To test the hypothesis that the folli cle -stimulating hor-mone(FSH)threshold in patients with elevated FSH levels in the early follicular phase(EFP)is higher than in con-trols.Pilot study.Academic hospital.Six patients with elevated EFP FSH(>10IU /L)and 13controls.Treat-ment with a GnRH agonist in the midlut eal phase before IV administration of recombinant FSH w as started in an ul-tra -low -dose step -up protocol.The FSH threshold was determined by the mean of FSH levels o f the above threshold value and the below thresh old value.Follicle -stimulating hormone threshold,FSH screening value,E 2,number of follicles.The FSH threshold in the elevated EFP FSH group was 6.75IU /L and was significantly higher than the FSH threshold of the controls(4.65IU /L).The FSH screening value on day 3was 12.0I U /L in the pa-tient group and 5.0IU /L in the contro ls.Estradiolwas significantly lower on the day that t he largest follicle was 18mm in the elevated EFP FSH group compared with controls(277vs.491pmol /L,respectively)-.On the day of hCG administration,the number of sm aller(10-13mm)follicles was equal but the number of larger(>14mm)follicles was higher in the control g roup compared with the elevated FSH group.In the contro l group,the basal FSH levels correlated highly with th e FSH threshold levels(r =0.8),but in the patients with elevated EFP FSH this correlation was absent.In normal wo men,basal FSH day 3values represent the ovarian thresh old for FSH.In womenwith elevated day 3FSH,the FSH threshold is higher but not as high as basal FSH values.We postulate that the FSH threshold in patients with elevated EFP FSH is higher be-cause of intraovarian factors.Basa l FSH overshoots the threshold,probably because of the l imited feedback by the ovary. To test the hypothesis that the folli cle -stimulating hor-mone (FSH) threshold in patients with elevated FSH levels in the early follicular phase (EFP) is higher than in con-trols. Pilot study. Academic patients. Patients with elevated EFP FSH (> 10 IU / L) and 13controls.Treat-ment with a GnRH agonist in the midlut eal phase before IV administration of recombinant FSH w as started in an ul-tra -low-dose step-up protocol. FSH threshold was determined by the mean of FSH levels of the above threshold value and the below thresh old value. Follicle -stimulating hormone threshold, FSH screening value, E 2, number of follicles. FSH threshold in the elevated EFP FSH group was 6.75 IU / L and was significantly higher than the FSH threshold of the controls (4.65 IU / L). The FSH screening value on day 3 was 12.0 IU / L in the pa-tient group and 5.0 IU / L in the conls ls.Estradiol was significantly lower on the day that t he largest follicle was 18mm in the elevated EFP FSH group compared with controls (277 the number of sm aller (10-13mm) follicles was equal but the number of larger (> 14mm) follicles was higher in the control g roup compared with the elevated FSH group.In the contro l group, the basal FSH levels correlated highly with th e FSH threshold levels (r = 0.8), but in the patients with elevated EFP FSH this correlation was absent.In normal wo men, basal FSH day 3 values represent the ovarian thresh old for FSH. women with elevated day 3FSH, the FSH threshold is higher but not as high as basal FSH values. We postulate that the FSH threshold in patients with elevated EFP FSH is higher be-cause of intraovarian factors. l FSH overshoots the threshold, probably because of the l imited feedback by the ovary.