ABC | Volume 115, Nº2, August 2020

Original Article Abellán-Huerta et al. Velocity-time integral of aortic regurgitation Arq Bras Cardiol. 2020; 115(2):253-260 VC, as they were not available for all patients and could cause overfitting. Model performance for predicting severe AR was evaluated by calibration (Hosmer-Lemeshow statistic) and discrimination (C-index) measures, both internally validated using the bootstrap resampling technique. The association between the VTI of AR and its severity was explored through multivariate analysis regardless of HR and DBP. We evaluated the relationship between VTIi and AR severity with a new logistic regression analysis. Confidence intervals (95%CI) were provided when appropriate. All probability values were 2-sided, and p-value<0.05 was statistically significant. Statistical analysis was performed using the SPSS software, v.18.0 (SPSS Inc., Chicago, Illinois). Results The original sample consisted of 65 patients with moderate or severe native AR in sinus rhythm. Proper Doppler alignment of the regurgitant jet could not be obtained for three patients, who showed very eccentric jets, and were thus excluded. Out of the remaining 62 participants, 40 (64.5%) presented moderate AR, and 22 (35.5%) had severe AR. Acute AR was diagnosed in 4 patients (6.5% of the sample). The consistency among the quantification determined by the two echocardiographers was κ =0.83. All patients included were Caucasian. Table 1 presents the baseline characteristics of the sample. As shown in Table 2, the VTI of the aortic regurgitant flow was higher in patients with moderate AR than in those with severe AR. The VTI range was 2.05 m [1.53–3.58 m] in the moderate AR group and 1.88 m [0.96–2.84 m] in the severe AR group. We found a significant and inverse correlation between VTI and HR [Pearson’s correlation coefficient (r p )=- 0.408, p=0.001]. Patients with severe AR presented lower LVEF, higher LVEDD and LVESD, as well as a larger ERO, RV, and VC. However, the proper measurement of these parameters was only possible in 62.9% of the sample for ERO, 67.7% for RV, and 72.6% for VC. We underline that we identified no statistically significant association between AR severity and PHT, even though we detected a trend for it. In the bivariate analysis (Table 3), VTI was inversely associated with AR severity. Besides, the classic severity variables related to the size and function of the left ventricle were associated with AR severity. In the multivariate analysis, the VTI value acted as a marker of AR severity regardless of LVEDD, left ventricular end-diastolic volume (LVEDV), Table 1 – Baseline characteristics Characteristic Total (n=62) Moderate AR (n=40) Severe AR (n=22) p-value Age (years) 68.5±14.9 68.6±14.2 66.1±15.5 0.299 Male 33 (53.2) 20 (50) 13 (59.1) 0.492 BMI (kg/m 2 ) 27.5±4.7 26.5±4 29.4±5.9 0.340 SBP (mmHg) 135.6±17.8 133.6±16.7 139.4±19.8 0.213 DBP (mmHg) 62.2±15.5 63.2±12.7 59.8±19.8 0.373 Heart rate 66.8±11.3 65.8±10.6 68.5±12.4 0.382 Arterial hypertension 45 (72.6) 29 (72.5) 16 (72.7) 0.985 Diabetes mellitus 11 (17.7) 8 (20) 3 (16.6) 0.530 Dyslipidemia 30 (48.4) 20 (50) 10 (45.5) 0.732 Active smokers 10 (16.1) 6 (15) 4 (18.2) 0.744 eGFR (mL/kg/1.73 m 2 ) 77.3 [40.3] 86.6 [42.3] 72.9 [34.6] 0.408 Hemoglobin 13.2±1.8 13.3±1.8 13.2±2 0.893 Beta-blockers 29 (46.8) 17 (42.5) 12 (54.5) 0.363 ACE inhibitors 19 (30.6) 11 (27.5) 8 (36.4) 0.469 ARA 16 (25.8) 13 (32.5) 3 (13.6) 0.104 DHP CCB 2 (3.2) 2 (5) 0 (0) 0.286 Non-DHP CCB 10 (16.1) 4 (10) 6 (27.3) 0.145 Amiodarone 2 (3.2) 1 (2.5) 1 (4.5) 1 Diuretics 31 (50) 17 (42.5) 14 (63.6) 0.111 Statins 26 (41.9) 19 (47.5) 7 (31.8) 0.231 Previous hospital admission for HF 16 (25.8) 9 (22.5) 7 (31.8) 0.422 ACE: angiotensin-converting enzyme; AR: aortic regurgitation; ARA: angiotensin receptor antagonists; BMI: body mass index; CCB: calcium channel blockers; DBP: diastolic blood pressure; DHP: dihydropyridine; eGFR: estimated glomerular filtration rate; HF: heart failure; SBP: systolic blood pressure. Continuous variables with normal distribution are expressed as mean±standard deviation, those with skewed distribution as median [interquartile range], and categorical variables as n (percentage). 255