OBJECTIVE To compare complications between healthy horses undergoing general anesthesia for ophthalmic versus non-ophthalmic procedures and identify potential risk factors for the development of complications. DESIGN Retrospective case series. ANIMALS 502 horses (556 anesthetic procedures). PROCEDURES Medical records from January 2012 through December 2014 were reviewed to identify horses undergoing general anesthesia. Signalment, body weight, drugs administered, patient positioning, procedure type (ophthalmic, orthopedic, soft tissue, or diagnostic imaging), specific procedure, procedure time, anesthesia time, recovery time, recovery quality, and postoperative complications were recorded. RESULTS Patients underwent general anesthesia for ophthalmic (n = 106), orthopedic (246), soft tissue (84), diagnostic imaging (110), or combined (10) procedures. Mean procedure, anesthesia, and recovery times were significantly longer for patients undergoing ophthalmic versus non-ophthalmic procedures. Excluding diagnostic imaging procedures, there was a significant positive correlation between surgery time and recovery time. Within ophthalmic procedures, surgery time, anesthesia time, and recovery time were significantly greater for penetrating keratoplasty versus other ophthalmic procedures. There was a significantly higher rate of postoperative colic following penetrating keratoplasty, compared with all other ophthalmic procedures. CONCLUSIONS AND CLINICAL RELEVANCE Results suggested that in healthy horses, duration of general anesthesia should be minimized to decrease the risk of postanesthetic complications. Judicious use of orally administered fluconazole is recommended for horses undergoing general anesthesia. For horses undergoing a retrobulbar nerve block during general anesthesia, use of the lowest effective volume is suggested.