoscopes, and blood pressure cuffs, have been suggested as possible vehicles for transmission of MRSA but there are insufficient data to determine if transmission occurs [66]. Gowns and clothing worn by HCWs may become contaminated with MRSA; whether this leads to transmission of the organism is not known.
CLINICAL MANIFESTATIONS — The clinical characteristics of CA-MRSA and HA-MRSA were compared in a prospective study of 1100 MRSA infections in Minnesota in the year 2000 [12]. CA-MRSA accounted for 12 percent of infections, occurred in younger patients (median 23 versus 68 years with HA MRSA overall and 30 versus 70 years if pediatric hospitals were excluded), and was much more likely to produce skin and soft tissue infections (75 versus 37 percent; odds ratio 4.25). Most important, 61 percent of CA-MRSA infections were initially treated exclusively with beta-lactams to which the isolates were not susceptible.
Community-associated MRSA — Patients infected with CA-MRSA frequently present with skin or soft tissue infections [12,16,31-33,69,70]. However, some patients have developed necrotizing pneumonia, necrotizing fasciitis, or rapidly fatal septicemia, Waterhouse-Friderichsen syndrome [71-73], endocarditis, or osteomyelitis [12,69,70,74,75]. Production of Panton-Valentine leukocidin by the offending strain appears to be responsible, at least in part, for the development of skin and soft tissue infections and necrotizing pneumonia. Other types of infection in adults include otitis media/externa, toxic shock syndrome, and bacteremia [12,76].
During the 2003 to 2004 influenza season, 17 cases of S. aureus community-acquired pneumonia (CAP) were reported to the Centers for Disease Control and Prevention from nine states; 15 cases were associated with MRSA [77]. All isolates had community-associated genetic characteristics; 12 of 13 available S. aureus isolates had the Panton-Valentine leukocidin gene. Influenza virus infection was also documented in 12 (71 percent) of cases. The median age of patients was 21 years. All patients were hospitalized and death occurred in five (29 percent); four of the deaths were in patients with MRSA infection.
The natural history of CA-MRSA colonization was evaluated in a prospective observational study of 812 soldiers [78]. Nasal swab samples were taken at baseline and eight to ten weeks later. The following findings were noted:
At initial testing, CA-MRSA colonization was present in 24 (three percent), nine of whom (38 percent) developed a soft tissue infection during the observation period. In contrast, the rate of colonization with methicillin-susceptible S. aureus at baseline was much higher (28 percent), but the rate of clinical infection was much lower (three percent).
Without any eradication efforts, the rate of CA-MRSA colonization fell to 1.6 percent at follow-up culture. Previous antibiotic use was a risk factor for colonization at initial testing.
Healthcare-associated MRSA — Between 30 and 60 percent of hospitalized patients who acquire MRSA will develop an overt MRSA infection [46]. Host factors associated with progression from colonization to infection include recent prior hospitalization, preceding surgical or wound debridement, and the number of invasive procedures.
The body sites most frequently affected by HA MRSA infection are wounds, skin, and the bloodstream, followed by the lower respiratory and urinary tracts (show figure 2). Infections at many other body sites have also been reported.
A possible role of enterotoxin-producing MRSA in antibiotic-associated diarrhea has been suggested since the 1990s when investigators reported the isolation of MRSA in stools of hospitalized patients with diarrhea in the absence of other known etiologic agents [79-81]. To further investigate the role of MRSA in antibiotic-associated diarrhea, a prospective study of 3590 stool samples from hospitalized patients with diarrhea who had stool submitted for C. difficile toxin assay, were additionally screened for MRSA [81]. MRSA was isolated from 321 (8.9 percent). Eleven patients met the case definition of nosocomial diarrhea, had negative stool assays for C. difficile and enteric pathogens (bacterial, viral and parasitic), and had enterotoxin-producing MRSA in their stool and detectable enterotoxin in stool filtrates, suggesting that enterotoxin-producing MRSA may cause nosocomial antibiotic-associated diarrhea. Further studies of the role of MRSA in causing antibiotic-associated diarrhea are warranted.
MRSA infections can cause considerable morbidity, and often result in prolongation of the hospital stay, and higher hospital costs than those incurred by patients with MSSA infections [82-84]. In one study, hospital costs attributable to serious MRSA infections were on average $5100 greater than those associated with similar infections caused by MSSA [83]. Mortality rates associated with HA MRSA are similar to, and in a few studies were higher than, those associated with MSSA infections [85].
In a prospective study of 134 patients with S. aureus ventilator-associated pneumonia (VAP), MRSA compared to MSSA was not associated with a worse prognosis when adjustments were made in differences in patient characteristics, initial ICU treatment, and time in the ICU [86].
SUMMARY AND RECOMMENDATIONS
Methicillin resistant Staphylococcus aureus (MRSA), once primarily a healthcare-associated (nosocomial) infection, is now an increasingly prevalent community-associated pathogen. (See "Epidemiology" above).
Community-associated (CA) MRSA isolates usually have different molecular and antimicrobial susceptibility characteristics from healthcare-associated (HA) MRSA. (See "Epidemiology" above).
There is an overall high rate of MRSA in nosocomial S. aureus infections (>30 percent) in the United States, Japan, and southern Europe. (See "Nosocomial MRSA infection" above).
Clusters or outbreaks of CA-MRSA infections have affected children in day care centers, native American communities in several states, athletic teams, military personnel, men who have sex with men in Los Angeles and San Francisco, and prison inmates in several states. (See "Community-associated MRSA infection" above).
There are three major reservoirs of HA MRSA: patients, healthcare workers (HCWs), and the inanimate environment. (See "Reservoirs" above).
MRSA is most commonly spread from one patient to another by HCWs whose hands or gloves have become transiently contaminated with MRSA. (See "Transmission" above).
Patients infected with CA-MRSA frequently present with skin or soft tissue infections, but may develop necrotizing pneumonia, necrotizing fasciitis, or rapidly fatal septicemia, endocarditis, or osteomyelitis. (See "Community-associated MRSA" above).
Between 30 and 60 percent of hospitalized patients who acquire MRSA will develop an overt MRSA infection. (See "Healthcare-associated MRSA" above).
2006-11-07 01:15:35
·
answer #1
·
answered by M.C. Clarence, M.D. 2
·
2⤊
0⤋