• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.5.2020.tde-22062021-090000
Document
Author
Full name
Maxim Goncharov
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2020
Supervisor
Committee
Jatene, Fabio Biscegli (President)
Atik, Fernando Antibas
Campos, Carlos Augusto Homem de Magalhães
Terra, Ricardo Mingarini
Title in Portuguese
Estudo de coorte observacional prospectivo multicêntrico internacional para a elaboração e validação de um modelo de predição de risco de mortalidade em pacientes de alto risco submetidos à cirurgia de alto risco submetidos à cirurgia de revascularização miocárdica: Projeto HiriSCORE
Keywords in Portuguese
Avaliação de risco
Estudos de validação
Fatores de risco
Mortalidade hospitalar
Procedimentos cirúrgicos cardiovasculares
Abstract in Portuguese
FUNDAMENTOS: Recentes evidências na área de cirurgia cardiovascular sugerem que há necessidade de novos modelos de predição de risco, além dos escores de risco tradicionais, especialmente dedicados para pacientes graves. OBJETIVO: Elaborar e validar o modelo de predição de mortalidade intrahospitalar após cirurgia de revascularização do miocárdica em pacientes de alto risco - HiriSCORE. MÉTODO: Estudo observacional de coorte prospectivo multicêntrico com a participação de 8 hospitais, onde um centro se localiza na China e sete no Brasil. Para isso um banco de dados foi criado na plataforma REDCap (https://redcap.hc.fm.usp.br). Foram incluídas 121 variáveis, dentre elas, características pré, intra e pós-operatórias, além dos desfechos clínicos e mortalidade hospitalar. A amostragem utilizou como critério de inclusão o risco de mortalidade de 5 a 10% pelo EuroSCORE II, em pacientes submetidos a cirurgia de Revascularização Miocárdica isolada. RESULTADOS: O total de dados coletados envolveu 19.786 pacientes cirúrgicos, dos quais 248 constituíram um grupo, submetidos à CRM, cujo escore de risco pelo EuroSCORE II de 5 a 10%. Após análise multivariada da regressão logística com Stepwise, pudemos distinguir 5 preditores importantes (p < 0,05): índice de massa corpórea > 30 kg/m2; clearance de creatinina < 30 ml/min/m2; lesão do tronco da artéria coronária esquerda > 50%; lesão das válvulas cardíacas não tratada cirurgicamente; glicemia pré-operatória > 150 mg/dL. A validação pela análise de discriminação pela curva ROC mostrou melhor desempenho para o HiriSCORE do que o EuroSCORE II e o STS (area under curve = 0,74, 0,50 e 0,67 respectivamente, p < 0,05), e pela análise de calibração também mostrou-se melhor acurácia de novo escore (observado menos predito = -0,2, + 6,7 e +11,4 respectivamente, p < 0,05). CONCLUSÕES: O modelo de predição de risco de mortalidade intra-hospitalar baseada na coorte dos pacientes graves, submetidos à CRM, HiriSCORE, foi elaborado e incluiu 5 variáveis perioperatórias. A validação do modelo mostrou melhores resultados na predição de mortalidade em comparação com os modelos EuroSCORE II e STS neste grupo de pacientes
Title in English
International observational multicenter prospective cohort study for the elaboration and validation of a mortality risk prediction model in high-risk patients undergoing myocardial revascularization surgery: HiriSCORE project
Keywords in English
Cardiovascular surgical procedures
Hospital mortality
Risk assessment
Risk factors
Validation studies
Abstract in English
Current evidences in the field of cardiovascular surgery along with discussions on the clinical application of traditional risk assessments indicate that there is a great need for predicting mortality risk in a group of high-risk patients. OBJECTIVE: To develop and validate an in-hospital mortality prediction model in high-risk patients undergoing coronary artery bypass grafting - HiriSCORE. METHOD: Observational multicenter prospective cohort study involving 8 hospitals, one from China and seven from Brazil. The database was created on the REDCap platform (https://redcap.hc.fm.usp.br). A total of 121 variables studied included pre, intra and post-operative characteristics, as well as clinical outcomes and hospital mortality. The sampling consisted of patients with high mortality risk in the range of 5 to 10% according to EuroSCORE II, who underwent isolated coronary artery bypass grafting. RESULTS: The total data obtained and analyzed were from 19 786 surgical cases. The results of analysis of the groups showed greater heterogeneity in clinical characteristics and mortality in the coronary artery bypass grafting cohort with the risk predicted by EuroSCORE II between 5 to 10%. This particular group contained 248 patients. Applying univariate Logistic Regression Analysis by Stepwise sampling allowed us to distinguish 5 important predictors (with p < 0.05). They were: body mass index > 30 kg/m2, creatinine clearance < 30 ml/min/m2, left main coronary artery stenosis > 50%, injury of the heart valves not treated surgically and preoperative serum glucose > 150 mg/dL. The analysis of discrimination by the ROC curve showed better results for HiriSCORE than EuroSCORE II or STS that is quantified by the area under the curve equal to 0.74, 0.50 and 0.67 respectively, p < 0.05). Additionally, calibration analysis showed better accuracy of new score (observed minus predicted equal to ?0.2, + 6.7 and +11.4, respectively, p < 0.05). CONCLUSIONS: The HiriSCORE cohort-based risk prediction model for intrahospital mortality was developed and showed better results in predicting mortality compared to the EuroSCORE II model in the high-risk group
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
MaximGoncharov.pdf (1.95 Mbytes)
Publishing Date
2021-06-25
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.