- Email: [email protected]
Plasma levels of ethinyl oestradiol and norgestrel during hormone replacement therapy
Maturitas, 2 (1980) 147-154
o Elsevier/North-Holland
147
Biomedical Press
PLASMA LEVELS OF ETHINYL OESTRADIOL
AND NORGESTREL
DURING
HORMONE REPLACEMENT THERAPY
LORRAINE DENNERSTEIN ‘, KEN FOTHERBY *, GRAHAM D. BURROWS 3, BETTY LABY 4 and CARL WOOD s r Department of Psychiatry, University of Melbourne, Melbourne; * Department of Steroid Biochemistry, Royal Postgraduate Medical School, London, U.K.; 3 Department of Psychiatry, University of Melbourne, Melbourne; 4 Department of Statistics, University of Melbourne, Melbourne; 5 Department of Obstetrics and Gynaecology, Monash University, Melbourne, Australia (Received 7 August 1979; accepted 14 April 1980)
Plasma levels of levonorgestrel and ethinyl oestradiol were estimated in six women who took part in a double blind cross-over trial of hormone replacement therapy. Over a 12 mth period, each woman received each of the following oral drugs for 3 mth: ethinyl oestradiol 50 I.tg, levonorgestrel 250 pg, combination of ethinyl oestradiol 50 c(g and levonorgestrel 250 c(g (Nordiol), and placebo. Plasma samples were collected at 2, 8 and 26 h after morning medication each month. Plasma levels of norgestrel were increased when ethinyl oestradiol was added in the Form of the combined preparation “Nordiol”. The addition of ethinyl oestradiol to norgestrel therapy also resulted in significant differences between patients in the plasma norgestrel levels. There was approximately a three-fold difference between individuals in the plasma levels of ethinyl oestradiol achieved on the same fixed dose. No simple relationship was evident between plasma levels of norgestrel or ethinyl oestradiol and any of the clinical features measured. (Key words: Ethinyl oestradiol, Norgestrel, Plasma levels)
INTRODUCTION
In order to determine the role of oestrogen and progestogen therapy at the menopause a clinical trial of these compounds was planned and has been described in detail [I 1. Briefly, 49 women who had undergone both hysterectomy and bilateral oophorectomy for benign conditions took part in a double blind cross-over trial. Over a 12 mth period, each woman received each of the following oral drugs daily for 3 mth: ethinyl oestradiol 50 pg, levonorgestrel 250 pg, combination of ethinyl oestradiol50 pg and levonorgestrel 250 pg (“Nordiol”), and placebo. The medication regimen was shortened only if there were intolerable symptoms. Drug taking was continuous, there being no drug free period between the different medications. Previous papers have reported the relationship of clinical features to plasma levels of endogenous hormones [2], the overall preference for Address for reprints: Dr. L. Dennerstein, Department P.O. Royal Melbourne Hospital, VIC, 3050, Australia.
of Psychiatry, Clinical Sciences Building, c/o
148
drugs received [I] and the effects of hormones on hot flushes [3], headaches [4], mood [5] and other somatic features [6]. The present study investigated plasma levels of norgestrel and ethinyl oestradiol and their relationship to clinical response. METHODS
Six of the 49 women volunteered to have plasma samples taken at the end of each month’s medication. Blood was collected in heparinized tubes by peripheral venipuncture at 2, 8 and 26 h following oral administration of the hormone. The usual morning dose was withheld prior to the 26 h sample. The samples were centrifuged and plasma separated and frozen in plain tubes at -2O’C. Radioimmunoassays for each hormone were performed in a single batch at the conclusion of the study. Ethinyl oestradiol
The total (free plus conjugated) ethinyl oestradiol in plasma was estimated by radioimmunoassay [7]. Antisera were raised in rabbits against ethinyl oestradiol3-hemisuccinate coupled to bovine serum albumin. The sensitivity (amount of unlabelled steroid needed to reduce percentage binding by 10%) was 8 pg. The coefficients of variation for nine replicate estimates of two plasma samples with a mean ethinyl oestradiol content of 80 and 40 pg/ml were 5.3 and 14%, respectively. Norgestrel
Norgestrel (13/3-ethyl-l7ar-ethinyl-l7/3-hydroxygon-4-on-3-one) is a potent synthetic progestational agent. There are two stereoisomeric forms of which only the d-isomer (levonorgestrel), corresponding in absolute stereochemical configuration to the natural steroid hormone, is biologically active.
Plasma levels of levonorgestrel were measured by radioimmunoassay as described by Warren and Fotherby [8] with the exception that a tritiated ligand was used instead of an iodinated one. Antibodies to levonorgestrel were raised in rabbits against the steroid 3-carboxymethoxime coupled to bovine serum albumin. Dextran-charcoal suspension was
149
used for the separation of bound and free labelled steroid. The sensitivity of the method was 50 pg levonorgestrel per ml plasma. The coefficient of variation for replicate analyses on the same sample varied from 14-16%. Clinical assessments At monthly interviews held on the same day as blood sampling, hot flushes and headache frequency over the previous month were recorded. Body weight was measured monthly. Mood was assessed by the Hamilton Rating for Depression (H.D.R.S.) [9]. Only the first 17 items on the scale were used. These items were rated on a O-2 or O-4 scale by the interviewer. The clinician also recorded on an ordinal scale women’s reports of changes in sexual desire (l-5 scale) and coital rate. RESULTS
Sample profile The six women ranged in age from 33-53 yr. Time since oophorectomy varied from l-3 yr. All of the women ceased other hormonal medication at least 2 wk prior to entering the clinical trial. Plasma levels of ethinyl oestradiol The mean plasma ethinyl oestradiol levels are listed in Tables I and II. There was no significant difference in plasma ethinyl oestradiol levels between months at any of the times of sampling. Ethinyl oestradiol levels during the placebo or norgestrel alone phases were below the sensitivity of the assay. Using the data from the 2 h samples an analysis of variance was carried out on the plasma ethinyl oestradiol assay results (Table III). No significant differences were found between ethinyl oestradiol given alone or when combined with norgestrel. There was also no significant difference between assays obtained after 1, 2 or 3 mth of therapy. There was a highly significant difference between patients. Paired t-test found no significant difference between the two medications studied in the half-lives obtained for either the periods 2-8 h (t = 1.93) or 8-26 h (t = 0.83). The relationships between plasma levels of ethinyl oestradiol and certain clinical
TABLE
I
Mean plasma
levels of ethinyl
oestradiol
(pg/ml)
after oral ethinyl
oestradiol.
Time of sampling
Number of observations Mean Standard deviation
2h
8h
26 h
18 1035 251.7
18 409 81.8
18 187 68.2
150 TABLE II Mean plasma levels of ethinyl oestradiol (pg/ml) after oral ethinyl oestradiol and norgestrel ‘Nordiol’. Time of sampling
Number of observations Mean Standard deviation
2h
8h
26 h
16 1083 151.9
16 594 52.9
16 213 63.7
Means are for 6 patients over 3 mth. (There were missing assays for 2 patients on 1 mth).
variables were plotted. The clinical features investigated were hot flush and headache frequency, the Hamilton Rating for Depression, sexual desire and coital rate. No simple relationship was evident between plasma levels of ethinyl oestradiol and any of these variables. Statistical analyses were not possible because of the limited number of observations per patient per treatment period. Plasma levels ofnorgestrel
The mean plasma norgestrel levels are listed separately for the two norgestrel containing preparations in Tables IV and V. There was no significant difference in norgestrel levels between month’s of therapy at any of the times of sampling. It is evident that the plasma levels of norgestrel were much higher when ethinyl oestradiol was taken concomitantly.
TABLE III Analysis of variance of plasma ethinyl oestradiol assays. Time of sampling 2 h Analysis of variance d.f. Between drugs Between patients Between months Drugs X patients Drugs X months Patients X months Residual Total
1
5 2 5 2 10 8 35
S.S.
M.S.
3 173 3 034 839 201 913 452 463 145 396 439 222 402 688 4 679 694
3 173 606 968 * 100 956 90 493 72 698 43 922 50 336
d.f. = degrees of freedom, S.S. = sum of squares, M.S. a mean squares, * = signiticant at 1% level. No other effects significant.
151 TABLE IV Mean plasma levels of norgestrel (pg/ml X 10m2) at different time intervals after oral norgestrel therapy. Time of sampling
Number of observations Mean Standard deviation
2h
8h
26 h
18 2318 813.3
18 904 212.4
18 460 111.1
TABLE V Mean plasma levels of norgestrel (pg/ml X 10M5)after norgestrel and ethinyl oestradiol. Time of sampling
Number of observations Mean Standard deviation
2h
8h
26 h
18 5882 571.4
18 4242 408.3
18 2860 415.9
TABLE VI Analysis of variance of plasma norgestrel assays. Time of sampling 2 h Analysis of variance
Between drugs Between patients Between months Drugs X patients Drugs X months Patients X months Residual Total
d.f.
S.S.
M.S.
1 5 2 5 2 10 10 35
114 315 300 10 254 358 97 289 8 834 421 1017 010 6 977 026 3915488 145 410 893
114315 300** 2050872* 48 645 1 766 884 * 508 505 697 703 391549
d.f. = degrees of freedom, S.S. = sum of squares, M.S. = mean squares, * = significant at 5% level, ** = significant at 1% level. No other effects significant.
152
The results of an analysis of variance are shown in Table VI. The difference in plasma levels of norgestrel was highly significant for the two preparations but was unrelated to the number of months of drug use. A separate analysis of variance for each preparation found no significant difference between patients in norgestrel levels when norgestrel was taken alone but a significant difference between patients when ethinyl oestradiol was taken concomitantly. Norgestrel levels during placebo and ethinyl oestradiol phases of the study were below the sensitivity of the assay. The relationships between plasma levels of norgestrel and certain clinical variables were assessed graphically. No simple relationship was evident between plasma levels of norgestrel and any of the following variables: hot flushes, headache frequency, Hamilton Rating for Depression, sexual desire, coital rate and body weight. In particular in this small sample there was no correlation between plasma levels and weight. There was no marked change in weight observed. The maximal weight change observed was 2 kg per month. DISCUSSION
The finding of an increase in norgestrel levels when ethinyl oestradiol is taken simultaneously concurs with previous research. Briggs [lo] reported that 50 1.18of ethinyl oestradiol increased the concentration of sex hormone binding globulin (SHBG) in Ijlasma. Subsequently Victor et al. [l l] demonstrated that SHBG was the main carrier protein for levonorgestrel and that there was a 2-6-fold rise in plasma levels of norgestrel when 50 ng of etMny1 oestradiol were taken daily. It was of interest that less patient variability in plasma levels of norgestrel was found when norgestrel was taken alone than when ethinyl oestradiol was added, reflecting a variable effect of this oestrogen. It is appreciated that there were only three points for the plotting of half-lives. Warren and Fotherby 18,121 showed that if the plasma concentrations of norgestrel after oral administration of this steroid were plotted semilogarithmically, the values fell approximately on a straight line (for the 2-8 h and 8-24 h periods) and this provided the rationale for the timing of the blood sampling in the present study. The calculated halflives of levonorgestrel for the period 2-8 h and 8-24 h were 2 h and 7 h, respectively. These values were considerably lower than those obtained in the present investigation even when norgestrel was administered alone. This difference may be accounted for by the fact that in the study of Warren and Fotherby male subjects were used as opposed to the female subjects used in the present study. The mean half-lives obtained in the present study for the period 2-8 h were 5.0 h (S.D. 0.42) after administration of norgestrel alone and 10.5 h (S.D. 2.13) for norgestrel administered with ethinyl oestradiol. For the period from 8-26 h the mean half-life of norgestrel administered alone was 18.5 h (S.D. 2.19) and for norgestrel administered with ethinyl oestradiol the mean value was 26.3 h (S.D. 6.54). Thus concomitant administration of ethinyl oestradiol caused a marked prolongation in the plasma half-life of norgestrel particularly for the period from 2-8 h. Oestrogen induced SHBG changes are the most likely explanation for this increase in the half-life of norgestrel and in the higher levels obtained in this study compared with previous studies of male volunteers.
153 A previous report of the correlation between plasma norgestrel levels and body weight [13] could not be confirmed in this study. The clinical parameters of sexual response, mood, headaches, and weight were chosen because they have been suggested to be the most frequent side effects of oral contraceptives containing norgestrel [ 141. No significant difference was evident in plasma ethinyl oestradiol levels or half-lives whether ethinyl oestradiol was taken alone or combined with norgestrel. Half-lives were calculated assuming that there was a linear decrease between 2 and 8 h and also another linear decrease from 8-26 h when the plasma ethinyl oestradiol concentration was plotted semilogarithmically against time. This rationale was based on the previous work of Fotherby [ 151. There is no way of knowing if there was a linear decrease in the particular patients studied other than by further pharmacokinetic studies of these women. The values obtained for half-lives in the present study did agree fairly closely with values obtained in previous studies. There was no significant difference between the values obtained for plasma ethinyl oestradiol levels each month suggesting that any inducing effect of ethinyl oestradiol would be mainly manifest in the first month of therapy. There was a three-fold interindividual variation in plasma levels of ethinyl oestradiol at the same oral dose of 50 pg of ethinyl oestradiol. This variation may partly explain the variation in plasma norgestrel observed when ethinyl oestradiol was administered concomitantly. CONCLUSION
The measurement of plasma levels of exogenous hormones administered to oophorectomised women found a considerable interindividual variation in plasma levels of ethinyl oestradiol and of norgestrel when ethinyl oestradiol was taken concomitantly. These findings are of interest in view of the standard clinical practice of prescribing fixed oral doses of steroids. ACKNOWLEDGEMENTS
We gratefully acknowledge the grants-in-aid for this project from Wyeth Pharmaceuticals, Organon Australia Pty. Ltd., and Allen and Hanbury’s. Many thanks to Drs. Bryan Hudson and Ken Sharpe for comments on the manuscript, to Kishore Shrimanker for the measurement of norgestrel and to James Akpoviroro for the measurement of ethinyl oestradiol. Tablets were supplied by Wyeth Pharmaceuticals. REFERENCES
[I] Dennerstein,
L., Burrows, G.D., Senior, J. and Wood, C. (1978) Hormone replacement therapy at the menopause - a double blind controlled study of women’s preferences. Aust. N.Z.J. Obstet. Gynaecol. 18,139-143. [2] Dennerstein, L., Wood, C., Hudson, B. and Burrows, G.D. (1978) Clinical features and plasma hormone levels after surgical menopause. Aust. N.Z.J. Obstet. Gynaecol. 18, 202-205. [3] Dennerstein, L., Burrows, G.D., Hyman, G. and Wood, C. (1978) Menopausal hot flushes: a double blind comparison of placebo, ethinyl oestradiol and norgestrel. Br. J. Obstet. Gynaecol. 85,852-856.
154 [4] Demrerstein, L., Laby, B., Burrows, G.D. and Hyman, G.J. (1978) Headache and sex hormone therapy. Headache 18,146-153. [5] DeMerstein, L., Burrows, G.D., Hyman, G. and Sharpe, K. (1979) Hormone therapy and affect. Maturitas 131,247-259. [6] Dennerstein, L., Burrows, G.D., Hyman, G.J. and Sharpe, K. (1979) Some clinical effects of hormone therapy in surgically castrated women. Submitted to Maturitas. [7] Akpoviroro, J. and Fotherby, K. (1980) Assay of ethinyl oestradiol in human serum. J. Steroid Biochem., in press. [8] Warren, R.J. and Fotherby, K. (1974) Radioimmunoassay of synthetic progestogens, norethisterone and norgestrel. J. Endocrinol. 62,605-608. [9] Hamilton, M.J. (1960) A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23,56-62. [lo] Briggs, M.H. (1975) Hormonal contraceptions and plasma sex hormone binding globulin. Contraception 12,149-153. [ll] Victor, A., Weiner, E. and Johansson, E.D.B. (1976) Sex hormone binding globulin: the carrier protein for d-norgestrel. J. Clin. Endocrinol. Metab. 43,244-247. [12] Warren, R.J. and Fotherby, K. (1975) Metabolism of d- and I-norgestrel in humans. Arzneim. Forsch. 25,964-965. [ 131 Victor, A. and Johansson, E.D.B. (1977) Contraceptive rings: self-administered treatment governed by bleeding. Contraception 16,137-147. [ 141 Gavin, J. (1974) Side effects of oral contraceptives containing norgestrel. In: The Second International Norgestrel Symposium, pp. 72-82. Editor: I. Fairweather. Excerpta Medica, Amsterdam. [ 151 Fotherby, K. (1974) Metabolism of synthetic steroids by animals and man. In: Pharmacological Models in Contraceptive Development, pp. 119-147. Editors: M.H. Brings and E. Diczfalusy. W.H.O. Symposium, Geneva.
o Elsevier/North-Holland
147
Biomedical Press
PLASMA LEVELS OF ETHINYL OESTRADIOL
AND NORGESTREL
DURING
HORMONE REPLACEMENT THERAPY
LORRAINE DENNERSTEIN ‘, KEN FOTHERBY *, GRAHAM D. BURROWS 3, BETTY LABY 4 and CARL WOOD s r Department of Psychiatry, University of Melbourne, Melbourne; * Department of Steroid Biochemistry, Royal Postgraduate Medical School, London, U.K.; 3 Department of Psychiatry, University of Melbourne, Melbourne; 4 Department of Statistics, University of Melbourne, Melbourne; 5 Department of Obstetrics and Gynaecology, Monash University, Melbourne, Australia (Received 7 August 1979; accepted 14 April 1980)
Plasma levels of levonorgestrel and ethinyl oestradiol were estimated in six women who took part in a double blind cross-over trial of hormone replacement therapy. Over a 12 mth period, each woman received each of the following oral drugs for 3 mth: ethinyl oestradiol 50 I.tg, levonorgestrel 250 pg, combination of ethinyl oestradiol 50 c(g and levonorgestrel 250 c(g (Nordiol), and placebo. Plasma samples were collected at 2, 8 and 26 h after morning medication each month. Plasma levels of norgestrel were increased when ethinyl oestradiol was added in the Form of the combined preparation “Nordiol”. The addition of ethinyl oestradiol to norgestrel therapy also resulted in significant differences between patients in the plasma norgestrel levels. There was approximately a three-fold difference between individuals in the plasma levels of ethinyl oestradiol achieved on the same fixed dose. No simple relationship was evident between plasma levels of norgestrel or ethinyl oestradiol and any of the clinical features measured. (Key words: Ethinyl oestradiol, Norgestrel, Plasma levels)
INTRODUCTION
In order to determine the role of oestrogen and progestogen therapy at the menopause a clinical trial of these compounds was planned and has been described in detail [I 1. Briefly, 49 women who had undergone both hysterectomy and bilateral oophorectomy for benign conditions took part in a double blind cross-over trial. Over a 12 mth period, each woman received each of the following oral drugs daily for 3 mth: ethinyl oestradiol 50 pg, levonorgestrel 250 pg, combination of ethinyl oestradiol50 pg and levonorgestrel 250 pg (“Nordiol”), and placebo. The medication regimen was shortened only if there were intolerable symptoms. Drug taking was continuous, there being no drug free period between the different medications. Previous papers have reported the relationship of clinical features to plasma levels of endogenous hormones [2], the overall preference for Address for reprints: Dr. L. Dennerstein, Department P.O. Royal Melbourne Hospital, VIC, 3050, Australia.
of Psychiatry, Clinical Sciences Building, c/o
148
drugs received [I] and the effects of hormones on hot flushes [3], headaches [4], mood [5] and other somatic features [6]. The present study investigated plasma levels of norgestrel and ethinyl oestradiol and their relationship to clinical response. METHODS
Six of the 49 women volunteered to have plasma samples taken at the end of each month’s medication. Blood was collected in heparinized tubes by peripheral venipuncture at 2, 8 and 26 h following oral administration of the hormone. The usual morning dose was withheld prior to the 26 h sample. The samples were centrifuged and plasma separated and frozen in plain tubes at -2O’C. Radioimmunoassays for each hormone were performed in a single batch at the conclusion of the study. Ethinyl oestradiol
The total (free plus conjugated) ethinyl oestradiol in plasma was estimated by radioimmunoassay [7]. Antisera were raised in rabbits against ethinyl oestradiol3-hemisuccinate coupled to bovine serum albumin. The sensitivity (amount of unlabelled steroid needed to reduce percentage binding by 10%) was 8 pg. The coefficients of variation for nine replicate estimates of two plasma samples with a mean ethinyl oestradiol content of 80 and 40 pg/ml were 5.3 and 14%, respectively. Norgestrel
Norgestrel (13/3-ethyl-l7ar-ethinyl-l7/3-hydroxygon-4-on-3-one) is a potent synthetic progestational agent. There are two stereoisomeric forms of which only the d-isomer (levonorgestrel), corresponding in absolute stereochemical configuration to the natural steroid hormone, is biologically active.
Plasma levels of levonorgestrel were measured by radioimmunoassay as described by Warren and Fotherby [8] with the exception that a tritiated ligand was used instead of an iodinated one. Antibodies to levonorgestrel were raised in rabbits against the steroid 3-carboxymethoxime coupled to bovine serum albumin. Dextran-charcoal suspension was
149
used for the separation of bound and free labelled steroid. The sensitivity of the method was 50 pg levonorgestrel per ml plasma. The coefficient of variation for replicate analyses on the same sample varied from 14-16%. Clinical assessments At monthly interviews held on the same day as blood sampling, hot flushes and headache frequency over the previous month were recorded. Body weight was measured monthly. Mood was assessed by the Hamilton Rating for Depression (H.D.R.S.) [9]. Only the first 17 items on the scale were used. These items were rated on a O-2 or O-4 scale by the interviewer. The clinician also recorded on an ordinal scale women’s reports of changes in sexual desire (l-5 scale) and coital rate. RESULTS
Sample profile The six women ranged in age from 33-53 yr. Time since oophorectomy varied from l-3 yr. All of the women ceased other hormonal medication at least 2 wk prior to entering the clinical trial. Plasma levels of ethinyl oestradiol The mean plasma ethinyl oestradiol levels are listed in Tables I and II. There was no significant difference in plasma ethinyl oestradiol levels between months at any of the times of sampling. Ethinyl oestradiol levels during the placebo or norgestrel alone phases were below the sensitivity of the assay. Using the data from the 2 h samples an analysis of variance was carried out on the plasma ethinyl oestradiol assay results (Table III). No significant differences were found between ethinyl oestradiol given alone or when combined with norgestrel. There was also no significant difference between assays obtained after 1, 2 or 3 mth of therapy. There was a highly significant difference between patients. Paired t-test found no significant difference between the two medications studied in the half-lives obtained for either the periods 2-8 h (t = 1.93) or 8-26 h (t = 0.83). The relationships between plasma levels of ethinyl oestradiol and certain clinical
TABLE
I
Mean plasma
levels of ethinyl
oestradiol
(pg/ml)
after oral ethinyl
oestradiol.
Time of sampling
Number of observations Mean Standard deviation
2h
8h
26 h
18 1035 251.7
18 409 81.8
18 187 68.2
150 TABLE II Mean plasma levels of ethinyl oestradiol (pg/ml) after oral ethinyl oestradiol and norgestrel ‘Nordiol’. Time of sampling
Number of observations Mean Standard deviation
2h
8h
26 h
16 1083 151.9
16 594 52.9
16 213 63.7
Means are for 6 patients over 3 mth. (There were missing assays for 2 patients on 1 mth).
variables were plotted. The clinical features investigated were hot flush and headache frequency, the Hamilton Rating for Depression, sexual desire and coital rate. No simple relationship was evident between plasma levels of ethinyl oestradiol and any of these variables. Statistical analyses were not possible because of the limited number of observations per patient per treatment period. Plasma levels ofnorgestrel
The mean plasma norgestrel levels are listed separately for the two norgestrel containing preparations in Tables IV and V. There was no significant difference in norgestrel levels between month’s of therapy at any of the times of sampling. It is evident that the plasma levels of norgestrel were much higher when ethinyl oestradiol was taken concomitantly.
TABLE III Analysis of variance of plasma ethinyl oestradiol assays. Time of sampling 2 h Analysis of variance d.f. Between drugs Between patients Between months Drugs X patients Drugs X months Patients X months Residual Total
1
5 2 5 2 10 8 35
S.S.
M.S.
3 173 3 034 839 201 913 452 463 145 396 439 222 402 688 4 679 694
3 173 606 968 * 100 956 90 493 72 698 43 922 50 336
d.f. = degrees of freedom, S.S. = sum of squares, M.S. a mean squares, * = signiticant at 1% level. No other effects significant.
151 TABLE IV Mean plasma levels of norgestrel (pg/ml X 10m2) at different time intervals after oral norgestrel therapy. Time of sampling
Number of observations Mean Standard deviation
2h
8h
26 h
18 2318 813.3
18 904 212.4
18 460 111.1
TABLE V Mean plasma levels of norgestrel (pg/ml X 10M5)after norgestrel and ethinyl oestradiol. Time of sampling
Number of observations Mean Standard deviation
2h
8h
26 h
18 5882 571.4
18 4242 408.3
18 2860 415.9
TABLE VI Analysis of variance of plasma norgestrel assays. Time of sampling 2 h Analysis of variance
Between drugs Between patients Between months Drugs X patients Drugs X months Patients X months Residual Total
d.f.
S.S.
M.S.
1 5 2 5 2 10 10 35
114 315 300 10 254 358 97 289 8 834 421 1017 010 6 977 026 3915488 145 410 893
114315 300** 2050872* 48 645 1 766 884 * 508 505 697 703 391549
d.f. = degrees of freedom, S.S. = sum of squares, M.S. = mean squares, * = significant at 5% level, ** = significant at 1% level. No other effects significant.
152
The results of an analysis of variance are shown in Table VI. The difference in plasma levels of norgestrel was highly significant for the two preparations but was unrelated to the number of months of drug use. A separate analysis of variance for each preparation found no significant difference between patients in norgestrel levels when norgestrel was taken alone but a significant difference between patients when ethinyl oestradiol was taken concomitantly. Norgestrel levels during placebo and ethinyl oestradiol phases of the study were below the sensitivity of the assay. The relationships between plasma levels of norgestrel and certain clinical variables were assessed graphically. No simple relationship was evident between plasma levels of norgestrel and any of the following variables: hot flushes, headache frequency, Hamilton Rating for Depression, sexual desire, coital rate and body weight. In particular in this small sample there was no correlation between plasma levels and weight. There was no marked change in weight observed. The maximal weight change observed was 2 kg per month. DISCUSSION
The finding of an increase in norgestrel levels when ethinyl oestradiol is taken simultaneously concurs with previous research. Briggs [lo] reported that 50 1.18of ethinyl oestradiol increased the concentration of sex hormone binding globulin (SHBG) in Ijlasma. Subsequently Victor et al. [l l] demonstrated that SHBG was the main carrier protein for levonorgestrel and that there was a 2-6-fold rise in plasma levels of norgestrel when 50 ng of etMny1 oestradiol were taken daily. It was of interest that less patient variability in plasma levels of norgestrel was found when norgestrel was taken alone than when ethinyl oestradiol was added, reflecting a variable effect of this oestrogen. It is appreciated that there were only three points for the plotting of half-lives. Warren and Fotherby 18,121 showed that if the plasma concentrations of norgestrel after oral administration of this steroid were plotted semilogarithmically, the values fell approximately on a straight line (for the 2-8 h and 8-24 h periods) and this provided the rationale for the timing of the blood sampling in the present study. The calculated halflives of levonorgestrel for the period 2-8 h and 8-24 h were 2 h and 7 h, respectively. These values were considerably lower than those obtained in the present investigation even when norgestrel was administered alone. This difference may be accounted for by the fact that in the study of Warren and Fotherby male subjects were used as opposed to the female subjects used in the present study. The mean half-lives obtained in the present study for the period 2-8 h were 5.0 h (S.D. 0.42) after administration of norgestrel alone and 10.5 h (S.D. 2.13) for norgestrel administered with ethinyl oestradiol. For the period from 8-26 h the mean half-life of norgestrel administered alone was 18.5 h (S.D. 2.19) and for norgestrel administered with ethinyl oestradiol the mean value was 26.3 h (S.D. 6.54). Thus concomitant administration of ethinyl oestradiol caused a marked prolongation in the plasma half-life of norgestrel particularly for the period from 2-8 h. Oestrogen induced SHBG changes are the most likely explanation for this increase in the half-life of norgestrel and in the higher levels obtained in this study compared with previous studies of male volunteers.
153 A previous report of the correlation between plasma norgestrel levels and body weight [13] could not be confirmed in this study. The clinical parameters of sexual response, mood, headaches, and weight were chosen because they have been suggested to be the most frequent side effects of oral contraceptives containing norgestrel [ 141. No significant difference was evident in plasma ethinyl oestradiol levels or half-lives whether ethinyl oestradiol was taken alone or combined with norgestrel. Half-lives were calculated assuming that there was a linear decrease between 2 and 8 h and also another linear decrease from 8-26 h when the plasma ethinyl oestradiol concentration was plotted semilogarithmically against time. This rationale was based on the previous work of Fotherby [ 151. There is no way of knowing if there was a linear decrease in the particular patients studied other than by further pharmacokinetic studies of these women. The values obtained for half-lives in the present study did agree fairly closely with values obtained in previous studies. There was no significant difference between the values obtained for plasma ethinyl oestradiol levels each month suggesting that any inducing effect of ethinyl oestradiol would be mainly manifest in the first month of therapy. There was a three-fold interindividual variation in plasma levels of ethinyl oestradiol at the same oral dose of 50 pg of ethinyl oestradiol. This variation may partly explain the variation in plasma norgestrel observed when ethinyl oestradiol was administered concomitantly. CONCLUSION
The measurement of plasma levels of exogenous hormones administered to oophorectomised women found a considerable interindividual variation in plasma levels of ethinyl oestradiol and of norgestrel when ethinyl oestradiol was taken concomitantly. These findings are of interest in view of the standard clinical practice of prescribing fixed oral doses of steroids. ACKNOWLEDGEMENTS
We gratefully acknowledge the grants-in-aid for this project from Wyeth Pharmaceuticals, Organon Australia Pty. Ltd., and Allen and Hanbury’s. Many thanks to Drs. Bryan Hudson and Ken Sharpe for comments on the manuscript, to Kishore Shrimanker for the measurement of norgestrel and to James Akpoviroro for the measurement of ethinyl oestradiol. Tablets were supplied by Wyeth Pharmaceuticals. REFERENCES
[I] Dennerstein,
L., Burrows, G.D., Senior, J. and Wood, C. (1978) Hormone replacement therapy at the menopause - a double blind controlled study of women’s preferences. Aust. N.Z.J. Obstet. Gynaecol. 18,139-143. [2] Dennerstein, L., Wood, C., Hudson, B. and Burrows, G.D. (1978) Clinical features and plasma hormone levels after surgical menopause. Aust. N.Z.J. Obstet. Gynaecol. 18, 202-205. [3] Dennerstein, L., Burrows, G.D., Hyman, G. and Wood, C. (1978) Menopausal hot flushes: a double blind comparison of placebo, ethinyl oestradiol and norgestrel. Br. J. Obstet. Gynaecol. 85,852-856.
154 [4] Demrerstein, L., Laby, B., Burrows, G.D. and Hyman, G.J. (1978) Headache and sex hormone therapy. Headache 18,146-153. [5] DeMerstein, L., Burrows, G.D., Hyman, G. and Sharpe, K. (1979) Hormone therapy and affect. Maturitas 131,247-259. [6] Dennerstein, L., Burrows, G.D., Hyman, G.J. and Sharpe, K. (1979) Some clinical effects of hormone therapy in surgically castrated women. Submitted to Maturitas. [7] Akpoviroro, J. and Fotherby, K. (1980) Assay of ethinyl oestradiol in human serum. J. Steroid Biochem., in press. [8] Warren, R.J. and Fotherby, K. (1974) Radioimmunoassay of synthetic progestogens, norethisterone and norgestrel. J. Endocrinol. 62,605-608. [9] Hamilton, M.J. (1960) A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23,56-62. [lo] Briggs, M.H. (1975) Hormonal contraceptions and plasma sex hormone binding globulin. Contraception 12,149-153. [ll] Victor, A., Weiner, E. and Johansson, E.D.B. (1976) Sex hormone binding globulin: the carrier protein for d-norgestrel. J. Clin. Endocrinol. Metab. 43,244-247. [12] Warren, R.J. and Fotherby, K. (1975) Metabolism of d- and I-norgestrel in humans. Arzneim. Forsch. 25,964-965. [ 131 Victor, A. and Johansson, E.D.B. (1977) Contraceptive rings: self-administered treatment governed by bleeding. Contraception 16,137-147. [ 141 Gavin, J. (1974) Side effects of oral contraceptives containing norgestrel. In: The Second International Norgestrel Symposium, pp. 72-82. Editor: I. Fairweather. Excerpta Medica, Amsterdam. [ 151 Fotherby, K. (1974) Metabolism of synthetic steroids by animals and man. In: Pharmacological Models in Contraceptive Development, pp. 119-147. Editors: M.H. Brings and E. Diczfalusy. W.H.O. Symposium, Geneva.