Methodological evaluation of dual-energy X-ray absorptiometry for Cobb angle measurement in females with idiopathic scoliosis: a reliability and validity study in Thailand.

Abstract

Comparative study. To evaluate the concurrent validity, test-retest reliability, and inter-rater reliability of Cobb measurement in the coronal plane via dual-energy X-ray absorptiometry (DEXA) images and plain radiographs in female patients with idiopathic scoliosis (IS). IS, which affects primarily females, is commonly monitored at least biannually with spine radiographs according to the Cobb method. DEXA is a safer imaging method because it involves less radiation exposure. Although DEXA has potential for assessing spinal alignment, its validity and reliability in measuring Cobb angles require further verification. A repeated-measures design was used to evaluate the test-retest reliability of measuring spinal alignment with DEXA. Eighty-six women aged 18-20 years with a diagnosis of IS, who had undergone spinal radiography within the previous 3 months, underwent two DEXA scans (DEXA1 and DEXA2) 1 week apart. Cobb angles were measured on radiographs, DEXA1, and DEXA2 by two independent assessors. Intraclass correlation coefficients (ICCs) were calculated to assess test-retest reliability and inter-rater reliability. Concurrent validity was examined using Pearson's correlation coefficients between Cobb angles obtained from radiographs and those from DEXA1 and DEXA2 images. Cobb angle measurements from DEXA images had excellent test-retest and nter-rater reliability (ICC>0.90) and very strong concurrent validity with radiographs (r >0.90, p <0.001). The mean difference in Cobb angles between radiographs and DEXA images ranged from 2.37° to 2.91°, whereby the curves were less severe on DEXA images. DEXA has consistent reliability and validity for evaluating spinal alignment and is potentially useful in monitoring curve progression in young populations.