Staphylococcal Cassette Chromosome mec (SCCmec) typing of clinical isolates of coagulase-negative staphylocci (CoNS) from a tertiary care hospital in New Delhi, India

Sir,

Coagulase-negative staphylococci (CoNS) have emerged as significant aetiological agent of health-care associated infections, particularly in patients with indwelling devices, prosthesis and implants and those subjected to invasive medical procedures. Rise in the incidence of infections with CoNS has also been attributed to the increase in hospitalized immunocompromised patients1. Methicillin resistance, mediated by mecA gene encoded low affinity penicillin-binding protein (PBP) 2a, has been observed in high proportion of CoNS isolates. Accurate detection of methicillin resistance among CoNS isolates is an important factor in guiding antibiotic therapy. The mecA gene is a part of the mobile genetic element Staphylococcal Cassette Chromososme mec (SCCmec), which acts as a vehicle for horizontal transfer of antibiotic resistance genes to organisms like Staphylococcus aureus12. Hospital-acquired infection associated methicillin resistant CoNS (MR-CoNS) serve as a reservoir of genetically diverse SCCmec types, each type being associated with a different antibiotic resistance pattern. The distribution of SCCmec types varies with host species of CoNS and with geographical locations2. Molecular characterization by SCCmec typing of MR-CoNS is an essential epidemiological tool for studying the evolution of these genetic elements and providing useful information regarding the antibiotic resistance pattern in staphylococci. There are limited studies reported from India on speciation and antibiotic susceptibility testing of MR-CoNS by phenotypic methods34. Thus, the present study was undertaken to determine the distribution of SCCmec types in MR-CoNS, isolated from patients admitted to a tertiary care hospital in north India and study its relation with antibiotic resistance.

The study was conducted in the Bacteriology laboratory, department of Microbiology, All India Institute of Medical Sciences (AIIMS), New Delhi, India. The study protocol was approved by the Ethics committee of the Institute. The study included 124 consecutive non-duplicate isolates of CoNS obtained from clinical specimens of patients admitted in various wards between May and October, 2012. Criteria for inclusion of a blood culture isolate of CoNS as clinically significant were isolation of the same strain in pure form from multiple blood culture samples of the same patient taken at different points of time or in presence of central venous catheters (CVCs), isolation of the same strain from culture of catheter segment and peripheral blood along with at least two of the following clinical criteria in absence of evidence of infection at any other site: temperature >38°C or <36°C; heart rate >90/min; respiratory rate >20/min; leucocyte count >12000/μl or <4000/μl or >10 per cent immature neutrophil granulocytes. For CoNS isolated from samples other than blood culture, only strains, which were isolated as pure growth from clinical samples of a patient, more than once, were considered as significant pathogens and included in the study5.

The CoNS isolates obtained from soft-tissue infection site and pus samples, blood and bones on culturing onto 5 per cent sheep blood agar plates, were screened for CoNS on the basis of colony morphology, Gram stain reaction, catalase, tube and slide coagulase tests. Species identification of CoNS was done based the panel of biochemical reactions described by Kloos and Bannerman6. All CoNS isolates were subjected to antimicrobial susceptibility testing on Mueller-Hinton agar using the Kirby-Bauer disc diffusion method as recommended by Clinical and Laboratory Standards Institute (CLSI)7 for the following antibiotics: penicillin (10 units), amoxicillin-clavulanic acid (20/10 μg), gentamicin (10 μg), amikacin (30 μg), doxycycline (30 μg), trimethoprim-sulphamethoxazole (1.25/23.75 μg), ciprofloxacin (5 μg), erythromycin (15 μg), clindamicin (2 μg), and linezolid (30 μg). Susceptibility to vancomycin was performed by microbroth dilution taking the minimum inhibitory concentration (MIC) interpretative criteria of ≤4, 8-16 and ≥ 32 μg/ml as sensitive, intermediate and resistant, respectively7. Staphylococcus aureus American Type Culture Collection (ATCC) 25923 and S. epidermidis ATCC 12228 were used as control strains.

Methicillin resistance in CoNS was determined by disc diffusion method using a cefoxitin disc (30 μg) with the interpretative criteria ≤ 24 mm (resistant) and ≥ 25 mm (susceptible) and by mecAgene PCR using the published primers mecA P4 (5’TCCAGATTACAACTTCACCAGG3’) and mecA P7 (5’CCACTTCATATCTTGTAACG3’)8. S. Aureus ATCC 33591 and S. epidermidis ATCC 12228 were used as positive and negative controls for PCR, respectively. Inducible resistance to clindamycin in erythromycin-resistant CoNS isolates was detected by D-test as per CLSI guidelines7. Multi-drug resistance was defined as resistance to three or more classes of antibiotics including β-lactams.

Isolates positive for mecA were further subjected to SCCmec typing (I-IV) by multiplex PCR which included four pairs of primers for locus A (exclusive to SCCmec type I), locus B (exclusive to SCCmec type II), locusD (common to SCCmec types I, II and IV) and locus E (exclusive to SCCmec type III) as described by Oliveira et al8. Identification and differentiation of SCCmec types was based on the separation of amplicons by agarose gel electrophoresis of the multiplex PCR products (Figure). The following S. aureus strains were used as controls for SCCmec typing: 665U (type I), UK EMRSA-16 (type II), UK EMRSA-1 (type III) and UK EMRSA-15 (type IV) (kindly provided by Dr Edet E. Udo, University of Kuwait, Kuwait).

Close

Figure

Multiplex PCR for SCCmec typing of CoNS. Lane 1: CoNS isolate harbouring SCCmec type III (243 bp). Lanes 2-4: CoNS isolates negative for mecA. Lane 5: CoNS isolate harbouring SCCmec type I (342 & 495 bp). Lane 6: CoNS isolate harbouring SCCmec type II (342 & 284 bp). Lane 7: CoNS isolate harbouring SCCmec type IV (342 bp). Lane 8: 100 bp DNA ladder.

Export to PPT

Potential risk factors for CoNS infection in hospitalized patients were evaluated by reviewing the medical records. Statistical analyses were performed using STATA version 11.2 software (STATA Corp LP, College Station, TX, USA) using the Chi-square test.

Of the 124 CoNS isolates, 79 (63.7%) were recovered from infected surgical site soft tissue samples and post-operative wound discharge/pus specimens, followed by blood (43, 34.7%), and bone tissues (2, 1.6%). Seven species accounted for all CoNS isolates: S. haemolyticus (n=45, 36.3%), S. epidermidis (n=23, 18.5%), S. schleiferi subspecies coagulans (n=18, 14.5%), S. schleiferi subspecies schleiferi (n=14, 11.3%), S. warneri (n=12, 9.7%), S. capitis (n=11, 8.9%) and S. intermedius (n=1, 0.8%).

Resistance to methicillin in CoNS was detected in 44 (35.5%) isolates by cefoxitin disc diffusion test. All these CoNS isolates also tested positive for mecA gene by PCR. All methicillin sensitive (MS-CoNS) isolates by cefoxitin disc diffusion assay were negative for mecA gene PCR.

Rates of methicillin resistance among all species of MR-CoNS were highest in S. haemolyticus (n=29, 64.4%) followed by S. epidermidis (n=6, 26.1%) and S. schleiferi (n=6, 18.7%) isolates. MR-CoNS showed higher level of resistance to all non β-lactam antimicrobials tested as compared to MS-CoNS, difference being significant for amikacin (18.2 vs 6.2%, P=0.04), ciprofoxacin (68.2 vs 37.5%, P<0.01) and clindamycin (36.4 vs 15%, P<0.01). Multi-drug resistance was observed in 72.7 per cent (32/44) of MR-CoNS isolates and was most frequently detected among methicillin-resistant S. haemolyticus isolates. Of all CoNS isolates, highest rates of antibiotic resistance were observed in S. hemolyticus. Resistance to erythromycin, ciprofloxacin, clindamycin, co-trimoxazole, doxycycline, amikacin and gentamicin were present in 51.1, 48.9, 31.1, 19.6, 20.0, 15.6 and 4.4 per cent of all S. haemolyticus isolates, respectively.

All CoNS isolates included in our study were susceptible to vancomycin and linezolid. All MS-CoNS were uniformly susceptible to penicillin and amoxicillin-clavulanic acid. None of erythromycin-resistant CoNS isolates expressed inducible resistance to clindamycin.

The 44 MR-CoNS isolates were tested for SCCmec types using the multiplex-PCR for SCCmec types I, II, III and IV and 27 (61.4%) isolates were identified with SCCmec type I, three (6.8%) with SCCmec type II, seven (15.9%) with SCCmec type III and five (11.4%) with SCCmec type IV. Two (4.5%) isolates of MR-CoNS were not typable using this method. Approximately 60 per cent of SCCmec type I positive CoNS isolates (16/27) belonged to S. haemolyticus species. The distribution of SCCmec types among methicillin-resistant S. haemolyticus, the most frequently isolated MR-CoNS species, indicated that 55.2 per cent harboured SCCmec I, 20.7 per cent SCCmec III, 13.8 per cent SCCmec IV and 6.9 per cent SCCmec II (Table I).

Table I Distribution of MR-CoNS according to sample types and distribution of SCCmec type among different MR-CoNS species (n=44)

An analysis of resistance rates was performed for the susceptibility profile of MR-CoNS to non-β-lactam antimicrobials among different SCCmec types (Table II). It was observed that isolates with SCCmec type I, the most frequently identified SCCmec type, exhibited high rates of resistance to ciprofloxacin, rifampicin, erythromycin, clindamycin, doxycycline and co-trimoxazole (70.3, 59.2, 51.7, 37.1, 33.3 and 25.9% of isolates, respectively).

Table II Distribution of SCCmec types in MR-CoNS isolates (n=44) according to resistance pattern to non β-lactams antibiotics

No significant association was observed between isolation of MR-CoNS and variables like duration of mean hospital stay, presence of prosthetic devices and intravenous catheters and history of admission to intensive care units (ICUs).

In our study, S. haemolyticus was the commonest species of CoNS followed by S. epidermidis. This is in contrast to studies from developed countries where S. epidermidis is the most frequently isolated species of CoNS from hospitalized patients, thereby highlighting the fact that epidemiology of health-care associated CoNS infections varies according to geographical locales29. Our finding was in concordance with a previous study from north India where S. haemolyticus was the commonest species of CoNS isolated from blood cultures4. The high prevalence of S. haemolyticus could be related to its ability to adapt to selective pressures such as antimicrobials and biocides present in the hospital environment10. Hospital-acquired CoNS act as reservoir of antibiotic resistance genes, with increase in resistance rates to multiple antibiotics being reported all over the world11. Our study also highlights a high prevalence of multi-drug resistance CoNS with clear dominance of resistance in S. haemolyticus.

In the present study, methicillin resistance was detected in a high percentage of CoNS isolates (34.4%). High rates of methicillin resistance ranging from 52 to 66 per cent in Indian isolates of CoNS from patients with health-care associated infections have been previously reported312. These studies have also reported high rates of penicillin resistance in MS-CONS isolates (81-100%). In contrast, none of our MS-CoNS isolates showed resistance to penicillin and amoxicillin-clavulanic acid, implying that these drugs remained the preferred antibiotics for treating infections with MS-CoNS in our setting. The difference in penicillin resistance in MS-CoNS isolates can be explained by the variations in antibiotic selection pressure caused by different antibiotic policies in different hospital settings.

In our study, MR-CoNS isolates were significantly more resistant to ciprofloxacin, clindamycin and amikacin than MS-CoNS isolates. In addition, more than 70 per cent of MR-CoNS isolates were multi-drug resistant. None of our CoNS isolates were resistant to vancomycin and linezolid. Given the high proportion of MR-CoNS resistant to the different non-β-lactam antibiotics, vancomycin and linezolid should be included in empirical therapy for patients with MR-CoNS infections. However, empirical usage of vancomycin and linezolid should be avoided in treating CoNS infections before recording results of methicillin resistance testing, since overusage of these antibiotics will lead to emergence of glycopeptide and oxazolidinone resistant CoNS strains. CoNS isolates with linezolid resistance and decreased susceptibility to vancomycin have already been reported13.

In our study, S. haemolyticus was the most frequent MR-CoNS with >60 per cent being methicillin resistant. Barros et al14 also observed that 87 per cent of S. haemolyticus isolates from various clinical samples were methicillin resistant. High methicillin resistance rates ranging from 75-90 per cent in S. haemolyticus isolates from blood cultures have also been reported15. High level of antimicrobial resistance in S. haemolyticus has been attributed to the significant number of insertion sequence elements, which represent hotspots for acquisition of antibiotic resistance genes15.

Among all MR-CoNS isolates, SCCmec type I was the commonest followed by SCCmec type III and 55 per cent of S. haemolyticus isolates harboured SCCmectype I. SCCmec types I and III have been previously reported as the predominant SCCmec types in hospital-acquired isolates of MR-CoNS2. SCCmec type III is harboured predominantly in methicillin-resistant S. epidermidis isolates and SCCmec type I is more common in methicillin-resistant S. haemolyticus isolates16. The predominance of SCCmec type I in our isolates could be attributed to the higher number of methicillin-resistant S. haemolyticus isolates compared to methicillin-resistant S. epidermidis. Low proportion of isolates for SCCmec elements such as IV can be explained by the fact that all our isolates were from patients with hospital-acquired infections and SCCmec type IV is predominantly detected in community-acquired MR-CoNS isolates.

The present study had limitations. The multiplex PCR used for SCCmec typing was capable of detecting only SCCmec types I-IV. With evolution of genetically diverse newer SCCmec elements, more than ten SCCmec types have now been reported in CoNS2. Non-typability of two isolates in our study was probably due to the fact that they harboured SCCmec genes different from types I-IV. Previous studies have demonstrated that majority of clinically significant MR-CoNS isolates from patients with health-care associated infections harbour SCCmec types I-IV17-19. In a study from Turkey, it was observed that more than 70 per cent of MR-CoNS isolates from blood culture harboured SCCmec types I-IV either alone or in combination18. Characterization of SCCmec gene of MR-CoNS isolates from a neonatal intensive care unit in Brazil also demonstrated a high prevalence of SCCmec types I-IV with an overall predominance of SCCmec type I17.

It has been have shown that the frequency of SCCmec among CoNS is increasing due to the wide dissemination of the microorganism in hospital-acquired infections, which is associated with the potential genetic transfer of resistance elements among S. aureus11. We observed a high frequency of SCCmec I and III in our isolates. Majority of MRSA clinical isolates, that have been previously characterized in our hospital, possessed SCCmec type III and were significantly more resistant to antibiotics19.

In conclusion, presence of a particular SCCmec type inhospital-acquired CoNS can be a predictor of antibiotic resistance pattern of the isolates. Since the pattern varies from one health-care setting to another, molecular typing of this genetically diverse group of organisms and establishing an association with antibiotic resistance need to be individualized for each laboratory where genotyping facilities are available. This can generate useful data elaborating the molecular epidemiology of nosocomial CoNS isolates particularly in countries like India where genotyping studies for drug-resistant CoNS are lacking. Since CoNS are capable of transmitting drug-resistance genes to organisms like S. aureus, molecular typing in adequately large number of CoNS isolates and prospectively monitoring their drug resistance can be helpful in designing antibiotic policies and hospital infection control strategies.