Sperm selection for ICSI: Shape properties do not predict the absence or presence of numerical chromosomal aberrations

Document Type

Journal Article

Publication Date



Human Reproduction








FISH; ICSI; Maturity; Morphology; Morphometry


Background: We hypothesize that the potential relationship between abnormal sperm morphology and increased frequency of numerical chromosomal aberrations is based on two attributes of diminished sperm maturity: (i) cytoplasmic retention and consequential sperm shape abnormalities; and (ii) meiotic errors caused by low levels of the HspA2 chaperone, a component of the synaptonemal complex. Because sperm morphology and aneuploidies were assessed in semen, but not in the same spermatozoa, previous studies addressing this relationship were inconclusive. We recently demonstrated that sperm shape is preserved following fluorescence in situ hybridization (FISH). Thus, we examined the shape and chromosomal aberrations in the same sperm. Methods: We performed phase contrast microscopy and FISH, using centromeric probes for chromosomes X, Y, 10, 11 and 17 in 15 men. The fluorescence and respective phase contrast images were digitized using the Metamorph program. We studied 1286 sperm (256 disomic, 130 diploid and 900 haploid sperm) by three criteria: head and tail dimensions, head shape and Kruger strict morphology. Furthermore, in each analysis, we considered whether disomic or diploid sperm may be distinguished from haploid sperm. Results: There was an overall, but not discriminative, relationship between abnormal sperm dimensions or shape and increased frequencies of numerical chromosomal aberrations. However, ∼68 of the 256 disomic, and four of 130 diploid sperm showed head and tail dimensions comparable with the most normal, lowest tertile of the 900 haploid spermatozoa. Considering all 1286 sperm, among those with the most regular, symmetrical shape (n = 367), there were 63 and five with disomic and diploid nuclei, respectively. In line with these findings, among the 256 disomic sperm, 10% were Kruger normal. Conclusions: Sperm dimensions or shape are not reliable attributes in selection of haploid sperm for ICSI. © European Society of Human Reproduction and Embryology 2004; all rights reserved.