Effect of postmenopausal oestradiol and dydrogesterone therapy on lipoproteins and insulin sensitivity, secretion and elimination in hysterectomised women

Maturitas 42 (2002) 233– 242 www.elsevier.com/locate/maturitas

Effect of postmenopausal oestradiol and dydrogesterone therapy on lipoproteins and insulin sensitivity, secretion and elimination in hysterectomised women N.A. Manassiev a, I.F. Godsland b,*, D. Crook c, A.J. Proudler b, M.I. Whitehead a, J.C. Stevenson b a Menopause Clinic, King’s College Hospital, London, UK Rosen Laboratories of the Wynn Institute, Endocrinology and Metabolic Medicine, Faculty of Medicine, Imperial College of Science, Technology and Medicine, Norfolk Place, London W2 1PG, UK c Department of Cardio6ascular Biochemistry, St. Bartholomew’s and the Royal London School of Medicine and Dentistry, London, UK b

Received 2 August 2001; received in revised form 11 March 2002; accepted 21 March 2002

Abstract Objecti6es: To investigate in depth the metabolic effects of oestradiol-17 b both alone and in combination with the progestagen dydrogesterone. Methods: Fifteen hysterectomised postmenopausal women were studied before treatment and after 24 weeks taking oestradiol-17 b alone (2 mg per day), then following a further 6 (oestrogen-alone phase) and 12 (oestrogen plus progestagen phase) weeks with inclusion of dydrogesterone (10 mg per day for days 17 – 28 of each 28 day cycle). Measurements at each visit included fasting serum lipid and lipoprotein concentrations, insulin sensitivity, secretion and elimination by modelling analysis of intravenous glucose tolerance test glucose, insulin and C-peptide concentrations, body fat distribution by dual-energy X-ray absorptiometry (DXA) and arterial function by carotid artery ultrasound. Results: Significant reductions were seen throughout in total and LDL cholesterol. The net reductions in total and LDL cholesterol by the end of the study were 5.8% (PB 0.05) and 18.4% (PB 0.001), respectively. HDL and HDL subfraction cholesterol concentrations rose during treatment with oestradiol alone, the rise being primarily in the HDL2 subfraction ( +21.6%, PB 0.001). Fasting serum triglycerides rose 30% on oestradiol treatment. These increases were unaffected by the addition of dydrogesterone. Insulin sensitivity, secretion and elimination, body fat distribution and arterial function were not significantly affected at any stage of the therapy. Conclusions: The small study sample and high variability in measures of glucose and insulin metabolism may have contributed to the absence of the expected significant improvement in these parameters. Orally administered oestradiol had beneficial effects on total, LDL and HDL cholesterol which were unaffected by the addition of dydrogesterone. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Oestradiol; Dydrogesterone; Hormone replacement therapy; Lipids; Insulin resistance; Insulin secretion

* Corresponding author. Tel.: + 44-171-594-3881; fax: +44-171-594-3875. E-mail address: [email protected] (I.F. Godsland). 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 2 ) 0 0 0 6 9 - 5

234

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242

1. Introduction Postmenopausal oestrogen replacement therapy, given when necessary with a progestagen for endometrial protection, is an established treatment for menopausal symptoms and prevention of osteoporosis. There continues to be widespread interest in the other potential health benefits of hormone replacement therapy (HRT), for example, cardiovascular disease (CVD). Theoretically, there may also be benefits with regard to diabetes, since oestrogens can reduce insulin resistance and, in experimental animals, increase pancreatic insulin secretion [1]. Postmenopausal women who take oestrogens have a lower incidence of CVD than those who do not [2]. Oestrogen-induced decreases in low density lipoprotein (LDL) cholesterol and increases in high density lipoprotein (HDL) cholesterol have been nominated as contributing factors [3]. This area remains controversial, however, particularly following recent reports that the combination of conjugated equine oestrogen and medroxyprogesterone acetate failed to lower coronary event rates or diminish angiographic progression in postmenopausal patients with CVD [4,5]. There is also conflicting evidence as to the incidence of diabetes amongst HRT users [6,7]. However, CVD is the principal cause of morbidity and mortality amongst diabetics, so it is noteworthy that the lower incidence of CVD amongst HRT users extends to women with diabetes [8]. Circumstantial evidence for the beneficial effects of HRT on CVD and diabetes continues to come from studies at the metabolic and cellular levels. To an extent that depends on the type, dose and route of administration of the steroids used, HRT can affect lipoprotein metabolism, endothelial function, glucose and insulin metabolism and body fat distribution in ways that would be expected to reduce the risks of both diseases. The manifold interrelated changes in metabolism and vascular function that take place at and following the menopause suggest a distinct menopausal metabolic syndrome [9], characterised by decreased levels of the HDL2 subfraction, increased levels of LDL, reduced pancreatic insulin secretion and peripheral insulin elimination, and

progressive reductions in insulin sensitivity and uptake of insulin by the liver. Oestrogen replacement can diminish most manifestations of the menopausal metabolic syndrome, but any accompanying progestagen may oppose these benefits. For example, both levonorgestrel and norethisterone can oppose oestrogen-induced increases in HDL [10] and both levonorgestrel [11] and medroxyprogesterone acetate [12,13] can oppose oestrogen-induced improvements in insulin sensitivity. Dydrogesterone is an isomer of 6-dehydroprogesterone which does not oppose the beneficial effects of oestrogen on HDL and LDL concentrations [14–17]. Moreover, there is growing evidence that improvements in insulin sensitivity, secretion and elimination induced by oestradiol17b are preserved in the presence of dydrogesterone [16,18,19]. In the present study, we have extended the investigation of the effects of dydrogesterone in combination with oestradiol to hysterectomised women. This has enabled the effects of oestradiol alone to be evaluated, as well as the effects of inclusion of dydrogesterone. We assessed a range of interrelated metabolic and physiologic parameters of the menopausal metabolic syndrome, including fasting serum lipid and lipoprotein levels, insulin sensitivity, secretion and elimination (by mathematical modelling analysis of intravenous glucose tolerance test (IVGTT) glucose, insulin and C-peptide levels), vascular function (carotid artery pulsatility index) and body fat distribution (measured by dual energy X-ray absorptiometry (DXA)).

2. Methods

2.1. Study population Participants were healthy, postmenopausal, hysterectomised white women, aged 40–65 years, attending the Menopause Clinic of King’s College Hospital. FSH levels were \ 35 IU l − 1. Women were within 20% of their ideal body weight, were non-smokers and consumed B 15 U of alcohol per week. No HRT steroids had been taken within 6 months prior to the study. Women with

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242

thyroid dysfunction, recent myocardial infarction, hypertension or other clinically significant metabolic, endocrine or gastrointestinal disease were excluded, as were women taking any medication likely to affect lipid or carbohydrate metabolism. Each participant gave signed, informed consent and local ethical committee approval for the study was obtained.

2.2. Study design This was an open study lasting 36 weeks and comprising of nine 28-day treatment cycles. For the first six cycles (24 weeks), subjects took 2 mg per day oestradiol (Zumenon: Solvay Pharmaceuticals). For a further three cycles (weeks 24– 36), subjects continued to take the same oestrogen, but combined with 10 mg dydrogesterone (Duphaston: Solvay Pharmaceuticals) for the last 14 days of each 28-day cycle. Fasting lipid measurements, an IVGTT and measurement of carotid artery pulsatility index were performed at baseline prior to treatment, after six cycles (24 weeks) of treatment with oestradiol alone, after 7.5 cycles (30 weeks) at the end of the oestrogen alone phase during the second cycle of combined therapy, and after nine cycles (36 weeks) at the end of the oestrogen plus progestagen phase of the third cycle of combined therapy. Body composition was measured at baseline, at the end of the oestrogen alone phase of treatment and during the combined phase of treatment, either at 7.5 or nine cycles of treatment.

2.3. Procedures Each volunteer was instructed to eat \ 200 g per day carbohydrate for 3 days prior to testing to minimise dietary-induced variation in the pancreatic insulin response to glucose and then to attend a metabolic day ward following a 12-h overnight fast. Height and weight were measured and details of exercise habits and alcohol consumption were taken. Blood pressure was recorded after 10 min bed rest in a semi-recumbent position, after which cannulae were inserted into ante-cubital veins of each arm, the cannula in the non-dominant arm being used for blood sampling. Blood samples for

235

measurement of fasting plasma glucose, insulin and C-peptide (lithium-heparin anti-coagulant) and serum lipid, lipoprotein and apolipoprotein levels (plain tubes with granules to assist clot retraction) were taken through this cannula. Additional samples were taken for a second measurement of fasting glucose, insulin and C-peptide. Samples were placed on ice immediately and separated, in the case of plasma within 30 min and for serum after 60 min had elapsed, for clot formation and retraction. Following fasting sampling, an intravenous glucose injection (0.5 g kg − 1 body weight as 50% dextrose solution) was given over 3 min via the opposite cannula, which was then withdrawn. Samples for measurement of glucose, insulin and C-peptide were taken at 3, 5, 7, 10, 15, 20, 30, 45, 60, 75, 90, 120, 150 and 180 min following commencement of the injection. On completion of the IVGTT at baseline visit 1, visit 2 and either visit 3 or visit 4, regional body fat mass was measured by DXA using a Lunar DPX scanner (Lunar Corp., Madison, WI). Carotid artery pulsatility index was measured at each visit as previously described [20].

2.4. Laboratory measurements Plasma glucose was measured within 24 h on samples stored at − 4 °C using a glucose oxidase method [21]. Plasma insulin and C-peptide were measured in batches on samples stored at − 20 °C by double antibody radioimmunoassay with materials supplied by Guildhay, Surrey, UK. Within- and between-batch coefficients of variation ranged between 2 and 3% (glucose), 4–6% (insulin) and 7–9% (C-peptide). Serum total cholesterol and triglycerides were measured by fully enzymatic methods. HDL cholesterol and HDL3 cholesterol were measured after sequential precipitation with heparin and manganese ions [22] and dextran sulphate [23], respectively. HDL2 cholesterol was calculated as the difference between HDL and HDL3 cholesterol. Very low density lipoprotein (VLDL) cholesterol and triglycerides were measured after separation by preparative ultracentrifugation at a solvent density of 1.006 g l − 1. Low density lipoprotein was calculated as the difference between total choles-

236

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242

terol and the sum of HDL and VLDL cholesterol. Apolipoproteins AI, AII and B were measured by immunoturbidimetry [24].

2.5. Data analysis Mean fasting plasma glucose, insulin and Cpeptide levels were derived from the two basal measurements and IVGTT incremental glucose, insulin and C-peptide areas (i.e. the area between the fasting level and the IVGTT concentration profile) were calculated by the trapezium rule (i.e. the sum of all areas consisting of time interval multiplied by the average of the concentrations at the beginning and end of that time interval). Measures of insulin sensitivity, secretion and elimination were derived by mathematical modelling analysis of the IVGTT glucose, insulin and C-peptide concentration profiles, using programs written in Fortran 77. Insulin sensitivity, Si and glucose effectiveness, Sg, were determined using the minimal model of glucose disappearance [25]. The relatively high glucose dose (0.5 g kg − 1) we employ provides for a sufficient endogenous insulin response in non-diabetic volunteers without recourse to additional augmentation of pancreatic insulin secretion. This is apparent in the high rate of model identification and good correlation with measures of insulin sensitivity derived from the euglycaemic clamp (r =0.92) that we obtain [26,27]. Insulin delivery characteristics were evaluated using the minimal model of post-hepatic insulin delivery [28], which provides measures of the fractional insulin elimination rate and the responsiveness of first and second phase post-hepatic insulin delivery to glucose, b1 and b2, respectively. A combined model of insulin and C-peptide delivery [29,30] was used to quantify fractional insulin and C-peptide elimination rates, a measure of the fraction of newly-secreted insulin that passes out of the liver, and basal and incremental insulin secretion during the IVGTT, in total and during the first and second phases of secretion. The models describing insulin delivery have been evaluated previously in both animals and humans [26,29,31]. For the minimal models of glucose disappearance and posthepatic insulin delivery, a valid analysis was one that converged

to a solution for which all parameters were positive and parameter fractional S.D. were B 100%. For the combined model of pancreatic insulin secretion, an analysis was considered valid if the fractional hepatic insulin throughput did not exceed 1.4 [11,32]. DXA-derived regional fat mass was defined with reference to anatomical bone landmarks, as previously described [33]. Android fat was taken as the total fat mass in the trunk region and gynoid fat as the total fat mass in the leg region. The android/gynoid fat mass ratio was calculated. For analysis, measurements made either at 6 or 12 weeks during the combined treatment part of the study (study weeks 30 and 36) were combined to provide a single, complete set of measurements made on all participants during the combined treatment part. Android/gynoid fat mass ratios were therefore available for baseline, oestrogenonly and combined treatment parts of the study. Statistical analyses were carried out using the SYSTAT statistical package (SYSTAT Inc, Evanston, IL). Triglyceride and VLDL levels were log-transformed prior to analysis. Normalising transforms were applied to insulin-related variables, as previously described [26]. Significant variation with treatment for each variable was identified by repeated measures analysis of variance and, where significant variation was found, post-hoc, pairwise comparisons were performed by linear contrasts.

3. Results Of the 20 patients entered into the study, three withdrew for reasons unrelated to the treatment, one withdrew because of breast tenderness and one declined to explain her withdrawal. Results for the remaining 15 patients who completed the study are given here. Of these women, one did not attend for her third visit and results for one woman’s fourth visit were excluded since she had been taking dydrogesterone for only 1 day at the time of testing. The women were aged between 49 and 65 years (mean 54.8) with an age at menopause of 41–56 years (mean 49.2). The number of years since menopause ranged from 0.5 to

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242

237

creases at the end of the study were 18.2 and 9.8%, respectively). There was no significant variation apparent on repeated measures analysis of variance in fasting or IVGTT glucose, insulin or C-peptide concentrations (Table 3). Neither was there any significant variation in the glucose elimination constant, k. For the minimal model of glucose disappearance there was one invalid modelling analysis, which was at visit 3. For the minimal model of post-hepatic insulin delivery, there were five invalid modelling analyses: two at visit 1, one at visit 2 and two at visit 4. For the combined model of pancreatic insulin secretion, there were two invalid analysis, both at visit 4. There was no significant variation apparent in any of the modelderived measures of insulin resistance, secretion or elimination (Table 4), neither was there significant variation in measures of carotid artery pulsatility and resistivity (Table 5).

15 (mean 6.2). Body mass index (BMI) ranged from 16.7 to 27.1 kg m − 2, android/gynoid fat mass ratio from 0.51 to 1.61, systolic blood pressure from 110–140 mmHg, diastolic blood pressure from 65 to 80 mmHg. BMI, android/gynoid fat mass ratio and blood pressure did not change during the course of the study (Table 1). Repeated measures analysis of variance revealed significant variation in all lipid, lipoprotein and apolipoprotein measures except HDL3 cholesterol, VLDL cholesterol and triglycerides and apolipoprotein B. Significant reductions were seen in total and LDL cholesterol after six cycles of treatment with oestradiol alone and this fall was sustained during combined oestradiol/dydrogesterone treatment, both during the oestrogen-alone and the oestrogen/progestagen phases of the treatment cycle (Table 2). The net fall in total and LDL cholesterol by end of the study was 5.8 and 18.4%, respectively. Fasting serum triglycerides rose 30% on treatment with oestradiol alone and there was a slight attenuation of this effect during the combined treatment (− 7.4 to − 3.7%). HDL and HDL subfraction cholesterol concentrations rose during treatment with oestradiol alone, the greatest rise (21.6%) being seen in the HDL2 subfraction. These increases were not diminished by inclusion of dydrogesterone. Apolipoproteins AI and AII were elevated compared with pretreatment levels throughout the study (overall in-

4. Discussion Twenty-four weeks of treatment with orally administered oestradiol resulted in changes in measures of lipid metabolism that would be expected to accompany a reversal of the metabolic effects of the menopause: HDL cholesterol, particularly in the HDL2 subfraction, increased and

Table 1 Mean (S.D.) BMI, android/gynoid fat mass ratio (A/G ratio), systolic and diastolic blood pressure in the15 women who completed the study Visit 1: pre-treatment

Body mass index (kg m−2) Android/gynoid fat mass ratio Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)

24.3 (3.0) 1.19 (0.37)

Visit 2: oestradiol alone, week 24

24.5 (3.2) 1.12 (0.44)

Visit 3: oestradiol+dydrogesterone, week 6 (oestrogen only phase, week 30) 24.6 (3.4) –

Visit 4: oestradiol+dydrogesterone, week 12 (combined phase, week 36) 24.4 (3.3) 1.14 (0.44)

118.7 (12.0)

121.5 (12.0)

115.1 (9.0)

120.8 (17.7)

73.0 (5.6)

72.4 (7.7)

74.2 (5.2)

71.8 (8.6)

238

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242

Table 2 Mean (S.D.) lipid, lipoprotein and apolipoprotein concentrations at successive time intervals in the 15 women who completed the study Visit 1: pre-treatment

Visit 2: oestradiol alone, week 24

Visit 3: oestradiol+dydrogesterone, week 6 (oestrogen only phase, week 30)

Total cholesterol 5.49 (1.01) 5.25 (1.16)† 5.28 (0.94)* (mmol l−1) LDL cholesterol 3.48 (0.88) 2.97 (1.00)§ 2.96 (0.82)§ (mmol l−1) Triglycerides 1.25 (−0.44, +0.67) 1.04 (−0.43, +0.73) 1.35 (−0.53, +0.87)† (mmol l−1) c HDL cholesterol 1.47 (0.30) 1.60 (0.40) 1.70 (0.34)† −1 (mmol l ) HDL2 0.51 (0.26) 0.62 (0.31) 0.67 (0.28)§ cholesterol (mmol l−1) HDL3 0.96 (0.12) 0.97 (0.17) 1.02 (0.16) cholesterol (mmol l−1) VLDL 0.18 (−0.08, +0.16) 0.20 (−0.11, +0.23) 0.18 (−0.08, +0.15) cholesterol (mmol l−1) c VLDL 0.30 (−0.14, +0.25) 0.32 (−0.17, +0.35) 0.34 (−0.14, +0.23) triglyceride (mmol l−1) c Apolipoprotein 142.9 (21.5) 162.9 (40.8) 170.3 (31.4)* AI (g l−1) Apolipoprotein 38.6 (5.1) 45.2 (8.3)* 44.6 (7.7)* AII (g l−1) Apolipoprotein 76.6 (22.9) 83.9 (19.2) 75.8 (18.0) B (g l−1)

Visit 4: oestradiol+dydrogesterone, week 12 (combined phase, week 36) 5.17 (0.96)* 2.84 (0.88)§ 1.30 (−0.46, +0.71)* 1.68 (0.33) 0.67 (0.26)§

1.01 (0.12)

0.16 (−0.08, +0.15)

0.33 (−0.16, +0.30)

168.9 (27.9)† 42.4 (6.4)* 69.6 (17.4)

c Back-transformed from log-transformed data. * PB0.05, significant differences compared with baseline. † PB0.01, significant differences compared with baseline. § PB0.001, significant differences compared with baseline.

LDL cholesterol decreased. In accord with the increases in HDL, there were significant elevations in the principal apolipoproteins of HDL, apoAI and apoAII. With regard to triglycerides, a reversal of the effects of the menopause would be expected to lead to a reduction in concentrations. The paradoxical increase seen in the present study is likely to be a pharmacological effect of the oral route of administration, since it is not seen with parenteral administration [34,35]. None of these changes were affected by a further 12 weeks inclusion of dydrogesterone in the HRT regimen. Dydrogesterone therefore behaves in combination

with oestrogen in relation to lipid metabolism very much like progesterone [36]. There were no statistically significant effects on glucose and insulin metabolism in our study. Previous studies with these steroids have demonstrated reductions in basal insulin concentrations and increases in basal C-peptide concentrations [16,18,19], with similar changes following oral glucose [16,19]. Such changes are consistent with improvements in both insulin sensitivity and secretion and, as such, are consistent with other studies of the effects of oestrogens [1]. In a recent report by Cucinelli et al., increased insulin sensi-

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242

tivity was detected, with both oestradiol alone and oestradiol in combination with dydrogesterone [19]. Moreover, increased hepatic insulin uptake was inferred from changes in insulin:C-peptide molar ratios. Several explanations may be considered to account for our findings. Possibly, the 24 weeks of therapy with oestradiol alone and the additional 12 weeks of oestradiol/dydrogesterone did not provide sufficient time for effects on glucose and insulin metabolism to develop. In accord with this, no change in body fat distribution was apparent, in contrast to both theoretical expectations [9] and findings from other studies [37]. Nevertheless, significant changes in glucose and insulin metabolism have been reported in earlier studies after 12 [19] or 24 [16] weeks and in the present study, effects on lipids and lipoproteins were clearly apparent. With regard to the methodology employed, Cagnacci et al. observed decreases in insulin concentrations and increases in C-peptide concentrations following oral but not intravenous glucose in response to transdermally administered oestradiol [38]. This could call into

239

question the suitability of the IVGTT as a procedure for detecting HRT-induced changes in glucose and insulin metabolism. However, we have previously demonstrated marked effects of different HRT regimens on glucose and insulin metabolism using the IVGTT [11,39], including improvements in insulin sensitivity, secretion and elimination [39]. There remains the possibility that larger numbers of study subjects or less heterogeneity in BMI and fat distribution would have been necessary to provide for an effective investigation of glucose and insulin metabolism. In the present study, a possible improvement in insulin sensitivity was apparent in the decline in mean IVGTT insulin response, with no accompanying change in glucose levels or glucose elimination rate. There were also possible improvements in insulin secretion apparent in the posthepatic insulin delivery parameter, b2, and in basal pancreatic insulin secretion which increased between 52 and 66% on treatment. Peripheral insulin elimination, ki, was also substantially increased. There was considerable within and between individual variation in

Table 3 Mean (S.D.) fasting and IVGTT glucose, insulin and C-peptide measures at successive time intervals in the 15 women who completed the study

Mean fasting glucose (mmol l−1) Mean fasting insulin (mU l−1)§ Mean fasting C-peptide (ng ml−1)§ IVGTT glucose inc AUC (mmol l−1 min−1) IVGTT insulin inc AUC (mU l−1 min−1)§ IVGTT C-peptide inc AUC (ng ml−1 min−1)§ k (min−1×10[2]) §

Visit 1: pre-treatment

Visit 2: oestradiol alone, week 24

Visit 3: oestradiol+dydro- Visit 4: oestradiol+dydrogesterone, week 6 gesterone, week 12 (oestrogen only (combined phase, week 36) phase, week 30)

5.21 (0.48)

5.11 (0.66)

5.02 (0.42)

4.95 (0.47)

4.19 (−2.25, +4.85) 5.13 (−2.98, +7.10) 6.13 (−3.45, +7.91)

6.80 (−4.14, +10.6)

1.49 (−0.51, +0.78) 1.47 (−0.52, +0.80) 1.46 (−0.58, +0.97)

1.50 (−0.58, +0.94)

506 (155)

552 (200)

543 (164)

493 (128)

2657 (−965, +1515) 286 (−107, +172)

2057 (−689, +942)

1963 (−1125, +2139)

1773 (−1193, +3521)

310 (−78, +103)

279 (−94, +141)

277 (−109, +180)

2.36 (−0.93, +1.52) 1.98 (−1.00, +2.00) 2.06 (−0.99, +1.91)

Back-transformed from log-transformed data.

2.26 (−0.87, +1.42)

0.45 (−0.18, +0.30) 0.069 (−0.027, +0.046) 0.015 (−0.005, +0.008) 0.82 1.99 0.75 1.30

Combined model of pancreatic insulin secretion (a) Insulin and C-peptide elimination Fractional hepatic insulin throughput, f § Insulin elimination rate, ki (min−1)§ C-peptide elimination rate, kc (min−1)§

(b) IVGTT pancreatic insulin secretion Net basal (pmol l−1 min−1)§ Net incremental (pmol l−1 min−1)§ Phase 1 incremental (pmol l−1 min−1) Phase 2 incremental (pmol l−1 min−1)†

§

†

3.27 (−1.59, +3.10) 4.06 (−2.42, +6.00)

Minimal model of post-hepatic insulin deli6ery b1 (mU l−1 min−1 mg−1 ml−1)§ b2 (mU l−1 min−2 mg−1 ml−1)§

Back-transformed from square-root-transformed data. Back-transformed from log-transformed data.

(−0.45, +0.97) (−0.97, +1.28) (0.60) (1.02, 1.21)

3.76 (−2.01, +2.77) 1.36 (−0.60, +1.08)

Minimal model of glucose disappearance SI (min−1 mU−1 l)† Sg (min−1)§

Visit 1: pre-treatment

1.36 2.73 0.81 1.95

(−0.68, +1.36) (−1.12, +1.46) (0.41) (1.41, 1.73)

0.59 (−0.23, +0.38) 0.135 (−0.078, +0.186) 0.22 (−0.008, +0.012)

3.30 (1.23, 1.96) 3.40 (−2.48, +9.18)

3.08 (−1.40, +1.82) 1.47 (−0.77, +1.60)

1.27 1.24 0.60 0.79

(0.76, 1.88) (−1.06, +1.58) (0.39) (0.85, 1.00)

0.73 (−0.37, +0.76) 0.119 (−0.072, +0.183) 0.017 (−0.006, +0.009)

3.36 (−1.83, +4.00) 5.54 (−3.82, +12.3)

3.23 (−1.69, +2.31) 1.35 (−0.70, +1.45)

Visit 2: oestradiol alone, Visit 3: oestradiol+dydroweek 24 gesterone, week 6 (oestrogen only phase, week 30)

Table 4 Mean (S.D.) IVGTT model-derived measures in the 15 women who completed the study

1.25 1.94 0.81 1.27

(0.77, 1.97) (−1.26, +1.85) (0.36) (1.23, 1.52)

0.61 (−0.25, +0.41) 0.100 (−0.052, +0.108 0.018 (−0.008, +0.014)

3.57 (1.65, 3.05) 5.32 (−4.08, +17.4)

3.52 (−2.38, +3.70) 1.27 (−0.85, +2.59)

Visit 4: oestradiol+dydrogesterone, week 12 (combined phase, week 36)

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242 240

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242

241

Table 5 Mean (S.D.) carotid artery pulsatility and resistivity indices at successive time intervals in the 15 women who completed the study

Pulsatility index Resistivity index

Visit 1: pre-treatment

Visit 2: oestradiol alone week 24

Visit 3: oestradiol+dydrogesterone, Visit 4: oestradiol+dydrogesterone, week 6 (oestrogen only phase, week 30) week 12 (combined phase, week 36)

0.91 (0.88)

0.88 (0.79)

0.83 (0.85)

0.90 (0.90)

0.56 (0.55)

0.56 (0.53)

0.55 (0.55)

0.57 (0.57)

these measures and with larger numbers, statistically significant changes might well have been found.

[6]

Acknowledgements [7]

This study was supported by Solvay Healthcare Ltd., Southampton, UK, the Heart Disease and Diabetes Research Trust and the Rosen Foundation. Professor Richard Bergman and Dr Richard Watanabe kindly provided the programme for measuring insulin secretion by the combined model. We thank the nursing, laboratory and clinical staff who contributed to this study.

[8]

[9]

[10]

References [1] Godsland IF. The influence of female sex steroids on glucose metabolism and insulin action. J Int Med 1996;240(Suppl 738):1 –65. [2] Stampfer M, Colditz G. Estrogen replacement therapy: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47 –63. [3] Bush T, Barrett-Connor E, Cowan L. Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the Lipid Research Clinics Program Followup Study. Circulation 1987;75:1102 –9. [4] Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, Vittinghoff E. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/ Progestin Replacement Study (HERS) Research Group. J Am Med Assoc 1998;280:605 –13. [5] Herrington DM, Reboussin DM, Brosnihan KB, Sharp PC, Shumaker SA, Snyder TE, Furberg CD, Kowalchuk GJ, Stuckey TD, Rogers WJ, Givens DH, Waters D. Effects of estrogen replacement on the progression of

[11]

[12]

[13]

[14]

coronary-artery atherosclerosis. New Engl J Med 2000;343:522 – 9. Manson JE, Rimm EB, Colditz GA, Willett WC, Nathan DM, Arky RA, Rosner B, Hennekens CH, Speizer FE, Stampfer MJ. A prospective study of postmenopausal estrogen therapy and subsequent incidence of non-insulindependent diabetes mellitus. Ann Epidemiol 1992;2:665 – 73. Gabal LL, Goodman-Gruen D, Barrett-Connor E. The effect of postmenopausal estrogen therapy on the risk of non-insulin-dependent diabetes mellitus. Am J Publ Health 1997;87:443 – 5. Psaty BM, Heckbert SR, Atkins D, Lemaitre R, Koepsell TD, Wahl PW, Siscovick DS, Wagner EH. The risk of myocardial infarction associated with the combined use of estrogens and progestins in postmenopausal women. Arch Intern Med 1994;154:1333 – 9. Spencer CP, Godsland IF, Stevenson JC. Is there a menopausal metabolic syndrome? Gynecol Endocrinol 1997;11:341 – 55. Crook D, Cust M, Gangar K, Worthington M, Hillard T, Stevenson J, Whitehead M, Wynn V. Comparison of transdermal and oral estrogen-progestin replacement therapy: effects on serum lipids and lipoproteins. Am J Obstet Gynecol 1992;166:950 – 5. Godsland I, Gangar K, Walton C, Cust M, Whitehead M, Wynn V, Stevenson J. Insulin resistance, secretion and elimination in postmenopausal women receiving oral or transdermal hormone replacement therapy. Metabolism 1993;42:846 – 53. Lindheim SR, Presser SC, Ditkoff EC, Vijod MA, Stanczyk FZ, Lobo RA. A possible bimodal effect of estrogen on insulin sensitivity in postmenopausal women and the attenuating effect of added progestin. Fertil Steril 1993;60:664 – 7. Cefalu WT, Wagner JD, Bell-Farroe AD, Wang ZQ, Adams MR, Toffolo G, Cobelli C. The effects of hormone replacement therapy on insulin sensitivity in surgically postmenopausal cynomolgus monkeys (Macaca fascicularis). Am J Obstet Gynecol 1994;171:440 – 5. Siddle NC, Jesinger DK, Whitehead MI, Turner P, Lewis B, Prescott P. Effects on plasma lipids and lipoproteins of postmenopausal oestrogen therapy with added dydrogesterone. Br J Obstet Gynaecol 1990;97:1093 – 100.

242

N.A. Manassie6 et al. / Maturitas 42 (2002) 233–242

[15] Schram JH, Boerrigter PJ, The TY. Influence of two hormone replacement therapy regimens, oral oestradiol valerate and cyproterone acetate versus transdermal oestradiol and oral dydrogesterone on lipid metabolism. Maturitas 1995;22:121 –30. [16] Crook D, Godsland IF, Hull J, Stevenson JC. Hormone replacement therapy with dydrogesterone progestin and estradiol-17b: effects on serum lipoproteins and on glucose tolerance during 24 month follow-up. Br J Obstet Gynaecol 1997;104:298 –304. [17] Gelfand M, Fugere P, Bissonnette F. Cardiovascular risk factors during sequentially combined 17 beta oestradiol and dydrogesterone (Femoston); results from a one-year study in postmenopausal women. Maturitas 1997;26:125 – 32. [18] Gaspard UJ, Wery OJ, Scheen AJ, Lefebvre PJ. Longterm effects of oral estradiol and dydrogesterone on carbohydrate metabolism in postmenopausal women. Climacteric 1999;2:93 –100. [19] Cucinelli F, Paparella P, Soranna L, Barini A, Cinque B, Mancuso S, Lanzone A. Differential effect of transdermal estrogen plus progestagen replacement therapy on insulin metabolism in postmenopausal women: relation to their insulinemic secretion. Eur J Endocrinol 1999;140:215 –23. [20] Gangar KF, Vyas S, Whitehead M, Crook D, Meire H, Campbell S. Pulsatility index in internal carotid artery in relation to transdermal oestradiol and time since menopause. Lancet 1991;338:839 –42. [21] Trinder P. Determination of blood glucose using an oxidase-peroxidase system with non-carcinogenic chromogen. J Clin Pathol 1969;22:158 –61. [22] Warnick GR, Albers JJ. A comprehensive evaluation of the heparin-manganese precipitation procedure for estimating high density lipoprotein cholesterol. J Lipid Res 1978;19:65 – 76. [23] Gidez LI, Miller GJ, Burnstein M, Slagle S, Eder HA. Separation and quantitation of subclasses of human plasma high density lipoproteins by a simple precipitation procedure. J Lipid Res 1982;23:1206 –23. [24] Mount JN, Kearney EM, Rosseneu M, Slavin BM. Immunoturbidimetric assays for serum apoproteins A1 and B using Cobas Bio centrifugal analyser. J Clin Pathol 1988;41:471 – 4. [25] Bergman RN, Ider YZ, Bowden CR, Cobelli C. Quantitative estimation of insulin sensitivity. Am J Physiol 1979;236:E667 – 77. [26] Walton C, Godsland IF, Proudler AJ, Felton C, Wynn V. Evaluation of four mathematical models of glucose and insulin dynamics with analysis of the effects of age and obesity. Am J Physiol 1992;262:E755 –62. [27] Swan J, Walton C, Godsland IF. Assessment of insulin sensitivity in man: a comparison of minimal model and euglycaemic clamp derived measures in health and heart

failure. Clin Sci 1994;86:317 – 22. [28] Toffolo G, Bergman R, Finegood D, Bowden C, Cobelli C. Quantitative estimation of beta cell sensitivity to glucose in the intact organism. Diabetes 1980;29:979 – 90. [29] Vølund A, Polonsky K, Bergman R. Calculated pattern of intraportal insulin appearance without independent assessment of C-peptide kinetics. Diabetes 1987;36:1195 – 202. [30] Watanabe R, Vølund A, Roy S, Bergman R. Prehepatic beta-cell secretion during the intravenous glucose tolerance test in humans: application of a combined model of insulin and C-peptide kinetics. J Clin Endocrinol Metab 1989;69:790 – 7. [31] Godsland I, Felton C, Proudler A, Wynn V. Evaluation of two methods for estimating pancreatic insulin secretion by modelling analysis of the intravenous glucose tolerance test. Diabetes Med 1991;8(Suppl 1):23A. [32] Walton C, Godsland I, Proudler A, Felton C, Wynn V. An apparently anomalous relationship between insulin and C-peptide concentrations in their initial response to intravenous glucose. Metabolism 1992;41:1210 – 4. [33] Ley C, Lees B, Stevenson J. Sex- and menopause-associated changes in body fat distribution. Am J Clin Nutr 1992;55:950 – 4. [34] Castelo-Branco C, Casals E, Sanllehy C, Duran M, Fortuny A, Vanrell JA. Effects of progestogen on lipids, lipoproteins and apolipoproteins during transdermal estrogen replacement therapy with and without medroxyprogesterone acetate. J Reprod Med 1996;41:833 – 8. [35] Bhathena R, Anklesaria B, Ganatra A, Pinto R. The influence of medroxyprogesterone acetate on the effects of transdermal oestradiol replacement therapy on plasma lipids. J Obstet Gynaecol 1998;18:141 – 5. [36] The Writing Group for the PEPI Trial. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women: the Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. J Am Med Assoc 1995;273:199 – 208. [37] Haarbo J, Marslew U, Gotfredsen A, Christiansen C. Postmenopausal hormone replacement therapy prevents central distribution of body fat after menopause. Metabolism 1991;40:1323 – 6. [38] Cagnacci A, Tuveri F, Cirillo R, Setteneri A, Melis G, Volpe A. The effect of transdermal 17-beta-estradiol on glucose metabolism of postmenopausal women is evident during the oral but not the intravenous glucose administration. Maturitas 1997;28:163 – 7. [39] Spencer CP, Godsland IF, Cooper AJ, Ross D, Whitehead MI, Stevenson JC. Effects of oral and transdermal 17 beta estradiol with cyclical norethindrone acetate on insulin sensitivity, secretion and elimination in postmenopausal women. Metabolism 2000;49:742 – 7.