Correspondence
Hormone-Replacement Therapy and Coagulation
N Engl J Med 1997; 337:200-202July 17, 1997
- Article
To the Editor:
Koh et al. (March 6 issue)1 report that the administration of conjugated estrogen alone or with progestin reduced plasminogen-activator inhibitor type 1 (PAI-1) levels in postmenopausal women, and the authors speculate that this may contribute to the protective effect of hormone-replacement therapy against coronary artery disease. Whether hormone-replacement therapy increases or decreases the risk of thromboembolic episodes is controversial. Several authors have reported an increased incidence of thromboembolic episodes and blood-coagulation changes, resulting in a hypercoagulable state in patients undergoing hormone-replacement therapy. On the other hand, the improvement in lipid patterns in patients undergoing hormone-replacement therapy may have the opposite effect (pertinent references are cited by Koh et al.).
Our own studies,2-4 based in part on long-term observation of 759 patients, indicate that progestin alone has no significant effect, whereas estrogen increases the production of blood coagulation factors I, II, VII, IX, and X and decreases the production of antithrombin III; these are all prothrombotic changes. On the other hand, there was also an increase in plasminogen production, plasminogen-activator activity, and fibrinolytic potential, all of which are antithrombotic factors. In the majority of patients, these changes appear to counteract each other. Serious thromboembolic episodes occurred in one patient in the control (untreated) group and three patients undergoing estrogen treatment. All four had minor preexisting blood-coagulation abnormalities and inadequate compensatory changes during estrogen therapy. Interestingly, all four also had type A or AB blood, a phenomenon that deserves further study. Studies of the connection between the A blood group and thromboembolic episodes have been reviewed.4
It appears that treatment with low-dose estrogen, alone or with progestin, poses little danger and may provide a benefit with respect to the incidence of thromboembolic episodes in postmenopausal women.5 However, in women with a history of blood-coagulation problems or type A or AB blood, or both, pretreatment blood-coagulation studies, as well as follow-up studies, may be justified.
5 ReferencesClara M. Ambrus, M.D., Ph.D.
Julian L. Ambrus, M.D., Ph.D.
State University of New York at Buffalo, Buffalo, NY 142031
Koh KK ,Mincemoyer R ,Bui MN , et al. Effects of hormone-replacement therapy on fibrinolysis in postmenopausal women. N Engl J Med 1997;336:683-690
Full Text | Web of Science | Medline2
Ambrus JL ,Mink IB ,Courey NG , et al. Progestational agents and blood coagulation. VII. Thromboembolic and other complications of oral contraceptive therapy in relationship to pretreatment levels of blood coagulation factors: summary of a ten-year study. Am J Obstet Gynecol 1976;125:1057-1062
Web of Science | Medline3
Ambrus JL. Estrogens and clotting factors. In: van Keep PA, Greenblatt RB, Albeaux-Fernet M, eds. Consensus on menopause research: a summary of international opinion. Lancaster, England: MTP Press, 1976:55-8.
4
Ambrus JL, Ambrus CM, Lillie MA. Endocrine effects on blood coagulation. In: Ludwig H, Genz HJ, eds. Blutgerinnung und Gefesswand. Stuttgart, Germany: FK Schattauer, 1981:197-209.
5
Ambrus JL ,Courey NG ,Browne BJ ,Mink IB ,Moore RH ,Ambrus CM . Progestational agents and blood coagulation. VIII. Effect of low-dose, alternate-day, estrogen-progestin combinations on blood coagulation factors in man, with a special note on the effect of freezing of blood samples. Am J Obstet Gynecol 1977;128:161-166
Web of Science | Medline
To the Editor:
On the basis of a study of the effects of transdermal estrogen therapy given for four weeks, Koh and others conclude that this route of estrogen administration does not modify PAI-1 levels in postmenopausal women. However, I believe this conclusion must be considered with great caution.
In contrast to the effects of oral estrogen on various lipid values, which plateau after four to six weeks of therapy,1 the effects of transdermal estrogen on lipids are usually not manifested until after six weeks and are not maximal until at least six months have elapsed.2,3 Hence, this study was terminated before the expected beneficial effects on lipids would occur (i.e., a reduction of total and low-density lipoprotein [LDL] cholesterol and triglyceride levels); only a 5 percent reduction in the total cholesterol level was observed in this study. Therefore, it is also possible that the limited duration of this study is the reason that an effect on PAI-1 was not observed.
3 ReferencesSusan Davis, M.D., Ph.D.
Jean Hailes Foundation for Women, Melbourne 3168, Australia1
Darling GM, Johns JA, Davis SR. Comparing the effects of combined oestrogen and progesterone therapy with the effect of simvastatin in postmenopausal women with hypercholesterolaemia. In: Proceedings of the 8th International Congress on the Menopause, Sydney, Australia, November 3-7, 1996 (in press). abstract.
2
Studd JW ,McCarthy K ,Zamblera D , et al. Efficacy and tolerance of Menorest compared to Premarin in the treatment of postmenopausal women: a randomised, multicentre, double-blind, double-dummy study. Maturitas 1995;22:105-114
CrossRef | Web of Science | Medline3
Samsioe G, Anders A, Gunvor B, Kristina S, Gabriele H. Transdermal estradiol lowers cholesterol in moderately hypercholesterolemic women. In: Proceedings of the 8th International Congress on the Menopause, Sydney, Australia, November 3-7, 1996 (in press). abstract.
Author/Editor Response
The authors reply:
To the Editor: Drs. Ambrus and Ambrus refer to their previous investigations of the effect of oral-contraceptive hormones on hemostatic factors in women of reproductive age. In their series,1 the risk of thromboembolic events in women taking oral contraceptives appeared to be low, as it was in three recently reported observational studies of postmenopausal women receiving hormone therapy, which we cited in our article. Thus, potentiation of fibrinolysis may provide protection against procoagulant effects of conventional dosages of estrogen administered to women, regardless of their age or hormonal status.
We know of no data on the importance of the A or AB blood group in determining the risk of thromboembolic episodes in postmenopausal women. Drs. Ambrus and Ambrus recommend pretreatment blood-coagulation studies for women with these blood types. If this recommendation were followed, approximately 45 percent of postmenopausal women would undergo testing for rare risk factors for thrombosis, an effort unlikely to be cost effective. Instead, physicians should consider the thromboembolic risks of hormone therapy in postmenopausal women with a personal or family history of venous thrombosis (these women should be tested for an inherited or acquired risk factor for thrombosis before therapy), gross obesity, cancer, or intercurrent illness leading to immobilization.2
Dr. Davis claims that one month of administration of transdermal estradiol may have been insufficient for favorable effects on lipoprotein cholesterol and PAI-1 levels to be apparent, in contrast to the rapid effects of oral estrogen therapy on these values. In the peer-reviewed article she cites,3 the administration of transdermal estradiol reduced LDL cholesterol levels by approximately 7 percent from base-line values but also lowered high-density lipoprotein (HDL) cholesterol levels by approximately 2 percent after 12 weeks of therapy in 46 women. The administration of oral conjugated estrogen lowered LDL cholesterol levels by approximately 10 percent and increased HDL cholesterol levels by approximately 15 percent at 12 weeks in 50 women randomly assigned to this therapy. There was no placebo group or untreated group for the comparison of changes in lipoprotein cholesterol levels over time in this study. In the preliminary study cited by Dr. Davis,4 LDL cholesterol levels after 10 months of treatment with transdermal estradiol in 21 women with moderate hypercholesterolemia were not statistically different from LDL levels in 25 untreated controls; HDL cholesterol levels were not altered by therapy. Thus, long-term transdermal application of estrogen appears to have less favorable effects on proatherogenic (LDL) and antiatherogenic (HDL) lipoproteins than long-term oral administration of estrogen. We know of no data from randomized trials regarding the long-term effects of the transdermal administration of estrogen on PAI-1 levels in postmenopausal women.
4 ReferencesKwang Kon Koh, M.D.
Richard O. Cannon, III, M.D.
National Heart, Lung, and Blood Institute, Bethesda, MD 208921
Ambrus JL ,Courey NG ,Browne BJ ,Mink IB ,Moore RH ,Ambrus CM . Progestational agents and blood coagulation. VIII. Effect of low-dose, alternate-day, estrogen-progestin combinations on blood coagulation factors in man, with a special note on the effect of freezing of blood samples. Am J Obstet Gynecol 1977;128:161-166
Web of Science | Medline2
The British Committee for Standards in Haematology
CrossRef | Web of Science | Medline3
Studd JW ,McCarthy K ,Zamblera D , et al. Efficacy and tolerance of Menorest compared to Premarin in the treatment of postmenopausal women: a randomised multicentre, double-blind, double-dummy study. Maturitas 1995;22:105-114
CrossRef | Web of Science | Medline4
Samsioe G, Anders A, Gunvor B, Kristina S, Gabriele H. Transdermal estradiol lowers cholesterol in moderately hypercholesterolemic women. In: Proceedings of the 8th International Congress on the Menopause, Sydney, Australia, November 3-7, 1996 (in press). abstract.
