- Email: [email protected]
Temporal Relationship Between Basal Body Temperature Nadir and Luteinizing Hormone Surge in Normal Women*
FERTILITY AND STERILITY
Vol. 27, No.7, July 1976 Printed in U.SA.
Copyright © 1976 The American Fertility Society
TEMPORAL RELATIONSHIP BETWEEN BASAL BODY TEMPERATURE NADIR AND LUTEINIZING HORMONE SURGE IN NORMAL WOMEN* NAOMI M. MORRIS, M.D.,* LOUIS E. UNDERWOOD, M.D.J WILLIAM EASTERLING, JR., M.D.
AND
Departments of Maternal and Child Health, Pediatrics, and Obstetrics and Gynecology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27514
Basal body temperature (BET) as a predictor of ovulation was assessed by examining the temporal relationship between the BET shift and the luteinizing hormone (LH) surge in individual cycles of 27 normal women. For 22 of the subjects, the LH surge occurred on the same day or within one day of the BET nadir. For the remaining five subjects, the surge fell within 2 days after or 3 days before the nadir. Despite the BET nadir's close temporal association with the LH surge, daily examination of BET for the purpose of predicting the day of ovulation during a given cycle is unsatisfactory. By 48 hours following the nadir, when one could usually be certain that temperature elevation had occurred, all subjects had already exhibited the LH surge.
Shift in basal body temperature (BBT) is used widely as an indicator of ovulation. However, there is scant documentation in individual normal women of the temporal relationship between the BBT shift and ovum release. In most reports of hormonal and clinical patterns of normal menstrual cycles, results have been pooled to produce a composite picture. Only a few observations on individual cycles have been reported. 1 - 9 Since the midcycle luteinizing hormone (LH) surge in serum is believed to precede ovulation by a brief, predictable period of time, 10 serial measurement of LH levels makes it possible to study Accepted February 5, 1976. *Supported by National Institute of Child Health and Human Development Grant 5 ROI HD0745402, The School of Public Health General Research Support, and the University Research Council. tReprint requests: Dr. Naomi Morris, Department of Maternal and Child Health, University of North Carolina, Chapel Hill, N. C. 27514. +Jefferson-Pilot Fellow in Academic Medicine, University of North Carolina.
the relationship between BBT and ovulation. In this report we assess the value of BBT measurement as a predictor of the day of ovulation by examining the temporal relationship between the BBT shift and the LH surge in individual menstrual cycles of 27 normal women. MATERIALS AND METHODS
The study group was originally composed of 43 normal women, most of whom were wives of university graduate students. The median age was 25 years. These subjects, recruited for a study of human sexual behavior patterns, used ovulation thermometers to measure BBT before arising each morning, and mailed temperature reports daily to the investigators. For 10 to 14 consecutive midcycle days, blood samples were obtained by a technician. For each subject, samples were drawn at the same hour each day between 9:00 A.M. and 4:00 P.M.
780
Vol. 27, No. 7
781
RELATIONSHIP BETWEEN BBT AND UI SURGE
Temperature cycles, graphed by one of the investigators by calendar day and bracketed by the menstrual bleeding periods, were reviewed by a gynecologist with a special interest in endocrinology and fertility. This investigator had no knowledge of the results of the LH determinations and had no contact with the women. The day of ovulation was estimated by (1) determining whether the temperature curve for the cycle was biphasic and discarding cycles in which it was not and (2) searching for approximately 14 days prior to menstrual bleeding for a temperature nadir followed by several days of rising temperature. When the day of the shift in BBT could not be determined, the ovulation day was recorded as unknown. Serum samples were aliquoted and stored at -20° C until assayed. LH levels were measured in duplicate by a double-antibody precipitation radioimmunoassay* which had a sensitivity of 1.2 miU/ml. The time of ovulation according to the serum LH surge was estimated independently by an endocrinologist who had no contact with the women and no access to the conclusions of the gynecologist. The day of the LH surge was identified when the peak LH level was at least 2 times higher than the LH level on any other day during the cycle. In most subjects the LH peak was 6 to 15 times higher than the next highest LH value. In all subjects, the peak LH level on the day of the surge was more than 4 times the mean level for the follicular and luteal phases of the cycle.
*The human pituitary LH for iodination (LER960), anti-human human chorionic gonadotropin, and human pituitary extract reference preparation (LER-907) used in these studies were provided by the Endocrinology Study Section of the National Institute for Arthritis, Metabolism and Digestive Diseases.
-
~
DAY
l
r-f
-3 -2 -I 0 +I +2
I
BBT NADIR
FIG. 1. Day on which the LH surge occurred in relation to the BBT nadir in individual menstrual cycles of 27 normal women.
RESULTS
It was necessary to discard the data derived from 16 of the 43 women who participated in the study, In eight women with a midcycle LH surge it was not possible to identify the nadir of the temperature curve: four seemed to have biphasic curves, but illness or missing data obscured midcycle events; the other four did not have clearly biphasic curves. Four women, whose cycle lengths were 27, 35, 35, and 43 days, had no detectable LH surge or shift in BBT. The LH surge was missed in four other women with long cycles who exhibited a shift in BBT 1 or more days after serial blood sampling was discontinued. In the remaining 27 subjects, it was possible to compare the timing of the BBT nadir with the LH surge (Fig. 1). For 22 ofthe 27 subjects (81%), the LH surge occurred on the same day or within 1 day of the BBT nadir. For the remaining five subjects, the surge fell within 2 days after or 3 days before the nadir.
DISCUSSION
The timing of the BBT nadir correlated fairly well with the LH surge. Studies conducted by others, describ-
782
MORRIS ET AL.
ing patterns for 18 individual cycles with BBT nadirs assigned by our method, collectively give a similar pattern, although the interval to the LH surge on either side of the nadir is extended to at least 4 days. 2 ' 5 Determining the time of the BBT shift was often difficult, since the nadir sometimes dropped below the mean temperature for the preceding 6 days. Small increases in temperature in the first 1 or 2 days after the nadir might, therefore, still result in temperatures at or below the mean follicular phase temperature. In our patients, the modal (most common) rise above the respective mean follicular phase temperatures within 48 hours was only 0.1° F. According to a World Health Organization definition 11 the shift in BBT to the hyperthermic phase of the cycle occurs within a period of 48 hours or less. The definition also requires that three consecutive daily temperatures be at least O.T C higher than the previous six daily temperatures. Had we strictly adhered to these or others of the many proposed criteria for assessment of BBT curves, our high correlation with the LH surge might have declined. Certainly the number of readable cycles would have been reduced. In our study, since temperature readings and blood samples were obtained only once within each 24-hour period, the temporal correlation between BBT nadir and LH surge may actually be higher than the data indicate. The basis of this supposition is that the LH peak may have occurred as much as 12 to 18 hours before or after the time of observations. The same reasoning, however, makes it possible that the separation between nadir and surge is actually 1 day longer than it appears. The data provide some insight into the limitation of the BBT nadir in determining the time of ovulation. In our subjects, temporal localization of the BBT nadir was determined by a medical spe-
July 1976
cialist who was aware of the timing of menstrual bleeding before and after the period of interest and was able to study the temperature values for the entire cycle. Opportunity for localization of the BBT nadir, therefore, was optimal. Without this spectrum of information, particularly without knowledge of the last half of the cycle, accuracy would have undoubtedly declined. There are additional factors which might lead to an overestimation of the predictive value of BBT when this clinical tool is to be utilized under more usual circumstances. Some of our subjects were selected for blood donation because they had shown definable shifts in BBT in previously recorded menstrual cycles. Nevertheless, we could not locate the day of the BBT shift during the cycle of interest in records from at least 12 of the 43 subjects. Furthermore, even when a clear shift was obvious, had we been using BBT to predict the period of fertility, it would have taken 48 hours or more to be certain that a shift had occurred. By this time all of our subjects had already exhibited the LH surge, as had 17 of the 18 subjects whose cycles have been reported by others. The subjects therefore were partly, if not wholly, through the period of maximal fertility. For both fostering conception and preventing pregnancy, BBT is an unsatisfactory signal.
Acknowledgments. Thanks are due to Charles Chase, Ruben Maness, Doretha Foushee, and Eyvonne Bruton for technical assistance.
REFERENCES 1. Taymore ML, Aono T, Pheteplace C, Page G:
Follicle-stimulating hormone and luteinizing hormone in serum during the menstrual cycle determined by radioimmunoassay. Acta Endocrinol (Kbh) 59:298, 1968
Vol. 27, No.7
RELATIONSHIP BETWEEN BBT AND LH SURGE
2. Midgley AR Jr, Jaffe RB: Human luteinizing hormone in serum during the menstrual cycle: determination by radioimmunoassay. J Clin Endocrinol Metab 26:1375, 1966 3. Zuspan FP, Zu .pan KJ: Ovulatory plasma amine (epinephrine and norepinephrine) surge in the woman. Am J Obstet Gynecol 117:654, 1973 4. Jewelewicz R, Dyrenfurth I, Warren M, Frantz AG, Vande Wiele RL: Effect of thryotropinreleasing hormone (TRH) upon the menstrual cycle in women. J Clin Endocrinol Metab 39:387, 1974 5. Dignam WJ, Parlow AF, Coyotupa J, Honda J, Hiroi M, Kinoshita K, Kushinsky S: Plasma estrogens and serum gonadotropins in a normal menstrual cycle. Obstet Gynecol 43:484, 1974 6. Goebelsmann U, Midgley AR Jr, Jaffe RB: Regulation of human gonadotropins. VII. Daily individual urinary estrogens, pregnanediol and serum luteinizing and follicle stimu-
7.
8.
9.
10.
11.
783
lating hormones during the menstrual cycle. J Clin Endocrinol Metab 29:1222, 1969 Abraham GE, Klaiber EL: Plasma immunoreactive estrogen and LH during the menstrual cycle. Am J Obstet Gynecol 108:528, 1970 Ross GT, Cargille CM, Lipsett MB, Rayford PL, Marshall JR, Strott CA, Rodbard D: Pituitary and gonadal hormones in women during spontaneous and induced ovulatory cycles. Recent Prog Horm Res 26:1, 1970 Vande Wiele RL, Bogumil J, Dyrenfurth I, Ferin M, Jewelewicz R, Warren M, Rizkallah T, Mikhail G: Mechanisms regulating the menstrual cycle in women. Recent Prog Horm Res 26:63, 1970 Edwards RG, Steptoe PC: Induction of follicular growth, ovulation, and luteinization in the human ovary. J Reprod Fertil [Suppl] 22:121, 1975 Report of a W.H.O. Scientific Group: Biology of fertility control by periodic abstinence. WHO Tech Rep Ser 360:1, 1967
Vol. 27, No.7, July 1976 Printed in U.SA.
Copyright © 1976 The American Fertility Society
TEMPORAL RELATIONSHIP BETWEEN BASAL BODY TEMPERATURE NADIR AND LUTEINIZING HORMONE SURGE IN NORMAL WOMEN* NAOMI M. MORRIS, M.D.,* LOUIS E. UNDERWOOD, M.D.J WILLIAM EASTERLING, JR., M.D.
AND
Departments of Maternal and Child Health, Pediatrics, and Obstetrics and Gynecology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27514
Basal body temperature (BET) as a predictor of ovulation was assessed by examining the temporal relationship between the BET shift and the luteinizing hormone (LH) surge in individual cycles of 27 normal women. For 22 of the subjects, the LH surge occurred on the same day or within one day of the BET nadir. For the remaining five subjects, the surge fell within 2 days after or 3 days before the nadir. Despite the BET nadir's close temporal association with the LH surge, daily examination of BET for the purpose of predicting the day of ovulation during a given cycle is unsatisfactory. By 48 hours following the nadir, when one could usually be certain that temperature elevation had occurred, all subjects had already exhibited the LH surge.
Shift in basal body temperature (BBT) is used widely as an indicator of ovulation. However, there is scant documentation in individual normal women of the temporal relationship between the BBT shift and ovum release. In most reports of hormonal and clinical patterns of normal menstrual cycles, results have been pooled to produce a composite picture. Only a few observations on individual cycles have been reported. 1 - 9 Since the midcycle luteinizing hormone (LH) surge in serum is believed to precede ovulation by a brief, predictable period of time, 10 serial measurement of LH levels makes it possible to study Accepted February 5, 1976. *Supported by National Institute of Child Health and Human Development Grant 5 ROI HD0745402, The School of Public Health General Research Support, and the University Research Council. tReprint requests: Dr. Naomi Morris, Department of Maternal and Child Health, University of North Carolina, Chapel Hill, N. C. 27514. +Jefferson-Pilot Fellow in Academic Medicine, University of North Carolina.
the relationship between BBT and ovulation. In this report we assess the value of BBT measurement as a predictor of the day of ovulation by examining the temporal relationship between the BBT shift and the LH surge in individual menstrual cycles of 27 normal women. MATERIALS AND METHODS
The study group was originally composed of 43 normal women, most of whom were wives of university graduate students. The median age was 25 years. These subjects, recruited for a study of human sexual behavior patterns, used ovulation thermometers to measure BBT before arising each morning, and mailed temperature reports daily to the investigators. For 10 to 14 consecutive midcycle days, blood samples were obtained by a technician. For each subject, samples were drawn at the same hour each day between 9:00 A.M. and 4:00 P.M.
780
Vol. 27, No. 7
781
RELATIONSHIP BETWEEN BBT AND UI SURGE
Temperature cycles, graphed by one of the investigators by calendar day and bracketed by the menstrual bleeding periods, were reviewed by a gynecologist with a special interest in endocrinology and fertility. This investigator had no knowledge of the results of the LH determinations and had no contact with the women. The day of ovulation was estimated by (1) determining whether the temperature curve for the cycle was biphasic and discarding cycles in which it was not and (2) searching for approximately 14 days prior to menstrual bleeding for a temperature nadir followed by several days of rising temperature. When the day of the shift in BBT could not be determined, the ovulation day was recorded as unknown. Serum samples were aliquoted and stored at -20° C until assayed. LH levels were measured in duplicate by a double-antibody precipitation radioimmunoassay* which had a sensitivity of 1.2 miU/ml. The time of ovulation according to the serum LH surge was estimated independently by an endocrinologist who had no contact with the women and no access to the conclusions of the gynecologist. The day of the LH surge was identified when the peak LH level was at least 2 times higher than the LH level on any other day during the cycle. In most subjects the LH peak was 6 to 15 times higher than the next highest LH value. In all subjects, the peak LH level on the day of the surge was more than 4 times the mean level for the follicular and luteal phases of the cycle.
*The human pituitary LH for iodination (LER960), anti-human human chorionic gonadotropin, and human pituitary extract reference preparation (LER-907) used in these studies were provided by the Endocrinology Study Section of the National Institute for Arthritis, Metabolism and Digestive Diseases.
-
~
DAY
l
r-f
-3 -2 -I 0 +I +2
I
BBT NADIR
FIG. 1. Day on which the LH surge occurred in relation to the BBT nadir in individual menstrual cycles of 27 normal women.
RESULTS
It was necessary to discard the data derived from 16 of the 43 women who participated in the study, In eight women with a midcycle LH surge it was not possible to identify the nadir of the temperature curve: four seemed to have biphasic curves, but illness or missing data obscured midcycle events; the other four did not have clearly biphasic curves. Four women, whose cycle lengths were 27, 35, 35, and 43 days, had no detectable LH surge or shift in BBT. The LH surge was missed in four other women with long cycles who exhibited a shift in BBT 1 or more days after serial blood sampling was discontinued. In the remaining 27 subjects, it was possible to compare the timing of the BBT nadir with the LH surge (Fig. 1). For 22 ofthe 27 subjects (81%), the LH surge occurred on the same day or within 1 day of the BBT nadir. For the remaining five subjects, the surge fell within 2 days after or 3 days before the nadir.
DISCUSSION
The timing of the BBT nadir correlated fairly well with the LH surge. Studies conducted by others, describ-
782
MORRIS ET AL.
ing patterns for 18 individual cycles with BBT nadirs assigned by our method, collectively give a similar pattern, although the interval to the LH surge on either side of the nadir is extended to at least 4 days. 2 ' 5 Determining the time of the BBT shift was often difficult, since the nadir sometimes dropped below the mean temperature for the preceding 6 days. Small increases in temperature in the first 1 or 2 days after the nadir might, therefore, still result in temperatures at or below the mean follicular phase temperature. In our patients, the modal (most common) rise above the respective mean follicular phase temperatures within 48 hours was only 0.1° F. According to a World Health Organization definition 11 the shift in BBT to the hyperthermic phase of the cycle occurs within a period of 48 hours or less. The definition also requires that three consecutive daily temperatures be at least O.T C higher than the previous six daily temperatures. Had we strictly adhered to these or others of the many proposed criteria for assessment of BBT curves, our high correlation with the LH surge might have declined. Certainly the number of readable cycles would have been reduced. In our study, since temperature readings and blood samples were obtained only once within each 24-hour period, the temporal correlation between BBT nadir and LH surge may actually be higher than the data indicate. The basis of this supposition is that the LH peak may have occurred as much as 12 to 18 hours before or after the time of observations. The same reasoning, however, makes it possible that the separation between nadir and surge is actually 1 day longer than it appears. The data provide some insight into the limitation of the BBT nadir in determining the time of ovulation. In our subjects, temporal localization of the BBT nadir was determined by a medical spe-
July 1976
cialist who was aware of the timing of menstrual bleeding before and after the period of interest and was able to study the temperature values for the entire cycle. Opportunity for localization of the BBT nadir, therefore, was optimal. Without this spectrum of information, particularly without knowledge of the last half of the cycle, accuracy would have undoubtedly declined. There are additional factors which might lead to an overestimation of the predictive value of BBT when this clinical tool is to be utilized under more usual circumstances. Some of our subjects were selected for blood donation because they had shown definable shifts in BBT in previously recorded menstrual cycles. Nevertheless, we could not locate the day of the BBT shift during the cycle of interest in records from at least 12 of the 43 subjects. Furthermore, even when a clear shift was obvious, had we been using BBT to predict the period of fertility, it would have taken 48 hours or more to be certain that a shift had occurred. By this time all of our subjects had already exhibited the LH surge, as had 17 of the 18 subjects whose cycles have been reported by others. The subjects therefore were partly, if not wholly, through the period of maximal fertility. For both fostering conception and preventing pregnancy, BBT is an unsatisfactory signal.
Acknowledgments. Thanks are due to Charles Chase, Ruben Maness, Doretha Foushee, and Eyvonne Bruton for technical assistance.
REFERENCES 1. Taymore ML, Aono T, Pheteplace C, Page G:
Follicle-stimulating hormone and luteinizing hormone in serum during the menstrual cycle determined by radioimmunoassay. Acta Endocrinol (Kbh) 59:298, 1968
Vol. 27, No.7
RELATIONSHIP BETWEEN BBT AND LH SURGE
2. Midgley AR Jr, Jaffe RB: Human luteinizing hormone in serum during the menstrual cycle: determination by radioimmunoassay. J Clin Endocrinol Metab 26:1375, 1966 3. Zuspan FP, Zu .pan KJ: Ovulatory plasma amine (epinephrine and norepinephrine) surge in the woman. Am J Obstet Gynecol 117:654, 1973 4. Jewelewicz R, Dyrenfurth I, Warren M, Frantz AG, Vande Wiele RL: Effect of thryotropinreleasing hormone (TRH) upon the menstrual cycle in women. J Clin Endocrinol Metab 39:387, 1974 5. Dignam WJ, Parlow AF, Coyotupa J, Honda J, Hiroi M, Kinoshita K, Kushinsky S: Plasma estrogens and serum gonadotropins in a normal menstrual cycle. Obstet Gynecol 43:484, 1974 6. Goebelsmann U, Midgley AR Jr, Jaffe RB: Regulation of human gonadotropins. VII. Daily individual urinary estrogens, pregnanediol and serum luteinizing and follicle stimu-
7.
8.
9.
10.
11.
783
lating hormones during the menstrual cycle. J Clin Endocrinol Metab 29:1222, 1969 Abraham GE, Klaiber EL: Plasma immunoreactive estrogen and LH during the menstrual cycle. Am J Obstet Gynecol 108:528, 1970 Ross GT, Cargille CM, Lipsett MB, Rayford PL, Marshall JR, Strott CA, Rodbard D: Pituitary and gonadal hormones in women during spontaneous and induced ovulatory cycles. Recent Prog Horm Res 26:1, 1970 Vande Wiele RL, Bogumil J, Dyrenfurth I, Ferin M, Jewelewicz R, Warren M, Rizkallah T, Mikhail G: Mechanisms regulating the menstrual cycle in women. Recent Prog Horm Res 26:63, 1970 Edwards RG, Steptoe PC: Induction of follicular growth, ovulation, and luteinization in the human ovary. J Reprod Fertil [Suppl] 22:121, 1975 Report of a W.H.O. Scientific Group: Biology of fertility control by periodic abstinence. WHO Tech Rep Ser 360:1, 1967