ABC | Volume 114, Nº1, January 2019

Original Article Silva et al. Risk stratification with reduced DAD Score Arq Bras Cardiol. 2020; 114(1):68-75 Table 1 – Main epidemiological, clinical and laboratory aspects of 71 HIV-positive patients n = 71 Males, n (%) 38 (53.52) Age, years 47.23 ± 9.36 Time of HIV diagnosis (years) 12 (6-17) Value of CD4 cells/mm 3 654.6 ± 308.3 Weight, Kg 73.14 ± 16.37 BMI, Kg/m 2 26.77 ± 5.21 Systolic pressure, mmHg 119.9 ± 15.47 Diastolic pressure, mmHg 75.97 ± 10.46 Total cholesterol, mg/dL 199.9 (171.2-244.9) LDL, mg/dL 126.4 ± 40.27 HDL, mg/dL 47.85 ± 14.36 Triglycerides, mg/dL 169 (96-232) Glycemia, mg/dL 100 (90.9-112.1) Glycemia > 100, n (%) 35/69(50.72) Triglycerides > 150, n (%) 38(53.52) Low HDL, n (%) 26(36.61) Triglycerides > 150 or total cholesterol > 200, n (%) 41(57.74) HDL: high-density lipoprotein; BMI: body mass index; LDL: low-density lipoprotein; n: number of subjects. being composed of male patients, median time to HIV infection diagnosis of 12 years (6-17), and CD4 lymphocyte count of 654, 6 ± 308.3 cells/mm³. The metabolic profile of these patients is shown in Table 1. We highlight the presence of alterations in triglycerides > 150 mg/dL or total cholesterol >200 mg/dL in 41 (57.74%) cases. Among the classic cardiovascular risk factors evaluated, the most frequent were dyslipidemia, physical inactivity and age in 53 (74.6%), 46 (64.78%) and 30 cases (42.25%), respectively (Figure 1). Increased waist circumference was found in 51 (71.83%) cases and metabolic syndrome, as defined by the IDF criteria, was found in 32 (45.07%) cases. Cardiovascular risk stratification was made by the Framingham 2008 and reduced DAD scores, and results are shown in Figure 2. The identification of high and very high risk was similar in both scores, differing in other categories, as Framingham 2008 showed 63,4% of low risk cases and DAD score revealed 54.9% intermediate risk cases. When the degree of agreement between scores was evaluated, an overall Kappa index of 0.318 was observed with p < 0.001. However, there was stronger agreement for patients classified as high risk, lower agreement for low risk patients, and no statistically significant difference for intermediate risk subjects (Table 2). Both scores showed a statistically significant and positive correlation with the medium-intimal thickening (Figure 3). The medium-intimal thickening (highest thickness) in these patients was 0.73 ± 0.14 and there were 28 cases of subclinical atherosclerosis (39.4%). Of these, 17 (60.7%) patients presented non-significant plaque, 6 (21.4%) only thickening, and 5 (17.8%) had both plaque and thickening. In patients classified as high risk, the occurrence of subclinical atherosclerosis was 77.8% for the Framingham 2008 score and 88.2% for the reduced DAD score. In patients classified as low or intermediate risk, the rate of subclinical atherosclerosis was higher for Framingham 2008, with 20% of patients classified as low risk presenting subclinical atherosclerosis (Figure 4). Among patients with subclinical atherosclerosis, 50% were classified as low or intermediate risk regardless of the score used. As for atherosclerosis stratified by Framingham 2008, 9/28 patients (32.1%) were classified as low risk and, by reduced DAD score, 12/28 (42.8%) were classified as intermediate risk (Figure 5). Analysis of discrimination of scores by comparison between ROC curves targeting subclinical atherosclerosis showed no significant difference between Framingham 2008 and reduced DAD scores (Figure 6), and the predictive accuracy is shown in Table 3. Discussion In this study, 71HIV-positive patients under regular follow‑up, diagnosed for more than one year, and under ART with immune reconstitution and viral suppressionwere evaluated. Most patients were classified as low risk by the Framingham 2008 score and intermediate risk by the reduced DAD score. There was a correlation between the medium‑intimal thickening and the scores, high agreement between patients classified as high risk by both scores, although no significant difference was observed in the ROC curve score discrimination analysis. Although subclinical atherosclerosis was observed in 88% of patients classified as high risk by the reduced DAD score, subclinical atherosclerosis was found in 62.5% of patients classified as intermediate risk by the Framingham 2008 score. Other authors have already compared risk stratification scores in HIV-positive patients, 8,14 but to our knowledge, this is the first study to apply the reduced DAD score and assess the degree of agreement with Framingham’s score. Checking the accuracy and applicability of this tool is important because it has been developed for HIV-infected patients, and unlike DAD full, it does not use ART-related factors, which makes its use more feasible. Regarding risk factors, it is important to highlight that, although diagnosis of dyslipidemia was reported in only 32.39% of patients, laboratory tests showed total cholesterol >200mg/dL and/or triglycerides >150 mg/dL in 57.74% of the cases identifying significant difference between the diagnosis reported by the patient and the laboratory verification of changes in lipid profile. When considering dyslipidemia or changes in LDL‑cholesterol, triglycerides or HDL-cholesterol, the frequency of dyslipidemia was 74.6%. These values reinforce the relevant presence of this risk factor in this population and the need for proper observation of criteria for diagnosis and treatment of these changes according to current guidelines. The frequency of subclinical atherosclerosis reported here (39.4%) is similar to data from Falcão et al., 15 who found 42.6% of subclinical carotid atherosclerosis. 15 Most patients evaluated were classified as low risk by 70