Reversal of an Unfavorable Effect of Hydrochlorothiazide Compared to Angiotensin Converting Enzyme Inhibitor on Serum Uric Acid and Oxypurines Levels by Estrogen-progestin Therapy in Hypertensive Postmenopausal Women.

Hyperuricemia is considered one of the risk factors for arterial hypertension in postmenopausal women. The recent analysis by Loeffler confirmed that, as indicated in the NHANES survey, uric acid is a powerful cardiovascular risk factor.Analysis of age and sex strata in a meta-analysis showed hyperuricemia to be independently predictive of new-onset hypertension with a stronger association in women. In women, a positive relationship was found between age and serum uric acid levels. This phenomenon can also be explained by impaired renal uric acid excretion progressing with age. Moreover, menopause is independently associated with higher serum uric acid levels, whereas postmenopausal hormone use is associated with lower uric acid levels among postmenopausal women.

According to the guidelines on the management of arterial hypertension, first-line drugs in hypertensive elderly patients also include thiazide diuretics. This is particularly justified in postmenopausal women at risk of osteoporosis because thiazides have a positive effect on calcium balance, which is disordered in postmenopausal women. The natriuretic effect of thiazide diuretics itself is also desired due to increased sodium sensitivity after menopause. Meta-analysis performed by Mussini et al. covering 60 studies with thiazide diuretics in antihypertensive therapy (from years 1946-2014) showed their superiority in reducing pulse pressure over other classes of antihypertensive drugs. However; at the same time, it is known that thiazides increase serum uric acid levels.It has been suggested that increase in serum uric acid caused by diuretic treatment may partially offset the benefits of blood pressure reduction. This has been confirmed in an analysis of data from the Systolic Hypertension in the Elderly Program (SHEP) that showed that increase in uric acid levels in hypertensive patients treated with diuretics was associated with smaller clinical benefits from antihypertensive therapy assessed as a risk for episodes of ischemic heart disease.

Since menopause depends to a great extent on age, it is difficult to separate the influence of uric acid levels and its renal clearance on the development of hyperuricemia. This is possible to achieve only if uric acid metabolism is compared in postmenopausal women receiving and not receiving estrogen-progestin therapy (EPT). A potential factor limiting the unfavorable effect of thiazides on uric acid levels in hypertensive women may be EPT. Mihmanli et al. demonstrated benefits from estrogen supplementation on the renal vessels. Experimental studies on rats showed that estrogens (through a non-genomic mechanism of action) have a vasodilating effect on the renal arteries due to the mechanism of endothelium generating nitric oxide by 17-B-estradiol; and through the "NO-dependent" and "NO-independent" mechanisms, estrogens have a suppressive effect on its vasoconstrictive activity. Moreover, Xiao et al. demonstrated the effect of estradiol on regression of hardening of the renal glomeruli.

Therefore, the aim of the study was to assess the effect of EPT on serum levels of uric acid and its precursors and on its renal excretion in postmenopausal women receiving antihypertensive treatment with ACEI or thiazide diuretics as well as to explain if EPT prevents hyperuricemia caused by thiazide diuretics in postmenopausal hypertensive patients.

Methods Subjects: Treatment-naive postmenopausal women with recently diagnosed essential hypertension (grade 1 or 2 according to ESH(European Society of Hypertension)/ESC(European Society of Cardiology) 2013 guidelines) were included in the study. They were assessed by a primary care physician or gynecologist as postmenopausal, with age ranging from 49 to 53 years old, of Caucasian race. The final menstrual period (FMP) was identified retrospectively. In order to determine this criterion, I used STRAW (Stages of Reproductive Aging Workshop) guidelines. According to the authors, a lack of menstruation for 12 months indicates clinical menopause and is referred to as "postmenopause". All the women had the expected postmenopausal increase in follicle-stimulating hormone concentrations (FSH 78.32 ± 8.73 IU/ml) and experienced flushes or other vasomotor symptoms associated with menopause. For this reason and due to hypertension, they were randomized to EPT and hypotensive treatment. Normotensive women with vasomotor symptoms received randomly either EPT or nothing. The exclusion criteria included: breast cancer in a first-degree relative, hyperplasia diagnosed by endometrial biopsy, history of thromboembolic diseases, current or history of use of estrogen-progestin therapy or contraceptives, diabetes, kidney failure, thyroid disease, and heart and other chronic diseases (secondary hypertension, atrial fibrillation). Results of mammography performed 12 months prior to the study were negative in all the screened women.

In the invitation letter, women with arterial hypertension were invited to take part in a study on hypertension management and menopause-related symptoms, or menopause-related symptoms only in normotensive women. Informed consent was obtained from the subjects prior to the study. The study was carried out in accordance with The Declaration of Helsinki and was also registered and approved by the local Ethics Committee at the University of Medical Sciences in Poznan with the registration number 1319/99. A physical examination and biochemical screening was performed in all women at baseline.

Study design: The screening phase included 140 women with vasomotor symptoms associated with menopause. They met inclusion, but not exclusion criteria, and were treatment-naive for hypertension. At two separate visits at a weekly interval (wash-out period), hypertension was defined as the mean of three office-based measurements between ≥140/90 mmHg and <180/110 mmHg. Subjects included in the study had no contraindications to transdermal estrogen-progestin therapy or hypotensive treatment. The study's control group consisted of 40 normotensive women (<140/90 mmHg, measured at two separate visits under similar conditions) who had also the possibility to receive transdermal estrogen-progestin therapy due to vasomotor symptoms associated with menopause.

At the beginning of the study, hypertensive women were assigned randomly (sealed envelopes) to two treatment groups: with a diuretic - hydrochlorothiazide 25 mg/day p.o. (n=50) or with an ACEI - perindopril 4 mg/day p.o. (n=50), and to a group receiving estrogen-progestin therapy (EPT+) and a group not receiving hormones (EPT-). Normotensive women were assigned randomly only to either EPT+ or EPT- groups. Estrogen-progestin therapy was given in the form of transdermal patches releasing 17β-estradiol (0.05 mg/24 hours) and norethisterone (0.25 mg/24 hours) Estracomb TTS®.

Control measurements were performed in all subjects after one year. At the end, 20 females did not complete the study (4 due to an adverse reaction of perindopril in the form of cough, 6 needed additional antihypertensive therapy - 2 with ACEI and 4 with HCTZ and 10 for no reason). All females from the normotensive group completed the trial.

Finally, groups of 20 women with hypertension in each treatment group and 20 women with normal blood pressure, treated with or without EPT, were chosen for statistical analyses in the order in which they had finished the 12-month follow-up of the study.

Measurements: Systolic (SBP) and diastolic (DBP) office-based blood pressure was measured after five minutes in sitting position by means of a validated automatic device (Omron 705 CP) placed on the arm with higher blood pressure. In each time, the average of three measurements was used. Ambulatory blood pressure measurements were performed in all study participants at entrance and after 6 and 12 months using an automatic device (MOBIL-O-GRAPH). The device was set up to perform a blood pressure recording every 15 minutes during the day and every 30 minutes at night, over a 24-hour period. 80% of the correct measurements were needed to validate the result.

Measurement of renal function: Renal plasma flow (RPF) was measured as the clearance of 125I-iodohippuran using a constant infusion technique with timed urine sampling [20]. Based on RPF, additional renal hemodynamic parameters: filtration fraction (FF) and renal vascular resistance (RVR) were estimated [21].

Laboratory tests: Serum and urine creatinine levels were determined with a spectrophotometric method [22] using an autoanalyzer (MEGA, Merck, Darmstadt, Germany) and the Mascot software package (Matrix Science, UK). Serum and urine uric acid levels were determined with an enzymatic uricase method using PAP 150 kits by bioMérieux SA (Marcy l'Etoile, France). Serum hypoxanthine and xanthine levels were determined with high-performance liquid chromatography (HPLC) [23] using a Hewlett-Packard 1050 system with UV detector by HP Inc. (Palo Alto, CA, USA).

Serum creatinine levels were determined using Jaffé reaction. Glomerular filtration rate (GFR) was calculated using the Cockcroft-Gault equation for women.

Statistical analysis: Student's t-test was used to compare continuous variables and the normality of distribution of an investigated feature in both study populations. T-test with the Cochran-Cox adjustment was used in cases where variances of analyzed variables in both study populations differed, or Wilcoxon Mann-Whitney test in cases where there were no normal distribution.

Baseline characteristics were compared using ANOVA (in the case of parametric variables) or by means of the Kruskal-Wallis test (in the case of nonparametric variables); subsequently, the different groups were compared in pairs. Means were compared by means of the Kruskal-Wallis test one-way analysis of ranks; then, the relevant groups were compared in pairs. If the groups differed significantly, multiple comparison Dunn test was applied.

Multiple linear regression models were constructed for selected parameters of the effects of HCTZ and EPT (women with arterial hypertension). Models with interactions were also developed. If β regression coefficients for interactions of independent variables were significant, these models were used to interpret the variability of the dependent variable. F-test verifies the statistical significance of all variables in the model. Similar multiple linear regression models were also constructed to show the effect of hypertension and EPT.

Values represent mean ± standard deviation (SD) if not stated otherwise. A p value < 0.05 was considered statistically significant.