Aterosclerosis carotídea subclínica en pacientes con infección por el virus de la inmunodeficiencia humanafactores de riesgo y relación con los modelos predictivos de riesgo cardiovascular de Framingham, SCORE y D:A:D

Résumé

Background and objectives: Cardiovascular disease (CVD) is an important cause of morbidity and mortality in people living with HIV (PLWH). These people have an increased risk of cardiovascular events which happen earlier in PLWH than in seronegative people. Cardiovascular risk (CVR) can be estimated by several predictor scales which are used in general population, and by some scales specifically designed for PLWH. The aims of this study were: to estimate the prevalence of subclinical carotid atherosclerosis (ScA) in our area; to study the ability of Framingham, SCORE and D:A:D predictor models to classify CVR and predict ScA; and to design a nomogram with a non-inferiority predictor ability than the three predictor models mentioned above. Material and methods: We designed a cross-sectional descriptive study from a prospective and multicentric data collection. PLWH above 18 years old who had been followed by the Infectious Diseases department were included. CVR was estimated by Framingham, SCORE and D:A:D equations. The presence of ScA was assessed by ultrasound, with a manual procedure protocol for the measurement of carotid intima-media thickness (cIMT) and carotid plaque detection. A binary logistic regression was performed and its results were applied to design a ScA predictor nomogram. ROC curves were used to assess the models&apos; predictive power and Cohen&apos;s Kappa coefficient was used to detect how accurate was the detection of ScA in the high CVR group. All the estimations were calculated by SPSS Statistics v24 and free software &apos;R&apos;. Results: Two hundred and thirty five patients were analysed. The cIMT median of all carotid segments was 0.75 mm (IQR 0.65 - 0.89 mm) and the cIMT median in left common carotid artery was 0.65 mm (IQR 0.57 - 0.77 mm). The ScA prevalence was 55.3% (CI 95% 48.9 - 61.5 %) and the carotid plaque prevalence was 14.9% (CI 95% 10.9 - 20.0 %). The D:A:D model was the one which classified less number of patients in the low CVR group than the other models. Regarding to the presence of ScA, the CVR was underestimated by all the predictive models (Framingham, SCORE and D:A:D) in a 22.6%, 26.4% and 8.1% of the patients respectively. The model&apos;s ScA predictive power was similar for all the equations and there were no significant statistical differences between them; the areas under curve were = 0.762, 0.779 and 0.768 respectively for each model. The classical CVR factors and the HIV infection duration were the main ScA associated risk factors. The independent predictors of ScA were: age ?50 years (OR = 3.054; CI 95% [1.54 - 6.04]; p < 0.001), total cholesterol (by mg/dL) (OR = 1.010; CI 95% [1.01 - 1.02]; p = 0.002), hypertension (OR = 2.759; CI 95% [1.23 - 5.98]; p = 0.010), and HIV infection duration (by year) (OR = 1.047; CI 95% [1.01 - 1.09]; p = 0.019). A nomogram was designed to predict ScA and its predictive power was not inferior than Framingham, SCORE and D:A:D models power. Four variables were used to make the nomogram: age ?50 years, hypertension condition, total cholesterol levels and HIV infection duration. Conclusions: ScA was elevated in PLWH in our area and the main risk factors were classical CVR factors and HIV infection duration. The prediction power of the Framingham, SCORE and D:A:D models was similar. However, D:A:D equation was better at estimating CVR related to ScA than the other models as well as demonstrated the best accuracy for detecting ScA in patients who were classified in the high CVR group. The positive correlation between the cIMT and CVR prognostic scales would support using carotid ultrasound in order to optimize CVR assessment and the preventive treatment of CVD in PLWH. Keywords: HIV, subclinical atherosclerosis, carotid intima-media thickness, atherosclerotic plaque, ultrasonography, carotid, cardiovascular risk, Framingham, SCORE, DAD.