Contraception
45523-532,
1992
A CROSS-OVER STUDY OF THREE ORAL CONTRACEPTIVESCONTAINING ETI-IINYLOESTRADIOLAND EITHER DESOGESTREL OR LEVONORGESTREL
Song Si’, Chen Jun-kang’, Li La-mei’, 1 2 3 4
Fan Bao-chuan2,
Yang Pei-juan’,
He Mei-li’,
H Rekers3 and K Fotherby4
Shanghai Institute of Planned Parenthood Research, Shanghai, China Shanghai First Maternity and Infant Hospital, Shanghai, China Organon International, Oss, The Netherlands Royal Postgraduate Medical School, London, UK
Address for correspondence: Dr K Fotherby, Road, London W12 ONN, UK
Royal
Postgraduate
Medical
School,
Ducane
ABSTRACT A randomised cross-over trial was performed to compare the pharmacodynamic actions of three low-dose oral contraceptives (OCs): Marvelon (150/.&g desogestrel (DSG)+ 3Opg ethinyloestradiol (EE)), Mercilon (150/&g DSG + 2Opg EE) and Microgynon (15O/Lg levonorgestrel (LNG) + 3opg EE). None of the OCs produced any significant changes in serum cholesterol, LDL-C and apoprotein B. Triglycerides were increased by the desogestrel OCs but not by Microgynon. The latter however increased the glucose and insulin responses to a glucose tolerance test whereas Marvelon and Mercilon had no effect. HDLC increased with Marvelon, was unchanged with Me&on and was decreased with Microgynon. Apoprotein AI1 was increased by all three OCs but only the DSG OCs increased apoprotein AI. All OCs produced similar increases in caeruloplasmin but the increase in SHBG was much greater with Marvelon and Mercilon than with Microgynon. Testosterone was reduced more with Microgynon than with the DSG OCs. Many of the changes reflect the strong anti-oestrogenic action of LNG on metabolic parameters compared to DSG. Except for the effect on HDL-C, there was little difference between Marvelon and Mercilon on metabolic parameters and this complements the findings from large-scale clinical trials of the hvo OCs. Mercilon, therefora provides a very satisfactory alternative to Marvelon.
INTRODUCTION OCs are used in China by about 9 million women (1). The most commonly used formulation is the Chinese No. 1 pill (norethisterone (NET) 0.6 mg + EE 35pg) and a norgestrel pill (dl-norgestrel 3OOpg + EE 3O/lg) is also available. DSG, a newly developed gestagen, appears to have some advantages over the older ones (2). Marvelon has been widely used during the past few years in Europe and is a very acceptable and efticacious formulation (3). More recently, trials (4) have been performed with Mercilon; the preliminary results appear to be similar to those obtained with Marvelon. It seemed likely that the lower dose pill would bc better suited to Chinese women who have a lower body weight than European women. Accordingly a cross-over study was performed comparing Marvelon and Mercilon with Microgynon. Submitted Accepted
for publication for publication
Copyright
0
1992
January 6, 1992 April 17, 1992
Butterworth-Heinemann
Contraception
524
SUBJECTS AND METHODS Design of investigation The 12 healthy volunteer subjects enrolled in the study were recruited from women attending the family planning clinic of Shanghai First Maternity and Infant Hospital and who had opted to use an OC. Informed consent was obtained. None of the subjects had been pregnant in the previous six months or had used steroidal contraceptives during the three months preceding the study. They had had regular menstrual cycles during the past six months of 25 to 36 days in length and duration of menses of 3 to 7 days. No other drugs were taken during the period of study. Medical and gynaecological examinations were performed prior to entering the study. The three OCs compared (Marvelon, A; Mercilon, B and Microgynon, C) are monophasic ones. A randomized cross-over design was used. The 12 cases were divided into 6 groups, i.e.: A+B+C, A-C+B, B-A-C, B+C+A, C+A-+B and C+B+A. Thus, each group contained 2 subjects. The women took the first tablet on day 5 of a menstrual cycle and continued to take one tablet daily for 21 days followed by 7 non-pill using days. Each OC was taken for 3 consecutive months with a month without pill-taking between each change-over when barrier methods were used.
Sampling
procedures Between days 6 and 9 (follicular phase) and days 21 and 24 (luteal phase) of the pretreatment cycle, the women attended the clinic in a fasting state and a blood sample (20 ml) was taken from an antecubital vein. The blood was allowed to clot, the serum separated and stored at -2O’C until assayed for lipids, sex hormone binding globulin (SHBG), caeruloplasmin (CP) and testosterone (T). This procedure was repeated on the morning of day 21 of each treatment cycle and on day 28 of each post-treatment cycle. Further blood samples were taken at the times indicated above during the luteal phase of the pretreatment cycle, during the third cycle of each treatment period and on the 28th day after the last tablet of the final period of treatment. From each blood sample, 1 ml was mixed with sodium fluoride and potassium oxalate and used for glucose assay; a further 1 ml was mixed with EDTA and used for estimation of glycosylated haemoglobin. On each of these days, an oral glucose tolerance test (GTT) using 75g glucose was performed and blood samples (3 ml) taken at 0.5, 1, 2 and 3 h after glucose administration for glucose and insulin assay. The following methods were used: Glucose: glucose oxidase method, interassay coefficient of variation (CV) 4%. Insulin: radioimmunoassay, CV less than 10%. Glycosylated haemoglobin Al: chromatographic method using Biorex-70 resin. SHBG: method of Rosner (S), CV 7.0%. CP: method of Schosinsky et al. (6), CV 6.0%. Testosterone: radioimmunoassay using WHO kits, CV less than 10%. Triglycerides (TG) and total cholesterol(TC): enzymatic methods using kits from Boehringer, Mannheim, CV 3.1% and 3.1%, respectively. HDL-C: phosphotungstic acid and magnesium chloride precipitation method using Boehringer kits, CV 2.9%. LDL-C: polyvinyl sulphate precipitation method using Boehringer kits, CV 2.6%. Apoproteins AI, AI1 and B: rocket immuno-electrophoresis (7), CV 3.0%, 2.8% and 2.9% respectively. Statistical significance
was tested using a paired t-test,
525
Contraception
RESULTS AU except one of the 12 women completed aII three periods of OC treatment. One woman on the regime C+A+B became pregnant during the non-treatment month after taking drug C and therefore did not complete treatments A and B. Since no significant differences were observed depending on the regimen, the results for each OC from the six regimens were combined. The characteristics of the subjects are shown in Table I. Table I. Subject characteristics
(mean ? SD, range in parentheses)
32.2 1.59 52.8 20.8 97.5 60.8
Age (Y) Height (m) Weight (kg) Body mass index Systolic BP (mmHg) Diastolic BP (mmHg)
? + + ? * +
2.4 0.5 4.8 2.0 7.5 2.9
(2634) (1.50-1.70) (46-60) (18.4-25.3) (90-110) (60-70)
Carbohydrate metabolism Mean values for the serum glucose and insulin concentrations pre, during and posttreatment are shown in Table II. Fasting levels of both glucose and insulin at all times were within the normal range and there were no effects from any of the three treatments. After glucose loading, significant increases at 1 h, 2 h and 3 h (P< 0.05) both in glucose and insulin occurred during treatment with Microgynon but not with Marvelon or MerciIon. Values for total and incremental areas under the blood concentration-time curves for glucose and insulin are shown in Table III. They were consistent with the results in Table II, only Microgynon showing significant increases in both total and incremental values for both glucose and insulin (PcO.05) compared to pretreatment values. However, the ratio for total areas of insulin to glucose were not changed signiticantly. No significant changes occurred in the glycosylated haemoglobin Al levels (pretreatment 6.4 2 0.7%; Marvelon 6.1 f 0.7%; Mcrcilon 6.3 + 0.9%; Microgynon 6.3 + 0.9%; posttreatment 6.2 + 0.8%). Table II. Serum glucose (mg/dl) and serum insulin @u/ml) concentrations during oral glucose tolerance test in treatment cycle 3 (Values are mean *SD, * denotes significant difference (PcO.05) from pretreatment values)
Time (h)
0
0.5
1
2
3
Pretreatment
G 83.026.5 I 13.2~2.4
129.9215.6 73.8215.8
112.42 18.9 60.8 ? 14.2
89.2 + 13.0 42.9 ? 18.7
73.62 10.1 19.9 ? 6.2
Marvelon
G 83.3k5.8 I 14.123.7
133.2k20.5 75.3220.0
117.6+21.9 68.1+ 18.0
96.0 + 14.4 46.8+ 18.0
73.72 11.7 21.72 10.4
Mercilon
G 89.9+ 4.6 I 13.42 2.8
128.1 f 13.8 74.9 + 16.3
118.6? 15.1 64.3 * 15.5
97.7k9.0 49.2? 15.4
73.5 f 6.9 20.7t8.9
Micro-
gynon
G 85.029.1 I 14.2k3.0
134.2k11.5 86.1225.4
124.9 ? 14.5’ 72.5?17.1*
106.92 11.3* 50.92 14.5’
81.9+8.2* 29.3? 16.8*
Posttreatment
G 87.0+ 6.8 I 12.6 YL 2.5
132.1+ 18.0 70.7 + 19.7
118.8223.5 58.3% 14.8
98.62 15.9 41.6 2 16.8
73.8 t 11.8 16.6 ? 6.7
526
Contraception
Table III. Areas under the serum glucose and insulin concentration-time curves after oral * denotes significant difference (P
45523-532,
1992
A CROSS-OVER STUDY OF THREE ORAL CONTRACEPTIVESCONTAINING ETI-IINYLOESTRADIOLAND EITHER DESOGESTREL OR LEVONORGESTREL
Song Si’, Chen Jun-kang’, Li La-mei’, 1 2 3 4
Fan Bao-chuan2,
Yang Pei-juan’,
He Mei-li’,
H Rekers3 and K Fotherby4
Shanghai Institute of Planned Parenthood Research, Shanghai, China Shanghai First Maternity and Infant Hospital, Shanghai, China Organon International, Oss, The Netherlands Royal Postgraduate Medical School, London, UK
Address for correspondence: Dr K Fotherby, Road, London W12 ONN, UK
Royal
Postgraduate
Medical
School,
Ducane
ABSTRACT A randomised cross-over trial was performed to compare the pharmacodynamic actions of three low-dose oral contraceptives (OCs): Marvelon (150/.&g desogestrel (DSG)+ 3Opg ethinyloestradiol (EE)), Mercilon (150/&g DSG + 2Opg EE) and Microgynon (15O/Lg levonorgestrel (LNG) + 3opg EE). None of the OCs produced any significant changes in serum cholesterol, LDL-C and apoprotein B. Triglycerides were increased by the desogestrel OCs but not by Microgynon. The latter however increased the glucose and insulin responses to a glucose tolerance test whereas Marvelon and Mercilon had no effect. HDLC increased with Marvelon, was unchanged with Me&on and was decreased with Microgynon. Apoprotein AI1 was increased by all three OCs but only the DSG OCs increased apoprotein AI. All OCs produced similar increases in caeruloplasmin but the increase in SHBG was much greater with Marvelon and Mercilon than with Microgynon. Testosterone was reduced more with Microgynon than with the DSG OCs. Many of the changes reflect the strong anti-oestrogenic action of LNG on metabolic parameters compared to DSG. Except for the effect on HDL-C, there was little difference between Marvelon and Mercilon on metabolic parameters and this complements the findings from large-scale clinical trials of the hvo OCs. Mercilon, therefora provides a very satisfactory alternative to Marvelon.
INTRODUCTION OCs are used in China by about 9 million women (1). The most commonly used formulation is the Chinese No. 1 pill (norethisterone (NET) 0.6 mg + EE 35pg) and a norgestrel pill (dl-norgestrel 3OOpg + EE 3O/lg) is also available. DSG, a newly developed gestagen, appears to have some advantages over the older ones (2). Marvelon has been widely used during the past few years in Europe and is a very acceptable and efticacious formulation (3). More recently, trials (4) have been performed with Mercilon; the preliminary results appear to be similar to those obtained with Marvelon. It seemed likely that the lower dose pill would bc better suited to Chinese women who have a lower body weight than European women. Accordingly a cross-over study was performed comparing Marvelon and Mercilon with Microgynon. Submitted Accepted
for publication for publication
Copyright
0
1992
January 6, 1992 April 17, 1992
Butterworth-Heinemann
Contraception
524
SUBJECTS AND METHODS Design of investigation The 12 healthy volunteer subjects enrolled in the study were recruited from women attending the family planning clinic of Shanghai First Maternity and Infant Hospital and who had opted to use an OC. Informed consent was obtained. None of the subjects had been pregnant in the previous six months or had used steroidal contraceptives during the three months preceding the study. They had had regular menstrual cycles during the past six months of 25 to 36 days in length and duration of menses of 3 to 7 days. No other drugs were taken during the period of study. Medical and gynaecological examinations were performed prior to entering the study. The three OCs compared (Marvelon, A; Mercilon, B and Microgynon, C) are monophasic ones. A randomized cross-over design was used. The 12 cases were divided into 6 groups, i.e.: A+B+C, A-C+B, B-A-C, B+C+A, C+A-+B and C+B+A. Thus, each group contained 2 subjects. The women took the first tablet on day 5 of a menstrual cycle and continued to take one tablet daily for 21 days followed by 7 non-pill using days. Each OC was taken for 3 consecutive months with a month without pill-taking between each change-over when barrier methods were used.
Sampling
procedures Between days 6 and 9 (follicular phase) and days 21 and 24 (luteal phase) of the pretreatment cycle, the women attended the clinic in a fasting state and a blood sample (20 ml) was taken from an antecubital vein. The blood was allowed to clot, the serum separated and stored at -2O’C until assayed for lipids, sex hormone binding globulin (SHBG), caeruloplasmin (CP) and testosterone (T). This procedure was repeated on the morning of day 21 of each treatment cycle and on day 28 of each post-treatment cycle. Further blood samples were taken at the times indicated above during the luteal phase of the pretreatment cycle, during the third cycle of each treatment period and on the 28th day after the last tablet of the final period of treatment. From each blood sample, 1 ml was mixed with sodium fluoride and potassium oxalate and used for glucose assay; a further 1 ml was mixed with EDTA and used for estimation of glycosylated haemoglobin. On each of these days, an oral glucose tolerance test (GTT) using 75g glucose was performed and blood samples (3 ml) taken at 0.5, 1, 2 and 3 h after glucose administration for glucose and insulin assay. The following methods were used: Glucose: glucose oxidase method, interassay coefficient of variation (CV) 4%. Insulin: radioimmunoassay, CV less than 10%. Glycosylated haemoglobin Al: chromatographic method using Biorex-70 resin. SHBG: method of Rosner (S), CV 7.0%. CP: method of Schosinsky et al. (6), CV 6.0%. Testosterone: radioimmunoassay using WHO kits, CV less than 10%. Triglycerides (TG) and total cholesterol(TC): enzymatic methods using kits from Boehringer, Mannheim, CV 3.1% and 3.1%, respectively. HDL-C: phosphotungstic acid and magnesium chloride precipitation method using Boehringer kits, CV 2.9%. LDL-C: polyvinyl sulphate precipitation method using Boehringer kits, CV 2.6%. Apoproteins AI, AI1 and B: rocket immuno-electrophoresis (7), CV 3.0%, 2.8% and 2.9% respectively. Statistical significance
was tested using a paired t-test,
525
Contraception
RESULTS AU except one of the 12 women completed aII three periods of OC treatment. One woman on the regime C+A+B became pregnant during the non-treatment month after taking drug C and therefore did not complete treatments A and B. Since no significant differences were observed depending on the regimen, the results for each OC from the six regimens were combined. The characteristics of the subjects are shown in Table I. Table I. Subject characteristics
(mean ? SD, range in parentheses)
32.2 1.59 52.8 20.8 97.5 60.8
Age (Y) Height (m) Weight (kg) Body mass index Systolic BP (mmHg) Diastolic BP (mmHg)
? + + ? * +
2.4 0.5 4.8 2.0 7.5 2.9
(2634) (1.50-1.70) (46-60) (18.4-25.3) (90-110) (60-70)
Carbohydrate metabolism Mean values for the serum glucose and insulin concentrations pre, during and posttreatment are shown in Table II. Fasting levels of both glucose and insulin at all times were within the normal range and there were no effects from any of the three treatments. After glucose loading, significant increases at 1 h, 2 h and 3 h (P< 0.05) both in glucose and insulin occurred during treatment with Microgynon but not with Marvelon or MerciIon. Values for total and incremental areas under the blood concentration-time curves for glucose and insulin are shown in Table III. They were consistent with the results in Table II, only Microgynon showing significant increases in both total and incremental values for both glucose and insulin (PcO.05) compared to pretreatment values. However, the ratio for total areas of insulin to glucose were not changed signiticantly. No significant changes occurred in the glycosylated haemoglobin Al levels (pretreatment 6.4 2 0.7%; Marvelon 6.1 f 0.7%; Mcrcilon 6.3 + 0.9%; Microgynon 6.3 + 0.9%; posttreatment 6.2 + 0.8%). Table II. Serum glucose (mg/dl) and serum insulin @u/ml) concentrations during oral glucose tolerance test in treatment cycle 3 (Values are mean *SD, * denotes significant difference (PcO.05) from pretreatment values)
Time (h)
0
0.5
1
2
3
Pretreatment
G 83.026.5 I 13.2~2.4
129.9215.6 73.8215.8
112.42 18.9 60.8 ? 14.2
89.2 + 13.0 42.9 ? 18.7
73.62 10.1 19.9 ? 6.2
Marvelon
G 83.3k5.8 I 14.123.7
133.2k20.5 75.3220.0
117.6+21.9 68.1+ 18.0
96.0 + 14.4 46.8+ 18.0
73.72 11.7 21.72 10.4
Mercilon
G 89.9+ 4.6 I 13.42 2.8
128.1 f 13.8 74.9 + 16.3
118.6? 15.1 64.3 * 15.5
97.7k9.0 49.2? 15.4
73.5 f 6.9 20.7t8.9
Micro-
gynon
G 85.029.1 I 14.2k3.0
134.2k11.5 86.1225.4
124.9 ? 14.5’ 72.5?17.1*
106.92 11.3* 50.92 14.5’
81.9+8.2* 29.3? 16.8*
Posttreatment
G 87.0+ 6.8 I 12.6 YL 2.5
132.1+ 18.0 70.7 + 19.7
118.8223.5 58.3% 14.8
98.62 15.9 41.6 2 16.8
73.8 t 11.8 16.6 ? 6.7
526
Contraception
Table III. Areas under the serum glucose and insulin concentration-time curves after oral * denotes significant difference (P
