This study compared the reproductive efficiency of dairy cows submitted to fixed-time AI (FTAI) protocols using GnRH and/or different esters of estradiol. A total of 1,029 lactating Holstein cows (358 primiparous and 671 multiparous; BCS 2.99 ± 0.01, 159.8 ± 3.4 DIM; 36.4 ± 0.29 kg/d of milk; LS means ± SEM) were housed in freestall barns and randomly assigned to 1 of 4 treatment groups using a completely randomized design with a 2 × 2 factorial arrangements of treatments. Cows were assigned to 1 of 2 treatments at the beginning of the FTAI protocol (GnRH [G] or estradiol benzoate (EB]) and 1 of 2 treatments at the end of the protocol (EB or estradiol cypionate [ECP]), resulting in: G-EB (n = 263), G-ECP (n = 277), EB-EB (n = 259) and EB-ECP (n = 230). All cows received an intravaginal progesterone (P4) implant at the start of the protocol (d 0) and it was removed on d 8. All cows also received PGF_2α (500 µg) on d 7 and d 8. At the end of the protocol, treatment with ECP was on d 8 or treatment with EB was on d 9 with FTAI on d 10. Statistical analyses were performed using Proc GLIMMIX of SAS 9.3. There were no interactions between treatments on pregnancies per AI (P/AI) at 30 and 60 d pregnancy diagnoses (P > 0.05). At 30 d, there was no effect of EB or GnRH at start of protocol (33.9 vs. 37.0%; P = 0.30) and no effect of treatment at end of protocol (33.1 vs. 37.9%; EB vs. ECP; P = 0.11). At 60 d, there were no treatment effects at start (29.8 vs. 32.2%; EB vs. GnRH; P = 0.43) or end (30.3 vs. 31.7%; EB vs. ECP; P = 0.63) of the protocol. For pregnancy loss (d 30 to 60 of pregnancy), there were no interactions (P > 0.05) and no effect of treatments at start (P = 0.72), however, there was an effect of treatment at end (8.6 vs. 15.7%; EB vs. ECP; P = 0.04) of the protocol. Independent of treatments, DIM tended to affect P/AI at 30 d (P = 0.08) and at 60 d (34.8 vs. 27.4%, low vs. high DIM; P = 0.01), but did not affect pregnancy loss (P = 0.05). We concluded that in lactating Holstein cows, the use of EB or GnRH at the beginning or EB or ECP at the end of the protocol provided same P/AI at 30 and 60 d after FTAI, despite that ECP, at the end of protocol, provided greater embryonic/fetal loss. Also, cows with less DIM had greater P/AI at 60 d. Thanks to CNPq, FAPESP, CAPES and VALLÉE.