The efficacy of sagittal cervical spine subtyping: Investigating radiological classification methods within 150...

The aim of this study is to (1) compare and contrast cervical subtype classification methods within an asymptomatic population, and (2) identify inter-methodological consistencies and describe examples of inconsistencies that have the potential to affect subtype classification and clinical decision-making.

A total of 150 asymptomatic 18-30-year-old participants met the strict inclusion criteria. An erect neutral lateral radiograph was obtained using standard procedures. The Centroid, modified Takeshima/Herbst methods and the relative rotation angles in cases of nonagreement were used to determine subtype classifications. Cohen's kappa coefficient (κ) was used to assess the level of agreement between the two methods.

Nonlordotic classifications represented 66% of the cohort. Subtype classification identified the cohort as, lordosis (51), straight (37), global kyphosis (30), sigmoidal (13), and reverse sigmoidal (RS) (19). Cohen's kappa coefficient indicated that there was only a moderate level of agreement between methods (κ = 0.531). Methodological agreement tended to be higher within the lordotic and global kyphotic subtypes whereas, straight, sigmoidal, and RS subtypes demonstrated less agreement.

This is the first study of its type to compare and contrast cervical classification methods. Subtypes displaying predominantly extended or flexed segments demonstrated higher levels of agreement. Our findings highlight the need for establishing a standardized multi-method approach to classify sagittal cervical subtypes.