If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Impact of micronised progesterone and medroxyprogesterone acetate in combination with transdermal oestradiol on cardiovascular markers in women diagnosed with premature ovarian insufficiency or an early menopause: a randomised pilot trial
# INSTITUTION WHERE THE WORK WAS UNDERTAKEN: King's College Hospital NHS Foundation Trust, Department of Obstetrics and Gynaecology, Denmark Hill, Brixton, London SE5 9RS
Monica Mittal
Correspondence
Corresponding author. Imperial College Healthcare NHS Trust, St Mary's and Hammersmith Hospitals, Praed Street, Paddington, London W2 1NY, UK
# INSTITUTION WHERE THE WORK WAS UNDERTAKEN: King's College Hospital NHS Foundation Trust, Department of Obstetrics and Gynaecology, Denmark Hill, Brixton, London SE5 9RS
Affiliations
Imperial College Healthcare NHS Trust, St Mary's and Hammersmith Hospitals, Department of Obstetrics and Gynaecology, Praed Street, London W2 1NY
Oxford University Hospitals NHS Foundation Trust, John Radcliffe Hospital, Department of Obstetrics and Gynaecology, Headley Way, Headington, Oxford OX3 9DU
Imperial College Healthcare NHS Trust and Chelsea and Westminster NHS Foundation Trust, Queen Charlotte's and Chelsea Hospital, Department of Obstetrics and Gynaecology, Du Cane Rd, White City, London W12 0HS
# INSTITUTION WHERE THE WORK WAS UNDERTAKEN: King's College Hospital NHS Foundation Trust, Department of Obstetrics and Gynaecology, Denmark Hill, Brixton, London SE5 9RS
Risk stratification of arterial disease can be performed with carotid femoral pulse wave velocity.
•
Menopause augments the age-dependent increase in arterial stiffness.
•
Hormone replacement therapy is the mainstay of management of women diagnosed with an early menopause and premature ovarian insufficiency.
•
Micronised progesterone/medroxyprogesterone acetate can both be used in this cohort.
•
Positive changes in traditional markers with micronised progesterone were not reflected in the carotid femoral pulse wave velocity measurements.
ABSTRACT
Objective
To compare the difference between micronised progesterone (MP) and medroxyprogesterone acetate (MPA) in combination with transdermal oestradiol (t-E2) on cardiovascular disease (CVD) risk markers in women diagnosed with an early menopause and premature ovarian insufficiency (EMPOI).
Background
The European Society for Cardiology has identified carotid femoral pulse wave velocity (cfPWV) as the gold standard cardiogenic biomarker for risk stratification of arterial disease. Menopause has been shown to augment the age-dependent increase in arterial stiffness, with hormone replacement therapy (HRT) being the mainstay of management of women diagnosed with EMPOI.
Study design
A pilot randomised prospective open-label trial. Women were randomised to either cyclical MP (Utrogestan® 200mg) or MPA (Provera® 10mg) in conjunction with t-E2 (Evorel® Patches 50mcg/day) for 12 months. Seventy-one subjects were screened, and baseline data are available for 57 subjects.
Main outcome measure
Carotid-femoral pulse wave velocity (cfPWV).
Results
PWV did not significantly change from baseline in either treatment arm. MP + t-E2 demonstrated a positive effect on traditional CVD markers, with a significant improvement seen in cardiac output (CO) (0.71±1.01mL/min, 95% CI 0.20 to 1.21) and reduction in diastolic blood pressure (DBP) (-3.43±6.31mmHg, 95% Cl -6.57 to -0.29) and total peripheral resistance (TPR) (-0.15±0.19mmHg⋅min⋅mL−1, 95% CI -0.24 to -0.05) after 12 months. MPA + t-E2, in contrast, did not demonstrate significant changes from baseline in traditional haemodynamic parameters.
Conclusion
The positive changes in traditional markers were not reflected in the cardiogenic biomarker, cfPWV, which has demonstrated a higher positive predictive value for cardiovascular events than traditional measurements.
Vascular aging can now be predicted through an assessment of carotid-femoral Pulse Wave Velocity (cfPWV), a non-invasive cardiogenic biomarker of aortic stiffness [
]. PWV is determined by the time taken for the arterial pulse pressure, generated by the systolic contraction of the heart, to propagate along the arterial tree [
]. The association between increased arterial stiffness and altered coronary perfusion is postulated to be through its effects on systolic blood pressure (SBP), subsequently increasing left ventricular afterload.
] has shown brachial pulse pressure (PP) to be a strong independent determinant of recurrent cardiac events and all-cause mortality in the general population. The risk of a major clinical event is said to increase by 10-40% for every 10mmHg increase in PP [
]. However, PP measurements and augmentation index (AI) are indirect surrogate measures of arterial stiffness which are influenced by factors related to cardiac function, such as heart rate (HR), limiting their interpretation [
European Network for Non-invasive Investigation of Large Arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications.
Associations among different functional and structural arterial wall properties and their relations to traditional cardiovascular risk factors in healthy subjects: a cross-sectional study.
on behalf of the Artery Society, the European Society of Hypertension Working Group on Vascular Structure and Function and the European Network for Noninvasive Investigation of Large Arteries. Expert consensus document on the measurement of aortic stiffness in daily practice using carotid-femoral pulse wave velocity.
]. PWV, however, is affected by the age and body mass index (BMI) of the study population. The structural composition of the arterial walls changes with advancing age, contributing to the increased PWV [
The menopause has been shown to augment the age-dependent increase in arterial stiffness. Oestrogen deficiency is associated with increased concentrations of proinflammatory cytokines, which through a sequence of pathways, have been implicated in the inhibition of endothelium dependent vasodilation and nitric oxide synthesis [
Studies have shown that women diagnosed with premature ovarian insufficiency (POI) have a higher incidence of total cardiovascular disease (CVD) (hazard ratio [HR] 1.61, 95% confidence interval [CI] 1.22 to 2.12, p=0.0007) [
on behalf of the collaborators of the Dutch Multidisciplinary Guideline Development Group on Cardiovascular Risk Management after Reproductive D. Cardiovascular disease risk in women with premature ovarian insufficiency: A systematic review and meta-analysis.
European Journal of Preventive Cardiology.2016; 23: 178-186
], irrespective of the cause. The Framingham Study documented an increased risk of coronary heart disease (CHD) in postmenopausal women, particularly in the age category 40-44 years [
]. However, this risk has been shown to be two-fold higher in women diagnosed with POI (HR 2.2, 95% CI 1.0 to 4.9) compared to an early menopause (EM) (HR 1.2, 95% CI 0.7 to 2.0) [
]. A sub-analysis of the Multi-Ethnic Study of Atherosclerosis reported a 4% lower risk of heart failure for every year of increased menopausal age (HR 0.96, 95% CI 0.94 to 0.99) [
] compared the lipid profile of 90 POI patients with 198 control subjects and demonstrated a significantly higher triglyceride level (mean difference 0.17 log mmol/L, 95% CI 0.06 to 0.29) and borderline lower high-density lipoprotein (HDL) cholesterol level in women with POI [
The mainstay of management of women diagnosed with POI or an early menopause (EM) (EMPOI) is hormone replacement therapy (HRT). Most international guidance documents recommend treatment until at least the natural age of the menopause [
]. The number of trials assessing the impact of HRT on cardiovascular events in women diagnosed with EMPOI are, however, limited. Kalantaridou et al., (2004) [
] assessed endothelial function in women with POI before and after 6 months of HRT (oral 0.625mg conjugated equine estrogen [CEE] with cyclical medroxyprogesterone acetate [MPA] 5mg). Women receiving HRT demonstrated a more than two-fold increase in flow-mediated dilation, comparable to the control group, but the findings are limited by the small sample size (n=18 versus controls n=20) and did not reach statistical significance [
] compared the impact of HRT (transdermal oestradiol [t-E2] 100-150mcg/day with cyclical vaginal progesterone 400mg/day for two weeks every month) with the combined oral contraceptive pill (COCP) (ethinylestradiol 30µg with 1•5mg norethisterone) on cardiovascular risk markers in women with POI. They concluded that HRT significantly lowered the mean 24-hour systolic and diastolic blood pressure and had a greater benefit on renal function than the COCP. The number who completed the study, however, is also small (n=18) [
The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial analysed the impact of oestrogen (CEE 0.625mg daily) alone or in combination with one of three progestogen regimens (MPA 2.5mg/day; MPA 10mg on days 1-12; or micronised progesterone [MP] 200mg on days 1-12) on heart disease risk factors (HDL cholesterol levels; SBP; serum insulin levels; and fibrinogen levels) in postmenopausal women aged between 45-64 years. They demonstrated that unopposed oestrogen produced the greatest beneficial effects, but the high rate of endometrial hyperplasia restricted its use to women without a uterus. For women with a uterus, the cyclic use of MP produced the most favourable cardiovascular effects [
The Writing Group for the PEPI Trial Effects of Estrogen/Progestin Regimens on heart Disease Risk Factors in Postmenopausal Women: The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial.
This pilot study aimed to assess the impact of MP and MPA in combination with t-E2 on the cardiovascular risk profile of women with EMPOI, through an analysis of both traditional surrogate markers and cfPWV, identified as the gold standard cardiogenic biomarker for risk stratification of arterial disease by the European Society for Cardiology [
To compare the CVD risk of MP versus MPA in combination with t-E2 in the management of women diagnosed with EMPOI.
2.2 Primary endpoint
The main outcome measure was cfPWV.
2.3 Secondary outcome measures
The secondary endpoints were HR, SBP, diastolic blood pressure (DBP), PP, AI, cardiac output (CO), stroke volume (SV), total peripheral resistance (TPR), total cholesterol, HDL, low density lipoprotein levels (LDL), triglyceride (TG) levels and cholesterol ratio (total cholesterol/HDL ratio).
2.4 Study design
Women under the age of 45-years diagnosed with EMPOI, with an intact uterus, were prospectively invited and recruited from two tertiary referral Reproductive Endocrine and Menopause Clinics between April 2013 and August 2015. Subjects fulfilling the inclusion and exclusion criteria (Table 1) were randomised into one of two treatment arms using a web-based computer randomisation software, Graph Pad. Both groups were prescribed 50mcg/day of t-E2 in the form of Evorel® Patches in conjunction with either cyclical MP (Utrogestan®) 200mg orally on days 15-26 of a 28-day cycle or, MPA (Provera®) 10mg orally on days 16-26 of a 28-day cycle. The medication was prescribed and administered in accordance with the Summary of Product Characteristics of each of the medications. The total duration of the study period was 12-months. The study was open labelled where neither the participant nor researcher were blinded to the randomisation secondary to limited funds.
Table 1Inclusion and exclusion criteria.
Inclusion Criteria
Exclusion Criteria
Females aged between 18 and up to 45-years of age
Age <18 or >45-years of age
Confirmed diagnosis of POI or EM
Pregnant or lactating females
Willingness to participate
Contraindication to the use of hormonal preparations
No concomitant co-morbidities that would contraindicate the use of hormonal preparations
Factors present in the medical history: that would contribute to an increased risk of CVD known thrombophilia known porphyria known liver disease known past or suspected breast cancer undiagnosed vaginal bleeding genital tract carcinoma
Smokers
Body mass index >35Kg/m2
The use of concomitant medications that could influence the results, such as anti-hypertensives
Known hypersensitivity to any of the active substances or excipients contained within Utrogestan®, Provera® or Evorel patches®; known allergy to peanuts or soya
Key: POI – premature ovarian insufficiency; EM – early menopause; CVD – cardiovascular disease.
Prior to randomisation, participants previously prescribed hormonal treatment were asked to stop it for a minimum of four weeks, designated a washout period, to enable a baseline assessment to be undertaken.
Measurements were undertaken at baseline and repeated at intervals of 3-, 6-, and 12-months.
2.5 Carotid-Femoral Pulse Wave Velocity
Carotid-femoral PWV was measured using an oscillometric technique to acquire the pulse waveform (Vicorder device, Skidmore Medical). Patients were placed into the supine position at approximately 450, in room temperature, and allowed to rest and re-climatise for 5-10 mins prior to assessment. A 10cm-wide cuff was placed over the upper thigh, in the region of the femoral pulse. Simultaneously, a 3cm-wide cuff was placed around the neck for the carotid pulse. The distance between the suprasternal notch and mid upper thigh cuff was measured and utilised in the assessment, as per the manufacturer's instructions. The cuffs were simultaneously inflated, and the pulse waveforms recorded. Measurements were made in duplicate/triplicate, and the mean values used for analysis.
2.6 Sample collection
Venous blood was collected for total cholesterol, HDL, LDL, and triglyceride levels. Blood was collected into a serum separating vacutainer with clot activator (BD Diagnostics, Plymouth, UK). All laboratory measurements were carried out at King's College Hospital, Haematology and Clinical Biochemistry Laboratories.
2.7 Statistical analysis and sample size
Univariate analysis was performed to compare age, BMI, and ethnicity between the two groups. Continuous variables were compared using paired t-test between visits and baseline for each group. Multiple logistic regression was undertaken to compare continuous variables adjusting for age, BMI, and ethnicity. A p-value of <0.05 was considered statistically significant. All analyses were performed using IBM SPSS Statistics version 22.0 (Statistical Package for Social Sciences, Chicago, USA).
We aimed to recruit 90 women in total to allow for a 10% drop out and loss to follow up. The sample size was calculated using the Altman nomogram [
] and based on changes in the AI in response to exposure to the two different progestogens assessed in the study. Baseline data were obtained from a reference population, Pulse Wave Analysis (PWA) and AI data as reported by McEniery et al., (2005) [
]. Based on the reported findings, we considered a change of 8% with Standard Deviation of 12 to be a clinically significant difference to detect. Using the Altman nomogram, this would give a standardised difference of 0.80. A sample of 80 women in two groups would detect a standardised difference of 0.80 with 80% power at the 5% level of significance.
3. Results
3.1 Patient characteristics
Seventy-one participants consented to the study. Five were excluded as they did not meet the study inclusion criteria (Table 1) (diabetes [n=1]; hereditary thrombophilia [n=3]; spontaneous pregnancy [n=1]). Fig. 1 outlines the study algorithm.
Fig. 1Study algorithm. *Side effects: no adverse effects were reported to the MHRA; **Loss to follow up: unable to contact the patient for a visit. The differences in numbers between the PWV and lipoprotein arm is secondary to either missing data or patients withdrawing consent at different points.
The baseline demographic characteristics and haemodynamic parameters for the two study populations did not differ (Table 2). The age of the participants ranged between 19 years to 44 years of age. Of these, 43 of the participants were ≤40 years of age (POI), and 23 were >40 years of age (EM).
Table 2Baseline demographic characteristics and haemodynamic parameters for the two study populations.
Category
PWV cohort
Lipoprotein profile cohort
MP + t-E2
MPA + t-E2
P-value
MP + t-E2
MPA + t-E2
P-value
Age (years)
35.75±6.52
37.97±6.02
0.53
35.75±6.52
37.97±6.02
0.17
BMI (Kg/m2)
25.23±5.17
24.90±4.21
0.79
25.23±5.17
24.90±4.21
0.79
Ethnicity
0.30
0.83
Asian
2
7
2
7
Black
11
7
12
7
White
13
13
14
13
Other
2
2
4
2
HR (bpm)
68.38±10.04
69.89±10.34
0.58
SBP (mmHg)
122.67±12.32
124.05±12.61
0.68
DBP (mmHg)
69.02±8.32
69.58±7.18
0.79
PP (mmHg)
53.69±7.33
54.96±8.57
0.55
AI (%)
20.73±7.35
19.87±5.62
0.63
CO (mL/min)
5.87±1.03
6.02±1.13
0.61
SV (ml)
89.03±12.48
89.62±16.98
0.88
TPR (mmHg⋅min⋅mL−1)
0.98±0.18
0.96±0.15
0.63
PWV (m/s)
6.39±1.72
6.39±1.17
1.00
Total cholesterol (mmol/l)
4.65±0.67
4.77±0.79
0.09
HDL (mmol/l)
1.74±0.42
1.54±0.40
0.40
LDL (mmol/l)
2.47±0.58
2.66±0.63
0.20
TG (mmol/l)
0.98±0.45
1.26±0.78
0.35
Cholesterol ratio (mmol/l)
2.79±0.67
3.31±1.05
0.55
Key: Data expressed as mean±standard deviation [SD]; t-E2 – transdermal oestradiol; MP – micronised progesterone; MPA – medroxyprogesterone acetate; BMI – body mass index; HR – heart rate; SBP – systolic blood pressure; DBP – diastolic blood pressure; PP – Pulse pressure (difference between the systolic and diastolic blood pressure); AI – Augmentation index (quantifies the extent of augmented pressure relative to the central pulse pressure, and is the difference between the second and first systolic peaks expressed as a percentage of the pulse pressure); CO – cardiac output; SV – stroke volume; TPR – total peripheral resistance; PWV – pulse wave velocity (speed of travel of the pulse along a specified arterial segment); HDL – high density lipoprotein; LDL – low density lipoprotein; TG – triglyceride; cholesterol ratio – total cholesterol/HDL.
PWV did not demonstrate significant changes from baseline for the duration of the study, in either treatment arm (Table 3).
Table 3Impact of MP and MPA in combination with t-E2 on the haemodynamic variables after 3-, 6- and 12-months, presented as mean difference (±SD), with 95% confidence intervals.
Variables
MP + t-E2
MPA + t-E2
Mean difference after 3-months (95% CI)
Mean difference after 6-months (95% CI)
Mean difference after 12-months (95% CI)
Mean difference after 3-months (95% CI)
Mean difference after 6-months (95% CI)
Mean difference after 12-months (95% CI)
HR
1.16±12.88 (-4.56 to 6.87)
7.69±12.80 (1.10 to 14.27)*
4.71±13.97 (-2.24 to 11.65)
2.89±12.67 (-2.72 to 8.51)
1.99±11.49 (-3.55 to 7.53)
4.03±18.07 (-5.98 to 14.03)
SBP
-0.65±10.20 (-5.18 to 3.87)
-3.61±7.39 (-7.41 to 0.19)
-1.69±7.78 (-5.56 to 2.18)
-2.14±11.64 (-7.30 to 3.02)
-2.42±9.78 (-7.13 to 2.30)
1.18±8.95 (-3.77 to 6.14)
DBP
-1.75±6.57 (-4.66 to 1.17)
-2.94±5.60 (-5.81 to -0.06)
-3.43±6.31 (-6.57 to -0.29)*
-2.08±7.18 (-5.26 to 1.10)
-2.12±6.86 (-5.42 to 1.19)
0.97±8.40 (-3.69 to 5.62)
PP
1.32±9.85 (-3.05 to 5.68)
-1.29±6.50 (-4.63 to 2.05)
1.83±4.69 (-0.50 to 4.17)
-0.32±8.63 (-4.15 to 3.51)
-0.42±7.98 (-4.27 to 3.43)
-0.80±6.43 (-4.36 to 2.76)
AI
-1.16±10.48 (-5.81 to 3.48)
-1.16±10.48 (-5.81 to 3.48)
2.14±6.52 (-1.21 to 5.49)
1.19±5.83 (-1.39 to 3.78)
0.73±6.12 (-2.21 to 3.68)
2.49±5.60 (-0.61 to 5.59)
CO
0.46±1.37 (-0.15 to 1.06)
0.34±0.95 (-0.14 to 0.83)
0.71± 1.01 (0.20 to 1.21)*
-0.07±1.37 (-0.68 to 0.54)
0.03±1.18 (-0.54 to 0.60)
-0.06±1.29 (-0.77 to 0.66)
SV
3.31±18.76 (-5.01 to 11.62)
-3.51±14.63 (-11.03 to 4.02)
3.54±12.07 (-2.47 to 9.54)
-3.10±16.07 (-10.22 to 4.03)
-0.68±16.69 (-8.73 to 7.36)
0.20±11.38 (-6.10 to 6.50)
TPR
-0.08±0.21 (-0.18 to 0.02)
-0.07±0.16 (-0.15 to 0.01)
-0.15±0.19 (-0.24 to -0.05)*
-0.02±0.18 (-0.10 to 0.06)
0.00±0.22 (-0.11 to 0.11)
0.04±0.24 (-0.09 to 0.18)
PWV
0.40±3.32 (-1.07 to 1.87)
0.22±1.38 (-0.49 to 0.93)
0.28±0.87 (-0.15 to 0.71)
-0.02±1.04 (-0.48 to 0.44)
-0.07±1.05 (-0.58 to 0.43)
0.64±1.62 (-0.26 to 1.54)
Total cholesterol
-0.16±0.45 (-0.37 to 0.06)
-0.13±0.50 (-0.37 to 0.11)
0.02±0.47 (-0.23 to 0.26)
-0.28±0.55 (-0.52 to -0.04)*
-0.21±0.51 (-0.46 to 0.04)
-0.23±0.51 (-0.51 to 0.06)
HDL
-0.07±0.17 (-0.15 to 0.01)
-0.03±0.21 (-0.13 to 0.07)
-0.09±0.20 (-0.20 to 0.01)
-0.12±0.46 (-0.32 to 0.08)*
-0.16±0.27 (-0.29 to -0.02)*
-0.11±0.21 (-0.22 to 0.01)
LDL
-0.04±0.35 (-0.20 to 0.13)
-0.01±0.35 (-0.17 to 0.16)
0.09±0.31 (-0.07 to 0.26)
-0.12±0.46 (-0.32 to 0.08)
-0.11±0.45 (-0.34 to 0.13)
-0.02±0.45 (-0.27 to 0.23)
TG
-0.12±0.49 (-0.34 to 0.11)
-0.16±0.59 (-0.45 to 0.12)
0.06±0.48 (-0.18 to 0.31)
-0.13±0.54 (-0.37 to 0.10)
0.17±0.73 (-0.19 to 0.53)
-0.21±0.37 (-0.42 to -0.01)*
Cholesterol ratio
0.03±0.30 (-0.11 to 0.18)
0.03±0.37 (-0.16 to 0.21)
0.18±0.30 (0.02 to 0.33)*
0.07±0.46 (-0.14 to 0.27)
0.25±0.67 (-0.10 to 0.59)
0.05±0.38 (-0.17 to 0.27)
Key: Data expressed as mean±standard deviation [SD]; t-E2 – transdermal oestradiol; MP – micronised progesterone; MPA – medroxyprogesterone acetate; HR – heart rate (bpm); SBP – systolic blood pressure (mmHg); DBP – diastolic blood pressure (mmHg); PP – Pulse pressure (mmHg); AI – Augmentation index (%); CO – cardiac output (mL/min); SV – stroke volume (ml); TPR – total peripheral resistance (mmHg⋅min⋅mL−1); PWV – pulse wave velocity (m/s); HDL – high density lipoprotein (mmol/l); LDL – low density lipoprotein (mmol/l); TG – triglyceride (mmol/l); cholesterol ratio – total cholesterol/HDL. *p<0.05.
MP + t-E2 demonstrated a positive effect on traditional CVD markers, with a significant improvement seen in CO (0.71±1.01mL/min, 95% CI 0.20 to 1.21, p=0.01) and reduction in DBP (-3.43±6.31mmHg, 95% Cl -6.57 to -0.29, p=0.03) and TPR (-0.15±0.19mmHg⋅min⋅mL−1, 95% CI -0.24 to -0.05, p=0.01) after 12-months duration compared to baseline. A significant increase in cholesterol ratio was seen after 12-months duration (0.18±0.30mmol/l, 95% CI 0.02 to 0.33, p=0.03) during the same study period.
MPA + t-E2 in contrast, did not demonstrate significant changes from baseline in HR, SBP, PP, AI, CO or TPR. Total cholesterol levels, however, were initially lowered after 3-months duration (-0.28±0.55mmol/l, 95% Cl -0.52 to -0.04, p=0.02). HDL levels demonstrated a significant decline after 3-months (-0.13±0.24mmol/l, 95% CI -0.23 to -0.02, p=0.02) and 6-months (-0.16±0.27mmol/l, 95% CI -0.29 to -0.02, p=0.03) duration from baseline. This difference was not maintained by 12-months duration (-0.11±0.21mmol/l, 95% CI -0.22 to 0.01, p=0.07). After 12-months duration, however, significant reductions were seen in the TG levels (-0.21±0.37mmol/l, 95% CI -0.42 to -0.01, p=0.04).
4. Discussion
MP + t-E2 resulted in a significant improvement in CO and reduction in DBP and TPR after 12-months of treatment. PWV, however, did not significantly change from baseline in either treatment arm over the course of the study, consistent with the neutral changes observed in the AI and PP readings over the same period.
No significant changes were observed in the lipoprotein profile of the MP + t-E2 treatment arm, in keeping with MP having a more selective effect on progesterone receptors [
]. Furthermore, no significant changes were demonstrated in the LDL levels, the serum levels of which have been closely correlated to the development of coronary artery disease, supporting the notion of a neutral impact of MP on CVD risk [
MPA + t-E2, in contrast, resulted in changes in the lipoprotein profile that included a reduction in total cholesterol levels at 3-months of treatment, HDL levels at 6-months and TG levels after 12-months of treatment. The TG lowering effect seen with MPA + t-E2 may have occurred secondary to its antagonistic effect on oestrogen stimulated hepatic TG synthesis or due to increased activity of the lipoprotein lipase enzyme [
]. MPA + t-E2, however, did not demonstrate significant changes in any of the other measured traditional haemodynamic parameters, including CO and TPR.
The findings overall, showed that MP and MPA given in combination with t-E2 did not adversely impact CVD risk markers as assessed by the gold standard cardiogenic biomarker, cfPWV when used in the management of women with EMPOI. Furthermore, MP + t-E2 resulted in a significant improvement in CO and reduction in DBP and TPR.
Central arterial stiffness has been shown to be an independent predictor of cardiovascular risk and all-cause mortality [
]. Invasive catheterisation of the ascending aorta remains the gold standard technique for central haemodynamic assessment, but the invasive nature of the technique has limited its clinical applicability. The analysis of peripheral waveforms with dedicated devices has provided the opportunity to assess central haemodynamic parameters non-invasively and has shown good correlation with direct central blood pressure measurements obtained during invasive catheterisation of the ascending aorta [
] demonstrated a strong correlation between non-invasive devices measuring cfPWV (Complior Analyse, PulsePen ET, PulsePen ETT, and SphygmoCor) and invasive aortic PWV measurements (r>0.83) [
]. The Vicorder system (Skidmore Medical Limited) used in this study has been calibrated to non-invasively record mean aortic and diastolic pressures and has also been validated against other devices used in this context [
]. Firstly, cfPWV assessments do not incorporate the ascending aorta in the path of travel. Secondly, cfPWV measurements include arterial segments in which the path of travel of the pulse can be in an opposite direction (brachiocephalic trunk and common carotid artery) to the thoracic aorta. Thirdly, cfPWV measurements include segments of the femoral artery in their evaluation. The muscular component of the femoral artery is greater than that of the aorta. This difference between muscular and elastic arteries can increase the PWV assessment in younger individuals. This difference, however, is reversed with advancing age, thus, cfPWV measurements may be overestimated in younger adults and underestimated in older individuals. Furthermore, PWV exponentially increases in aortic arteries with increasing age, but only weakly and linearly increases with age within the muscular arteries of the lower limbs [
Overall, cfPWV is deemed to be a highly reproducible, non-invasive emerging cardiogenic biomarker of arterial stiffness and thus, cardiovascular risk stratification [
European Network for Non-invasive Investigation of Large Arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications.
], however, its clinical application is limited by population-based differences seen within the vascular behaviour in varying physiological and pathological conditions [
Pulse Wave Velocity as Marker of Preclinical Arterial Disease: Reference Levels in a Uruguayan Population Considering Wave Detection Algorithms, Path Lengths, Aging, and Blood Pressure.
The European Society of Hypertension have stated that a PWV of 12m/s can signify vascular damage and/or cardiovascular risk. The baseline PWV for both treatment arms and the change from baseline throughout the study duration was approximately 50% lower than the single cut-off value used by the European Society of Hypertension. This single value, however, cannot be extrapolated to all populations, as it can underestimate the level of risk in young subjects and overestimate the level of risk in older individuals [
Pulse Wave Velocity as Marker of Preclinical Arterial Disease: Reference Levels in a Uruguayan Population Considering Wave Detection Algorithms, Path Lengths, Aging, and Blood Pressure.
]. More directed ranges accounting for age and ethnicity are therefore needed for clinical application of this cardiogenic biomarker in everyday practice [
] concluded that a 1m/s increase in PWV is associated with a 14% (95% CI 9 to 20%), 15% (95% CI 9 to 21%) and 15% (95% CI 9 to 21%) increased risk of total cardiovascular events, cardiovascular mortality, and all-cause mortality, respectively when controlled for risk factors [
]. In this study, the PWV for both treatment arms changed by <1m/s from baseline, consistent with no significant changes observed in this parameter throughout the study.
] conducted a meta-analysis of 11 longitudinal studies including 5,648 subjects with a mean follow up duration of 45-months. The studies included both men and women of broad age categories, the youngest of which were 42.6±11.2 years. They found that the relative risk of cardiovascular events increased by 8.8% (n=3285; 95% CI 1.04 to 1.14) with a 10mmHg increase in central systolic pressure, 13.7% (n=4778; 95% CI 1.06 to 1.22) with a 10mmHg increase in central PP and 31.8% (n=1326; 95% CI 1.09 to 1.59) with a 10% absolute increase in central AI. A 10% increase in central AI was also correlated to a 38.4% (n=569; 95% CI 1.19 to 1.61) increased risk of all-cause mortality [
]. This study did not demonstrate significant changes in PP, AI or SBP, in either treatment arm from baseline, with absolute levels differing by <10mmHg and <10% change in AI at 12-months, consistent with the neutral changes observed in the PWV readings over the same study period.
Hypertension, overall, is an important risk predictor for CVD. Current practice estimates a greater prediction of all cardiovascular events with raised SBP, but both systolic and diastolic hypertension being independent predictors of adverse cardiovascular outcomes [
]. This study did not identify changes in the SBP over the course of the research, consistent with no changes observed in SV and AI. The DBP, however, did demonstrate a significant reduction after 12-months duration (-3.43±6.31mmHg, 95% Cl -6.57 to -0.29, p=0.03) in the MP + t-E2 treatment arm. This could reflect the younger cohort included in this study and support the more favourable impact of MP in combination with t-E2 when traditional surrogate markers of CVD risk are assessed.
The HR increased by 4-5 beats per minute after 12-months duration from baseline in both treatment arms, with a significant increase recorded in the MP + t-E2 (7.69±12.80bpm, 95% CI 1.10 to 14.27, p=0.03) treatment arm at 6-months duration. Cardiac function, such as HR, has been shown to influence the parameters PP and AI [
European Network for Non-invasive Investigation of Large Arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications.
], with this difference potentially providing an explanation for the fluctuations seen in the PP and AI readings over the course of the study.
Epidemiological studies, such as the Framingham cohort studies in the USA, have identified several CVD risk factors including the lipid profile. Risk stratification models to estimate a patient's 10-year risk of developing CVD incorporate lipid parameters in their equation. Both the Framingham-based equations [
] consider total cholesterol, HDL cholesterol and the ratio of total cholesterol to HDL to be the strongest predictors. The Framingham Offspring study followed a cohort of individuals for 20-years and identified that any combination of low levels of HDL cholesterol, high levels of LDL cholesterol and high levels of TG were associated with an increased risk of CVD [
]. The National Clinical Guideline Centre for Cardiovascular Risk Assessment for primary prevention of CVD, however, recommend using the QRISK2 risk assessment tool to assess cardiovascular risk for the primary prevention of CVD in people aged ≤84 years which incorporates the cholesterol ratio, total cholesterol/HDL in its calculation [
LDL is the predominant cholesterol-carrying lipoprotein and main atherogenic lipoprotein. Epidemiological data has suggested that isolated low HDL levels is a strong and independent risk factor for CVD [
]. HDL particles may act as a protective factor against atherosclerosis through several biological mechanisms including the prevention of endothelial activation, inflammation, and oxidative stress, as well as enhancing nitric oxide production and promoting cholesterol efflux to reduce lesion formation and maintain barrier integrity [
Linton MF, Yancey PG, Davies SS, Jerome WG, Linton EF, Song WL, Doran AC and Vickers KC. The Role of Lipids and Lipoproteins in Atherosclerosis. www.endotext.org Last updated 03.01.2019. Last accessed 04.12.2020.
A subsidiary analysis of a clinical trial in Australia, New Zealand, and Finland, looking at the impact of Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID study), (n=9014 aged between 31-75 years with recorded acute coronary syndrome), acknowledged that all lipid parameters were linked with future coronary events. They concluded that the total cholesterol/HDL ratio, LDL/HDL ratio or apolipoprotein B/apolipoprotein A1 ratio were more superior in predicting the time to coronary event than a single lipid measure [
The ESCARVAL Study Group found that HDL levels, total cholesterol/HDL ratios and TG/HDL ratios were better predictors for mortality and CVD than other lipid parameters commonly used in clinical practice [
]. They demonstrated a positive association with the TG/HDL ratio and an inverse association with total cholesterol, HDL, and LDL on age-adjusted mortality rates (deaths/10,000 person-years) [
]. The subfractions of HDL were not measured as they have not shown an added benefit in the identification of persons at risk above the measurement of standard HDL.
The main strength of the study is in its’ aim to bridge the gap in knowledge regarding the impact of HRT on surrogate markers of CVD through an analysis of both traditional markers and cfPWV in this cohort of women. The study, despite highlighting several findings is not without its limitations.
The main caution that needs to be considered when interpreting non-invasively measured cfPWV readings is the calculation of path length between the carotid and femoral sites which can significantly influence PWV, with differences of up to 30% being cited. The distance is commonly measured by one of four different methodologies: (i) the direct distance between the carotid and femoral sites; (ii) the distance between the sternal notch and femoral sites; (iii) the subtracted distance between the carotid and sternal notch from the total distance; or (iv) the subtracted distance between the carotid and sternal notch from the sternal notch and femoral site [
]. The expert consensus document on the measurement of aortic stiffness recommended using 80% of the direct distance between the common carotid artery and common femoral artery [
on behalf of the Artery Society, the European Society of Hypertension Working Group on Vascular Structure and Function and the European Network for Noninvasive Investigation of Large Arteries. Expert consensus document on the measurement of aortic stiffness in daily practice using carotid-femoral pulse wave velocity.
]. This study, however, utilised the distance between the sternal notch and femoral site to calculate the distance between the two points as per the manufacturer's instructions for the device, which is largely dependent on body habitus, potentially introducing an error in the PWV estimation.
Furthermore, the PWV measurements were undertaken at different points within the cyclical hormonal medication phases of oestrogen only or oestrogen and progesterone combined. Each visit occurred after a set period plus or minus two weeks to allow flexibility for the trial subjects. Hormonal influences can occur directly on the arterial wall physiology, affecting blood pressure and vascular reactivity through oestrogen driven activation of endothelial nitric oxide synthase activity. Adkisson et al., (2010) [
] postulated cyclic variations within the haemodynamic parameters, with lower central and systemic blood pressure (approximately 4mmHg) readings in the late follicular and early luteal phases of the menstrual cycle mirrored by a proportionate increase in oestrogen levels at this time, and a subsequent increase in the bioavailability of nitric oxide [
]. The progestogen component may antagonise the oestrogen mediated responses.
The main limitation is the small sample size which did not achieve the intended power calculation. The number of recruits who voluntarily decided to leave the study or were lost to follow up, however, appears consistent across the two treatment arms; this, however, does not detract from needing larger studies to demonstrate significance. Furthermore, the loss of subjects is unlikely to be solely attributable to the widely reported progestogen related adverse effects as loss is seen across both treatment arms. The largest fall in numbers occurred after the initial 3-month period. This needs to be considered when interpreting the findings.
The open-label design means that neither the participants nor the investigators were blinded to the study arms. This was due to the limitation in funding that restricted the option of packaging the medication identically.
5. Conclusion
MP combined with t-E2 was shown to have a more favourable effect on traditional surrogate markers of CVD risk in women with EMPOI, with changes observed in CO, DBP and TPR, when compared to MPA + t-E2. No significant differences, however, were noted when the cardiogenic biomarker, cfPWV, was assessed. Further studies are needed to establish reference values for cfPWV to help guide assessment of CVD risk and evaluate the response to hormone replacement in women with EMPOI.
Contributors
Monica Mittal conducted the study and analysed the data.
Carmel McEniery checked the manuscript for intellectual content.
Prasanna Raj Supramaniam analysed the data and checked the manuscript for intellectual content.
Linda Cardozo checked the manuscript for intellectual content.
Mike Savvas checked the manuscript for intellectual content.
Nick Panay checked the manuscript for intellectual content.
Haitham Hamoda formulated the hypothesis, oversaw the manuscript, and checked the manuscript for intellectual content.
All authors approved the final submission.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Ethical approval
Study approval was obtained by the Research and Development Department at King's College Hospital and ethical approval was granted by the London and GTAC Ethics Committee (REC Number: 12/LO/1957; EudraCT Number: 2012-004511-30 https://www.clinicaltrialsregister.eu/ctr-search/search?query=2012-004511-30) on the 16th January 2013. All patients gave informed written consent prior to data collection.
Provenance and peer review
This article was not commissioned and was externally peer reviewed.
Research data (data sharing and collaboration)
There are no linked research data sets for this paper. Data will be made available on request.
Declaration of competing interest
The authors declare that they have no competing interests.
Acknowledgements
I would like to extend my gratitude to all respondents, without whose cooperation I would not have been able to conduct this study.
European Network for Non-invasive Investigation of Large Arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications.
Associations among different functional and structural arterial wall properties and their relations to traditional cardiovascular risk factors in healthy subjects: a cross-sectional study.
on behalf of the Artery Society, the European Society of Hypertension Working Group on Vascular Structure and Function and the European Network for Noninvasive Investigation of Large Arteries. Expert consensus document on the measurement of aortic stiffness in daily practice using carotid-femoral pulse wave velocity.
on behalf of the collaborators of the Dutch Multidisciplinary Guideline Development Group on Cardiovascular Risk Management after Reproductive D. Cardiovascular disease risk in women with premature ovarian insufficiency: A systematic review and meta-analysis.
European Journal of Preventive Cardiology.2016; 23: 178-186
Effects of Estrogen/Progestin Regimens on heart Disease Risk Factors in Postmenopausal Women: The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial.
Pulse Wave Velocity as Marker of Preclinical Arterial Disease: Reference Levels in a Uruguayan Population Considering Wave Detection Algorithms, Path Lengths, Aging, and Blood Pressure.
Linton MF, Yancey PG, Davies SS, Jerome WG, Linton EF, Song WL, Doran AC and Vickers KC. The Role of Lipids and Lipoproteins in Atherosclerosis. www.endotext.org Last updated 03.01.2019. Last accessed 04.12.2020.