Antimicrobial Resistance Pattern among Community-Acquired Gram-Positive and Gram-Negative Bacterial Bloodstream Isolates in India

Blood stream infections (BSI) are serious clinical events with life threatening consequences. This scenario has worsened by the emergence of drug-resistant pathogens, making it difficult for clinicians to design optimal therapy-regimen for effective patient care. Antimicrobial resistance (AMR) may arise from various innate bacterial mechanisms: spontaneous mutations, efflux pump, porin loss or by acquisition of mobile genetic elements [1]. In addition, a large number of patients in the community are being catered by private medical practitioners leading to 20-50% of antibiotic misuse or overuse; this remains another major cause for increasing AMR within the community pathogens [2].


Introduction
Blood stream infections (BSI) are serious clinical events with life threatening consequences. This scenario has worsened by the emergence of drug-resistant pathogens, making it difficult for clinicians to design optimal therapy-regimen for effective patient care. Antimicrobial resistance (AMR) may arise from various innate bacterial mechanisms: spontaneous mutations, efflux pump, porin loss or by acquisition of mobile genetic elements [1]. In addition, a large number of patients in the community are being catered by private medical practitioners leading to 20-50% of antibiotic misuse or overuse; this remains another major cause for increasing AMR within the community pathogens [2].
The most efficient way to monitor changing drug resistance

Results
Antimicrobial susceptibility data of 4811 culture positive patient isolates (1825, gram positive; 2986, gram negative) were available for the study analysis. Figure 1 depicts the changing antimicrobial resistance trend for each organism against each drug over the period of 4 yrs. Figure 2 shows the ESBL data for each of the gram negative organism (except bukholderia species) against each drug over the period of 4 yrs. This data has not been explained in the text (except, E. coli) due to its small sample size.

Streptococcus species
Eighty-six isolates were included in the study analysis. All isolates collected were susceptible to vancomycin; more than 90% isolates were susceptible to clindamycin and β-lactam agents. Tetracyline showed relatively weaker antimicrobial activity (32% resistance) in comparison of other antimicrobials tested ( Table 1).

Annals of Clinical and Laboratory Research
ISSN 2386-5180

2015
Depicts the changing antimicrobial resistance trend (in percentage) for each organism against each drug over the period of 4 yrs.
Three strains of entercococcus could not be identified at the species level. Hence, their individual resistance data was not shown; but considered in the pooled analysis. Since total number for individual species are small, only number of resistant organims are indicated. c One strain of streptococcus could not be identified at the species level and hence, their resistance data is not shown; but considered in the pooled analysis. Shows the ESBL data for each of the gram negative organism (except bukholderia species) against each drug over the period of 4 yrs. This data has not been explained in the text (except, E. coli) due to small sample size. The number in brackets next to the year indicates the number of ESBL isolates detected in that year.

Annals of Clinical and Laboratory Research
ISSN 2386-5180  (9) 100 (9) 100 (9) 100 (9) 100 (9) 55.5 (5) 100 (9) ESBL (15) 93.3 (14) 86.6 (13) 93.3 (14) 73.3 (11) 80 (12) 93.3 (14) 93.3 (14) A. lwoffii (62)       Since total number for individual species are small, only number of resistant organims are indicated. c 10 isolates of Enterobacter could not be speciated at the species level. Hence, their individual resistance data was not shown; but considered in the pooled analysis. d 16 isolates of pseudomonas could not be speciated at the species level. Hence, their individual resistance data was not shown; but considered in the pooled analysis. e 15 isolates of Salmonella could not be speciated at the species level. Hence, their individual resistance data was not shown; but considered in the pooled analysis.

Salmonella species
A total 1361 isolates were included in the study analysis. High-antimicrobial activity (>90%) was recorded to all tested antibiotics except fluroquinolones (11-20% resistance, Table 2). Of the fluoroquinolone susceptible isolates, 65% isolates were phenotypically confirmed to be nalidixic acid resistant (data not shown) indicating reduced fluoroquinolone susceptibility.

Discussion
Primary surveillance studies from different regions in India have documented an increase in antimicrobial resistance among important bloodstream pathogens, both in the hospital and community settings. In this retrospective study, we document the antimicrobial resistance pattern of different pathogens associated with bloodstream infections.

Gram positive organisms
Of the 372 S. aureus isolates identified in this study, 29% (n=108) isolates were methicillin resistant; a prevalence similar to that reported by the Indian National Surveillance of Antimicrobial Resistance group (24-31%) [4]. of the total MRSA isolates, 42% (n=45) and 0% (n=0) isolates conferred constitutiveand inducible-clindamycin resistance respectively (data not shown); raising concerns on its efficacy for use during empirical therapy. Hence, glycopeptides (particularly vancomycin) are often considered as the drug of choice for empirically treating MRSA infections, thereby posing a risk for possible increase in MIC creep. In this four-year surveillance period, 4% (n=4) MRSA isolates also reported high level vancomycin resistance (MIC >16 µg/mL) while one of them was simultaneously found to be daptomycin non-susceptible. This co-resistance could be mediated due to thickened cell wall, changes in cellular metabolism, and enhanced cell wall turnover that could interfere with the antimicrobial action of glycopeptides. Also, in case of S. aureus isolates, vancomycin, teicoplanin and linezolid reported high antimicrobial activity of 96.3%, 100% and 100% respectively; a finding similar to other studies [4][5][6][7][8][9][10]. On the contrary, reduced susceptibility was reported by Dubey et al. (vancomycin, 44.9%; teicoplanin, 44.7%; linezolid, 76.6%) [5].
Another gram positive pathogen of increasing concern, Enterococcus spp. recorded high level resistance to most antimicrobials (>50%) tested; with 80.4% isolates exhibiting ciprofloxacin resistance, similar to that reported by Kapoor et al. [11]. Among the glycopeptides, 12.1% isolates were resistant to vancomycin, a frequency similar to that reported by other Indian studies (12-16.9%) [10,12]. This increase in vancomycin resistance could be attributed to the vanA genotype or the acquisition of vanB mobile genetic determinant [13,14]. Although, its presence was not assessed in this study, recent studies from India have reported the involvement of vanA and vanB resistance mechanism in such vancomycin resistant isolates [15].

Gram negative organisms
Gram-negative pathogens are other major causative agents associated with CA-BSI; conferring high-level resistance to most antimicrobials including carbapenems. Resistance among Acinetobacter spp. seems to be a major problem in India, both in nosocomial-and community-settings. In this study, a correlation existed between Acinetobacter spp. and carbapenem resistance; with A. baumannii recording high level resistance in comparison to non-A. baumannii isolates (imipenem, 14.2% vs 0%; meropenem, 80.1% vs 0%; p<0.05, data not shown). This finding was similar to a study from South India, wherein 75% A. baumannii and 22% non-A. baumannii isolates conferred carbapenem resistance [16]. Recent studies in India have molecularly confirmed that acquisition of transposable plasmids like IMP, VIM and OXA could be major mechanism responsible for conferring carbapenem resistance [17,18]. The best example, a case report from Pune, wherein the patient infected with the community acquired panresistant A. baumannii (metallo-β-lactamase IMP-1 producer) strain led to fulminating septicemia, and death of the patient [19].
Another important group of organisms, E. coli and Klebsiella spp. have shown increased resistance to most tested antibiotics in the study. Low level resistance was recorded by E. coli against carbapenems (~6%), whereas higher resistance levels (9-45%) have been recorded across the country [11,20,21]. Also, higher degree of carbapenem resistance was conferred by Klebsiella spp. in comparison to E. coli (34-41% vs 5-6%) and in comparison to other studies from India [11,20,21].
Carbapenems were the most active antimicrobial agents against both, Pseudomonas spp. (46%) and enterobacter spp. (80%); but showed dramatic decrease in susceptibility to aminoglycosides, β-lactam agents, fluoroquinolones, tetracycline and combination of Trimethoprim/Sulfamethoxazole (<50%, Table 2). Similar findings have been reported by various studies across India, both in community-and nosocomial settings [22][23][24]. Increase in resistance is usually being attributed to the irrational and rampant use of broad-spectrum antibiotics by doctors, leading to increase in resistance (antibiotic selection pressure). But a recent study by Kothari et al. suggests that there exists a tremendous load of ESBL and/or AmpC in the community in absence of any direct selection pressure indicating that these genes are widely distributed in the environment [25]. Other studies from India have also documented the existence of SHV-and TEMβ-lactamases contributing to high level drug resistance [24,26]. This may result in significant increase in carbapenem resistance within the community, thereby limiting treatment options.
Resistance among Burkholderia spp. isolates was similar to that reported in other regions of India [27]. High level ceftazidime resistance (70.5%) was recorded, possibly due to overproduction/mutation/deletion of penicillin binding protein 3 [28]. Meropenem and trimethoprim/sulfamethoxazole were found to have high susceptibility, 86% and 80% respectively; a finding similar to that reported by Behera et al. [27]. Besides these, BSI caused by Salmonella spp. was found to be highly prevalent (45.5% [1361/2986], in this study). Salmonella spp. recorded high susceptibility to fluoroquinolones (ciprofloxacin, 80%), whereas large variation has been noted (54-95%) across other studies (either based in a community-or nosocomialsetting [29,30]. Also, Salmonella reported high susceptibility to ceftriaxone (99.9%), similar to that reported by studies in the hospital setting [30].
Although, this study would be first of kind study from India as there is no systematic national surveillance programme, our study has some limitations: 1) Being a standalone diagnostic laboratory we do not receive detailed clinical history for every patient, hence we cannot make present any data in terms of nosocomials and community based infections, and hence the study should be considered as community-based. 2) Also, with no systematic surveillance implemented in the country, this data may not be a true representation but could be considered as a snapshot the actual scenario.

Conclusion
Levels of antimicrobial resistance in community acquired bloodstream infections are increasing among some clinically relevant pathogens in India, most notably A. baumannii, K. pneumonia, E. coli and S. aureus. This report shows that glycopeptides and carbapenems remain important tools in the treatment of difficult-to-treat gram positive and gram negative infections respectively.