New EOS Imaging Protocol Allows a Substantial Reduction in Radiation Exposure for Scoliosis Patients.

STUDY

DESIGN

Prospective.

OBJECTIVE

To evaluate the reliability of three-dimensional (3D) spinal models from Micro Dose EOS x-rays compared to standard, Low Dose EOS x-rays utilized for evaluating patients with adolescent idiopathic scoliosis (AIS).

SUMMARY OF BACKGROUND DATA: There is a strong suggestion that radiation exposure to scoliosis patients can be further reduced.

METHODS

Sixty AIS patients who received biplanar, posteroanterior, and lateral standard Low Dose spine x-rays in our EOS imaging unit (∼0.33 mGy) as part of routine care also underwent an additional set of new reduced "Micro Dose" EOS x-rays (∼0.05 mGy) using a recently developed protocol. Two measurers created 3D models of the images using sterEOS software (Low Dose x2, Micro Dose x2). From this 3D modeling software, coronal Cobb angles, sagittal (T1-T12, T4-T12, L1-L5, L1-S1), and apical axial rotation measurements were obtained. Intraclass correlations (ICCs) and standard error of measurement (upper bound of 95% confidence interval) for the differences between Low Dose and Micro Dose measurements were compared. Interrater reliability was assessed on standard two-dimensional (2D) radiographic measurements.

RESULTS

The ICCs were rated as "substantial" to "almost perfect" for Low Dose 3D, Micro Dose 3D, and 2D measures (range 0.78-0.99). The calculated measurement error was not significantly different between groups except for intrarater error on 3D L1-L5 lordosis (2.9° Micro Dose vs. 2.2°, p = .04), interrater 3D rotation of the lumbar apex (2.6° Micro Dose vs. 1.7°, p = .03), and 2D T12-sacrum lordosis (4.6° Micro Dose vs. 3.4°, p = .04).

CONCLUSIONS

Although statistically significant differences in average measurement error were observed in lordosis and lumbar apex rotation, these differences are not believed to be clinically significant. The Micro Dose images have slightly less clarity qualitatively, yet the critical 2D and 3D measures of the curvature were reliably measured with error of measurement comparable to standard radiologic techniques.

LEVEL OF EVIDENCE: Level I, Diagnostic.