Is methicillin-resistant Staphylococcus aureus involved in community acquired skin and soft tissue infections?: Experience from a tertiary care centre in MumbaiRS Phakade1, G Nataraj1, SK Kuyare1, US Khopkar2, PR Mehta1
1 Department of Microbiology, Seth G. S. Medical College and KEM Hospital, Mumbai, India
2 Department of Dermatology, Seth G. S. Medical College and KEM Hospital, Mumbai, India
Background: To improve the empiric antimicrobial therapy of community-acquired (CA) skin and soft tissue infections (SSTIs), it is necessary to generate data on the current spectrum and susceptibility profile of associated bacteria. CA methicillin-resistant Staphylococcus aureus (CA MRSA) is increasingly being reported in SSTIs in India and globally. Aims: The present study was undertaken to determine the bacterial profile of CA-SSTIs, to know the contribution of MRSA in these infections, to determine inducible clindamycin resistance in S. aureus and to compare the resistance patterns of isolates from hospital-acquired (HA) SSTIs. Materials and Methods: Eight hundred and twenty patients with CA SSTIs were prospectively studied. Pus samples were cultured and antimicrobial susceptibility pattern determined. Inducible clindamycin resistance was detected by D-test. Laboratory records were analyzed retrospectively to generate data on HA SSTIs. Results: 619 isolates were recovered in CA-SSTIs, of which S. aureus (73%) and Streptococci (12%) were the most common. Pseudomonas aeruginosa (28%) and Acinetobacter spp (18%) were the predominant HA-SSTI pathogens. Susceptibility of CA S. aureus to antibiotics tested was, penicillin (6%), co-trimoxazole (20%), ciprofloxacin (37%), cefazolin (100%), erythromycin (84%), clindamycin (97%), gentamicin (94%) and fusidic acid (95%). No MRSA was found in CA SSTIs whereas 45% of HA S. aureus strains were methicillin-resistant. HA strains demonstrated significantly higher resistance as compared to their CA counterparts (P<0.001). D test was positive in 22% of CA S. aureus tested. Conclusions: In CA SSTIs, methicillin-susceptible S. aureus is the predominant pathogen. Penicillinase-resistant penicillins, clindamycin and erythromycin in that order can be used as suitable antimicrobials for empiric therapy. D test should be carried out routinely. No CA MRSA was detected in the present series.
Keywords: Community-acquired methicillin resistant Staphylococcus aureus, D-test, pyoderma
Skin and soft tissue infections (SSTIs) are common medical problems. They are divided into primary and secondary pyodermas and necrotizing infections.  The organisms causing these infections differ based on the type of lesion. S. aureus is the most common cause of carbuncle and folliculitis. Streptococci are associated with impetigo and cellulitis.  There is considerable variation in the reported antibiotic susceptibility pattern of the organisms isolated from SSTIs with a trend towards increasing resistance. , The treatment of patients attending the outpatient department with SSTIs is usually empirical. In order to define or improve empirical antibiotic therapy, it is important to generate data on the current spectrum and susceptibility profile of bacteria associated with community acquired (CA) SSTIs. The emergence of methicillin-resistant S. aureus (MRSA) in community-acquired settings, CA-MRSA, has further added to the therapeutic dilemma. MRSA as a cause of CA-SSTIs has been reported in a few studies from India. ,, This prospective study was carried out primarily to determine the bacterial spectrum of CA-SSTIs, their antimicrobial susceptibility profile, and the contribution of CA-MRSA in these infections and secondarily to compare the bacterial spectrum and susceptibility profile of CA and hospital-acquired (HA) SSTIs.
The study was initiated after institutional ethics committee approval. Written informed consent was obtained from patients.
Prospective study for the community-acquired arm and a retrospective analysis for the hospital-acquired arm.
Department of Microbiology of a tertiary care teaching hospital in Mumbai.
CA-SSTIs were studied from July 2007 to July 08 and HA-SSTIs from July to December 2007 were retrospectively analyzed.
Patients with CA-SSTIs for the prospective arm and isolates recovered during the same period from patients with suspected HA-SSTIs for the retrospective arm.
A skin and soft tissue infection was considered community-associated/acquired if the infection arose outside the healthcare setting and there was no history of surgery, hospitalization, catheterization, any other invasive procedure in the previous year, antibiotic usage in the previous two months or MRSA isolation in the past. 
A skin and soft tissue infection was considered hospital-associated/acquired if the infection occurred during the patient's hospital stay and/or the specimen was collected after 48 h of admission.
Prospective study (community-acquired)-Pus (preferred specimen) or swab was collected aseptically from all patients at their first visit either on an outpatient basis or in case of hospitalized patients, within 48 h of admission. One specimen was collected per patient. The specimens were processed as per standard protocol for the isolation of aerobic bacteria.  The specimens were inoculated on 5% sheep blood agar and MacConkey's agar which were used as isolation media and mannitol salt agar was used as a selective medium for S. aureus. Any isolate was further characterized up to species level. Antimicrobial susceptibility testing (AST) was performed by the Kirby Bauer disc diffusion method as per Clinical Laboratory Standards Institute (CLSI) standards.  All the isolates of S. aureus were tested for susceptibility against penicillin, oxacillin, cefazolin, cotrimoxazole, ciprofloxacin, erythromycin, clindamycin, gentamicin, vancomycin, linezolid and fusidic acid. Methicillin resistance of S. aureus isolates was confirmed by using oxacillin screen agar. Inducible resistance to clindamycin in S. aureus was detected by 'D' test, as per CLSI guidelines.  Extended spectrum beta lactamase (ESBL) production in E. coli and Klebsiella spp was detected as per CLSI guidelines.  S. aureus ATCC 25923, E. coli ATCC 25922, and P. aeruginosa ATCC 27853 were used as quality control strains for AST.
Retrospective study (hospital-associated): A retrospective analysis of laboratory data was carried out to determine the bacterial spectrum and susceptibility pattern of isolates recovered from non-duplicate patients with suspected hospital-associated SSTIs during 6 months of the same study period.
The results were analyzed.
To determine the significance of the differences obtained between groups, Chi square test was used. P value <0.05 was considered significant.
Eight hundred and twenty patients with CA-SSTIs were included in the study. Patients' age ranged from 1.5 years to 87 years. One hundred and fifteen patients were <15 years of age, 664 were in the age group of 15-60 years and 41 were >60 years of age. Of 820 patients, 525 were males (64%) and 295 were females (34%). Impetigo and boils were common in children whereas abscesses, cellulitis and ulcers were more common in adults. Age or gender-based association with pathogen was not seen.
The lesion-wise distribution of pathogens is shown in [Table 1]. Abscess (338/820) followed by cellulitis (221/820) were common presentations. Of the 820 samples, 592 were culture-positive (72.2%). A single isolate was obtained from 565 samples and 27 samples yielded two isolates each, giving a total of 619 isolates. S. aureus was the commonest isolate (451/619, 72.86%) followed by Streptococci (72/619, 12%). Amongst Gram-negative bacilli, P. aeruginosa accounted for 50/619 (8%) of total isolates. Amongst the 566 hospital-acquired (HA) isolates, P. aeruginosa (158/566, 28%) and Acinetobacter spp (100/566, 18%) were the predominant pathogens.
All the CA S. aureus strains were susceptible to oxacillin which indicates their susceptibility also to cloxacillin, dicloxacillin and nafcillin.  Only 6% of the strains were susceptible to penicillin. All the strains were susceptible to cefazolin, vancomycin and linezolid. Their susceptibility to other antibiotics was: Co-trimoxazole (20%), ciprofloxacin (37%), erythromycin (84%), clindamycin (97%), gentamicin (94%) and fusidic acid (95%). The susceptibility of HA S. aureus strains to the antibiotics was: Penicillin (0%), erythromycin (14%), clindamycin (31%), gentamicin (27%), and ciprofloxacin (2%). Inducible clindamycin resistance was detected in 22% (13/60) of erythromycin-resistant clindamycin-susceptible CA S. aureus strains and in 37.5% (3/8) of HA strains.
The sensitivity of CA and HA P. aeruginosa strains to ceftazidime was 86% and 13%, to netilmicin it was 96% and 5%, ciprofloxacin 88% and 10%, piperacillin 94% and 18% and amikacin 100% and 19% respectively.
ESBL production was not seen in CA E. coli and K. pneumoniae strains but it was seen in 57% (25/44) of E. coli and in 30% (14/46) K. pneumoniae strains isolated from the hospital.
A significantly higher resistance in HA strains was observed for any antibiotic tested for any pathogen (P<0.001).
Amidst increasing reports of MRSA in CA-SSTIs, the present prospective study tried to determine the same in a large number of patients attending a tertiary care teaching hospital. , Patients visiting this hospital are from in and around Mumbai. Centre for disease control (CDC, USA) criteria were strictly applied for classifying SSTIs as CA.
In the present study, methicillin-sensitive S. aureus was the predominant CA-SSTI pathogen. S. aureus has been the commonest pathogen associated with skin and soft tissue infections globally, for many decades. The frequency of association however varies in different regions.  The results of the present study are similar to those reported in the literature. 
No CA-MRSA was detected in the present series. Absence of CA-MRSA may be a reflection of strict inclusion criteria, regional distribution or antibiotic usage. The methodology used for detecting methicillin resistance in S. aureus was as per standard protocol.  During the same period, MRSA was reported from HA strains and from outpatient department (OPD) patients who had history of hospitalization and/or invasive procedure in the previous year. Such patients were not considered to have CA-SSTI in this study. Two corollaries can be drawn from these findings. One, patients availing of outpatient services need not necessarily have only community-acquired infection as per definition. Two, MRSA can be associated with infections in patients attending OPD with prior healthcare history and therefore their specimens should be cultured and antimicrobial susceptibility determined to individualize therapy. Epidemiologic criteria as followed in the present study are not considered as the gold standard for discriminating CA- and HA-MRSA strains and therefore that is a limitation of the present study. Molecular typing of staphylococcal cassette chromosome mec along with the presence or absence of Panton Valentine Leucocidin (PVL gene) are considered as more accurate discriminators of CA and HA-MRSA strains. , With other reports from India highlighting the prevalence of CA-MRSA, a strict surveillance needs to be maintained to detect the emergence of these strains in the community. , A high prevalence, if detected, will then require a change in empiric therapy.
A retrospective analysis of HA-SSTIs revealed that the bacterial spectrum differed markedly in community-acquired and hospital-associated infections. Whereas S. aureus accounted for 73% of CA strains, it accounted for only 17% of HA strains. Gram-negative organisms dominated the spectrum of HA-SSTIs with P. aeruginosa being the most common. In a similar study comparing bacterial profiles and susceptibility pattern of CA and HA infections, S. aureus was found to be the commonest pathogen in both HA and CA infections.  In contrast to the CA setting, more than half of the hospital-associated strains of S. aureus were methicillin-resistant. The proportion of strains resistant to any antimicrobial was significantly higher in all hospital-acquired isolates as compared to community-acquired isolates.
As compared to a previous study from this institute on CA-SSTIs in 2005,  resistance to ciprofloxacin has increased from 17-63% (P<0.0001), to gentamicin from 0.4-6.43% (P=0.16) while resistance to erythromycin (which is representative of other macrolides such as azithromycin)  has significantly decreased from 43-15% (P<0.0001). The high resistance to ciprofloxacin and co-trimoxazole may be attributed to their increased usage in the community since these antibiotics are available in oral form. Quinolones are excreted in human sweat. This may explain the rapidity with which quinolone resistance has emerged in Staphylococci. 
S. aureus strains demonstrated good sensitivity to clindamycin and erythromycin. Inducible clindamycin resistance was detected in 22% of erythromycin-resistant clindamycin-susceptible CA S. aureus strains. This is significant since treatment failure has been detected in D-test-positive patients treated with clindamycin. ,, It is recommended that these antimicrobials be used only after susceptibility testing results indicate so.
S. aureus was the most common organism isolated from both abscess as well as cellulitis. This is consistent with the current knowledge about the causative organisms for these infections.  Gram-negative organisms, especially P. aeruginosa were found to be associated with lesions such as cellulitis and ulcers. Based on these findings, anti-staphylococcal cover would probably suffice for lesions such as boils, carbuncles and abscess whereas therapy for cellulitis and ulcer would need additional Gram-negative cover if culture sensitivity is not routinely performed. Though the association between a particular pathogen and type of SSTI has long been established, analysis was carried out to determine whether there was any change in trend. S. aureus has replaced Streptococci as the most common cause of cellulitis.  In the present series all patients with SSTIs underwent surgical drainage or debridement and received systemic antimicrobial therapy.
The strengths of this study are a large sample size (820 patients), lesion-wise distribution of pathogens, determining change in the trends of antimicrobial susceptibility pattern, a wide range of antimicrobials tested and detection of inducible clindamycin resistance in S. aureus strains. The limitation of this study is that molecular typing of S. aureus strains was not performed to validate the epidemiological characterization of MRSA strains as CA or HA. The other shortcoming is that the study was focused on a population consulting at only one hospital from India.
To summarize, the antimicrobial of choice for treatment of SSTIs should include a penicillinase-resistant penicillin, such as cloxacillin. Since strains also demonstrated good sensitivity to clindamycin and erythromycin, these could be used as second-line therapy or in those with known penicillin allergy. D- test should be routinely performed on all strains of S. aureus to avoid treatment failure due to inducible clindamycin resistance. High susceptibility to fusidic acid (95%), a commonly used local antimicrobial makes it a suitable antimicrobial for topical application. The poor sensitivity to ciprofloxacin (37%) and co-trimoxazole (20%) makes them unsuitable for empirical therapy of SSTIs.
Resistance to antimicrobials is reported both in community as well as hospital pathogens. Surveillance for antibiotic resistance as well as communication of surveillance data is essential to improve empiric antibiotic therapy.  Based on the findings of the present study, anti-MRSA antibiotics are not warranted as empiric therapy of CA-SSTIs. Their inappropriate use will lead to selection of resistant strains.