Study design: Retrospective review.
Objective: The aim of this study was to describe thoracic kyphosis (TK) in a normal asymptomatic population and to evaluate the association between TK magnitude and its shape.
Summary of background data: Understanding spinal anatomy requires a three-dimensional appreciation of the spine's shape, morphology, and proportions. The customary definition of TK is the angle between T4 and T12. However, little is known on the actual shape of TK in adults.
Methods: Asymptomatic volunteers were recruited; demographic data along with full-body standing radiographs were recorded. Radiographic data such as T1-12 and T4-12 angles were collected. Maximum TK and vertebral orientation/tilt were also collected, in addition to cumulative TK and Centered Kyphosis at T7. The cohort was stratified by T1-12 value (<40°, 40°-60°, and>60°) and comparisons and regressions were performed afterward.
Results: One hundred nineteen subjects were included (average age 50.8 yrs, 81 female). Mean T1-12 kyphosis was 49.5°, mean T4-12 kyphosis 41.5°, and mean maximum TK was 52.6°. T1 was the most anteriorly tilted vertebra, L1 the most posteriorly tilted; T7 was horizontal, independently of T1-12 value or age. Cumulative kyphosis analysis revealed that the apex of kyphosis was located at T6-T7. Regression analysis predicting the value and the percentage of T1-7 both yielded T1-12 as a predictor (Adj. r = 0.32, Adj. r = 0.13).
Conclusion: Changes in kyphosis distribution in an asymptomatic population suggest that TK is not a simple circle arc: with low TK, 2/3 of the kyphosis is located in the upper part and when TK increases, the distribution of kyphosis will be symmetric around T7. It is possible to predict the amount of kyphosis in the upper part using total kyphosis value. This could help estimate preoperative compensation and predict reciprocal change.
Level of evidence: 3.