Intracytoplasmic sperm injection (ICSI) is a relatively new treatment for human male-related infertility (1992) and for the production of transgenic animals (1995). However, ICSI bypasses many natural biological processes such as sperm maturation, interaction within the female genital tract, sperm capacitation, interaction with oocyte vestments, and sperm membrane fusion with the oocyte. With the widespread use of this technology, its potential adverse outcomes need to be ascertained. It is theoretically possible that ICSI may cause specific problems through injury to the sperm or egg or injection of damaged or defective sperm. Here, we determined if ICSI has a long-term effect on mouse growth, behavior, and health. Female CD1 mice were superovulated and oocytes were injected with frozen-thawed spermatozoa (without cryoprotector or chelating agent) obtained from CD1 mice epididymes (Moreira et al. 2004 Biol. Reprod. 71, in press). Embryos were cultured 24 h in KSOM, and 2-cell embryos were transferred into CD1 females. Fifty-six mice (36 males and 20 females) produced by ICSI and 41 control mice (18 males and 23 females) obtained from in vivo-fertilized mice were analyzed. On week 20, animals were submitted to the following behavior tests: locomotor activity (open field), exploratory/anxiety behavior (elevated plus maze, open field), and spatial memory (free-choice exploration paradigm in Y maze). Comparison between groups was made using analysis of variance followed by least significant difference post hoc test. Postnatal weight gain of female mice produced by ICSI was heavier than for their control counterparts from 10 weeks on (P < 0.01). Males produced by ICSI showed more anxiety and lower locomotion in the p-maze and the Y-maze tests (P < 0.05), but no significant differences were found in the open-field test. Also, no differences were found in spatial memory or in the habituation pattern. Anatomopathological analysis of animals at 16 months of age showed some large organs (heart, lung, and liver; P < 0.01) and an increase in pathologies (15% of animals produced by ICSI presented some solid tumors in lung, dermis of back, or neck). Loss of imprinting is one of the most common epigenetic changes associated with the development of a wide variety of tumours. An association between some imprinting disorders, rare tumors, and ICSI has recently been reported in humans (Wittermer et al. 2004 Med. Sci. 20, 352). We are now analyzing the epigenetic modifications that may be induced by our ICSI protocol and whether the sperm DNA fragmentation that may take place during sperm freezing before the ICSI procedure might not only affect postnatal development, growth, and physiology, but also increase the risk of tumors in adult animals. Our data suggest that our ICSI method produces mice with sex-dimorphic alterations in aberrant growth and anxiety, as well as with a higher probability of developing a solid tumor.