Haemostatic and metabolic impact of estradiol pills and drospirenone-containing ethinylestradiol pills vs. levonorgestrel-containing ethinylestradiol pills: A literature review.

The European Journal of Contraception & Reproductive Health Care

ISSN: 1362-5187 (Print) 1473-0782 (Online) Journal homepage: http://www.tandfonline.com/loi/iejc20

Haemostatic and metabolic impact of estradiol pills and drospirenone-containing ethinylestradiol pills vs. levonorgestrel-containing ethinylestradiol pills: A literature review Iñaki Lete, Nathalie Chabbert-Buffet, Christian Jamin, Stefano Lello, Paloma Lobo, Rossella E. Nappi & Axelle Pintiaux To cite this article: Iñaki Lete, Nathalie Chabbert-Buffet, Christian Jamin, Stefano Lello, Paloma Lobo, Rossella E. Nappi & Axelle Pintiaux (2015) Haemostatic and metabolic impact of estradiol pills and drospirenone-containing ethinylestradiol pills vs. levonorgestrel-containing ethinylestradiol pills: A literature review, The European Journal of Contraception & Reproductive Health Care, 20:5, 329-343, DOI: 10.3109/13625187.2015.1050091 To link to this article: http://dx.doi.org/10.3109/13625187.2015.1050091

Published online: 26 May 2015.

Submit your article to this journal

Article views: 542

View related articles

View Crossmark data

Citing articles: 7 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=iejc20 Download by: [University of Exeter]

Date: 30 September 2017, At: 10:11

The European Journal of Contraception and Reproductive Health Care, 2015; 20: 329–343

Downloaded by [University of Exeter] at 10:11 30 September 2017

REVIEW

Haemostatic and metabolic impact of estradiol pills and drospirenonecontaining ethinylestradiol pills vs. levonorgestrel-containing ethinylestradiol pills: A literature review Iñaki Lete∗, Nathalie Chabbert-Buffet †, Christian Jamin‡, Stefano Lello§ , Paloma Lobo #, Rossella E. Nappi^ and Axelle Pintiaux $ ∗Department of Obstetrics and Gynaecology, Araba University Hospital,Vitoria-Gasteiz, Spain, †Department of Obstetrics and Gynecology, APHP Tenon Hospital, Pierre and Marie Curie University, Paris, France, ‡169, boulevard Haussmann, Paris, France, §Gynecology Unit, Madonna delle Grazie Clinic,Velletri, Rome, Italy, #Department of Obstetrics and Gynecology, Infanta Sofia University Hospital, San Sebastián de los Reyes, Madrid, Spain, ^Research Center for Reproductive Medicine and Unit of Gynecologic Endocrinology and Menopause, IRCCS Policlinico San Matteo, Department of Clinical, Surgical, Diagnostic and Pediatric Sciences, University of Pavia, Pavia, Italy, and $Department of Obstetrics and Gynecology, Citadelle Hospital, University of Liège, Liège, Belgium

ABSTRACT

Objective Since its introduction 50 years ago, the contraceptive pill has continuously evolved to decrease the risk of venous thromboembolism (VTE) associated with its use. An increased risk of VTE still remains, however. Other concerns, such as effects on lipid and carbohydrate metabolism, have also been reported. In this study we compared two reference combined oral contraceptives (COCs) containing ethinylestradiol (EE)/levonorgestrel (LNG) and EE/drospirenone (DRSP) with COCs containing estradiol (E2) (estradiol valerate [E2V]/ dienogest [DNG] and E2/nomegestrol acetate [NOMAC]). They were evaluated according to their influence on recognised haemostatic and metabolic markers. Methods A literature search of the MEDLINE/PubMed database was conducted for head-to-head studies. EE/LNG was chosen as the comparator pill. Results The haemostatic impact of E2 pills and EE/LNG has been extensively compared, in contrast to that of EE/DRSP and EE/LNG. Changes in haemostatic and metabolic marker levels between EE/LNG and E2V/DNG were generally not statistically significant. E2/NOMAC showed statistically significantly favourable results on haemostatic markers and had a neutral effect on carbohydrate and lipid metabolism when compared with EE/LNG.

Correspondence: Iñaki Lete, Department of Obstetrics and Gynaecology, Araba University Hospital, Jose Atxotegi, 01009 Vitoria, Spain. Tel: ⫹ 34 945 007107. E-mail: [email protected]

© 2015 The European Society of Contraception and Reproductive Health DOI: 10.3109/13625187.2015.1050091

Haemostatic impact of different oral contraceptives

Lete et al.

Conclusion E2/NOMAC exhibits less haemostatic and metabolic impact than EE/LNG and other COCs, suggesting that it may be a promising candidate to reduce residual VTE risk associated with COC use. Confirmation from a well-powered prospective clinical trial is, however, needed. K E Y WO R D S

Combined oral contraceptive; Estradiol; Ethinylestradiol; Haemostatic markers; Markers of lipid and carbohydrate metabolism; Progestin; Venous thromboembolism


I N T RO D U C T I O N

When choosing a contraceptive method many factors must be taken into account and discussed with the patient: in particular, efficacy, tolerability, additional health benefits, and whether or not the patient can/ will use an alternative method. The risk of venous thromboembolism (VTE) is still a prominent element of the equation. Currently available combined oral contraceptives (COCs) are associated with an overall 2.8-fold increased residual risk of VTE. Large epidemiological studies have shown that women taking low-dose third-generation or drospirenone (DRSP)-containing COCs were exposed to a higher thrombotic risk than those taking secondgeneration COCs1–3. This could be the result of the higher ‘total estrogenicity’ of newer COCs, which may cause more pronounced changes in coagulation4. Although the absolute risk remains low, VTE is still a real concern5,6. It is estimated that 100 million women worldwide use oral contraceptives1. Consequently, even the slightest increase in the incidence of side effects will affect a large population of women, including the vulnerable group of new users, especially during the first months of use6,7. The use of COCs is not the sole contributing factor to VTE in the general population, however.The disease is known to be multifactorial or multigenic, and its occurrence increases when more than one risk factor is present8,9. Risk factors include both acquired (e.g. major surgery, trauma, cancer therapies, stroke, acquired thrombophilia such as lupus, autoimmune diseases, advancing age, obesity, pregnancy, postpartum period, hormone replacement therapies, long-distance airline travel, immobility) and congenital (e.g. factor V Leiden, genetic thrombophilia, antithrombin deficiency, protein C deficiency, protein S deficiency, anatomic characteristics) risk9,10. Risk factors for VTE have different odds ratios; COCs belong to the group of moderate risk factors for VTE (Table 1)11. 330

Most of the clinical risk factors for VTE are, however, uncommon in women of reproductive age10. In COC users, there is a strong interaction between hereditary defects and the risk of VTE10. Nevertheless, it has been shown that most users of COCs who develop VTE do not have a recognised hereditary coagulation problem: genetic risk factors contribute to only about 30% of the family history of VTE12. It also appears that in women with underlying coagulation thrombophilia, a family history of VTE is a poor indicator, meaning routine testing prior to COC use would not be a solution10. It is therefore difficult to put together a possible vulnerability profile that could predispose users of COCs to VTE events. As discussed above,VTE tends to occur when multiple risk factors, including genetic and environmental, are present simultaneously. COC use, as well as surgery, immobility or pregnancy, could be considered triggering factors. The use of COCs interacts with personal vulnerabilities such as congenital or acquired thrombophilia9. It causes changes in coagulation that modify the prothrombotic state induced by pre-existing genetic alterations in a supra-additive manner. Subsequent effects lead to a considerable increase in thrombosis potential, which in prone individuals may exceed the thrombosis threshold and lead to VTE4. Because COC users have varying degrees of vulnerability to VTE, not all will exceed the thrombosis threshold. It appears that there is a sense of uncertainty associated with the use of COCs. Consequently, proper and comprehensive information that helps women make their final decision (independently or shared with health care practitioners) about the most suitable COC to use is critical in the patient– physician relationship13. Although the absolute risk is low, VTE may be severe, even fatal.VTE events in healthy young women are particularly unacceptable, especially in modern societies.The use of COCs with limited metabolic and vascular impact on the body – a major goal of

The European Journal of Contraception and Reproductive Health Care


Lete et al.

Table 1 Risk factors for VTE11. Strong risk factors (odds ratio ⬎ 10)

Fracture (hip or leg) Hip or knee replacement Major general surgery Major trauma Spinal cord injury Arthroscopic knee surgery Central venous lines Chemotherapy Congestive heart or respiratory failure Hormone replacement therapy Malignancy Oral contraception Paralytic stroke Pregnancy, postpartum Previous VTE Thrombophilia Bed rest ⬎ 3 days Immobility due to sitting (e.g., prolonged car or air travel) Increasing age Laparoscopic surgery (e.g., cholecystectomy) Obesity Pregnancy, antepartum Varicose veins


Moderate risk factors (odds ratio 2–9)

Weak risk factors (odds ratio ⬍ 2)

contraceptive research for many years – may help reduce uncertainty. Such pills would reduce the likelihood of crossing the risk threshold, leading to improved safety (Figure 1).

To review the influence of the contraceptive pill on haemostatic and metabolic markers, we decided to focus our attention on four specific COCs: the second-generation ethinylestradiol (EE)/levonorgestrel

VTE risk level Thrombosis threshold

High

WOMEN WITH HIGH VULNERABILITY PROFILES*

Low

WOMEN WITH LOW VULNERABILITY PROFILES

High

Acquired risk factors

Figure 1 The dynamic vulnerability profile model in oral contraception.


331



Lete et al.

(LNG) combination, which is the European Medicines Agency (EMA)-recommended comparator pill to be used to investigate VTE risk14; the fourth-generation pill containing EE and the spironolactone analogue DRSP, the most widely used pill, but which, according to epidemiological data, is associated with a significantly higher risk of venous thrombosis than COCs with LNG6; the valerate ester of 17β-estradiol (E2V)/dienogest (DNG) combination in a quadriphasic regimen, the first natural estrogen pill introduced in clinical practice15; and the monophasic E2/nomegestrol acetate (NOMAC) pill, the other estrogen pill and sole member of its group according to the World Health Organization (WHO) classification16 ̶ a COC containing the progesterone derivative NOMAC and natural estrogen structurally identical to the major estrogen produced by the ovaries of healthy nonpregnant women.

METHODS

The choice of haemostatic and metabolic markers To evaluate the haemostatic and metabolic risk associated with the use of COCs, adequate markers are needed to help determine the metabolic profile and potential impact on health. Haemostatic markers that may be affected by EE-containing pills

Table 2 Haemostatic changes during COC use (adapted from reference 17). Factor

The specific mechanisms underlying the increased risk of VTE associated with the use of COCs are still relatively unclear. They include a direct effect of estrogens on the vascular wall, changes in factors that promote endothelial dysfunction, and changes in coagulation factors17. Recent data suggest that the concentration and function of haemostatic proteins are not the only factors determining VTE risk. Cells and cellular components associated with (acute) inflammation and platelets also play a rate-limiting role in thrombus formation and can also affect thrombotic risk18,19. The occurrence of VTE seems to involve the combination of stasis and hypercoagulability17. Normally, coagulation is controlled by several inhibitors that limit clot formation near the damaged vessel wall, thus avoiding thrombus propagation17. However, if the activity of one of the coagulation factors or the activity 332

of one of the naturally occurring inhibitors is modified, the balance can be interrupted, leading to thrombus formation17. Several haemostatic factors which belong to procoagulant, anticoagulant or fibrinolytic pathways have been shown to be affected by the use of COCs (Table 2). Although their concentrations usually remain within the normal range, alterations in their levels may have synergistic or opposing effects on the risk of VTE17. In the pill, EE and progestin have specific effects on haemostasis. EE modifies estrogen-sensitive haemostatic factors and hepatic proteins, while these effects are modulated by the type of progestin20,21. Shortly after intake, EE increases procoagulants and decreases anticoagulants, in particular protein S22. By contrast, intake of progestins alone was shown to increase both protein S and fibrinolytic potency22. EE-containing COCs increase the activity of coagulation factors I (fibrinogen), II, VII, VIII, IX, X and XI by about 30 to 50%23 and decrease the activity of naturally occurring anticoagulants like protein S by approximately 30 to 40%21. Due to the decreased protein S activity, one important effect of COCs on blood coagulation is the development of an acquired

Procoagulant factors Fibrinogen, V, VII, VIII, IX, X, XII XI von Willebrand factor Anticoagulant proteins Antithrombin Protein C Protein S Resistance to APC Markers of thrombin formation F 1 ⫹ 2, TAT, fibrinopeptide A, D-dimer Fibrinolytic factors TAFI, PAI-1, PAI-2, t-PA

Change during COC use  ⫽ or  ⫽   ⫽ or        

, increase;   , decrease; ⫽ , no change (vs. non-use of COCs). TAT, thrombin–antithrombin complex; TAFI, thrombinactivatable fibrinolysis inhibitor; PAI, plasminogen activator inhibitor; t-PA, tissue plasminogen activator.




resistance to activated protein C (APC), a prothrombotic phenotype17,22. To balance this hypercoagulability, upregulation of fibrinolysis occurs, as highlighted by higher levels of d-dimer and prothrombin fragment 1 ⫹ 2 (F 1 ⫹ 2) (Figure 2 and Table 2)22. The prothrombotic effect of COCs can be considered to be related to the total estrogenicity of the formulation (‘total estrogenicity’ rises with increasing dose of estrogen but decreases with increasing antiestrogenic activity of the progestin compound)4. Being both affected by EE levels and the androgenic activity of progestin, the hepatic protein sex hormone-binding globulin (SHBG) was proposed as a marker for estrogen potency and also as a marker for predicting thrombotic risk in COC users24. Its levels are indeed correlated to changes in the APC sensitivity ratio, which is, in turn, correlated to the risk of VTE4,20. Generally, COCs with higher total estrogenicity are associated with a higher thrombotic risk20. Which haemostatic markers really matter? The number of accepted surrogate markers of VTE risk is limited, but several have been proposed14. Concentrations of prothrombin F 1 ⫹ 2 and d-dimer reflect overall activation of the coagulation and fibrinolytic

Lete et al.

systems. Increased concentrations of these factors are a measure of thrombotic tendency25. Under normal conditions, the coagulation and fibrinolytic pathways sustain low-grade activation. In the prethrombotic state, increased concentrations of activation markers, such as F 1 ⫹ 2, thrombin–antithrombin complex and d-dimer, are above normal levels25. F 1 ⫹ 2 and d-dimer are both important clinical markers of thrombotic risk26. The measurement of d-dimer has become a cornerstone in the diagnosis of patients with suspected VTE and plays a pivotal role in the detection of hypercoagulable states26. The median concentration of F 1 ⫹ 2 is a dynamic marker, considered to be the most representative of changes in coagulation, whereas median change in d-dimer concentration is the dynamic marker for the fibrinolytic system25.They may illustrate different pharmacological effects, possibly related to VTE risk14. To help determine the pill’s haemostatic profile, other haemostatic variables are needed, such as acquired APC resistance and protein S levels. Accurate profiling of the haemostatic impact of a pill should include measurement of the following markers, which are already recommended by health authorities but not always used in clinical trials14: (1) F 1 ⫹ 2; (2) d-dimer; (3) acquired APC resistance; (4) protein S.

PROCOAGULATORY MARKERS: prothrombin F 1+2, fibrinogen, factor II, factor VIII, nAPCr

COAGULATION

ANTICOAGULATORY MARKERS: free protein S protein S activity, APC, antithrombin,TFP1

FIBRINOLYSIS MARKERS: plasminogen, D-dimers

FIBRINOLYSIS

ANTIFIBRINOLYSIS MARKER: PAI-1

Figure 2 The dynamic balance of haemostasis. Haemostasis is a complex process resulting from a subtle balance between procoagulatory, anticoagulatory, fibrinolysis and antifibrinolysis markers. COCs have multiple effects on this dynamic balance. They alter procoagulant, anticoagulant and fibrinolytic pathways. APC (activated protein C); nAPCr, normalised APC ratio; TFP1, tissue factor pathway inhibition 1; PAI-1: plasminogen activator inhibitor-1.


333


Lete et al.


Other metabolic risks and their markers The hormonal components of COCs have effects on inflammation, lipid and carbohydrate metabolism. Some of the metabolic changes, such as increased lowdensity lipoprotein cholesterol (LDL-C), triglycerides (TG), insulin resistance and insulin levels, and decreased levels of high-density lipoprotein cholesterol (HDL-C), are similar to those observed in individuals who are at increased risk of coronary heart disease27. A few articles have reported cases of women who have developed pancreatitis, with elevated blood plasma TG concentrations, secondary to COC-induced hyperlipidaemia28,29. The effects on levels of unfavourable lipoproteins are, in fact, determined by the type of progestin27. Metabolic changes could be due, in part, to the ability of androgenic progestins to counteract estrogen-induced changes in levels of LDL-C and HDL-C30–33. COCs have also been associated with subclinical disturbances in carbohydrate metabolism, including impaired glucose tolerance and insulin resistance, which may increase the risk of type 2 diabetes and vascular disease33. Glucose intolerance and hyperinsulinaemia are well-known risk factors for cardiovascular disease34. Studies evaluating the effects of estrogens on carbohydrate metabolism have, however, reported contradictory findings33. Insulin sensitivity could be affected by the estrogen component, whereas the magnitude of the response can be modified by the dose and type of progestin21,27,35. Decline in insulin sensitivity is greater with a progestin with androgenic activity than with a progestin with antiandrogenic properties36. Progestins have also been shown to be responsible for impaired glucose tolerance (the degree of impairment depending on both the type and the dose), whereby their androgenic properties may also be involved in the condition34,37. In addition, change in insulin response or in glucose tolerance occurs with COC and not with transdermal therapy34. Overall, measuring LDL-C, TG and HDL-C markers of lipid metabolism could help better characterise the profile of COCs. In addition, insulin response and glucose levels can also be used as markers of the effect of COCs on carbohydrate metabolism. Literature search A literature search was conducted using the MEDLINE/ PubMed database. It spanned the years from 1990 to 334

2014. Additional resources were identified using reference lists from the obtained articles. Inclusion criteria were: COC, haemostasis, lipid metabolism, carbohydrate metabolism, LNG, DRSP, E2V, DNG, E2, and NOMAC. Other criteria for inclusion were: published, prospective studies and English language. Head-tohead studies evaluating the effects of EE/LNG, EE/ DRSP, E2V/DNG and E2/NOMAC on haemostasis, lipid and carbohydrate metabolism were preferentially searched. We specifically focused on the following markers: F 1 ⫹ 2, d-dimer, acquired APC resistance, protein S; LDL-C, HDL-C (or total cholesterol), TG; insulin response, glucose levels. Only studies in healthy women were included. R E S U LT S

Haemostatic profile of COCs In 2005, the EMA recommended that for any new COC the appropriate comparators to be used to investigate VTE risk should be either a pill containing EE and LNG or a pill containing EE and desogestrel (DSG)14. In 2013, women using COCs containing LNG were confirmed to have the lowest risk of VTE, along with women using COCs containing norethisterone or norgestimate38. LNG-containing COCs LNG is probably one of the most widely used progestins in COCs and its combination with EE has been extensively studied. Prior to 2000, around 14 prospective studies directly comparing the haemostatic effects of LNG-containing COCs and third-generation COCs (containing DSG or gestodene [GSD]) were published39. A review of these studies was designed, but, owing to considerable differences in inclusion and exclusion criteria, methodology and statistical analysis, a metaanalysis could not be conducted39. Nevertheless, it was possible to identify whether consistent trends existed: no differential effects of DSG/GSD- and LNGcontaining COCs were found for established risk markers of VTE39. The effects on haemostasis of a COC containing 150 μg DSG and a COC containing 150 μg LNG in combination with 30 μg EE were also investigated for two consecutive cycles40. Prothrombin F 1 ⫹ 2 concentrations increased significantly during




use of both COCs: 0.4 nmol/l with EE/LNG vs. 0.3 nmol/l with EE/DSG (mean; p ⫽ 0.33). This confirmed previous results obtained with EE/LNG and showed that EE/DSG had the same effect on this marker of thrombin formation. By contrast, in another study, the activity of the anticoagulant pathways in plasma from users of DSG-containing COCs was more extensively impaired than in plasma from users of LNG-containing COCs41. As previously mentioned in the section ‘Which haemostatic markers really matter?’, of the four haemostatic markers we recommended to evaluate, only protein S and APC sensitivity ratio were assessed in this study. Compared with EE/ LNG, DSG-containing COCs caused a significant decrease in total (p ⬍ 0.005) as well as free protein S (p ⬍ 0.0001), and more pronounced APC resistance in both activated partial thromboplastin time (p ⫽ 0.02) and endogenous thrombin potential-based APC resistance tests (p ⬍ 0.0001)41. More recently, the effects of COCs on haemostatic biomarkers were confirmed to vary by progestin, with LNG-based COCs causing a less pronounced effect than COCs releasing DSG: ⫹ 170 vs. ⫹ 592 pmol/l (F 1 ⫹ 2); ⫹ 88.3 (0.47) vs. ⫹ 168.5 ng/ml (0.90 nmol/l) (d-dimer); ⫺ 12.8% vs. ⫺ 19.5% (protein S); and ⫺ 0.14 vs. ⫺ 0.29 (APC resistance), respectively42. Overall, this could explain epidemiological results showing that COCs with DSG, or GSD, were associated with a significantly higher risk of VTE compared with COCs with LNG6. In 2003, the Oral Contraceptive and Hemostasis Study Group prospectively investigated seven COCs, differing in EE dose and progestin type, in 707 healthy women for six cycles43. LNG-containing COCs were compared with COCs containing DSG, norgestimate or GSD. Changes in haemostatic markers were lower with lower EE doses. In women using LNG-containing COCs, prothrombin F 1 ⫹ 2 levels were lower than in those using DSG-, norgestimate- or GSD-containing COCs. In parallel, significantly lower levels of protein S (free and total) and an increased APC sensitivity ratio were found in the latter group compared with the LNG group. It is thought that the reduction of EE in COCs results in small changes in haemostatic markers, suggesting a potentially more favourable haemostatic profile43,44. In terms of EE dose, the EMA states that the epidemiological data did not suggest a lower risk of VTE for COCs containing DSG or GSD with 20 μg EE, compared with those containing 30 μg EE45.

Lete et al.

In 2013, the EMA confirmed that the known risk of VTE with all low-dose COCs (EE ⬍ 50 μg) was small38. In addition, a recent Cochrane systematic review and network meta-analysis reported a doserelated effect of EE for GSD, DSG and LNG, with higher doses being associated with an increased risk of thrombosis46. DRSP-containing COCs To our knowledge, EE/DRSP and EE/LNG have not been directly compared to date. EE 20 μg/DRSP 3 mg and EE 20 μg/DSG 150 μg were studied and found to increase the levels of the activation markers thrombin, fibrin turnover F 1 ⫹ 2 and d-dimer. The mean increase (from baseline to cycle 7) in F 1 ⫹ 2 was less pronounced with EE/ DRSP than with EE/DSG, but the difference in means was not statistically different. For d-dimer, the mean increase was greater with EE/DRSP than with EE/ DSG (96.00 [0.51] vs. 80.74 ng/ml [0.43 nmol/l]; the difference in mean change was not statistically significant). Anticoagulatory factors (protein S and APC resistance) decreased slightly in both treatment groups. Both EE/DRSP and EE/DSG produced similar increases in the activation markers of thrombin and fibrin turnover, while they decreased anticoagulatory markers47. The lack of a head-to-head comparison with the recommended comparator EE/LNG is, however, disadvantageous to EE/DRSP. E2V/DNG combination To assess changes in haemostatic parameters, the quadriphasic COC E2V/DNG (E2V 3 mg on days 1–2; E2V 2 mg/DNG 2 mg on days 3–7; E2V 2 mg/DNG 3 mg on days 8–24; E2V 1 mg on days 25–26; placebo on days 27–28) was compared with COCs containing high doses of EE (30 or 40 μg) and various doses of LNG (50, 75, 125 or 150 μg)48,49. Absolute changes in prothrombin F 1 ⫹ 2 (primary endpoint) from baseline to cycle 3 were not statistically different between E2V/DNG and EE 30 μg/LNG 150 μg48. A significantly smaller increase in d-dimer levels (primary endpoint) for E2V/DNG than for EE/LNG was observed (38.9% vs. 157.9%, respectively; p ⫽ 0.01), but mean levels in both groups remained within the reference interval48. No statistically significant differences were found between the two COCs for protein


335


Lete et al.

S and APC resistance (secondary endpoints), although the mean APC sensitivity ratio showed a statistically significant greater increase in the EE/LNG group. Increases in the levels of SHBG were observed in the E2V/DNG and EE 30 μg/LNG 150 μg groups but were less marked in women who received E2V/ DNG48. This result was not statistically different.


E2/NOMAC combination The effects of the monophasic E2/NOMAC pill on haemostasis were evaluated using two LNG-containing COCs as comparators: EE 20 μg/LNG 100 μg and EE 30 μg/LNG 150 μg33,50. Mean changes in prothrombin F 1 ⫹ 2 (primary endpoint) were significantly smaller and close to zero with E2/NOMAC treatment compared with EE 20 μg/LNG 100 μg (⫺ 0.02 vs. ⫹ 0.08 nmol/l; p ⬍ 0.01), as was the case with antithrombin levels. These results suggest a minimal effect of the E2/ NOMAC COC on thrombin turnover. Increases in free protein S levels and protein S activity were lower with E2/NOMAC than with EE 20 μg/LNG 100 μg (4.38% vs. 9.02%, respectively, and 5.09% vs. 7.29%, respectively), reaching significance for free protein S levels (p ⬍ 0.05). Mean changes in APC resistance (normalised ratio) increased with both treatments but to a lesser extent with E2/NOMAC than with EE 20 μg/LNG 100 μg (⫹ 0.20 vs. ⫹ 0.46; p ⬍ 0.01). In general, E2/NOMAC induced significantly less change on the overall activity of the coagulation system and on APC resistance, indicating a lower activation of coagulation tendency. The impact on fibrinolysis was significantly different between the two pills: d-dimer levels increased during EE 20 μg/LNG 100 μg treatment but decreased during E2/NOMAC treatment (⫹ 43 [0.23] vs. ⫺ 53 ng/ml [0.28 nmol/l]; p ⬍ 0.001). SHBG levels increased after the use of both COCs but were not significantly different between the two pills. The overall effect of EE/LNG on SHBG levels is likely to be a consequence of an increase induced by the EE component, counterbalanced by a decrease induced by the androgenic (or antiestrogenic) effects of LNG. By contrast, the overall effect of E2/NOMAC on SHBG levels most probably reflects only the effects of the E2 component, as the progestin is not androgenic50. In summary, E2/NOMAC induces fewer changes on coagulation and fibrinolysis markers compared with EE 20 μg/LNG 100 μg, the contraceptive pill 336

considered today’s best comparator. This finding was confirmed in another study comparing E2/NOMAC and EE 30 μg/LNG 150 μg33. Lipid profile of COCs The effect of the pill on levels of unfavourable lipids seems to be related to the androgenic properties of the progestins. For example, LNG-containing COCs have a less favourable effect on the lipid profile in comparison with DNG-containing COCs. DNG is less androgenic than LNG, but the EE and LNG doses in EE/LNG pills seem to play a major role51. Contraceptives containing third-generation progestins have less metabolic impact than LNG-containing COCs, because progestins of the second generation are more potent, have androgenic activity and induce changes in the lipid profile which lead to atherosclerosis52. A decrease in HDL-C and an increase in LDL-C and TG were found with both EE 20 μg/LNG 100 μg and EE 30 μg/LNG 150 μg. Results were related to progestin dose. A smaller median 1-year change was seen in the 20 μg EE group, compared with the 30 μg EE group, indicating a smaller impact on lipid profile with the lower LNG dose preparation: adverse lipids (e.g. LDL-C,TG) were increased and the positive lipid HDL-C decreased but to a lesser degree than in the 30 μg EE group53. The effects of 3 mg DRSP in combination with 20 μg EE and EE 20 μg/DSG 150 μg on lipid markers were also tested: an increase in HDL-C and a decrease in LDL-C were observed with both COCs47.Although the differences were not significant between the two groups, the greater increase in HDL-C (16% vs. 11%) and the greater decrease in LDL-C (⫺ 18% vs. ⫺ 10%) were associated with the EE 20 μg/DRSP 3 mg pill. This may reflect the lack of androgenic activity with DRSP compared with DSG43. As with EE/LNG COCs, the levels of TG were increased with both pills: ⫹ 78% with EE/DRSP and ⫹ 40% with EE/DSG after seven cycles47. Similarly, EE/DRSP and EE/LNG preparations were compared, and HDL-C levels were found to be significantly increased by 9% and 23% in patients treated with EE 30 μg/DRSP 3 mg and EE 20 μg/ DRSP 3 mg, respectively, whereas the levels decreased by 12% in patients treated with EE 30 μg/150 μg LNG (p ⬍ 0.05). In parallel, TG levels showed a stronger




increase in the DRSP group than in the LNG group (p ⬍ 0.05)54. By contrast, LDL-C fell slightly in the DRSP group (p ⬎ 0.05)54. The effects of E2V and DNG in combination on lipid markers were studied for seven cycles in comparison with a triphasic regimen of EE/LNG (EE 30 μg/LNG 50 μg on days 1–6, EE 40 μg/LNG 75 μg on days 7–11, EE 30 μg/LNG 125 μg on days 12–21 and placebo on days 22–28)49. Mean HDL-C increased by 7.9% with E2V/DNG and decreased by 2.3% with EE/LNG (p ⫽ 0.055). Mean LDL-C decreased by 6.5% with E2V/DNG and by 3.0% with EE/LNG (p ⫽ 0.458).The difference between mean baseline and cycle 7 TG levels was not significant between the two treatment groups. The influence of the new contraceptive E2/ NOMAC was also assessed. Total cholesterol was unchanged with both COCs: ⫹ 1.3% with E2/ NOMAC and ⫺ 0.7% with EE/LNG (p ⫽ 0.69)33. However, in contrast to EE 30 μg/LNG 150 μg treatment, the E2 pill had no effect on HDL-C, LDL-C and TG after six cycles (1.6% vs. ⫺ 13.1%, p ⬍ 0.001, ⫺ 0.5% vs. 6.8%, p ⫽ 0.046, and 7.5% vs. 17.0%, p ⫽ 0.008 for the E2 pill and the EE pill, respectively). In conclusion, the E2/NOMAC combination does not induce changes in lipid metabolism, whereas EE 30 μg/LNG 150 μg tends to decrease HDL-C and increase LDL-C – changes considered unfavourable with regard to cardiovascular risk. Overall, these findings are supported by previous evidence suggesting that progestins with androgenic properties may reverse the lipid benefit of estrogen, whereas progesterone-like derivatives may be more likely to preserve these benefits55. Carbohydrate profile of COCs The effect on insulin secretion and glucose tolerance of several COCs was studied in various preparations, including those containing LNG, DSG or norethindrone. LNG combinations with EE increased secondphase pancreatic insulin secretion by 60 to 90% but did not affect insulin half-life. By comparison, COCs containing DSG increased insulin half-life by 28% without affecting insulin secretion56. The antiandrogenic progestin DRSP was compared with the androgenic progestin LNG. Glucose levels were studied in women receiving EE 30 μg/LNG 150 μg or 3 mg DRSP combined with various doses of

Lete et al.

EE (15 μg, 20 μg or 30 μg), for six months. On treatment, fasting glucose was unchanged in all groups, but the area under the curve (AUC) for glucose tolerance increased for all formulations. Although not statistically significant between groups, the EE 30 μg/DRSP group had a 19% worsening of glucose tolerance21,54. Other trials with COCs containing LNG or DSG have shown similar findings57–59. Changes in carbohydrate metabolism with the use of E2V/DNG were also studied49. This COC was compared with a triphasic regimen of EE/LNG. Results showed that the AUCs for insulin and glucose remained relatively unaffected by both COCs but were statistically not significant. The non-androgenic progestin NOMAC combined with E2 showed significantly less impact than the comparator pill EE 30 μg/LNG 150 μg after six cycles. EE 30 μg/LNG 150 μg was associated with an increased insulin response and higher glucose levels compared with the E2/NOMAC combination: the percentage changes in AUC for insulin were ⫹ 26.4 vs. ⫺ 0.9 (p ⬍ 0.001), respectively, and the percentage changes in AUC for glucose were 13.9 vs. 6.1 (p ⫽ 0.002), respectively33. These results are in accordance with the finding that non-androgenic progestins have a neutral effect on carbohydrate metabolism34. DISCUSSION

Managing or containing risk has become a hallmark of modern society, in particular in the field of health care60. Society seeks to control, or at least reduce, the risk of harm, focusing on those who are particularly vulnerable. Unfortunately, since risk can never be fully managed, there is always an element of uncertainty. To reduce haemostatic and metabolic risks associated with the use of COCs, a solution might be to provide all women with a pill that minimises metabolic changes.This is especially important when considering the vulnerable group of new users, consisting of young and healthy women. Over the years, the pill has constantly evolved. Efforts were made to reduce thrombotic risk by reducing the dose of EE and synthesising new progestins. Progestins were modified to minimise their androgenic effects and impact on carbohydrate and lipid metabolism61–63. However, third-generation COCs are less able to counterbalance the prothrombotic effect of


337



Lete et al.

EE compared with second-generation progestins – even those using a very low dose of EE4,64. As for newer progestins such as DRSP, they do not have androgenic properties but may not sufficiently oppose the effect of EE on certain haemostatic markers. In women taking EE/DRSP, the levels of the activation markers of thrombin and fibrin turnover F 1 ⫹ 2 and d-dimer increase, while anticoagulatory factors such as protein S and APC resistance decrease. By contrast, LNG, which has residual androgenic properties, can antagonise the EE-induced rise in prothrombin F 1 ⫹ 2, and the EE-dependent reduction in total and free protein S. Compared with third-generation COCs and COCs containing DRSP, pills containing LNG are considered to be associated with the lowest risk of VTE,1–3 probably because their total estrogenicity is lower than that of third-generation and DRSP pills24. To our knowledge, no head-to-head comparison has ever been conducted between EE/DRSP and EE/LNG formulations. Of the ‘natural estrogen generation’ pills, the E2/ NOMAC preparation has more statistically significantly favourable results on haemostatic markers compared with E2V/DNG (Table 3), which could mean that E2/ NOMAC has less pronounced effects on the vascular

system than its counterpart. This, however, has to be proved in epidemiological studies. When compared with the EE/LNG formulation, E2/NOMAC induced significantly less change in both the overall activity of the coagulation system and fibrinolysis markers, suggesting a lower impact on haemostasis than conventional COCs.A large European post-authorisation safety study has been designed to confirm this. In addition, E2/NOMAC does not affect lipid metabolism, especially TG. This is particularly interesting, given that acute hypertriglyceridaemia has recently been shown to induce platelet hyperactivity in women with polycystic ovary syndrome and in healthy women65. One limitation of our review may be that, based on the literature, we chose a few clinical markers of thrombotic risk among a variety of haemostatic variables (Table 4). The relationship between the observed haemostatic changes and VTE risk has not yet been established. However, the haemostatic variables we decided to use as markers of thrombotic risk are on the list of biological variables that, according to the EMA, may reflect different pharmacological effects, possibly related to VTE risk22. In addition, these variables appeared to be commonly used to evaluate the risk of VTE in studies we reviewed.

Table 3 Effects of E2-based COCs (E2V/DNG and E2/NOMAC) and DRSP-containing COCs on haemostasis, lipid and carbohydrate metabolism. EE/LNG was used as the comparator. Marker Haemostatic markers F 1⫹2 D-dimer APC ratio Protein S (activity) Markers of lipid metabolism HDL-C (or total cholesterol) LDL-C TG Markers of carbohydrate metabolism Insulin response (AUC) Glucose levels (AUC)

EE/DRSP 54

E2V/DNG 48,49

E2/NOMAC 33,50

⫺ ⫺ ⫺ ⫺

NS *†   NS NS

*   *‡   *  

* NS *

NS NS NS

NS *   *  

NS ⫺

NS NS

*   *  

NS

vs. EE 30 μg/LNG 150 μg. Another study showed a non-significant difference vs. the comparator49. ‡Another study showed a non-significant difference vs. EE 30 μg/LNG 150 μg33. *, *, significant relative increase vs. EE/LNG’s effects on the marker evaluated;  significant relative decrease vs. EE/LNG’s effects on the marker evaluated; ⫺, no existing data; NS, not significant. †Significant

338



Lete et al.

Table 4 Head-to-head studies evaluating effects of EE/LNG, EE/DRSP, E2V/DNG and E2/NOMAC on haemostasis, lipid and carbohydrate metabolism. This table also includes two studies that assessed effects of EE/DSG combinations on relevant markers. Relevant marker studied


Study and treatment group Oelkers et al. 199554 EE 15 μg, 20 μg, 30 μg/DRSP 3 mg (n ⫽ 20 women in each group) EE 30 μg/LNG 150 μg (n ⫽ 20) Middeldorp et al. 2000*40 EE 30 μg/LNG 150 μg EE 30 μg/DSG 150 μg Klipping & Marr 200547 EE 20 μg/DRSP 3 mg (n ⫽ 29) EE 20 μg/DSG 150 μg (n ⫽ 30) Klipping et al. 201148 E2V 1 mg, 2 mg, 3 mg/DNG 2 mg, 3 mg, placebo (n ⫽ 16) EE 30 μg/LNG 150 μg (n ⫽ 16) Junge et al. 201149 E2V 1 mg, 2 mg, 3 mg/DNG 2 mg, 3 mg, placebo (n ⫽ 30) EE 30 μg ,40 μg/LNG 50 μg, 75 μg, 125 μg (n ⫽ 28) Gaussem et al. 201150 E2 1.5 mg/NOMAC 2.5 mg (n ⫽ 45) EE 20 μg/LNG 100 μg (n ⫽ 45) Ågren et al. 201133 E2 1.5 mg/NOMAC 2.5 mg (n ⫽ 60) EE 30 μg/LNG 150 μg (n ⫽ 58)

Carbohydrate Haemostasis Lipid metabolism metabolism Acquired APC Insulin F 1 ⫹ 2 D-dimer resistance Protein S HDL-C LDL-C TG response Glucose NA

NA

NA

NA

⫹

⫹

⫹

⫺

⫹

⫹

⫺

⫺

⫺

NA

NA

NA

NA

NA

⫹

⫹

⫹

⫹

⫹

⫹

⫹

NA

NA

⫹

⫹

⫹

⫹

NA

NA

NA

NA

NA

⫹

⫹

⫹

⫹

⫹

⫹

⫹

⫹

⫹

⫹

⫹

⫹

⫹

NA

NA

NA

NA

NA

⫹

⫹

⫹

⫹

⫹

⫹

⫹

⫹

⫹

*33 women were recruited in this study. ⫹, assessed; ⫺, not assessed; NA, not appropriate.

At this point, it is useful to note that only a wellpowered and well-conceived prospective trial can accurately assess the risk of VTE or other adverse cardiovascular events associated with the use of any particular COC. Nonetheless, determining the effects of COCs on haemostasis parameters and other markers of thrombotic risk provides critical information because of the physiological association of such markers with thrombogenesis. Because of the low incidence of VTE in COC users, it takes several years and requires

many women to develop VTE before clinical data become available. In the meantime, markers of VTE risk can be helpful and cannot be overlooked66. The effects of COCs on metabolic factors, including haemostatic markers, can be modulated by the type and dose of estrogen and progestin20,67. EE contained in most COCs has been shown to increase hepatic protein levels more than E2, in particular proteins involved in haemostasis.This is the result of an estrogen rebound in blood levels, which has a strong impact on


339



Lete et al.

hepatic proteins. This extensive resorption of EE has been associated with an increased risk of VTE3,68,69. By contrast, the natural E2 causes no iterative liver processing64. Interestingly, no androgenic properties are required to counterbalance weak estrogenic effects of E2 at the hepatic level70,71, which is particularly appealing considering that androgenic progestins exert an adverse influence on lipid and carbohydrate metabolism. In this regard, when considering COCs with limited impact on lipid and carbohydrate metabolism, it appears that E2 combined with NOMAC has less impact than COCs containing LNG, and possibly E2V/DNG. Third-generation progestins have been shown to have better lipid profiles and promote less insulin resistance compared with second-generation progestins, but surprisingly, so far, they have not been found to reduce the risk of stroke and myocardial infarction. They may also be associated with an increased risk of VTE72. The use of progestin-only pills could solve the thrombotic complications, but these pills have poor cycle control, and some formulations are associated with lower contraceptive efficacy than COCs10,63,73. In summary, our search for a COC that would affect women’s haemostatic profile as little as possible, while providing convenience and acceptability, indicated that E2/NOMAC might be the most promising candidate (Table 3). In this preparation, both the estrogen and progestin components of the COC balance their actions in a ‘natural’ way20. E2 is less potent than EE at inducing the synthesis of estrogen-sensitive liver proteins.The physiological E2 dose in the E2/NOMAC combination provides cycle control and neutral metabolic impact. To ensure limited effects on the body, it has been coupled with NOMAC, which is referred to as a ‘pure progestational compound’. The pharmacological properties of NOMAC resemble those of progesterone to reduce or eliminate undesirable androgenic and mineralocorticoid effects on the body20,43,74. NOMAC is highly antigonadotropic and effectively inhibits ovulation inhibition and, in combination with a weak E2V dose, provides moderate endometrial atrophy20. In clinical practice, it is most important to avoid prescribing COCs to women at elevated risk of VTE. The WHO Medical Eligibility Criteria for

340

Contraceptive Use should serve as a first point of guidance for prescribers. Women who have a higher risk of VTE due to obesity, smoking, family history of VTE or cardiovascular disease should undergo a personal risk assessment and be advised appropriately. The hormonal contraceptive methods with the lowest VTE risk are progestin-only contraceptives75. For those women at a low baseline risk of VTE, the attempt to minimise the risk of VTE associated with the use of COCs can be further bolstered by the selection of a pill with the smallest impact on haemostasis. This should help avoid ‘crossing the VTE risk threshold’ in as many women as possible, including those with a high unknown congenital vulnerability to VTE. Contraceptive pills that provide limited metabolic effects on the body could contribute to further reduce the residual VTE risk currently observed with the use of COCs.

AC K N OW L E D G E M E N T S

The manuscript was prepared with writing and editorial assistance from Jean-Louis Carsol and Daniel Loniewski, both employed by Publicis Life Brands International. Declaration of interest: IL has a financial relationship (lecturer, member of advisory boards and/or consultant) with Bayer, MSD and Teva. NC-B shares expertise as a lecturer, member of advisory boards and/or consultant with HRA Pharma, Gedeon Richter and Teva/Theramex, without personal gain; she currently supervises a PhD student in receipt of a grant by HRA Pharma (October 2012–October 2015). CJ is a consultant with HRA Pharma, MSD and Teva/Theramex. SL is the recipient of grants and honoraria from MSD, Teva and Pfizer Inc. PL has received lecture fees, is a member of advisory boards and/or consultant with Bayer, Effick, HRA Pharma, MSD and Teva/Theramex. REN has had financial relationships (lecturer, member of advisory boards and/or consultant) with Bayer Pharma, Eli Lilly, Gedeon Richter, HRA Pharma, MSD, Novo Nordisk, Pfizer Inc., Shionogi and Teva. AP shares expertise as a lecturer, member of advisory boards and/or consultant with Bayer, Abbott, Amgen and Teva/Theramex, without personal gain.



Lete et al.


REFERENCES

1. van Hylckama Vlieg A, Helmerhorst FM, Vandenbroucke JP, et al. 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. 2. Kemmeren JM, Algra A, Grobbee DE. Third generation oral contraceptives and risk of venous thrombosis: Meta-analysis. BMJ 2001;323:131–9. 3. Jick SS, Hernandez RK. Risk of non-fatal venous thromboembolism in women using oral contraceptives containing drospirenone compared with women using oral contraceptives containing levonorgestrel: Casecontrol study using United States claims data. BMJ 2011;342:d2151. 4. Tchaikovski SN, Rosing J. Mechanisms of estrogeninduced venous thromboembolism. Thromb Res 2010; 126:5–11. 5. Drife J. Oral contraception and the risk of thromboembolism: What does it mean to clinicians and their patients? Drug Saf 2002;25:893–902. 6. Lidegaard Ø, Løkkegaard E, Svendsen AL, et al. Hormonal contraception and risk of venous thromboembolism: national follow up study. BMJ 2009:339:b2890. 7. World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.Venous thromboembolic disease and combined oral contraceptives: Results of international multicentre case-control study. Lancet 1995;346:1575–82. 8. Zöller B, García de Frutos P, Hillarp A, et al.Thrombophilia as a multigenic disease. Haematologica 1999;84:59–70. 9. Kottke-Marchant K. The role of coagulation in arterial and venous thrombosis. In Askari AT and Lincoff AM eds. Contemporary cardiology: Antithrombotic drug therapy in cardiovascular disease. Heidelberg, London, New York: Humana Press 2010:19–38. 10. Hannaford PC. Epidemiology of the contraceptive pill and venous thromboembolism. Thromb Res 2011;127: S30–4. 11. Anderson FA Jr, Spencer FA. Risk factors for venous thromboembolism. Circulation 2003;107(23 Suppl. 1): I9–16. 12. Bezemer ID, van der Meer FJ, Eikenboom JC, et al. The value of family history as a risk indicator for venous thrombosis. Arch Intern Med 2009;169:610–15. 13. Dehlendorf C, Levy K, Kelley A, et al. Women’s preferences for contraceptive counseling and decision making. Contraception 2013;88:250–6. 14. Guideline on clinical investigation of steroid contraceptives in women. London: European Medicines Agency. Committee for Medicinal Products for Human Use. EMEA/CPMP/EWP/519/98 Rev 1.

15. Ahrendt HJ, Makalová D, Parke S, et al. Bleeding pattern and cycle control with an estradiol-based oral contraceptive: A seven-cycle, randomized comparative trial of estradiol valerate/dienogest and ethinyl estradiol/ levonorgestrel. Contraception 2009;80:436–44. 16. WHO Collaborating Centre for Drug Statistics Methodology. Accessed on 26 May 2014 from: www. whocc.no/atc_ddd_index/?code ⫽ G03AA14 17. Martinelli I, Bucciarelli P, Mannucci PM. Thrombotic risk factors: Basic pathophysiology. Crit Care Med 2010;38(2 Suppl.):S3–9. 18. Versteeg HH, Heemskerk JWM, Levi M, et al. New fundamentals in hemostasis. Physiol Rev 2013;93:327–58. 19. von Bruhl ML, Stark K, Steinhart A, et al. Monocytes, neutrophils, and platelets cooperate to initiate and propagate venous thrombosis in mice in vivo. J Exp Med 2012;209:819–35. 20. Chabbert-Buffet N, Gerris J, Jamin C, et al. Toward a new concept of ‘natural balance’ in oral estroprogestin contraception. Gynecol Endocrinol 2013;29:891–6. 21. Sitruk-Ware R, Nath A. Metabolic effects of contraceptive steroids. Rev Endocr Metab Disord 2011;12:63–75. 22. Rott H. Contraception, venous thrombosis and biological plausibility. Minerva Med 2013;104:161–7. 23. Winkler UH. Blood coagulation and oral contraceptives. A critical review. Contraception 1998;57:203–9 24. Odlind V, Milsom I, Persson I, et al. Can changes in sex hormone binding globulin predict the risk of venous thromboembolism with combined oral contraceptive pills? Acta Obstet Gynecol Scand 2002;81:482–90. 25. Ariëns RA, de Lange M, Snieder H, et al. Activation markers of coagulation and fibrinolysis in twins: Heritability of the prethrombotic state. Lancet 2002;359:667–71. 26. Pabinger I, Ay C. Biomarkers and venous thromboembolism. Arterioscler Thromb Vasc Biol 2009;29:332–6. 27. Stanczyk FZ, Archer DF, Bhavnani BR. Ethinyl estradiol and 17β-estradiol in combined oral contraceptives: Pharmacokinetics, pharmacodynamics and risk assessment. Contraception 2013;87:706–27. 28. Parker WA. Estrogen-induced pancreatitis. Clin Pharm 1983;2:75–9. 29. Lee J, Goldberg IJ. Hypertriglyceridemia-induced pancreatitis created by oral estrogen and in vitro fertilization ovulation induction. J Clin Lipidol 2008;2(1):63–6. 30. Gaspard UJ. Metabolic effects of oral contraceptives. Am J Obstet Gynecol 1987;157:1029–41. 31. Krauss RM, Burkman RT Jr.The metabolic impact of oral contraceptives. Am J Obstet Gynecol 1992;167:1177–84. 32. Taneepanichskul S, Phupong V. Influence of a new oral contraceptive with drospirenone on lipid metabolism. Gynecol Endocrinol 2007;23:347–50.


341



Lete et al.

33. Ågren UM, Anttila M, Mäenpää-Liukko K, et al. Effects of a monophasic combined oral contraceptive containing nomegestrol acetate and 17β-oestradiol compared with one containing levonorgestrel and ethinylestradiol on haemostasis, lipids and carbohydrate metabolism. Eur J Contracept Reprod Health Care 2011;16:444–57. 34. Sitruk-Ware R. Progestins and cardiovascular risk markers. Steroids 2000;65:651–8. 35. Godsland IF, Crook D, Simpson R, et al. The effects of different formulations of oral contraceptive agents on lipid and carbohydrate metabolism. N Engl J Med 1990;323:1375–81. 36. Cagnacci A, Ferrari S, Tirelli A, et al. Insulin sensitivity and lipid metabolism with oral contraceptives containing chlormadinone acetate or desogestrel: A randomized trial. Contraception 2009;79:111–16. 37. Baird DT, Glasier AF. Hormonal contraception. N Engl J Med 1993;328:1543–9. 38. European Medicines Agency. Benefits of combined hormonal contraceptives (CHCs) continue to outweigh risk — CHMP endorses PRAC recommendation. EMA/709120/2013. London: European Medicines Agency. 39. Winkler UH. Hemostatic effects of third- and secondgeneration oral contraceptives: Absence of a causal mechanism for a difference in risk of venous thromboembolism. Contraception 2000;62(2 Suppl.):11S–20S. 40. Middeldorp S, Meijers JC, van den Ende AE, et al. Effects on coagulation of levonorgestrel- and desogestrelcontaining low dose oral contraceptives: A cross-over study. Thromb Haemost 2000;84:4–8. 41. Tans G, Curvers J, Middeldorp S, et al. A randomized cross-over study on the effects of levonorgestrel- and desogestrel-containing oral contraceptives on the anticoagulant pathways. Thromb Haemost 2000;84:15–21. 42. Nappi RE, Paoletti AM, Volpe A, et al. Multinational, multicentre, randomised, open-label study evaluating the impact of a 91-day extended regimen combined oral contraceptive, compared with two 28-day traditional combined oral contraceptives, on haemostatic parameters in healthy women. Eur J Contracept Reprod Health Care 2014;19:285–94. 43. Oral Contraceptive and Hemostasis Study Group. The effects of seven monophasic oral contraceptive regimens on hemostatic variables: Conclusions from a large randomized multicenter study. Contraception 2003; 67:173–85. 44. Jespersen J, Endrikat J, Düsterberg B, et al. A 1-year study to compare the hemostatic effects of oral contraceptive containing 20 microg of ethinylestradiol and 100 microg of levonorgestrel with 30 microg of ethinylestradiol and 100 microg of levonorgestrel. Contraception 2005;72: 98–104.

342

45. CPMP public assessment report. Combined oral contraceptives and venous thromboembolism. London: European Medicines Agency. Committee for Medicinal Products (CPMP). EMEA/CPMP/2201/01/en/Final. 46. de Bastos M, Stegeman BH, Rosendaal FR, et al. Combined oral contraceptives: Venous thrombosis. Cochrane Database Syst Rev 2014;3:CD010813. 47. Klipping C, Marr J. 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;71:409–16. 48. Klipping C, Duijkers I, Parke S, et al. Hemostatic effects of a novel estradiol-based oral contraceptive: An openlabel, randomized, crossover study of estradiol valerate/ dienogest versus ethinylestradiol/levonorgestrel. Drugs R D 2011;11:159–70. 49. Junge W, Mellinger U, Parke S, et al. Metabolic and haemostatic effects of estradiol valerate/dienogest, a novel oral contraceptive: A randomized, open-label, single-centre study. Clin Drug Investig 2011;31: 573–84. 50. Gaussem P, Alhenc-Gelas M, Thomas JL, et al. Haemostatic effects of a new combined oral contraceptive, nomegestrol acetate/17β-estradiol, compared with those of levonorgestrel/ethinyl estradiol. A double-blind, randomised study. Thromb Haemost 2011;105:560–7. 51. Wiegratz I, Lee JH, Kutschera E, et al. Effect of dienogest-containing oral contraceptives on lipid metabolism. Contraception 2002;65:223–9. 52. Krintus M, Sypniewska G, Kuligowska-Prusinska M. Effect of second and third generation oral contraceptives on C-reactive protein, lipids and apolipoproteins in young, non-obese, non-smoking apparently healthy women. Clin Biochem 2010;43:626–8. 53. Endrikat J, Klipping C, Cronin M, et al. An open label, comparative study of the effects of a dose-reduced oral contraceptive containing 20 microg ethinyl estradiol and 100 microg levonorgestrel on hemostatic, lipids, and carbohydrate metabolism variables. Contraception 2002; 65:215–21. 54. Oelkers W, Foidart JM, Dombrovicz N, et al. Effects of a new oral contraceptive containing an antimineralocorticoid progestogen, drospirenone, on the renin-aldosterone system, body weight, blood pressure, glucose tolerance, and lipid metabolism. J Clin Endocrinol Metab 1995; 80:1816–21. 55. Sitruk-Ware R. Pharmacology of different progestogens: The special case of drospirenone. Climacteric 2005; (8 Suppl. 3):4–12. 56. Godsland IF, Walton C, Felton C, et al. Insulin resistance, secretion, and metabolism in users of oral contraceptives. J Clin Endocrinol Metab 1992;74:64–70.




57. Knopp RH, Broyles FE, Cheung M, et al. Comparison of the lipoprotein, carbohydrate, and hemostatic effects of phasic oral contraceptives containing desogestrel or levonorgestrel. Contraception 2001;63:1–11. 58. Bergink EW, Kloosterboer HJ, Lund L, et al. Effects of levonorgestrel and desogestrel in low-dose oral contraceptive combinations on serum lipids, apolipoproteins A-I and B and glycosylated proteins. Contraception 1984;30:61–72. 59. Song S, Chen JK, Yang PJ, et al. A cross-over study of three oral contraceptives containing ethinyloestradiol and either desogestrel or levonorgestrel. Contraception 1992;45:523–32. 60. Hall D. The sociology of risk and social demographic change. PSC Discussion Papers Series 2001;15:1-12. Available at: http://ir.lib.uwo.ca/pscpapers/vol15/ iss12/1. 61. Dhont M. History of oral contraception. Eur J Contracept Reprod Health Care 2010;15:S12–18. 62. Gronich N, Lavi I, Rennert G. Higher risk of venous thrombosis associated with drospirenone-containing oral contraceptives: A population-based cohort study. CMAJ 2011;183:E1319–25. 63. Batur P, Elder J, Mayer M. Update on contraception: Benefits and risks of the new formulations. Cleve Clin J Med 2003;70:668–90. 64. Sitruk-Ware R. Pharmacological profile of progestins. Maturitas 2008;61:151–7. 65. Aye MM, Kilpatrick ES, Aburima A, et al. Acute hypertriglyceridemia induces platelet hyperactivity that is not attenuated by insulin in polycystic ovary syndrome. J Am Heart Assoc 2014;3:e000706. 66. van Vliet HA, Rosendaal FR, Rosing J, et al. Sex hormone-binding globulin: An adequate surrogate marker

Lete et al.

67.

68.

69.

70.

71. 72.

73.

74. 75.


for venous thromboembolism in women using new hormonal contraceptives. Contraception 2009;79:328–9. Blanco-Molina MA, Lozano M, Cano A, et al. Progestinonly contraception and venous thromboembolism. Thromb Res 2012;129:e257–62. Mashchak CA, Lobo RA, Dozono-Takano R, et al. Comparison of pharmacodynamic properties of various estrogen formulations. Am J Obstet Gynecol 1982;144: 511–18. van Vliet HA, Frolich M, Christella M, et al. 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;20:563–8. Schindler AE, Campagnoli C, Druckmann R, et al. Classification and pharmacology of progestins. Maturitas 2008;61:171–80. Coelingh Benninck HJT. Are all estrogens the same? Maturitas 2008;61:195–201. Lidegaard O, Nielsen LH, Skovlund CW, et al. Venous thrombosis in users of non-oral hormonal contraception: Follow-up study, Denmark 2001–10. BMJ 2012;344:e2990. Mantha S, Karp R, Raghavan V, et al. Assessing the risk of venous thromboembolic events in women taking progestin-only contraception: A meta-analysis. BMJ 2012;345:e4944 Lello S. Nomegestrol acetate: pharmacology, safety profile and therapeutic efficacy. Drugs 2010;70:541–59. Bitzer J, Amy JJ, Beerthuizen R, et al. Statement on combined hormonal contraceptives containing third- or fourth-generation progestogens or cyproterone acetate, and the associated risk of thromboembolism. J Fam Plann Reprod Health Care 2013;39:156–9.

343

Sleeping pills and insomnia.

Spleen Pills.

Sleeping pills and insomnia.

Sweet and sour pills.

Plague Pills.

Sleeping pills and insomnia.

"Pills" and the air passages: a continuum.

King of the pills.

Editorial: Pills for pimples.

Pills for personal problems.

Letter: Too many pills.

Death pills from pesticide.

'IQ pills' land in trouble.

Silver-Coated Pills-A Caution.

Study finds sleeping pills overprescribed.

Sleeping pills for psychiatric patients.

Progestin-only pills for contraception.

Editorial: Pills over the counter.

Pharmacogenetics of taste: turning bitter pills sweet?

Improvement interventions are social treatments, not pills.

Two techniques to make swallowing pills easier.

Renal infarction due to oral contraceptive pills.

Pricey pills for an even pricier problem.

Estrogen potency of oral contraceptive pills.