Tf. Kruger et al., A PROSPECTIVE-STUDY ON THE PREDICTIVE VALUE OF NORMAL SPERM MORPHOLOGY AS EVALUATED BY COMPUTER (IVOS), Fertility and sterility, 66(2), 1996, pp. 285-291
Objective: To evaluate the IVOS (Hamilton Thorne Research Version 2.1
Dimension Program, Beverly, MA) system's ability to predict fertilizat
ion in vitro in a prospective study. Design: A prospective clinical st
udy. Setting: Hospital-based academic ART program. Patients: Eighty pa
tients from the IVF-GIFT program were evaluated. The same semen sample
was analyzed on a day-to-day basis by both laboratory (manual method)
and the computerized system for percentage normal morphology, concent
ration/mL, motility, and forward progression. Only patients with two o
r more metaphase II (MII) oocytes available were allowed into the stud
y and excluded where the male partner had antisperm antibodies or qual
ified for intracytoplasmic sperm injection (< 500,000 motile spermatoz
oa obtained after glass wool separation). Statistical Analysis: Logist
ic regression analysis was used to study predictors of fertilization i
n vitro. Results: Three hundred thirty-eight oocytes were obtained fro
m 80 patients of which 239 fertilized. The logistic regression analysi
s of the manual method (percentage normal morphology) and IVOS indicat
ed that both were predictors of fertilization. Sperm morphology as eva
luated by IVOS in patients with < 10 x 10(6) motile spermatozoa/mL ret
rieved after swim-up was a significant predictor of fertilization as w
as the number of oocytes obtained. Thus, the more oocytes obtained in
the lower morphological groups, the better the chance of fertilization
. The fertilization rate in the morphology group 0% to 4% normal forms
was 45.6% (37/81) and in the group > 14% normal forms was 85.2% (69/8
1). Conclusions: It was shown that in patients where less than or equa
l to 10 x 10(6) motile spermatozoa were obtained, the role of morpholo
gy (evaluated by IVOS) as well as the number of oocytes were important
predictors of fertilization. The computer can assist to identify thes
e patient with a poor prognosis for fertilization.