Among Enterobacteriaceae, the production of beta-lactamases (ESBLs) is the leading cause of multi-resistance. The first cases of ESBL-producing Enterobacteriaceae (E-ESBL) infections were described in the 1980s and subsequently experienced worldwide dissemination.

Since the turn of the century, the prevalence of E-ESBL infections, especially among Escherichia coli (E. coli) and Klebsiella pneumoniae (K. pneumoniae) has increased dramatically.

The emergence of multidrug-resistant Enterobacteriaceae is currently a real public health problem. The European Antimicrobial Resistance Surveillance Network evaluated, among clinical strains, the rate of resistance to 3rd generation cephalosporins (C3G) at 9.5% for E. coli and 28% for K. pneumoniae. Numerous studies have shown that bacterial colonization is the prerequisite for the occurrence of many infections.

However, the existence of prior colonization does not seem to be the only risk factor for the occurrence of a secondary infection. Therefore, in patients with multidrug-resistant Gram-negative bacillus gastrointestinal carriage there appear to be factors associated with the onset of infection. Several studies have examined the risk factors associated with E-ESBL-related infections in both community-based and healthcare-associated / nosocomial infections. Two main risk factors seem to be associated with E-ESBL infections: prior antibiotic therapy and the existence of invasive devices.

A recent study, carried out on 1288 patients and aimed at validating a predictive score for the occurrence of ESBL-E bacteremia, demonstrated 5 factors associated with the appearance of E-ESBL-linked bacteremia. These factors were: (i) a history of colonization / infection with ESBL-E, (ii) age ≥ 43 years, (iii) recent hospitalization in a region with a high prevalence of ESBL-E, (iv) antibiotic therapy ≥ 6 days in the previous 6 months and (v) the existence of a chronic vascular access.

Recently, a retrospective case-control study conducted in the United States by Augustine et al. Suggested that 5% of cases of bacteremia were related to ESBL-E.

Few studies have looked at risk factors for infection in patients known to be colonized by the digestive system. In a retrospective case-control study, conducted outside the intensive care unit and including pediatric and adult patients, the authors identified 2 factors associated with the occurrence of Ec-ESBL infection in previously colonized patients. These two factors were the prior use of antibiotics with β-lactam antibiotics and β-lactamase inhibitor (s), and urinary catheterization.

In intensive care hospital patients, the occurrence of ESBL-producing enterobacteriaceae infection appears to be a rare event, including in colonized patients.

The work of Ruppé et al. showed a direct link between relative fecal abundance of EScher-producing Escherichia coli and prior antibiotic intake.

This work also demonstrated a link between the value of the relative fecal abundance in Ec-ESBL and the occurrence of a urinary tract infection linked to the same clone. In particular, the authors found that women with a low relative fecal abundance rate (≤ 0.1%) had no risk of developing an Escherichia coli urinary tract infection. Conversely, the risk increased with the relative fecal abundance of Escherichia coli, but with a positive predictive value limited to 57% for relative fecal abundances between 10 and 100%.

Inclusion Criteria:

• Adult patient (≥ 18 years old) hospitalized at the Paris Saint-Joseph Hospital Group or in the intensive care unit of Avicenne hospital, Necker Enfants Malades hospital, Center Sud Francilien, detected as a carrier of enterobacteriaceae in the digestive system ESBL producers
• Patient affiliated to a health insurance plan
• French-speaking patient
• Patient living at home, in EHPAD or retirement home
• Patient or Relative able to give free, informed and express consent

Exclusion Criteria:

• Known patient colonized rectally with ESBL-producing enterobacteria and subjected to antibiotic pressure other than beta-lactams
• Patient participating simultaneously in other intervention research that may interfere with the objectives of the study
• Patient under guardianship or curatorship
• Patient deprived of liberty
• Pregnant or breastfeeding woman