FERTILITY AND STERILITY
Vol. 53, No.1, January 1990
Copyright 1990 The American Fertility Society
Printed on acid-free paper in U.S.A.
Influence of a triphasic oral contraceptive preparation on plasma lipids and lipoproteins* Suzanne Lussier-Cacan, Ph.D.t:j:
A. Christine Nestruck, Ph.D.t Hubert Arslanian, M.D.1f Marianne Xhignesse, M.D. t
Jean Davignon, M.D.t Michael E. Kafrissen, M.D.§ Alcide Chapdelaine, M.D.1f
Clinical Research Institute of Montreal (IRCM), and Maisonneuve-Rosemont Medical Center, Montreal, Quebec, Canada
The influence of a triphasic oral contraceptive preparation on plasma lipid, lipoprotein, and apolipoprotein levels was studied in 20 women during 12 treatment cycles. Multiple blood samples representing all phases of the therapeutic cycle as well as posttherapy were obtained. Total and low-density lipoprotein (LDL) cholesterol fluctuated transiently in the earlier part of the study but after 9 and 12 cycles of therapy did not differ from baseline. Cyclic elevations in total cholesterol corresponding to changes in LDL cholesterol were noted twice. Total high-density lipoprotein (HDL) cholesterol remained remarkably stable over the entire study while HDL 2 cholesterol decreased and HDL3 cholesterol increased. Triglycerides (total and lipoprotein fractions) increased during treatment and fell to baseline levels within one posttreatment cycle. Very low-density lipoprotein (VLDL) cholesterol was also elevated during the study. Apolipoprotein (apo) AI, apo All, and apo Brose under therapy, the latter increase producing a lowered LDL cholesterol/ apo B ratio. Apolipoprotein E showed a temporary decrease early in the study but otherwise remained unchanged. Fertil Steril 53:28, 1990
Oral contraceptives (OCs) are known to effect changes in plasma lipids and in the lipoprotein profile that appear to be a function of dosage of both the estrogen and progestogen components and of the type of progestational agent as well as the formulation (monophasic versus multiphasic) used. 1-5 Several studies have been conducted concerning the influence of medium- and low-dose combined OCs, but to date, little conclusive evidence is avail-
able on the effects of the new triphasic preparations. This report describes the lipid, lipoprotein, and apolipoprotein profile of 20 young women during 12 cycles of therapy with a triphasic OC preparation. The study adhered to a strictly controlled protocol. Further, the importance of appropriately timed blood sampling to ensure proper evaluation of the results is discussed.
Received May 1, 1989; revised and accepted September 13, 1989. • Supported by Ortho Pharmaceutical CorPoration, Raritan, New Jersey. t Hyperlipidemia and Atherosclerosis Research Group (IRCM), Montreal. i Reprint requests: Suzanne Lussier-Cacan, Ph.D., Clinical Research Institute, 110 Pine Avenue West, Montreal, Que, Canada H2W 1R7. 1f Research Center, Maisonneuve-Rosemont Medical Center, Montreal. § Ortho Pharmaceutical CorP., Raritan, New Jersey.
MATERIALS AND METHODS
28
Lussier-Cacan et al.
Triphasic OCs/lipids and lipoproteins
Subjects
Their ages ranged from 18 to 38 years (mean ±standard deviation [SD]: 29.3 ± 5.8) and all were of normal weight, expressed as a body mass index (BMI) < 25.8.6 Baseline plasma glucose and insulin levels were normal. There was no excessive use of alcohol, five women (25%) were smokers, and the majority engaged in light to moderate exercise. Fertility and Sterility
The subjects had had no steroid hormonal therapy for at least 90 days before study admission and had no disorders commonly accepted as contraindications to OC therapy. They were normolipidemic with plasma cholesterol levels ranging from 106 to 216 mg/dL and triglyceride from 24 to 115 mg/dL. Subject enrollment took place within a 3month period starting in February and the protocol was completed during the month of May of the following year. The study was conducted at Maisonneuve Rosemont Medical Center in Montreal and was approved by the Institutional Ethics Committee. Subjects gave their written informed consent. Protocol
OC Preparation
This was an open 14-cycle study, which included one pretherapy cycle, 12 cycles of OC treatment, and 1 posttherapy cycle. The medication was administered as a 21-day on-medication and 7-day off-medication cyclic regimen, the first pill being taken on day 5 from the beginning of menses. The combined, triphasic preparation consisted of a constant dose of 35 J.Lg ethinyl estradiol with three sequential 7-day dosages of norethindrone at 0.5 mg, 0.75 mg, and 1.0 mg (Ortho 7/7/7, Ortho Pharmaceutical Corporation, Raritan, NJ). Blood Sampling
Blood samples (n = 16) were collected after a 12 to 14 hour fast and at time intervals that were representative of either the proliferative and the luteal phases of the pre and posttherapy cycles (samples 1, 2, 15, and 16), or of the different progestin doses and the 7-day off-medication period of the 1st and 6th therapy cycles (samples 3, 4, 5, 6, 9, 10, 11, and 12). Samples 7, 8, 13, and 14 were withdrawn on the last four days on medication of the therapy cycles 3, 4, 9, and 12. Total cholesterol (C), triglyceride (TG), high-density lipoprotein cholesterol (HDLC), total apolipoprotein (apo) B, and apo AI levels were measured in every sample. During pretherapy (samples 1 and 2) and at samples 8 (cycle 3) and 14 (cycle 12) under therapy, a complete lipoprotein evaluation was carried out including the determinations of very low-density lipoprotein cholesterol (VLDL-C) and VLDL triglycerides, low-density lipoprotein cholesterol (LDL-C), HDL2 - and HDL3C, VLDL apo B, LDL apo B, apo All, and apo E. Vol. 53, No.1, January 1990
Methods
Separation of Lipoproteins
Lipoproteins were separated under standard conditions by a combination of ultracentrifugation at d = 1.006 g/mL and heparin-manganese precipitation of the apo B-containing lipoproteins in the d = 1.006 infranatant, according to the Lipid Research Clinics protocol. 7 High-density lipoprotein3-cholesterollevels were determined after precipitation of HDL2 with dextran sulfate and the difference between total HDL-C and HDL3-C represented HDL 2-C. 8 In samples without ultracentrifugation, precipitation of the apo B-containing lipoproteins in total plasma yielded HDL-CT values. In two concurrent studies, there was a very good correlation between HDL-C values obtained after precipitation of the d = 1.006 ultracentrifugation infranatant (HDL-Cu) and those obtained from total plasma (r = 0.96; HDL-CT = 0.98 [HDLCu] + 2.33; n = 158). In the samples without ultracentrifugation, LDL-C was derived from the equation of Friedewald et al, using values for total C, TG, and HDL-C obtained from total plasma (HDL-CT): LDL-CF = Total C - (HDL-CT + TG/ 5). 9 There was a high correlation between LDL determinations made on fractions obtained by ultracentrifugation (LDL-Cu) and those obtained by the equation of Friedewald et al. (r = 0.99; LDL-CF = l[LDL-Cu] + [-0.27]; n = 158). Analytical Procedures
Plasma and lipoprotein C10 and TG 11 were measured enzymatically, on an automated analyzer (Abbott Bichromatic Analyzer 100, Abbott Laboratories, Pasadena, CA). Apo B (total and LDL apo B) 12 as well as apo AI and AII 13 were measured by electroimmunoassay. Total apo E was determined by radioimmunoassay using monoclonal antibodies.14 Glucose was measured enzymatically (Boehringer Mannheim, Mannheim, FRG) and insulin by radioimmunoassay (1251-Insulin Chromacode, BioMega Diagnostic, Montreal, Que., Canada). Statistical Analyses
The pretherapy determinations were averaged and used as baseline in the evaluation of treatment. However, for comparison of before and after therapy measurements, these samples were considered individually. One-way analysis of variance for re· peated measures on one factor was used to assess Lussier-Cacan et al.
Triphasic OCsflipids and lipoproteins
29
SAMPLE
225 ..... 200 ;:; ..... 175
....E"' ...I
0
c.: 1.1.1
1-
(I)
1.1.1 ...I
0
::c u
150 125 100 75
8
6
2
14
12
16
I
*
0
TOTAL-C
!
*
*
~ *
50
. . J ......
25 0
I
!
PRE
LDL-C (F)
i
k
00
~
&
&
HOL-C (T)
I i
2
3
4
5
6
7
6
9
10
11
12
CYCLE the significance of differences obtained with contraceptive therapy. Single variable regression analysis was used to calculate coefficients of correlation.
RESULTS Lipids and Lipoproteins
The results for total C, LDL-C, and HDL-C throughout the study are displayed graphically in Figure 1, and are provided for cycles 3 and 12 in Table 1. There were no statistically significant differences from baseline values except for transient changes that occurred in the 1st and 6th cycles. These. changes consisted of decreases in total Table 1
Figure 1 Plasma total and lipoprotein cholesterol levels during 12 cycles of OC therapy. High-density lipoprotein cholesterol was determined from precipitation on total plasma. Values for the LDLC reported in this figure were obtained by using the equation of Friedewald et al. 9 Significant difference from baseline: *P < 0.001; +P < 0.01; •P
Vol. 53, No.1, January 1990
Copyright 1990 The American Fertility Society
Printed on acid-free paper in U.S.A.
Influence of a triphasic oral contraceptive preparation on plasma lipids and lipoproteins* Suzanne Lussier-Cacan, Ph.D.t:j:
A. Christine Nestruck, Ph.D.t Hubert Arslanian, M.D.1f Marianne Xhignesse, M.D. t
Jean Davignon, M.D.t Michael E. Kafrissen, M.D.§ Alcide Chapdelaine, M.D.1f
Clinical Research Institute of Montreal (IRCM), and Maisonneuve-Rosemont Medical Center, Montreal, Quebec, Canada
The influence of a triphasic oral contraceptive preparation on plasma lipid, lipoprotein, and apolipoprotein levels was studied in 20 women during 12 treatment cycles. Multiple blood samples representing all phases of the therapeutic cycle as well as posttherapy were obtained. Total and low-density lipoprotein (LDL) cholesterol fluctuated transiently in the earlier part of the study but after 9 and 12 cycles of therapy did not differ from baseline. Cyclic elevations in total cholesterol corresponding to changes in LDL cholesterol were noted twice. Total high-density lipoprotein (HDL) cholesterol remained remarkably stable over the entire study while HDL 2 cholesterol decreased and HDL3 cholesterol increased. Triglycerides (total and lipoprotein fractions) increased during treatment and fell to baseline levels within one posttreatment cycle. Very low-density lipoprotein (VLDL) cholesterol was also elevated during the study. Apolipoprotein (apo) AI, apo All, and apo Brose under therapy, the latter increase producing a lowered LDL cholesterol/ apo B ratio. Apolipoprotein E showed a temporary decrease early in the study but otherwise remained unchanged. Fertil Steril 53:28, 1990
Oral contraceptives (OCs) are known to effect changes in plasma lipids and in the lipoprotein profile that appear to be a function of dosage of both the estrogen and progestogen components and of the type of progestational agent as well as the formulation (monophasic versus multiphasic) used. 1-5 Several studies have been conducted concerning the influence of medium- and low-dose combined OCs, but to date, little conclusive evidence is avail-
able on the effects of the new triphasic preparations. This report describes the lipid, lipoprotein, and apolipoprotein profile of 20 young women during 12 cycles of therapy with a triphasic OC preparation. The study adhered to a strictly controlled protocol. Further, the importance of appropriately timed blood sampling to ensure proper evaluation of the results is discussed.
Received May 1, 1989; revised and accepted September 13, 1989. • Supported by Ortho Pharmaceutical CorPoration, Raritan, New Jersey. t Hyperlipidemia and Atherosclerosis Research Group (IRCM), Montreal. i Reprint requests: Suzanne Lussier-Cacan, Ph.D., Clinical Research Institute, 110 Pine Avenue West, Montreal, Que, Canada H2W 1R7. 1f Research Center, Maisonneuve-Rosemont Medical Center, Montreal. § Ortho Pharmaceutical CorP., Raritan, New Jersey.
MATERIALS AND METHODS
28
Lussier-Cacan et al.
Triphasic OCs/lipids and lipoproteins
Subjects
Their ages ranged from 18 to 38 years (mean ±standard deviation [SD]: 29.3 ± 5.8) and all were of normal weight, expressed as a body mass index (BMI) < 25.8.6 Baseline plasma glucose and insulin levels were normal. There was no excessive use of alcohol, five women (25%) were smokers, and the majority engaged in light to moderate exercise. Fertility and Sterility
The subjects had had no steroid hormonal therapy for at least 90 days before study admission and had no disorders commonly accepted as contraindications to OC therapy. They were normolipidemic with plasma cholesterol levels ranging from 106 to 216 mg/dL and triglyceride from 24 to 115 mg/dL. Subject enrollment took place within a 3month period starting in February and the protocol was completed during the month of May of the following year. The study was conducted at Maisonneuve Rosemont Medical Center in Montreal and was approved by the Institutional Ethics Committee. Subjects gave their written informed consent. Protocol
OC Preparation
This was an open 14-cycle study, which included one pretherapy cycle, 12 cycles of OC treatment, and 1 posttherapy cycle. The medication was administered as a 21-day on-medication and 7-day off-medication cyclic regimen, the first pill being taken on day 5 from the beginning of menses. The combined, triphasic preparation consisted of a constant dose of 35 J.Lg ethinyl estradiol with three sequential 7-day dosages of norethindrone at 0.5 mg, 0.75 mg, and 1.0 mg (Ortho 7/7/7, Ortho Pharmaceutical Corporation, Raritan, NJ). Blood Sampling
Blood samples (n = 16) were collected after a 12 to 14 hour fast and at time intervals that were representative of either the proliferative and the luteal phases of the pre and posttherapy cycles (samples 1, 2, 15, and 16), or of the different progestin doses and the 7-day off-medication period of the 1st and 6th therapy cycles (samples 3, 4, 5, 6, 9, 10, 11, and 12). Samples 7, 8, 13, and 14 were withdrawn on the last four days on medication of the therapy cycles 3, 4, 9, and 12. Total cholesterol (C), triglyceride (TG), high-density lipoprotein cholesterol (HDLC), total apolipoprotein (apo) B, and apo AI levels were measured in every sample. During pretherapy (samples 1 and 2) and at samples 8 (cycle 3) and 14 (cycle 12) under therapy, a complete lipoprotein evaluation was carried out including the determinations of very low-density lipoprotein cholesterol (VLDL-C) and VLDL triglycerides, low-density lipoprotein cholesterol (LDL-C), HDL2 - and HDL3C, VLDL apo B, LDL apo B, apo All, and apo E. Vol. 53, No.1, January 1990
Methods
Separation of Lipoproteins
Lipoproteins were separated under standard conditions by a combination of ultracentrifugation at d = 1.006 g/mL and heparin-manganese precipitation of the apo B-containing lipoproteins in the d = 1.006 infranatant, according to the Lipid Research Clinics protocol. 7 High-density lipoprotein3-cholesterollevels were determined after precipitation of HDL2 with dextran sulfate and the difference between total HDL-C and HDL3-C represented HDL 2-C. 8 In samples without ultracentrifugation, precipitation of the apo B-containing lipoproteins in total plasma yielded HDL-CT values. In two concurrent studies, there was a very good correlation between HDL-C values obtained after precipitation of the d = 1.006 ultracentrifugation infranatant (HDL-Cu) and those obtained from total plasma (r = 0.96; HDL-CT = 0.98 [HDLCu] + 2.33; n = 158). In the samples without ultracentrifugation, LDL-C was derived from the equation of Friedewald et al, using values for total C, TG, and HDL-C obtained from total plasma (HDL-CT): LDL-CF = Total C - (HDL-CT + TG/ 5). 9 There was a high correlation between LDL determinations made on fractions obtained by ultracentrifugation (LDL-Cu) and those obtained by the equation of Friedewald et al. (r = 0.99; LDL-CF = l[LDL-Cu] + [-0.27]; n = 158). Analytical Procedures
Plasma and lipoprotein C10 and TG 11 were measured enzymatically, on an automated analyzer (Abbott Bichromatic Analyzer 100, Abbott Laboratories, Pasadena, CA). Apo B (total and LDL apo B) 12 as well as apo AI and AII 13 were measured by electroimmunoassay. Total apo E was determined by radioimmunoassay using monoclonal antibodies.14 Glucose was measured enzymatically (Boehringer Mannheim, Mannheim, FRG) and insulin by radioimmunoassay (1251-Insulin Chromacode, BioMega Diagnostic, Montreal, Que., Canada). Statistical Analyses
The pretherapy determinations were averaged and used as baseline in the evaluation of treatment. However, for comparison of before and after therapy measurements, these samples were considered individually. One-way analysis of variance for re· peated measures on one factor was used to assess Lussier-Cacan et al.
Triphasic OCsflipids and lipoproteins
29
SAMPLE
225 ..... 200 ;:; ..... 175
....E"' ...I
0
c.: 1.1.1
1-
(I)
1.1.1 ...I
0
::c u
150 125 100 75
8
6
2
14
12
16
I
*
0
TOTAL-C
!
*
*
~ *
50
. . J ......
25 0
I
!
PRE
LDL-C (F)
i
k
00
~
&
&
HOL-C (T)
I i
2
3
4
5
6
7
6
9
10
11
12
CYCLE the significance of differences obtained with contraceptive therapy. Single variable regression analysis was used to calculate coefficients of correlation.
RESULTS Lipids and Lipoproteins
The results for total C, LDL-C, and HDL-C throughout the study are displayed graphically in Figure 1, and are provided for cycles 3 and 12 in Table 1. There were no statistically significant differences from baseline values except for transient changes that occurred in the 1st and 6th cycles. These. changes consisted of decreases in total Table 1
Figure 1 Plasma total and lipoprotein cholesterol levels during 12 cycles of OC therapy. High-density lipoprotein cholesterol was determined from precipitation on total plasma. Values for the LDLC reported in this figure were obtained by using the equation of Friedewald et al. 9 Significant difference from baseline: *P < 0.001; +P < 0.01; •P
