The main finding of this study was that Cdis declined 11% on an average while estradiol levels increased from 0.1 nmol/l to 6.2 nmol/l (mean) concentration (62-fold) in their IVF cycles. Correlation analyses showed that this decline was not due to the higher diameter of the carotid artery, high blood pressure or dyslipidemia. We conclude, that the estradiol level is likely to contribute to the function of carotid artery and blood pressure and despite of these hormonal changes, the stimulation protocol is not hazardous to women’s vasculature. All the changes in Cdis are variations of normal.
Estrogen is likely to play a role in arterial distensibility, but the available evidence suggests that its effect depends on the baseline estrogen level and magnitude of the estrogen concentration changes. In normal menstrual cycle the estrogen change is proportionately moderate (three to four-fold) from low baseline levels. This results in 10% increase in Cdis (1) . In pregnancy from the first to third trimester the estrogen change is from high baseline to even ten-fold higher concentration in the third trimester, which in turn decreased carotid artery elasticity by one third (36%)(3) (4) . In menopause, the change is from low to very low estrogen and HRT rises it close to premenopausal level, with an improvement of 30 to 40% in Cdis with replaced estrogen(6) .
In the current study setting estrogen changes are exaggerated compared to natural menstrual cycle, since levels at the ovulatory peak are up to 60-fold higher than during suppression and up to 10-fold higher than in natural cycle. Interestingly, in this setting carotid artery elasticity declined, implying that an IVF cycle is very different from natural cycle in terms of carotid artery elasticity. This phenomenon, however, is very similar that seen in course of normal pregnancy. It may be speculated that beyond certain estrogen levels, its effect is opposite to those seen at low or moderate levels. It has been previously speculated, that this phenomenon can be adaptation to the altered hemodynamics occurring during pregnancy(3) and also, that there are less estrogen receptors occurring in the carotid arteries and this could explain the difference in Cdis compared to other arteries(9) .
It has been published earlier(10), that there is a complex relation of estrogen with arterial health, higher estradiol was associated with smaller Cdia, but lower Cdis. That study was made with postmenopausal women, but the findings are in concordance with ours. However, among published studies the results have been controversial. (11). There are also studies made with postmenopausal women with HRT and its effect to the Cdis. One of them (6) discovered, that carotid arterial compliance and Cdis were significantly higher and stiffness index values lower with HRT compared to placebo treatment. Two other studies had similar findings (12) (13). A study made with perimenopausal women (14) and a study made with postmenopausal women(15) found out, that there was no effect on Cdis with HRT versus placebo using.
The clinical implications of the present results can be seen via stimulation protocol’s impact to vascular health. The huge variation in estradiol levels during a long agonist IVF treatment gives an opportunity to study this phenomenon and effect to vasculature in vivo. During the stimulation, the exposure for high estradiol is after all quite short. In general, low Cdis is associated with greater stroke risk(16) . In our study, Cdis lowered during the highest estradiol level. Still, the Cdis values stayed within the normal limits (17) and Cdis lowered only -0.26 SD (Figure 1). The change of Cdis can be detected between visits 2 and 3, but this finding seems to be only modest and is likely to be clinically less significant. We speculate that if there are plenty of other risk factors for stroke, then this phenomenon could additionally have unfavorable effect. Furthermore, although in short term the values of Cdis do not differ much from baseline values, hormonal status may contribute to Cdis and this might have consequences to vascular health in long term. This study shows, that the stimulation protocol is not hazardous to vasculature, if there are no earlier cardiovascular problems. Further studies are needed to clarify more this phenomenon. It has been described earlier, that there will be regional heterogeneity in elastic behavior of the arterial tree with aging and hypertension, there will be vascular remodeling (18) ). It has also been shown, that carotid artery stiffens during pregnancy(4). This finding is in line with ours.
The widening of the arteries is related to a worse cardiovascular profile, active coronary plague volume and cardiovascular events(19) . In our study, Cdia was highest during the low estrogen level and then diminished during the high estrogen level, and this could be considered as a beneficial finding related to estrogen when considering the cardiovascular health. It has been reported (20), that arterial lumen widens with age in normotensive postmenopausal women and that might compensate age-related increase in arterial stiffness. The mechanism may be different in young women. (21), that menopause may increase aortic stiffness, which probably causes the rise in systolic blood pressure and this may lead to a slight dilatation of common carotid artery. They suggested that aortic elasticity diminishes due to menopause and aging.
Many studies have shown(6)(22) (23) (24) , that estrogen deprivation may increase blood pressure. Also in our study, blood pressure was at its highest during the low estradiol level and lowered during the high estradiol level, which could be considered as an advantageous finding, even when the change in blood pressure was only 3 mmHg between visits 2 and 3 (Table I). That probably does not have that much clinical significance, but a trend can be seen. Conversely it has been shown, that brachial arterial blood pressure did not change significantly during the menstrual cycle in young women(1).
In our study, lipid profile was at its worst during the low estradiol level in visit 2 and then got better during the high estradiol level, so total cholesterol and LDL were higher during the low estradiol level and then improved during the IVF stimulation. During such a short period this phenomenon does barely have any clinical significance, but there was an association between Cdis and lipids. Diet barely has an impact to those findings while this short period of observation.
The validity of this study can be considered good since all women underwent the same IVF protocol in one unit and they served as their own controls. Laboratory tests were analyzed blindly and the Cdis and Cdia were measured without the information of IVF cycle phase. The cause of underlying infertility unlikely plays a role in arterial elasticity as all women went through the same treatment protocol. One limitation of this study was a small study population. Limited statistical power in analyses can affect our results, and when interpreting the results of this study, a possibility for type two error should be taken into consideration. Considering the methods, Cdis-assessment is likely to be sensitive, because we could detect several previously known and clinically relevant associations between Cdis and systolic blood pressure, diastolic blood pressure, dyslipidemia and obesity.