The aim of this work was to evaluate the efficiency of IVF and intracytoplasmic sperm injection (ICSI) when few eggs available for insemination. A total of 601 women (group A, mean age 31.2 +/- 2.8 years) who were undergoing a total of 671 assisted reproduction cycles donated their excess oocytes to 694 patients (group B, mean age 41.0 +/- 0.2) for 1606 replacement cycles. Each recipient received three to five eggs. The recipients were divided into two groups depending on the insemination method used (IVF, group B1; or ICSI, group B2); ICSI patients were then subdivided into two further groups based on the semen parameters: B2A adequate for IVF and B2B only suitable for ICSI. The results showed that, when comparing A versus B and B1 versus B2, no significant differences were found in terms of pregnancy (28.0 versus 24.1% and 25.5 versus 21.4%), implantation (15.6 versus 14.9% and 15.9 versus 13.1%) and miscarriage (15.4 versus 20.5% and 17.9 versus 26.3) rates respectively. Comparing subgroups B2A and B2B, no significant differences were found in terms of pregnancy (20.0 versus 21.9%), implantation (14.4 versus 12.7%) and miscarriage rates (18.2 versus 28.6%) respectively. In conclusion, ICSI does not seem to yield better outcomes.