Summary and Discussion. Objectives of the thesis Since the introduction of the first contraceptive pill in 1959, the development of new hor- monal contraceptives has focused on maintaining the benefits of oral contraceptives while reducing their adverse effects. Four approaches have been used to optimize the risk-benefit profile: (i) lowering of the steroid dose; (ii) development of new formulas and schedules of administration; (iii) development of new steroids and (iv) development of new routes of administration. The first objective of this thesis was to compare the multiphasic schedule of administration of oral contraceptives with the classic monophasic schedule of admin- istration in terms of contraceptive effectiveness, bleeding pattern and discontinuation. The second objective was to predict the thrombotic risk of oral contraceptives containing the new steroid drospirenone by comparing the thrombin generation-based APC-resistance in users of pills containing drospirenone with the APC-resistance in users of pills containing other progestogens. We also focused on the biological basis of acquired APC-resistance in oral contraceptive users by studying the two main determinants of the thrombin generation- based APC-resistance test, free protein S and tissue factor pathway inhibitor free antigen. In addition, we tested the usefulness of sex hormone binding globulin as a new marker for the thrombotic risk of a hormonal contraceptive. The tಗird objective was to estimate the thrombotic risk of contraceptives which administer steroids vaginally, transdermally or intrauterine by assessing the effect of these contraceptives on thrombin generation-based APC-resistance. At East we evaluated whether varying levels of estradiol and progesterone during a natural menstrual cycle are associated with differences in APC-resistance. New formulas and schedules of administration In chapter 2 and chapter 3 we systematically reviewed the literature for randomized controlled trials comparing biphasic or triphasic oral contraceptives with monophasic oral contraceptives and planned to perform a meta-analysis for the outcomes contraceptive effectiveness, bleeding pattern and discontinuation. In chapter 4 we conducted a systematic review of trials comparing biphasic and triphasic oral contraceptives. The literature search yielded twenty-one trials comparing triphasic regimens with monophasic regimens, one trial comparing biphasic regimens with monophasic regimens and two trials comparing triphasic regimens with biphasic regimens. These studies provided insufficient evidence to determine whether the multiphasic approach differs from the monophasic approach in contraceptive effectiveness, bleeding patterns and discontinuation rates. Nor was it possible to adequately compare the triphasic approach with the biphasic approach. Overall, the reporting of the study methods was limited and the methodological quality of the studies were poor 1-10. Pooling of the data on contraceptive effectiveness or bleeding in a meta-analysis was gener- ally not possible due to (i) differences in progestogen type, progestogen dosage and estrogen Chapter 12 Summary, Discussion and Recommendations dosage of the studied contraceptive pills and (ii) differences between the trials in measuring, analyzing and reporting the data on cycle disturbances. The sample sizes of the individual trials were too small to detect differences in contraceptive effectiveness. Some trials included in the reviews reported favorable bleeding patterns, i.e. less spotting, breakthrough bleeding or amenorrhea, in users of triphasic oral contraceptives than in users of monophasic or biphasic oral contraceptives. However, in most of these trials, the progestogen type differed between the two studied oral contraceptives. Since the type of progestogen is thought to affect cycle control, the observed differences in bleeding pattern might be explained by the differences in progestogen content rather than the phasic approach 11,12. Further, several trials used the proportion of all cycles with spotting, breakthrough bleeding or amenorrhea as effect measure. This measure might give a distorted impression as one does not know whether a few women had all the cycles with bleeding problems or many women had a few cycles with bleeding problems. No differences in intermenstrual bleeding and amen- orrhea were observed between biphasic and monophasic oral contraceptives. The three approaches did not differ significantly in the number of women who discontinued due to bleeding disturbances and other minor side effects which can be considered as an indicator of how women tolerated the contraceptive method. In the absence of proven advantages of multiphasic oral contraceptives, the greater com- plexity of the multiphasic approach, and the higher costs of multiphasic oral contraceptives we recommend monophasic pills as the first choice for women starting oral contraceptive use. At first prescription, monophasic pills containing 30 yg estrogen are preferred over monophasic pills containing 20 yg estrogen since the latter cause more bleeding problems and discontinuation 13. Large, adequately reported, high-quality, randomized controlled trials that compare triphasic and monophasic oral contraceptives with identical proges- togens are needed to determine whether triphasic pills differ from monophasic pills in contraceptive effectiveness, bleeding patterns and continuation rates. Future studies should follow the recommendations of the Hormonal Contraceptives Trial Methodology Consensus Conference on recording of bleeding patterns 14 and the CONSORT guidelines on reporting of randomized controlled trials 2,3. Given the limited availability and use of biphasic oral contraceptives, additional trials comparing the biphasic approach with the monophasic approach are of low-priority. Nevertheless, all new contraceptive preparations require com- parison of effectiveness, frequent side effects and beneficial effects with the gold standard in adequately reported, high-quality randomized controlled trials. The presence of only two studies which were also of low methodological quality made it impossible to adequately compare triphasic pills with biphasic pills and to recommend one of the two approaches. Considering the sparse use of biphasic oral contraceptives, additional comparative trials are of low importance. In chapter 5 we investigated the frequency of quality assessment of randomized con- analysis. We included new systematic reviews of at least five trials of therapeutic or preven- tive interventions that appeared in issue 2, 2003, of the Cochrane Database of Systematic Reviews and all systematic reviews in the 2002 issues of the general and internal medicine journals Annals of Internal Medicine, BMJ, JAMA, and Lancet. Trial quality was assessed in all Cochrane reviews and most of the paper reviews. However, only half of the reviews incorporated the results of the quality assessment in the analysis, with no substantial dif- ▇▇▇▇▇▇▇ between ▇▇▇▇▇▇▇▇ and paper reviews. These findings were confirmed by a similar literature study 15. Since isolated quality assessments without incorporation of its results in the analysis are futile, we advise authors, peer-reviewers, and editors not to exclusively focus on whether quality assessment has been performed, but also to concentrate on incorporation of quality assessments in the analysis of systematic reviews. New steroids In chapter 6 we assessed the thrombotic profile of a new combined oral contraceptive con- taining drospirenone by comparing the resistance to APC in users of drospirenone-pills with the resistance to APC in users of pills containing second- or third-generation progestogens or cyproterone acetate. In addition, we measured the resistance to APC in women who switched from a drospirenone-containing pill to a second-generation oral contraceptive containing levonorgestrel or who switched from a second- or third-generation oral contraceptive to the drospirenone-containing pill. The resistance to APC was determined by quantifying the effect of APC on thrombin generation (thrombin generation-based or ETP-based APC- resistance test). The study demonstrated that women using oral contraceptives containing drospirenone were more resistant to APC than women using second-generation oral contra- ceptives containing levonorgestrel. The APC-resistance in users of drospirenone-containing pills was similar to the APC-resistance in users of pills with an increased risk of thrombosis, i.e. pills containing desogestrel or cyproterone acetate 16-19. In the group of women who switched from oral contraceptive type the resistance to APC altered correspondingly, so it increased when switching from a levonorgestrel-pill to a drospirenone-pill and decreased when switching from drospironone-pill to a levonorgestrel-pill. This study confirmed that the thrombin generation-based APC-resistance test discriminates well between oral contracep- tives with a high thrombotic risk (desogestrel and cyproterone acetate) and with a lower risk of thrombosis (levonorgestrel) 16-20. The higher resistance to APC in users of drospirenone- containing pills compared to users of levonorgestrel-containing pills suggests an increased risk of thrombosis for drospirenone-containing oral contraceptives. The predicted increased risk of thrombosis associated with oral contraceptives containing drospirenone was confirmed in two independent studies and contradicted in two studies sponsored by the manufacturer of the drospirenone-containing pill. A population based case-control study including 1524 patients with a venous thrombosis of the leg or pulmonary Chapter 12 Summary, Discussion and Recommendations embolism and 1760 controls observed an 1.7-fold increased risk of thrombosis (95% CI 0.7 to 3.9) for oral contraceptives containing drospirenone compared to oral contraceptives con- taining levonorgestrel 21. A national cohort study comprising 10.4 million women-years of observation found the risk of thrombosis associated with contraceptive pills containing dro- spirenone to be increased by a factor 1.64 (95% CI 1.27 to 2.10) compared to contraceptive pills containing levonorgestrel 22. No difference in the risk of thrombosis between oral con- traceptives containing drospirenone and oral contraceptives containing levonorgestrel was observed in a cohort study of 16.534 women using drospirenone-containing pills followed for 28.621 women-years of observation and 15.428 women using levonorgestrel-containing pills followed for 31.415 years of observation (Hazard ratio drospirenone versus levonorg- estrel 1.0; 95% CI 0.6 to 1.8) 23. A cohort study of 22.429 users of drospirenone-containing oral contraceptives followed for 14.081 women-years and 44.858 users of oral contracep- tives containing 'other progestogens' followed for 27.575 women-years observed no differ- ence in thrombotic risk between the preparations (Hazard ratio drospirenone versus other progestogens 0.9; 95% CI 0.5 to 1.6) 24. However, the sample size of the study by ▇▇▇▇▇▇ et al. was too small to detect a difference in thrombotic risk between drospirenone-containing and levonorgestrel-containing pills and the confidence interval did not exclude the 1.6 to 1.7-fold increased risk observed in the two studies mentioned above. In order to demonstrate a doubling of risk of venous thrombosis between two different oral contraceptives about 600.000 women should be followed for 1 year 18. In the study by ▇▇▇▇▇▇ et al. the thrombotic risk of drospirenone-containing pills was compared with the thrombotic risk of pills contain- ing 'other progestogens' which most likely also include progestogens with an increased risk of thrombosis. The primary comparator should have been an oral contraceptive with a low risk of thrombosis, i.e. containing the second-generation progestogen levonorgestrel. In addition, both studies where funded by the manufacturer of the drospirenone-pill 23,24. Studies sponsored by pharmaceutical companies are more likely to have outcomes favoring the sponsor than studies funded by other sources 9,10. An example of the impact of sponsor- ship on contraceptive research outcomes emerges from a meta-analyses by ▇▇▇▇▇▇▇▇ et al. which observed that the difference in thrombotic risk between second- and third-generation oral contraceptives was less in trials sponsored by the pharmaceutical industry compared to studies that were financed with public funds or through charities 19. No differences in contraceptive effectiveness, bleeding pattern and minor adverse effects have been observed between oral contraceptives containing drospirenone or levonorgestrel 25. Suggested beneficial effects on weight, acne, hirsutism and premenstrual syndrome of the drospirenone-pill in comparison with a levonorgestrel-pill were not confirmed by Cochrane reviews 26-29. In the absence of proven advantages of the drospirenone-containing oral contraceptive, an increased risk of thrombosis, and the higher costs of the drospirenone-pill we advise not to prescribe this oral contraceptive as a first choice for women starting oral contraceptives. Due to the low incidence of venous thrombosis in users of hormonal contraceptives, assessment of the thrombotic risk necessitates studies including hundreds of thousands of women or case-control studies. In a development program for market authorization of a new hormonal contraceptive evaluation of the thrombotic risk in a clinical trial is virtually impos- sible. We confirmed that the thrombin generation-based APC-resistance test discriminates well between oral contraceptives with a high and with a low risk of thrombosis. In agreement with the European Medicines Agency Guideline on Clinical Investigation of Steroid Contra- ceptives in Women we advocate that before a new hormonal contraceptive is marketed the effect of the preparation on the resistance to APC measured with the thrombin generation- based APC-resistance test is determined to estimate the thrombotic risk 30. Preferably, the new hormonal contraceptive is compared with a combined oral contraceptive with the low- est thrombotic risk, i.e. a pill containing the second-generation progestogen levonorgestrel. In chapter 7 we focused on the biological basis of acquired APC-resistance in oral contraceptive users and the different effects of various types of oral contraceptives on the resistance to APC by studying the two main determinants of the thrombin generation-based APC-resistance test, free protein S and tissue factor pathway inhibitor (TFPI) free antigen 31. In this study we observed that oral contraceptives containing different progestogens had different effects on free protein S and TFPI free antigen. Women using third-generation oral contraceptives containing desogestrel, which are known to double the risk of thrombosis compared to second-generation oral contraceptives containing levonorgestrel, had lower free protein S and TFPI free antigen levels than women using oral contraceptives contain- ing levonorgestrel 16-19. Free protein S and TFPI antigen levels in users of drospirenone- containing oral contraceptives were lower than in users of levonorgestrel-containing oral contraceptives and similar to those in users of third-generation oral contraceptives. Women using oral contraceptives containing cyproterone acetate, which are known to increase the thrombotic risk 4-fold compared to oral contraceptives containing levonorgestrel, had the lowest free protein S and TFPI free antigen levels 20. Low free protein S and low TFPI free antigen levels were associated with an increased resistance to APC. These results indicate that the observed differences in APC-resistance induced by oral contraceptives containing different progestogens and the differences in thrombotic risk can at least in part be explained by different effects of oral contraceptives on free protein S and TFPI free antigen. The low free protein S and TFPI free antigen levels in women using oral contracep- tives with a high thrombotic risk is in concordance with other studies on the effect of oral contraceptives on protein S and TFPI 32-36. We did not investigate the mechanism for the decrease in free protein S and TFPI free antigen caused by oral contraceptives. ▇▇▇▇▇▇▇▇ et al. explained the differences in
Appears in 2 contracts
Sources: Not Applicable, Not Applicable
Summary and Discussion. Objectives of the thesis Since the introduction of the first contraceptive pill in 1959, the development of new hor- monal contraceptives has focused on maintaining the benefits of oral contraceptives while reducing their adverse effects. Four approaches have been used to optimize the risk-benefit profile: (i) lowering of the steroid dose; (ii) development of new formulas and schedules of administration; (iii) development of new steroids and (iv) development of new routes of administration. The first objective of this thesis was to compare the multiphasic schedule of administration of oral contraceptives with the classic monophasic schedule of admin- istration in terms of contraceptive effectiveness, bleeding pattern and discontinuation. The second objective was to predict the thrombotic risk of oral contraceptives containing the new steroid drospirenone by comparing the thrombin generation-based APC-resistance in users of pills containing drospirenone with the APC-resistance in users of pills containing other progestogens. We also focused on the biological basis of acquired APC-resistance in oral contraceptive users by studying the two main determinants of the thrombin generation- based APC-resistance test, free protein S and tissue factor pathway inhibitor free antigen. In addition, we tested the usefulness of sex hormone binding globulin as a new marker for the thrombotic risk of a hormonal contraceptive. The tಗird objective was to estimate the thrombotic risk of contraceptives which administer steroids vaginally, transdermally or intrauterine by assessing the effect of these contraceptives on thrombin generation-based APC-resistance. At East we evaluated whether varying levels of estradiol and progesterone during a natural menstrual cycle are associated with differences in APC-resistance. New formulas and schedules of administration In chapter 2 and chapter 3 we systematically reviewed the literature for randomized controlled trials comparing biphasic or triphasic oral contraceptives with monophasic oral contraceptives and planned to perform a meta-analysis for the outcomes contraceptive effectiveness, bleeding pattern and discontinuation. In chapter 4 we conducted a systematic review of trials comparing biphasic and triphasic oral contraceptives. The literature search yielded twenty-one trials comparing triphasic regimens with monophasic regimens, one trial comparing biphasic regimens with monophasic regimens and two trials comparing triphasic regimens with biphasic regimens. These studies provided insufficient evidence to determine whether the multiphasic approach differs from the monophasic approach in contraceptive effectiveness, bleeding patterns and discontinuation rates. Nor was it possible to adequately compare the triphasic approach with the biphasic approach. Overall, the reporting of the study methods was limited and the methodological quality of the studies were poor 1-10. Pooling of the data on contraceptive effectiveness or bleeding in a meta-analysis was gener- ally not possible due to (i) differences in progestogen type, progestogen dosage and estrogen Chapter 12 Summary, Discussion and Recommendations dosage of the studied contraceptive pills and (ii) differences between the trials in measuring, analyzing and reporting the data on cycle disturbances. The sample sizes of the individual trials were too small to detect differences in contraceptive effectiveness. Some trials included in the reviews reported favorable bleeding patterns, i.e. less spotting, breakthrough bleeding or amenorrhea, in users of triphasic oral contraceptives than in users of monophasic or biphasic oral contraceptives. However, in most of these trials, the progestogen type differed between the two studied oral contraceptives. Since the type of progestogen is thought to affect cycle control, the observed differences in bleeding pattern might be explained by the differences in progestogen content rather than the phasic approach 11,12. Further, several trials used the proportion of all cycles with spotting, breakthrough bleeding or amenorrhea as effect measure. This measure might give a distorted impression as one does not know whether a few women had all the cycles with bleeding problems or many women had a few cycles with bleeding problems. No differences in intermenstrual bleeding and amen- orrhea were observed between biphasic and monophasic oral contraceptives. The three approaches did not differ significantly in the number of women who discontinued due to bleeding disturbances and other minor side effects which can be considered as an indicator of how women tolerated the contraceptive method. In the absence of proven advantages of multiphasic oral contraceptives, the greater com- plexity of the multiphasic approach, and the higher costs of multiphasic oral contraceptives we recommend monophasic pills as the first choice for women starting oral contraceptive use. At first prescription, monophasic pills containing 30 yg estrogen are preferred over monophasic pills containing 20 yg estrogen since the latter cause more bleeding problems and discontinuation 13. Large, adequately reported, high-quality, randomized controlled trials that compare triphasic and monophasic oral contraceptives with identical proges- togens are needed to determine whether triphasic pills differ from monophasic pills in contraceptive effectiveness, bleeding patterns and continuation rates. Future studies should follow the recommendations of the Hormonal Contraceptives Trial Methodology Consensus Conference on recording of bleeding patterns 14 and the CONSORT guidelines on reporting of randomized controlled trials 2,3. Given the limited availability and use of biphasic oral contraceptives, additional trials comparing the biphasic approach with the monophasic approach are of low-priority. Nevertheless, all new contraceptive preparations require com- parison of effectiveness, frequent side effects and beneficial effects with the gold standard in adequately reported, high-quality randomized controlled trials. The presence of only two studies which were also of low methodological quality made it impossible to adequately compare triphasic pills with biphasic pills and to recommend one of the two approaches. Considering the sparse use of biphasic oral contraceptives, additional comparative trials are of low importance. In chapter 5 we investigated the frequency of quality assessment of randomized con- analysis. We included new systematic reviews of at least five trials of therapeutic or preven- tive interventions that appeared in issue 2, 2003, of the Cochrane Database of Systematic Reviews and all systematic reviews in the 2002 issues of the general and internal medicine journals Annals of Internal Medicine, BMJ, JAMA, and Lancet. Trial quality was assessed in all Cochrane reviews and most of the paper reviews. However, only half of the reviews incorporated the results of the quality assessment in the analysis, with no substantial dif- ▇▇▇▇▇▇▇ between ▇▇▇▇▇▇▇▇ and paper reviews. These findings were confirmed by a similar literature study 15. Since isolated quality assessments without incorporation of its results in the analysis are futile, we advise authors, peer-reviewers, and editors not to exclusively focus on whether quality assessment has been performed, but also to concentrate on incorporation of quality assessments in the analysis of systematic reviews. New steroids In chapter 6 we assessed the thrombotic profile of a new combined oral contraceptive con- taining drospirenone by comparing the resistance to APC in users of drospirenone-pills with the resistance to APC in users of pills containing second- or third-generation progestogens or cyproterone acetate. In addition, we measured the resistance to APC in women who switched from a drospirenone-containing pill to a second-generation oral contraceptive containing levonorgestrel or who switched from a second- or third-generation oral contraceptive to the drospirenone-containing pill. The resistance to APC was determined by quantifying the effect of APC on thrombin generation (thrombin generation-based or ETP-based APC- resistance test). The study demonstrated that women using oral contraceptives containing drospirenone were more resistant to APC than women using second-generation oral contra- ceptives containing levonorgestrel. The APC-resistance in users of drospirenone-containing pills was similar to the APC-resistance in users of pills with an increased risk of thrombosis, i.e. pills containing desogestrel or cyproterone acetate 16-19. In the group of women who switched from oral contraceptive type the resistance to APC altered correspondingly, so it increased when switching from a levonorgestrel-pill to a drospirenone-pill and decreased when switching from drospironone-pill to a levonorgestrel-pill. This study confirmed that the thrombin generation-based APC-resistance test discriminates well between oral contracep- tives with a high thrombotic risk (desogestrel and cyproterone acetate) and with a lower risk of thrombosis (levonorgestrel) 16-20. The higher resistance to APC in users of drospirenone- containing pills compared to users of levonorgestrel-containing pills suggests an increased risk of thrombosis for drospirenone-containing oral contraceptives. The predicted increased risk of thrombosis associated with oral contraceptives containing drospirenone was confirmed in two independent studies and contradicted in two studies sponsored by the manufacturer of the drospirenone-containing pill. A population based case-control study including 1524 patients with a venous thrombosis of the leg or pulmonary Chapter 12 Summary, Discussion and Recommendations embolism and 1760 controls observed an 1.7-fold increased risk of thrombosis (95% CI 0.7 to 3.9) for oral contraceptives containing drospirenone compared to oral contraceptives con- taining levonorgestrel 21. A national cohort study comprising 10.4 million women-years of observation found the risk of thrombosis associated with contraceptive pills containing dro- spirenone to be increased by a factor 1.64 (95% CI 1.27 to 2.10) compared to contraceptive pills containing levonorgestrel 22. No difference in the risk of thrombosis between oral con- traceptives containing drospirenone and oral contraceptives containing levonorgestrel was observed in a cohort study of 16.534 women using drospirenone-containing pills followed for 28.621 women-years of observation and 15.428 women using levonorgestrel-containing pills followed for 31.415 years of observation (Hazard ratio drospirenone versus levonorg- estrel 1.0; 95% CI 0.6 to 1.8) 23. A cohort study of 22.429 users of drospirenone-containing oral contraceptives followed for 14.081 women-years and 44.858 users of oral contracep- tives containing 'other progestogens' followed for 27.575 women-years observed no differ- ence in thrombotic risk between the preparations (Hazard ratio drospirenone versus other progestogens 0.9; 95% CI 0.5 to 1.6) 24. However, the sample size of the study by ▇▇▇▇▇▇ et al. was too small to detect a difference in thrombotic risk between drospirenone-containing and levonorgestrel-containing pills and the confidence interval did not exclude the 1.6 to 1.7-fold increased risk observed in the two studies mentioned above. In order to demonstrate a doubling of risk of venous thrombosis between two different oral contraceptives about 600.000 women should be followed for 1 year 18. In the study by ▇▇▇▇▇▇ et al. the thrombotic risk of drospirenone-containing pills was compared with the thrombotic risk of pills contain- ing 'other progestogens' which most likely also include progestogens with an increased risk of thrombosis. The primary comparator should have been an oral contraceptive with a low risk of thrombosis, i.e. containing the second-generation progestogen levonorgestrel. In addition, both studies where funded by the manufacturer of the drospirenone-pill 23,24. Studies sponsored by pharmaceutical companies are more likely to have outcomes favoring the sponsor than studies funded by other sources 9,10. An example of the impact of sponsor- ship on contraceptive research outcomes emerges from a meta-analyses by ▇▇▇▇▇▇▇▇ et al. which observed that the difference in thrombotic risk between second- and third-generation oral contraceptives was less in trials sponsored by the pharmaceutical industry compared to studies that were financed with public funds or through charities 19. No differences in contraceptive effectiveness, bleeding pattern and minor adverse effects have been observed between oral contraceptives containing drospirenone or levonorgestrel 25. Suggested beneficial effects on weight, acne, hirsutism and premenstrual syndrome of the drospirenone-pill in comparison with a levonorgestrel-pill were not confirmed by Cochrane reviews 26-29. In the absence of proven advantages of the drospirenone-containing oral contraceptive, an increased risk of thrombosis, and the higher costs of the drospirenone-pill we advise not to prescribe this oral contraceptive as a first choice for women starting oral contraceptives. Due to the low incidence of venous thrombosis in users of hormonal contraceptives, assessment of the thrombotic risk necessitates studies including hundreds of thousands of women or case-control studies. In a development program for market authorization of a new hormonal contraceptive evaluation of the thrombotic risk in a clinical trial is virtually impos- sible. We confirmed that the thrombin generation-based APC-resistance test discriminates well between oral contraceptives with a high and with a low risk of thrombosis. In agreement with the European Medicines Agency Guideline on Clinical Investigation of Steroid Contra- ceptives in Women we advocate that before a new hormonal contraceptive is marketed the effect of the preparation on the resistance to APC measured with the thrombin generation- based APC-resistance test is determined to estimate the thrombotic risk 30. Preferably, the new hormonal contraceptive is compared with a combined oral contraceptive with the low- est thrombotic risk, i.e. a pill containing the second-generation progestogen levonorgestrel. In chapter 7 we focused on the biological basis of acquired APC-resistance in oral contraceptive users and the different effects of various types of oral contraceptives on the resistance to APC by studying the two main determinants of the thrombin generation-based APC-resistance test, free protein S and tissue factor pathway inhibitor (TFPI) free antigen 31. In this study we observed that oral contraceptives containing different progestogens had different effects on free protein S and TFPI free antigen. Women using third-generation oral contraceptives containing desogestrel, which are known to double the risk of thrombosis compared to second-generation oral contraceptives containing levonorgestrel, had lower free protein S and TFPI free antigen levels than women using oral contraceptives contain- ing levonorgestrel 16-19. Free protein S and TFPI antigen levels in users of drospirenone- containing oral contraceptives were lower than in users of levonorgestrel-containing oral contraceptives and similar to those in users of third-generation oral contraceptives. Women using oral contraceptives containing cyproterone acetate, which are known to increase the thrombotic risk 4-fold compared to oral contraceptives containing levonorgestrel, had the lowest free protein S and TFPI free antigen levels 20. Low free protein S and low TFPI free antigen levels were associated with an increased resistance to APC. These results indicate that the observed differences in APC-resistance induced by oral contraceptives containing different progestogens and the differences in thrombotic risk can at least in part be explained by different effects of oral contraceptives on free protein S and TFPI free antigen. The low free protein S and TFPI free antigen levels in women using oral contracep- tives with a high thrombotic risk is in concordance with other studies on the effect of oral contraceptives on protein S and TFPI 32-36. We did not investigate the mechanism for the decrease in free protein S and TFPI free antigen caused by oral contraceptives. ▇▇▇▇▇▇▇▇ Kemmeren et al. explained the differences in
Appears in 1 contract
Sources: Thesis