IJCS | Volume 31, Nº4, July / August 2018

360 Gomes et al. Atherosclerosis and hypertension in pregnancy Int J Cardiovasc Sci. 2018;31(4)359-366 Original Article the number of complicated pregnancies, 12 even after the normalization of blood pressure levels after childbirth. Nonetheless, Romundstad et al. 13 questioned whether such an association would be a factor of ambiguity, because pre-gestational characteristics – especially obesity, hypertension and dyslipidemia – would attenuate the effect that the gestational hypertensive disorder has on the late cardiovascular outcome. Unfortunately, most studies have insufficient evidence, such as limited sample size and clinical follow-up. Considering that atherosclerosis is a gradual process that starts in childhood, the aim of this study was to verify the association between carotid atherosclerosis inmenopausal women who had pregnancy-induced hypertension. Methods Acase-control studywas carried out, with a population of women aged between 45 and 65 years, who had had menstrual irregularities or interruption in the last year. Women receiving hormone replacement therapy, those with chronic inflammatory conditions or any previously diagnosed conditions with high cardiovascular risk or heart disease were excluded from the analysis. The sample was calculated based on the systematic review performed by Brown et al., 14 using as reference a hypertension exposure during pregnancy of around 8% and Odds Ratio (OR) to increase the risk of atherosclerosis of 2.28. For a paired study with a one- tailed hypothesis test, we calculated at least 116 cases and 348 controls in order to obtain a 95% level of significance and 80% of test power with a ratio of one case for three controls. Controls were obtained from the same database, and were paired by age group. All women underwent carotid ultrasound with the same examiner; the carotid intima-media thickness (CIMT) was quantified and the presence of carotid plaques was assessed. For image acquisition, a high- resolution device (EnVisor, Philips) was used with a 12.3 MHz linear transducer. The data were recorded for subsequent analysis using the QLAB-Intima Media Thickness (QLAB-IMT, Philips) software. The presence of carotid atherosclerosis was defined when the CIMT was greater than 1 mm (mean values obtained in the analyzed segments of the right and left carotid arteries) and / or the presence of atheroma plaque. Atheroma plaque was defined as: (1) localized parietal structure with a thickness greater than 1.5 mm; (2) protrusion into vessel lumen > 0.5mmor; (3) thickness > 1.5-fold the adjacent CIMT, according to theMannheim Carotid Intima-Media Thickness and Plaque Consensus. 15 The cases consisted of women who had carotid atherosclerosis and the controls, of women who did not have this alteration at the ultrasonographic assessment. The independent variable was pregnancy-induced hypertension, considered as the self-reported information of blood pressure increase during pregnancy. According to Diehl et al., 16 this information shows good accuracy (specificity of 96% and sensitivity of 79.6%) for the antecedents of pregnancy-induced hypertension, even 24.5 years after the pregnancy. Other variables were considered, namely: blood pressure, income, smoking, type 2 diabetes mellitus, family history of coronary artery disease (CAD), bodymass index (BMI), number of pregnancies, preterm birth, low birth-weight offspring, fasting glycemia, total cholesterol (CT), high-density lipoprotein cholesterol (HDL-cholesterol), low-density lipoprotein cholesterol (LDL-cholesterol), triglycerides and ultrasensitive C-reactive protein (us-CRP). The study was approved by the Research Ethics Committee of Complexo Hospitalar Hospital Universitário Oswaldo Cruz/ Pronto-Socorro Cardiológico de Pernambuco under CAAE number 55361416.0.0000.5192 and Opinion number 1,593,189 of June 16, 2016. Statistical analysis The results were expressed as percentages for categorical variables and as statistical measures such as means, standard deviation and medians, when indicated, for numerical variables. The association between the occurrence of carotid atherosclerosis and the categorical variables was performed using Pearson’s chi-square test, whereas the non-paired Student’s t test was used to compare carotid atherosclerosis in relation to numerical variables. Cox regression analysis was performed to evaluate the influence of covariates on carotid atherosclerosis development in the menopausal period. The strength of the association between the categorical variables was evaluated using the odds ratio (OR) with the respective confidence interval. The normality hypothesis verification was performed using the Kolmogorov-Smirnov test. The level of significance used in the statistical test decisions was 5% and the intervals had 95% of confidence. The Statistical Package for the Social Sciences (SPSS) version 21 was the statistical program used for the statistical calculations.