Prediction of the potentially fertile period by urinary hormone measurements using a new home-use monitor: comparison with laboratory hormone analyses

K. Tanabe1,4, N. Susumu1, K. Hand2, K. Nishii3, I. Ishikawa3 and S. Nozawa1

1 Department of Obstetrics and Gynecology, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-Ku, Tokyo 160-8582, Japan, 2 Unipath Ltd, Priory Business Park, Bedford, UK and 3 Mitsui Pharmaceuticals Inc., 12–2, Nihonbashi 3-Chome, Chuo-Ku, Tokyo 103-0027, Japan


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: The study compared a new urinary hormone monitoring system, Clearview Primera Fertility Monitor® (CPFM), with laboratory hormone analyses in the prediction of the potentially fertile period. METHODS: Thirty healthy female volunteers provided blood and early morning urine samples for one cycle. Serum oestradiol, progesterone and luteinizing hormone (LH), and urinary LH and oestrone-3-glucuronide (E3G) were measured. The fertility status of volunteers; Low, High or Peak, was collected from monitors and compared with the hormone measurements. RESULTS: There was agreement between the first day of peak fertility and the urinary LH peak day in 65.6% of cycles and detection 1 or 2 days before the urinary LH peak day in 24.1 and 6.9% of cycles respectively. In 58.6% of cycles the system detected up to 5 days of increased fertility prior to the urinary LH peak day. Warning days of the urinary LH peak were similarly determined using defined thresholds of E3G and oestradiol providing up to 5 days warning in 82.8 and 96.6% of cycles respectively. CONCLUSIONS: The system can provide couples attempting to conceive with information about the potentially fertile days in the cycle in order that they may time intercourse. It also has potential for use in evaluation and treatment of infertile couples.

Key words: fertile period/LH/LH surge/oestradiol/urinary metabolite


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The potentially fertile period of a woman's cycle is the time during which sexual intercourse may lead to pregnancy. The length of the potentially fertile period is dependent upon the life span of the gametes within the female reproductive tract and its position within the cycle is governed by the time of ovulation. The development and release of a mature oocyte at ovulation is controlled by the pituitary gonadotrophins; FSH and LH, and gonadal sex hormones; oestrogen and progesterone (Martinez et al.1995Go). Oestradiol production by the developing follicle stimulates the cervical mucus glands to produce mucus that supports the survival and transport of sperm (Moghissi et al.1972Go), and gives rise to the surge in LH which precedes ovulation (Pauerstein et al.1978Go; Burger, 1989Go). Estimates of the length of the potentially fertile period are largely based upon two studies (Barret and Marshall, 1969; Wilcox et al.1988Go). These studies have been the subject of several analyses (Schwartz et al., 1980Go; Royston, 1982Go; Wilcox et al.1995Go; Dunson et al.1999Go). The conclusions of these analyses are supportive of a potentially fertile period which is typically 6 days long ending on the day of ovulation.The temporal relationships between ovulation and defined hormonal markers were first established by the WHO Task Force on Methods for the Determination of the Fertile Period (WHO, 1980aGo,bGo). The first of these studies estimated the median time (95% confidence limits) from the first significant rise in serum LH to ovulation to be 32 h (range 23.6–38.2 h). It has also been demonstrated that accurately measured urinary LH can be used to identify ovulation. A rapid urinary LH test has been used as a reference point to monitor changes in ovarian and uterine form and blood flow (Bourne et al.1996aGo,bGo). The study showed the median time from a positive LH test to follicle rupture to be 25 h. The temporal relationship between a positive ClearPlan One Step test (home ovulation predictor kit) and follicle rupture has been studied in detail (Collins, 1996Go). The time from a positive result to rupture ranged from 24–48 h (median 32 h) with 6 hourly testing and scans, and with daily early morning urine testing the range was 8–48 h (median 24 h). These results coincide well with the results of the WHO studies. Metabolites of oestradiol can be detected in urine. Oestrone-3-glucuronide (E3G) is one of the principal metabolites and is quantitatively important (Collins et al.1981Go). A correlation has been shown between concentrations of oestradiol present in plasma and concentrations of E3G present in early morning urine (Branch et al.1982Go; Catalan et al.1989Go). The use of urinary E3G measurement has been shown to be successful in the identification of the potentially fertile period. Based on the use of peak concentrations of urinary LH as a marker for ovulation it was concluded that a rise in the concentration of E3G of 50% over the mean of the previous three values could be used to locate the start of the potentially fertile period (between peak LH day minus 3 and peak LH day minus 7) in over 90% of the cycles studied (Aldercreutz et al., 1982). Other studies have shown the use of E3G to be 83 and 89% successful in delineating the potentially fertile period (WHO 1983Go; Schiphorst et al.1985Go).

The studies described above show that the measurement of LH alone will result in the prediction and detection of the two most potentially fertile days of the cycle. The evidence suggests that a test based on the monitoring of both urinary LH and E3G could be used to delineate the potentially fertile period and provide additional warning days of ovulation, on which intercourse may lead to pregnancy, compared with the use of LH alone. A new system has been developed to identify those days on which a couple should have intercourse if a woman is to maximise her chances of conception. The Clearview Primera Fertility Monitor (CPFM) (Mitsui Pharmaceuticals Inc., Tokyo, Japan), which has been designed for home-use, identifies the potentially fertile period of a woman's cycle by monitoring changes in the levels of LH and E3G in urine. (the CPFM is marketed in the USA as ClearPlan Easy® Fertility Monitor, Unipath Diagnostics Co., Princeton, NJ, USA). The aim of this study was to test whether CPFM could provide information about changes in cycle fertility comparable with that deduced from laboratory measurements. In order to exclude, at this stage, potential errors resulting from use of the system by ordinary women at home, CPFM testing was performed in real time in a laboratory. The monitor LH and E3G measurements and corresponding fertility status were compared with serum oestradiol, serum LH, urinary E3G and urinary LH measurements in a population of normal healthy women.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The CPFM system
The system comprises a hand-held monitor and disposable dual-assay urine test sticks. The test sticks simultaneously detect LH and E3G in early morning urine. The LH assay is a classical sandwich assay and as the concentration of LH in the urine increases then the intensity of the line formed on the test stick increases. The E3G assay is a competition assay and as the concentration of E3G increases then the corresponding line intensity decreases. The monitor optically measures the intensity of the lines that form on the test sticks after sampling. The corresponding signal is measured in percentage transmission units (%T). As the concentration of LH increases the associated signal increases, as the concentration of E3G increases the associated signal decreases.

The system will delineate three levels of fertility according to changes detected in the concentrations of LH and E3G. Low fertility is displayed when the probability of conception is low and these hormones are at a baseline concentration. Low fertility will be displayed from day 1 of a woman's cycle until rises above the baseline concentrations are detected. The change from Low fertility to High fertility is triggered by detection of elevated E3G concentrations, typically at concentrations between 20–30 ng/ml. The display of High fertility indicates that the woman is approaching a period of maximum fertility. High fertility is also displayed for one day after Peak fertility. The change from High fertility to Peak fertility is triggered by the detection of an LH surge, typically with a concentration >30 IU/l. The display of Peak fertility indicates that ovulation is imminent. Peak fertility is displayed on the day of the LH surge (CPFM peak day) and on the following day. Subsequently High fertility will be displayed for 1 day prior to a return to Low fertility. The changes in fertility are displayed on the monitor's LCD.

The transition from Low to High to Peak fertility is indicated by an increasing number of solid bars on the display; one bar (Level I) for Low fertility, two bars (Level II) for High fertility and three bars (Level III) for Peak fertility. In addition an egg symbol is displayed on Peak fertility days. The %T units are not displayed by the monitor; the monitor uses a built in rule-set to convert the detected changes in LH and E3G %T into the fertility status.

To use the system, on the morning after the start of menses the woman presses a button on the monitor to indicate that this is day 1 of her cycle. She then simply looks at her monitor every morning to check whether she needs to do a test. To perform a test she holds a test stick in her urine stream for 3 s. She then inserts the test stick into the monitor which reads the test stick after 5 min. On completion of the test the monitor informs the woman of her fertility status through a simple display.

Study recruitment
Normal healthy women were recruited onto the study. A brief medical history was collected from the volunteers who were screened for the following selection criteria: Age 20–40 years; regular menstrual cycles of length 25–34 days; no history of ovulatory disorders; no use of medication which would interfere with ovarian function; not breastfeeding; not pregnant.

Prior to admission all volunteers gave their informed consent and it was explained that they were free to leave the study at any time.

Study protocol
Blood and early morning urine samples were collected from volunteers for one cycle according to the schedule presented in Table IGo.


View this table:
[in this window]
[in a new window]
 
Table I. Schedule for the collection of blood and early morning urine samples
 
All blood samples were analysed for serum oestradiol, progesterone and LH. All urine samples were analysed for LH and E3G. Serum oestradiol and progesterone were measured by radioimmunoassay (RIA) (Tatron Laboratories Inc., Tokyo, Japan). The sensitivity of the serum oestradiol assay was 10 pg/ml. Intra-assay coefficients of variation (CVs) were 9.6, 8.4 and 8.7% and inter-assay CVs were 7.0, 6.1 and 5.1% at low, medium and high concentrations respectively. The sensitivity of the serum progesterone assay was 0.2 ng/ml. Intra-assay CVs were 6.9 and 5.0% and inter-assay CVs were 5.9 and 7.8% at low and high concentrations respectively. Serum and urinary LH were measured by RIA (SPAC-S LH kit; Daiichi Radioisotope Laboratories Ltd., Tokyo, Japan). The standards had been calibrated against the WHO 1st IRP LH 68/40. The sensitivity of the serum LH assay is 0.2 mIU/ml. Intra-assay CVs were 3.2, 9.1 and 4.8% and inter-assay CVs were 4.1, 5.9 and 3.2% at low, medium and high concentrations respectively. The sensitivity of the urinary LH assay is 0.2 mIU/ml. Intra-assay CVs were 6.2, 8.0 and 6.8% and inter-assay CVs were 7.1, 7.0 and 6.2% at low, medium and high concentrations respectively. Urinary E3G was measured by time-resolved fluorescence immunoassay (Unipath Ltd. in collaboration with Wallac OY, Turku, Finland). The sensitivity of the assay was 0.2 ng/ml. Intra-assay CVs were 4.9, 2.3 and 2.2% and inter-assay CVs were 4.4, 2.3 and 2.2% at low, medium and high concentrations respectively.

In this study the volunteers did not test their urine themselves using the monitor, instead they provided early morning urine samples for testing in the laboratory on the day of collection. Two types of CPFM measurements were made. Firstly the fertility status on that day, Low, High or Peak, was determined by testing the collected urine with a test stick and reading the fertility status using a standard CPFM allocated to each volunteer. Secondly the corresponding %T value for LH and E3G was measured by completing a second test on each of the collected samples. A monitor linked to a computer and adapted to provide the %T readings via the computer was used for this purpose.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The 30 volunteers recruited onto the study were monitored for one cycle. They had a mean age of 28.6 years with 10th and 90th percentiles of 22.5 and 35.5 years respectively and their mean cycle length during the study was 29.2 days with 10th and 90th percentiles of 25 and 32 days respectively.

For the purposes of this study the day of peak urinary LH concentration was used as the reference day. A peak in the urinary LH concentration was observed in 29/30 cycles. One cycle was monitored up until day 28 of a 31 day cycle and no urinary LH peak was recorded. In all 29 cycles with a urinary LH peak, the peak was accompanied by a corresponding peri-ovulatory rise in the concentrations of serum oestradiol. A mid-luteal phase peak in progesterone was observed in 28/29 of the cycles. A possible reason for the absence of the progesterone peak in one of the cycles was that sample collection had stopped only 5 days after the urinary LH peak day and therefore the peak was not observed.

Figure 1Go shows, on the days relative to the urinary LH peak day, the changes in mean serum oestradiol, LH, and progesterone concentrations, urinary E3G and LH concentrations, urinary LH and E3G %T measurements (adapted CPFM monitors), and percentage of cycles displaying High or Peak fertility on each day (standard CPFM monitors). Samples were not collected on all the cycle days (Table IGo) however the data in Figure 1Go are plotted against cycle day relative to the urinary LH peak day and the mean values may not include measurements from all volunteers. Table IIGo lists the number of measurements included on each day relative to the urinary LH peak and the standard errors of the mean are included in Figure 1Go.



View larger version (27K):
[in this window]
[in a new window]
 
Figure 1. Graphs showing changes in concentration of serum and urinary LH, and serum progesterone (P), CPFM-measure LH and oestrone 3-glucuranide (E3G) (from adapted monitors), serum oestradiol and urinary E3G, and the percentage of cycles displaying either CPFM High or Peak fertility status (from standard monitors), on days relative to the urinary LH peak day. Values are means (± SEM), the number of measurements are listed in Table IIGo.

 

View this table:
[in this window]
[in a new window]
 
Table II. Number of serum and urine samples used to calculate the mean value on each day relative to the urinary LH peak day.
 
The peak mean serum LH concentration occurred on the same day as the peak mean urinary LH concentration (assigned as day 0) (Figure 1aGo). The changes in the mean urinary E3G concentration were very similar to those observed in the mean serum oestradiol concentration (Figure 1cGo). The mean peak concentrations occurred on days 0 and –1 respectively and there were rises in both the mean urinary E3G and mean serum oestradiol in the mid-luteal phase. The corresponding mean CPFM LH and E3G signals (Figure 1bGo) showed similar changes to the mean serum and urinary concentrations with both the peak mean LH and E3G signals occurring on day 0 the same day as the peak mean urinary LH and E3G concentrations measured by radioimmunoassay and fluorescence immunoassay respectively.

Table IIIGo shows that there was 65.6% agreement between the CPFM Peak day and the urinary LH peak day. The CPFM Peak day was observed either 1 day or 2 days before the urinary LH peak day in 24.1 and 6.9% of cycles respectively. In one cycle the CPFM showed peak status on day 9, 7 days prior to the urinary LH peak observed on day 16. A corresponding small peak in the urinary LH concentration was observed on this day. Similarly Table IVGo shows that there was 55.2% agreement between the urinary LH peak day and the serum LH peak day. The serum LH peak day as observed either 1 day or 2 days before the urinary LH peak day in 37.9 and 3.4% of cycles respectively. In one cycle the serum LH peak was observed on day 18, 4 days prior to the urinary LH peak observed on day 22. After day 18 samples were collected on alternate days. This may have affected accurate detection of the urinary and serum LH peak and is the likely reason for the greater difference observed.


View this table:
[in this window]
[in a new window]
 
Table III. Difference between the CPFM Peak day and the urinary LH peak day (CPFM peak day–urinary LH peak day)
 

View this table:
[in this window]
[in a new window]
 
Table IV. Difference between the serum LH peak day and the urinary LH peak day (serum LH peak day – urinary LH peak day)
 
Figure 1cGo shows the mean serum oestradiol and mean urinary E3G concentrations on days relative to the urinary LH peak day, together with the percentage of cycles displaying CPFM High or peak fertility status on these days. The changes in mean urinary E3G concentration correspond closely to the changes in mean serum oestradiol concentration on these days. In addition, as the mean concentrations of the hormones rise there is a similar increase in the percentage of cycles displaying High or peak fertility status on these days.

An oestradiol increase to concentrations >100 pg/ml was considered as the periovulatory oestradiol rise (Leidenberger, 1992Go). Similarly an E3G rise day was assigned using a threshold of 20 ng/ml. The days on and after the serum oestradiol rise day up until the urinary LH peak day were considered as warning days and up to 5 days warning was detected in 96.6% of cycles (Table VGo). Similarly, the number of days on and after the E3G rise day up until the urinary LH peak day were considered as warning and up to 5 days warning were detected in 82.8% of cycles (Table VIGo). In 58.6% of cycles CPFM showed either High or peak fertility for up to 5 days prior to the urinary LH peak day (Table VIIGo).


View this table:
[in this window]
[in a new window]
 
Table V. Warning of the urinary LH peak provided by the serum oestradiol rise day (day of urinary LH peak – oestradiol rise day)
 

View this table:
[in this window]
[in a new window]
 
Table VI. Warning of the urinary LH peak provided by the E3G rise day (day of urinary LH peak –E3G rise day)
 

View this table:
[in this window]
[in a new window]
 
Table VII. Warning of the urinary LH peak provided by the monitor (number of High and Peak fertility days prior to the urinary LH peak)
 
Warning of the CPFM first Peak day was defined as the number of High fertility days preceding it. One cycle had no CPFM Peak day (only Low and High days displayed). Of the remaining 29 cycles CPFM provided up to 5 days of High fertility prior to the CPFM Peak day in 72.4% of cases (Table VIIIGo). A total of 69% of cycles had between 1 and 5 days of High fertility prior to the Peak fertility day with only 3.4% of cycles having no High days. The median number of High and peak days was 8.


View this table:
[in this window]
[in a new window]
 
Table VIII. Warning of the CPFM peak day (number of High fertility days prior to the peak day)
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
It has recently been shown that the highest probability of pregnancy results from intercourse on the 2 days prior to ovulation and that the potentially fertile period begins ~5 days prior to ovulation ending on the day of ovulation (Dunson et al.1999Go). There is, therefore, a 6 day window where intercourse may lead to pregnancy. It is important that couples having difficulty conceiving are able to have intercourse during these potentially fertile days in the woman's cycle. Existing methods used to time intercourse are unable to identify all of these days. The shift in basal body temperature occurs too late to be of predictive value in the current cycle (Dunson et al., 1996) and the surge in urinary LH can typically only provide 1 to 2 days advanced warning of ovulation (Collins, 1996Go). CPFM offers increased benefits over existing methods; by informing the woman that her fertility is high (High fertility) it provides more warning of her 2 day period of maximum fertility (Peak fertility), and in addition intercourse on the High fertility days themselves may lead to pregnancy. The number of potentially fertile days is dependent upon the length of time that spermatozoa can survive in the reproductive tract such that they can fertilize the ovum when it is released. The appearance of sperm supportive mucus is controlled by the production of oestradiol which is produced in increasing quantities as the follicle develops towards ovulation (Burger, 1989Go). Detection of the rise in the concentrations of the urinary metabolite E3G can provide information about the changing concentrations of oestradiol during the follicular phase of the cycle. This information can be used by couples to identify the most potentially fertile days in the cycle and allow them to time intercourse accordingly.

This study has shown good agreement between the CPFM Peak day and the urinary LH peak as detected by laboratory assay. It is not suprising that in 24.1% of cycles CPFM displayed Peak fertility the day before the urinary LH peak. This is a result of the difference in days between the LH surge, detected by CPFM, and the actual peak in urinary LH concentrations assigned retrospectively. In two cycles urinary LH began to rise 2 days before levels peaked and CPFM detected the start of the rise in LH. Neither is it suprising to see that in 37.9% of cycles peak serum LH concentrations were reached 1 day before peak urinary LH concentrations. This is a result of the time taken for changes in serum hormone concentrations to be detected in the urine, particularly when measurements are taken once during a 24 h period. In one cycle (3.4%) CPFM detected a false positive LH surge declaring Peak fertility 7 days prior to the observed urinary LH peak.

Another study was recently undertaken comparing the home-use performance of CPFM against transvaginal ultrasound. Results show that in 123 out of 135 ovulatory cycles with a CPFM peak day, ovulation was correctly predicted within 2 days following the CPFM Peak day (Behre et al.2000Go). Detection of the LH surge can at best identify the two most potentially fertile days of the cycle and therefore has limited opportunities for the timing of intercourse. This study has shown that the rise in E3G closely follows the rise in serum oestradiol and the number of days warning of the urinary LH peak day provided by CPFM coincide well with the warning provided by the quantitative measurement of serum oestradiol and E3G concentrations. CPFM was able to show High fertility for up to 5 days prior to the urinary LH peak day in 58.6% of cycles and High fertility for up to 5 days prior to Peak fertility in 72.4% of cycles. It should be noted that, as the urinary LH peak is an indirect marker of ovulation, the 5 days of High fertility detected by the monitor cannot be compared exactly with the five potentially fertile days prior to ovulation estimated from analysis of the Wilcox data (Dunson et al., 1999Go). Typically follicle rupture has been observed to occur 24 h after detection of the urinary LH surge (Collins, 1996Go). These additional days of potential fertility identified by CPFM, compared with LH tests alone, increase the opportunities for maximising conception.

For 27.6% of subjects the number of High fertility days displayed by the system was greater than five. It remains to be seen whether experiencing more than 5 days of High fertility has any bearing on a couples ability to also have intercourse on Peak fertility days when the chances of conception are at a maximum.

Here we present the results of a study designed to test whether CPFM can provide information about changes in cycle fertility which is comparable with that deduced from laboratory measurements. Results of a home-use study (Behre et al., 2000Go) have shown that women can use the system at home to predict ovulation and therefore the days of highest potential fertility. These results demonstrate that the CPFM will detect changes in urinary levels of LH and E3G which coincide well with laboratory measurements in the definition of the potentially fertile period. The system, which has been designed for home use, will allow couples to use the information to time intercourse for the best prospects of natural conception. In addition the system shows potential for use in the evaluation and treatment of the infertile couple.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors would like to express their gratitude to all the volunteers who participated in the study and the staff at the Iryouhoujinshadan Seikoukai NS Clinic. The study was funded by Unipath Ltd., Bedford, UK and Mitsui Pharmaceuticals Inc. Tokyo, Japan.


    Notes
 
4 To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Tokyo Dental College Ichikawa General Hospital, 11–13, Suagno 5-Chome, Ichikawa-City, Chiba 272–8513, Japan. E-mail: ktanabe{at}tdc.ac.jp Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Adlercreutz, H., Brown, J., Collins, W. et al. (1982) The measurement of urinary steroid glucuronides as indices of the fertile period on women. J. Steroid Biochem., 17, 695–702.[ISI][Medline]

Barrett, J.C. and Marshell, J. (1969) The risk of conception of different days of the menstrual cycle. Pop. Studies, 23, 455–461.[ISI]

Behre, H.M., Kuhlage, L., Gassner, C. et al. (2000) Prediction of ovulation by urinary hormone measurements with the home use ClearPlan Fertility Monitor: comparison with transvaginal ultrasound scans and serum hormone measurements. Hum. Reprod., 15, 2478–2482.[Abstract/Free Full Text]

Bourne, T.H., Hagström, H.-G., Granberg, S. et al. (1996a) Ultrasound studies of vascular and morphological changes in the human uterus after a positive self-test for the urinary luteinising hormone surge. Hum. Reprod., 11, 369–375.[Abstract]

Bourne, T.H., Hagström, H.-G., Hahlin, M. et al. (1996b) Ultrasound studies of vascular and morphological changes in the human corpus luteum during the menstrual cycle. Fertil. Steril., 65, 753–758.[ISI][Medline]

Branch, C.M., Collins, P.O. and Collins, W.P. (1982) Ovulation prediction: changes in the concentrations of urinary estrone-3-glucuronide, estradiol-17ß-glucuronide and estriol-16{alpha}-glucuronide during conceptional cycles. J. Steroid. Biochem., 16, 345–347.[ISI][Medline]

Burger, H.G. (1989) Estradiol: The physiological basis of the fertile period. Int. J. Gynaecol. Obstet., 1, 5–9.

Catalan, R., Castellanos, J.M., Palomino, T. et al. (1989) Correlation between plasma estradiol and estrone-3-glucuronide in urine during the monitoring of ovulation induction therapy. Int. J. Fertil., 34, 271–275.[ISI][Medline]

Collins, W.P. (1996) Indicators of potential fertility: scientific principles. In Bonnar, J. (ed.) Natural Conception Through Personal Hormone Monitoring. The Parthenon Publishing Group, New York, pp. 13–33.

Collins, W.P., Branch, C.M. and Collins, P.O. (1981). Ovulation prediction and detection by the measurement of steroid glucuronides. In Cortes-Prieto, J., Campos de Paz, A. and Neces-e-Castro, M. (eds.) Research on Fertility and Sterility. MPT Ltd., Lancaster, pp. 19–33.

Dunson, D.B., Baird, D.D., Wilcox, C.R. et al. (1999) Day specific probabilities of clinical pregnancy based on two studies with imperfect measures of ovulation. Hum. Reprod., 14, 1835–1839.[Abstract/Free Full Text]

Leidenberger, F. (1992) Clinical endocrinology for gynecologists. Springer, Berlin, pp. 160–166.

Martinez, A.R., Zinaman, M.J., Jennings, V.H. et al. (1995) Prediction and detection of the fertile period: the markers. Int. J. Fertil., 40, 139–155.[ISI]

Moghissi, K.S., Syner, F.N. and Evans, T.N. (1972) A composite picture of the menstrual cycle. Am. J. Obstet. Gynecol., 114, 405–418.[ISI][Medline]

Pauerstein, C.J., Eddy, C.A., Croxatto, H.D. et al. (1978) Temporal relationships of estrogen, progesterone and luteinising hormone levels to ovulation in women and infra-human primates. Am. J. Obstet. Gynecol., 130, 876–886.[ISI][Medline]

Royston, J.P. (1982) Basal body temperature, ovulation and the risk of conception, with special reference to the lifetimes of sperm and egg. Biometrics, 38, 397–406.[ISI][Medline]

Schwartz, D., MacDonald, P.D.M. and Heuchel, V. (1980) Fecundability, coital frequency and the viability of ova. Pop. Studies, 34, 397–400.[ISI]

Schiphorst, L.E.M., Collins, W.P. and Royston, J.P. (1985) An estrogen test to determine the times of potential fertility in women. Fertil. Steril., 44, 328–334.[ISI][Medline]

WHO Task Force on Methods for Determination of the Fertile Period (1980a) Temporal relationships between ovulation and defined changes in the concentration of plasma estradiol-17ß, luteinising hormone, follicle stimulating hormone and progesterone. I. Probit analysis. Am. J. Obstet. Gynecol., 138, 383–390.[ISI][Medline]

WHO Task Force on Methods for Determination of the Fertile Period (1980b) Temporal relationships between ovulation and defined changes in the concentration of plasma estradiol-17ß, luteinising hormone, follicle stimulating hormone and progesterone. II. Histologic dating. Am. J. Obstet. Gynecol., 139, 886–895.[ISI]

WHO Task Force on Methods for Determination of the Fertile Period (1983) Temporal relationships between indices of the fertile period. Fertil. Steril., 39, 647–655.[ISI][Medline]

Wilcox, A.J., Weinberg, C.R., O'Connor, J.F. et al. (1988) Incidence of early loss of pregnancy. N. Eng. J. Med., 319, 189–194.[Abstract]

Wilcox, A.J., Weinberg, C.R. and Baird, D.D. (1995) Timing of sexual intercourse in relation to ovulation. N. Eng. J. Med., 333, 1517–1521.[Abstract/Free Full Text]

Submitted on December 29, 2000; accepted on April 19, 2001.