Inducible clindamycin resistance in Staphylococcus aureus: A study from North IndiaV Gupta, P Datta, H Rani, J Chander
Department of Microbiology, Government Medical College Hospital, Chandigarh - 160 030, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.57393
Source of Support: None, Conflict of Interest: None
Background : The resistance to antimicrobial agents among Staphylococci is an increasing problem. The resistance to macrolide can be mediated by msr A gene coding for efflux mechanism or via erm gene encoding for enzymes that confer inducible or constitutive resistance to macrolide, lincosamide and Type B streptogramin. Aim : The present study was aimed to find out the percentage of Staphylococcus aureus having inducible clindamycin resistance (iMLS B ) in our geographic area using D-test. Also, we tried to ascertain the relationship between Methicillin-resistant Staphylococcus aureus (MRSA) and inducible clindamycin resistance, association of these iMLS B isolates with community or nosocomial setting and treatment options for these iMLS B isolates. Settings and Design : A total of 200 non-duplicate Staphylococcus aureus isolates from various clinical samples from both outdoor and indoor patients were studied. Materials and Methods : Susceptibility to routine antimicrobial agents was carried out using Kirby Bauer method. Methicillin resistance was detected by oxacillin disc on Mueller Hinton agar (MHA) supplemented with 2% NaCl. D-test was performed on all erythromycin-resistant and clindamycin-sensitive Staphylococcus aureus strains to detect inducible clindamycin resistance. Results : Among 200 Staphylococcus aureus strains, 50 (25%) were found to be MRSA and 36 were D-test positive. Also, MRSA isolates showed both higher inducible resistance and constitutive resistance to clindamycin as compared to Methicillin-sensitive Staphylococcus aureus (MSSA). Out of 36 isolates of Staphylococcus aureus showing inducible clindamycin resistance, 24 were from the outpatient department and 12 were recovered from indoor patients. All isolates of Staphylococcus aureus showed 100% sensitivity to vancomycin and linezolid. Conclusions : Clindamycin is kept as a reserve drug and is usually advocated in severe MRSA infections depending upon the antimicrobial susceptibility results. We have reported a higher incidence of iMLS B from both community (66.67%) as well as hospital (33.33%) setup. Therefore clinical microbiology laboratory should report inducible clindamycin resistance routinely.
Keywords: Clindamycin, D-test, erythromycin, Staphylococcus aureus
Erythromycin (a macrolide, ERY) and clindamycin (a lincosamide, CLI) represent two distinct classes of antimicrobial agents that act by binding to the 50s ribosomal subunit of bacteria to inhibit its protein synthesis. Macrolide resistance in Staphylococcus aureus is by diverse mechanisms. The resistance to macrolide can arise by efflux mechanism, classically mediated by msr A gene. Another mechanism is via erm gene, which encodes enzymes that confer inducible or constitutive resistance to macrolide, lincosamide and Type B streptogramin (MLS B resistance).  This resistance mechanism can be constitutive, where r RNA methylase is always produced (cMLS B ) or can be inducible where methylase is produced only in the presence of an inducing agent (iMLS B ). ERY is an effective inducer whereas CLI is a weak inducer. In vitro Staphylococcus aureus isolates with constitutive resistance are resistant to both ERY and CLI whereas those with inducible resistance are resistant to ERY and appear sensitive to CLI (iMLS B ).  If clindamycin is used for treatment of such an isolate (iMLS B ), selection for constitutive erm mutants occurs which may lead to clinical failure. This inducible MLS B resistance can be detected by a simple disc approximation test, commonly referred to as D-test. For this test, an ERY (15μg) disc is placed 15-26 mm (edge to edge) from a CLI (2 μg) disc in a standard disc diffusion test. Following incubation, a flattening of the zone in the area between the discs where both drugs have diffused indicates that the organism has inducible clindamycin resistance. 
In the present study Staphylococcus aureus isolates were tested with ERY and CLI separately and by disc approximation test using D-test to find out the percentage of Staphylococcus aureus having inducible clindamycin resistance in our geographic area. Also, we tried to ascertain the relationship between Methicillin-resistant Staphylococcus aureus (MRSA) and inducible clindamycin resistance, association of these iMLS B strains with community or nosocomial setting and lastly, treatment options for these iMLS B isolates. 
This study included 200 non-duplicate isolates of Staphylococcus aureus isolated from samples of pus/wound swab, respiratory tract, high vaginal swab and body fluids derived from both outdoor and indoor patients of our hospital over a time period of six months (January 2008-June 2008). Staphylococcus aureus were identified using standard microbiological procedures. All Staphylococcus aureus isolates were tested by routine Kirby Bauer method for the following antibiotics - erythromycin (15 μg), clindamycin (2 μg), ciprofloxacin (5 μg), amoxycillin/ clavulanic acid (20/10 μg), gentamicin (10 μg), vancomycin (30 μg), linezolid (30 μg), trimethoprim/sulfamethoxazole (1.25/23.75 μg) (Hi - Media, Mumbai, India). Methicillin resistance in Staphylococcus aureus was detected by oxacillin disc (1 μg), (Hi - Media, Mumbai, India) on Mueller Hinton agar plate supplemented with 2% NaCl and incubated at 35° C for 24 h.  All erythromycin-resistant and clindamycin-sensitive Staphylococcus strains were further tested by D-test for finding inducible clindamycin resistance. On Mueller Hinton agar, standard recommendations for inoculum preparation and inoculation were followed. ERY disc was placed at a distance of 15 mm (edge to edge) from CLI disc. Following overnight incubation at 37° C, appearance of CLI zone close to ERY disc was noted.  The interpretation was done as follows: 
Quality control (QC) of the ERY and CLI discs was performed with Staphylococcus aureus ATCC25923 according to the standard disc diffusion QC procedure. Additional QC was performed with separate in-house selected Staphylococcus aureus strains that demonstrated positive and negative D-test reactions.
Among the 200 Staphylococcus aureus strains, 50 (25%) were found to be MRSA. [Table 1] shows the percentage of MRSA and Methicillin-sensitive Staphylococcus aureus (MSSA) showing inducible resistance to clindamycin (iMLS B phenotype), MS phenotype (resistant to ERY and sensitive to CLI but no induction) and those strains showing combined resistance and sensitivity to both ERY and CLI.
Total of 36 Staphylococcus aureus strains from our sample size of 200 were D-test-positive i.e. these 36 isolates showed in vitro sensitivity to clindamycin but will be in vivo-resistant to clindamycin. Also MRSA isolates showed both higher inducible resistance (20%) and constitutive resistance (46%) to clindamycin as compared to MSSA (inducible -17.3% and constitutive -10%) whereas MS phenotype was predominant among MSSA. Similarly, isolates sensitive to both ERY and CLI were mostly MSSA.
The treatment options for the isolates which were iMLS B showed all these 36 isolates to be 100% sensitive to vancomycin and linezolid, moderately sensitive to gentamicin and amoxycillin/clavulanic acid whereas sensitivity was least to cotrimoxazole and ciprofloxacin [Table 2].
In addition, the isolates of Staphylococcus aureus showing inducible clindamycin resistance were found to have different sources. Isolates from the patients who presented in the outpatient department without any history of recent hospitalization were considered as community-acquired while isolates from the patients in wards or critical care areas after 48 h of hospitalization without any previous history of similar type of lesion at the time of hospitalization were considered as nosocomial. The number of isolates recovered from patients from the outpatient department was 24 and the isolates from patients admitted in different wards or critical care areas of the hospital were 12 in number. This reveals that inducible clindamycin resistance is both community-associated as well as hospital-acquired in our geographic location.
For any clinical microbiology laboratory, the differentiation of erm-mediated inducible MLS B (iMLS B phenotype) isolates from isolates with msrA-mediated (MS phenotype) resistance is a critical issue because of the therapeutic implications of using clindamycin to treat a patient with an inducible clindamycin-resistant Staphylococcus aureus isolate. In recent times, clindamycin has become an excellent drug for some Staphylococcal infections, particularly skin and soft tissue infections and as an alternative in penicillin-allergic patients.  Also, clindamycin has good oral bioavailability making it a good option for outpatient therapy and changeover after intravenous antibiotics. 
Since the iMLS B resistance mechanism is not recognized by using standard susceptibility test methods and its prevalence varies according to geographic location, D-test becomes an imperative part of routine antimicrobial susceptibility test for all clinical isolates of Staphylococcus aureus. Failure to identify iMLS B resistance may lead to clinical failure of clindamycin therapy. Conversely, labeling all erythromycin-resistant Staphylococci as clindamycin-resistant prevents the use of clindamycin in infections caused by truly clindamycin-sensitive Staphylococcal isolates. Hence, Clinical and Laboratory Standards Institute (CLSI) recommends routine testing of all Staphylococcal isolates for iMLS B . 
There are studies which reveal higher constitutive resistance in comparison to inducible resistance. Fiebelkorn et al. in their study found that out of 114 erythromycin-resistant Staphylococcus aureus isolates, 39 demonstrated constitutive resistance to clindamycin while 33 showed inducible resistance.  Various authors have highlighted the relationship of MRSA and MSSA with different phenotypes of CLI and ERY-resistant isolates. In Turkey, Azap et al. reported a higher percentage of inducible CLI resistance in MRSA compared to MSSA (5-7% vis-ΰ-vis 3.6% in MSSA).  Schmitz et al. in the European SENTRY surveillance program found that 93% of the macrolide-resistant MRSA isolates and 44% of the macrolide-resistant MSSA isolates displayed a constitutive MLS B resistance phenotype.  Azap et al. and Delialioglu et al. found that constitutive phenotype is predominant over inducible phenotype in MRSA isolates and MS phenotypes are only found in MSSA. , Kader et al. also reported 53% cMLS B and 43% iMLS B (i.e. higher constitutive resistance) amongst MRSA.  From India, Gadepalli et al., reported that constitutive resistance is significantly higher than inducible phenotype in both MRSA and MSSA. In MRSA constitutive resistance was 38% and inducible resistance was 30%; whereas in MSSA constitutive resistance was 15% and inducible was 10%.  Likewise in our study MRSA isolates had higher constitutive resistance (46%) compared to inducible resistance (20%). But, in our MSSA isolates constitutive resistance (10%) was lower than inducible resistance (17.3%). This difference is significant and can be due to difference in bacterial susceptibility in different geographical areas and also due to varying antimicrobial subscribing patterns of physicians. In addition none of the studies have found MS phenotype among MRSA isolates. We isolated 16% MS phenotype from MRSA. These differences highlight the variations and importance of inducible clindamycin resistance in our geographical setting.
The presence of inducible clindamycin resistance both in nosocomial and community setting is also important. We have reported higher incidence of iMLS B from community (66.67%) than from hospital (33.33%). This may be because of the fact that being an oral drug, clindamycin has been increasingly subscribed by the physicians in outdoor clinical setting due to increasing incidence of community-acquired MRSA.
We have isolated a significant number of MRSA and MSSA isolates with in vitro inducible clindamycin resistance at our institute. This is alarming and raises concern that clindamycin treatment failures may occur with the infection caused by these isolates. There are reports of inducible clindamycin resistance in coagulase-negative Staphylococcus (CoNS) also. Fiebelkorn et al. have also reported inducible clindamycin resistance in CoNS. In their study, amongst 100 CoNS isolates, 82 were erythromycin-resistant. Out of these 82 erythromycin-resistant strains, 57 demonstrated constitutive clindamycin resistance and 25 demonstrated inducible resistance.  Schreckenberger et al. in their study found 14% and 35% inducible clindamycin resistance in CoNS from an inner city hospital and a suburban community hospital respectively.  So, clinical microbiology laboratories should report inducible clindamycin resistance. Moreover, clinicians need to be educated regarding this issue.
The drugs which are recommended for treatment of MRSA-associated infections are vancomycin, linezolid, co-trimoxazole, tetracycline, rifampicin in combination with co-trimoxazole or tetracycline and clindamycin. Resistance against co-trimoxazole, tetracycline and rifampicin has also increased these days. There are many factors which are considered before prescribing clindamycin. The prescription of clindamycin depends upon site of infection, prevalence of MRSA, antimicrobial susceptibility results, whether the patient is admitted in the hospital or is an outpatient and clinician's own experience. The pattern of macrolide resistance in Staphylococci varies in different regions. Depending upon this the prescription rate will not be uniform in different regions. To the best of our knowledge there is no substantial data regarding clindamycin prescription from India. In our hospital setting also, the clinicians consider the above mentioned factors while prescribing clindamycin. It is kept as a reserve drug and is usually advocated in severe in-patient MRSA infections depending upon the antimicrobial susceptibility results. Further, by using clindamycin, use of vancomycin can be avoided. However, expression of inducible resistance to clindamycin could limit the effectiveness of this drug. In such cases, vancomycin and linezolid are the drugs which are considered for therapy. There are reports of decreased vancomycin susceptibility amongst MRSA i.e. VISA (vancomycin-intermediate Staphylococcus aureus) and VRSA (vancomycin-resistant Staphylococcus aureus). In our study we did not find any isolate showing resistance to vancomycin and linezolid. Currently, VRSA is not widespread, but it could well be the next "superbug". 
[Table 1], [Table 2]