The present study was conducted in the Department of Microbiology, Indira Gandhi Medical College, Shimla, Himachal Pradesh, India. The study included 350 non-duplicate isolates of S. aureus from various clinical samples of indoor (IPD) and outdoor (OPD) patients received over a period of nine months (March 2014 to November 2014). Various clinical samples included pus, blood, urine, tracheal aspirates and cerebrospinal fluid (CSF), and distribution of S. aureus strains from origin of recovery is shown in [Table/Fig-1]. The organism was identified by conventional laboratory methods such as colony morphology, catalase test, slide and tube coagulase test and standard biochemical reactions [12].

Antibiotic sensitivity testing was done by Kirby Bauer disc diffusion method and methicillin resistance was identified by using cefoxitin (30 μg) disc and interpreted as per CLSI guidelines [9]. Antibiotic discs used were penicillin G (10 units), cotrimoxazole (25 μg), tetracycline (30 μg), erythromycin (15 μg), clindamycin (2 μg), gentamicin (10 μg), linezolid (30 μg) and teicoplanin (30 μg). All isolates were subjected to inducible clindamycin resistance testing by CLSI recommended D test on Mueller Hinton agar by keeping erythromycin (15μg) disc and clindamycin (2 μg) disc at 15 mm apart (edge to edge) [9]. Blunting of the circular zone of inhibition around clindamycin disc towards erythromycin disc indicated the presence of iMLS_B resistance and was reported as resistant to clindamycin. Quality control (QC) for erythromycin and clindamycin discs was done by using S. aureus ATCCC 25923 according to standard disc diffusion QC procedure. We interpreted the results as follows:

Statistical analysis was performed by using Epi info version 3.3.2, and P-values of < 0.05 were considered statistically significant.

Out of total 350 S.aureus isolates, 158 (45.14%) were from outpatients and 192 (54.85%) from inpatients. Eighty two isolates (23.42%) were MRSA. There was statistically significant difference in MRSA between inpatients (76.83%) and outpatients (23.17%) (p=0.0001). Overall erythromycin resistance was seen in 137 (39.14%) and clindamycin resistance in 108 (30.85%) isolates. Susceptibility to both drugs was found in 213 (60.85%) S.aureus examined. MS phenotype, iMLS_B and cMLS_B phenotypes were seen in 29 (8.28%), 48 (13.71%) and 60 (17.14%) isolates respectively. Clindamycin resistance, both inducible and constitutive, was significantly higher (p<0.05) in MRSA strains (28.39% and 29.62%) compared to MSSA strains (9.29% and 13.38%). Findings of D test are shown in [Table/Fig-2]. Detection of iMLS_B and cMLS_B phenotypes were greater in inpatients (19.79% and 23.43% respectively) than in outpatients (6.34% and 9.4% respectively), which was statistically significant (p<05). Penicillin, cotrimoxazole, tetracycline and gentamicin resistance was seen in 24 (70.57%) 187 (53.42%), 98 (28%) and 35 (10%) isolates respectively. No resistance was detected to teicoplanin and linezolid.

Resistance to most antibiotics used in the treatment of staphylococcal infections is an increasing problem and treatment options have become more limited. The changing pattern in antibiotic susceptibility has led to renewed interest in the use of clindamycin. Good oral bioavailability, low cost, excellent tissue penetration and the fact that it accumulates in abscesses makes it a good option to treat staphylococcal infections. It is a useful drug in the treatment of both methicillin sensitive and resistant staphylococcal strains [13]. It is indicated in skin and soft tissue infections, paediatric staphylococcal infections and patients allergic to beta lactam antibiotics [14]. Therapeutic failures caused by iMLS_B resistant strains are now being reported commonly [15,16]. Routine antimicrobial sensitivity testing can detect cMLS_B phenotypes but iMLS_B resistance is missed if erythromycin and clindamycin discs are placed at non adjacent sites.

Methicillin resistance was identified in 23.42% isolates of S. aureus, in concordance with other studies done in India [17,18]. High rate of methicillin resistance is noted among S. aureus isolates in developed nations [19]. There was statistically higher prevalence of MRSA in inpatients (76.83%) than in outpatients (23.17%) (p=0001). Significant difference of MRSA presence between inpatients and outpatients is reported in literature [20]. A high occurrence of MRSA in inpatient settings can be explained by the fact that organisms develop resistance in closed environment of hospitals and health care facilities due to selection pressure and their convenience in spreading between patients via the health care workers and instruments. Differences in the prevalence rate of MRSA among countries globally and different regions with in a country emphasize the importance of generating a local resistance data to guide empirical therapy.

Among 350 isolate studied, 39.14% were resistant to erythromycin which is comparable to an Indian study done by Deotale et al., [21]. In our study, cMLS_B phenotype predominated (29.62%MRSA; 13.38% MSSA) followed by iMLS_B (28.39% MRSA; 9.29% MSSA) and MS phenotypes (13.58% MRSA; 6.69%MSSA) which is in concordance with a study done in Greece [22].

Incidence of MLS_B phenotypes varies significantly by geographical regions. There are studies which reveal higher constitutive resistance in comparison to inducible resistance in S. aureus isolates. Fiebelkorn et al in their study found that out of 114 erythromycin resistant S.aureus isolates 39 (34.12%)were cMLS_B while 33 (28.94%) were iMLS_B [2]. In our study we found that 30.85% S. aureus isolates were clindamycin resistant; 13.71% were inducible and 17.14% constitutive resistant phenotypes. Angel et al., reported 23.24% inducible clindamycin resistance with no constitutive resistance, and Devdas et al., reported ICR in 6% and cMLS_B resistance in 8% of S. aureus isolates, where as Deotale et al., found 3.6% constitutive and 14.5% inducible clindamycin resistance [21,23,24]. Therefore, studies depict a wide variation in incidence of clindamycin resistance among clinical isolates of S.aureus in different geographic areas.

Relationship of MRSA and MSSA with different resistant phenotypes has been studied by different authors. In Europe, there is a high incidence (93%) of constitutive resistance in MRSA where as the ICR predominates in MSSA strains [25]. Azap et al., reported a high percentage of inducible resistance in MRSA (5-7%) than in MSSA (3.7%) [26]. From India, Debdas et al., reported that constitutive resistance was higher than inducible resistance in both MRSA and MSSA. In MRSA cMLS_B was 23% and iMLS_B was 18%; whereas in MSSA cMLS_B was 3% and iMLS_B two percent [24]. Likewise in our study, constitutive resistance was significantly higher than ICR both in MRSA and MSSA. A study from south India reported cMLS_B in 32% and iMLS_B in 14.2% staphylococcus isolates similar to the present study whereas contrary to our study all iMLS_B phenotypes were MRSA [27]. Prevalence of clindamycin resistance among clinical isolates of S.aureus in various Indian studies is shown in [Table/Fig-3]. The difference in various resistant phenotypes in literature can be due to the difference in bacterial susceptibility in different geographical regions and also due to varying antimicrobial prescribing patterns of clinicians.

There was a significant difference in detection of ICR in inpatient (19.79%) and outpatient isolates (6.34%) and in MRSA (28.39%) and MSSA (9.29%) isolates. This finding may have implications that outpatients with MRSA infections can reasonably be offered clindamycin treatment option due to the lower likely hood of these strains exhibiting iMLS_B resistance. Two hundred seven (77.23%) MSSA strains were found susceptible to clindamycin as opposed to 46.68% MRSA, suggesting a possible role of treatment with clindamycin in such strains. The occurrence of clindamycin resistance varies with geographic area, even with the same city, and methicillin susceptibility [11]. Hence, it should be determined at individual settings. We demonstrate ICR in a high percentage of MRSA and MSSA isolates, more so in inpatient setting, at our institute. This is alarming and raises concern that clindamycin treatment failure may occur without prior testing for inducible resistant phenotypes. Without the D test we would have wrongly reported 48 (13.71%) S. aureus isolates out of total 350 as clindamycin sensitive. Studies like ours should be necessary to generate local sensitivity data which help in guiding empiric therapy and formulate institutional antibiotic policy.

Treatment of staphylococcal infections has always been a challenge for the treating physician, particularly in the backdrop of changing resistance pattern. Keeping the mode of action, side effects and pharmacokinetics in mind of certain drugs like vancomycin and linezolid, clindamycin should be considered for the treatment of severe and resistant staphylococcal infections. Different studies done across the globe show that prevalence of inducible clindamycin resistance varies from place to place. Therefore, we recommend that whenever clindamycin is intended for treatment of staphylococcal infection the clinical microbiology laboratory should test the isolated organism for iMLS_B by D test, before clindamycin susceptibility is reported. Present study giving a magnitude of clindamycin resistance among clinical isolates of S. aureus from this region of the country will help clinicians choose an appropriate therapy.