A prospective multicenter study was carried out between 2017 and 2021 to assess (1) the appropriateness of the empirical treatment to the local guidelines of urinary source E. coli bacteremia, (2) the appropriateness of empirical treatment to antibiotic sensitivity results, and (3) the degree of error in the local guidelines regarding the antibiotic sensitivity reported in acute care hospitals enrolled in the VINCat program.During the study period, 79.0% of the empirical treatments analyzed complied with the guidelines and 88.1% were appropriate in view of the in vitro activity of the isolated strain. The rate of appropriateness rose from 73.8% in 2017 to 81.0% in 2021 (p<0.001). The degree of error in the recommendations regarding the in vitro activity of the isolated strains was 5.9% and remained stable during the study period. Antibiotic families correctly prescribed according to the guidelines were third-generation cephalosporins (54.9%), carbapenems (16.8%) and combinations of penicillins and beta-lactamase inhibitors (16.4%). Of the 8009 E. coli strains, 19.0% were ESBL producers, 36.8% were resistant to quinolones, and 0.5% were resistant to carbapenems.The broad implementation of an antimicrobial stewardship program with quality indicators of antibiotic use improved compliance to local guidelines in the empiric treatment of urinary tract E. coli bacteremia. The degree of error in local guidelines was low but higher in more complex hospitals and in health-care associated infections. Guidelines need to be constantly updated with the use of epidemiological data, rapid diagnostic tests and the analysis of patient risk factors specific to each geographical area.