In previous studies, a new IVF method of intracytoplasmic morphologically selected sperm injection (IMSI) was introduced, based on motile sperm organellar morphology examination (MSOME). It was concluded that microinjection of morphologically selected sperm cells with strictly normal nucleus, defined by MSOME, improves IVF-ICSI outcome. The aim of the present study was to confirm this conclusion in new, enlarged study groups. Comparison between 80 couples, who underwent an IVF-IMSI trial, with matched couples, who underwent a standard IVF-ICSI procedure, confirmed that pregnancy rate following IVF-IMSI was significantly higher, and abortion rate significantly lower than in the routine IVF-ICSI (60.0 versus 25.0%, and 14 versus 40% respectively, P <or= 0.05). Another comparison was performed between matched IMSI groups, where embryos were obtained from microinjection by spermatozoa with a morphologically normal nucleus ('best' group, n = 70) and a 'second best' group was selected, where embryos were obtained from microinjection of spermatozoa with minimal morphological impairment, since no other sperm cells were available. It was confirmed that microinjection by 'second best ' spermatozoa result in significantly lower pregnancy and delivery rates and significantly higher abortion rates than microinjection with 'best' spermatozoa (25.7 versus 58.2%, P <or= 0.01; 17.1 versus 52.8%, P <or= 0.01, and 33.3 versus 9.7%, P <or= 0.02 respectively). The present study has strengthened previous conclusions.