Vitamin D Deciency in Premenopause: Endothelial and Diastolic Functions With Hyperparathyroidism

Introduction: Vitamin D deciency(VDD) is a common condition in all ages, especially in winter, and causes several adverse outcomes, including cardiovascular diseases. Although there is conicting information about the causes of cardiac events in VDD, the commonly accepted cause is a pathogenic relationship with hyperparathyroidism. Aim: We aim to demonstrate the effects of hyperparathyroidism, if present, on endothelial and diastolic functions using echocardiography, carotid and brachial tissue Doppler imaging(tDi) in premenopausal women with VDD independent from other confounding risk factors. Methods: Our study is a cross-sectional, observational study investigating premenopausal women aged 18-50 who applied to internal medicine. The patients were divided into two groups according to their parathyroid hormone(PTH) levels. While PTH levels are within the normal range in 41 patients, it was high in 29 patients. Study groups are analyzed for basal characteristics, standard echocardiographic evaluation, carotid, and brachial artery tDi features. Results: There were no signicant differences in age, height, weight, body mass index(BMI), and diastolic blood pressures(BP) between the groups(all p values>0.05). Systolic BPs were signicantly higher in the secondary hyperparathyroidism, no patients diagnosed with hypertension(Mean systolic BP 124±8, p=0.020). The rate of severe VDD was 78%. In terms of diastolic parameters, prolongation of A wave durations and decreased E/A ratio were observed in the secondary hyperparathyroidism(p values are 0.043 and 0.031, respectively). Carotid and brachial tDi; carotid intima-media thickness(IMT) is signicantly higher in secondary hyperparathyroidism(p=0.032). Conclusion: We showed that high PTH levels have adverse effects on diastolic functions(A wave, E/A ratio) and endothelial functions(carotid IMT) in premenopausal women with VDD, even in the absence of other confounding risk factors. We can state that our study's ndings will contribute to the literature and are predictive for future studies.


Introduction
Vitamin D(VitD) is an important steroid hormone in bone and mineral physiology with effects on calcium and phosphorus metabolism (1). By binding to VitD receptors in the whole body, including endothelium, vascular smooth muscle cells, and cardiomyocytes, it also affects many vital functions such as homeostasis, autoimmunity, and in ammatory interleukins synthesis, cell proliferation and differentiation, and blood pressure regulation (2)(3)(4)(5)(6)(7). The prevalence of VDD can be at 30-50% (8,9). The main risk factors are high altitude, winter season, low sun exposure, restricted dietary intake, postmenopause, and advanced age (9,10).
Although its pathophysiology is not fully known, VDD has been linked with various cardiovascular outcomes; hypertension, myocardial infarction complications, arterial stiffness, myocardial brosis, systolic and diastolic dysfunction (1,(11)(12)(13)(14). Hyperstimulation of the renin-angiotensin-aldosteron system(RAAS) and sympathetic nervous system, consequently increasing systemic in ammation, uidelectrolyte balance distortions, and continuing these conditions in a vicious circle, can be considered as the underlying mechanism (12,14,15). Sunbul M. Et al. showed that basal VitD values might adversely associated with left ventricular basal global longitudinal strain values. VitD therapy may have positive effects on myocardial deformation (16).
Secondary hyperparathyroidism can be seen at rates 18-25% in VDD, which increases with age (17).
Secondary hyperparathyroidism may be associated with disease severity in congestive heart failure, especially in older, due to bone loss and osteoporosis (18). There may be a relationship between PTH levels and arterial stiffness, coronary atherosclerosis, brachial ow-mediated dilatation(bFMD) (adversely), especially in primary hyperparathyroidism, and a regression in pulse wave velocity(PWV) with parathyroidectomy (19)(20)(21)(22). Although this nding could not be reached with primary hyperparathyroidism in all studies, improvements in PWV were also observed using cinacalcet; a calcimimetic agent that provides a decrease in PTH levels in chronic renal failure and secondary hyperparathyroidism (23,24).
The relationship between VDD and hyperparathyroidism with arterial stiffness has been shown one by one, as mentioned above (11,14,19,22). Pirro et al. investigated independent associations between VitD and PTH with arterial stiffness in the postmenopausal stage with normal kidney functions. In this study, an inverse relationship, albeit weak, was found between VitD levels and PWV. This relation was signi cantly associated with PTH levels regardless of existing risk factors and factors involving bone formation (25).
The risk of cardiovascular disease increases in the postmenopausal period, especially with hormonal changes (26). Menopause can be considered as a confounding risk factor. In the light of these pieces of information and taking into account the missing points, our study aims to reveal the differences that secondary hyperparathyroidism will create on endothelial and diastolic functions, if present, using echocardiography, carotid, and brachial tDi in premenopausal women with VDD.

Materials And Methods
Our study is a cross-sectional, observational study, includes premenopausal women aged 18-50 who applied to internal disease and were diagnosed with VDD. Our study was carried out in Bilecik City in the period November-December 2020. Bilecik, located at an approximate altitude of 500 meters and a latitude of 40°N, is a small city in Turkey. During the study period, the average temperature in Bilecik city was determined as 8-9 degrees Celsius(maximum 13°C-minimum 2°C).
The fasting routine blood samples(including PTH levels) were investigated of these patients' internal medicine records. In evaluating VitD, the serum circulating 25-hydroxyvitamin D[25(OH)D] level was measured. VDD was de ned as a serum 25-hydroxyvitamin D level of < 20 ng/ml, and severe VDD was also de ned as a serum 25-hydroxyvitamin D level of < 10 ng/ml (27). Patients diagnosed with VDD without replacement therapy were evaluated using essential echocardiographic evaluation, brachial and carotid tDi. Local ethics committee approval was obtained.
Patients diagnosed with diabetes mellitus, hypertension, hyperlipidemia, coronary artery disease, congestive heart failure, peripheral artery disease, moderate-severe heart valve disease, cardiomyopathies, thyroid dysfunction, chronic obstructive pulmonary disease, malignancy, rheumatological disease, active infection, kidney failure, liver disease, hormonal diseases(affecting calcium level, including primary hyperparathyroidism), obesity(BMI > 30), drug usage(including betablockers and medications affect calcium level), menopause, pregnancy, and heavy alcohol users and smokers were excluded from the study. A total of 70 patients were included in the survey by excluding patients with low echogenicity.
Patients were divided into two groups according to their PTH levels. While PTH levels were in the normal reference range(< 65 pg/ml) in 41 patients, it was high in 29 patients. Groups were compared regarding basal demographic characteristics, standard echocardiographic evaluation, carotid and brachial artery tDi features.

Image Recordings And Analysis
Standard echocardiographic measurements; left ventricular ejection fraction(EF) with modi ed Simpson method, tDi measurements, and cardiac valve evaluations performed with transthoracic echocardiography(EPIQ 7 echocardiography device(Philips, Amsterdam, Netherlands)) with reference of 2015 ASE(American Society of Echocardiography) guideline (28). In echocardiographic tDi evaluations, global systolic and diastolic ventricular function indicator, the myocardial performance index(MPI), also known as Tei Index, is also calculated with the formula '[Isovolumetric contraction time(IVCT) +

Isovolumetric relaxation time(IVRT)]/Ejection Time(ET)' for both left ventricle lateral and septal parts(29).
Carotid systolic and diastolic diameters, IMT, and bFMD performed with duplex ultrasound(Toshiba Sonolayer SSA-270 A equipped with a 7.5 Mhz linear array transducer, Toshiba Medical Systems, Japan). The distal 1 cm of each common carotid artery's far wall and plaque-free segments were used for carotid IMT measurements. The mean carotid IMT value was derived from the measurements made in both carotid arteries' locations (30).
All ultrasonic examinations were made with continuous electrocardiography in a semi-dark room. Brachial artery measurements were made by keeping the patient's left arm steady in a horizontal plane, approximately 2 cm above the elbow, in the longitudinal plane at the diastolic phase. Basal measurements were taken after 15 minutes of rest in the fasted state. The appropriate sphygmomanometer cuff located in the patient's upper arm in ated to the level of 200 mmHg and waited for 5 minutes, and recordings will be continued until 2 minutes after cuff de ation. The maximum rate of change, which is also named as peak bFMD was calculated in a percentage(%) with the formula '[(maximum diameter -basal diameter)/basal diameter]*100'(31, 32).

Statistical Analysis
Statistical analysis was performed by the SPSS version 20.0(IBM, Armonk, New York, United States). The normality of distribution was examined by using both the Shapiro-Wilk W test. Descriptive statistical methods, including percent and mean ± standard deviation(SD) or median (interquartile range[IQR]), were used to provide the basic features of the data. Mann-Whitney U test was used for non-normally distributed continuous variables for normal PTH levels group and secondary hyperparathyroidism group. A p-value less than 0.05 was accepted as signi cant.

Results
There were no signi cant differences in age, height, weight, BMI, and diastolic BP between the groups(all p values > 0.05). The secondary hyperparathyroidism group's mean systolic BP was signi cantly higher than the group without hyperparathyroidism, but not high enough to diagnose hypertension(Mean systolic BP 124 ± 8, 118 ± 12, respectively; p = 0.020) (33). There were no signi cant differences in laboratory parameters, except for vitamin B12, which were signi cantly lower in the secondary hyperparathyroidism group, but the current values were within the reference range(Mean values 335 ± 107, 426 ± 181, respectively; p = 0.019). Baseline characteristics and laboratory results of the groups are shown in table 1. VitD and PTH levels are also presented in Fig. 1.
VitD levels were relatively low in the secondary hyperparathyroidism compared to the group with normal PTH levels(mean values 8.0 ± 4.0, 9.3 ± 7.8,respectively; p = 0.376). The rate of severe VDD(< 10 ng/ml) in our study's total population was 78%.
Echocardiographic and peripheral tDi ndings are shown in table 2. Some signi cant differences related to diastolic parameters are observed, which will explain in the discussion section in detail. Other echocardiographic parameters were similar, including EF(p values > 0.05). Regarding carotid and brachial tDi, only a signi cant difference was observed according to carotid IMT(p = 0.032).

Discussion
VDD is a common condition that can be seen in all age groups worldwide, especially in winter, and causes several adverse outcomes, including cardiovascular diseases (34). The removal of inhibitory effects on the RAAS system and proin ammatory cytokines can be accepted as the primary mechanism for cardiovascular involvement in VDD (12,14,15,34). Hyperparathyroidism cases secondary to VDD are not uncommon and lead to adverse cardiac outcomes (17,35).
Menopause may be considered a risk factor for VDD and, consequently, secondary hyperparathyroidism (25,36). It is also known that cardiovascular risk increases in menopause with hormonal changes (26,36). In postmenopausal women, Pirro M. et al. showed that VitD insu ciency was associated with arterial stiffness weakly. The main in uential factor was PTH level, independent of other risk factors (25). However, possible cardiac risks due to the menopause nature can be considered a confounding factor (26,36). Therefore regardless of the confounding factors, we aimed to reveal the differences that secondary hyperparathyroidism will create on endothelial and diastolic functions, if present, using echocardiography, carotid, and brachial tDi in premenopausal women with VDD.
In our study, in baseline characteristics, systolic BP is higher in the secondary hyperparathyroidism group than the group with normal PTH levels(p = 0.020). Neurohormonal effects of hyperparathyroidism may cause the difference (37). A positive correlation was found between PTH level and systolic BP, albeit standard PTH value, in a study investigating the relation of endothelial functions with VitD and PTH levels. Our ndings can be evaluated in this route (38).
As stated before, secondary hyperparathyroidism prevalence is approximately 18-25% according to VitD levels, secondary hyperparathyroidism prevalence in severe VDD was also shown as 33%. In Gómez-Alonso C et al. study, the autumn-winter months covered 60% of the study period in 40°N latitude in a city 400 meters above sea level (18). However, Islam MZ. et al. showed that this prevalence could vary in the range of 16%-89%, according to ethnic differences (39). Secondary hyperparathyroidism was observed with a rate of 41% in our study. Ethnic differences, characteristics of the city, and the time interval may have in uenced these differences.
In the secondary hyperparathyroidism group, A wave durations were longer, and E/A ratios were lower than the group with normal PTH levels(p values are 0.043 and 0.031, respectively). It can be accepted that the prolongation of the A wave duration is effective in decreasing the E/A ratio. In Pandit A. et al. study investigating VDD and left ventricular diastolic dysfunction in a population of 67% women, there was no signi cant relationship between vitamin D levels and diastolic parameters(Deceleration time, e' wave, E/A, and E/e' ratios) (40). In Pilz S. et al. study, which evaluated approximately 600 patients over 65 years of age, the prevalence of diastolic dysfunction might increase seasonally in VDD. Even in these seasons, VDD lost its signi cance when adjusted for age and cardiovascular risks (41).
In 'the fth Tromso study,' myocardial systolic contraction rates were signi cantly lower in the secondary hyperparathyroidism group(approximately 100 patients), even after adjustments for covariables(for lateral s' and septal e', s' waves; p < 0.05). In the follow-up of these patients 6-12 months later, in case of persistently elevated PTH, a similar trend was observed in myocardial systolic contraction velocities, but the signi cance is lost. This study's difference from our study was that the mean VitD values after the relevant unit conversions (nmol/l to ng/ml) were 17.2 ± 5.0 in the high PTH group and 18.7 ± 5.4 ng/ml in the normal PTH group(p > 0.05). Also, no signi cant difference was found in diastolic parameters(E,A waves, IVRT, pulmonary vein atrial reversal ow) in this study (42).
The study's design differences, the population, the city where the study was conducted, and some other factors may have been in uential in these differences. Suppose we interpret the ndings of diastolic parameters in our study; there may be additional impairment in diastolic functions due to PTH levels in VDD, even if there is no confounding factor. Lim S. et al. showed VDD is associated with endothelial dysfunction and coronary atherosclerosis (43). In diabetic nephropathy patients, VDD is signi cantly linked with carotid IMT, but no signi cant relation was found for chronic renal failure (44,45). Studies conducted with chronic renal failure and secondary hyperparathyroidism found a signi cant relationship between carotid thickness and PTH levels (46,47). In Yankouskaya L. et al. study, which investigates women with arterial hypertension, the effects on endothelial dysfunction were explained by the pathogenic relationship between VDD and PTH levels, not a single factor (38).
Our study carotid IMT mean values are 0.6 ± 0.1 mm in secondary hyperparathyroidism and 0.5 ± 0.1 mm in the group with normal PTH levels(p = 0.032), no other signi cant difference related to carotid and brachial artery. We can say that hyperparathyroidism is an independent risk factor for carotid IMT in VDD.
As a result, we observed a signi cant relationship between PTH levels and endothelial(carotid IMT) and diastolic functions(A wave duration, E/A ratio) in premenopausal women with VDD, independent of all other risk factors. Therefore, patients should be evaluated in terms of possible future cardiac risks by paying attention to PTH levels in diagnosing VDD. We can state that our study's ndings are predictive for future studies since the signi cant ndings were detected independently of confounding risk factors and even in young patients. The subject can be clari ed with multicentric studies and advanced subgroup analyses that can be done with many patients in further studies.

Limitations
A limited number of patients could be included in our study. Our study population may not fully re ect the target population due to the pandemic process, the inclusion of patients in a limited time frame(it may have helped clarify the ndings.), and the recruitment of only patients referred to cardiology.
The difference between the vitamin B12 levels among the study groups may have affected the results, even though the vitamin levels were in the reference range.

Conclusion
Our study showed that parathormone has adverse effects on diastolic functions(A wave, E/A ratio) and endothelial functions(carotid IMT) in premenopausal women with VDD, independent of other confounding risk factors. We can state that our results will contribute to the literature since con icting information in the literature and confounding factors are included in these studies.
As a result of our study's ndings, the importance of PTH levels in VDD should not be forgotten. An awareness should be created regarding possible cardiac risks in high PTH in VDD.

Declarations
Con ict Of Interest and Funding No con ict of interest and nancial support. Special thanks to internal medicine department for their helpfulness.

Availability Data and Material
This study's data are available on request from the corresponding author(Onur Akhan, akhanonur@gmail.com), not publicly available due to privacy or ethical restrictions.