Background:

Determining which spinal levels to instrument during surgical treatment of Lenke Type 1 adolescent idiopathic scoliosis (AIS) depends on the reducibility of the primary and secondary curve patterns. This reducibility can be evaluated in several ways, with the most popular being radiographs in bending for moderate thoracic and lumbar curvatures.

Hypothesis:

Side-bending radiographs will alter the choice of the lowest instrumented vertebra (LIV) for the surgical treatment of AIS.

Methods:

Thirteen experienced French spine surgeons were invited to perform surgical planning on 23 patients based on stereoradiographs with and without (standing) side-bending views. The surgical planning was repeated a second time to assess the intra- and inter-rater reliability. Variations in the choice of LIV were analyzed for each evaluation.

Results:

The intra-rater reliability was moderate to substantial. The inter-rater reliability was low to moderate. The study compared 879 surgical plans. Selective fusion was chosen in 0.3% of the plans. The median LIV was L2. The availability of side bending views changed the plan in 39% of cases. However, 36% of the plans were changed in the control (test-retest) condition. No significant difference was found between the variations with side-bending radiographs and “control” variations (p > 0.05).

Conclusion:

The use of radiographs in bending has no significant effect on the LIV choice in this study. This result is derived from statistically robust analysis made possible by one of the largest datasets available on this topic. Large inter-rater variability was observed and will be explored further in a future study.

Level of evidence:

II; non-randomized controlled comparative study.