K. Farish1, C. D. Fletcher¹, D. M. Hart², F. Al. Azzawi², H. I, Abdalla² and C. E. Gray³

Department of BiochemistryStohhil^l General Hospital, Glasgow G21, Department of Gyneacology², Stohhill General Hospital, Glasgow G21 and Department of Clinical Biochemistry³, Glasgow Royal Infirmary, Glasgow 04, Scotland.

Abstract

Scrum lipoproteins were measured over a period of (i months in 14 oophorectomised women treated with oestrogen implants (50 mg oestradiol-17β and 17 oophorectomised women treated with oestrogen/ testosterone implants (50 mg oestradiol-17β, 100 mg testosterone). Boll) types of implant caused only minimal changes in lipoprotein metabolism. Low density lipoprotein (LDL) cholesterol decreased with both types of implant and high density lipoprotein (HDL) cholesterol rose with the oestrogen implants. HDL subfractions were also measured. The oestrogen implants caused a transient rise in HDL₂ cholesterol levels at 2 months and a slower rise in HDL₃ cholesterol. The oestrogen/testosterone implants had no effect on HDL fractions. The results indicate that hormone implants do not cause the profound changes in lipoproteins associated with oral hormone therapy.

Considerable interest has !>een shown in the lipid altering properties of hormone replacement therapy in post-menopausal women because of the link between lipoproteins and cardiovascular disease (Gordon et al. 1977; Miller et al. 1977). Oestrogens, when administered orally have been shown to cause significant alterations in lipoproteins (Tikkanen et al. 1978; Silfverstolpe et al. 1980). However. Fahraeus et al. (1982) found that the changes in lipoproteins caused by oestradiol- 17β taken orally were not in evidence when it was administered as a cream and Buckmaii et al. (1980) found that parenteral depo-oestradiol cypionate had little effect on serum lipoproteins. To date there has only been one detailed report dealing with the effects on lipoproteins of oestrogens administered as a subcutaneous (sc) implant. Brook et al. (1982) studied 3 women for a period of 12 weeks and noted profound changes in serum lipoproteins, particularly in high density lipoprotein (HDL) and its subclasses.

In this study we describe changes in lipoproteins over a period of 6 months in 31 oophorectomised women who were receiving hormone replacement therapy by way of sc implants. Two types of implant were used, one containing only oestradiol- 17(5, the other an oestradiol/testosterone mixture, since it has been suggested (Studd et al. 1977) that the inclusion of testosterone improves libido in postmenopausal women.

Patients and methods

Thirty-one women attending menopausal clinics at the Western Infirmary and Stobhill Hospital. Glasgow who were suffering from climacteric symptoms were treated with hormone implants. They were aged between 36 and 54 years (mean age 46.4 years) and all had undergone hysterectomy and bilateral oophorectomy for non-malgi- nant conditions. At least 6 weeks had elapsed post-operation prior to commencing treatment and informed consent was obtained in all cases before implant inser-tion. None of the women had received any hormone therapy prior to commencing treatment nor were taking any drug liable to interfere with lipid metabolism. None of the women had renal or hepatic abnormalities before or during treatment as indicated by routine biochemical tests.

The women were randomly divided into two groups. Fourteen women were given oestrogen implants (50 mg oestradiol-17β) and the remaining 17 were given oestrogen/testosterone implants (50 mg oestradiol-17β, 100 mg testosterone).

Blood was obtained, after a 14 h fast, prior to insertion of the implant and 2, 4 and 6 months post implantation. Serum was separated by centrifuging at 1000 × g for 10 min at 4C,C. An aliquot of serum was stored at 4°C for a maximum of 5 days prior to lipoprotein analysis and a further aliquot stored at — 20°C for steroid analysis.

Serum lipoproteins were separated into their density classes by ultracentrifugation (Airfuge, Beckman Instruments Ltd., High Wycombe, F.ngland) as previously described (Parish ct al. 1983). In addition, total HDL and HI)I, subfractions were measured in 10 of the women who had oestrogen implants and 9 who had oestrogen/ testosterone implants using the method described by Eyre et al. (1981). Cholesterol concentrations were esti-mated manually using an enzymatic technique (Ailain et al. 1974). Total serum triglyceride concentrations were also quantitated enzymatically (Bucolo & David 1973) using a Gemini centrifugal analyser (Electro-Neucleonics Inc. Breda, The Netherlands).

Serum oestradiol was measured by a radioimmuno-assay (RIA) which used a rabbit antiserum raised against oestradiol-6-O-carboxymethyIoxime-BSA, [3H]oestradiol and a double antibody separation. The sensitivity of the system was 100 pmol/1 and the inter-assay CV was 11%. Serum testosterone was also measured by RIA using a double antibody system employing a rabbit antiserum raised against testosterone-3-O-carboxymethyloxime- BSA and an [_I25I]histamine conjugate of testosterone-3- O-carboxymethyloxime as radioligand. The assay had a sensitivity of 0.5 nmol/l and an inter-assay precision of 10%.

The results obtained were analysed by standard statis-tical techniques. Lilliefors’ test (Conover 1975) was used to ascertain whether the sample data were normally distributed. Lipoprotein concentrations during treatment were compared with baseline values using Student’s paired Mesis when the data were normally distributed and Wileoxon's matched pairs signed ranks tests when they were not.

Results

Oestradiol and testosterone levels for the 31 women are shown in table 1. In both groups oestradiol levels rose in the first 2 months to prc-mcno- pausal levels at which they remained for the rest of the trial. Testosterone levels remained constant in the oestrogen-only group for the whole of the 6 months. In the oestrogen/testosterone group, they rose in the first 2 months then gradually fell to pre-treatment levels at b months.

Table 1. Oestradiol and testosterone levels (mean ± SEM) in postmenopausal women treated with hormone implants.

Lipoprotein levels for the two groups of women are shown in tables 2 and 3. Neither implant had any marked effects on lipoproteins. The oestrogen implants caused a reduction in low density lipoprotein (I.D1.) cholesterol and a small increase in HDI. cholesterol. The only significant change caused by the oestradiol/testosterone implants was a reduction in LDL levels.

Table 2. Lipoprotein levels (mean ± SD) in postmenopausal women treated with oestrogen-only implants.

Table 3. SLipoprotein levels (mean ± sn) in postmenopausal women treated with oestrogcn/testosterone implants.

Table 4 shows total HDL, HDL² and HDL³ levels for 10 of the women who had oestrogen implants. There was a temporary elevation of HDL² cholesterol levels at 2 months and a slower but more sustained increase in HDL³ cholesterol levels. The levels of total HDL and the HDL subfractions were unaffected by the oestrogen/testosterone implants (table 5).

Table 4. Total HDL and HDL, stdif'iactions (mean ± sn) in postmenopausal women treated with oestrogen-onlv implants.

Table 5. Total HDL and HDL subtractions (mean ± sn) in postmenopausal women treated with oestrogen./tcstosterone implants.

Discussion

Oral natural estrogens have been shown to elevate HDL cholesterol levels and reduce LDL cholesterol levels (Tikkanen et al. 1978; Silfverstolpc et al. 1980; Fahraeus et al. 1982). However, these reports were contradictory regarding their effects on triglyceride and very low density lipoprotein (VLDL) cholesterol. Androgens reduce HDL and VLDL cholesterol and increase LDL cholesterol (Furman & Howard 1962; Solyon 1971).

We have shown that implants do not produce the profound alterations in serum lipids produced by oral therapy. The only significant alterations were the lowering of LDL cholesterol by both types of implant and the increasing of HDL cholesterol by the oestrogen implants and none of these changes were large. Fahraeus et al. (1982) investigating the effects of parenteral oestradiol-17β, produced similar results in that they found a gradual decrease in LDL cholesterol levels. However, they found no changes in HDL cholesterol levels.

The effects of natural oestrogens and androgens on HDL suhfractions are not well documented. Brook et al. (1982) reported oestradiol-17β implants caused marked increases in both HDL₂ and HDL₃ cholesterol over a period of 3 months. This contrasts with our finding of small increases in HDL subtractions. However, Brook et al. (1982) studied only 3 women and used implants containing double the amount of oestradiol-17β used in the present study. We found the addition of testosterone to the implants to have only a slight effect on the HDL subfraction changes, preventing the small rises produced by oestrogen alone.

It is not as yet clear why parenteral oestrogen therapy should differ so markedly from oral therapy in its effects since the oestradiol levels we found with the implants were comparable to those produced by oral therapy (Lind et al. 1979). A likely explanation cited by previous workers (Buck- man et al. 1980; Fahraeus et al. 1982) is that oral oestrogens lead to peaks in intrahepatic and plasma concentrations and these cause the marked changes in lipoprotein levels.

In a recent study carried out on a similar group of bilaterally oophorectomised women using exactly the same therapy regimens, Dow et al. (1983) found that both types of implant were equally effective in reducing the severity of psychological, somatic and vasomotor symptoms and significantly improved libido. Their results indicated that the additional use of testosterone offers no advantage over oestradiol alone in the treatment of sexually unresponsive women. There is little to choose between the two types implant with regard to their effects on the lipoprotein risk factors for coronary heart disease. Although the oestrogen implants caused a small increase in HDL cholesterol not in evidence with the combined implant, this was mainly due to a slow rise in HDL₃ cholesterol, the rise in the anti-atherogenic HDL₂ fraction being transient. In addition, testosterone supplementation appeared to slightly enhance the LDL chole-sterol-lowering effect of oestradiol.

In conclusion, we have confirmed that oestrogen implants affect lipoproteins in a similar manner, but to a much lesser extent than oral oestrogen therapy and that the addition of testosterone to the implants, although of dubious clinical value, has little further effect on lipoprotein status.

References

Allain C C, Poon L S, Chan CSC, Richmond W & Fu P C (1974): Enzymatic determination of total serum cholesterol. Clin Chem 20: 470-475. 

Brook J G, Aviram M, Oettingcr M & Sharf M (1982): The effects of oestrogen implants on high density lipoprotein and its subfractions in women in their pre-mature menopause. Maturitas 4: 257-265. 

Buckman M T, Johnson J, Ellis H, Srivastava L & Peake G T (1980): Differential lipemie and protéinémie re-sponse to oral ethinyl estradiol and parenteral estradiol cypionate. Metabolism29: 803-805. 

Bucolo G & David H (1973): Quantitative determination of serum triglycerides by the use of enzymes. Clin Chem 19:476-482. 

Conover W J (1975): Practical nonparametric statistics, p 302-305. John Wiley & Sons, New York. 

Dow M G T, Hart D M Sc Forrest C A (1983): Hormonal treatments of sexual unresponsiveness in post-menopausal women: a comparative study. Br J Obstet Gynaecol 90: 361—366. 

Eyre J, Hammett F & Miller N E (1981): A micro-method for the rapid ultracentrifugal separation of human plasma high density lipoprotein subfractions HDLj and HDL3. Clin Chim Acta 114: 225-231. 

Fâhraeus L, Larsson-Cohn U & Wallentin L (1982): Lipoproteins during oral and cutaneous administration of oestradiol-17(1 to menopausal women. Acta Endocrinol (Copenh) 101: 597-601. 

Farish E, Fletcher C D, Hart D M, Kitchener H 8c Sharpe G I. M (1983): A long term study of the effects of norcthisterone on lipoprotein metabolism in meno-pausal women. Clin Chim Acta 132: 193-198. 

Furman RH & Howard R P (1962): Effects of androste- rone and triiodothyronine on serum lipids and lipo-proteins, nitrogen balance and related metabolic phe-nomena in subjects with normal and decreased thyroid function, with hyperglyceridemia and/or hyperchole-sterolemia. Metabolism 11: 76-93. 

Gordon T, Castelli W P, Hjortland M C, Kannel W B & Davrber T R (1977): High density lipoprotein as a t* protective factor against coronary heart disease. Am J Med 62:707-714. 

Lind T, Cameron F. C, Hunter W M, Leon C, Moran P K, Oxley A, Gerrard J & Lind U C G (1979): A prospective, controlled trial of six forms of hormone replacement therapy given to postmenopausal women. Br J Obstet Gynaecol 86, Suppl 3: 1-29.* 

Miller N E, Forde O H, Thelle D S & Mjos O (1977) : The Tromso heart study. High-density lipoprotein and coronary heart disease: a prospective case-control study. Lancet 1: 965—968. 

Silfverstolpe G, Gustafson A, Samsioe G & Svanborg A (1980): Lipid metabolic studies in oophorectomized women. Effects induced by two different estrogens on serum lipids and lipoproteins. Gynecol Obstet Invest II: 161-169. 

Solyon A (1971): Effects of androgens on serum lipids and lipoproteins. Lipids 7: 100-105. 

Studd J W W, Collins W P, Chakravarti S, Newton J R, Oram D & Parsons A (1977): Oestradiol and testosterone implants in the treatment of psychosexual problems in the postmenopausal woman. Br J Obstet Gynaecol 84: 314—315. 

Tikkanen M J, Nikkila E A & Vartiainen E (1978): Natural oestrogen as an effective treatment for type II hyperlipoproteinaemia in postmenopausal women. Lancet 2: 490-491.