Recent studies suggest an association between enteric colonization with extended-spectrum β-lactamase–producing Enterobacteriaceae (ESBL-pE) and the occurrence of deep surgical-site infections (SSIs) among abdominal surgery patients, particularly colorectal surgery patients.^{Reference Dubinsky-Pertzov, Temkin and Harbarth1,Reference Apisarnthanarak, Kondo and Mingmalairak2} Screening for ESBL-pE and personalized ertapenem use was associated with reduction in the incidence of SSIs caused by ESBL-pE.^{Reference Nutman, Temkin and Harbarth3} Despite the potential benefit, screening all patients for ESBL-pE carriage may be considered an excess cost in resource-limited settings. Selective screening for high-risk groups with ESBL-pE carriage may be an alternative approach to surveillance screening and antimicrobial stewardship. This cohort study was conducted to evaluate the risk factors for ESBL-pE carriage in abdominal surgery patients at a tertiary-care center in Thailand.

A cohort study was performed at Thammasat University Hospital (TUH), a 750-bed, tertiary-care hospital in Pratum Thani, Thailand, over a 26-month period from February 1, 2017, to April 1, 2019. At this hospital, ~200 patients undergo abdominal surgery annually, and during the study period, SSIs occurred in 14%. Of these SSIs, 80% were superficial SSIs and 20% were deep or organ-space SSIs.^{Reference Apisarnthanarak, Kondo and Mingmalairak2} Notably, ESBL-pE SSIs were detected in ~15.6% of these patients.^{Reference Apisarnthanarak, Kondo and Mingmalairak2} All patients who underwent clean-contaminated, contaminated, or dirty abdominal surgical procedures were enrolled and screened for ESBL-pE colonization by rectal swab culture within 1 day prior to surgery. Data collection included baseline demographic and clinical characteristics, underlying comorbidities, presence of multiple comorbidities (defined as ≥3 comorbidities), previous hospitalization(s), type of surgical procedure, history and type of antibiotic exposure within 3 months prior to surgery, prior history of ESBL colonization ≥3 months prior to surgery, as well as American Society of Anesthesiology (ASA) risk class. Patients were excluded if they had documented infection(s) at the time of surgery. The institutional review board at TUH approved this study.

Enteric surveillance swabs (Becton Dickinson Diagnostics, Sparks, MD) were obtained from each patient within 1 day prior to surgery. Each rectal swab was processed in real time, inoculated into tryptic soy broth containing a 30-μg ceftriaxone disk (bioMérieux, Marcy-l’Étoile, France), and incubated at 37°C. Within 48 hours of inoculation, 100-μL broth samples with visible turbidity were plated on sheep blood agar and MacConkey agar with a 30-μg ceftriaxone disk and incubated at 37°C overnight. All recovered isolates within the zone of inhibition underwent routine identification using the Vitek 2 automated system (bioMérieux). Antimicrobial susceptibility testing was performed using the Clinical and Laboratory Standards Institute Interpretive Guidelines (disk diffusion test, bioMérieux), and confirmed by ESBL double-disk synergy testing for Enterobacteriaceae.^{Reference Boo, Ng and Lim4}

All analyses were performed using SPSS version 19 software (IBM, Armonk, NY). The χ² or the Fisher exact test was used to compare categorical variables. Mann-Whitney U tests were used for continuous data. All P values were 2-tailed, and a P value < .05 was considered statistically significant. To determine factors associated with ESBL-pEnterobacteriaceae carriage, variables that had a significant level of P < .20 or variable with a priori significance in univariate analysis were entered into multivariate logistic regression models. Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were calculated.

Among 360 patients who underwent abdominal surgery, 234 (65%) patients underwent clean-contaminated surgical procedures, 90 (25%) underwent contaminated surgical procedures, and 36 (10%) underwent dirty surgical procedures. At baseline, 129 patients (35.8%) had preoperative enteric colonization with ESBL-pE; baseline demographic characteristics were similar between the patients with and without ESBL-pE (Table 1). By multivariate analysis, preoperative enteric colonization with ESBL-pE was associated with antibiotic exposure within 3 months of surgery (adjusted odd ratios [aOR], 5.4; 95% CI, 1.4–24.5), multiple comorbidities (aOR, 4.6; 95% CI, 1.24–44.2), and dirty surgical wound classification (aOR, 3.69; 95% CI, 1.19–54.9).

Table 1. Demographic, Clinical Characteristics Among 360 Patients Screened for Preoperative Enteric Colonization With Extended-Spectrum β-Lactamase (ESBL)-Producing Enterobacteriaceae

Note. GFR, glomerular filtration rate; ASA, American Society of Anesthesiology.

^a History on steroid or immunosuppressive agents more than 2 weeks before surgery, underlying human immunodeficiency virus.

^b ≥3 comorbidities; clean-contaminated included abdominal procedures that include hepatobiliary tract with contamination; contaminated surgery included abdominal procedures that include colorectal and hepatobiliary tract surgery with evidence of contamination; and dirty surgery included any abdominal surgery with evidence of massive contamination (eg, perforated colonic obstruction).

In this study cohort, 35.8% of abdominal surgery patients had ESBL-pE carriage. This high prevalence may reflect reported ESBL-pE carriage in community and healthcare-associated settings among Thai populations as well as the report of injudicious use of antibiotics in community settings in this region.^{Reference Apisarnthanarak, Kiratisin, Saifon, Kitphati, Dejsirilert and Mundy5-Reference Thamlikitkul, Tangkoskul and Seenama9} The prior exposure to antibiotics, within 3 months prior to surgery, as a risk factor for ESBL-pE carriage is consistent with earlier studies reporting the risk of ESBL-pE in community and healthcare settings.^{Reference Apisarnthanarak, Kiratisin, Saifon, Kitphati, Dejsirilert and Mundy5,Reference Apisarnthanarak, Kiratisin and Mundy6} Together, multiple comorbidities, recent antibiotic use, and dirty wound classification are key factors that surgeons should consider as risks for deep SSIs by ESBL-pE. The presence of these key factors may identify patients for enteric screening for ESBL-pE carriage or for carbapenem empiric antibiotic prophylaxis prior to abdominal surgical procedures to reduce risk the of deep SSI.

This study has several limitations. It was performed at a single center with endemic ESBL-pE. It is plausible that the study findings are not generalizable to other regions where ESBL-pE is nonendemic. Second, given the small sample size, we were not able to identify other potential risk factors or specific antibiotic relationships. Lastly, the enteric colonization of ESBL-pE was performed using rectal swab culture, which has a <80% sensitivity.^{Reference Jazmati, Jazmati and Hamprecht10} Hence, misclassification bias may have occurred with underestimation of the prevalence ESBL-producing Enterobacteriaceae carriage. Despite these limitations, we identified key factors associated with rectal carriage of ESBL-pE that can help inform selective screening for ESBL-pE. Additional studies to evaluate selective screening for ESBL-pE carriage, together with personalized antibiotic prophylaxis among at risk abdominal surgeries, and studies to evaluate the cost effectiveness of these interventions in ESBL-pE prevalence regions are warranted.