Antimicrobial resistance (AMR) poses a threat to the treatment of infectious conditions and leads to an increase in morbidity, mortality, disease transmission, and healthcare costs. In 2019, bacterial AMR was associated with ∼4.95 million deaths globally, with the highest rates of mortality reported in sub-Saharan Africa and South Asia. Reference Murray, Ikuta and Sharara1 Disparities in AMR morbidity and mortality are multifactorial. High rates of AMR often require the use of more expensive, “last resort” antibiotics, which presents a significant challenge in low- and middle-income countries (LMICs). 2,Reference Singh, Das, Singh, Gajamer, Pradhan, Lepcha and Tiwari3 Despite significant progress with AMR surveillance in recent years, LMICs in South and Southeast Asia continue to struggle with the establishment and expansion of regulatory programs due to technical, financial, and logistical constraints in healthcare infrastructure. Reference Gandra, Alvarez-Uria, Turner, Joshi, Limmathurotsakul and van Doorn4
India has one of the highest burdens of bacterial infection in the world, and its population consumed more antibiotics than any other country from 2000 to 2010. Reference Singh, Das, Singh, Gajamer, Pradhan, Lepcha and Tiwari3,Reference Laxminarayan and Chaudhury5,Reference Van Boeckel, Gandra, Ashok, Caudron, Grenfell, Levin and Laxminarayan6 Various highly resistant microorganisms with novel resistance mechanisms, such as the New Delhi metallo-β-lactamase enzymes in Enterobacteriaceae, were initially isolated in India and are now recognized worldwide. Reference Laxminarayan and Chaudhury5 Due to the gravity of this situation, India’s Ministry of Health and Family Welfare declared AMR a national health priority and, in conjunction with the World Health Organization (WHO), developed the National Action Plan on Antimicrobial Resistance (NAP-AMR) for 2017–2021. 7 Six strategic priorities were identified in this action plan, including strengthening of knowledge and evidence through AMR surveillance, reduction in the incidence of infection through effective prevention and control policies, and the optimization of antimicrobial use in healthcare, animal domestication, and food production. 7
Significant AMR to firstline therapeutic agents (ie, β-lactams, cotrimoxazole, and fluoroquinolones) among gram-negative microorganisms in India has led to their replacement by broad-spectrum agents (ie, carbapenems and aminoglycosides). AMR has now been documented in urinary pathogens in Indian hospitals servicing both urban and rural populations, and decreasing effectiveness of “last resort” antimicrobial agents like carbapenems is an alarming phenomenon. Reference Singh, Das, Singh, Gajamer, Pradhan, Lepcha and Tiwari3,Reference Bent, Nallamothu, Simel, Fihn and Saint8–Reference Patel, Soni, Bhagyalaxmi and Patel11 Susceptibility testing in commensal microorganisms indicate that AMR in India is widespread in rural communities outside the tertiary hospital setting. Reference Mathai, Chandy and Thomas12–Reference Purohit, Lindahl, Diwan, Marrone and Lundborg14 Specifically, Singh et al Reference Singh, Das, Singh, Gajamer, Pradhan, Lepcha and Tiwari3 found multidrug resistance in >25% of the commensal Escherichia coli isolated from patients in rural villages of northeast India. Furthermore, researchers identified variation in resistance patterns between villages within the same state, suggesting a greater degree of diversity in resistance mechanisms than initially anticipated. Reference Singh, Das, Singh, Gajamer, Pradhan, Lepcha and Tiwari3 Although these studies of commensal organisms provide valuable approximations, they likely underestimate the resistance rates among pathogenic microorganisms isolated from clinical specimens. Patients from rural areas admitted to tertiary-care centers in India often already harbor pathogens that are resistant to available antimicrobial agents.
Although many reports have documented AMR among clinical isolates in the tertiary-care setting in India, there is a paucity of similar data from the primary healthcare centers (PHCs) and rural communities in India. Reference Mathai, Chandy and Thomas12 Because these data have suggested that primary healthcare settings in rural areas might be playing a more significant role in the pathogenesis of AMR than previously appreciated, we conducted a pilot survey as part of the first phase of a larger study (1) to characterize the resistance profiles of infections caused by sentinel, antimicrobial-resistant pathogens causing infections in the rural setting and (2) to estimate the extent of the AMR among rural patients referred from a primary to a tertiary-care center for advanced care.
Methods
We carried out a retrospective prevalence survey of urine, wound, and blood cultures obtained from patients admitted to the the Jagadguru Shivarthreeshwara Hospital in Karnataka, South India, during January and February 2021. The hospital is a tertiary-care facility with 1,800 general medical beds and ∼260 critical care beds. It caters to the healthcare needs of the rural population in the Mysore district and the population of Mysore city.
The study population included patients aged >18 years who were referred to the hospital from surrounding rural village with infections acquired in the urinary tract, wounds, or bloodstream. We focused on isolates from patients with urine, wound, and blood infections because of the relatively high predictive value for positive cultures from these anatomic sites.
Patients were enrolled in the study after informed consent if they (1) lived in a rural area, (2) acquired their infections in the community before referral to the tertiary hospital, and (2) had not been hospitalized previously.
The case definitions for obtaining cultures and selecting study participants were as follows:
-
1. Urinary tract infection: any patient over the age of 18 who presented with an oral temperature ≥38°C, burning on micturition, dysuria, urine analysis with pyuria and a midstream urine specimen that yielded growth on culture.
-
2. Bacteremia: any patient over the age of 18 who presented with signs and symptoms suggestive of systemic infection and had a positive blood culture.
-
3. Wound infection: any patient over the age of 18 with cultures obtained from skin abscesses or from purulent discharge from the wound.
All cases were ascertained by chart review. We obtained approval from the Institutional Review Board at the University of Florida and the Local Ethics Committee at the study hospital.
Bacterial identification and antimicrobial susceptibility testing (AST) were carried out on the VITEK 2 fully automated system (BioMérieux, Marcy-l’Étoile, France) that uses AST cards based on the broth microdilution minimum inhibitory concentration platform according to the benchmarks established by the Clinical and Laboratory Standards Institute (CLSI) in the United States. The Jagadguru Shivarthreeshwara Hospital Microbiology Laboratory, where our study isolates were cultured, provides microbiology services for the hospital but is also a national reference laboratory that has links with the World Health Organization (WHO), the CDC, and clinical laboratories across countries in the British Commonwealth.
Results
During the study period, 100 urine cultures, 102 blood cultures, and 102 wound cultures yielded growth.
Urine
Escherichia coli was the most frequently isolated pathogen (47%), followed by Enterococcus spp (13%) and Candida spp (15%) (Table 1). A resistance rates >60% were detected for cephalosporins, quinolones, and penicillin (Table 2). Resistance rates >25% among urinary pathogens were also documented for aminoglycosides. A relatively high rate of resistance to carbapenems (>25%) was also documented. Colistin was sensitive in 95% of strains of Acinetobacter baumannii, Enterobacter spp, Escherichia coli, and Klebsiella pneumoniae.
Positive Cultures by Cultures by Organism and Culture Site
a Morganella morganii, Pantoea agglomerans, Proteus spp, and Providencia spp
b Achromobacter denitrificans and Burkholderia cepacia.
Overall Resistance Profiles of Wound, Blood, and Urinary Pathogens by Antimicrobial Class
Note: The remaining percentages for each drug class reflect intermediate susceptibility.
Blood
The most frequently isolated microorganisms were coagulase-negative Staphylococcus spp (22%), K. pneumoniae (19%), and E. coli (15%) (Table 1). Resistance rates were >65% for cephalosporins, quinolones, and penicillin (Table 2). High resistance rates were also noted for aminoglycosides (48%) and carbapenems (41%). Most (88.5%) of the isolates also displayed intermediate sensitivity to colistin.
Wounds
Staphylococcus aureus was the most frequently isolated organism (22.4%), followed by Klebsiella pneumoniae (14%), whereas coagulase-negative Staphylococcus, E. coli, and Pseudomonas aeruginosa each comprised 13% of the wound isolates (Table 1). Almost one-fifth of all organisms isolated were multidrug resistant (17.1%), and the highest rates of resistance were attributed to quinolones (65%), penicillins (47%), and cephalosporins (45%) (Table 2). The resistance rates for aminoglycosides and carbapenems were 18% and 13%, respectively.
The antimicrobial resistance profiles of the most common microorganisms isolated from wounds, bloodstream, and urine in the study population are summarized in Tables 3a and 3b.
Percentage Resistance to Available Antimicrobial Agents Among Common Gram-Negative Microorganisms Isolated From Wounds, Blood, and Urine Obtained From Patients in Rural Areas
Note. TMP-SMX, trimethoprim-sulfamethoxazole.
Percentage Resistance to Available Antimicrobial Agents Among Common Gram-Positive Microorganisms Isolated From Wounds, Blood, and urine Obtained From Patients in Rural Areas
Note. TMP-SMX, trimethoprim-sulfamethoxazole.
Discussion
High rates of antimicrobial resistance portend poor patient outcomes. Our study confirmed a high prevalence of resistance among clinical isolates at a tertiary-care hospital serving a large catchment of rural communities. Similar to surveillance data on healthcare-associated infections published by the US Centers for Disease Control and Prevention (CDC), 15 E. coli, Enterococcus spp, and Candida spp were the 3 most common causes of symptomatic UTIs in this population. However, the CDC data were drawn from critical care patients with indwelling urinary catheters in US tertiary-care hospitals, and the specimens in the current study were obtained from symptomatic patients from rural areas. Previous studies have shown E. coli to be the most common pathogen in both rural and urban populations. We found high rates of resistance of urinary pathogens to aminoglycosides, as has been described in previous studies. Reference Balachandra, Sawant, Huilgol, Vithya, Kumar and Prasad16
Other studies have shown that E. coli isolates from the community tend to have high overall resistance to ampicillin (75%), nalidixic acid (73%), and trimethoprim-sulfamethoxazole (59%). 17 In a 2019 study performed in rural South India, ∼3% of Enterobacteriaceae were resistant to nonertapenem carbapenems, whereas resistance to ertapenem was as high as 4 times that of other carbapenems. Reference Sekar, Srivani, Kalyanaraman and Thenmozhi18 In India, from 2008 to 2018, E. coli resistance to third-generation cephalosporins increased from 71% to 88% and fluoroquinolone resistance increased from 83% to 89%, whereas carbapenem resistance increased from 9% to 41%. 19,Reference Datta, Wattal, Goel, Oberoi, Raveendran and Prasad20 Similar to the resistance patterns cited from other studies, we detected a resistance rate >60% for cephalosporins and quinolones. Additionally, a relatively high rate of resistance (>25%) to carbapenems has also been documented.
In a prior study comparing AMR in rural and urban areas in South India, most isolates from patients in urban centers are gram-negative pathogens; E. coli and Klebsiella spp are the most common, and S. aureus comprises only 4% of specimens. By comparison, E. coli and S. aureus were the most common pathogens in rural areas (30%). More than 25% of all K. pneumoniae and Pseudomonas spp isolates showed resistance to imipenem, and a similar proportion of all E. coli, Klebsiella spp and Pseudomonas spp isolates were resistant to fourth-generation cephalosporins. Reference Balachandra, Sawant, Huilgol, Vithya, Kumar and Prasad16
In this study, S. aureus (including MRSA) was the most common pathogen causing wound infections, with ∼20% of all bacteria being multidrug resistant (ie, resistant to ≥2 antimicrobial agents). Coagulase-negative staphylococci constituted ∼36% of all wound and blood cultures, likely representing colonization rather than true infection. Gram-negative microorganisms, including E. coli and Pseudomonas spp, were also leading causes of wound infections as well as bloodstream infections, in which nearly 40% of isolates were resistant to >1 carbapenem. The large number of coagulase-negative staphylococci isolated from blood cultures were almost certainly a result of blood-culture contamination because they represent the most common blood-culture contaminants. Reference Hall and Lyman21 Studies have shown that the mortality rate is higher in patients who are infected by A. baumannii. In a previous study of ICU patients with bloodstream infections, the mortality rate was twice as high among those with gram-negative MDR infections by E. coli, K. pneumonia, and Acinetobacter compared to those infected with non–drug-resistant pathogens. Reference Gandra, Mojica and Klein22 In the current study, Acinetobacter spp constituted ∼10% of all bloodstream infections.
In the tertiary-care setting, risk factors associated with AMR include frequent use of invasive devices, prescription of empiric antimicrobials, critically ill patient population, high patient census, and opportunities for cross transmission during close contact between ICU personnel and their inpatient population. These AMR risk factors have long been characterized in myriad published studies of tertiary-care centers in economically developed countries. Reference Chen, Li and Beiersmann23–Reference Sotto, De Boever, Fabbro-Peray, Gouby, Sirot and Jourdan28 A combination of these risk factors with a failure to identify patients infected or colonized with AMR pathogens is the usual reason for increased resistance in the healthcare setting.
In rural areas, antimicrobial overprescribing represents one of the most significant contributors to AMR because patients in these areas are less likely to have in situ urinary catheters, intravascular catheters, or other invasive medical devices. Because patients in rural settings in India usually present to the primary care setting in the first instance; thus, it is highly probable that rural populations already harbor resistant strains of Staphylococcus spp, Enterobacteriaceae, or nonfermenters, such as Pseudomonas spp and Acinetobacter spp. This reality then leads to the following question: Is the emergence of AMR among strains of healthcare-associated pathogens seen in tertiary-care centers in India a consequence of prescribing practices in rural areas? Unfortunately, published data pertaining to this hypothesis are scarce. It remains unknown whether unique risk factors peculiar to the rural setting exist, such as use of antimicrobials as growth promoters in farming and other agricultural endeavors. Hence, there is a need for a paradigm change from focusing on tertiary-care hospitals to focusing on primary care settings.
This study had several limitations. Its design was descriptive; there was no collection of data regarding clinical or epidemiologic risks factors or patient outcomes. By intention, however, we pilot-tested a project to ascertain the presence of AMR in this tertiary-care setting. In the next phase of our project, we plan to perform a prospective, epidemiologic study that characterizes AMR in rural populations to understand attributable risk factors and clinical surrogate markers.
Most of rural India’s population visits government primary healthcare centers for their medical needs. Guidelines to treat common infections are based on recommendations that do not consider the local antibiograms. Rural primary healthcare centers lack the infrastructure to perform cultures and sensitivities. Furthermore, laboratory technicians often lack the expertise to perform these tests.
The NAP-AMR strategy prioritizes the strengthening of microbiology laboratories and establishing standards for AMR surveillance in humans, animals, food, and the environment. 7 Although the Indian Council of Medical Research has initiated a network to provide surveillance data at the national level, very little is known about AMR in rural communities and PHCs. Future studies should examine the contribution of prescription practices, healthcare-seeking behaviors, and antimicrobial use in agriculture to the locally high rates of antimicrobial resistance.
Antimicrobial resistance affects the socioeconomic growth, food security, environmental health, and long-term financial viability of communities. 29,30 Unless AMR is tackled at the level of the primary healthcare center, its severe social and economic effects will continue to push India’s citizens into poverty. This pilot study was the first phase of a planned long-term endeavor to characterize resistance in healthcare centers throughout the rural and urban communities of India.
In conclusion, the current study has demonstrated a high prevalence of resistance among clinical isolates obtained from urine, blood, and wound cultures at a tertiary-care hospital serving a large catchment of rural communities. Specifically, high resistance rates (>45%) to quinolones, penicillin, and cephalosporins were evident among all three types of cultures. The gradual replacement of what used to be considered firstline therapeutic agents by broad-spectrum agents, such as carbapenems and aminoglycosides, represents an ongoing challenge in India. This issue has led to poor antimicrobial stewardship, especially in primary healthcare settings where resources are scarce and antimicrobial overprescribing is less subject to established antimicrobial stewardship programs. Our findings revealed another alarming problem regarding these broad-spectrum antimicrobials that were once viewed as last resort agents. Among blood and urinary pathogens, resistance rates to both aminoglycosides and carbapenems were high (>25%). Thus, despite being considered last-resort agents, these agents might become increasingly ineffective in the treatment of resistant infections under current prescription practices. AMR is not a phenomenon restricted to tertiary-care centers, and it represents a significant challenge in rural settings. There is a relative lack of literature focusing on AMR in rural healthcare settings in India. This study highlights the need to conduct future studies investigating the contribution of prescription practices, healthcare-seeking behaviors, and antimicrobial use in agriculture to the alarming rates of AMR in rural settings.
Acknowledgments
Financial support
No financial support was provided relevant to this article.
Competing interest
All authors report no conflicts of interest relevant to this article.
Open access
