Study design and selection of participants
Full-thickness vaginal wall biopsies and whole blood samples were collected from 32 women undergoing surgery for benign gynecological reasons at the University of Debrecen, Faculty of Medicine, Department of Obstetrics and Gynecology between 05/17/2017 and 04/20/2018 . The study protocol complied with the Declaration of Helsinki principles and was approved by University of Debrecen, Faculty of Medicine, Department of Obstetrics and Gynecology, and all experimental methods and protocols were approved by the Hungarian National Institutional Review Medical Research Council (approval no. 7239-3/2017EIUG). All women signed written informed consent before participating in our research.
Exclusion criteria included women with osteoporosis, cancer, zinc deficiency, endometriosis, pregnancy, immunological and connective tissue diseases, recent use of vaginal or systemic hormone replacement therapy, and women with a prior pessary or intrauterine device (IUD) use. None of the participants indicated taking zinc or copper supplements. None of the patients used nicotine containing products or consumed excess amounts of alcohol. We prospectively collected relevant demographic and clinical data into a predefined database.
Biological sample collection and processing
We have collected peripheral whole blood samples from the patients into 10 ml capacity BD Vacutainer® K2E (EDTA) plastic tube. We have divided each sample into five separate containers and stored them at − 20 °C for further analysis .
We have collected tissue specimens of vaginal wall biopsies in a standardized fashion from women undergoing abdominal or vaginal hysterectomy for benign gynecologic reasons [20, 21]. Tissues were transferred to polypropylene tubes and stored at − 70 °C for further analysis. Every participant received a unique study identification number after enrolment. We have transferred the coded samples to the laboratory and stored them for a maximum of 1 year. After thawing, the samples were processed immediately. During the laboratory measurements, the investigators were blinded to the persons providing the samples.
We have standardized the specimen collection because vaginal wall composition might vary throughout the vagina, and full-thickness biopsies of the anterior vaginal wall were cut from the midline area of the vault with Metzenbaum scissors as previously described . We carefully avoided any crush injury to the site of the vaginal wall biopsy.
A validated analytical balance was used to weigh samples of vaginal tissues (Precisa ES225-SMDR, Precisa Gravimetrics AG, Switzerland). The samples were moved into glass beakers and placed into a drying cabinet to dry until constant weight. After measuring the dry mass of the samples, a pre-treatment of wet digestion was carried out before the elemental analysis. For the mineralization, we added 4 ml of 65% (m/m) nitric acid (Sigma-Aldrich, St. Louis, MO, USA) to the samples, and beakers were heated until complete dissolution. After cooling back to room temperature, an additional 1 ml of 30% (m/m) hydrogen peroxide (Sigma-Aldrich, St. Louis, MO, USA) was added. The resulted transparent solutions were transferred into volume calibrated plastic test tubes utilizing an ultrasound bath and diluted up to 10 ml with 0.1 M nitric acid prepared in ultrapure water (MilliQ, Millipore System, Merck, Germany) .
Whole blood samples were also treated with wet digestion at atmospheric pressure before the analysis: 1 ml of the blood was transferred into glass beakers and heated on an electric hot plate along with 5 ml concentrated nitric acid and 1 ml hydrogen peroxide. Before adding the hydrogen peroxide, 1 ml of ultrapure water was pipetted to the acid digested dry samples to avoid a too heavy a reaction. When the intense reaction generated by the peroxide stopped, the samples were poured into plastic tubes by continuously washing the beakers with 0.1 M nitric acid and diluted up to 10 ml final volume .
We have kept all samples in polypropylene tubes in which they were diluted and stored at 4 °C in a refrigerator until measurement.
Certified reference materials
Certified reference material of an artificial clinical control (Seronorm™, Sero AS, Billingstad, Norway) was used to validate the elemental analytical method. Trace metals were measured within the acceptance range given in the certification (relative standard deviation, RSD < 5%).
The elemental analysis of the pre-treated tissue and blood samples was carried out by inductively coupled plasma optical emission spectrometry (ICP-OES 5100, Agilent Technologies, Santa Clara, CA, USA). The measurements were conducted in SVDV (Synchronous Vertical Dual View) mode, gaining intensity data from the axial and radial view, simultaneously. An automatic sample introduction was applied (SPS 4, Agilent Technologies, Santa Clara, CA, USA), and the samples were measured in a randomized design . We have performed measurements to generate a five-point calibration curve for the quantitative analysis of copper and zinc. We have diluted the calibration solutions from a multi-element standard of 1000 mg/l (ICP standard IV, Merck, Germany) with 0.1 M nitric acid in ultrapure water. We have expressed trace element concentration of vaginal tissues in milligrams per kilogram (mg/kg), and trace element concentration of whole blood samples in milligrams per liter (mg/l) .
We have used SigmaStat (Systat Software, San Jose, CA, USA)/SPSS software (IBM, Armonk, NY, USA) for statistical analysis. We have performed Student's t-test to evaluate the differences in the mean levels of trace elements. The correlations between trace element concentrations in the blood / vaginal tissue samples and patients' age were assessed using the Pearson correlation coefficient. Multiple linear regression was calculated to predict vaginal tissue zinc and copper levels based on age, menopausal status, number of vagina deliveries and copper/zinc blood levels in which the vaginal tissue copper and zinc level was the dependent variable, the age, menopausal status, number of vagina deliveries and copper/zinc blood levels were independent variables. We have considered the differences significant if the p value was less than α = 0.05. Numerical results are presented as estimated mean ± standard deviation unless otherwise specified. Sample size calculation was performed with G*Power 22.214.171.124 Statistical Software for Windows (Heinrich Heine University, Düsseldorf, Germany). The following parameters were used: effect size of 0.3, the number of tested predictors of 1 and a total number of predictors of 4, and R2 of 0.4 (based on preliminary data). Sample size of 29 women is needed in order to achieve a power of 80% with a level of significance 5%.