Study design, Setting and Participants
This was a retrospective pregnant women cohort study, studying association of copper level with preeclampsia. The study was conducted in obstetrics department of Foshan Chancheng Central Hospital, Foshan city Guangdong province, China, which obtain the Joint Commission International (JCI) Hospital Accreditation 6th ed. Pregnant women with induced labour, fetal death in uterus, serious birth defects, age at delivery Less than 18 years or more than 42 years were not included in the cohort study.
The population of this study was consecutive participants, which pregnancy registered, checked up and delivered in our hospital among on August 1, 2019 to November 30, 2019. The Hospital Ethics Committee approved this study and waived informed consent because data were unidentified. The study protocols were registered in Chinese Clinical Trial Registry (Registration number: ChiCTR2000029643).All reporting followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.[21]
2512 women were enrolled in the study. 335 were excluded for the following reasons: 281 had missing information on all four trace element,5 had chronic hypertension prior to pregnancy,3 had T2DM prior to pregnancy,20 had twin pregnancy,1 had BMI missing,19 had BMI≥30,1 age<18,5 age>42.2177 participants were remained quantity. We excluded the extreme value of copper ≤1%(9.32-13.08,n=22)and > 99%(21.13-25.98,n=21).A total of 2134 participants were included in the final analysis(Figure 1).
Data Collection
We extracted socio-demographic characteristics and delivery from electronic medical records system. Information on prepregnancy height and weight, smoking and drinking history,education level, past medical history, last menstrual period, parity history, method of conception were gathered by the physicians at the first visit for pregnancy registered. And the first visit for pregnancy registered usually opens at 8–12weeks of gestation in this city. In Guangdong province, China, the typical schedule of routine antenatal visits are at weeks 8–12,week 16,week 20,week 24,week 28 and thereafter every second week until birth.
Prepregnancy height and weight refers to the height and weight at the first visit for pregnancy registered in 8-12weeks of gestation, which self-reported by pregnant women. Prepregnancy body mass index (BMI) was calculated as weight (kg) divided by height squared(m2).Smoking and drinking refer to no smoking and no drinking in pregnancy, which self-reported by pregnant women at admission to the hospital and before delivery. In this study, all women self-reported no smoking and drinking in pregnancy. Education level included middle school or below, high school, college or above. Parity is divided into nulliparous and multiparous. Method of conception included natural conception, assisted reproductive technology (ART) conception. Gestational age (weeks) was calculated from the first day of last menstrual period (LMP) until the date of delivery. Women with irregular menstrual cycles or unsure of their LMP, we based on their first trimester ultrasound examination to corrected gestational age.
Copper sampling and measurement
Copper concentration were measured in maternal whole blood in 12–27(+6) week of pregnancy, from women declaring themselves to be fasting. In this study, copper concentrations were measured prior to the date of diagnosis pre-eclampsia. Blood samples were collected from the antecubital vein, using heparin anti-coagulation Vacuum blood vessel collection. Whole blood copper concentration were measured by flame atomic absorption spectrometer (BH5100S, BOHUI, Beijing, China) with copper - zinc composite element hollow cathode lamp at 3 mA current.
Both before and after the detection, we have internal quality control. Related reagents (trace element calibration solution, batch numbe1070719) were provided by BOHUI, Beijing, China. We respectively conducted calibration before and after testing the sample, with two concentrations of the calibration solution. Copper concentration of the calibration solution were 4.6mg/L and 1mg/L, respectively.Correlation coefficients for calibration curves of copper were greater than 0.999. All measurement on copper concentration in maternal whole blood were performed in the Center for Clinical laboratory of Foshan Chancheng Central Hospital. The respective reference intervals of copper concentration in our laboratory included were 7.12-21.29umol/L.
Outcomes
Preeclampsia was the outcome of this study, obtained directly from electronic medical records system. Preeclampsia was defined according to report of the American College of Obstetricians and Gynecologists’ Task Force on Hypertension in Pregnancy.[22]Preeclampsia was defined as newly diagnosed hypertension and proteinuria occurring after 20 weeks of gestation. Hypertension was defined as systolic≥140mmHg or diastolic≥90mmHg,2 occasions,4h apart in previously normotensive woman. Proteinuria was defined as≥300 mg/24 hour urine collection, or protein/creatinine≥0.3,or or dipstick reading =1+. Severe pre-eclampsia case were considered if they had at least one of the following symptoms: 1) Systolic blood pressure≥160mmHg or diastolic≥110mmHg, 2 occasions, 4h apart on bedrest;2)Thrombocytopenia(blood platelet<100×109/L);3)Liver function tests ≥ 2×normal or severe persistent right upper quadrant or epigastric pain;4)Serum creatinine concentration >1.1 mg/dL or doubling of creatinine in the absence of other renal disease;5)Pulmonary edema;6)New-onset cerebral or visual symptoms. All other cases were diagnosed mild pre-eclampsia. Early onset and late onset pre-eclampsia defined as delivery before 34 and≥34weeks, respectively. In this study, all cases of pre-eclampsia occurred after 28 weeks, with 5 cases of early onset and 52 cases of late onset.
Statistical Analysis
Frequency and percentage were used in categorical variables, such as education level, parity, and method of conception, delivery way. Normally distributed variables are presented as means and standard deviations (SD). Non-normally distributed variables are presented as median and interquartile range (IQR) .We compared categorical variables, normally and non-normally distributed variables between Normal and Pre-eclampsia women, used Pearson Chi-square, ANOVA and Kruskal-Wallis Rank sum test. Scatter plot with bar were used to display the median, IQR, and range of four trace element, and Kruskal-Wallis Rank sum test were used to determine differences in their distributions in Normal and Pre-eclampsia women.
We used Logistic regression analyses to estimate odds ratios and 95% confidence intervals (CIs) for pre-eclampsia in relation to trace element, unadjusted model and multivariable adjusted model. Copper concentrations were treated as continuous variables scaled to per1 SD increase and also as categorical variables by quintiles to explore potential nonlinear dose-response relationships on preeclampsia. P values for trend were obtained by coding concentration categories as ordinal variables in the regression models. To assess potential confounders, we adjusted for age at delivery (smooth),prepregnancy body mass index(smooth),education level (middle school or below, high school, college or above),parity(nulliparous, multiparous), and method of conception(natural conception, ART conception). We also evaluated subgroup analysis of copper with preeclampsia. Furthermore, age at delivery, prepregnancy BMI, education level, parity and method of conception were considered as potential effect measure modifiers. We conducted interaction test of copper with preeclampsia and showed P values for interaction under multivariable adjusted model.
We did sensitivity analysis to investigate the robustness of the findings in line with missing data. Nearly all clinical factors were similar in 2231 patients with available data on copper and the 281 patients with missing data on copper (Table S1 in the Supplementary Appendix).It shows that the missing data has little effect on the overall research conclusion to some extent. Given the differences in the baseline characteristics between eligible participate in the two groups, propensity-score matching was used to identity a cohort of patients with similar age at delivery and prepregnancy BMI. Matching was performed with the use of a 1:4 matching protocol Without replacement (greedy-matching algorithm), with a caliper width equal to 0.01 of the standard deviation of the logit of the propensity score. The results were compared in the two different samples, which show that the results are relatively robust (Table S2 in the Supplementary Appendix).
Data management and analyses were performed with the statistical software package R (http: //www.R-project. org, The R Foundation) and Empower Stats (http:// www.empowerstats.com,X&Y Solutions, Inc. Boston, MA).And 2-sided P values less than 0.05 were defined statistically significant.