P-637 Evaluation of a digitally integrated ovulation prediction kit with proof of ovulation – a pilot study

Abstract

Abstract Study question Is a digital ovulation prediction kit (OPK) with integrated progesterone measurements a useful tool in predicting and proving ovulation? Summary answer The digitally integrated OPK with progesterone integration may be a helpful tool for couples to identify the woman’s fertile window and detect anovulation. What is known already Apart from detecting the coming fertile window (i.e. the period of the highest probability of conception during the menstrual cycle, including the 5 days previous to and the day of ovulation itself), knowledge whether the likely ovulation indeed occurs, is crucial for achieving pregnancy. As women’s hormone profiles show intra- and interindividual variations, an individual prediction of the fertile window is a promising approach for couples, particularly in early premature ovarian insufficiency and perimenopause. Information whether the presumed ovulation has actually happened is not provided by most cycle apps. Study design, size, duration Healthy women, aged 21 to 44 years, who had not been taking hormonal contraceptives for at least three months were recruited via social media, paper postings and hand-outs. Women with infertility for more than 24 months were excluded. Up to three cycles were measured in every woman in case of a lacking pregnancy. Participants/materials, setting, methods Luteinizing hormone (LH) and Follicle-stimulating hormone (FSH) in first morning urine were gaged via lateral flow immunoassay test strips (LFIA) (RunBio Biotechnology Ltd.). Three algorithms of the smartphone application Pearl Fertility (iOS 1.0.3-1.0.5, Colorimetrix GmbH) were applied to predict the day of ovulation. Photographic visual control of the LH test strips, vaginal ultrasound close to the predicted ovulation and progesterone (PdG LFIA) were used to confirm ovulation, human chorionic gonadotropin (HCG LIFA) to confirm pregnancy. Main results and the role of chance 36 cycles of 19 women recruited for this pilot study were started. 6 participants dropped out or were excluded, leaving 25 evaluable cycles in 13 healthy women, average age 37 years (range 30 to 44), who had not been taking hormonal contraceptives for at least three months. 76 % of the analysable cycles had corresponding prediction of ovulation by algorithm and by visual control of the measured maximum value of LH (OVV). OVV corresponded with the day of ovulation determined retrospectively by length of luteal phase in 75 % (Bland-Altman plot (acceptable deviation Δ≤|2| days)). Prediction of ovulation differed by 2.12±3.13 days (mean±SD) between OVV and prospective cycle-length-based data (Wilcoxon signed-rank test (p < 0,05)). The PdG LIFA suggested ovulation in 84 % of the cycles with presumed ovulation by postovulatory PdG measurement. 6 of the participating women got pregnant during the study. Analysis of cumulative rate of pregnancy with 95 % confidence interval resulted in 25±21 %, 40±19 % and 40±17 % within the first, second and third cycle. Limitations, reasons for caution In order to make valid statements about the benefit of digitally integrated OPKs, further research with a larger population is needed. Perimenopausal women and patients with irregular cycles may specifically benefit from individual self-monitoring. Wider implications of the findings The knowledge of the fertile window is expected to help fertile couples to achieve pregnancy and proof of ovulation or anovulation may detect infertility due to anovulation sooner, thereby reducing the interval until diagnosis and treatment. Trial registration number 'not applicable'