Introduction: The early evaluation of the spine in children is desirable because it is at this stage of development that the greatest changes in the body structures occur. Objective: To determine the test-retest, intra- and inter-rater reliability of the Flexicurve instrument for the evaluation of spinal curvatures in children. Method: Forty children ranging from 5 to 15 years of age were evaluated by two independent evaluators using the Flexicurve to model the spine. The agreement was evaluated using Intraclass Correlation Coefficients (ICC), Standard Error of the Measurement (SEM), and Minimal Detectable Change (MDC). Results: In relation to thoracic kyphosis, the Flexicurve was shown to have excellent correlation in terms of test-retest reliability (ICC2,2=0.87) and moderate correlation in terms of intra-(ICC2,2=0.68) and inter-rater reliability (ICC2,2=0.72). In relation to lumbar lordosis, it was shown to have moderate correlation in terms of test-retest reliability (ICC2,2=0.66) and intra- (ICC2,2=0.50) and inter-rater reliability (ICC=0.56). Conclusion: This evaluation of the reliability of the Flexicurve allows its use in school screening. However, to monitor spinal curvatures in the sagittal plane in children, complementary clinical measures are necessary. Further studies are required to investigate the concurrent validity of the instrument in order to identify its diagnostic capacity.

BACKGROUND: The adapted arcometer has been validated for use in adults. However, its suitability for use in children can be questioned given the structural differences present in these populations. OBJECTIVE: To verify the concurrent validity, repeatability, and intra- and inter-reproducibility of the adapted arcometer for the measurement of the angles of thoracic kyphosis and lumbar lordosis in children. METHOD: Forty children were evaluated using both sagittal radiography of the spine and the adapted arcometer. The evaluations using the arcometer were carried out by two trained evaluators on two different days. In the statistical treatment, the intraclass correlation coefficient (ICC), Pearson's product moment correlation, Spearman's rho, the paired t test, and Wilcoxon's test were used (α=.05). RESULTS: A moderate and significant correlation was found between the x-ray and the adapted arcometer regarding thoracic kyphosis, but no correlation was found regarding lumbar lordosis. Repeatability and intra-evaluator reproducibility of the thoracic kyphosis and lumbar lordosis were confirmed, which was not the case of inter-evaluator reproducibility. CONCLUSION: The adapted arcometer can be used to accompany postural alterations in children made by the same evaluator, while its use for diagnostic purposes and continued evaluation by different evaluators cannot be recommended. Further studies with the aim of adapting this instrument for use in children are recommended.