Validity and reliability of surface measurement of the thoracic khyphosis and lumbar lordosis in age of growth with a pocket inclinometer

Introduction: The gold standard in the measurement of sagittal curves of the spine is the radiographic Cobb method. Over the last 30 years no-invasive approaches have been developed for the study of sagittal spine deformity. The aim of the present study is to evaluate the reliability and validity of a heavy pointer pocket-size inclinometer, IncliMed, to measure the surface thoracic kyphosis and lumbar lordosis in adolescent with no curvatures of the spine or with idiopathic scoliosis, Scheuermann’s kyphosis or postural kyphosis. Material and methods: In a first group of 128 patients, the sagittal curves of the spine were measured according to the IncliMed technique and these values compared with those obtained by the same examiner according to the Cobb radiological technique. In a second group of 139 patients measurements of the surface thoracic kyphosis and lumbar lordosis were taken on the same day for each subject by two examiners (P/S); in 30 subjects measurements were repeated by examiners according to the IncliMed technique at the distance of 28 days, as a mean. The linear regression analysis and the Bland and Altman criteria were used for the data analysis (p<0.05). Results: The linear regression analysis in the measurement of kyphosis showed a strong correlation and overlap of radiological and IncliMed technique, because the linear regression coefficient almost equal to 1 (b=0.897;p=0.000). The mean relative difference between measurements (-0.3°) confirmed the absence of systematic difference techniques (p=0.58). For lumbar lordosis, the results were not satisfactory (b=0.526), so we introduced a variable of the original radiological measurement with improved the linear regression coefficient (b=0.820, p=0.000). Both for the kyphosis and the lordosis , the variability interobserver (PvsS) in the surface measurement with the IncliMed method was ± 11°, according to the repeatability coefficient. The mean difference of measurements was 0° for kyphosis and -0.4° for lordosis, excluding the presence of a systematic error between P and S. The intraobserver variability (P vs. P, S vs. S) in the measurement of kyphosis and lordosis was, according to the coefficient of repeatability, respectively ± 11° and ± 12° to the observer P, and ± 9° and ± 11° to the observer S. The mean differences measures were near 0°. Conclusions: although the Cobb technique remains the "gold standard" for evaluation of sagittal curves of the spine, IncliMed is an excellent complement, representing an opportunity to reduce the use of radiological examinations in subjects during the age of growth.