Can We Screen for Limb Length Discrepancy on Spinal Radiographs of Patients With Adolescent Idiopathic Scoliosis?

Background: The aim of this study was to determine whether height differences in the levels of the iliac crests and femoral heads on erect spinal radiographs can be used as indirect measurements for the screening and surveillance of limb length discrepancy (LLD) in patients with scoliosis.

Methods: Whole body posteroanterior (PA) and lateral erect radiographs of patients with adolescent idiopathic scoliosis (AIS) were retrospectively reviewed. Patients with congenital, syndromic, and neuromuscular scoliosis were excluded. A direct measurement of each limb was taken from the highest point of the femoral head to the middle of the tibial plafond; any difference between the sides was recorded as the LLD. In addition, the PACS Software tool was used to measure femoral head height difference (FHHD) and iliac crest height difference (ICHD). Pearson's correlation, linear regression, and Bland-Altman plots were used to determine the relationships between LLD and FHHD, and LLD and ICHD.

Results: Radiographs of 141 patients (92 women, 49 men) with an average age of 12.0±2.65 years were analyzed. A strong correlation (r=0.730, P <0.001) was found between LLD and FHHD; the correlation between LLD and ICHD was weaker (r=0.585, P <0.001). The Bland-Altman analysis showed good agreements of LLD with FHHD and ICHD. Linear regression analysis predicted an LLD of ≤10 mm based on an FHHD of ≤11.5 mm or an ICHD of ≤15.3 mm.

Conclusions: FHHD and ICHD on spinal PA radiographs can be used for the screening and monitoring of LLD in patients with AIS with FHHD being the preferred indirect measurement. These measurements are readily learned and quick to perform. The FHHD and ICHD can be measured on any erect scoliosis PA radiograph. Therefore, these proxy measurements can be used to screen and monitor for LLD in patients with AIS.

Level of evidence: Level III.