Introduction

Worldwide, methicillin-resistant Staphylococcus aureus (MRSA) strains are among the most commonly isolated bacteria in patients requiring hospitalization or with significant healthcare exposure (HA-MRSA) [1–3]. The presence of the mecA gene, which induces resistance to almost all β-lactams is probably one of their most important characteristics [4]. At the turn of the 20th century, the first reports of community-associated MRSA (CA-MRSA) infections among healthy individuals (with no identifiable risk factors for HA-MRSA infections) and among injection drug users, incarcerated people, and athletes were published [5, 6].

Strains of MRSA are more frequently associated with skin and soft tissue infections, but more invasive infections, including bone and joint infections, also occur [4, 7]. In fact, MRSA has been identified as one of the most common causes of bone and joint infections [8, 9]. During the last decade CA-MRSA strains have been reported to be responsible for osteomyelitis or septic arthritis [10]. We sought to review systematically the available evidence in order to identify the incidence, characteristics, and outcomes of patients with CA-MRSA in bone and joint infections.

Materials and methods

Data sources

The studies included in this review were retrieved from searches performed in PubMed and Scopus (up to May 2012), using the search terms “bone and joint infections,” “osteomyelitis,” “septic arthritis,” “spondylodiscitis,” “spondylitis,” “bursitis,” “discitis” in combination with the terms “community-associated methicillin-resistant Staphylococcus aureus” or “community-associated MRSA.” References of the retrieved articles and relevant reviews were also hand-searched.

Study selection criteria

Studies reporting data on the incidence, characteristics, and outcomes of patients with CA-MRSA bone and joint infections were included in this review. Therefore, studies of any design that enrolled patients of any age could be included. Prosthetic joint infections were also eligible. Abstracts of scientific conferences and studies published in languages other than English, Italian, German, Greek, French, and Spanish were excluded. Studies not fulfilling the definition of CA-MRSA infection (as described below) were excluded, even if the titles indicated that they were reporting data on patients with CA-MRSA bone and joint infection.

Definitions

A case of CA-MRSA bone or joint infection was defined as disease compatible with osteomyelitis or septic arthritis, in which MRSA was cultured from blood, synovial fluid, or bone biopsy. A culture from wound or abscess was eligible in cases of clinically and/or radiographically diagnosed bone or joint infections complicated with abscess formation or fistula. The culture should have been taken in an outpatient setting or within 48 h after hospital admission, and with none of the following healthcare risk factors: use of broad spectrum antibiotics during the previous 6 months, recent hospitalization, residence in a long-term care facility, dialysis, surgery 1 year before the onset of illness or permanent indwelling catheter or percutaneous medical device [11]. Moreover, the definition was broadened to include cases in which the molecular typing methods (pulsed-field gel electrophoresis [PFGE], multi locus sequence typing [MLST] or other techniques) provided evidence of a community-associated strain. This was allowed in order to study the potential penetration of MRSA strains with “community-associated” characteristics into the hospital environment.

Results

Epidemiology incidence of CA-MRSA bone/joint infections

A population-based surveillance program in Atlanta (GA, USA) and Baltimore (MD, USA) and a hospital laboratory sentinel surveillance of 12 hospitals in Minnesota (USA) performed in 2001 and 2002 showed that the annual incidence of invasive CA-MRSA infections was 25.7 cases per 100,000 in Atlanta and 18.0 per 100,000 in Baltimore. Bone and joint infections were responsible for 2.8 % of these cases in Atlanta, 5 % in Baltimore and 6 % in Minnesota. In both Atlanta and Baltimore CA-MRSA were more common in patients aged <2 years; in Atlanta CA-MRSA infections were more common among blacks than whites in all age groups [12]. In the Active Bacterial Core surveillance system during 2004–2005 the annual incidence of invasive CA-MRSA infections ranged between 1.6 and 29.7 per 100,000 among different regions. In this surveillance, the higher incidence was seen in patients aged >65 years (8.9 per 100,000) and the lower in patients aged 2–17 years (0.6-0.8 per 100,000) [13]. Osteomyelitis accounted for 8.1 % of all cases. Finally, data from a prospective surveillance in Sweden showed that during the period 2003–2005 the annual incidence of CA-MRSA-invasive infections was 16.6 per 100,000; bone and joint infections accounted for 43 % of the cases [14].

Selected studies

The searches performed in PubMed and Scopus generated a total of 130 and 233 search results respectively. The process of study selection is shown in details in Fig. 1. A total of 67 studies were included in the review [5, 15–80].

Fig. 1
figure 1

Flow diagram of the detailed process of selection of articles for inclusion in the review

Full size image

Case reports of CA-MRSA bone/joint infections

Table 1 summarizes the data available from patients described in case reports of CA-MRSA bone and joint infections; data for 45 patients from 35 reports were available [15, 17, 19, 24, 25, 27, 31, 34, 36, 38, 39, 41, 42, 44–48, 51, 52, 54–57, 61, 62, 65–68, 71, 73, 80]. The majority of the patients were young (25 out of 45 [56 %] were children ≤14 years, 30 out of 45 [67 %] were younger than 30 years old and almost all [44 out of 45, 98 %] were younger than 65 years). Most of them were male (31 out of 45, 69 %) with no known risk factors for CA-MRSA infections except for a history of skin and soft tissue infections (12 out of 42, 29 %). The median duration of symptoms prior to the diagnosis or hospitalization was 6 days. Fever, local tenderness, and articular disability were the main symptoms. Osteomyelitis was the main diagnosis (either alone [32 out of 45, 71 %] or in combination with septic arthritis [4 out of 45, 9 %]). Long bones were mainly affected followed by vertebrae. Diagnosis was confirmed by X-rays, computed tomography (CT), magnetic resonance imaging (MRI), or scintigraphy. Further microbiological studies for the identification of toxins (Panton–Valentine leukocidin [PVL] positive in 14 out of 17) and typing (according to SCCmec type and MLST) was performed in 38 % of the cases. SCCmec IV was the predominant or sole type in all studies.

Table 1 Main characteristics, interventions and outcomes of patients with community-acquired methicillin-resistant Staphylococcus aureus bone and joint infections
Full size table

The majority of patients had bacteremia (27 out of 45, 60 %) with or without local complications, including abscesses, pyomyositis, and deep venous thrombosis. Sixteen patients had systemic complications, either on admission or during treatment; 13 patients had pulmonary complications (mainly septic emboli and pleural effusion), 4 patients had central nervous system and 3 cardiac involvement.

The empirical treatment (before the culture results were available) was provided for 29 cases; in 16 of them (55 %) the empirical regimen did not include an antibiotic effective against MRSA. Following treatment that employed surgical interventions in 24 patients (59 %) and antibiotics in all patients (mainly vancomycin [36 out of 42, 88 %] and in some cases clindamycin, fusidic acid, linezolid, fosfomycin, trimethoprim/sulfamethoxazole or teicoplanin) for a median duration of 8 weeks (range 1–30), the majority of patients were cured (39 out of 44, 87.5 %), while 3 patients died (8 %). Two patients died during the first week of the hospital stay; 1 had chronic kidney disease and diabetes and 1 developed respiratory insufficiency due to septic emboli. The last patient died on week 5: a 7-year-old girl who developed respiratory insufficiency. Four recurrences were reported, one of which was finally cured, and the other resulted in chronic osteomyelitis. Local complications at the end of treatment included pathological fractures in 7 patients and chronic osteomyelitis in 3.

Case series of CA-MRSA bone/joint infections

We identified 33 case series that included 510 patients with CA-MRSA bone and joint infections [16, 18, 20–23, 26, 28–30, 32–35, 37, 43, 49, 50, 53, 58–60, 63, 64, 69, 70, 72, 74–79]. The majority of the case series included described patients with bone and joint infections due to several pathogens and also included patients with MRSA osteoarticular infections. Therefore, specific data regarding patients with MRSA infections were limited (Table 2). Most of the studies included were performed on children (488 out of 510 patients, 96 %) [16, 18, 20–23, 26, 29, 30, 32–35, 37, 49, 50, 58–60, 63, 64, 69, 70, 72, 76–79], and only 5 included adults [28, 43, 53, 74, 75]. In studies that provided data, bacteremia was present in most of the patients (45–100 %) [18, 21, 22, 34, 35, 49, 50, 59, 60, 63, 72], while DVT was the major complication in 6 of them (4 % in the largest study that provided data [16], up to 100 % in small case series [22, 26, 34, 35, 49, 59, 63, 72, 76]. PVL was present in 166 out of 172 (97 %) of isolates in which it was tested. SCCmec IV USA300 was the predominant clone in the few studies that provided relevant data. One study from France reported that the European clone ST 80 was isolated. Surgical interventions were required in 202 out of 236 (86 %) of patients for whom data were available. Vancomycin and clindamycin were the most commonly prescribed antibiotics. Four patients died (4 out of 370, 1 %) and 16 out of 126 (13 %) developed chronic osteomyelitis.

Table 2 Characteristics and outcomes of patients with CA-MRSA bone and joint infections in published case series
Full size table

Discussion

There are limited data regarding the annual incidence of CA-MRSA bone and joint infections. The available evidence shows that it varies according to the location and age of the studied population. Thus, the incidence of invasive CA-MRSA ranged from 1.6 to 29.7 cases per 100,000, while bone and joint infections accounted for 2.8 to 43 % of invasive CA-MRSA infections. In addition, surveillance studies showed that these infections affect mainly patients <2 years of age; the distribution of infections in other age populations depends on the study site and race. However, it seems that black race is associated with a higher risk of infections than white.

The published evidence from case reports and case series suggested that the majority of patients of CA-MRSA bone and joint infections were cured. A total of seven deaths were reported; 3 in case reports and 4 in case series. On the other hand, complications were relatively common; abscesses, pyomyositis, DVT, and chronic osteomyelitis were the most common local complications. Although systemic complications were also frequent (39 % in case reports), this did not seem to affect mortality. As expected, prolonged antibiotic treatment and hospitalization were employed in most of the cases.

Vancomycin was the most frequently used antibiotic in both case reports and case series. However, clindamycin monotherapy was also used effectively in case series as well as in combination with or following initial vancomycin treatment [16, 49, 50, 69, 72]. In these case series, inducible resistance to clindamycin was not detected. Vancomycin is the treatment of choice for MRSA bone and joint infections [81, 82]. However, specific data regarding the most effective treatment option for CA-MRSA infections are not available. Linezolid, teicoplanin, and daptomycin are alternative options that have been used effectively for the treatment of MRSA bone and joint infections [83–87].

S. aureus is the most commonly isolated pathogen from bone and joint infections [4, 7, 81, 82]. During the last decade, several studies showed that S. aureus infections increased, and the relative frequency of MRSA increase was more prominent than that of methicillin-susceptible S. aureus (MSSA), both in the community and in hospitals [88–90]. Therefore, it seems prudent to include in the empirical antibiotic regimen a drug that is effective against MRSA. In the limited available data in this review, approximately 50 % of patients did not receive appropriate empirical treatment. Fortunately, this did not seem to have an impact on mortality. However, complications were frequent. Owing to the limited data available, we were unable to identify whether or not the use of appropriate treatment could be associated with fewer complications. A recent review reported that MRSA osteomyelitis in children was associated with more DVTs than MSSA osteomyelitis [91].

Most of the patients included in the present review did not have risk factors for CA-MRSA infection, in accordance with previously published studies [92, 93]. Moreover, a site of entry was not identified, suggesting a hematogenous shedding of MRSA. Hematogenous osteomyelitis of long bones is most commonly seen in children, as was the case in this review. In addition, the propensity of CA-MRSA toward younger individuals has been confirmed in this review. The great majority in both case reports and case series were children. On the contrary, we have previously reported that the majority of published cases of CA-MRSA pneumonia were young adults [92, 93]. Other differences between the two reports included the higher severity of complications and mortality reported in patients with pneumonia.

One study provided data for 5 patients with CA-MRSA prosthetic joint infections [43]. All infections developed early after joint replacement. All patients required replacement of the affected joint and prolonged antibiotic therapy. The isolated strains had identical antibiograms. Molecular characterization of two of them revealed a PFGE type that was identical to the USA300 clone, SCCmec type IV. This report provides further evidence that “community-associated” strains have been identified as a cause of healthcare-associated or hospital-associated infections [94–96].

In conclusion, the currently limited available evidence from case reports and studies with different study designs and patients’ characteristics suggests that the incidence of bone and joint infections caused by CA-MRSA varies and depends on the geographic region (even within the same country), age, and race of the population. A propensity toward younger individuals is evident. The currently available treatment options seem to be adequately effective. However, the increase in the number of CA-MRSA infections both in the community and in the hospital among patients with no known risk factors requires increased awareness for early recognition and treatment.