High success rates have been reported for the use of intracytoplasmic sperm injection (ICSI) in alleviating essentially andrological infertility. However, neither the relationship between any of the sperm parameters and the result of ICSI nor the minimal sperm requirements for ICSI have been investigated so far. In this paper, our objective was therefore to study the relationship between three basic sperm parameters (total sperm count, sperm motility and morphology) and the outcome of ICSI by retrospective analyses of fertilization, embryo development and pregnancy rates in 966 micro-injection cycles, performed with ejaculated semen. The results showed that there was no important influence from either the type or the extent of sperm impairment on the outcome of ICSI. Even in the most extreme cases of male-factor infertility, where cryptozoospermia or total astheno- or total teratozoospermia was diagnosed in the initial semen sample, high fertilization and pregnancy rates were obtained by ICSI. Only one condition had a strongly negative influence on the result of ICSI: where an immotile (presumably dead) spermatozoon was injected into the oocyte. Thus the only ultimate criterion for successful ICSI is the presence of at least one living spermatozoon per oocyte in the pellet of the treated semen sample used for micro-injection.