Influência de fatores clínicos e operacionais na reprodutibilidade das tomografias de coerência óptica para diagnóstico e seguimento de glaucoma
Data
2023
Autores
Masera, Mariana Costa 
Orientadores
Tavares, Ivan Maynart
Tipo
Dissertação de mestrado
Título da Revista
ISSN da Revista
Título de Volume
Resumo
Objetivo: O objetivo do presente estudo foi avaliar e comparar a reprodutibilidade das medidas de espessura da camada de fibra nervosa da retina (CFNR), obtidas pelos tomógrafos de coerência óptica (OCT) Spectralis (Heidelberg Engineering, Heidelberg, Alemanha) e o RTVue (Optovue Inc, Fremont, EUA), em pacientes saudáveis, suspeitos e com glaucoma. Avaliou-se também a influência da gravidade da doença, idade, área do disco óptico e operador sobre a reprodutibilidade das medidas de espessura da CFNR. Metodologia: Cento e um olhos de 57 pacientes foram incluídos neste estudo transversal. Três varreduras consecutivas de espessura da CFNR foram obtidas de cada participante, por dois operadores diferentes. A reprodutibilidade foi avaliada utilizando-se coeficiente de correlação intraclasse (ICC) e coeficiente de variação (CV). Para avaliar a influência da área do disco, da gravidade da doença, idade e operador na reprodutibilidade dos instrumentos, foi construído um modelo de regressão univariada, utilizando a diferença entre a maior e menor medida, por olho, como variável dependente. Resultados: Os parâmetros com maior ICC foram a espessura média da CFNR tanto para o RTVue (ICC=0,97) quanto para o Spectralis (ICC=0,99). O setor global também apresentou o menor CV para o RTVue (CV=2,11%) e para o Spectralis (CV=0,87%). Para o RTVue, a análise univariada não mostrou influência da idade (P=0,56), área do disco (P=0,24), gravidade da doença (P=0,69) e operador (P=0,08), na reprodutibilidade da espessura global da CFNR. Para o Spectralis, a análise univariada não mostrou influência da idade (P=0,47), área do disco (P=0,20), gravidade da doença (P=0,12) e operador (P=0,85) na reprodutibilidade da espessura global da CFNR. Conclusão: Ambos os instrumentos apresentaram excelente reprodutibilidade, que não foi afetada por fatores clínicos ou operacionais. Portanto, conclui-se que a reprodutibilidade do instrumento é robusta em diferentes cenários clínicos.
Objective: The aim of the present study was to evaluate and compare the reproducibility of measurements of retinal nerve fiber layer (RNFL) thickness obtained by Spectralis optical coherence tomography (OCT) (Heidelberg Engineering, Heidelberg, Germany) and RTVue (Optovue Inc, Fremont, USA) in healthy and glaucoma patients. The influence of disease severity, age, optic disc area and operator on the reproducibility of RNFL thickness measurements was also evaluated. Methodology: One hundred and one eyes of 57 patients were included in this crosssectional study. Three consecutive RNFL thickness scans were obtained from each participant by two different operators. Reproducibility was evaluated using intraclass correlation coefficient (ICC) and coefficient of variation (CV). To assess the influence of disc area, disease severity, age and operator on the reproducibility of the instruments, a univariate regression model was built, using the difference between the highest and lowest measurement, per eye, as the dependent variable. Results: The parameters with the highest ICC were the mean RNFL thickness for both RTVue (ICC=0.97) and Spectralis (ICC=0.99). The global sector also had the lowest CV for RTVue (CV=2.11%) and Spectralis (CV=0.87%). For RTVue, univariate analysis showed no influence of age (P=0.56), disc area (P=0.24), disease severity (P=0.69) and operator (P=0.08), on the reproducibility of the global RNFL thickness. For Spectralis, univariate analysis showed no influence of age (P=0.47), disc area (P=0.20), disease severity (P=0.12) and operator (P=0.85) on the reproducibility of the global RNFL thickness. Conclusion: Both instruments showed excellent reproducibility. Clinical and operational factors did not affect its reproducibility. Therefore, it is concluded that the reproducibility of the instrument is robust in different clinical scenarios.
Objective: The aim of the present study was to evaluate and compare the reproducibility of measurements of retinal nerve fiber layer (RNFL) thickness obtained by Spectralis optical coherence tomography (OCT) (Heidelberg Engineering, Heidelberg, Germany) and RTVue (Optovue Inc, Fremont, USA) in healthy and glaucoma patients. The influence of disease severity, age, optic disc area and operator on the reproducibility of RNFL thickness measurements was also evaluated. Methodology: One hundred and one eyes of 57 patients were included in this crosssectional study. Three consecutive RNFL thickness scans were obtained from each participant by two different operators. Reproducibility was evaluated using intraclass correlation coefficient (ICC) and coefficient of variation (CV). To assess the influence of disc area, disease severity, age and operator on the reproducibility of the instruments, a univariate regression model was built, using the difference between the highest and lowest measurement, per eye, as the dependent variable. Results: The parameters with the highest ICC were the mean RNFL thickness for both RTVue (ICC=0.97) and Spectralis (ICC=0.99). The global sector also had the lowest CV for RTVue (CV=2.11%) and Spectralis (CV=0.87%). For RTVue, univariate analysis showed no influence of age (P=0.56), disc area (P=0.24), disease severity (P=0.69) and operator (P=0.08), on the reproducibility of the global RNFL thickness. For Spectralis, univariate analysis showed no influence of age (P=0.47), disc area (P=0.20), disease severity (P=0.12) and operator (P=0.85) on the reproducibility of the global RNFL thickness. Conclusion: Both instruments showed excellent reproducibility. Clinical and operational factors did not affect its reproducibility. Therefore, it is concluded that the reproducibility of the instrument is robust in different clinical scenarios.