ABC | Volume 114, Nº1, January 2019

Original Article Otto et al. Aortic prosthesis mismatch in public health system Arq Bras Cardiol. 2020; 114(1):12-22 Table 2 – Postoperative Doppler-Echocardiographic Data According to PPM Groups (176 patients- Definition# 2) No PPM 30.4% (54) Moderate PPM 36.2% (64) Severe PPM 33.4% (58) p LVEF % 57 ± 14% 60 ± 14% 58 ± 14% 0.75 Vmax Ao cm/s 273 ± 15 306 ± 25 335 ± 18 < 0.002 Peak Gradient (mmHg) 30 ± 14 ┼ 37 ± 14 ┼* 45.1 ± 20 * < 0.0001 Mean Gradient (mmHg) 18 ± 8 ┼ 21 ± 8 ┼* 28 ± 13 * <0.0001 EOA cm² 1.78 ± 0.43 ┼ 1.3 ± 0.2 ┼* 0.52 ± 0.1 * < 0.0001 EOA/BSA cm²/m² 1.05 ± 0.17 ┼ 0.73 ± 0.06 ┼* 0.51 ± 0.1 * < 0.0001 VTI LVOT/VTI Ao valve 0.49 ± 0.1 ┼ 0.41 ± 0.07 ┼* 0.33 ± 0.08 * < 0.0001 LVOT diameter (cm) 2.15 ± 0.3 ┼ 2.02 ± 0.24 ┼* 1.92 ± 0.22 * 0.04 LV mass index g/m² 115 ± 42 119 ± 38 117 ± 35 0.84 LA index volume ml/m² 32 ± 12 33 ± 10 33 ± 12 0.72 Ascending Aorta cm 3.6 ± 0.72 3.6 ± 0.74 3.5 ± 0.56 0.29 ┼ p < 0.05 between no PPM and moderate PPM. * p < 0.05 between moderate and severe PPM. LVEF: left ventricle ejection fraction; EAO: effective orifice area; BSA: body surface area; VTI: velocity time integral; LVOT: left ventricle outflow tract; Ao: aortic; LV: left ventricle; LA: left atrium. NA: not available. The prevalence of severe PPM in this studywas 33%compared to up to 20%previously described. 6,15 Oliveira et al. described lower prevalence of PPM in Brazilian patients with small aortic annulus (16.8%). However, their cutoff points for definition of PPMwas different from the present study. 9 Another significant finding is that degenerative aortic stenosis is the main cause of aortic valve disease in our study (50%), but rheumatic etiology remains high, compared to data reported in developed countries (19%). 5,7 In addition, rheumatic etiology is independently associated with the risk of severe PPM. PPM Characteristics Similarly to other studies, 15,16 patients with severe PPM had larger BSA and BMI 15 and smaller LVOT diameters. 6,15 Patients with severe PPM in our study were younger compared to those in previous studies and mostly males. 5,7,8,15 This finding could be explained by the inclusion of aortic regurgitation in our study, to explain male gender as an independent variable, and by the significant proportion of patients with rheumatic etiology, to explain the predominance of younger individuals with PPM. 18,19 Determinants of Severe PPM A very important application of our findings is in the identification of independent preoperative variables which determine the risk of severe PPM. From these variables we built a predictive model that enables the identification of individual risk for development of severe PPM. This model can be used to identify patients at high risk for severe PPM prior to AVR and to implement preventive strategies. 18,19 A larger BMI (> 25 kg/m 2 ), male gender, smaller LVOT diameter (< 2.1 cm), younger age (≤ 60 years) and rheumatic etiology were determinants of high risk for severe PPM. Based on the predictive model proposed in this study, preventive strategies should be contemplated, including aortic root enlargement and implantation of prosthetic valves with superior hemodynamic performance with surgical or transcatheter procedure 11,15,18, 19 In this study population, transcatheter implantation is controversial because the procedure is approved for high surgical risk in patients, who are usually older and at a higher level of frailty. This study also raises the importance for improving the hemodynamic performance of the prosthetic valves implanted in the Brazilian Public Health System . However, we must consider the costs of using stentless prostheses in the public health system, which may have a negative impact cost to treat the population more comprehensively. Potential Limitations and Strengths of the Study This was a retrospective study with limited data available of the iEOA in part of the population. It is important to emphasize that it was possible to obtain the indexed effective orifice data - the main parameter for the differentiation of PPM - in only 55% of the study population. Hence, the 45% of patients with missing data could generate bias and increase the prevalence of severe PPM. The type of prosthesis used was not found in some patients, in spite of being exhaustively searched in medical records. No long-term echocardiographic and clinical follow-up data was available to assess the effect of PPM on outcomes. However, this is the first study to show a high frequency of PPM in AVR performed in the Brazilian Public Health System. In addition, our study was able to build a mathematical model to predict PPM and find preoperative independent variables related to the implantation of small prosthesis. Further studies are needed to apply and validate this model in other populations. 15