Confiabilidade de dois métodos radiográficos para determinação da maturação esquelética por meio de imagens das vértebras cervicais e concordância com imagens de mão e punho
Carregando...
Data
2018-05-17
Autores
Lima, Juliana Ferreira
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O objetivo deste estudo foi avaliar a confiabilidade de dois métodos radiográficos de determinação da maturação esquelética das vértebras cervicais e a concordância com o método baseado nas imagens radiográficas de mão e punho. Foram analisadas imagens radiográficas digitais cefalométricas laterais e de mão e punho de 120 indivíduos, com idades variando entre 6 e 18 anos. Para o estudo da repetibilidade e reprodutibilidade, as imagens radiográficas foram interpretadas por 4 examinadores (2 radiologistas e 2 ortodontistas) em dois momentos distintos de acordo com os métodos de Hassel e Farman (VC1) e Baccetti et al. (VC2). Para o estudo de concordância, o método de Fishman também foi realizado pelos quatro examinadores. Para as análises estatísticas foi utilizada estatística Kappa (ҡ), com ponderação linear, seguida do cálculo dos intervalos de confiança (IC95%) e Coeficiente de Correlação Intraclasse (CCI). Os resultados da análise intra-examinador indicaram calibração “quase perfeita” para os quatro examinadores (ҡ>0,81) em todos os métodos de análise. Os limites inferiores dos IC estiveram todos acima de 0,71 para a análise inter-examinador. Os valores de CCI para reprodutibilidade inter-examinador foram considerados “excelentes” para os limites inferiores de todos os IC (≥0,91). Embora a proporção total de diagnósticos concordantes (acurácia) obtida nos métodos VC1 e VC2 tenha sido, respectivamente, 42% e 50% para os ortodontistas e 42% e 40% para os radiologistas, o cruzamento de todos os IC95% não mostrou diferença significativa. Concluímos que os dois métodos de determinação da maturação esquelética das vértebras cervicais são confiáveis, pois apresentaram excelente repetitividade e reprodutibilidade. Estes dois métodos apresentaram concordância substancial com o método que utiliza imagens radiográficas da mão e punho e, além disso, ortodontistas e radiologistas os interpretaram com a mesma precisão.
The objective of this study was to evaluate the reliability of two radiographic methods to determine the skeletal maturation of the cervical vertebrae and the concordance with the method based on hand and wrist radiographic images. Were analyzed lateral cephalogram, hand and wrist digital radiographic images of 120 individuals, ranging from 6 to 18 years. For the study of repeatability and reproducibility, the images were interpreted by 4 examiners (2 radiologists and 2 orthodontists) on distinct moments, according to the methods of Hassel e Farman (CV1) and Bacetti et al. (CV2). For the concordance study, the Fishman method was also applied. Statistical analysis was performed using Kappa (ҡ), with linear weighting, followed by the calculation of confidence intermission (CI95%) and the Intraclass Correlation Coefficient (ICC). The intra-examiner analysis results indicated ‘almost perfect’ calibration for the examiners (ҡ>0,81) in all methods. The lower limits of Cis were all above 0.71 for the inter-examiner analysis. ICC values for inter-examiner reproducibility were considered ‘excellent’ for the lower limits of all Cis (≥0.91). Although the total proportion of concordant diagnosis (accuracy) obtained on CV1 and CV2 methods was 42% and 50% for orthodontists and 42% and 40% for radiologists, the crossing of all Cis95% showed no significant difference. Both methods of determination of skeletal maturation of the cervical vertebrae are reliable, since they presented excellent repeatability and reproducibility. These methods showed substantial concordance with the method using radiographic images of the hand and wrist. Moreover, orthodontists and radiologists interpreted them with the same precision
The objective of this study was to evaluate the reliability of two radiographic methods to determine the skeletal maturation of the cervical vertebrae and the concordance with the method based on hand and wrist radiographic images. Were analyzed lateral cephalogram, hand and wrist digital radiographic images of 120 individuals, ranging from 6 to 18 years. For the study of repeatability and reproducibility, the images were interpreted by 4 examiners (2 radiologists and 2 orthodontists) on distinct moments, according to the methods of Hassel e Farman (CV1) and Bacetti et al. (CV2). For the concordance study, the Fishman method was also applied. Statistical analysis was performed using Kappa (ҡ), with linear weighting, followed by the calculation of confidence intermission (CI95%) and the Intraclass Correlation Coefficient (ICC). The intra-examiner analysis results indicated ‘almost perfect’ calibration for the examiners (ҡ>0,81) in all methods. The lower limits of Cis were all above 0.71 for the inter-examiner analysis. ICC values for inter-examiner reproducibility were considered ‘excellent’ for the lower limits of all Cis (≥0.91). Although the total proportion of concordant diagnosis (accuracy) obtained on CV1 and CV2 methods was 42% and 50% for orthodontists and 42% and 40% for radiologists, the crossing of all Cis95% showed no significant difference. Both methods of determination of skeletal maturation of the cervical vertebrae are reliable, since they presented excellent repeatability and reproducibility. These methods showed substantial concordance with the method using radiographic images of the hand and wrist. Moreover, orthodontists and radiologists interpreted them with the same precision
Descrição
Palavras-chave
Intensificação de imagem radiográfica, Determinação da idade pelo esqueleto, Ossos da mão, Vértebras cervicais, Digital radiography, Skeletal age determin, Hand bones