MATHEMATICAL MODEL OF AGING IN COVID-19; [MATEMATI^KI MODEL STARENJA KOD KOVIDA-19]

Abstract

Background: The aim was examination of the intimamedia thickness of carotid arteries in COVID-19 infection. Methods: In 50 patients, the thickness of the intimomedial complex (IMT) in the common carotid arteries was measured. The values were compared with the control group in 2006-9. The condition of the lungs was assessed by ultrasound score (It score) (0-42) as mild (0-14) or mediumsevere (15-28) Covid. IMT thickening risk factors and the value of fibrinogen, IL-6 and CRP were recorded. Two IMT prediction models were formed. The socio-epidemiological model predicts the development of IMT based on epidemiological factors. Apart from these factors, the second model also includes the values of the mentioned biomarkers. Results: It score 20±6, IMT values right: median 0.99 mm, p25=0.89, p75=1.14; left: 1±0.22 mm. Control: IMTright: median 0.7 mm, p25=0.68 mm; p75=0-9 mm; left: median=0.75 mm, p25=0.6 mm, p75=1.0 mm. The group/control difference is highly significant. Epide mio - logical model: logit (IMT)= 4.463+(2.021+value for GEN)+(0.055x AGE value)+(-3.419x RF value)+(-4.447x SM value)+(5.115x HTA value)+(3.56x DM value)+ (22.389x LIP value)+(24.206x CVD value)+(1.449x other value)+(-0.138x It score value)+(0.19xBMI value). Epidemiological-inflammatory model: logit (IMT)=5.204+ (2.545x GEN value)+(0.076x AGE value)+(-6.132x RF value)+(-7.583x SM value)+(8.744x HTA value)+(6.838x DM value)+(25.446x LIP value)+(28.825x CVD value)+ (2.487x other value)+(-0.218xIt score value)+(0.649x BMI value) +(-0.194x fibrinogen value)+(0.894x IL-6 value)+(0.659x CRP value). Values for both models Exp(B)=4.882; P of sample=0.83; logit=-0.19; OR= 23.84; model accuracy for the first model 87% and for the second 88%; Omnibus test of the first model c2=34.324; p=0.000; reliability coefficient -2LogLH=56.854; Omnibus test of the second model c2=39.774; p=0.000; and -2LogLH=51.403. Conclusions: The ageing of blood vessels in COVID-19 can be predicted. © 2023 Sciendo. All rights reserved.