ABC | Volume 113, Nº4, October 2019

Original Article Cruz et al. Surgical mitral valvuloplasty in paediatric patients Arq Bras Cardiol. 2019; 113(4):748-756 Table 3 – Description of the immediate and late outcomes (and combined outcomes) of 54 children and adolescents with rheumatic mitral disease undergoing mitral valvuloplasty Variables n (%) Immediate outcome (n = 54) None 37 (68.6) CHF 01 (1.9) Sepsis 01 (1.9) Mitral lesion* 04 (7.4) Others 04 (7.4) Bleeding and others 02 (3.7) CHF and mitral lesion * 01 (1.9) Others and CHF 01 (1.9) Mitral lesion* and others 02 (3.7) Mitral lesion* and bleeding 01 (1.9) Late outcome (n = 54) None 38 (70.4) CHF 01 (1.9) Reoperation/valve replacement (up to 30 days) 03 (5.6) Mitral lesion* 11 (20.4) Others 01 (1.9) *mitral lesion: mitral stenosis and/or residual mitral insufficiency; n: number of participants; CHF: congestive heart failure. but did not maintain significance in the multivariate analysis. In their study of 122 mitral repairs, Kim et al., 14 after univariate and multivariate analysis, found preoperative pulmonary hypertension as the only independent risk factor for death (HR: 3.75 95%CI: 1.21–11.57; p = 0.022), but this study was carried out with adults with a mean age of 48.9 ± 11.5 years. The results in this study may differ from those found in the literature due to the sample size, as well as the age: the younger the sample, the higher the mortality rate and the higher the early rate of valve failure. 15-17 In addition, postoperative TI was also a predictor of univariate analysis in this study, both in the immediate postoperative period and in the late postoperative period, but it lost significance in the multivariate analysis. No associations were found in the literature reviewed, but, like PH, TI is a criterion that should be valued. Although they were not independent predictors of outcomes, both variables were associated with unfavourable outcomes, both in the immediate and late postoperative period, with statistical significance, and should be used as initial points for surgical indication and follow-up in these patients. Yakub et al. 13 described residual mitral regurgitation ≥2 crosses as a predictor of valve failure and reoperation. In our study, the presence of moderate or significant residual MI in the postoperative period was associated with outcome both in the immediate and late postoperative period, in the univariate analysis, with statistical significance. Of the three children who underwent valve replacement, one presented moderate to severe mitral regurgitation even after attempted repair by MVP technique and the other two developed moderate to severe aortic regurgitation and underwent aortic valve replacement before completing sixty postoperative days. This data is in agreement with the literature in other studies as well, such as in those by Silva et al. 11 and Severino et al., 15 always being a marker for reoperation. It is known that the lateMVP results also depend on good coaptation of the cusps, which can be obtained through association of surgical techniques and reassessed at the end of the repair, whenever possible with intraoperative transesophageal echocardiography, 18 which is not available in the hospital in which the study was carried out. The mean duration of ECC found was 62.7 ± 18.8 minutes and the duration of anoxia was 50.0 ± 15.7 minutes, showing an association with the outcomes in the immediate postoperative period. In fact, in the literature, duration of CPB has been described as an independent predictor for cardiac surgeries, usually due to inflammatory factors in the bloodstream. Thus, it is well established that a CPB time longer than 90 minutes is associated with a more complicated postoperative period. The study by Talwar et al. 16 showed duration of CPB of 47.6 ± 11.9 and duration of anoxia of 37.2 ± 12.8, and when assessed for association with early death or reoperation, no significance was found. 16 This divergence with the literature may be due to the longer duration of both CPB and anoxia found in our sample. The ECC time found in this study may have been influenced by variables that were not studied, such as anatomical differences, anterior leaflet involvement, thickening of the ribs, calcifications and papillary muscle involvement. Non‑standardization of intraoperative and echocardiographic records, as well as the retrospective nature of the study, prevented the analysis of these data. The duration of VAD use in the immediate postoperative period was found as a predictor for outcome in this study. Silva et al. 11 also described the use of VAD, finding OR of 1.47 (95% CI 0.32-6.83), but they did not present statistical association in their sample. Other variables may have influenced this divergence in results, such as longer duration of CPB and anoxia in the study by Silva et al. 11 The use of drugs for clinical optimization of CHF, due to the volume overload generated by mitral annular dilatation, also did not present significance and no description of the study of this variable was found in the reviewed literature, except in the study by Silva et al., 11 in which there is only a description of the use of anticongestive medications in 40% of patients in the preoperative period, not presenting a statistical analysis of this data. The variables age, weight, height, duration of illness until surgery, NYHA preoperative classification, presence of atrial fibrillation and use of drugs were not significant in this study. In the studies by Talwar et al. 16 and Kalfa et al., 17 these data were evaluated descriptively without statistical analysis and using only residual MI, reoperation, valve replacement and mortality as the outcome, without considering other variables such as bleeding, CHF and sepsis as possible outcomes. 752