The length of the fertile window is associated with the chance of spontaneously conceiving an ongoing pregnancy in subfertile couples

Human Reproduction Vol.22, No.6 pp. 1652 1656, 2007 Advance Access publication on April 20, 2007 doi:10.1093/humrep/dem051 The length of the fertile window is associated with the chance of spontaneously conceiving an ongoing pregnancy in subfertile couples M.J.Keulers 1,4, C.J.C.M.Hamilton 1, A.Franx 2, J.L.H.Evers 3 and R.S.G.M.Bots 2 1 Department of Obstetrics and Gynaecology, Jeroen Bosch Hospital, s-hertogenbosch, The Netherlands; 2 Department of Obstetrics and Gynaecology, St Elisabeth Hospital, Tilburg, The Netherlands; 3 Department of Obstetrics and Gynaecology, Research institute GROW, Academic Hospital Maastricht and Maastricht University, The Netherlands 4 Correspondence address. Department of Obstetrics and Gynaecology, Jeroen Bosch Hospital, PO Box 90153, 5211 NL s-hertogenbosch, The Netherlands. Tel: þ0031-0652067998; Fax: þ0031-0736998650; E-mail: m.keulers@haio.umcn.nl BACKGROUND: The period in each menstrual cycle during which sexual intercourse can result in conception is called the fertile window. Although the fertile window closes on the day of ovulation, little is known about the moment it opens. We defined the first day of normal sperm mucus interaction as the opening of the fertile window. We hypothesized that length of the fertile window varies between couples and that the number of days the fertile window is open is related to the time to spontaneous conception. METHODS: Serial post-coital tests and sperm mucus penetration tests were performed to detect the first normal sperm mucus interaction day. Ovulation was confirmed by serial ultrasound. Using Cox regression analysis, we determined whether the fertile window length was associated with time to ongoing pregnancy. This association was expressed in fecundability ratios (FR). RESULTS: The fertile window length was determined in 410 subfertile couples. The fertile window length varied among couples from <1 to >5 days. The FR increased with increasing fertile window length and varied between 0.11 (95% CI: 0.03 0.45) for a fertile window of 1 day, to 2.4 (95% CI: 1.1 5.2) for a fertile window of 5 days or more. CONCLUSIONS: The longer the fertile window in subfertile couples, the higher is the probability of spontaneously conceiving an ongoing pregnancy. Keywords: fertile window; sperm mucus interaction; fertility Introduction Sexual intercourse can only result in conception if it occurs on or during the 5 days preceding ovulation (Wilcox et al., 1995, 2000; Dunson et al., 2001; Bigelow et al., 2004). Although this fertile window, by definition, closes on the day of ovulation (Wilcox et al., 1995), precise estimates of the variation in length of the fertile window are lacking. Conception is unlikely if cervical mucus quality is poor, since spermatozoa will not penetrate it (Dunson et al., 2001; Francavilla et al., 2002; Stanford et al., 2003). Since cervical mucus quality improves towards ovulation (Billings et al., 1972), the first day of normal sperm mucus interaction determines the opening and the length of the fertile window. We hypothesized that the fertile window length, defined as the interval from the first day of normal sperm mucus interaction to ovulation, varies among couples and that the length of the fertile window is related to the occurrence of spontaneous ongoing pregnancy. Patients and Methods Subfertile couples referred between 1998 and 2003 for a basic fertility work-up in our fertility centre in Tilburg, The Netherlands, were studied prospectively. The fertility work-up was performed according to a strict protocol and consisted of cycle monitoring by ultrasound, ultrasonographic examination of the uterus and ovaries, semen analysis, serial assessment of sperm mucus interaction and diagnostic laparoscopy with dye testing. This study was approved by the Institutional Review Board. Couples with a regular menstrual cycle with an average length between 25 and 35 days were included. Couples with anovulation, or in whom tubal pathology (one or both sided tubal obstruction), moderate or severe endometriosis (mafs ¼ modified American Fertility Society (mafs) scoring system, which is a Method of assessment of the incidence, extent and severity of adhesions from laparoscopic video recordings stage III/IV), uterine abnormalities, or severe male subfertility (a total count of,1 10 6 morphologically normal and motile spermatozoa) was diagnosed, were excluded from the study. Ultrasound ovulation detection and cervical mucus sampling were started at least 5 days before ovulation, as estimated by menstrual history. For example, if the shortest cycle length was 28 days, the first mucus sampling started at day 9. Ultrasound scanning and cervical mucus aspiration were performed at each visit, at 1 to 2 day intervals until follicle rupture was confirmed by ultrasound. Sperm mucus interaction was assessed by the post-coital test (PCT). The PCT was performed within 48 h following sexual intercourse. In case of an 1652 # The Author 2007. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Fertile window length and probability of spontaneous pregnancy abnormal PCT result or if couples did not have intercourse in this period, the sperm mucus interaction was assessed by the sperm mucus penetration test (SMPT). The SMPT is an accepted in vitro sperm mucus interaction test (Kremer and Kroeks, 1975; WHO laboratory manual, 1992). To perform the SMPTs, all mucus samples were stored at 78C until detection of ovulation. Then, a new semen sample was obtained, to perform SMPTs on all mucus samples simultaneously. The SMPT was performed in capillary tubes and the (homologous) sperm mucus interaction was assessed after 2 h at 4 cm distance of the sperm reservoir (Kremer and Kroeks, 1975). Normal sperm mucus interaction was defined as the presence of at least one progressively motile spermatozoon per high power microscope field in either the PCT or SMPT. In several studies it has been demonstrated that normal sperm mucus interaction by this definition is related to the likelihood of spontaneous conception (Oei et al., 1995; Glazener et al., 2000). The first day of the fertile window was defined as the day of sexual intercourse resulting in the first normal PCT, or the day of the first normal SMPT result. Closure of the fertile window was defined as the day of ovulation (Wilcox et al., 1995). The day of ovulation was detected by serial ultrasound (Vermesh et al., 1987; Guermandi et al., 2001). Couples in whom both the PCT and SMPT remained abnormal until ovulation, were considered to have a fertile window length of one day or less. Couples were not monitored in weekends. As a consequence, we had difficulty in determining the fertile window length accurately when the first day of normal sperm mucus interaction or ovulation was detected on a Monday. We could not rule out that the fertile window might have started or ended already on the preceding Saturday or Sunday. In these couples, we were only able to determine the minimum fertile window length. We evaluated the minimum fertile window length in these couples separately from the couples in whom we did observe the complete fertile window. Couples were monitored for the occurrence of a spontaneous ongoing pregnancy (defined as a viable fetus at first trimester ultrasound) within one year of initial assessment. The fertile window length was determined once in each couple, in the first cycle of their fertility work-up. Time to pregnancy, i.e. the time interval from the first day of the first cycle of the fertility work-up, to the first day of the last menstrual period preceding an ongoing pregnancy, was the primary endpoint of the study. Follow-up ended when fertility treatment started, a spontaneous conception occurred, or at one year after the initial assessment. The fertile window length was related to the time interval to the conception of an ongoing pregnancy using Cox Regression analysis and Kaplan Meier analysis. A Kaplan Meier curve was constructed separately for those couples with a complete fertile window, and for those couples in whom only the minimum fertile window could be determined. The primary outcome was fecundability ratio (FR), as a measure for the relationship of the fertile window length with the time to ongoing pregnancy. The FR is equivalent to the hazard ratio, and is a relative rate that takes time to event into account. It describes the effect of an explanatory variable, in this case the fertile window length, on the chance of ongoing pregnancy. To define the FR for different lengths of the fertile window, the complete fertile window length most frequently observed was used as reference group. For this fertile window length, the FR was set at 1.0. We defined separate FR s for couples in whom the complete fertile window length and couples in whom the minimum fertile window length had been determined. Both the complete and minimum fertile window length group were analysed separately in the Cox regression model. To determine if differences in clinical characteristics were confounding the FR s of the fertile window length, we performed a multivariable Cox regression analysis. We included the following clinical characteristics in the model: female age; duration and type of subfertility; and sperm volume, concentration, motility and morphology. Results A total of 410 couples were eligible for the study. Baseline characteristics are shown in Table 1, and did not differ between the group with a complete fertile window length and the group with a minimum fertile window length (data not shown). In all couples, sperm mucus assessment started at least 5 days before the day on which ovulation was confirmed. In 83.0% of couples with a complete fertile window length, sperm mucus assessment and ultrasound were performed on a daily basis. The complete fertile window could be determined in 212 couples, whereas in the remaining 198 couples the first day of the fertile window (n ¼ 58) or the day of ovulation (n ¼ 140) was detected on a Monday. In these 198 couples the minimum fertile window length was related to occurrence of pregnancy. In 165 couples follow-up ended before the 12-month evaluation period due to fertility treatment (i.e. active censoring). Their cycles until the moment of censoring were included in the final analysis. Twenty-one couples were lost to follow-up. Within 12 months of initial assessment, a spontaneous ongoing pregnancy was observed in 116 couples (28.3%). The average conception rate per month was 4.3%. The length of the fertile window in the complete fertile window group varied between 1 day to.5 days (Fig. 1). The Kaplan Meier analysis showed a very strong relationship of the complete fertile window length with the occurence of an ongoing pregnancy within the first year after initial assessment (Fig. 2A). The Kaplan Meier chart for the minimum fertile window group showed a similar trend (Fig. 2B). The most frequently observed complete fertile window length was 3 days. Thus, the group of couples with this fertile window length was taken as the reference group in the Cox regression models. The FRs for the various fertile window lengths are given in Table 2 and ranged from 0.11 to 2.4. Also in the minimum fertile window group, the length of the fertile window was associated with the probability of ongoing pregnancy and FRs varied between 0.28 and 2.5. In multivariable Cox analysis, the fertile window length remained a strong determinant of the chance for the spontaneous conception of an ongoing pregnancy (Table 2). Table 1: Baseline characteristics of the study population (n ¼ 410) Female age (years) a 31.2 (20 40) Male age (years) a 34.3 (24 46) Duration of subfertility (years) b 1.9 (1 7) Primary subfertility (%) 62.7 Sperm characteristics b Volume (ml) 2.7 (1 8) Concentration (10 6 /ml) 35.0 (1 200) Motility (%) 32.7 (5 98) Morphology (%) 24.0 (3 64) a Mean (range); b Median (range). 1653

Keulers et al. Figure 1: Flow chart showing the distribution of the included couples over the complete and minimal fertile window lengths Besides the fertile window length, female age (FR: 0.93 per year; 95% CI: 0.89 0.98) and sperm concentration (FR: 1.08 per 1 10 6 ; 95% CI: 1.02 1.14) were significantly related with ongoing pregnancy in the first 12 months. Discussion This prospective study shows that the fertile window length, based on ultrasonographically timed testing of the sperm mucus interaction, varies among couples from,1 day to.5 days, and is related to the probability of a spontaneous conception of an ongoing pregnancy in the next 12 months. Although this relationship was strongest for the complete fertile window length group, it was also observed in those couples for whom we could only estimate a minimum fertile window length. These findings disprove the paradigm that the fertile window is a fixed period of 5 or 6 days, ending on the day of ovulation (Wilcox et al., 1995, 2000; Dunson et al., 2001; Bigelow et al., 2004). Our study was performed in a clinical setting. Since monitoring was not available at weekends, a complete fertile window length could be determined in only 212 of the 410 couples. In order to avoid bias from imprecise estimates of the fertile window length in the other couples, we classified and analysed the latter separately as couples with a minimum fertile window length. However, our initial hypothesis could be accepted on the basis of the findings in these couples too. We acknowledge that the number of visits needed to determine the fertile window length by ultrasound is considerable. Although ultrasound is an accurate and widely accepted method for ovulation detection, urinary LH-testing is a reliable alternative and may reduce the number of visits required for determining the fertile window length (Vermesh et al., 1987; Guermandi et al., 2001). A clinical limitation of the fertile window length is that its assessment requires a substantial effort by fertility professionals, and might pose a burden on subfertile couples. It is widely believed that repeated PCTs are unacceptable for many couples. In this study.80% of all couples accepted the PCTs on a daily base. The predictive value of the PCT has been evaluated as poor in a systematic review (Oei et al., 1995), but this apparently was due to disregarding the effect of the duration of subfertility. The accuracy of the PCT in predicting pregnancy is related to the duration of subfertility (Glazener et al., 2000). The clinical usefulness of the PCT is affected by incorrect timing in the cycle (Griffith and Grimes, 1990). Nevertheless, the PCT has vanished from many fertility clinics. Our data strongly suggest that the fertile window length has more potential to predict spontaneous conception than a single and poorly timed PCT, the reintroduction of which is not to be recommended. Alternative and less cumbersome methods to acquire information about the fertile window length would be of great benefit to its clinical usefulness. Our study has some potential limitations. First, we omitted to control for the time interval between intercourse and ovulation, which has been shown to be an important marker of the chance to conceive (Wilcox et al., 1995, 2000; Bigelow et al., 2004). Also, the frequency of intercourse and the quality of the cervical mucus around ovulation are related to pregnancy chance (Wilcox et al., 1995; Bigelow et al., 2004). In our clinical study, we did not control for these factors either. Second, the first ultrasound was timed on the basis of the length of the previous cycles, which is not necessarily a precise estimate of the length of the cycle under investigation. Nevertheless, the first ultrasound and cervical mucus examination were performed at least 5 days prior to ovulation in all couples. Therefore, we are not concerned that the fertile window length has been underestimated in a significant proportion of the couples. In conclusion, we showed that the length of the fertile window varies considerably among subfertile couples and 1654

Fertile window length and probability of spontaneous pregnancy Figure 2: Kaplan Meier charts showing the association between the fertile window length and the time to conception of an ongoing pregnancy in 212 couples in whom the complete fertile window length could be determined (Fig. 2A), and in 198 couples in whom the minimum fertile window length was determined (Fig. 2B). ¼ couples with a fertile window length of five days or more (n ¼ 17 for the complete fertile window length),. ¼ couples with a fertile window length of four days (n ¼ 47 for the complete fertile window length, n ¼ 6 for the minimum fertile window length), BBBBB¼ couples with a fertile window length of three days (n ¼ 58 for the complete fertile window length, n ¼ 50 for the minimum fertile window length), ¼couples with a fertile window length of two days (n ¼ 34 for the complete fertile window length, n ¼ 69 for the minimum fertile window length), ¼ couples with a fertile window length of one day or shorter (n ¼ 54 for the complete fertile window length, n ¼ 73 for the minimum fertile window length). 1655

Keulers et al. Table 2: Univariable and multivariable Cox Regression analysis Univariable analysis Multivariable analysis a FR b 95% CI c P-value FR 95% CI P-value 1 day(n¼54) 0.11 0.03 0.45 0.00 0.12 0.03 0.53 0.00 2days(n ¼ 34) 0.50 0.20 1.25 0.14 0.56 0.22 1.41 0.29 3days(n ¼ 58) 1 1 4days(n ¼ 47) 1.7 0.93 3.00 0.09 1.9 1.04 3.4 0.04 5 days(n ¼ 17) 2.4 1.1 5.2 0.02 2.3 1.06 5.0 0.04 At least 1 day(n¼73) 0.28 0.12 0.63 0.00 0.34 0.15 0.77 0.01 At least 2 days (n ¼ 69) 0.67 0.36 1.3 0.22 0.71 0.37 1.3 0.29 At least 3 days (n ¼ 50) 1.2 0.67 2.2 0.52 1.2 0.65 2.2 0.58 At least 4 days (n ¼ 6) 2.5 0.74 8.5 0.14 1.9 0.54 6.4 0.33 Female age (years) 0.92 0.88 0.97 0.00 0.93 0.89 0.98 0.01 Duration of subfertility (years) 0.86 0.71 1.03 0.10 0.87 0.71 1.06 0.16 Primary subfertility (%) 0.84 0.58 1.21 0.35 0.80 0.54 1.18 0.25 Sperm characteristics Volume (ml) 1.02 0.87 1.19 0.64 1.05 0.89 1.22 0.58 Concentration (10 6 ml 21 ) 1.10* 1.04 1.16 0.00 1.08* 1.02 1.14 0.01 Motility (%) 1.01 1.00 1.02 0.18 1.00 0.98 1.01 0.47 Morphology (%) 1.01 1.00 1.03 0.08 1.01 0.99 1.02 0.24 a In multivariable analysis potential confounders were introduced: female age, duration of subfertility, type of subfertility and sperm characteristics; b FR, fecundability ratio (odds ratio); c CI, confidence interval; *FR for sperm concentrations from 1 to 20 10 6 ml 2 1. Higher concentrations have the same FR as concentrations of 20 10 6 ml 21. that it is closely related with the chance of spontaneous conception of an ongoing pregnancy. The concept of the fertile window length merits evaluation for its potential as a fertility marker in other populations. Acknowledgements No sources of funding were involved in this study. We are indebted to the staff of the laboratory of the St. Elisabeth Hospital, who performed the SMPTs; Fulco van der Veen, Nicholas Macklon and Kathy Tucker for their valuable comments on previous drafts of this paper. References Bigelow JL, Dunson DB, Stanford JB et al. Mucus observations in the fertile window: a better predictor for conception than timing of intercourse. Hum Reprod 2004;19:889 92. Billings EL, Brown JB, Billings JJ et al. Symptoms and hormonal changes accompanying ovulation. Lancet 1972;1:282 4. Dunson DB, Sinai I, Colombo B. The relationship between cervical secretions and the daily probabilities of pregnancy: effectiveness of the Two Day Algorithm. Hum Reprod 2001;16:2278 82. Francavilla F, Romana R, La Verghetta G et al. Interactive effect of sperm and cervical mucus quality on postcoital test outcome: analysis from an andrological point of view. Int J Androl 2002;25:236 42. Glazener CM, Ford WC, Hull MG. The prognostic power of the post-coital test for natural conception depends on duration of infertility. Hum Reprod 2000;5:1953 7. Griffith CS, Grimes DA. The validity of the postcoital test. Am J Obstet Gynecol 1990;162:615 20. Guermandi E, Vegetti W, Bianchi MM et al. Reliability of ovulation test in infertile women. Obstet Gynecol 2001;1:92 6. Kremer J, Kroeks MV. Modifications of the in-vitro spermatozoal penetration test by means of the sperm penetration meter. Acta Eur Fertil 1975;6: 377 80. Oei SG, Helmerhorst FM, Keirse MJ. When is the post-coital test normal? A critical appraisal. Human Reprod 1995;10:1711 4. Stanford JB, Smith KR, Dunson DB. Vulvar mucus observations and the probability of pregnancy. Obstet Gynecol 2003;101:1285 93. Vermesh M, Kletzky OA, Davajan V et al. Monitoring techniques to predict and detect ovualtion. Fertil Steril 1987; 2:259 64. WHO laboratory manual for the examination of human semen and sperm cervical mucus interaction. 3rd edn., Cambridge, UK: Cambridge University Press, 1992 Wilcox AJ, Dunson DB, Baird DD. The timing of the fertile window in the menstrual cycle: day specific estimates from a prospective study. BMJ 2000;321:1259 62. Wilcox AJ, Weinberg CR, Baird DD. Timing of sexual intercourse in relation to ovulation. Effects on the probability of conception, survival of the pregnancy, and sex of the baby. N Engl J Med 1995;333:1517 21. Submitted on October 1, 2006; resubmitted on January 24, 2007; accepted on January 26, 2007 1656