One of the major sources of success in embryo transfer is timing of AI relative to the LH surge and ovulation. The aim of this study was to compare the embryo production following superovulation during a PGF2alpha (control cycle) or a CIDR-B synchronized cycle (CIDR-B cycle). CIDR-B (CIDR-B ND, Virbac, Carros, France) was inserted on Day 11 of a previously synchronized cycle and left for 5 days. A total dose of 350 microg FSH was administered (eight injections i.m. for 4 days; first on Day 13, decreasing doses) and PGFalpha analog (750 microg i.m.: Uniandine ND, Schering-Plough, Levallois-Perret, France) injected at the time of third FSH injection. Artificial inseminations were performed 12 and 24 h after standing estrus (Day 0). Embryos were collected on Day 7. Luteinizing hormone was measured by EIA (Reprokit Sanofi, Libourne, France) from blood samples collected every 3 h for 36 h, starting 24 h after PGF2alpha (control cycle) or 12 h after CIDR-B removal (CIDR-B cycle). The effects of treatment group and interval between the LH peak and AI (two classes, < 10 and > or = 10 h) on embryo production and quality were analyzed by ANOVA. No effect of treatment was observed on embryo production variables. The intervals between the end of treatment and onset of estrus and between end of treatment and LH surge were greater in heifers treated during a control than a CIDR-B cycle, respectively (45.5 +/- 1.4 versus 31.9 +/- 0.7; 42.0 +/- 1.6 versus 31.0 +/- 1.5; P < 0.05), but maximal LH and estradiol concentrations, at the preovulatory surge were similar in control and CIDR-B synchronized heifers. The numbers of viable and Grade I embryos were significantly increased (P < 0.01) when animals had an interval from LH peak to first AI > or = 10 h (7.2 +/- 0.9 and 3.5 +/- 0.6) when compared to shorter intervals (4.2 +/- 1.1 and 2.0 +/- 0.7) whereas total number of embryos was unchanged (11.8 +/- 1.4 versus 10.3 +/- 1.8). It is concluded that late occurrence of LH peaks in relation to estrous behavior is associated with a lower embryo quality when first AIs are performed systematically 12 h after standing estrus. Further studies are needed to know if results may be improved when making AI at a later time after standing estrus or if LH assays are useful to better monitor AI time.