Background: To investigate whether cumulus cell removal after a 3 h co-incubation of gametes can affect the outcomes of in vitro fertilization. Methods: A prospective randomized sibling-oocyte study was performed in which sibling oocytes obtained from each patient were randomly assigned to either a 3 h or 20 h group (cumulus cells removed at 3 h or 20 h after insemination, respectively). Same origin embryos (either 3 h or 20 h) were transferred. The study participants were patients < 38 years old and with infertility due to tubal factors. The study outcomes included fertilization, embryo quality, and birth status. Results: The study enrolled 172 patients, from whom 2275 oocytes were retrieved (1139 oocytes for the 3 h group and 1136 oocytes for the 20 h group). A total of 134 patients received embryo transfers (74 patients in the 3 h group and 60 patients in the 20 h group), and there were 54 live births (32 in the 3 h group and 22 in the 20 h group). When compared with patients in the 20 h group, patients in the 3 h group produced a larger number of optimal quality embryos, but had higher rates of polyspermy and low birth weight newborns. The two groups showed no differences in their rates of normal fertilization, pregnancy, and live birth. Conclusions: When compared with results obtained using a traditional cumulus cell removal protocol, early cumulus cell removal has both advantages and disadvantages. Further studies, and especially long-term outcome studies on newborns, need to be performed.