Antimicrobial Susceptibility Trends among Pathogens Isolated from Blood: A 6-Year Retrospective Study from a Tertiary Care Hospital in East Sikkim, India

Introduction

Episodes of bloodstream infection (BSI) is a cause of grave concern to a health care facility. BSI complicated by drug-resistant organisms are known to cause increased morbidity, mortality, and hospital expenditure.^1-3 The clinical and epidemiological impact of such antimicrobial resistance (AMR) is a worrisome trend posing severe limitations and challenges to clinicians and health care policymakers alike.⁴⁶

In the epidemiological triad (organism, host, and environment) of BSI, the role of infecting organism is often the most complex to determine, and stringent identification based upon varying parameters is of utmost importance in distinguishing a commensal flora from a pathogen.⁷ Accordingly, adherence to guidelines helps in avoiding unnecessary treatment of colonization in the absence of infection. Although antibiotic usage exerts selective pressure on the selection of AMR organisms, it is the poor/lack of infection prevention and control practices that facilitate the dissemination of these organisms.⁴

The AMR among pathogens, particularly in a hospital setting, is increasing and variations in local trends and epidemiology of BSI require constant surveillance.^2,8,9 The study aims to address the paucity of information regarding the AMR status of blood-borne pathogens from this region. The objective of this study is to understand the local epidemiology of BSI and the status of antibiotic resistance by evaluating the distribution and AMR of the blood-borne organisms isolated over a 6-year duration.

Materials and Methods

Results

Over a period of 6 years, 13,091 blood samples were cultured and 1,340 (10.2%) positive BSIs were reported. ►Fig. 1 represents the year-wise distribution of BSI isolates. ►Table 1 shows the distribution of organisms; the majority of the isolates were from medicine wards (58.9%), followed by intensive care unit (ICU) patients under medicine (11.2%) and pediatrics (10.2%). Fungal isolates were from patients admitted under neonatal intensive care unit and pediatric intensive care unit (53.8%). The maximum bacterial BSIs (32.1%, n = 418) were reported from patients < 20 years of age. Patients < 1 year of age constituted 20.8% (n = 279) of total positive samples, whereas, 79.2% (n = 1,061) were of mean age 43.7 ± 21.5 (SD) years. Male to female ratio was 1.09:1.

Close

Fig. 1

Percentage distribution of organisms isolated from positive blood culture (n = 1,340) over a duration of 6 years.

Export to PPT

Secondary BSI in 14.02% of patients had comparable organisms isolated from the respiratory tract (122, 64.8%), urine (36, 19.1%), and pus (30, 15.6%). Significant polymicrobial growth constituted 1.6% (n = 21) of LCBSI organisms.

Most common isolates among the gram-positive bacteria were coagulase-negative Staphylococci (CoNS; 383, 73.9%), followed by Staphylococcus aureus (97, 18.7%), Enterococcus species (30, 5.7%), and Streptococcus species (8, 1.5% [two Streptococcus pyogenes; three Streptococcus agalactiae; one each of Streptococcus pneumoniae, Streptococcus anginosus, and Aerococcus species]).

Most common isolates among the gram-negative bacteria were Escherichia coli (257, 32.8%), followed by Klebsiella species (175, 22.3%), Salmonella species (122, 15.2%), Acinetobacter species (54, 6.9%), Pseudomonas species (48, 6.1%), Burkholderia cepacia (32, 4.1%), Enterobacter species (26, 3.3%), and Serratia species (15, 1.9%). Other less frequently isolated organisms were Sphingomonas species (8, 1.02%), Proteus species (8, 1.02%), Citrobacter species (7, 0.9%), Aeromonas species (5, 0.6%), Achromobacter species (4, 0.5%), Morganella species (4, 0.5%), Cronobacter sakazakii (3, 0.4%), and one (0.1%) isolate each of Elizabethkingia species, Chryseobacterium species, Shigella species, and Alcaligenes faecalis.

The most common organisms isolated from the ICUs and the wards were Klebsiella species (n = 122) and CoNS (n = 298), respectively. On evaluation of organism’s likelihood of isolation from patients in the ICUs or the wards (odds ratio [OR] 95% confidence interval [CI]), Klebsiella species (OR: 6.5, CI: 4.6–9.2, p < 0.0001), Enterobacter species (OR: 6.0, CI: 2.5–14.3, p < 0.0001), Enterococcus species (OR: 2.5, CI: 1.2–5.1], p = 0.02), Acinetobacter species (OR: 1.9, CI: 1.1–3.3, p = 0.03), and Pseudomonas species (OR: 2.0, CI: 1.1–3.6, p = 0.02) were likely from the ICU patients whereas CoNS, S. aureus, E. coli, and Salmonella species were more likely from the ward patients.

►Table 2 represents the antibiotic resistance of the commonly isolated gram-negative bacteria. E. coli, Klebsiella, and Enterobacter species had similar resistance pattern with > 60% resistance to ampicillin, cefuroxime, ceftriaxone, and cefepime. Acinetobacter species displayed high resistance to almost all antibiotics tested, except, tigecycline and colistin. Salmonella paratyphi A was the frequent isolate among Salmonella species (59%). Only nalidixic acid-resistance (84.4%) was prominent among the Salmonella species.

Antibiogram of gram-positive organism indicates high resistance to benzylpenicillin, erythromycin, and ciprofloxacin as represented in ►Table 3. Vancomycin-resistance was lower in Staphylococcus species than Enterococcus species. Among enterococci, vancomycin resistance was seen only in Enterococcus faecalis (80%) and Enterococcus faecium (20%), whereas high-level gentamicin (HLG) resistance was observed higher in E. faecium (53.8%) than in E. faecalis (46.1%). Enterococcus casseliflavus and Enterococcus gallinarum isolates were vancomycin and HLG-sensitive isolates.

Fungi constituted 2.9% of positive cultures. They were caused by yeast-like organisms such as Candida albicans and Candida tropicalis. Candida species were commonly isolated from ICU patients (82.1%) with a mean age of 41.9 ± 23.7 (SD) years and a male to female ratio of 1.05:1. Neonates contributed to 43.6% of total fungal BSI (►Table 1).

On the year-wise evaluation of MDRI of the commonly isolated organism for a duration of 6 years, it is evident that there is a significant rise in resistance pattern of all isolates, particularly A. baumannii complex, P. aeruginosa, and S. aureus (►Fig. 2). A. baumannii complex and P. aeruginosa had the highest number of multidrug resistances throughout the duration of the study period.

Close

Fig. 2

Multiple antibiotic resistance index of commonly isolated organisms from bloodstream infections over a duration of 6 years.

Export to PPT

Discussion

Similar to several studies, gram-negative bacilli were the common isolates (60.2%) in our study.^8,13-15 Gram-positive cocci were the commonly isolated organisms at the beginning of our study in 2013, with CoNS as the frequently isolated group.¹⁶ However, by 2014 onward, upward trend of gram-negative bacilli isolation increased when compared with a decline in the frequency of CoNS isolation. Overall, a gradual decrease in isolation of both types of organisms, with gram-positive cocci more so, could perhaps be attributed to a better understanding of sample collection, infection control, and prevention.^9,13

In our study, the rise in total antibiotic resistance of S. aureus is of concern. Till 2017, S. aureus had the lowest MDR index values, but by 2018, it had abruptly risen to become the organism with the third-highest values. This rise could perhaps be attributed to the majority of isolates developing resistance to many of the first-line antibiotics such as fluoroquinolones, erythromycin, and penicillin.^9,14 On the contrary, few gram-positive cocci in our study were vancomycin-resistant, which is similarly reported by Zhu et al.⁹

Similar to the findings by Djuric et al, the overall resistance among gram-negative isolates was observed highest to β-lactams, trimethoprim-sulfamethoxazole, and aminoglycosides.¹³ The MDRI of commonly isolated gram-negative bacilli was alarmingly high and remained consistently high throughout our study period. The resistance pattern is comparable to reports indicating a predictable rise of MDR organisms in the future.^4,6

Carbapenem resistance in our study was highest for Acinetobacter species (40.7%), also reported by Kumar et al.¹⁵ Carbapenem resistance of > 20% was detected in Pseudomonas species and Burkholderia species, whereas the lowest resistance was among the Enterobacteriaceae, which is in contrast to the study by Datta et al¹⁷ Nevertheless, close monitoring and improvement of infection control should be implemented to avoid the emergence of carbapenem-resistant Enterobacteriaceae (CRE).⁴ All the gram-negative isolates in our study were mostly sensitive to colistin, which is the drug of choice in treating CRE. However, according to the recent CLSI guidelines, broth microdilution is the only acceptable method for detecting colistin resistance and Vitek 2 GN-AST cards use agar dilution method for antibiotic sensitivity testing.¹¹ Consequently, the importance of the colistin susceptibility reports from our study remains uncertain.

Though the MDR indices of E. coli and Klebsiella species were fairly constant, it was consistently high throughout the duration of the study. E. coli was the most common gram-negative organism causing BSI in our study and it is also implicated as a frequent organism causing BSI in Europe and in the SENTRY surveillance program.^4,18

In our study, the MDRI of A. baumannii complex has been the highest. It exhibited resistance to mostly all the antibiotics, except tigecycline and colistin, which is also reported elsewhere.^13,15 Surveillance studies in Europe have revealed that Acinetobacter species have an alarmingly high level of AMR to routinely used antibiotic groups of fluoroquinolones, aminoglycosides, and carbapenems.⁴

Though the isolation rate of Pseudomonas species in this study is less (3.6%), its resistance to many antibiotics (excluding colistin) increased significantly over 6 years. P. aeruginosa is an organism responsible for the majority of health care-associated infections, with its intrinsic and acquired resistance to several antibiotics adding pressure to a preexisting global AMR burden.⁴

The current situation of BSI in India is rather complex to quantify as regional variations abound.¹⁴ The Centers for Disease Control and Prevention and SCOPE project (Surveillance and Control of Pathogens of Epidemiological Importance) in the United States monitor BSI via a nationwide network of hospitals, similar to the SENTRY Antimicrobial Surveillance Program, the European Antimicrobial Resistance Surveillance Network, Central Asian and Eastern European Surveillance of Antimicrobial Resistance network, and the Asian-Pacific Research Foundation for Infectious Diseases of South Korea.^4,7,13,18,19 The SCOPE nationwide collective effort is also adopted by Brazil.⁸ Such surveillance programs are the need of the hour in India as there are inadequate databases and no repository of resistant pathogens that streamline active surveillance programs (to guide clinical decisions on the use of antimicrobials).²⁰ Hopefully, the Antimicrobial Resistance Surveillance and Research Network and National Action Plan on Antimicrobial Resistance will provide some solutions in tackling the problem of emerging AMR pathogens collective for the Indian subcontinent.²⁰

Conclusion

A progressive increase in multidrug (≥3 antibiotics)-resistant organisms is posing several challenges to our hospital infection-control operations. More studies are required to comprehend the etiology of such AMR organisms to aid appropriate antibiotic usage, control, and contain the spread of AMR organisms in the future.