Prediction of Cardiovascular Diseases in Women With Rheumatoid Arthritis.

Aim: To develop prognostic models for arterial hypertension (AH) and atherosclerosis based on studying the totality and significance of traditional and disease-mediated risk factors (RFs) in women with rheumatoid arthritis (RA).

Material and methods: 223 female patients with RA aged 54.9±2.1 years were evaluated at the premises of the polyclinic of the Gulla Municipal Hospital #4 (Barnaul), the "Health Center", the City Rheumatology Department of the polyclinic, and the Hospital Department in 2016-2019. Statistical analysis was performed using Excel Microsoft Office 2007, Statistica 6.0 and 10.0, and SigmaPlot 12.5 software packages. Multivariate regression analysis was used for studying the attributes influencing the development of AH and atherosclerosis in RA and for constructing predictive models. ROC analysis was used to determine the quality of the developed models. Differences were considered statistically significant at p<0.05.

Results: The following RFs predominating in the onset of disease were identified: traditional (hyperglycemia, obesity, increased diastolic BP (DBP), tachycardia, dyslipidemia); disease-mediated (ESR, fibrinogen, C-reactive protein (CRP), rheumatoid factor, cyclic citrullinated peptide antibodies, moderate and high DAS-28 activity), and psychosocial (stress, anxiety, depression, sleep disorders). The highest RF incidence and their combinations were determined with a RA duration of more than a year: traditional (obesity, hyperglycemia, increased systolic BP (SBP)), and decreased glomerular filtration rate; and disease-mediated (prednisolone treatment). A highly sensitive model for AH screening was developed that included a combination of RFs: disease-mediated (RA duration, CRP); traditional (improper diet, low physical activity, history of early cardiovascular diseases, increased SBP and DBP, preeclampsia and/or eclampsia, early menopause, older age, dyslipidemia); psychosocial (anxiety, depression), and a high salt-taste threshold. A highly sensitive model was developed for probable prediction of multifocal atherosclerosis in RA in women. The model includes a complex of risk factors: disease-mediated (RA activity by DAS-28, CRP, fibrinogen, ESR, dose-dependent prednisolone treatment); traditional (AH, SBP, waist circumference, heart rate, early menopause, preeclampsia and/or eclampsia, age 55 years and older, dyslipidemia); and psychosocial (sleep disorders, depression).

Conclusion: Algorithms for early prevention of AH and atherosclerosis were developed with consideration of identified predictors and proposed prediction models for women with RA.