A 2-year, randomized, comparative, placebo-controlled study on the effects of raloxifene on lipoprotein(a) and homocysteine

Maturitas 41 (2002) 105– 114 www.elsevier.com/locate/maturitas

A 2-year, randomized, comparative, placebo-controlled study on the effects of raloxifene on lipoprotein(a) and homocysteine Raimond G.V. Smolders a, Tatjana E. Vogelvang a, Velja Mijatovic a, W. Marchien van Baal a, Simone J.M. Neele b, J. Coen Netelenbos b, Peter Kenemans a, Marius J. van der Mooren a,* a

Department of Obstetrics and Gynecology, Institute for Cardio6ascular Research, VU Uni6ersity Medical Centre, P.O. Box 7057, 1007 MB Amsterdam, The Netherlands b Department of Endocrinology, Institute for Cardio6ascular Research, VU Uni6ersity Medical Centre, Amsterdam, The Netherlands Received 17 May 2001; received in revised form 13 August 2001; accepted 14 September 2001

Abstract Objecti6es: Lipoprotein(a) (Lp(a)) and homocysteine (Hcy) are independent cardiovascular risk factors, which have been shown to be lowered by hormone replacement therapy (HRT). In this 2-year study, the long-term effects of raloxifene (Rlx) in two doses, on Lp(a) and Hcy, were studied and compared with the effects of continuously combined hormone replacement therapy (ccHRT). Methods: In a prospective, randomized, double-blind, placebo-controlled 2-year study, 95 healthy, non-hysterectomized, early postmenopausal women, received daily either oral Rlx 60 mg (N= 24) or 150 mg (N=23), ccHRT (conjugated equine estrogens 0.625 mg plus medroxyprogesterone acetate 2.5 mg; N =24) or placebo (N=24). Fasting serum Lp(a) and plasma Hcy concentrations were measured at baseline and at 6, 12 and 24 months. Results: The mean individual changes compared to baseline after 24 months were for Lp(a): Rlx 60: −5%, Rlx 150: −7%, ccHRT: −34%, placebo: +1% and for Hcy: Rlx 60: − 3%, Rlx 150: − 4%, ccHRT: −4%, placebo: + 6%. ANCOVA was significant for Lp(a) under ccHRT versus placebo (P = 0.001) and for Lp(a) under ccHRT versus each of the two Rlx groups (PB0.05). Conclusions: Long-term treatment with Rlx was not as effective as ccHRT in lowering Lp(a). Although not significant and without an obvious dose-related response, the Hcy values showed the same trend for each treatment arm, which is in line with data reported earlier. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Raloxifene; Hormone replacement therapy; Lipoprotein(a); Homocysteine; Prospective; Placebo; Long-term

* Corresponding author. Tel.: + 31-2044-43244; fax: +312044-44422. E-mail address: [email protected] (M.J. van der Mooren).

1. Introduction Cardiovascular disease (CVD) is not only an important health risk for men, it is the leading

0378-5122/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 5 1 2 2 ( 0 1 ) 0 0 2 8 0 - 8

106

R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

cause of death for women over 60 as well. Conversely, for women before menopause, the risk for CVD is negligible when compared to their age-matched counterparts [1]. This observation preceded the reports that showed that post menopausal estrogen-use decreased cardiovascular risk [2]. A growing number of mechanisms have been reported to be involved [3]. Lipoprotein(a) (Lp(a)) is a complex of apolipoprotein(a) and low-density lipoprotein (LDL), linked through a disulfide bond. It is present in the arterial wall, where it contributes to the formation of atherosclerotic plaques [4]. After both surgical and natural menopause Lp(a) serum levels rise [5,6]. Furthermore, postmenopausal women have higher Lp(a) levels than agematched men [6]. This suggests the involvement of female sex hormones. In 1969, high serum homocysteine (Hcy) was identified as being a causal agent in the development of thromboembolic disease [7]. More recent epidemiological data has shown a relationship, not only between high plasma Hcy concentrations and CVD, but also between mildly elevated plasma Hcy concentrations and CVD [8]. Homocysteine is lower in premenopausal women than in men and postmenopausal women, suggesting a relation with female sex hormones [9,10]. Epidemiologically, both Lp(a) and Hcy have proven to be independent risk factors for CVD [11]. The Heart and Estrogen/Progestin Replacement Study (HERS), which is the first randomized placebo-controlled secondary prevention trial with hormone replacement therapy (HRT) and hard clinical endpoints, showed no overall effect after on average 4.1 years of HRT [12]. This is in contrast to the majority of other studies on this subject and may emphasize that not all mechanisms related to CVD are modified advantageously in all circumstances. It seems necessary, not only to identify, but also to quantify the impact of estrogen and estrogen/progestogen use on the different mechanisms that together determine the individual risk for CVD. A recent publication from the HERS emphasizes the importance of Lp(a) as a predictor of CVD [13].

An alleged risk of long-term estrogen use is the increased incidence of breast cancer [14]. This has led to the development of a new class of compounds, referred to as selective estrogen receptor modulators (SERMs). Due to the organ- or tissue-specific estrogenic or anti-estrogenic effects, these substances might become an important alternative for long-term estrogen use. Almost two decades ago, raloxifene (Rlx) (a second generation SERM) was synthesized. It showed estrogenic effects, such as a favorable change in bone turnover balance and in the lipid profile [15]. It also showed no increased induction of endometrial hyperplasia or polyps nor is it associated with an increased risk of endometrial cancer [15]. A recent report suggested a decreased risk of breast cancer after 3 years of raloxifene use [16]. Raloxifene’s most important side effect is the increased number of venous thrombo-embolic complications in the first half-year of treatment [16]. Moreover, in contrast to HRT, it increases the incidence of hot flushes, which makes raloxifene unsuitable for the treatment of menopausal vasomotor symptoms. Nevertheless, it might prove to be a more acceptable long-term treatment option for preventing the diseases associated with postmenopausal estrogen deficiency, such as osteoporosis and cardiovascular disease. Both Rlx and HRT have been shown to lower Lp(a) and Hcy [5,17–23]. However, at the onset of this study, there were no reports on studies that had tested for the differences, if any, between Rlx and conjugated equine estrogens continuously combined with medroxyprogesterone acetate (ccHRT), which is the most widely used form of ccHRT. These differences remain to be specified and confirmed against placebo in longterm trials. In this paper, we present the data of a randomized double-blind placebo-controlled trial of a 24-month treatment with raloxifene 60 or 150 mg daily versus 0.625 mg conjugated equine estrogens continuously combined with 2.5 mg medroxyprogesterone acetate and versus placebo. We examined the effect of these treatments on serum concentrations of Lp(a) and plasma concentrations of Hcy in healthy early postmenopausal women.

R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

2. Material and methods

2.1. Subjects Ninety-five healthy non-hysterectomized early postmenopausal women were recruited through advertisements in the local newspapers and subsequently enrolled in this 2-year study, which was performed at the outpatient clinics of the Department of Obstetrics and Gynecology and the Department of Endocrinology. The study conforms to the principles outlined in the declaration of Helsinki and was approved by the Institutional Review Board of the VU University Medical Centre. Written informed consent was obtained from each participant before study entry. Participants were between 47 and 60 years old, had a body mass index (BMI) between 18 and 31 kg/m2, had had their last menstrual period 6– 24 months before randomization and had no indication requiring estrogen replacement therapy (ERT). Follicle-stimulating hormone (FSH) concentration was \ 30 IU/l in each participant. Exclusion criteria included a history of metabolic or endocrinological disease, estrogendependent neoplasia, proven gluten sensitivity, an untreated malabsorption syndrome, alcohol or drug abuse, as well as clinically relevant abnormalities in laboratory tests of renal and hepatic function. Women who had previously participated in any study investigating raloxifene were also excluded. Eligible women were assigned in a double-blind fashion to raloxifene HCl 60 mg (Rlx 60) or raloxifene HCl 150 mg (Rlx 150); (Eli Lilly and Co., Indianapolis, IN) or to continuously combined hormone replacement therapy (ccHRT; conjugated equine estrogens 0.625 mg/day, Wyeth-Ayerst, Philadelphia, PA, combined with medroxyprogesterone acetate, 2.5 mg/day, Upjohn Co., Kalamazoo, MI), or to placebo. Since raloxifene HCl and ccHRT were different in shape and size (tablets and capsules, respectively), there were two separate placebos identical in appearance to the raloxifene HCl and ccHRT to maintain the blinding (double-dummy). Study medication and placebo were packaged according to a random-number table (block-size: 8) and

107

assigned to the participants in sequence. Subjects were instructed to take one tablet and two capsules of the blinded study medication each morning. Of the 95 participants initially enrolled, 12 women dropped out before the 6-month visit and were therefore excluded from the analysis: one woman in the Rlx 60 mg group, three in the Rlx 150 mg group, six in the ccHRT group and two women in the placebo group.

2.2. Blood collection Samples were taken at baseline and after 6, 12 and 24 months. Blood collection was scheduled between 08:00 and 10:00 h after at least 10 h of fasting and non-smoking. We used both plain tubes and tubes containing ethylenediaminetetraacetic acid (EDTA) from the Vacutainer® system. The EDTA tubes (for the Hcy measurement) were immediately placed on ice and centrifuged within 1 h. The plain tubes (for the Lp(a) measurement) were allowed to clot for 1 h. Centrifuging was carried out for 30 min at 3000×g and 4 °C (EDTA) or 20 °C (plain tubes). The plasma was divided into polypropylene vials, snap-frozen and stored at − 70 °C until analysis.

2.3. Assays Serum Lp(a) concentrations were determined using a commercially available enzyme-linked immunoassay (Innotest, Innogenetics, NV Zwijndrecht, Belgium). The intra-coefficient of variation was 4.5%. Cross-reactivity with plasminogen and LDL was tested for this assay and this could not be found up to concentrations of 500 mg/dl. Plasma Hcy is defined as plasma total homocysteine measured as the sum of all Hcy subfractions in plasma, including free and protein-bound forms. This was measured using high performance liquid chromatography with fluorescence detection, according to Fiskerstrand et al. [24]. The lower limit of detection was 0.5 mmol/l and the intra-assay coefficient of variation for homocysteine within the normal range was 3.5%. All samples of a given subject were analyzed in a single run.

108

R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

FSH concentrations were measured in serum using a microparticle enzyme immune assay (IMX, Abbott Laboratories, Chicago, IL). Estradiol levels were determined using the estradiol double anticorps (I125 RIA) from DPC (Diagnostic Products Corp.).

2.4. Statistics Statistical analysis was performed using the Statistical Package for the Social Sciences PC + 4.0 (SPSS Inc., Chicago, IL). The data set comprised all the women who had at least one follow-up visit after randomization. For the women who withdrew from the study before the 24-month visit, the last-observation-carried-forward procedure was applied for the missing values. Values are given as mean9S.D. or as median and range. We compared baseline measurements between groups using standard parametric and non-parametric tests where applicable. Analyses of covariance (ANCOVA), for repeated measurements, with baseline as a constant co-variate, were used for comparisons between the four groups. Correlations between variables were calculated with Pearson’s or Spearman’s correlation coefficient. A two-tailed P B0.05 was accepted as the level of significance, except for the correlation analyses, where a two-tailed P B 0.01 was considered significant.

3. Results Table 1 shows the descriptive characteristics at baseline. After randomization, smokers were not equally divided among groups within our definition of chance (one way ANOVA, P = 0.02). No significant differences in baseline values for Lp(a) or Hcy were detected (one way ANOVA, P = 0.28 and P= 0.54). At baseline, there was an almost significant difference in body mass index (BMI) between the groups (Table 1); therefore, BMI was also used as a constant covariate. As a result of a statistically significant difference between the four groups in smoking habits and since smoking, as a dichotomous variable could not be introduced as a covariate, we re-analyzed the ANCOVA for the non-smoking women. The results were similar (data not shown). Baseline correlation-coefficients were calculated for Lp(a) and Hcy, as well as for Lp(a) or Hcy and FSH, time since menopause, estradiol, cholesterol and blood pressure. No significance was found.

3.1. Lipoprotein(a) ANCOVA with baseline values and BMI as constant covariates, over 12 and 24 months showed a significant difference among the groups (P B0.01). Further analyses showed that this difference could be explained by the changes occurring in the Rlx groups, separately and combined,

Table 1 Descriptive characteristics of the four groups at baseline

Age (years) Duration of amenorrhoea (months) Body mass index (kg/m2) Blood pressure: systolic (mmHg) Blood pressure: diastolic (mmHg) Smokers N (%) Serum cholesterol (mmol/l) Serum FSH (U/l) Serum E2 (pmol/l)

Rlx 60 (N= 24)

Rlx 150 (N =23)

ccHRT (N =24)

Placebo (N =24)

P-value*

51.09 3.2 13.09 5.5 23.892.9 118915 78 9 11 5 (21) 6.1 9 1.1 86.5930.2 9.5 (5–285)

51.3 9 2.2 13.1 95.3 26.09 3.6 124 911 79 97 10 (44) 6.2 90.8 87.7 9 30.7 16.0 (5–338)

50.8 9 2.4 12.3 9 5.0 25.0 92.4 126 9 19 82 9 9 2 (9) 6.0 9 1.0 91.2 9 31.3 18.0 (5–648)

51.8 9 3.1 13.1 9 4.8 26.1 94.3 127 915 82 9 8 10 (42) 5.8 9 0.9 92.4 9 37.4 7.0 (5–55)

0.68 0.94 0.07 0.18 0.30 0.02 0.73 0.91 0.47

* One-way ANOVA or Kruskal–Wallis or Fisher’s exact test for between-group differences. Values are given as number (N), with percentage in parentheses, mean 9 SD or as median (range). Rlx 60, raloxifene 60 mg; Rlx 150, raloxifene 150 mg; ccHRT, conjugated equine estrogens 0.625 mg and medroxyprogesterone acetate 2.5 mg; FSH, follicle-stimulating hormone; E2, 17b-estradiol.

R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

Fig. 1. Mean percentage change from baseline in fasting serum Lp(a) concentrations after 24 months; , placebo; Rlx 150; , ccHRT.

compared to the HRT group (P B0.05), as well as to the changes in the ccHRT group when compared to the placebo group (P =0.001). Compared to placebo after 24 months, Lp(a) serum concentrations had decreased in all groups; by 6% in both the Rlx 60 group, by 8% in the Rlx 150 group and by 34% in the ccHRT group (Fig. 1). Baseline values and mean individual changes from baseline are given in Table 2. To assess the influence of the last-observationcarried-forward procedure, we repeated the analysis on only the values of the patients who completed the study. This showed comparable results: (ANCOVA overall P =0.03 on 12 months and P =0.02 on 24 months).

109

, Rlx 60;

,

groups (P= 0.12 and P = 0.11, respectively). In between-group ANCOVAs, Rlx 60 compared to placebo showed a significant difference (P = 0.03; Rlx 150 or HRT versus placebo ANCOVA, P = 0.07). Compared to placebo after 24 months, Hcy had decreased 9% both in the Rlx 60 group and the Rlx 150 group and 10% in the ccHRT group (Fig. 2). Baseline values and mean percentage changes from baseline are given in Table 3. The analysis on only the values of the patients who completed the study had a comparable outcome for the overall analysis but no significant differences for the between group ANCOVAs.

4. Discussion

3.2. Homocysteine ANCOVA with baseline values and BMI as constant covariates, over 12 and 24 months showed no significant difference among the

This report shows that 24 months of treatment with Rlx lowers Lp(a) less effectively than treatment with continuously combined conjugated equine estrogens plus medroxyprogesterone ace-

102 (6–1440) 159 (3–1575) 90 (12–957) 142 (22–1360)

24 23 23 24

N

−8.19 34.6 −3.49 33.6 −18.0923.6 10.9 9 24.4

% Change from baseline at 6 months 23 20 17 23

N

−7.79 25.9 17.9 9 87.4 −26.59 27.7 −6.596.5

% Change from baseline at 12 months

22 16 14 20

N

−5.1927.1 −7.29 21.5 −33.6929.8 0.79 17.0

% Change from baseline at 24 months

22 16 13 19

N

Values are given as median and range in mg/l, as number (N) or as mean individual percentage change from baseline with S.D. Rlx 60, raloxifene 60 mg; Rlx 150, raloxifene 150 mg; ccHRT, conjugated equine estrogens 0.625 mg and medroxyprogesterone acetate 2.5 mg.

Rlx 60 Rlx 150 ccHRT Placebo

Baseline

Table 2 Serum concentrations and percentage change of Lp(a)

110 R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

tate. Hcy was found to decrease under all active treatments to the same extent. However, we only observed a significant effect on Hcy in the Rlx 60 group when compared to the placebo group. In our analysis, we used the last-observationcarried-forward procedure to resemble, as closely as possible, the clinical intention-to-treat situation. Furthermore, it enabled us to include more baseline values and all of the 6-month values, therefore providing a stronger correction for the differences at baseline. This is an accepted method to prevent loss of data in an ANCOVA analysis. However, if an initial effect would decrease in spite of continuation of the treatment, the effect would be statistically overestimated instead of underestimated. To evaluate if this was the case in the values of the completers, we performed a completers’ analysis. Since the outcome was highly similar, we felt our approach was justified. Lp(a) has no known physiological function. This unique lipoprotein consists of two main parts: the first is a protein particle, which resembles low density lipoprotein (LDL) and is covalently bound to the second, the apolipoprotein(a) (apo(a)) particle. The LDL resembling particle is likely to give Lp(a) its atherogenic properties, while the apo(a) particle may inhibit fibrinolysis, due to its structural similarity to plasminogen [25]. There is no information available on the mechanisms by which Rlx interacts with Lp(a). On the other hand, for HRT it has been hypothesized that the estrogen-induced upregulation of the hepatic LDL receptor results in an increase in Lp(a) degradation [26].

111

Hcy is the demethylated intermediate in the metabolism of the methyl donor methionine, which is an essential amino acid. Hcy can act as a methyl acceptor and be remethylated to methionine or transulferated and excreted in the urine. Besides being prothrombotic, Hcy is toxic to the endothelium [27,28]. Furthermore, it is thought to increase vascular smooth muscle cell collagen production and to decrease the availability of nitric oxide [29,30]. The way in which Rlx or HRT modulates these mechanisms remains to be clarified. For HRT, possible mechanisms include a change in the anabolism/catabolism ratio, as well as an increase in kidney methionine synthase activity or changes in the methionine transamination [31]. The first prospective studies to show the reduction of Lp(a) by HRT appeared in the 1990s [5,17,19,32–37]. In the Postmenopausal Estrogen/ Progestin Intervention (PEPI) study, Espeland et al. showed a Lp(a) reduction of : 20% in women treated for 3 years with ccHRT [38]. Although the changes vary, in general the findings of the articles cited are comparable with the results presented here. Mijatovic et al., who showed the Lp(a) lowering effect of conjugated equine estrogens, also found a significant reduction with raloxifene [21]. The results included a significant 30% reduction during raloxifene 150 mg treatment, as well as a 35% decrease of serum Lp(a) during ERT. Although the variations in calculation methods and the wide (normal) range of Lp(a) make quantitative comparisons in these small groups haphazard, differences in age, menopausal status

Fig. 2. Mean percentage change from baseline in fasting plasma homocysteine concentrations after 24 months; , placebo; 60; , Rlx 150; , ccHRT.

, Rlx

R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

112

Table 3 Plasma concentrations and percentage change of total homocysteine

Rlx 60 Rlx 150 ccHRT Placebo

Baseline

N

% Change from baseline N at 6 months

% Change from baseline at 12 months

N

% change from baseline at 24 months

N

8.8 92.5 9.0 92.2 10.3 96.8 8.9 9 3.3

24 23 24 24

−6.79 17.0 −2.6914.3 −3.3914.8 4.8916.0

−6.99 13.2 −5.29 9.3 −2.3916.7 1.3 918.8

22 17 14 20

−3.49 17.3 −3.79 17.5 −4.1920.6 5.5 9 16.4

22 16 13 19

23 20 17 23

Values are given as mean concentration in mmol/l9 S.D., as number (N) or as mean percentage change from baseline (percentage change). Rlx 60, raloxifene 60 mg; Rlx 150, raloxifene 150 mg; ccHRT, conjugated equine estrogens 0.625 mg and medroxyprogesterone acetate 2.5 mg.

and consequently endogenous estrogen levels at baseline might explain the differences in results with this particular study. Several studies have been carried out on Hcy under HRT or ERT [22,39– 42]. These articles describe a 10–20% reduction in Hcy during HRT or ERT, where our results show a non-significant 9.6% decrease during HRT compared to placebo. Moreover, we found an 8.9 and 9.2% change compared to placebo in the Rlx 60 and 150 mg groups, respectively. The only available article reporting a 24-month Rlx result showed a significant reduction of over 15% when Rlx 150 was used [20]. In a more recent 6-month study, Walsh et al. showed a significant 6– 8% reduction in Hcy during HRT as well as during Rlx 60 and 120 mg treatment [23]. These changes are comparable with our findings. An apparent difference between most of the literature and our present data is the difference in baseline values. For both Lp(a) and Hcy the subjects in our study had low baseline values. Since several articles showed a relatively more pronounced effect on high levels of both Lp(a) and Hcy, this may explain the discrepancy found in statistical significance [31]. With regard to the study by Walsh et al., the larger number of subjects (N= 390) supplied this study with more power to reach statistical significance [23]. Although not the aim of this study, the question remains how to use these markers in clinical practice [8,11]. This issue becomes even more pressing when put in perspective with both the main results of the HERS trial, as well as with its recent data on Lp(a) [12,13]. To be able to employ

these markers when making the decision whether or not to start or continue HRT or Rlx, more information is needed. Firstly, the effect on morbidity and mortality of modulating Lp(a) or Hcy with HRT or Rlx has to be confirmed and determined quantitatively in a prospective randomized controlled setting with hard clinical endpoints. Secondly, a clinically relevant correlation between increments in serum values and relative risk has to be demonstrated. Thirdly, to facilitate the first two points, we have to gain more insight into the biochemical role of Lp(a) and Hcy in CVD and the way in which this is influenced by sex hormones or SERMs, for this might provide the clues needed to discriminate the women who will benefit from HRT or Rlx use from those who will not benefit from their use. In conclusion, during 24 months of treatment, raloxifene at doses of 60 and 150 mg/day did not decrease Lp(a) as effectively as a current regime ccHRT. In line with previous observations, the decrease in Hcy, albeit not reaching significance in the present study, was roughly the same in all treatment modes. The prospects for using Rlx as an alternative for HRT in order to avert the deleterious effects of estrogen deficiency on the cardiovascular system, will depend on how important any modulation of Lp(a) and Hcy will prove to be for the prevention of CVD.

Acknowledgements The authors thank Ieteke Peters-Muller, M.Sc. (Department of Obstetrics and Gynecology) for

R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

statistical calculations. This research project was supported by Eli Lilly and Company® which markets raloxifene, through a grant from the Biocare Foundation (Grant No. 95-32.3).

References [1] Wenger NK, Speroff L, Packard B. Cardiovascular health and disease in women. New Engl J Med 1993;329:247 –56. [2] Stampfer MJ, Colditz GA, Willett WC, et al. Postmenopausal estrogen therapy and cardiovascular disease. New Engl J Med 1991;325:756 –62. [3] Van Baal WM, Kooistra T, Stehouwer CDA. Cardiovascular disease risk and hormone replacement therapy: a review based on randomised, controlled studies in postmenopausal women. Curr Med Chem 2000;7:499 –517. [4] Smith EB, Keen GA, Grant A. Factors influencing the accumulation in fibrous plaques of lipid derived from low density lipoprotein. I. Relation between fibrin and immobilization of apo B-containing lipoprotein. Atherosclerosis 1990;84:165 – 71. [5] Bruschi F, Meschia M, Soma M, et al. Lipoprotein(a) and other lipids after oophorectomy and estrogen replacement therapy. Obstet Gynecol 1996;88:950 –4. [6] Heinrich J, Sandkamp M, Kokott R, Schulte H, Assmann G. Relationship of lipoprotein(a) to variables of coagulation and fibrinolysis in a healthy population. Clin Chem 1991;37:1950 – 4. [7] McCully KS. Vascular pathology of homocysteinemia: implications for the pathogenesis of arteriosclerosis. Am J Pathol 1969;56:111 –28. [8] Boushey CJ, Beresford SA, Omenn GS, Motulsky AG. A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes. J Am Med Assoc 1995;274:1049 – 57. [9] Andersson A, Brattstrom L, Israelsson B, et al. Plasma homocysteine before and after methionine loading with regard to age, gender, and menopausal status. Eur J Clin Invest 1992;22:79 – 87. [10] Hak AE, Polderman KH, Westendorp IC, et al. Increased plasma homocysteine after menopause. Atherosclerosis 2000;149:163 – 8. [11] Ridker PM, Glynn RJ, Hennekens CH. C-reactive protein adds to the predictive value of total and HDL cholesterol in determining risk of first myocardial infarction. Circulation 1998;97:2007 –11. [12] Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. J Am Med Assoc 1998;280:605 –13. [13] Shlipak MG, Simon JA, Vittinghoff E, et al. Estrogen and progestin, lipoprotein(a), and the risk of recurrent coronary heart disease events after menopause. J Am Med Assoc 2000;283:1845 –52.

113

[14] Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52705 women with breast cancer and 108411 women without breast cancer. Lancet 1997;350:1047 – 59. [15] Delmas PD, Bjarnason NH, Mitlak BH, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. New Engl J Med 1997;337:1641 – 7. [16] Cummings SR, Eckert S, Krueger KA, et al. The effect of raloxifene on risk of breast cancer in postmenopausal women— results from the MORE randomized trial. J Am Med Assoc 1999;281:2189 – 97. [17] Soma MR, Osnago-Gadda I, Paoletti R, et al. The lowering of lipoprotein (a) induced by estrogen plus progesterone replacement therapy in postmenopausal women. Arch Intern Med 1993;153:1462 – 8. [18] The Writing Group for the PEPI Trial. Effects of estrogen or estrogen/progestin regimes on heart disease risk factors in postmenopausal women: the Postmenopausal Estrogen/Progestin Intervention (PEPI) trial. J Am Med Assoc 1995;273:199 – 208. [19] Mijatovic V, Kenemans P, Netelenbos JC, et al. Oral 17 beta-estradiol continuously combined with dydrogesterone lowers serum lipoprotein(a) concentrations in healthy postmenopausal women. J Clin Endocrinol Metab 1997;82:3543 – 7. [20] Mijatovic V, Netelenbos C, Van der Mooren MJ, et al. Randomized, double-blind, placebo-controlled study of the effects of raloxifene and conjugated equine estrogens on plasma homocysteine levels in healthy postmenopausal women. Fertil Steril 1998;70:1085 – 9. [21] Mijatovic V, Van der Mooren MJ, Kenemans P, De Valk de Roo G, Netelenbos C. Raloxifene lowers serum lipoprotein(a) in healthy postmenopausal women: a randomized, double-blind, placebo-controlled comparison with conjugated equine estrogens. Menopause 1999;6:134 – 7. [22] Van Baal WM, Smolders RGV, Van der Mooren MJ, Teerlink T, Kenemans P. Hormone replacement therapy and plasma homocysteine levels. Obstet Gynecol 1999;94:485 – 91. [23] Walsh BW, Paul S, Wild RA, et al. The effects of hormone replacement therapy and raloxifene on C-reactive protein and homocysteine in healthy postmenopausal women: a randomized, controlled trial. J Clin Endocrinol Metab 2000;85:214 – 8. [24] Fiskerstrand T, Refsum H, Kvalheim G, Ueland PM. Homocysteine and other thiols in plasma and urine: automated determination and sample stability. Clin Chem 1993;39:263 – 71. [25] Marcovina SM, Koschinsky ML. Lipoprotein(a) as a risk factor for coronary artery disease. Am J Cardiol 1998;82:57U – 66U. [26] Snyder ML, Hay RV, Whitington PF, Scanu AM, Fless GM. Binding and degradation of lipoprotein(a) and LDL by primary cultures of human hepatocytes. Comparison with cultured human monocyte-macrophages and fibroblasts. Arterioscler Thromb 1994;14:770 – 9.

114

R.G.V. Smolders et al. / Maturitas 41 (2002) 105–114

[27] Harker LA, Ross R, Slichter SJ, Scott CR. Homocystineinduced arteriosclerosis. The role of endothelial cell injury and platelet response in its genesis. J Clin Invest 1976;58:731 – 41. [28] Hajjar KA. Homocysteine-induced modulation of tissue plasminogen activator binding to its endothelial cell membrane receptor. J Clin Invest 1993;91:2873 – 9. [29] Majors A, Ehrhart LA, Pezacka EH. Homocysteine as a risk factor for vascular disease. Enhanced collagen production and accumulation by smooth muscle cells. Arterioscler Thromb Vasc Biol 1997;17:2074 –81. [30] Upchurch GR Jr, Welch GN, Fabian AJ, et al. Homocyst(e)ine decreases bioavailable nitric oxide by a mechanism involving glutathione peroxidase. J Biol Chem 1997;272:17012 –7. [31] Van der Mooren MJ, Mijatovic V, Van Baal WM, Stehouwer CDA. Hormone replacement therapy in postmenopausal women with specific risk factors for coronary artery disease. Maturitas 1998;30:27 – 36. [32] Van der Mooren MJ, Demacker PN, Thomas CM, Borm GF, Rolland R. A 2-year study on the beneficial effects of 17 beta-oestradiol-dydrogesterone therapy on serum lipoproteins and Lp(a) in postmenopausal women: no additional unfavourable effects of dydrogesterone. Eur J Obstet Gyn Reprod Biol 1993;52:117 –23. [33] Kim CJ, Jang HC, Cho DH, Min YK. Effects of hormone replacement therapy on lipoprotein(a) and lipids in postmenopausal women. Arterioscler Thromb 1994;14:275 – 81. [34] Sacks FM, McPherson R, Walsh BW. Effect of postmenopausal estrogen replacement on plasma Lp(a) lipoprotein concentrations. Arch Intern Med 1994;154: 1106 –10.

[35] Mendoza S, Velazquez E, Osona A, Hamer T, Glueck CJ. Postmenopausal cyclic estrogen – progestin therapy lowers lipoprotein [a]. J Lab Clin Med 1994;123:837 – 41. [36] Taskinen MR, Puolakka J, Pyorala T, et al. Hormone replacement therapy lowers plasma Lp(a) concentrations. Comparison of cyclic transdermal and continuous estrogen – progestin regimens. Arterioscler Thromb Vasc Biol 1996;16:1215 – 21. [37] Haines C, Chung T, Chang A, et al. Effect of oral estradiol on Lp(a) and other lipoproteins in postmenopausal women. A randomized, double-blind, placebo-controlled, crossover study. Arch Intern Med 1996;156:866 – 72. [38] Espeland MA, Marcovina SM, Miller V, et al. Effect of postmenopausal hormone therapy on lipoprotein(a) concentration. PEPI Investigators. Postmenopausal estrogen/ progestin interventions. Circulation 1998;97:979 – 86. [39] Van der Mooren MJ, Wouters MG, Blom HJ, et al. Hormone replacement therapy may reduce high serum homocysteine in postmenopausal women. Eur J Clin Invest 1994;24:733 – 6. [40] Van der Mooren MJ, Demacker PN, Blom HJ, De Rijke YB, Rolland R. The effect of sequential three-monthly hormone replacement therapy on several cardiovascular risk estimators in postmenopausal women. Fertil Steril 1997;67:67 – 73. [41] Mijatovic V, Kenemans P, Jakobs C, et al. A randomized controlled study of the effects of 17beta-estradiol-dydrogesterone on plasma homocysteine in postmenopausal women. Obstet Gynecol 1998;91:432 – 6. [42] Mijatovic V, Kenemans P, Netelenbos C, et al. Postmenopausal oral 17beta-estradiol continuously combined with dydrogesterone reduces fasting serum homocysteine levels. Fertil Steril 1998;69:876 – 82.