Avaliação do ROPScore como preditor de retinopatia da prematuridade em neonatos prematuros de muito baixo peso: estudo coorte
Carregando...
Arquivos
Data
2017-02-03
Autores
Lucio, Kellen Cristiane do Vale [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Introdução e Objetivo: A Retinopatia da prematuridade (ROP) é a principal causa de deficiência visual permanente na infância. A identificação precoce da forma grave da ROP pode prevenir a cegueira. O ROPScore é um sistema de pontuação desenvolvido para otimizar a triagem e calcular o risco de ROP em neonatos pré-termo. Este estudo objetivou avaliar a acurácia deste algoritmo como preditor de ROP em uma coorte brasileira. Métodos: Realizou-se um estudo coorte prospectivo de 220 neonatos pré-temo com PN ≤1500 g e / ou IG ≤ 32 semanas. O ROPScore foi aplicado na 6ª semana de vida nas 181 crianças que sobreviveram até a 45ª semana de idade gestacional corrigida. Cada criança foi classificada como Sem-ROP, ROP em qualquer estágio ou ROP grave. Curvas ROC foram utilizadas para determinar os melhores valores de sensibilidade e especificidade do escore na população estudada. Resultados: A média de PN foi de 1271,6g ± 354,6 e a IG média foi de 29,2 ± 2,26 semanas. Dos 181 prematuros estudados, 32 (17,6%) desenvolveram ROP. O melhor ponto de corte do ROPscore para sensibilidade e especificidade foi 16,0 para ROP em qualquer estágio e 16,6 para ROP grave. A área sob a curva ROC para prever ROP em qualquer estágio foi 0,937 (IC 95%: 0,888-0,986; P <0,0001), e para prever ROP grave, 0,962 (IC 95% 0,931-0,993; P <0,0001). A sensibilidade para ROP em qualquer estágio foi 87,5% e para ROP grave foi 95,4%. Utilizando o ROPscore, 130 pré-termos não precisariam ser avaliados com a mesma frequência, diminuindo em 71,8% o número total de exames necessários para detectar ROP. Conclusão: O ROPScore foi um método auxiliar útil para a triagem de ROP, otimizou os exames e teve precisão para identificar ROP grave.
Background and Objective: Retinopathy of prematurity (ROP) is the leading cause of childhood permanent visual impairment. Early identification of severe form of ROP can prevent blindness. The ROPScore is a scoring system, developed to optimize the screening and calculate risk of ROP in preterm infants. This study aimed to evaluate the accuracy of this algorithm as predictor of ROP in a Brazilian cohort. Methods: A prospective cohort of 220 preterm infants with BW ≤1500 g and/or GA ≤32 weeks was conducted. The ROPScore was applied in the 6th week of life in 181 infants that survived who survived until the 45th week of corrected gestational age. Each infant was categorized as having no-ROP, any stage ROP or severe ROP. Receivers operating characteristic (ROC) curves were used to determine the best sensitivity and specificity values of the score in the population studied. Results: Mean BW was 1271,6 g ± 354.6 and the mean GA was 29,2 ± 2,2 weeks. Of 181 studied preterm, 32 (17,6%) developed ROP. The best cutoff point for sensitivity and specificity was established as 16.0 for any stage ROP and 16.6 for severe ROP. The area under the ROC curve to predict ROP at any stage was 0,937 (95% CI: 0,888 to 0,986; P <0,0001), and to predict severe ROP, 0,962 (95% CI 0,931 to 0,993; P <0,0001). The sensitivity for any stage ROP was 87,5% and for severe ROP was 95,4%. Using ROPscore, 130 pre-terms did not need to be evaluated at the same frequency, reducing the total number of examinations required to detect ROP by 71,8%. Conclusion: The ROPScore was a useful adjunct screening tool for ROP, optimized the examinations and had accuracy for identify severe ROP.
Background and Objective: Retinopathy of prematurity (ROP) is the leading cause of childhood permanent visual impairment. Early identification of severe form of ROP can prevent blindness. The ROPScore is a scoring system, developed to optimize the screening and calculate risk of ROP in preterm infants. This study aimed to evaluate the accuracy of this algorithm as predictor of ROP in a Brazilian cohort. Methods: A prospective cohort of 220 preterm infants with BW ≤1500 g and/or GA ≤32 weeks was conducted. The ROPScore was applied in the 6th week of life in 181 infants that survived who survived until the 45th week of corrected gestational age. Each infant was categorized as having no-ROP, any stage ROP or severe ROP. Receivers operating characteristic (ROC) curves were used to determine the best sensitivity and specificity values of the score in the population studied. Results: Mean BW was 1271,6 g ± 354.6 and the mean GA was 29,2 ± 2,2 weeks. Of 181 studied preterm, 32 (17,6%) developed ROP. The best cutoff point for sensitivity and specificity was established as 16.0 for any stage ROP and 16.6 for severe ROP. The area under the ROC curve to predict ROP at any stage was 0,937 (95% CI: 0,888 to 0,986; P <0,0001), and to predict severe ROP, 0,962 (95% CI 0,931 to 0,993; P <0,0001). The sensitivity for any stage ROP was 87,5% and for severe ROP was 95,4%. Using ROPscore, 130 pre-terms did not need to be evaluated at the same frequency, reducing the total number of examinations required to detect ROP by 71,8%. Conclusion: The ROPScore was a useful adjunct screening tool for ROP, optimized the examinations and had accuracy for identify severe ROP.
Descrição
Palavras-chave
Retinopatia da prematuridade, Fatores de risco, Escore de predição, ROPScore, Retinopathy of prematurity, Risk factors, Prediction score