ABC | Volume 114, Nº1, January 2019

Original Article Determinants of Aortic Prosthesis Mismatch in a Brazilian Public Health System Hospital: Big Patients or Small Prosthesis? Maria Estefania Otto, 1 Fernando Antibas Atik, 1 Marcelo do Nascimento Moreira, 1 Luiz Carlos Madruga Ribeiro, 1 Bianca Corrêa Rocha de Mello, 1 Joyce Gomes Elias Lima, 1 Maiara Sanchez Ribeiro, 1 Ana Carolina Pereira Matos Domingues, 1 Reyna Pinheiro Calzada, 1 Armindo Jreige Jr., 1 Larissa Lucas Schloicka, 1 P hilippe Pibarot 2 Instituto de Cardiologia do Distrito Federal (ICDF), 1 Brasília, DF – Brazil Québec Heart & Lung Institute - Valvular Heart Diseases, 2 Quebec – Canada Mailing Address: Maria Estefania Otto • Instituto de Cardiologia do Distrito Federal (ICDF) - Aos 02 Bloco B Apto 604. Postal Code 70660-022, Brasília, DF – Brazil E-mail: mariaestefaniaotto@gmail.com Manuscript received October 19, 2018, revised manuscript February 06, 2019, accepted March 10, 2019 DOI: 10.5935/abc.20190231 Abstract Background: Prosthesis-patient mismatch (PPM) is associated with worse outcomes. Objective: Determine the frequency and evaluate preoperatory variables independently associated with severe PPM in a tertiary hospital focused on Public Health Care. Methods: A total of 316 patients submitted to aortic valve replacement, who had echocardiography performed within the first 30 days after surgery, were retrospectively analyzed. The indexed effective orifice area (iEOA) of the prosthesis was used to classify the patients into three groups, according to PPM, considering bodymass index (BMI): severe PPM (iEOA) <0.65 cm 2 /m 2 ), mild to moderate PPM (iEOA, 0.65 cm 2 /m 2 – 0.85 cm 2 /m 2 ) and without PPM (iEOA > 0.85 cm 2 /m 2 ) for a BMI < 30 kg/m 2 and severe PPM (iEOA) < 0.55 cm 2 /m 2 ), mild to moderate (iEOA, 0.55 cm 2 /m 2  – 0.70 cm 2 /m 2 ) and without PPM (iEOA > 0.7 cm 2 /m 2 ) for a BMI > 30 kg/m 2 . Statistical significance was considered when p < 0.05. Results: iEOA was obtained in 176 patients. The frequency of severe and moderate PPM was 33.4% and 36.2%, respectively. Severe PPM patients were younger and had larger BMI, but smaller left ventricular outflow tract diameter (LVOTD). The independent variables used to predict severe PPMwere male gender, BMI > 25 kg/m 2 , age < 60 years, LVOTD< 21 mm, and rheumatic etiology with an area under the ROC curve of 0.82. Conclusion: The frequency of severe PPM is high in a Brazilian population representative of the Public Health System, and it is possible to predict PPM from preoperative variables such as rheumatic valvular disease, gender, BMI, age and LVOTD. (Arq Bras Cardiol. 2020; 114(1):12-22) Keywords: Heart Valve Prosthesis/surgery; Size Perception; Body Mass Index; Preoperative Care; Postoperative Care; Echocardiography/methods. Introduction The concept of prosthesis-patient mismatch (PPM) after aortic valve replacement (AVR) occurs when the indexed effective orifice area (iEOA) of the inserted prosthesis is too small in relation to patient body size. 1 PPM was first described in 1978, 2 and its negative impact on morbidity, mortality and left ventricular reverse remodeling has been established. 3-6 Transprosthetic gradients in patients with PPM varies with cardiac output, which in turn is determined by body surface area (BSA), and the relation of iEOA and pressure gradient is curvilinear. Therefore, iEOA smaller than 0.85 cm²/m² generates higher gradients with possible consequences to the left ventricle (LV). 2 The incidence of PPM is variable and ranges from20‑70% for moderate and 5-20% for severe PPM. 2,3 Severe PPM has been associated with a 1.8-fold increase in mortality. 3 Many studies have reported an impact of PPM on early 7,8 and late mortality, 5-8 especially in patients with pre-existing LV dysfunction. 5,7 PPMwas also associated with reduced functional capacity, less regression of LV mass and accelerated bioprosthetic valve degeneration. 6 Several factors were associated with the occurrence of severe PPM, including: advanced age, 3 female gender, 4 large body surface area (BSA) and body mass index (BMI), presence of diabetes, hypertension, small aortic valve annulus (< 21 mm), 5 and bioprosthesis implantation. 6 There are few studies on the incidence and impact of PPM in Brazil. Oliveira et al. observed that 17% of patients with EOA < 0.75 cm 2 /m 2 showed no increased mortality during a 10-year follow-up. 9 There are some interesting features specific to the Brazilian population, such as the higher prevalence of rheumatic fever, a large proportion of patients with a small BSA, and implantation of prosthesis with iEOAs not reported according to the normal reference values provided by the medical society guidelines and recommendations. 1,10 Furthermore, the analysis of preoperative factors that predict the occurrence of PPM is essential for its prevention. 11,12 12