Intracytoplasmic sperm injection (ICSI) is currently being used clinically in horses, but little information is available on factors affecting its success. We have conducted research trials and evaluated data from our clinical ICSI caseload, currently over 450 cases per year, to provide information in this area. In the following summaries, blastocyst data are reported per injected oocyte; clinical data are from 2013 and 2014. Clinically, when immature follicles were aspirated, the number of follicles aspirated per mare decreased significantly with mare age, from 16.2 at age 12–15 to 8.9 at age 24–25; however, maturation and blastocyst rates of recovered oocytes did not differ significantly. Immature oocytes shipped to the laboratory by referring veterinarians yielded a significantly higher blastocyst rate than that for immature oocytes aspirated at the laboratory (27 v. 21%, respectively). Shipped oocytes recovered from stimulated preovulatory follicles yielded a higher blastocyst rate per oocyte than did shipped immature oocytes (39 v. 27%, respectively), but provided fewer mature oocytes per aspiration (1.0 v. 4.5 for immature). From research data, administration of gentamicin and ampicillin to mares before immature oocyte aspiration did not affect blastocyst rate. Holding the aspirate for ~1.5 h at ambient temperature (26 to 33°C) was associated with a blastocyst rate of 32%; however, holding for 2 h at 32°C yielded only 16% blastocysts v. 23% for control. Blastocysts (18%) were obtained from oocytes recovered in the nonbreeding season (December and January). Holding oocytes at room temperature overnight before maturation did not affect blastocyst rate (25 to 34%), nor did inclusion of zinc in the maturation medium (18 to 31%). Diluting and refreezing semen to increase doses available for ICSI did not affect blastocyst rate (23 to 27%); blastocysts (13%) were also obtained after injection of nonmotile sperm. Significant differences in cleavage and blastocyst rates were identified among stallions. For one stallion, use of density gradient plus swim-up before ICSI increased cleavage (49 v. 18%) and blastocyst rates (11 v. 0%) compared to density gradient alone. Blastocyst production was not affected by the amount of glucose added to a human embryo culture medium (0 or 5 mM added glucose on Days 0–5, then 10 or 20 mM added glucose; 31 to 46% blastocysts). Replacement of 10% FBS in embryo culture medium with a mixture of FBS, human serum replacement, and equine follicular fluid lowered blastocyst rate (15 v. 37% for FBS alone). Clinically, embryos were vitrified or were shipped to embryo transfer centers for transfer. There was no significant difference in ongoing pregnancy rate for embryos from shipped immature v. preovulatory oocytes (54 and 69%). For immature oocytes, embryos developing to blastocyst on Day 10 or 11 had a lower ongoing pregnancy rate after transfer (40 and 0%) than did those developing on Days 7 to 9 (51 to 75%). This work was supported by the Link Equine Research Endowment Fund, Texas A&M University, and by the Clinical Equine ICSI Program, Texas A&M University.