SEARCH

SEARCH BY CITATION

Keywords:

  • oestrogen replacement therapy;
  • blood coagulation;
  • venous thrombosis;
  • protein C;
  • antithrombin III

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The effects of post-menopausal hormone therapy (HRT) on blood coagulation in elderly women are not well defined. We studied associations of HRT use with levels of natural anticoagulant proteins in a cross-sectional study of 3393 women ≥ 65 years of age participating in the Cardiovascular Health Study. Protein C antigen and antithrombin were measured in all users (n = 230 unopposed oestrogen; 60 oestrogen/progestin) and a comparison group of 196 age- and race-matched non-users. Compared with non-users, oestrogen use was associated with higher protein C (4·80 vs. 4·30 µg/ml, P < 0·01). Results were similar for oestrogen/progestin (P > 0·05). In both user groups, antithrombin was lower than in non-users (109% for each vs. 115% in non-users, P < 0·001). Adjustment for factors related to prescription of HRT and to anticoagulant protein levels had little impact on the results. For antithrombin, associations with HRT were larger for thinner Caucasian women and black women. Venous thrombosis from HRT may be mediated partly by alterations in antithrombin, but not protein C concentrations. This study extends previous observations to older women, the group at highest risk of venous thromboembolism. Studies of HRT-induced alterations in anticoagulant function in relation to the occurrence of thrombosis with HRT are required.

Post-menopausal hormone replacement therapy (HRT) increases the risk of venous thromboembolism (Daly et al, 1996; Grodstein et al, 1996; Jick et al, 1996; Grady et al, 2000). The mechanisms underlying this association are unclear; however, lower levels of protein C and antithrombin deficiency are risk factors for first venous thrombosis (Koster et al, 1995), so these factors may be involved. Also, in an in vitro system, changes in antithrombin concentration within the normal range are a major determinant of thrombin production (Butenas et al, 1999), suggesting that modest drug effects on this protein might have clinical implications.

The literature on associations of oral HRT with these proteins is not conclusive and consists of small trials and one large cross-sectional study. In the experimental studies, oral HRT was associated with lower antithrombin and higher, lower or unchanged protein C (Caine et al, 1992; Kroon et al, 1994; Scarabin et al, 1997; van Baal et al, 2000). Two other trials, including one larger study, evaluated transdermal oestradiol plus oral progestin and reported no change in protein C with treatment (The Writing Group for the Estradiol Clotting Factors Study, 1996; Scarabin et al, 1997). In the large cross-sectional analysis from the Atherosclerosis Risk in Communities (ARIC) study, in middle-aged post-menopausal women, protein C antigen was higher with both unopposed oestrogen and combined therapy compared with non-use. Women on unopposed oestrogen had the highest concentrations. Antithrombin levels were lower in both types of HRT users (Nabulsi et al, 1993). Women in older age groups, who are at the highest risk of venous thrombosis (Goldhaber, 1994), have not been studied sufficiently. Age is a risk factor for venous thrombosis, and there is a modest decline in protein C, but no change in antithrombin with ageing in women (Sakkinen et al, 1998).

We studied the cross–sectional associations of oestrogen use with concentrations of antithrombin and protein C in female participants of the Cardiovascular Health Study, a cohort study of 5888 men and women aged 65 and over. Associations of HRT with cardiac risk factors and coagulation factors were reported previously (Manolio et al, 1993; Cushman et al, 1999a). Associations of protein C and antithrombin with cardiac risk factors and coagulation factors were also reported (Sakkinen et al, 1998). In the latter report, among 26 HRT users, we observed lower concentrations of antithrombin, but not protein C. That study lacked sufficient power to determine the independence of associations from other risk factors, so this analysis included data on 290 HRT users.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Subjects The Cardiovascular Health Study is a longitudinal population-based cohort of 5888 men and women recruited in 1989–90 or 1992–3 from four field centres (Fried et al, 1991). Subjects for this analysis were selected from among 3393 women participating in the 1992–3 examination. All participants provided informed consent with procedures approved by the institutional review committees at each site.

Women were excluded if they had a history of clinical vascular diseases, were taking warfarin or transdermal oestrogen or did not have a plasma sample available from the 1992–3 examination. Vascular disease was defined as confirmed baseline presence of cardiovascular, cerebrovascular or peripheral vascular disease using previously published criteria (Psaty et al, 1995) or as incident vascular disease events identified between the baseline and 1992–3 examination years. There were 19 HRT users (17 oestrogen, two combined therapy) who were excluded for lack of a blood sample. After exclusions, there were 230 women taking unopposed oestrogen and 60 taking oestrogen with progestin. A comparison group of 196 women was selected randomly from among women not currently using hormones. These women were frequency matched with users on 5-year age strata, clinic site and race.

Definitions Hormone use was established by medication inventory. Duration of current or past use, hysterectomy and oophorectomy status and smoking status were assessed by interview. Ankle brachial index, waist–hip ratio and body mass index were measured (Fried et al, 1991). Hypertension was defined as absent, borderline (systolic pressure 140–160 mmHg or diastolic pressure 90–95 mmHg) or present (systolic pressure > 160 mmHg or diastolic pressure > 95 mmHg or self-reported hypertension in association with the use of antihypertensive medication). Diabetes was defined as fasting glucose over 7·7 mmol/l or use of insulin or oral hypoglycaemic agents.

Laboratory analyses A fasting morning blood sample was collected in 1992–3 with minimal stasis as described previously (Cushman et al, 1995). Samples were stored in a central laboratory at the University of Vermont, at −70°C, and analysed in batches in 1997. Protein C was measured by immunoassay (Howard et al, 1988) with a coefficient of variation of 14·4%. Antithrombin was measured using a chromogenic assay with a coefficient of variation of 7·0% (Cushman et al, 1995).

Statistical analysis SPSS 6·1.3 was used for data analysis, using the Cardiovascular Health Study database from the 1992–3 examination, updated on 3 October 1996. The characteristics of each user group and non-users were compared using chi-squared tests or Fisher's exact test for categorical data and t-tests for continuous data. As protein C had a skewed distribution, crude associations between hormone use and the coagulation factors were determined by the Wilcoxon rank sum test. To assess the impact of other factors on relationships between hormone use and coagulation factors, for each user group (oestrogen alone; oestrogen plus progestin), multivariate linear models were developed using each coagulation factor as the dependent variable, with hormone user status (yes or no) forced into the model and covariates entered in stepwise fashion (P-value for entry 0·10, for removal 0·15). A priori-determined subgroup analyses assessed differences in associations by race and body mass index, a known risk factor for venous thrombosis (Jick et al, 1996; Goldhaber et al, 1997).

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Subjects

Characteristics of the hormone user groups have been reported previously and are summarized briefly in Table I (Cushman et al, 1999a). Most oestrogen or combined therapy users were taking conjugated equine oestrogens as Premarin (92% and 90% respectively). Among combined therapy users, 24 out of 62 (39%) were taking continuous progestin, and the remainder used various cyclical dosing regimens. Self-reported duration of use was long for both current user groups: 18·4 years for oestrogen and 12·0 years for combined therapy.

Table I.   Characteristics of study groups.
Mean (SD), frequency (%) or median (interquartile range)Oestrogen (n = 230)Oestrogen + progestin (n = 60)Non-users (n = 196)
  • *

    P < 0·05;

  • **

    P < 0·01;

  • ***

    P < 0·001; P-values are for comparing each user group with the control group.

Age (years)72·9 (4·0)72·6 (3·9)72·8 (3·9)
Years of use18·4 (11·6)12·0 (9·0)NA
Race (white)198 (86%)57 (95%)172 (88%)
Income ($12 000–34,999/≥ $35 000)117 (56%)/58 (28%)*28 (51%)/24 (44%)***88 (49%)/43 (24%)
Smoking (current)21 (9%)7 (12%)24 (13%)
Body mass index25·8 (4·0)***25·4 (4·7)**27·5 (5·3)
Hypertension (borderline/yes)35 (15%)/82 (36%)8 (13%)/17 (28%)28 (14%)/82 (42%)
Hysterectomy202 (89%)***5 (8%)***66 (34%)
Oophorectomy120 (57%)***5 (8%)*37 (21%)
Protein C (µg/ml)4·80 (3·90–5·80)*4·55 (3·43–5·30)4·30 (3·70–5·40)
Antithrombin (U/ml)1·09 (1·03–1·17)**1·09 (1·02–1·17)**1·15 (1·07–1·22)

Hormone use and anticoagulant proteins

Table I also shows the median protein C and antithrombin concentrations by hormone use status. Protein C antigen was higher in both types of hormone users than in non-users, but this difference was statistically significant only for unopposed oestrogen (4·80 µg/ml vs. 4·30 µg/ml, P < 0·01). This difference represented an 11·6% higher level. Conversely, antithrombin concentrations were lower in both types of hormone users than in non-users (109% in each user group vs. 115% in non-users, P < 0·001). This difference represented a 5·2% lower level. These results were confirmed statistically using a more stringent analysis (three-way analysis of variance with Bonferroni-corrected P-values for multiple comparisons). For further illustration of the intergroup differences, combining the two user groups, Fig 1 depicts the distribution of antithrombin according to user status.

image

Figure 1.  Frequency distribution of antithrombin concentration by hormone use status. Histograms of the distribution of antithrombin concentration by hormone use status illustrate higher values for antithrombin concentration in hormone non-users (A) compared with users (B).

Download figure to PowerPoint

Associations between HRT use and concentrations of the anticoagulants, with and without adjustment for confounders, are shown in Table II. Only those variables that were retained in the stepwise selection procedure remained in the final regression models. For protein C, with adjustment for low-density lipoprotein cholesterol (LDLc), the difference between oestrogen users and non-users was 0·20 µg/ml higher than in bivariate analysis. Adjustment had little impact on the antithrombin–hormone use relationships.

Table II.   Crude and adjusted differences in the anticoagulant proteins by hormone type compared with non-users.
Study markerCrude differenceAdjusted differenceVariables entering adjusted model
  • *

    P < 0·05,

  • **

    P < 0·01,

  • ***

    P < 0·001.

  • Hormone use status (yes/no) was forced into each model, with potential confounders allowed to enter in stepwise fashion (P-value for entry 0·10; for removal, 0·05). Variables considered were current age, race (white vs. other) diabetes, HDLc, LDLc, body mass index, smoking, osteoporosis, age at menopause, hysterectomy status and income level.

Protein C (µg/ml)
 Oestrogen0·38*0·58***LDL
 Oestrogen/progestin−0·11−0·04LDL
Antithrombin (%)
 Oestrogen−5***−6***LDL, hysterectomy
 Oestrogen/progestin−3**−3**Race, body mass index

Associations of HRT use with anticoagulant proteins did not differ on the basis of duration of use (data not shown). Further, because the incidence of venous thrombosis may be highest during the early months of HRT use, we assessed levels of protein C and antithrombin among women using it for < 1 year. Among these seven women (longest duration of use 7 months), the geometric mean antithrombin and protein C concentrations were 1·04 U/ml and 5·94 µg/ml respectively.

Among non-using women, 67/196 (34%) reported past oestrogen use for a median duration of 3 years (1 month−34 years). Protein C and antithrombin concentrations in these past users were 4·68 µg/ml and 1·13 U/ml respectively. Exclusion of past users from the control group had trivial effects on the differences in protein C and antithrombin concentrations between non-users and users (data not shown).

Subgroup analyses

Considering only unopposed oestrogen users and non-users, among black women, protein C concentration did not differ by use status: 4·66 µg/ml in users vs. 4·85 µg/ml in non-users (P = 0·35). The corresponding values in white women were 4·92 µg/ml and 4·40 µg/ml (P = 0·0005). This apparent race–drug use interaction was not statistically significant. There were no differences in associations of HRT use with protein C antigen by body mass index (data not shown).

Because antithrombin levels were identical comparing oestrogen and oestrogen/progestin users, the two user groups were combined for subgroup analyses by race and body mass index. Adjusting for LDLc, hysterectomy status and body mass index, there was an apparent race difference in associations, as shown in Fig 2. Within each race, antithrombin was significantly lower in users; however, there was a larger difference between users and non-users for blacks compared with whites (P for race by use interaction = 0·01). Specifically, the absolute difference comparing black users and non-users was −0·19 U/ml, whereas this value was −0·04 U/ml for whites. Black non-users had the highest antithrombin concentration, 1·25 U/ml. Compared with white participants, black participants were nearly twice as likely to have hypertension, and four times as likely to have diabetes. Among the 37 black participants with hypertension, respective antithrombin concentrations in users compared with non-users were 1·06 U/ml and 1·25 U/ml, whereas these values were 1·05 U/ml and 1·27 U/ml among blacks without hypertension. There were too few black diabetics (n = 6) for further analysis.

image

Figure 2.  Antithrombin concentration by hormone use and race. Oestrogen and combined therapy users were combined for analysis. Values were adjusted for LDLc, hysterectomy status and body mass index. Within each race group, the difference between users and non-users was statistically significant (P ≤ 0·001). The P-value for the interaction of hormone use and race was 0·01. The axis for antithrombin concentration was truncated to correspond to the normal range.

Download figure to PowerPoint

Because of the racial differences in antithrombin by hormone use status, blacks and whites were separated for the analysis of effect modification by body mass index. Among blacks, there was no difference in the association of antithrombin and hormone use by body mass index (data not shown). However, among white women, the difference in antithrombin concentration between hormone users and non-users was most apparent among women in the lowest tertile of body mass index (Table III). Non-users in the lowest body mass index tertile had the highest antithrombin concentration. Users in this tertile had an absolute value that was 7% lower than non-users. Differences between users and non-users in the top two tertiles of body mass index were less than half of this. These findings were not observed in the analysis of waist–hip ratio.

Table III.   Antithrombin concentration by hormone use and body mass index among Caucasian women.
 Mean antithrombin concentration ± SE (%) in tertiles of body mass index
Tertile 1 (21·6–22·2 kg/m2)Tertile 2 (22·3–26·0 kg/m2)Tertile 3 (26·1–32·2 kg/m2)
  1. Oestrogen and combined therapy users were combined for analysis. Values were adjusted for LDLc and hysterectomy status. The P-value for the interaction of hormone use and body mass index was 0·055.

Users109 ± 1·1110 ± 1·8111 ± 1·4
n = 102n = 77n = 65
Non-users116 ± 1·6113 ± 2·1113 ± 1·4
n = 45n = 57n = 65
P-value0·0010·270·26

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

In this cross-sectional study of long-term HRT users aged 65 years and older, unopposed oestrogen and combined therapy were associated with higher protein C and lower antithrombin concentrations. These findings are consistent with those of others in some smaller prospective studies and in one study of younger women (Nabulsi et al, 1993). The present findings extend previous findings to an older age group, which is at higher risk of venous thrombosis (Goldhaber, 1994). The long duration of hormone use in this study suggests that hormone-induced changes reported previously in short-term experimental studies may be stable over time into older age.

The association of HRT use with higher protein C suggests that HRT does not increase the risk of venous thrombosis directly through a protein C-dependent mechanism. However, we did not measure protein C activity or activated protein C resistance, and it is possible that oral HRT may influence these. Moreover, the possibility of a gene–environment interaction of factor V Leiden and post-menopausal hormone therapy in increasing the risk of venous thromboembolism must be considered, based on findings of such an interaction in younger women taking oral contraceptives (Vandenbroucke et al, 1994) and similar findings among HRT users in a recent case–control study (Lowe et al, 2000).

The association of hormone use with lower antithrombin is consistent with an increased risk of venous thrombosis from HRT. Even though the difference between users and non-users was small, changes in antithrombin, even within the physiologically normal range, may contribute importantly to the control of thrombin generation (Butenas et al, 1999). Also, among HRT non-users in this population, the central 95% of the distribution of antithrombin was 0·93–1·33 U/ml, suggesting regulation within a narrow range. Other simultaneous HRT-induced changes in coagulation, such as reduced protein S (Caine et al, 1992; The Writing Group for the Estradiol Clotting Factors Study, 1996) and increased factor VII (Kroon et al, 1994; Cushman et al, 1996; The Medical Research Council's General Practice Research Framework, 1996), may contribute to the risk of thrombosis, although some reports have suggested reduced factor VII with HRT (Scarabin et al, 1993; The Writing Group for the Estradiol Clotting Factors Study, 1996). Any relative influence of simultaneous coagulation profile changes from HRT, including increased protein C and reduced antithrombin, cannot be determined without a prospective study linking these changes to clinical outcomes.

Our findings suggest differences in associations of HRT with antithrombin by race. The higher prevalence of hypertension among black participants did not appear to explain this difference, nor did differences in body mass index comparing blacks with whites. Regardless, the observed race differences were based on a relatively small number of black users, so the finding requires confirmation.

Obesity is a risk factor for venous thrombosis (Jick et al, 1996; Goldhaber et al, 1997), but we are not aware of data describing an interaction between being overweight and HRT in enhancing this risk. Among white women, comparing HRT users with non-users, the difference in antithrombin concentration was largest in thinner women. This suggests a hypothesis that, through antithrombin lowering, HRT may enhance the risk of thrombosis to a greater degree in normal compared with overweight women. There is no direct association of obesity with antithrombin concentration (Sakkinen et al, 1998). However, obesity is associated with higher endogenous oestrogen levels (Cauley et al, 1989), so any added effect of exogenous hormones on coagulation in larger women might be expected to be blunted. We do not have endogenous oestrogen measures available in this study population for further exploration of this question.

There are differences between these results and colleagues. First, we did not observe significantly higher protein C antigen in combined therapy users, as was demonstrated in the ARIC study (Nabulsi et al, 1993). The low number of combined therapy users and the relatively high analytical CV for protein C may provide explanations for this. Secondly, two experimental studies assessing transdermal oestrogen with progestin reported no change in protein C (The Writing Group for the Estradiol Clotting Factors Study, 1996). This difference probably relates to the route of administration. Our results agree with another short-term experimental study of protein C and oral HRT in 29 women, although the increase in protein C was only observed at a higher oestrogen dose (1·25 mg of conjugated oestrogens) than was typically used here (Caine et al, 1992).

Mechanisms for HRT effects on anticoagulant proteins may include altered hepatic synthesis or clearance or, for antithrombin, chronic consumption related to coagulation activation from hormones (Caine et al, 1992). There are data supporting and refuting the latter hypothesis. One trial reported increased fragment 1–2 with hormones (Caine et al, 1992), and another reported increased d-dimer (Koh et al, 1997). These findings support ongoing consumption as a possible explanation for antithrombin lowering from HRT. However, in the same population studied here, there was no association of long-term hormone use with prothrombin fragment 1–2 or d-dimer concentrations, even though factor VIIc was higher among unopposed oestrogen users (Cushman et al, 1996, 1999a). Higher protein C with oestrogen use may represent a compensatory increase in response to coagulation activation, but may also relate to the apparent proinflammatory effects of oestrogen (Cushman et al, 1999a,b; van Baal et al, 1999), as protein C is an acute phase reactant.

As in other studies of the haemostasis effects of HRT, the main limitation of this study is the cross-sectional design. The findings may not be generalizable: healthier women may have been prescribed oestrogen (Matthews et al, 1996) and may have other health-conscious behaviours that could have influenced the results. Multivariate methods were used to adjust for confounding factors such as income that may relate to healthy user bias; however, misclassification of these factors and lack of control for unknown confounders must be considered. A healthy user bias related to withholding HRT among women felt to be at risk of venous thrombosis might be considered. However, this is not likely because HRT therapy in these women predated reports of HRT as a risk factor for venous thrombosis (Daly et al, 1996; Grodstein et al, 1996; Jick et al, 1996). Type II error may have influenced subgroup analyses; these findings require confirmation. Overall, the main findings concur with the existing literature, suggesting a small impact of the limitations mentioned, and increasing confidence in the other findings.

In conclusion, a modest influence of HRT on antithrombin, along with other coagulation proteins, may translate to clinical thrombosis. However, clinical application of these measures in women considering HRT would be premature. Future studies should address the effects of biochemical alterations from hormones on subsequent thrombosis risk.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The study was supported by NIH contracts NO1-HC-85079-85086 (L.H.K., B.M.P.), AHA New Hampshire−Vermont Affiliate 9606258S (M.C.) and K08-HL-03618 (M.C.). We are grateful to our colleagues at the participating institutions of the Cardiovascular Health Study: Forsyth County, NC – Bowman Gray School of Medicine of Wake Forest University: Gregory L. Burke, Sharon Jackson, Alan Elster, Walter H. Ettinger, Curt D. Furberg, Gerardo Heiss, Dalane Kitzman, Margie Lamb, David S. Lefkowitz, Mary F. Lyles, Cathy Nunn, Ward Riley, John Chen, Beverly Tucker; EKG Reading Center – Bowman Gray School of Medicine: Farida Rautaharju, Pentti Rautaharju; Sacramento County, CA – University of California, Davis: William Bommer, Charles Bernick, Andrew Duxbury, Mary Haan, Calvin Hirsch, Lawrence Laslett, Marshall Lee, John Robbins, Richard White; Washington County, MD – The Johns Hopkins University: M. Jan Busby-Whitehead, Joyce Chabot, George W. Comstock, Linda P. Fried, Joel G. Hill, Steven J. Kittner, Shiriki Kumanyika, David Levine, Joao A. Lima, Neil R. Powe, Thomas R. Price, Jeff Williamson, Moyses Szklo, Melvyn Tockman; MRI Reading Center – The Johns Hopkins University: R. Nick Bryan, Norman Beauchamp, Carolyn C. Meltzer, Naiyer Iman, Douglas Fellows, Melanie Hawkins, Patrice Holtz, Michael Kraut, Grace Lee, Larry Schertz, Cynthia Quinn, Earl P. Steinberg, Scott Wells, Linda Wilkins, Nancy C. Yue; Allegheny County, PA – University of Pittsburgh: Diane G. Ives, Charles A. Jungreis, Laurie Knepper, Peg Meyer, Roberta Moyer, Anne Newman, Richard Schulz, Vivienne E. Smith, Sidney K. Wolfson; Echocardiography Reading Center (Baseline) – University of California, Irvine: Hoda Anton-Culver, Julius M. Gardin, Margaret Knoll, Tom Kurosaki, Nathan Wong; Echocardiography Reading Center (Follow-Up) – Georgetown Medical Center: John Gottdiener, Eva Hausner, Stephen Kraus, Judy Gay, Sue Livengood, Mary Ann Yohe, Retha Webb; Ultrasound Reading Center – Geisinger Medical Center: Daniel H. O'Leary, Joseph F. Polak, Laurie Funk; Central Blood Analysis Laboratory – University of Vermont: Russell P. Tracy, Edwin Bovill, Elaine Cornell; Respiratory Sciences – University of Arizona–Tucson: Paul Enright; Coordinating Center – University of Washington–Seattle: Alice Arnold, Annette L. Fitzpatrick, Bonnie K. Lind, Richard A. Kronmal, David S. Siscovick, Lynn Shemanski, Will Longstreth, Patricia W. Wahl, David Yanez, Paula Diehr, Maryann McBurnie, Chuck Spiekerman, Scott Emerson, Cathy Tangen, Priscilla Velentgas; NHLBI Project Office: Diane E. Bild, Robin Boineau, Teri A. Manolio, Peter J. Savage, Patricia Smith.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  • Butenas, S., Van'T Veer, C., Mann, K.G. (1999) ‘Normal’ thrombin generation. Blood, 94, 21692178.
  • Caine, Y.G., Bauer, K.A., Barzegar, S., Ten Cate, H., Sacks, F.M., Walsh, B.W., Schiff, I., Rosenberg, R.D. (1992) Coagulation activation following estrogen administration to postmenopausal women. Thrombosis and Haemostasis, 68, 392395.
  • Cauley, J.A., Gutai, J.P., Kuller, L.H., LeDonne, D., Powell, J.G. (1989) The epidemiology of serum sex hormones in postmenopausal women. American Journal of Epidemiology, 129, 11201131.
  • Cushman, M., Cornell, E.S., Howard, P.R., Bovill, E.G., Tracy, R.P. (1995) Laboratory methods and quality assurance in the Cardiovascular Health Study. Clinical Chemistry, 42, 264270.
  • Cushman, M., Yanez, D., Psaty, B.M., Fried, L.P., Heiss, G., Lee, M., Polak, J.F., Savage, P.J., Tracy, R.P. (1996) Correlates of fibrinogen and coagulation factors VII and VIII in the elderly: results from the Cardiovascular Health Study. American Journal of Epidemiology, 143, 665676.
  • Cushman, M., Meilahn, E.N., Psaty, B.M., Kuller, L.H., Dobs, A.S., Tracy, R.P. (1999a) Hormone replacement therapy, inflammation, and hemostasis in elderly women. Arteriosclerosis Thrombosis and Vascular Biology, 19, 893899 .
  • Cushman, M., Legault, C., Barrett-Connor, E., Stefanick, M.L., Kessler, C., Judd, H.L., Sakkinen, P.A., Tracy, R.P. (1999b) Effect of postmenopausal hormones on inflammation-sensitive proteins: the Postmenopausal Estrogen/Progestin Interventions (PEPI) study. Circulation, 100, 717722.
  • Daly, E., Vessey, M.P., Hawkins, M.M., Carson, J.L., Gough, P., Marsh, S. (1996) Risk of venous thromboembolism in users of hormone replacement therapy. Lancet, 348, 977980.DOI: 10.1016/s0140-6736(96)07113-9
  • Fried, L.P., Borhani, N.O., Enright, P., Furberg, C.D., Gardin, J.M., Kronmal, R.A., Kuller, L.H., Manolio, T.A., Mittelmark, M.B., Newman, A., O'Leary, D.H., Psaty, B., Rautaharju, P., Tracy, R.P., Weiler, P.G. & the CHS Research Group. (1991) The Cardiovascular Health Study: design and rationale. Annals of Epidemiology, 1, 263276.
  • Goldhaber, S. (1994) Epidemiology of pulmonary embolism and deep vein thrombosis. In: Haemostasis and Thrombosis (ed. by A.Bloom, C.Forbes, D.Thomas & E. Tuddenham), pp. 13271333. Churchill Livingstone, New York.
  • Goldhaber, S.Z., Grodstein, F., Stampfer, M.J., Manson, J.E., Colditz, G.A., Speizer, F.E., Willett, W.C., Hennekens, C.H. (1997) A prospective study of risk factors for pulmonary embolism in women. JAMA, 277, 642645.
  • Grady, D., Wenger, N.K., Herrington, D., Khan, S., Furberg, C., Hunninghake, D., Vittinghoff, E., Hulley, S. for the Heart and Estrogen/Progestin Replacement Study Research Group. (2000) Postmenopausal hormone therapy increases risk for venous thromboembolic disease: The Heart and Estrogen/Progestin Replacement Study. Annals of Internal Medicine, 132, 689696.
  • Grodstein, F., Stampfer, M.J., Goldhaber, S.Z., Manson, J.E., Colditz, G.A., Speizer, F.E., Willett, W.C., Hennekens, C.H. (1996) Prospective study of exogenous hormones and risk of pulmonary embolism in women. Lancet, 348, 983987.DOI: 10.1016/s0140-6736(96)07308-4
  • Howard, P., Bovill, E., Mann, K., Tracy, R. (1988) A monoclonal antibody-based immunoassay for measurement of protein C in plasma. Clinical Chemistry, 34, 324330.
  • Jick, H., Derby, L.E., Wald Myers, M., Vasilakis, C., Newton, K.M. (1996) Risk of hospital admission for idiopathic venous thromboembolism among users of postmenopausal oestrogens. Lancet, 348, 981983.DOI: 10.1016/s0140-6736(96)07114-0
  • Koh, K.K., Mincemoyer, R., Bui, M.N., Csako, G., Pucino, F., Guetta, V., Waclawiw, M., Cannon, R.O. (1997) Effects of hormone-replacement therapy on fibrinolysis in postmenopausal women. New England Journal of Medicine, 336, 683690.
  • Koster, T., Rosendaal, F.R., Briet, E., Van der Meer, F.J.M., Colly, L.P., Trienekens, P.H., Poort, S.R., Reitsma, P.H., Vandenbroucke, J.P. (1995) Protein C deficiency in a controlled series of unselected outpatients: an infrequent but clear risk factor for venous thrombosis (The Leiden Thrombophilia Study). Blood, 85, 27562761.
  • Kroon, U., Silfverstolpe, G., Tengborn, L. (1994) The effects of transdermal estradiol and oral conjugated estrogens on haemostasis variables. Thrombosis and Haemostasis, 71, 420423.
  • Lowe, G., Woodward, M., Vessey, M., Rumley, A., Gough, P., Daly, E. (2000) Thrombotic variables and risk of idiopathic venous thromboembolism in women aged 45–64 years: relationships to hormone replacement therapy. Thrombosis and Haemostasis, 83, 530535.
  • Manolio, T.A., Furberg, C.D., Shemanski, L., Psaty, B.M., O'Leary, D.H., Tracy, R.P., Bush, T.L. for the CHS Collaborative Research Group. (1993) Associations of postmenopausal estrogen use with cardiovascular risk factors in older women. Circulation, 88, 21632171.
  • Matthews, K.A., Kuller, L.H., Wing, R.R., Meilahn, E.N., Plantinga, P. (1996) Prior use of estrogen replacement therapy, are users healthier than nonusers? American Journal of Epidemiology, 143, 971978.
  • Nabulsi, A.A., Folsom, A.R., White, A., Patsch, W., Heiss, G., Wu, K.K., Szklo, M. for the Atherosclerosis Risk in Communities Study Investigators. (1993) Association of hormone-replacement therapy with various cardiovascular risk factors in postmenopausal women. New England Journal of Medicine, 328, 10691075.
  • Psaty, B.M., Kuller, L.H., Bild, D., Burke, G.L., Kittner, S.J., Mittelmark, M., Price, T.R., Rautaharju, P.M., Robbins, J. (1995) Methods of assessing prevalent cardiovascular disease in the Cardiovascular Health Study. Annals of Epidemiology, 5, 270277.DOI: 10.1016/1047-2797(94)00092-8
  • Sakkinen, P.A., Cushman, M., Psaty, B.M., Kuller, L.H., Bajaj, S.P., Sabharwal, A.K., Boineau, R., Macy, E., Tracy, R.P. (1998) Correlates of antithrombin, protein C, protein S, and TFPI in a healthy elderly cohort. Thrombosis and Haemostasis, 80, 134139.
  • Scarabin, P.-Y., Plu-Bureau, G., Bara, L., Bonithon-Kopp, C., Guize, L., Samama, M.M. (1993) Haemostatic variables and menopausal status: influence of hormone replacement therapy. Thrombosis and Haemostasis, 70, 584587.
  • Scarabin, P.Y., Alhenc-Gelas, M., Plu-Bureau, G., Taisne, P., Agher, R., Aiach, M. (1997) Effects of oral and transdermal estrogen/progesterone regimens on blood coagulation and fibrinolysis in postmenopausal women: a randomized controlled trial. Arteriosclerosis Thrombosis and Vascular Biology, 17, 30713078.
  • The Medical Research Council's General Practice Research Framework. (1996) Randomised comparison of oestrogen versus oestrogen plus progestogen hormone replacement therapy in hysterectomised women. BMJ, 312, 473478.
  • The Writing Group for the Estradiol Clotting Factors Study. (1996) Effects on haemostasis of hormone replacement therapy with transdermal estradiol and oral sequential medroxyprogesterone acetate: a 1-year double-blind, placebo-controlled study. Thrombosis and Haemostasis, 75, 476480.
  • Van Baal, W.M., Kenemans, P., Van Der Mooren, M.J., Kessel, H., Emeis, J.J., Stehouwer, C.D.A. (1999) Increased C-reactive protein levels during short-term hormone replacement therapy in healthy postmenopausal women. Thrombosis and Haemostasis, 81, 925928.
  • Van Baal, W.M., Emeiss, J.J., Ven Der Mooren, M.J., Kessel, H., Kenemans, P., Stehouwer, C.D. (2000) Impaired procoagulant–anticoagulant balance during hormone replacement therapy? A randomised, placebo-controlled 12-week study. Thrombosis and Haemostasis, 83, 2934.
  • Vandenbroucke, J.P., Koster, T., Briet, E., Reitsma, P.H., Bertina, R.M., Rosendaal, F.R. (1994) Increased risk of venous thrombosis in oral-contraceptive users who are carriers of factor V Leiden mutation. Lancet, 344, 14531457.