Abstract Objective: We aimed to analyze whether patent foramen ovale (PFO) closure reduces the risk of stroke, assessing also some safety outcomes after the publication of a new trial. Introduction: The clinical benefit of closing a PFO has been an open question, so it is necessary to review the current state of published medical data in regards to this subject. Methods: MEDLINE, EMBASE, CENTRAL/CCTR, SciELO, LILACS, Google Scholar and reference lists of relevant articles were used to search for randomized controlled trials (RCTs) that reported any of the following outcomes: stroke, death, major bleeding or atrial fibrillation. Six studies fulfilled our eligibility criteria and included 3560 patients (1889 for PFO closure and 1671 for medical therapy. Results: The risk ration (RR) for stroke in the “closure” group compared with the “medical therapy” showed a statistically significant difference between the groups, favouring the “closure” group (RR 0.366; 95%CI 0.171–0.782, P=0.010). There was no statistically significant difference between the groups regarding the safety outcomes, death and major bleeding, but we observed an increase in the risk of atrial fibrillation in the “closure” group (RR 4.131; 95%CI 2.293–7.443, P<0.001). We also observed that the larger the proportion of effective closure, the lower the risk of stroke. Conclusion: This meta-analysis found that stroke rates are lower with percutaneously implanted device closure than with medical therapy alone, being these rates modulated by the rates of hypertension, atrial septal aneurysm and effective closure. The publication of a new trial did not change the scenario in the medical literature.

Abstract Objective: We aimed to determine whether patent foramen ovale closure reduces the risk of stroke, also assessing some safety outcomes. Introduction: The clinical benefit of closing a patent foramen ovale after a cryptogenic stroke has been an open question for several decades, so that it is necessary to review the current state of published medical data in this regard. Methods: MEDLINE, EMBASE, CENTRAL/CCTR, SciELO, LI-LACS, Google Scholar and reference lists of relevant articles were searched for randomized controlled trials that reported any of the following outcomes: stroke, death, major bleeding or atrial fibrillation. Five studies fulfilled our eligibility criteria and included 3440 patients (1829 for patent foramen ovale closure and 1611 for medical therapy). Results: The risk ratio (RR) for stroke in the "device closure" group compared with the "medical therapy" showed a statistically significant difference between the groups, favouring the "device closure" group (RR 0.400; 95% CI 0.183-0.873, P=0.021). There was no statistically significant difference between the groups regarding the safety outcomes death and major bleeding, but we observed an increase in the risk of atrial fibrillation in the "device closure group (RR 4.000; 95% CI 2.262-7.092, P<0.001). We also observed that the larger the proportion of effective closure, the lower the risk of stroke. Conclusion: This meta-analysis found that stroke rates are lower with percutaneously implanted device closure than with medical therapy alone, being these rates modulated by the rates of effective closure.

ResumoObjetivo:Verificar se os modelos de estratificação da complexidade em cirurgias de cardiopatias congênitas atualmente disponíveis (RACHS-1, escore básico de Aristóteles e escore de mortalidade do STS-EACTS) se adequam ao nosso serviço, determinando o de melhor acurácia em discriminar a mortalidade hospitalar.Métodos:Procedimentos em pacientes menores de 18 anos foram alocados nas categorias propostas pelos modelos de estratificação da complexidade. O desfecho de mortalidade hospitalar foi calculado para cada categoria dos três modelos. Análise estatística foi realizada para verificar se as categorias apresentavam distintas mortalidades dentro de cada modelo. A capacidade discriminatória dos modelos foi determinada pelo cálculo de área sob a curva ROC e uma comparação entre as curvas dos três modelos foi realizada.Resultados:360 pacientes foram alocados pelos três modelos. Houve diferença estatisticamente significante entre as mortalidades das categorias propostas pelos modelos de RACHS-1 (1) - 1,3%, (2) - 11,4%, (3) - 27,3%, (4) - 50%, (P<0,001); escore básico de Aristóteles (1) - 1,1%, (2) - 12,2%, (3) - 34%, (4) - 64,7%, (P<0,001); e escore de mortalidade do STS-EACTS (1) - 5,5%, (2) - 13,6%, (3) - 18,7%, (4) - 35,8%, (P<0,001). Os três modelos tiveram semelhante capacidade discriminatória para o desfecho de mortalidade hospitalar pelo cálculo da área sob a curva ROC: RACHS-1- 0,738; STS-EACTS- 0,739; Aristóteles- 0,766.Conclusão:Os três modelos de estratificação da complexidade atualmente disponíveis na literatura tiveram utilidade com distintas mortalidades entre as categorias propostas, com semelhante capacidade discriminatória para o desfecho de mortalidade hospitalar.

AbstractObjective:To determine whether stratification of complexity models in congenital heart surgery (RACHS-1, Aristotle basic score and STS-EACTS mortality score) fit to our center and determine the best method of discriminating hospital mortality.Methods:Surgical procedures in congenital heart diseases in patients under 18 years of age were allocated to the categories proposed by the stratification of complexity methods currently available. The outcome hospital mortality was calculated for each category from the three models. Statistical analysis was performed to verify whether the categories presented different mortalities. The discriminatory ability of the models was determined by calculating the area under the ROC curve and a comparison between the curves of the three models was performed.Results:360 patients were allocated according to the three methods. There was a statistically significant difference between the mortality categories: RACHS-1 (1) - 1.3%, (2) - 11.4%, (3)-27.3%, (4) - 50 %, (P<0.001); Aristotle basic score (1) - 1.1%, (2) - 12.2%, (3) - 34%, (4) - 64.7%, (P<0.001); and STS-EACTS mortality score (1) - 5.5 %, (2) - 13.6%, (3) - 18.7%, (4) - 35.8%, (P<0.001). The three models had similar accuracy by calculating the area under the ROC curve: RACHS-1- 0.738; STS-EACTS-0.739; Aristotle- 0.766.Conclusion:The three models of stratification of complexity currently available in the literature are useful with different mortalities between the proposed categories with similar discriminatory capacity for hospital mortality.