Performance de escores de risco cardiovascular na predição de mortalidade dez anos após Síndrome Coronariana Aguda
Carregando...
Arquivos
Data
2017-08-21
Autores
Petek, Amanda Aparecida [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Introdução: Os escores de risco cardiovascular foram desenvolvidos, em sua maioria, em população aparentemente sem doença cardiovascular (DCV), portanto as suas conclusões dizem respeito à prevenção primária, não se aplicando a pacientes com doença coronariana diagnosticada. Além disso, os escores prognósticos utilizados após síndromes coronarianas agudas (SCAs) avaliaram a mortalidade apenas a curto prazo. Objetivos: O objetivo do presente estudo é avaliar a performance do escore de risco de Framingham (ERF) e do escore proposto pela American College of Cardiology e da American Heart Association (ACC/AHA) em predizer a mortalidade em pacientes dez anos após SCA. Casuística e métodos: Trata-se de um estudo de coorte retrospectivo que incluiu pacientes com idade ≥ 18 anos, com SCA, que estiveram internados na Unidade de Terapia Intensiva Coronariana (UTI) do Hospital das Clínicas de Botucatu – UNESP, no período de janeiro de 2005 a dezembro de 2006, e que apresentavam os exames para o cálculo dos escores de risco cardiovascular. Foram excluídos os pacientes cujo desfecho final em dez anos não foi obtido ou aqueles não aceitaram participar do estudo. Os ERF e o ER ACC/AHA foram calculados e a mortalidade avaliada dez anos após a SCA. Resultados: Foram avaliados 451 pacientes com SCA internados na UTI coronariana. Destes 122 foram excluídos pois a mortalidade 10 anos após a internação na UTI não foi obtida. Logo, 329 pacientes com idade média de 62,913,0 anos foram estudados. Dentre esses pacientes, 58,4% eram homens e 44,4% morreram no período de 10 anos após a internação. Destes 63% apresentaram SCA sem supradesnivelamento do segmento ST enquanto 37% apresentaram IAM com supradesnivelamento do segmento ST. Quando comparamos esses dois grupos de SCA, a mortalidade foi semelhante (p=0,97). A mediana do ERF foi de 16 (14-18) %, e do ER ACC/AHA foi 18,5 (9,1-31,6). Os pacientes que evoluíram ao óbito eram mais velhos, e apresentavam maiores valores de creatinina, ureia e potássio, e menores valores de colesterol total na admissão da UTI. Além disso, é interessante observarmos que os pacientes que evoluíram ao óbito apresentaram maiores valores dos escores. No entanto, quando classificamos os pacientes em alto risco cardiovascular, apenas o ER ACC/AHA foi associado com a mortalidade. A área sob a curva (AUC) da associação do ERF com a mortalidade em 10 anos é 0,6307, com IC95%: 0,5708-0,6905, e valor de p <0,001. O ponto de corte do ERF associado com maior mortalidade é > 14,3% com sensibilidade de 74,66%, especificidade de 42,62 %, valor preditivo positivo de 50,93 % e o valor preditivo negativo de 67,83 %. Em relação ao ER ACC/AHA a AUC é 0,7015, com IC95% 0,6455-0,7576 e valor de p <0,001. O ponto de corte do escore que está associado com maior mortalidade é > 23,56%, com sensibilidade de 50,00%, especificidade de 71,04%, valor preditivo positivo de 57,94% e valor preditivo negativo de 64,04%. Na análise de regressão logística tanto o ERF (OR: 1,127; IC95%: 1,056-1,204; p<0,001) quanto o ER ACC/AHA (OR: 1,043; IC95%: 1,025-1,060; p<0,001) foram associados com a mortalidade em 10 anos em pacientes com SCA, após serem ajustados pela creatinina e potássio. Conclusão: Apesar de não apresentarem boa performance, tanto o ERF quanto o ER ACC/AHA foram associados com a mortalidade em 10 anos. No entanto, o ER ACC/AHA apresentou maior AUC e apenas os pacientes classificados com alto risco por esse escore estiveram associados com a mortalidade em 10 anos. Além disso, quando comparados os pontos de corte desses escores para nossa população, que já apresentou evento coronariano, o ERF apresentou melhor sensibilidade enquanto o ER ACC/AHA apresentou melhor especificidade na predição da mortalidade. Logo, sugerimos que os mesmos podem ser aplicados em conjunto para a predição de mortalidade nos pacientes com SCA prévia.
Introduction: The cardiovascular risk scores were developed, mostly, in population apparently without cardiovascular disease (CVD), therefore its conclusions envolve primary prevention, they shall not be applied to patients with diagnosed coronary disease. In addition, the prognostic scores used after acute coronary syndromes (ACS) evaluated the mortality only in a shortterm period. Goals: The goal of the current study is to evaluate the performance of the Framingham Risk Score (FRS) and the score proposed by the American College of Cardiology and the American Heart Association (ACC/AHA) in forecasting the mortality in patients ten years after ACS. Case studies and methods: It is a retrospective cohort study which includes patients in the age of ≥ 18 years old, with ACS, who were admitted to the Coronary Care Unit (CCU) of the Hospital das Clínicas of Botucatu Medical School – UNESP, in the period from January/2005 to December/2006, and that presented the exams for the calculation of the cardiovascular risk scores. Patients which their final outcome in ten years weren´t obtained or those who weren´t accepted in the study were eliminated. The FRS and ACC/AHA risk score were calculated and the mortality was evaluted ten years after ACS. Results: Four hundred and fifty-one patients with ACS admitted to the Coronary Care Unit were evaluated. One hundred and twenty-two were eliminated from those patients because the 10-year mortality after CCU admission wasn´t obtained. Hence, 329 patients in the average age of 62,913,0 years old were studied. Among those patients, 58,4% were men and 44,4% died in the period of 10 years after admission. Sixty-three per cent of those patients showed ACS without ST-segment elevation while thirtyseven per cent showed MI with ST-segment elevation. When we compared those two ACS groups, the mortality was similar (p=0,97). The average of the FRS was 16 (14-18) %, and the average of the ACC/AHA risk score was 18,5 (9,1-31,6). The patients who progressed to death were older, and showed higher values of creatinine, urea and potassium, and lower values of the total cholesterol in the CCU admission. Furthermore, it´s interesting to note that the patients who progressed to death showed higher values of scores. However, when we classify the patients in high cardiovascular risk, only the ACC/AHA risk score was associated to the mortality. The area under the curve (AUC) from the association of FRS and the 10-year mortality is 0,6307, with IC95%: 0,5708-0,6905, and the value of p <0,001. The FRS cut-off point associated to higher mortality is > 14,3% with sensibility of 74,66%, specificity of 42,62 %, positive predictive value of 50,93 % and negative predictive value of 67,83 %. In relation to the ACC/AHA risk score is 0,7015, with IC95% 0,6455-0,7576 and value of p <0,001. The score cut-off point which is associated to higher mortality is > 23,56%, with sensibility of 50,00%, specificity of 71,04%, positive predictive value of 57,94% and negative predictive value of 64,04%. In the logistic regression analysis both the FRS (OR: 1,127; IC95%: 1,056-1,204; p<0,001) and the ACC/AHA risk score (OR: 1,043; IC95%: 1,025-1,060; p<0,001) were associated to the mortality in 10 years in patients with ACS, after being adjusted by the creatinine and potassium. Conclusion: Both the FRS and the ACC/AHA risk score were associated to the 10-year mortality. However, the ACC/AHA risk score showed higher AUC and only the patients who were classified in high risk by that score were associated to the 10-year mortality. In addition, when the scores cut-off points for our population, the one who showed coronary occurences, were compared, the FRS showed better sensibility while the ACC/AHA risk score showed better specificity in the mortality prediction. Hence, we suggested that they can be applied concurrently for the mortality prediction in patients with previous ACS.
Introduction: The cardiovascular risk scores were developed, mostly, in population apparently without cardiovascular disease (CVD), therefore its conclusions envolve primary prevention, they shall not be applied to patients with diagnosed coronary disease. In addition, the prognostic scores used after acute coronary syndromes (ACS) evaluated the mortality only in a shortterm period. Goals: The goal of the current study is to evaluate the performance of the Framingham Risk Score (FRS) and the score proposed by the American College of Cardiology and the American Heart Association (ACC/AHA) in forecasting the mortality in patients ten years after ACS. Case studies and methods: It is a retrospective cohort study which includes patients in the age of ≥ 18 years old, with ACS, who were admitted to the Coronary Care Unit (CCU) of the Hospital das Clínicas of Botucatu Medical School – UNESP, in the period from January/2005 to December/2006, and that presented the exams for the calculation of the cardiovascular risk scores. Patients which their final outcome in ten years weren´t obtained or those who weren´t accepted in the study were eliminated. The FRS and ACC/AHA risk score were calculated and the mortality was evaluted ten years after ACS. Results: Four hundred and fifty-one patients with ACS admitted to the Coronary Care Unit were evaluated. One hundred and twenty-two were eliminated from those patients because the 10-year mortality after CCU admission wasn´t obtained. Hence, 329 patients in the average age of 62,913,0 years old were studied. Among those patients, 58,4% were men and 44,4% died in the period of 10 years after admission. Sixty-three per cent of those patients showed ACS without ST-segment elevation while thirtyseven per cent showed MI with ST-segment elevation. When we compared those two ACS groups, the mortality was similar (p=0,97). The average of the FRS was 16 (14-18) %, and the average of the ACC/AHA risk score was 18,5 (9,1-31,6). The patients who progressed to death were older, and showed higher values of creatinine, urea and potassium, and lower values of the total cholesterol in the CCU admission. Furthermore, it´s interesting to note that the patients who progressed to death showed higher values of scores. However, when we classify the patients in high cardiovascular risk, only the ACC/AHA risk score was associated to the mortality. The area under the curve (AUC) from the association of FRS and the 10-year mortality is 0,6307, with IC95%: 0,5708-0,6905, and the value of p <0,001. The FRS cut-off point associated to higher mortality is > 14,3% with sensibility of 74,66%, specificity of 42,62 %, positive predictive value of 50,93 % and negative predictive value of 67,83 %. In relation to the ACC/AHA risk score is 0,7015, with IC95% 0,6455-0,7576 and value of p <0,001. The score cut-off point which is associated to higher mortality is > 23,56%, with sensibility of 50,00%, specificity of 71,04%, positive predictive value of 57,94% and negative predictive value of 64,04%. In the logistic regression analysis both the FRS (OR: 1,127; IC95%: 1,056-1,204; p<0,001) and the ACC/AHA risk score (OR: 1,043; IC95%: 1,025-1,060; p<0,001) were associated to the mortality in 10 years in patients with ACS, after being adjusted by the creatinine and potassium. Conclusion: Both the FRS and the ACC/AHA risk score were associated to the 10-year mortality. However, the ACC/AHA risk score showed higher AUC and only the patients who were classified in high risk by that score were associated to the 10-year mortality. In addition, when the scores cut-off points for our population, the one who showed coronary occurences, were compared, the FRS showed better sensibility while the ACC/AHA risk score showed better specificity in the mortality prediction. Hence, we suggested that they can be applied concurrently for the mortality prediction in patients with previous ACS.
Descrição
Palavras-chave
Síndrome Coronariana Aguda, Escore de Risco Cardiovascular de Framingham, Escore de Risco Cardiovascular do American College of Cardiology e da American Heart Association (ACC/AHA), Mortalidade