The population of cumulus–oocyte complexes (COCs) used in OPU–IVP (ovum pick-up combined with in vitro production of embryos) is largely homogeneous due to repeated sessions resulting in the elimination of dominant and atretic follicles, especially when a 3–4 day interval is used. However, on average only 20% of the COCs develop to the blastocyst stage (Merton et al. 2003 Theriogenology 59, 651–674). Different blood flow changes within the follicle wall influence the fate of the follicles, and detectable blood flow and vasculature are associated with follicle viability. Furthermore, blood flow in follicles may be involved in not only selection of the dominant follicle but also early follicular development including follicular recruitment (Miyamoto et al. 2006 J. Reprod. Dev. 52, 153–160). However, no data are available regarding the quality of the COCs collected from follicles with or without blood flow. The purpose of this study was to determine whether qualitative perifollicular blood flow changes can be used to predict the developmental competence of COCs collected during repeated OPU sessions. Lactating Holstein cows were used as oocyte donors. After dominant follicle removal, OPU was performed twice weekly employing a 7.5-MHz transducer (GE 8C-RS) of an ultrasound scanner (GE Logiq Book). Follicle size and Doppler characteristics (color flow imaging) were recorded by transvaginal ultrasonography just before COC collection. Due to technical limitations for measurement of blood flow in small individual follicles, only the presence or absence of blood flow was assessed for each follicle. When a clearly visible blue or red spot (blood flow) was detected in the follicle wall, it was considered as a follicle with detectable blood flow. Follicles with or without detectable blood flow from each individual cow were aspirated separately. After morphological classification of COCs, standard protocols for IVP were used for embryo production. Cleavage and blastocyst rates were recorded at Day 3 and Day 8, respectively. In total, 464 (246 with and 218 without detectable blood flow) follicles e3 mm were aspirated. The percentage of follicles with detectable blood flow increased depending on follicle size (3 mm: 28.7, 4 mm: 48.4, 5 mm: 50.5, 6 mm: 62.5, 7 mm: 64.8, 8 mm: 71.4, and 9 mm: 76.9). Cleavage rates for COCs stemming from follicles with or without detectable blood flow did not show differences, 45.5% (35/77) and 56.7% (38/67), respectively. The rates of blastocyst formation were also similar in COCs originating from follicles with and without detectable blood flow, 16.9% (13/77) and 14.9% (10/67), respectively. These results show that perifollicular blood flow increases during early follicular growth. Within the detection limits of this study, differences in perifollicular blood flow during repeated OPU sessions twice weekly did not seem to be predictive of oocyte competence. We acknowledge the Ruthe Research Farm, Germany, for providing the animals, and Masterrind GmbH, Germany, for donating the semen.