Sage Journals: Discover world-class research

Abstract

This study was undertaken to evaluate the impact of progestins as part of low-estrogen (ethinyl estradiol [EE2] ≤35 μg) combined oral contraceptives (COCs) on hemostatic variables. One hundred ninety-five healthy women took oral contraceptives with following formulations: 35 EE2/norgestimate (NGM), 35 EE2/cyproterone acetate, 35 EE2/norethisterone, 30 EE2/levonorgestrel, 30 EE2/drospirenone (DRSP), 20 EE2/gestodene, and 20 EE2/DRSP, for 6 months. Hemostatic assays (prothrombin time, activated partial thromboplastin time, fibrinogen, resistance to activated protein C ratio, protein C, protein S, factor VIII [FVIII], antithrombin, plasminogen, α2-antiplasmin, inhibitor of plasminogen activator type 1 [PAI-1] and d-dimers) were performed in 3 time points: at baseline, after 3, and 6 cycles. For each formulation, results were compared according to baseline values, intergroup analysis, and the amount of estrogen or progestin component. Most of the variables were changed except FVIII. Significant difference between oral contraceptives was found in antithrombin, protein C, protein S activities, and PAI-1 values, but changes were mostly within reference range.

Keywords

blood coagulation factors clinical pharmacology gynecology and obstetrics

Introduction

There are numerous combined oral contraceptives (COCs) present at the market. Past studies established that COCs with estrogen component lower than 50 μg of ethinyl estradiol (EE2) have lower risk of venous thromboembolic events (VTEs).^1,2 Consequently, new forms of COCs were developed containing 20 to 35 μg EE2 and new progestins that can provide effective contraception with less breakthrough hemorrhaging. Today, the fourth generation of progestins (drospirenone [DRSP]) is used in parallel with first (cyproterone acetate [CPA], norethisterone [NET]), second (levonorgestrel [LNG]), and third generation (gestodene [GSD], norgestimate [NGM], desogestrel [DSG]). However, the risk of VTE, although reduced, is still present because progestins alone or in combination with estrogen are independent risk factor.³ Estimated odd ratios for the risk of VTE in COCs users differ for progestins from 3.6 for LNG to 7.3 for DSG, compared with nonusers.⁴ The effects of progestins, other than pregnancy prevention, are used to treat acne vulgaris, ovarian cysts, metrorrhagias, premenstrual symptoms, regulation of menstrual cycles, and so on.⁵ Nevertheless, pharmacokinetics of progestins is not clear, despite their wide use. Most studies have examined progestins in COCs. It is determined that the activity of progestins is mostly dependant on type, dosage, receptor affinity (toward estrogen, progesterone, androgenic, glucocorticoid, and mineralocorticoid receptors), distribution and storage in tissues (especially fat), manner of intake (orally or parenterally), and liver first-pass effect. Cumulative effect is intrapersonal, interpersonal, and ethnic diversity in response to progestins.^6

–9

After lowering EE2, number of studies showed controversial results for the COCs of second and third generation—some pointed the absence of significant difference in thrombogenic tests in the risk of VTE,¹⁰ but others explained, procoagulatory impact on hemostasis and therefore increased risk of VTE, through biological activity of progestins.¹¹ One of the possible ways is affinity for binding sex hormone binding globulin (SHBG). Specifically, estrogens cause increase in SHBG levels, while progestins do the opposite—decrease in SHBG levels. Therefore, it was suggested that the hemostatic effect of COCs, as cause for increased VTE risk, are presented with sum of estrogenic (from EE2) and antiestrogenic (from progestins) characteristics.¹²

In this work, the aim is to determine the effect of COCs with ≤35 μg EE2 and progestins: CPA (2 mg), triphasic NET (50/75/100 μg), LNG (150 μg), GSD (75 μg), NGM (25 μg), and DRSP (3 mg) on hemostasis using tests of coagulation and fibrinolytic pathway, in accordance to the consent committee conclusion that COC with minimal hemostatic changes is considered best option for women¹³ and therefore the least risk for VTE development.

Materials and methods

Study Participants

The study included 227 women from the northern part of Croatia, referred by their gynecologists for blood testing on voluntary basis to Cakovec County Hospital, in the period from June 2009 to February 2011. Women should be at least 3 months free from COC use; a parental consent was required for underage participants. Dropout rate was 14%, so 195 participants were included in per protocol analysis. The mean age of the study participants was 23.51 (range 15-43) years, mean body mass index [BMI] was 21.8, and 58 (30%) of the participants were smokers.

Study Design

Study was performed in 3 time points in prospective manner—baseline (before taking of COCs), after 3 and 6 cycles. The gynecologists suggested several COCs from the market to the participants who selected one for use. The participants were asked to fill out a standardized questionnaire concerning their medical history, body height and weight were measured, and blood samples were obtained for laboratory testing. After the third and sixth cycle, participants stated which COC was used.

The following hemostatic assays were performed in all 3 time points: prothrombin time (PT), activated partial thromboplastin time (aPTT), fibrinogen, resistance to activated protein C ratio (aPCR), protein C activity, protein S activity, factor VIII (FVIII) activity, antithrombin activity, plasminogen activity, α2-antiplasmin activity, inhibitor of plasminogen activator type 1 (PAI-1), and d-dimers. Only routine tests, that are in common use by clinicians, were chosen to assess hemostasis to gain more practical value of this study.

Study protocol was approved by the ethics committee of Faculty of Pharmacy and biochemistry, University of Zagreb and by the local hospital ethics committee. An informed consent in writing was obtained from all study participants.

Samples

Blood samples were collected between 8.00 and 10.00 am after overnight fast from antecubital vein into 2 plastic tubes containing 1/10 volume of 3.2% sodium citrate and thoroughly mixed several times before double centrifugation (10 min at 2500g). Within 1 hour, 1-mL aliquots of platelet-depleted plasma for further testing in series were frozen and stored in plastic tubes (Eppendorf micro tubes) at −25°C for a maximum of 4 weeks. Before testing, they were thawed at 37°C within 10 minutes.

Assays for Hemostatic Variables

All tests were performed on a Behring Coagulation System (BCS) automated analyzer using reagents manufactured by Siemens Healthcare Diagnostics, Germany, as follows: PT (Thromborel S, reference range 10-15 seconds), aPTT (Actin FS, reference range 23.0-31.9 seconds) and fibrinogen (Multifibren U, reference range 2.1-4.0 g/L) were determined in fresh citrated plasma; FVIII activity (FVIII chromogenic assay, reference range 50%-150%), aPCR ratio (ProC AcR, reference value ratio >1.8), protein C activity (Berichrom Protein C, reference range 70%-140%), protein S activity (Protein S Ac, reference range 59%-118%), and antithrombin activity (Berichrom ATIII reference range 75%-125%) were determined in thawed plasma; d-dimers [d-dimer Innovance, reference value <550 μg/L Fibrinogen equivalent unit (FEU)] were determined routinely in fresh citrated plasma; plasminogen (plasminogen, reference range 75%-150%), α2-antiplasmin (Berichrom α2-antiplasmin, reference range 80%-120%) and PAI-1 (Berichrom PAI, reference range 2.0-7.0 U/mL) were determined in thawed plasma.

Statistical Analysis

For statistical analysis MedCalc version 12.2.1.0 statistical software was used. Significance for all hemostatic variables were analyzed by COCs group with Friedman analysis of variance (ANOVA) test, overall intergroup analysis was performed with Kruskal-Wallis test, and comparison of 2 COC groups with Mann-Whitney rank sum test.Determined significance level was P < .05. Relative change percentage was calculated according to formula (cycle value − baseline value)/baseline value × 100.

Results

Initial interview and sampling was carried out for 227 women, but 32 of them discontinued the study. Reasons are 21 women prematurely discontinued COCs without any additional explanation, 4 women did not take any dose of oral contraceptives because they changed their minds, 6 women did not follow the protocol correctly (eg, did not present for testing on time), and 1 woman became pregnant before starting COCs. A total of 195 participants finished the whole study.

Study participants were comparable for age and BMI, which did not change significantly by the end of study.

Results are presented as values for hemostatic tests grouped according to COC formulations: 35 EE2/NGM, 35 EE2/CPA, 35 EE2/NET, 30 EE2/LNG, 30 EE2/DRSP, 20 EE2/GSD, and 20 EE2/DRSP. Intragroup and intergroup analysis was performed. Because only a small number of participants have taken 35 EE2/CPA, their results are shown in tables but not commented.

General Tests

The PT values showed a decrease, and they were significantly changed in all groups after third and sixth cycle. Relative change was the highest at 6% in 30 EE2/LNG group after third cycle, but only in the 35 EE2/NET group the values continued to decrease after 6 cycles. Values for aPTT test were also decreased and significantly changed after 3 cycles in all groups, but only 20 EE2/GSD, 35 EE2/NET, and 30 EE2/LNG groups kept the significance of change after sixth cycle. Other groups showed tendency toward baseline values (Table 1).

Table 1.

Assay Values for General Tests (PT, aPTT, and Fibrinogen) in 3 Measurement Points—Baseline, After 3, and 6 Cycles.

Type of COCs	N	Baseline	Cycle 3	Cycle 6
Type of COCs	N	(Mean ± Standard Deviation)
PT, seconds
35 EE2/NGM	52	12.34 ± 0.65	11.71 ± 0.75^a	11.70 ± 0.90^a
35 EE2/CPA	9	12.38 ± 0.68	11.91 ± 0.73	11.64 ± 0.75^a
20 EE2/GSD	35	12.55 ± 0.70	11.89 ± 0.73^a	11.90 ± 0.72^a
35 EE2/NET	30	12.51 ± 0.56	11.75 ± 0.69^a	11.66 ± 0.85^a
30 EE2/LNG	17	12.42 ± 0.78	11.66 ± 0.71^a	11.86 ± 1.03^a
30 EE2/DRSP	24	12.33 ± 0.66	11.61 ± 0.60^a	11.80 ± 0.78^a
20 EE2/DRSP	28	12.48 ± 0.86	11.76 ± 0.72^a	11.73 ± 0.81^a
aPTT, seconds
35 EE2/NGM	52	28.04 ± 1.87	27.01 ± 1.87^a	27.49 ± 2.28
35 EE2/CPA	9	29.21 ± 2.95	27.63 ± 3.05	27.68 ± 2.47
20 EE2/GSD	35	28.94 ± 2.93	26.73 ± 2.60^a	27.09 ± 2.96^a
35 EE2/NET	30	29.28 ± 1.95	27.49 ± 1.76^a	28.06 ± 2.24^a
30 EE2/LNG	17	28.55 ± 2.46	26.12 ± 2.39^a	27.27 ± 3.21^a
30 EE2/DRSP	24	28.38 ± 2.12	26.39 ± 2.30^a	27.34 ± 2.81
20 EE2/DRSP	28	28.92 ± 2.06	27.13 ± 0.08^a	28.34 ± 2.81
Fibrinogen (g/L)
35 EE2/NGM	52	2.86 ± 0.42	3.03 ± 0.47	3.04 ± 0.74
35 EE2/CPA	9	2.74 ± 0.59	2.80 ± 0.41	2.78 ± 0.46
20 EE2/GSD	35	2.81 ± 0.52	3.13 ± 0.52^a	3.27 ± 0.80^a
35 EE2/NET	30	2.62 ± 0.50	2.83 ± 0.59^a	2.78 ± 0.62
30 EE2/LNG	17	2.92 ± 0.57	3.16 ± 0.63	3.14 ± 0.64
30 EE2/DRSP	24	2.79 ± 0.42	3.04 ± 0.54	3.04 ± 0.62
20 EE2/DRSP	28	2.83 ± 0.42	3.04 ± 0.44	3.02 ± 0.58

Abbreviations: aPPT, activated partial thromboplastin time; COC, combined oral contraceptive; N, number of participants; PT, prothrombin time; EE2, ethinyl estradiol; DRSP, drospirenone; NGM, norgestimate; CPA, cyproterone acetate; GSD, gestodene; LNG, levonorgestrel; NET, norethisterone.

^a P < .05 vs baseline values.

Results for fibrinogen vary between groups. Significant increase after 3 cycles was shown in groups 35 EE2/NET and 20 EE2/GSD, of which only one had significance even after 6 cycles.

Coagulation Pathway Tests

Antithrombin results showed slight but notable decrease in only 2 groups—in 20 EE2/GSD group at both the time points and in 35 EE2/NGM group after the sixth cycle. All other groups of COCs did not present significant change during time of the study (Table 2).

Table 2.

Assay Values for Coagulation Pathway Tests (Antithrombin, Protein C, Protein S, aPCR, and FVIII) in 3 Measurement Points—Baseline, After 3, and 6 cycles.

Type of COCs	N	Baseline	Cycle 3	Cycle 6
Type of COCs	N	(Mean ± Standard Deviation)
Antithrombin (%)
35 EE2/NGM	52	101.42 ± 7.52	98.48 ± 10.03	95.91 ± 9.52^a
35 EE2/CPA	9	100.87 ± 8.56	103.16 ± 8.13	92.93 ± 9.79
20 EE2/GSD	35	104.86 ± 8.71	101.03 ± 9.15^a	100.91 ± 11.37^a
35 EE2/NET	30	99.66 ± 8.04	99.98 ± 10.24	96.10 ± 12.59
30 EE2/LNG	17	102.85 ± 8.24	97.54 ± 11.17	100.27 ± 12.39
30 EE2/DRSP	24	102.69 ± 9.36	100.55 ± 10.55	97.59 ± 7.76
20 EE2/DRSP	28	104.13 ± 10.43	101.44 ± 10.41	102.77 ± 9.55
Protein C (%)
35 EE2/NGM	52	99.9 ± 15.89	114.75 ± 16.59^a	112.57 ± 17.43^a
35 EE2/CPA	9	108.92 ± 18.52	117.79 ± 29.81	122.67 ± 31.50
20 EE2/GSD	35	107.06 ± 18.19	117.02 ± 16.54^a	117.58 ± 13.88^a
35 EE2/NET	30	100.37 ± 19.01	107.79 ± 16.73^a	108.30 ± 17.33^a
30 EE2/LNG	17	105.49 ± 15.28	117.59 ± 20.75^a	114.85 ± 21.33^a
30 EE2/DRSP	24	106.70 ± 20.82	124.60 ± 18.77^a	122.59 ± 20.95^a
20 EE2/DRSP	28	102.21 ± 15.44	123.15 ± 20.06^a	127.77 ± 21.63^a
Protein S (%)
35 EE2/NGM	52	90.74 ± 15.29	68.23 ± 13.59^a	69.28 ± 14.00^a
35 EE2/CPA	9	83.88 ± 12.87	69.69 ± 11.57^a	65.10 ± 7.64^a
20 EE2/GSD	35	91.51 ± 15.29	79.38 ± 13.95^a	84.63 ± 19.54^a
35 EE2/NET	30	88.57 ± 15.61	69.92 ± 14.97^a	74.62 ± 22.47^a
30 EE2/LNG	17	93.85 ± 16.87	82.56 ± 8.67	83.42 ± 18.36^a
30 EE2/DRSP	24	88.90 ± 14.26	65.98 ± 19.13^a	73.79 ± 16.32^a
20 EE2/DRSP	28	90.45 ± 12.70	67.24 ± 17.56^a	70.74 ± 16.76^a
aPCR (ratio)
35 EE2/NGM	52	4.19 ± 1.01	3.61 ± 0.88^a	3.51 ± 0.85^a
35 EE2/CPA	9	4,02 ± 1.02	3.44 ± 0.63	3.69 ± 0.98
20 EE2/GSD	35	4.36 ± 0.97	3.80 ± 0.85^a	3.89 ± 1.04^a
35 EE2/NET	30	4.07 ± 1.26	3.33 ± 1.07^a	3.46 ± 1.59^a
30 EE2/LNG	17	3.92 ± 1.05	3.62 ± 0.79	3.37 ± 0.74^a
30 EE2/DRSP	24	4.39 ± 1.02	3.63 ± 0.88^a	3.57 ± 0.71^a
20 EE2/DRSP	28	4.48 ± 1.08	3.86 ± 0.93^a	3.50 ± 0.97^a
FVIII (%)
35 EE2/NGM	52	107.28 ± 39.03	107.49 ± 39.17	110.05 ± 42.98
35 EE2/CPA	9	92.77 ± 35.46	102.44 ± 39.74	97.26 ± 24.89
20 EE2/GSD	35	95.27 ± 39.91	89.89 ± 22.47	96.25 ± 32.76
35 EE2/NET	30	97.54 ± 32.99	103.89 ± 27.05	99.92 ± 29.65
30 EE2/LNG	17	84.45 ± 23.96	88.08 ± 21.09	85.71 ± 23.65
30 EE2/DRSP	24	95.93 ± 31.43	102.33 ± 37.09	97.79 ± 33.01
20 EE2/DRSP	28	90.90 ± 29.50	103.23 ± 29.36	103.46 ± 40.69

Abbreviations: N, number of participants; aPCR, resistance to activated protein C ratio; FVIII, factor VIII; COC, combined oral contraceptive; EE2, ethinyl estradiol; DRSP, drospirenone; NGM, norgestimate; CPA, cyproterone acetate; GSD, gestodene; LNG, levonorgestrel; NET, norethisterone.

^a P < .05 vs baseline values.

Values for protein C activity revealed that all groups of COCs are changed. Elevated values were the most prominently changed for COCs containing DRSP 17% to 20%. Similar are the data for protein S, but here decreased values by 25% were present in COCs with DRSP and NGM. Another difference was shown by 30 EE2/LNG group, which gained significant result after sixth cycle comparing to the baseline time point.

Activated PCR ratio was measured by aPTT-based test. In study it showed significant difference in decreased values for all groups after 3 cycles, except 30 EE2/LNG, which reached significance after sixth cycle, and it had the lowest change among groups by 14%. All COC formulations further lowered values, measured at time point after sixth cycle, except 35 EE2/NET and 20 EE2/GSD groups. For the FVIII activity, none of the groups showed notable increase in values in either time point.

Fibrinolytic Pathway Tests

Increase values were recorded in all COC groups for plasminogen activity, and they stayed significantly changed after 6 cycles. All changes were greater than 25%, but the most prominent was 37% change in 20 EE2/DSP group, which further increased after sixth cycle to 44% (Table 3).

Table 3.

Assay Values for Fibrinolytic Pathway Tests (Plasminogen, d-dimers, α2-antiplasmin, inhibitor of PAI-1) in 3 measurement points—Baseline, After 3, and 6 Cycles.

Type of COCs	N	Baseline	Cycle 3	Cycle 6
Type of COCs	N	(Mean ± Standard Deviation)
Plasminogen (%)
35 EE2/NGM	52	92.86 ± 8.68	123.71 ± 14.50^a	126.08 ± 17.17 ^a
35 EE2/CPA	9	96.38 ± 4.59	118.43 ± 24.91^a	122.67 ± 17.96^a
20 EE2/GSD	35	65.85 ± 14.13	116.46 ± 17.15^a	118.73 ± 15.30^a
35 EE2/NET	30	91.92 ± 8.48	121.20 ± 18.57^a	127.50 ± 18.83^a
30 EE2/LNG	17	94.85 ± 8.57	124.51 ± 16.13^a	122.79 ± 14.13^a
30 EE2/DRSP	24	97.33 ± 11.60	125.69 ± 13.13^a	120.72 ± 19.06^a
20 EE2/DRSP	28	93.30 ± 10.39	127.88 ± 16.37^a	134.74 ± 18.71^a
d-dimers (μg/L FEU)
35 EE2/NGM	52	258.09 ± 137.63	433.99 ± 328.69^a	358.22 ± 292.51^a
35 EE2/CPA	9	207.31 ± 73.70	363 ± 233.46	315.78 ± 188.97
20 EE2/GSD	35	217.52 ± 83.33	350.92 ± 262.59^a	339.58 ± 257.69^a
35 EE2/NET	30	248.12 ± 271.57	421.29 ± 415.79^a	375.06 ± 379.09^a
30 EE2/LNG	17	215.65 ± 75.44	474.92 ± 332.11^a	360.80 ± 215.68^a
30 EE2/DRSP	24	219.99 ± 120.09	408.32 ± 372.84^a	416.33 ± 381.69^a
20 EE2/DRSP	28	212.76 ± 128.98	446.79 ± 359.29^a	445.12 ± 275.62^a
α2-antiplasmin (%)
35 EE2/NGM	52	102.79 ± 8.40	108.55 ± 8.68^a	105.48 ± 7.75^a
35 EE2/CPA	9	101.30 ± 4.77	107.74 ± 12.14^a	107.50 ± 10.61^a
20 EE2/GSD	35	105.88 ± 11.16	106.11 ± 9.59	108.91 ± 10.34
35 EE2/NET	30	101.74 ± 10.54	109.25 ± 11.05^a	111.28 ± 10.46^a
30 EE2/LNG	17	104.89 ± 9.24	107.46 ± 10.35	107.88 ± 8.35
30 EE2/DRSP	24	102.48 ± 10.03	109.22 ± 9.78^a	108.89 ± 7.97^a
20 EE2/DRSP	28	101.80 ± 8.67	109.06 ± 7.16^a	111.57 ± 6.61^a
PAI-1 (U/mL)
35 EE2/NGM	52	4.12 ± 1.05	3.89 ± 1.25	3.44 ± 0.87
35 EE2/CPA	9	3.89 ± 1.04	3.77 ± 1.40	3.33 ± 0.76
20 EE2/GSD	35	4.68 ± 1.31	4.43 ± 1.44	4.16 ± 1.08
35 EE2/NET	30	4.09 ± 1.08	3.63 ± 1.16	3.36 ± 0.97^a
30 EE2/LNG	17	4.09 ± 0.95	3.38 ± 0.86^a	3.54 ± 0.87
30 EE2/DRSP	24	4.14 ± 1.13	4.10 ± 0.79	3.57 ± 0.75
20 EE2/DRSP	28	4.32 ± 1.40	3.47 ± 0.88^a	3.58 ± 1.07^a

Abbreviations: N, number of participants; PAI-1, inhibitor plasminogen activator type 1; COC, combined oral contraceptive; EE2, ethinyl estradiol; DRSP, drospirenone; NGM, norgestimate; CPA, cyproterone acetate; GSD, gestodene; LNG, levonorgestrel; NET, norethisterone.

^a P < .05 vs baseline values.

d-dimers values expressed the highest percentage of increased relative change among all assays. All groups changed significantly after third and sixth cycle with regard to the baseline values. Changes were increased by 61% for 20 EE2/GSD after third cycle until increase by 110% in 20 EE2/DRSP group after the sixth cycle. Only DRSP-containing COCs kept high values between third and sixth cycle, all others have decreased in their change relative to the baseline value.

Most of the groups had significant elevation in α2-antiplasmin activity test values, except in 20 EE2/GSD and 30 EE2/LNG groups. The highest change was after 6 cycles by 9% in 35 EE2/NET and 20 EE2/DRSP groups.

The PAI-1 values were significantly decreased just in 20 EE2/DRSP group in both measured time points. All other groups also showed tendencies to lower the values comparing to baseline but significance was reached for 30 EE2/LNG only after 3 cycles and for 35 EE2/NET group after 6 cycles.

Intergroup Results Comparison

Overall intergroup analysis was performed after 6 cycles of COCs, and they confirmed that, although most of hemostatic tests are influenced by COCs, several of them changed significantly among themselves (Table 4). Significance was present in values for:

Table 4.

Intergroup Analysis of Difference Between Combined Oral Contraceptives (COCs) in Hemostatic Variables Values Measured After Sixth Cycle.

	All Formulations of COCs	35 EE2/NGM vs 35 EE2/NET	30 EE2/LNG vs 30 EE2/DRSP	20 EE2/GSD vs 20 EE2/DRSP	20 EE2/DRSP vs 30 EE2/DRSP
	P Value
PT	.7592	.7287	.8013	.2928	.7066
aPTT	.3808	.3237	.9262	.0609	.1988
Fibrinogen	.1115	.1931	.6055	.1130	.9707
Antithrombin	.0302^a	.9271	.4120	.3469	.0517
Protein C	.0015^a	.1921	.1729	.0434^a	.4517
Protein S	.0006^a	.2703	.0762	.0097^a	.6267
aPCR ratio	.2189	.3402	.7307	.0678	.5629
FVIII	.3848	.3916	.2779	.4805	.9415
Plasminogen	.0332^a	.8061	.7408	.0009^a	.0193^a
d-dimers	.4618	.2918	.8738	.0599	.2188
A2-antiplasmin	.0625^a	.0243^a	.8738	.0679	.2153
PAI-1	.0416^a	.7653	1.0000	.0427^a	.9123

Abreviations: FVIII, factor VIII; PAI-1, inhibitor plasminogen activator type 1; aPCR, resistance to activated protein C ratio; COC, combined oral contraceptive; aPTT activated partial thromboplastin time; PTT, prothrombin time; EE2, ethinyl estradiol; DRSP, drospirenone; NGM, norgestimate; GSD, gestodene; LNG, levonorgestrel.

^a P < .05.

antithrombin activity for 20 EE2/DRSP versus 35 EE2/NGM, 35 EE2/NET,

protein C activity for 35 EE2/NGM versus 30 EE2/DRSP, 20 EE2/DRSP, for 35 EE2/NET versus 20 EE2/GSD, 20 EE2/DRSP, 30 EE2/DRSP, and for 30 EE2/LNG versus 20 EE2/DRSP,

protein S activity for 20 EE2/GSD versus 35 EE2/NGM, 35 EE2/NET, 30 EE2/DRSP, 20 EE2/DRSP, and for 30 EE2/LNG versus 35 EE2/NGM, 20 EE2/DRSP, and

PAI-1 level for 20 EE2/GSD versus 35 EE2/NET, 35 EE2/NGM, 20EE2/DRSP.

To establish the difference between COCs, we have made comparison between COCs with the same amount of EE2 or progestins, after 6 cycles. In 35 EE2/NGM versus 35 EE2/NET group significant difference was only in α2-antiplasmin test—P value was .0243.

There was no significant difference between any tests of hemostasis comparing 30 EE2/LNG and 30 EE2/DRSP.

The 20 EE2/GSD versus 20EE2/DRSP showed significant difference in 4 measured tests: protein C, protein S, plasminogen, and PAI-1 with P values .0434, .0097, .0009, and 0.0427, respectively.

Comparison of 20EE2/DRSP versus 30EE2/DRSP gave us insight into hemostatic changes influenced by EE2 amount difference—and the plasminogen activity values were notably different (P value = .0193).

Discussion

Our study confirmed changes in hemostatic variables with low-estrogen COCs. Most of results per group were changed after 3 cycles in PT, aPTT, protein C, protein S, aPCR, plasminogen, d-dimers, and α2-antiplasmin tests. After the sixth cycle, most of variables still changed significantly but showed tendency toward baseline values. However, in d-dimer and protein S tests some participants had pathological values after 3 cycles, and these values remained after sixth cycle. In d-dimers test, 42 participant (21%) had values above reference range (>550 μg/L FEU), 10 of them had values higher than 1100 μg/L FEU. Participants had no other condition that could influence the result, at the time. Half of these participants were either in 35 EE2/NGM or in 20 EE2/DRSP group. Further intergroup analysis confirmed similar previous findings^14,15 that increased d-dimer levels have no significant difference between groups and therefore are not influenced by estrogen amount or progestin type.

Nineteen participants (9.7%) had protein S values lower than 50%, and most of them were in 35 EE2/NGM and 20 EE2/DRSP groups (7 participants in each group). Significant intergroup difference for 30 EE2/LNG group versus 35 EE2/NGM and 20 EE2/DRSP group supports earlier studies^14,16 that DRSP-containing COCs have similar effect on protein S as the third-generation oral contraceptives like 35 EE2/NGM compared to the second-generation 30 EE2/LNG oral contraceptive.

Analysis between groups of oral contraceptives pointed out difference in protein S, protein C, antithrombin, and PAI-1 values. Decreased protein S levels have an influence on increasing resistance to activated protein C, and it is considered as one of contributing mechanisms for increased risk of VTE.¹⁷ Although we used an aPTT-based aPCR test, which is considered as less sensitive,^18,19 we established good correlation in both measured time points (P < .05) between protein S and aPCR values but not intergroup significant difference.

Protein C activity as part of the same anticoagulant system was significantly changed, and it also presented intergroup difference for example between 35 EE2/NET and DRSP containing COCs, that could be explained with biological activities of progestin and their interaction with estrogen component.²⁰ Norethisterone has estrogenic and androgenic activity, and it showed the least increase in protein C activity compared to DRSP-containing contraceptives with antiestrogenic and antiandrogenic properties.⁶

Decreased antithrombin activity in third generation of COCs shown in previous studies^14,21,22 coincides with our results which showed significant decrease in 20 EE2/GSD in both time points compared to baseline and 35 EE2/NGM after sixth cycle but the values remained within the reference range. However, Winkler et al,²³ established significant decrease in antithrombin activity in 30 EE2/LNG group, which we could not confirmed.

The PAI-1 levels were significantly changed in 20 EE2/DRSP group after 3 and 6 cycles, and 30 EE2/LNG after third and 35 EE2/NET after sixth cycle. Studies so far gave opposite results concerning PAI-1 activity.^23

–26 Estrogen influence on PAI-1 synthesis and its clearance in liver,²⁷ but polymorphism of PAI-1 locus and environmental impact may have greater influence.²⁸

Possible distinction between progestins was established by pairwise comparison of COCs containing same amount of EE2. The 30 EE2 group showed no difference between LNG and DRSP; 35 EE2 group showed significant difference between NGM and NET only concerning α2-antiplasmin but the values remained in reference values, and group with least EE2 showed significant difference between GSD and DRSP in protein C, protein S, PAI-1, and plasminogen, though difference found in plasminogen values can be, partly, explained with difference in baseline values. Biological properties of GSD and DRSP are distinct in androgen and glucocorticoid activity as well as high SHBG affinity of GSD,⁶ resulting in less prominent changes compared to DRSP. This is particularly found in association of SHBG values and factors of protein C pathway. Elevation in SHBG is associated with increased resistance to activated protein C. Progestin that has more binding affinity for SHBG (such as GSD) reduces the effect of estrogen component in the COC formulation. Drospirenone does not bind to SHBG and therefore does not affect the estrogen effect on the hemostatic system. However, the effect of progestin on the value of SHBG in plasma is only one of the established modes of their action, while the effects of progestin on other steroid receptors and their possible impact on the hemostatic system are not fully understood.^17,29,30

Drospirenone is a novel progestin derived from spirolactone available in market in 2 formulations of 20 EE2 or 30 EE2, with antimineralocorticoid and antiandrogenic properties.⁶ Antimineralocorticoid property prevents transactivation of mineralocorticoid receptors, thus reducing aldosterone effect—resulting in water and sodium excretion and consequently slight decrease in body weight and blood pressure.³¹ Antiandrogenic property potentiates the use of DRSP for treatment in women with acne. However, DRSP has no binding affinity for SHBG, and therefore it does not decrease estrogen impact on SHBG levels and hemostasis.^12,32 Therefore, the overall thrombotic risk for DRSP formulations depends mostly on the amount of estrogen. Comparison of 2 formulations in our study showed notably increased plasminogen activity for 20 EE2/DRSP, suggesting its higher profibrinolytic response. None of other hemostatic variables were significantly changed. According to a study by Kluft et al,²⁴ only significant difference between 2 oral contraceptives persisted in protein S activity.

Factor VIII activity is only a hemostatic variable that was not changed by any type of oral contraceptives. That implies a different route for increasing resistance to activated protein C, in aPTT-based aPCR test.

Conclusion

Combined oral contraceptives represent sum of estrogen and specific progestin properties resulting in different disturbances of hemostatic variables affecting both coagulation and fibrinolytic pathways. Significance of these changes, within reference range, to increased risk of VTE is still unclear. In this study, we did not establish any major difference among COC formulations. The major difference in laboratory testing was shown in d-dimer and protein S assays. These changes in test values of both assays should be taken into consideration by clinicians while evaluating clinical signs and symptoms of venous thromboembolism.

Our findings also suggest slightly favorable hemostatic profile of 20 EE2/GSD over 20 EE2/DRSP. More studies comparing DRSP formulations are needed.

Footnotes

Authors’ Note

The study was supported by Siemens Healthcare Diagnostics GmbH, Austria.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Sabra

Bonnar

Hemostatic system changes induced by 50 micrograms and 30 micrograms estrogen/progestogen oral contraceptives. Modification of estrogen effects by levonorgestrel. J Reprod Med. 1983;28(suppl 1):85–91.

Gerstman

Piper

Tomita

Ferguson

Stadel

Lundin

. Oral contraceptive estrogen dose and the risk of deep venous thromboembolic disease. Am J Epidemiol. 1991;133(1):32–37.

Barsoum

Heit

Ashrani

Leibson

Petterson

Bailey

. Is progestin an independent risk factor for incident venous thromboembolism? A population-based case-control study. Thromb Res. 2010;126(5):373–378.

van Hylckama Vlieg

Helmerhorst

Vandenbroucke

Doggen

Rosendaal

. The venous thrombotic risk of oral contraceptives, effects of oestrogen dose and progestogen type: results of the MEGA case-control study. BMJ. 2009;339:b2921.

Haider

D'Souza

. Non-contraceptive benefits and risks of contraception. Best Pract Res Clin Obstet Gynaecol 2009;23(2):249–62.

Schindler

Campagnoli

Druckmann

. Classification and pharmacology of progestins. Maturitas. 2003;46(suppl 1):S7–S16.

Stanczyk

. All progestins are not created equal. Steroids. 2003;68(10-13):879–890.

Shi

Fotherby

. Pharmacokinetics of norethisterone in humans. Contraception. 1987;35(5):465–475.

Goldzieher

Stanczyk

. Oral contraceptives and individual variability of circulating levels of ethinyl estradiol and progestins. Contraception. 2008;78(1):4–9.

10.

Winkler

UH.

Hemostatic effects of third- and second-generation oral contraceptives: absence of a causal mechanism for a difference in risk of venous thromboembolism. Contraception. 2000;62(suppl 2):11S–20S; discussion 37S-38S.

11.

Schindler

. Differential effects of progestins on hemostasis. Maturitas. 2003;46(suppl 1):S31–S37.

12.

van Vliet

Frolich

Christella

. Association between sex hormone-binding globulin levels and activated protein C resistance in explaining the risk of thrombosis in users of oral contraceptives containing different progestogens. Hum Reprod. 2005;20(2):563–568.

13.

The European Consensus Development Conference 2002: Sex Steroids and Cardiovascular Diseases. On the route to combined evidence from OC and HRT/ERT. Maturitas. 2003;44(1):69–82.

14.

The effects of seven monophasic oral contraceptive regimens on hemostatic variables: conclusions from a large randomized multicenter study. Contraception. 2003;67(3):173–185.

15.

Klipping

Marr

. Effects of two combined oral contraceptives containing ethinyl estradiol 20 microg combined with either drospirenone or desogestrel on lipids, hemostatic parameters and carbohydrate metabolism. Contraception. 2005;71(6):409–416.

16.

van Vliet

Bertina

Dahm

. Different effects of oral contraceptives containing different progestogens on protein S and tissue factor pathway inhibitor. J Thromb Haemost. 2008;6(2):346–351.

17.

Tchaikovski

Rosing

. Mechanisms of estrogen-induced venous thromboembolism. Thromb Res. 2010;126(1):5–11.

18.

Alhenc-Gelas

Plu-Bureau

Guillonneau

. Impact of progestagens on activated protein C (APC) resistance among users of oral contraceptives. J Thromb Haemost. 2004;2(9):1594–600.

19.

Curvers

Thomassen

Nicolaes

. Acquired APC resistance and oral contraceptives: differences between two functional tests. Br J Haematol. 1999;105(1):88–94.

20.

Kemmeren

Algra

Meijers

. Effect of second- and third-generation oral contraceptives on the protein C system in the absence or presence of the factor VLeiden mutation: a randomized trial. Blood. 2004;103(3):927–933.

21.

Lippi

Manzato

Brocco

Franchini

Guidi

. Prothrombotic effects and clinical implications of third-generation oral contraceptives use. Blood Coagul Fibrinolysis. 2002;13(1):69–72.

22.

Tans

Curvers

Middeldorp

. A randomized cross-over study on the effects of levonorgestrel- and desogestrel-containing oral contraceptives on the anticoagulant pathways. Thromb Haemost. 2000;84(1):15–21.

23.

Winkler

Rohm

Hoschen

. An open-label, comparative study of the effects of a dose-reduced oral contraceptive containing 0.02 mg ethinylestradiol/2 mg chlormadinone acetate on hemostatic parameters and lipid and carbohydrate metabolism variables. Contraception. 2010;81(5):391–400.

24.

Kluft

Endrikat

Mulder

Gerlinger

Heithecker

. A prospective study on the effects on hemostasis of two oral contraceptives containing drospirenone in combination with either 30 or 20 microg ethinyl estradiol and a reference containing desogestrel and 30 microg ethinyl estradiol. Contraception. 2006;73(4):336–343.

25.

Machado

de Melo

Maia

Jr Cruz

. Effect of a continuous regimen of contraceptive combination of ethinylestradiol and drospirenone on lipid, carbohydrate and coagulation profiles. Contraception. 2010;81(2):102–106.

26.

Wiegratz

Lee

Kutschera

Winkler

Kuhl

. Effect of four oral contraceptives on hemostatic parameters. Contraception. 2004;70(2):97–106.

27.

Hoetzer

Stauffer

Greiner

Casas

Smith

DeSouza

. Influence of oral contraceptive use on endothelial t-PA release in healthy premenopausal women. Am J Physiol Endocrinol Metab. 2003;284(1):E90–E95.

28.

de Lange

Snieder

Ariens

Spector

Grant

. The genetics of haemostasis: a twin study. Lancet. 2001;357(9250):101–105.

29.

Africander

Verhoog

Hapgood

. Molecular mechanisms of steroid receptor-mediated actions by synthetic progestins used in HRT and contraception. Steroids. 2011;76(7):636–652.

30.

Fleischer

van Vliet

Rosendaal

Rosing

Tchaikovski

Helmerhorst

. Effects of the contraceptive patch, the vaginal ring and an oral contraceptive on APC resistance and SHBG: a cross-over study. Thromb Res. 2009;123(3):429–435.

31.

Oelkers

. Drospirenone, a progestogen with antimineralocorticoid properties: a short review. Mol Cell Endocrinol. 2004;217(1-2):255–261.

32.

Sitruk-Ware

Nath

. The use of newer progestins for contraception. Contraception 2010;82(5):410–417.

Comparison of the Impact of Four Generations of Progestins on Hemostatic Variables

Abstract

Keywords

Introduction

Materials and methods

Study Participants

Study Design

Samples

Assays for Hemostatic Variables

Statistical Analysis

Results

General Tests

Coagulation Pathway Tests

Fibrinolytic Pathway Tests

Intergroup Results Comparison

Discussion

Conclusion

Footnotes

Authors’ Note

Declaration of Conflicting Interests

Funding

References