Devoted to diagnostic and interventional spine imaging and therapeutics


Thoracic Kyphosis in the Absence of Vertebral Fracture: Age Related changes and the Normal Development of Extreme Curvature 2004

Bartynski, Walter, M.D., Heller, Matthew, M.D., Grahovac, Stephen, M.D., Rothfus, William, M.D.

While it is known that the thoracic kyphotic angle increases with age, the reason for this change is not understood.(1) Recently it has been shown that the thoracic kyphotic angle is not only greater in the elderly (>65 year old) but demonstrates a bimodal distribution in both men and women suggesting a population at risk for vertebral compression fracture. (2) The purpose of this project was to evaluate the thoracic kyphotic angle in young, mid aged and elder men and women without vertebral compression to assess the age related changes in thoracic kyphosis and determine at what point the bimodal features develop.

Materials and Methods:
Thoracic kyphotic angle was measure d by Cobb angle on standing lateral erect chest x-rays in 221 patients. Four age ranges were separately evaluated to include: 18-35 year old (63 pts), 36-50 year old (57 pts), 51-65 year old (49 pts) and >65 year old (52 pts). Only patients with normal appearing spines were included and patients with thoracic scoliosis, vertebral compression, vertebral body angulation or congenital anomaly were excluded. Results were tabulated, graphically plotted and analyzed.

The average thoracic kyphotic angle in the four groups increased with advancing age to include: 18-35 yr - 27 deg; 36-50 yr - 31 deg; 51-65 yr - 36 deg and > 65 yr - 41 deg. While both men and women demonstrated an increased kyphotic angle with age, women developed a greater increase in kyphotic angle and at an earlier age than men.

In the 18-35 yr and 36-50 yr groups, men and women demonstrated a similar average kyphotic angle.
In the 51-65 yr group, men and women separated with a significantly greater average kyphotic angle in women (38.5 deg) vs. men (33.5 deg).

In the elder group (>65 yr) a further increase in kyphotic angle for women (45.5 deg) and men (38.9 deg) was noted. In addition, the elder population demonstrated a clear bimodal (non-standard) distribution with a lower mode of 29 deg (similar to the average kyphotic angle in young patients) and an upper mode of 50 deg.

In men, the distribution began as standard, progressively shifted higher slightly with age and became clearly bimodal in the >65 year age group (p<0.05) with a lower mode of 29 deg and an upper mode of 48 deg.

Women were surprisingly bimodal in the initial younger 18-35 age group (p<0.01), developed a mixed normal distribution like pattern in the 36-50 and 51-65 age ranges and then became bimodal again (p<0.05) in the >65 year age range with a lower mode of 29 deg and an upper mode of 54 deg.

The thoracic kyphotic angle increases with age but not symmetrically and significant differences exist between men and women. In men, a normal distribution of the kyphotic angle is present in the young but shifts with age to become bimodal in the elderly. In women the distribution of the kyphotic angle begins as bimodal in the young, becomes symmetric and greater than in men in late middle age and re-emerges as bimodal in the elderly.
Considering the biomechanical effects of severe thoracic kyphosis, this bimodal distribution in the elderly suggests a population at risk for vertebral compression fracture and could help explain the earlier and greater incidence of vertebral compression in women. Extreme thoracic kyphosis in the absence of vertebral compression must also be taken into consideration when planning treatment of vertebral compression fractures with either vertebroplasty or kyphoplasty techniques.


1. Fon GT et al. AJR Am J Radiology 1980; 134:979-983
2. Heller M, Bartynski WS et al, RSNA 2003 Chicago, Il