Colonization With Antibiotic-Resistant Bacteria in a Hospital and Associated Communities in Guatemala: An Antibiotic Resistance in Communities and Hospitals (ARCH) Study

Abstract

Background We estimated the prevalence of colonization with extended-spectrum cephalosporin-resistant Enterobacterales (ESCrE) and carbapenem-resistant Enterobacterales (CRE) from a hospital and associated communities in western Guatemala. Methods Randomly selected infants, children, and adults (<1, 1–17, and ≥18 years, respectively) were enrolled from the hospital (n = 641) during the coronavirus disease 2019 (COVID-19) pandemic, March to September 2021. Community participants were enrolled using a 3-stage cluster design between November 2019 and March 2020 (phase 1, n = 381) and between July 2020 and May 2021 (phase 2, with COVID-19 pandemic restrictions, n = 538). Stool samples were streaked onto selective chromogenic agar, and a Vitek 2 instrument was used to verify ESCrE or CRE classification. Prevalence estimates were weighted to account for sampling design. Results The prevalence of colonization with ESCrE and CRE was higher among hospital patients compared to community participants (ESCrE: 67% vs 46%, P < .01; CRE: 37% vs 1%, P < .01). Hospital ESCrE colonization was higher for adults (72%) compared with children (65%) and infants (60%) (P < .05). Colonization was higher for adults (50%) than children (40%) in the community (P < .05). There was no difference in ESCrE colonization between phase 1 and 2 (45% and 47%, respectively, P > .05), although reported use of antibiotics among households declined (23% and 7%, respectively, P < .001). Conclusions While hospitals remain foci for ESCrE and CRE colonization, consistent with the need for infection control programs, community prevalence of ESCrE in this study was high, potentially adding to colonization pressure and transmission in healthcare settings. Better understanding of transmission dynamics and age-related factors is needed.

Risk factors for colonization with extended-spectrum cephalosporin-resistant and carbapenem-resistant Enterobacterales among hospitalized patients in Guatemala: An Antibiotic Resistance in Communities and Hospitals (ARCH) study

“…In contrast to CRE, colonization with ESCrE was not associated with any factors of a patient's current hospital stay beyond the expected positive association with ceftriaxone use and negative association with carbapenem use (ceftriaxone selects for ESCrE and carbapenems likely suppress ESCrE transmission or detection). Previous work with this same study population documented frequent use of broad-spectrum β-lactam antibiotics, including third-generation cephalosporins and carbapenems [22] . The strongest correlate of ESCrE colonization was hospitalization 30 days before study hospitalization but length of stay was not.…”

“…The prevalence of ESCrE and CRE colonization for the entire population was 72.3% and 34.6%, respectively. The distribution of ESCrE and CRE by age groups is published elsewhere [22] . The frequency of ceftriaxone and carbapenem administration among all patients was 22% and 11%, respectively.…”

“…One potential explanation for this correlation is that antibiotic use during the previous hospitalization led to a higher probability of ESCrE colonization (consistent with Table 2 ); however, our methods did not adequately document this previous hospital-based antibiotic selection pressure. The non-significant association between length of stay and ESCrE colonization may indicate that a high baseline saturation of community colonization (46% [22] ), especially among previously hospitalized patients, does not further increase the likelihood of ESCrE colonization through transmission events occurring as hospital length of stay increases; however, antibiotic use in the hospital (e.g. ceftriaxone) occurs with sufficient frequency to select for ESCrE colonization in this population, resulting in a higher overall colonization prevalence during hospital stays.…”

“…Detailed laboratory methods are described elsewhere [22] . Briefly, stool samples were collected in stool cups and transported at 4°C to the laboratory where swabs (FLOQSwabs) were used to transfer subsamples into eSwab liquid Amies medium, supplemented with sterile glycerol (20% v/v) and were transported with dry ice to a UVG central campus laboratory for processing.…”

“…Beyond antibiotic administration, less is known about other factors that could contribute to colonization before or during hospitalization. Earlier work showed that the prevalence of ESCrE and CRE colonization within a regional hospital in Guatemala (67% and 37%, respectively) is relatively high compared with other low- and middle-income countries [20] , [21] , [22] and very high compared with high-income countries [ 23 , 24 ]. This high prevalence offers an opportunity to detect potential contributions from the community to colonization in hospitalized patients.…”