Background: the quality of the gametes used for an intracytoplasmic microinjection of spermatozoïde is a significant factor which can influence pregnancy rates.
Aim: To assess the effect of conventional sperm parameters, origin of spermatozoa and oocyte quality on pregnancy rate in ICSI.
Methods: A retrospective and comparative study of 500 women who underwent ICSI cycle during the study period from January 2004 to December 2005. Conventional sperm parameters (count, motility and morphology) and oocyte quality (mature and immatures oocytes) was compared in two groups of patients: Those achieving a pregnancy: The "pregnancy+" group and those failing to have a pregnancy: The "pregnancy-" group.
Results: Among the conventional sperm parameters, only spermatozoa count after preparation was significantly higher in "pregnancy+" group (p=0,02). We found significantly more pregnancies in ejaculated and epididymal sperm groups than in the testicular one (p<10-3). The number of oocyte retrieved was significantly higher in "pregnancy+" group (13,9±7 vs 10,6± 7,6 ; p<10-3) with mainly mature oocyte (metaphaseII) (9,1±5,5 vs 6,6±5,4 ; p<10-3). Analysis of ROC curve and logistic regression study show that for mature oocyte, the most discriminative cut-off for predicting pregnancy is 4 (OR=2,1 ; LR+=6,7 ;IC[1 ;1,4] ; p0,009).
Conclusion: Conventional parameters of ejaculated sperm have almost no influence on pregnancy rates in ICSI. Testicular sperm seem to have worse results. The number of oocytes retrieved and the proportion of mature oocytes (metaphaseII) affect markedly the prognosis.