Estudo da reprodutibilidade da ressonância magnética na avaliação biométrica do colo uterino na segunda metade da gestação e análise comparativa com ultrassonografia transvaginal

• Autor(a) principal: Brandão, Rosieny Souza [UNIFESP]
• Data de Publicação: 2010
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UNIFESP
• Texto Completo: http://repositorio.unifesp.br/handle/11600/9535
• topic: Ultrassonografia transvaginal; Gravidez; Imagem por ressonância magnética; Colo de útero; Magnetic resonance imaging; Transvaginal ultrasound; Pregnancy; Cervix uteri

Objective: Verify the reproducibility of magnetic resonance imaging to measure the cervical length through the analysis of intra and interobserver variability in the second half of pregnancy and compare the transvaginal ultrasound and magnetic resonance imaging in the assessment of cervical length. Methods: Crosssectional observational study that included 42 women with gestational ages between 20 and 39 weeks initially submitted to transvaginal ultrasound by a single observer to measure the cervical length. After the scan, patients underwent magnetic resonance imaging by independent observers in a double-blind for the analysis of intra and interobserver variability. Results: There was a distribution of measurements of the cervix measured by means of different methods by Gaussian distribution model (Komolgorov-Smirnov test Z). There was good correlation (intraclass correlation coefficient) for the measurements performed by magnetic resonance single investigator 0.995 (95% CI: 0.991 to 0.997, p <0.001) and by two different examiners 0.990 (95% CI: 0.995, p <0.001). In calculating the paired t test no significant difference between the mean measurements of the cervix for the analysis of intra and interobserver variability (p = 0.200 and p = 0.257, respectively) confirmed by Bland-Altman (-1.07% and 0.74%; respectively) which showed a small average difference percentage. No significant variation was observed between measurements taken by transvaginal ultrasound and magnetic resonance imaging in the paired t test (p=0.191) and Bland-Altman (-0.17%). Conclusions: The results indicate good reproducibility of magnetic resonance imaging to measure the length cervical and no significant difference between transvaginal ultrasound and magnetic resonance imaging in the estimation of cervical biometry.