Detection of Shiga toxin-producing Escherichia coli (STEC) in complex sample matrices remains challenging. In an attempt to improve detection, nonselective and selective enrichment broths were compared as follows: (1) trypticase soy broth (TSB) was compared with TSB plus novobiocin, vancomycin, rifampicin, bile salts, and potassium tellurite (TSB-NVRBT) for supporting growth of STEC in pure culture; (2) E. coli broth (EC), TSB, and TSB plus bile salts (mTSB) were compared for enrichment of STEC O26, O45, O103, O104, O111, O121, O145, and O157 (STEC-8) in inoculated cattle fecal samples; (3) EC, TSB, and mTSB were compared for the detection of STEC-8 in inoculated cattle fecal samples. Fecal samples were inoculated with wild-type STEC-8 or nalidixic acid- or rifampicin-resistant derivatives of the same strains at 100, 1000, or 10,000 colony-forming units per gram (CFU/g) of feces. In pure culture, the mean STEC CFU/mL following enrichment in TSB was 1.17 log10 greater than that in TSB-NVRBT (P < 0.05). In inoculated fecal samples, EC enrichment yielded growth of STEC-8 (6.42 log10 CFU/g) that was significantly greater than in TSB (6.23 log10 CFU/g; P < 0.05), and numerically but not significantly greater than in mTSB (6.37 log10 CFU/g; P = 0.60). Wild-type STEC strains were detected in 43.8 % (21/48) of the samples enriched in EC and mTSB compared to 27.1 % (13/48) of the samples enriched in TSB (P = 0.15). Overall, STEC grew significantly better when enriched in EC compared to TSB. Modification of TSB by the addition of bile salts improved the growth and detection of STEC compared to TSB alone.