|Home | About | Journals | Submit | Contact Us | Français|
Staphylococcus aureus is one of the most common pathogens in community- and hospital-associated infections and frequently causes severe and intractable infections in osteoarticular tissues. S. aureus isolates that are resistant to methicillin (meticillin) (MRSA isolates) and intermediately resistant to vancomycin (VISA isolates) have emerged. In this report, we described two patients, one female and one male, diagnosed with septic arthritis due to S. aureus infections. A total of 13 MRSA isolates were obtained from these two patients. All but one isolate belonged to the VISA group. All seven isolates from the female patient were determined to be community associated, multilocus sequence type (MLST) 59, and staphylococcal cassette chromosome mecA (SCCmec) type IV; had direct repeat units (DRUs) of nine repeats; were spa type t437; and were susceptible to sulfa and quinolone antibiotics. The other six isolates, from the male patient, were determined to be hospital associated, MLST 239, and SCCmec type III; had DRUs of 14 repeats; were spa type t037; and were resistant to sulfa and quinolone antibiotics. All 13 MRSA isolates were in agr group I, were pvl negative, and showed no evidence of any association between vancomycin resistance and autolysis.
Staphylococcus aureus, which causes severe infections in the skin and soft tissue and in the cardiovascular, osteoarticular, and respiratory systems, is one of the most common and important pathogens in both community-associated and hospital-associated infections (7, 24, 31). Methicillin (meticillin)-resistant S. aureus (MRSA) emerged in the 1960s and has maintained a prevalence of 60 to 80% since the 1980s (15, 31). Vancomycin-intermediate S. aureus (VISA) strains have also emerged in the past 10 years due to frequent use of glycopeptide antibiotics for treatment of MRSA infections (20).
Hematogenous and surgery-related septic arthritides are severe infections of the osteoarticular system, and S. aureus is the most common causative pathogen (1, 10). Although glycopeptides, including vancomycin and teicoplanin, are commonly used to treat MRSA septic arthritis (44), poor penetration of antibiotics into the periarticular space and biofilm formation on prosthesis make it very difficult to treat septic arthritis caused by S. aureus (17, 36).
We had reported a patient with septic arthritis caused by VISA isolates which were resistant to autolysis in the Triton X-100 lysis assay due to increased thickness of the cell wall (32). Although several molecular typing methods, such as pulsed-field gel electrophoresis (PFGE) (35, 47), staphylococcal cassette chromosome mecA (SCCmec) typing (23, 27), and multilocus sequence typing (MLST) (12), have been developed for typing, few molecular epidemiology studies of S. aureus strains from patients with osteoarticular infections have been reported. In this study, we applied these methods to delineate the molecular types and phenotypes of S. aureus isolates from two patients with septic arthritis of distinct sources.
The first patient (patient A) was a 67-year-old female admitted to the Tri-Service General Hospital (TSGH), Taipei, Taiwan, on 12 October 2001 due to painful swelling of the left knee joint. She had received for the first time total knee replacement of the left knee joint about 3 years before admission. There was no evidence of any previous MRSA infection. The patient also had no history of hospitalization, surgery, dialysis, or long-term stay in any health care facility within 1 year of admission. There was also no use of an indwelling catheter or percutaneous medical device (e.g., tracheostomy tube or Foley catheter) at the time of admission. The first MRSA isolate (isolate no. 109) was isolated from synovial fluid of the left knee joint on 13 October 2001; therefore, the patient was considered to have septic arthritis due to MRSA infection. The patient had undergone a total of six operations, including debridement of devitalized tissue through arthrotomy, revision of the left total knee prosthesis, and split-thickness skin graft, from 17 October 2001 to 17 January 2003. Six additional MRSA isolates (isolates 1962, 2357, 2358, 2459, 2482, and 2516) were obtained from synovial fluid and arthrotomy wounds of the left knee from 23 October 2002 to 8 January 2003. The patient was treated with vancomycin and teicoplanin for several courses (from 19 October to 17 November 2001 and from 8 January to 8 February 2003 for vancomycin therapy and from 18 to 30 November 2001 and from 24 October to 10 December 2002 for teicoplanin therapy) and had recovered without any sequela.
The second patient (patient B) was a 19-year-old male who had had four admissions and surgical operations due to osteomyelitis of the left tibia and septic arthritis of the left knee joint at one local hospital from August 2000 to September 2001. Vancomcyin therapy for several courses from September 2000 to October 2001 was instituted at that hospital without any improvement. The patient was transferred to the TSGH on 23 October 2001, and arthroscopic debridement and revision of the left knee prosthesis were performed on 30 October 2001. Six consecutive and unduplicated MRSA isolates (isolates 188, 218, 257, 204, 205, and 243) were recovered from six different synovial fluid samples and knee wound tissue cultures from 2 to 21 November 2001.
Identification of S. aureus isolates was achieved by standard techniques. All 13 isolates were determined to be gram-positive cocci with β-hemolysis on 5% sheep blood agar and gave positive results for catalase, coagulase, DNase, and mannitol fermentation tests. They were confirmed to be MRSA and VISA by the presence of the mecA gene (12, 23) and by antimicrobial susceptibility tests, including Etest and the disc diffusion test, according to the 2009 CLSI protocols (5, 6). These isolates were also examined for inducible macrolide-lincosamide-streptogramin resistance by a disk diffusion test (D test) (41) and VISA by inoculation of each isolate on brain heart infusion (BHI) agar containing 4 μg/ml vancomycin, as described previously (32).
Genomic DNA was extracted from each isolate and digested with the SmaI restriction enzyme (Promega Corp., Madison, WI). PFGE was performed using a contour-clamped homogeneous electric field apparatus (DR-III; Bio-Rad, Hercules, CA) as described by Tenover et al. (47). Isolates with identical PFGE profiles were designated the same type, and isolates that differed by one or two bands were assigned to a subtype (47).
Multiplex PCR (M-PCR) for detection and identification of SCCmec, which carries the mecA gene responsible for methicillin resistance, was performed using genomic DNA from each MRSA isolate as a template, as described previously (23, 27). The SCCmec elements contained both the genes encoding the cassette chromosome recombinase (ccr genes) and the mec gene complex (23). A total of 14 primers were used for the M-PCR: 2 for detection of the mecA gene, 4 for identification of the four classes of the mec gene complex (classes A to D), and 8 for identification of the five types of ccr genes (ccr types 1 to 5) (27). SCCmec types I to V of MRSA strains were identified by comparing the M-PCR banding patterns of the isolates to those of the following reference strains: ATCC 10442 (SCCmec type I), N315 (SCCmec type II), 85/2082 (SCCmec type III), MW2 (SCCmec type IVa), WIS (SCCmec type V), and TSGH-17 (SCCmec type VT) (23, 27, 34).
The following seven housekeeping genes were amplified from each S. aureus isolate, as described previously (12): the carbamate kinase (arcC), shikimate dehydrogenase (aroE), glycerol kinase (glp), guanylate kinase (gmk), phosphate acetyltransferase (pta), triosephosphate isomerase (tpi), and acetyl coenzyme A acetyltransferase (yqiL) genes. The amplified products were then sequenced, and the allelic number of each gene was determined by comparing its sequence to those of the known alleles in the S. aureus MLST database.
The X region of the spa gene contains variable numbers of 21- to 27-bp repeats, with the 24-bp repeat being the most common (14). The X region of each MRSA isolate was amplified by PCR with primers 1095F (5′-AGACGATCCTTCGGT GAGC-3′) and 1517R (5′-GCTTTTGCAATGTCATTTACTG-3′) (46). The amplified products were sequenced, and the sequences were analyzed by the Ridom StaphType software program (version 1.4; Ridom, GmbH, Wurzburg, Germany [http://spa.ridom.de/index.shtml]), which automatically determines the repeat profile and the spa type of each isolate (18).
Amplification of the hypervariable region, which contains variable numbers of 40-bp direct repeat units (DRUs), was performed by PCR with primers orf145 and IS431mec (45). The number of DRUs in each MRSA isolate was determined by the size of the amplified fragment according to the following equation: no. of DRUs = (size of PCR product − 171)/40 (45).
Amplification of the pvl gene was done with primers luk-PV-1 (5′-ATCATTAGGTAAAATGTCTGGACATGATCCA-3′) and luk-PV-2 (5′-GCATCAASTGTATTGGATAGCAAAAGC-3′) as described by Lina et al. (29). The reference strain TSGH-17 was used as the positive control, and strain 85/2082 was used as the negative control.
The agr sequences were amplified with the following primers: Pan (5′-ATGCACATGGTGCACATGC-3′), agr1 (5′-GTCACAAGTACTATAAGCTGCGAT-3′), agr2 (5′-TATTAC TAATTGAAAAGTGGCCATAGC-3′), agr3 (5′-GTAATGTAATAGCTT GTATAATAATACCCAG-3′), and agr4 (5′-CGATAATGCCGT AATACCCG-3′). These primers allow amplification of a 441-bp DNA fragment from agr group I, a 575-bp fragment from agr group II, a 323-bp fragment from agr group III, and a 659-bp DNA fragment from agr group IV strains, as described previously (16).
Isolates suspected to be VISA were grown on screening plates containing vancomycin and then subjected to population analysis to find subpopulations of cells that were resistant to vancomycin. Briefly, 50 μl of the starting cell suspension of each isolate and its serial dilutions was spread on BHI agar plates containing 1 to 10 μg/ml of vancomycin in 1-μg/ml increments. The starting cell suspension was prepared by diluting the fresh bacterial culture in BHI broth to an optical density at 580 nm (OD580) of 0.3. The plates were incubated at 37°C for 48 h. The number of vancomycin-resistant cells in each of the 50-μl aliquots was plotted semilogarithmically (32).
Cells were grown in BHI broth to mid-logarithmic phase, pelleted, washed twice in ice-cold water, and then resuspended in the Triton X-100 lysis buffer (0.05 M Tris-HCl and 0.05% Triton X-100, pH 7.2). The decrease in OD620 value due to cell lysis was determined at 30-min intervals for 4 h, and the percentage of surviving cells, expressed as remaining OD620, was plotted as described previously (32).
A total of 13 MRSA isolates were recovered from the two patients with septic arthritis. Their demographics and antimicrobial susceptibility results as determined by disk diffusion and Etest are shown in Table Table1.1. All isolates harbored the mecA gene and were resistant to penicillin, erythromycin, clindamycin, and gentamicin but were susceptible to vancomycin and teicoplanin. Since patient A had no previous hospitalizations, the MRSA infection was considered community associated (CA-MRSA). All seven MRSA isolates from this patient were also susceptible to trimethoprim-sulfamethoxazole (SXT) and ciprofloxacin. The MRSA infection of patient B was considered hospital associated (HA-MRSA) because the patient had had four hospitalizations due to osteomyelitis of the left tibia and septic arthritis of the left knee joint. All six MRSA isolates from this patient were resistant to SXT and ciprofloxacin. All but one isolate (no. 109) grew after 48 h of incubation on screening agar plates containing 4 μg/ml vancomycin. All isolates were determined by Etest to be vancomycin intermediate, with MICs ranging from 2 to 8 μg/ml, and were the noninduction phenotype as determined by D test.
All seven MRSA isolates from patient A had the same PFGE pattern. The six MRSA isolates (four of which are shown in Fig. Fig.1)1) from patient B also had the same PFGE pattern, but this pattern was distinct from that of patient A (Fig. (Fig.1).1). The results of other molecular typing methods, including SCCmec typing, MLST, spa typing, DRU determination, and agr typing, are shown in Table Table2.2. All seven MRSA isolates from patient A had nine DRUs; were determined to be SCCmec type IV, MLST 59, and spa type t437; and were in agr group I. The six MRSA isolates from patient B had 14 DRUs; were determined to be SCCmec type III, MLST 239, and spa type t037; and were in agr group I. All 13 MRSA isolates were negative for pvl.
All seven MRSA isolates (no. 2357, 2358, 109, 1962, 2459, 2482, and 2516) from patient A were analyzed (Fig. (Fig.2).2). Strain no. 109 was determined to be a vancomycin-susceptible S. aureus (VSSA) isolate, as its growth was completely inhibited by vancomycin at 4 μg/ml. Isolate no. 1962 was found to be an S. aureus isolate heterogeneously and intermediately resistant to vancomycin (h-VISA) since its growth was completely inhibited by vancomycin at 5 μg/ml. The growths of isolates 2358 and 2357 were completely inhibited by vancomycin at 6 μg/ml. These two isolates, as well as isolates 2459, 1962, and 2516, were determined to be VISA. The growths of isolates 2459, 1962, and 2516 were not completely inhibited until the vancomycin concentrations reached 8 μg/ml. It is notable that isolates 2357, 2358, 2459, 2482, and 2516 were obtained from synovial fluids after patient A had completed several courses of glycopeptide therapy. Population analysis of MRSA isolates (no. 188, 204, 218, and 257) from patient B has been described previously (32). Isolates 188, 204, and 218 were classified as VISA, with vancomycin MICs of 8 μg/ml, and isolate no. 257 was determined to be h-VISA, with a vancomycin MIC of 4 μg/ml (32). Surprisingly, isolate no. 257 was isolated several weeks after discontinuation of glycopeptide therapy.
The seven MRSA isolates (no. 109, 1962, 2357, 2358, 2459, 2482, and 2516) from patient A exhibited different autolysis patterns in Triton X-100 lysis solution, as shown in Fig. Fig.3.3. Isolate no. 2516 was the isolate most resistant to Triton X-100-induced autolysis, with 47% surviving cells, and isolate no. 2358 was the most sensitive isolate, with only 22% surviving cells after 4 h of treatment. Isolates 109, 2482, 1962, 2459, and 2359 had 43%, 36%, 28%, 26%, and 26% surviving cells, respectively, at the end of the 4-hour Triton X-100 treatment. Triton X-100-induced autolysis of the six isolates from patient B has been reported (32).
S. aureus, especially MRSA, is a common pathogen in humans. According to a surveillance study of antibiotic resistance from 1987 to 1997 by the Centers for Disease Control and Prevention, the prevalence of MRSA in the United States was increased from 22.8% to 56.2% (31). The Taiwan Surveillance of Antimicrobial Resistance I and II (TSAR I and II) studies showed that approximately 60% of the 400 S. aureus isolates from 65 hospitals were MRSA (34). MRSA strains were originally found in hospital settings, but CA-MRSA has emerged from patients without prior history of exposure to a hospital environment (15, 19).
In this study, we demonstrated two distinct molecular types and antibiotic susceptibility patterns of successive MRSA isolates from two patients with septic arthritis. A total of seven MRSA isolates from a 67-year-old female patient with hematogenous septic arthritis of the left knee joint were isolated. These isolates were considered to be CA-MRSA because the patient had no previous history of exposure to any hospital environment. A CA-MRSA clone of USA300 (MLST 8 with an arginine catabolic mobile element and the gene encoding Panton-Valentine leukocidin) has been reported to disseminate from the community setting to the hospital setting in nine San Francisco area medical centers from 2004 to 2005, which blurs its distinction as a community-associated pathogen (30). On the basis of the molecular and epidemiologic results reported by Liu et al. (30), classification of MRSA isolates into CA-MRSA and HA-MRSA groups by molecular characters seems more reliable than the use of demographic data.
Colonization of MRSA on skin or in nostrils was suspected to be the origin of this infection in the female patient. In a study of isolates from Taiwan, the nasal and skin carriage rates of MRSA strains in community- and health care facility-related populations were found to be 3.6% and 11%, respectively (33), and Huang and Platt reported that colonization of S. aureus appeared to be the event preceding subsequent infections, as 60 (29%) of 209 patients who harbored MRSA developed a total of 90 episodes of MRSA infections (22). According to the study by Fridkin et al., in three large U.S. communities, CA-MRSA accounted for 8 to 20% of all MRSA infections (15). Most (1,266/1,647 [77%]) CA-MRSA infections involved skin and soft tissue, and only 1% (15/1,647) of them caused septic arthritis.
Prosthetic infections are difficult problems for surgeons and patients receiving surgical intervention and prosthesis implantation. A large 15-year prospective study involving 10,735 patients after hip or knee replacements found that the incidences of deep prosthetic infections were 0.57% among patients with hip replacement and 0.86% among those with knee replacement (42). The most common causative organism is coagulase-negative staphylococcus (36%), followed by S. aureus (25%), among which MRSA accounted for 4% of the infections. Most (64%) of the prosthesis infections occurred within the first year after joint replacement (42). The young male patient in this study developed osteomyelitis in the left tibia and subsequently septic arthritis in the left knee joint several months after surgical intervention. After removal of infected prosthesis and devitalized tissue and appropriate antimicrobial regimens, the prosthetic infection was eradicated.
There were relatively few molecular epidemiology studies of S. aureus isolates from patients with osteoarticular infections. In the study reported by Chen et al. (4), 11 (84.3%) MRSA isolates from 13 children with community-associated osteoarticular infections were found to be MLST 59. In our previous study of 11 CA-MRSA isolates from the TSGH, Taipei, Taiwan, in 2002, three (20%) were found to be SCCmec type IV, and the other eight isolates (53%) were SCCmec type VT (48). In the present study, the seven CA-MRSA isolates from patient A were determined to be pulsotype A, SCCmec type IV, MLST 59, and spa type t437; had DRUs of nine repeats; were in agr group I; and were susceptible to SXT and ciprofloxacin. The six HA-MRSA isolates from patient B were pulsotype B, SCCmec type III, MLST 239, and spa type t037; had DRUs of 14 repeats; were in agr group I; and were resistant to non-ß-lactam antibiotics. Whether these isolates are among those that have been reported remains to be determined, as previous isolates were typed by only one or two methods (8, 13).
Staphylococcal protein A is a cell wall-anchored protein of S. aureus. It can interact with several host components (39) and is a virulence factor in septic arthritis (38). The polymorphic X region in the spa gene has proved to be sufficiently stable for typing, with discriminatory power comparable to that of phage typing, PFGE, and restriction fragment length polymorphism typing (14). We demonstrated that the seven CA-MRSA isolates from patient A were spa type t437 and that the six HA-MRSA isolates from patient B were spa type t037. Neither of these two types has been found in other osteoarticular infections, as determined by a literature search of the PubMed database.
Panton-Valentine leukocidin was initially discovered by Van deVelde in 1894 due to its ability to lyse leukocytes and was named after Sir Philip Noel Panton and Francis Valentine because of their discovery in soft tissue infections in 1932 (3, 40). Panton-Valentine leukocidin has been found to be associated with necrotizing pneumonia (11, 28), scalded skin syndrome (21), wounds (21), bacteremia (21), and osteoarticular infections (9). The above-mentioned MLST 59 CA-MRSA strain isolated from children with osteoarticular infections in Taiwan harbored the pvl gene (4). In two clinical studies, pvl+ MRSA strains were shown to be more significantly associated with an enhanced inflammatory response characterized by higher erythrocyte sedimentation rates and C-reactive protein levels, positive blood cultures, the presence of adjacent myositis, severe sepsis, and septic shock (2, 9). Holmes et al. (21) reported that the frequency of pvl in S. aureus strains associated with clinical diseases in England and Wales was 1.6% (8/515) and that most of the pvl+ MRSA isolates were sequence type 80 and SCCmec type IVc. In another study, involving 37 S. aureus isolates from patients with osteomyelitis and septic arthritis in Lyon, France, no isolate was found to harbor pvl (9). In the present study, none of 13 MRSA isolates harbored the pvl gene.
The agr gene of S. aureus is involved in quorum sensing. It also upregulates the production of several secreted virulence factors and downregulates the production of some cell-associated virulence factors (25). Most clinical S. aureus isolates have been determined to be agr group I isolates, but some CA-MRSA isolates were found to belong to agr group III (37). Little is known about the relationship between the expression profiles of virulence factors and agr groups. All 13 MRSA isolates in this study were classified as agr group I isolates.
Clinical failure with vancomycin therapy had been reported for a Japanese patient with staphylococcal pneumonia caused by h-VISA (20). Suppression of cell wall turnover rate and a decrease in autolysis rate had been observed in VISA isolates (26, 43). We had also demonstrated that four VISA isolates from patient B in this study had increased cell wall thickness and reduction in the rate of autolysis by Triton X-100 lysis buffer (32). Among the seven MRSA isolates from patient A, the first isolate was found to be VSSA (isolate 109; vancomycin MIC, 2 μg/ml), one isolate was determined to be h-VISA (isolate 1962; vancomycin MIC, 4 μg/ml), and the other five isolates were found to be VISA (isolates 2357, 2358, 2459, 2482, and 2516; vancomycin MICs, 8 μg/ml). Since these five isolates were obtained after the patient had undergone several courses of glycopeptides therapy, it is likely that the vancomycin resistance phenotype was induced. The VSSA isolate was not found to be more susceptible to Triton X-100-induced autolysis than these five VISA isolates. Whether resistance to glycopeptide antibiotics and autolysis in S. aureus strains is associated with factors such as wall thickness and specific agr gene groups remains to be investigated.
This work was supported by grants NSC 94-2320-B-016-010 and NSC 95-2320-B-016-023-MY3 from the National Science Council, Taiwan.
We thank Chao-Hung Lee for assistance with the manuscript.
Published ahead of print on 16 September 2009.