Correlation of cervical endplate strength with CT measured subchondral bone density

Abstract

Cervical interbody device subsidence can result in screw breakage, plate dislodgement, and/or kyphosis. Preoperative bone density measurement may be helpful in predicting the complications associated with anterior cervical surgery. This is especially important when a motion preserving device is implanted given the detrimental effect of subsidence on the postoperative segmental motion following disc replacement. To evaluate the structural properties of the cervical endplate and examine the correlation with CT measured trabecular bone density. Eight fresh human cadaver cervical spines (C2-T1) were CT scanned and the average trabecular bone densities of the vertebral bodies (C3-C7) were measured. Each endplate surface was biomechanically tested for regional yield load and stiffness using an indentation test method. Overall average density of the cervical vertebral body trabecular bone was 270 +/- 74 mg/cm3. There was no significant difference between levels. The yield load and stiffness from the indentation test of the endplate averaged 139 +/- 99 N and 156 +/- 52 N/mm across all cervical levels, endplate surfaces, and regional locations. The posterior aspect of the endplate had significantly higher yield load and stiffness in comparison to the anterior aspect and the lateral aspect had significantly higher yield load in comparison to the midline aspect. There was a significant correlation between the average yield load and stiffness of the cervical endplate and the trabecular bone density on regression analysis. Although there are significant regional variations in the endplate structural properties, the average of the endplate yield loads and stiffnesses correlated with the trabecular bone density. Given the morbidity associated with subsidence of interbody devices, a reliable and predictive method of measuring endplate strength in the cervical spine is required. Bone density measures may be used preoperatively to assist in the prediction of the strength of the vertebral endplate. A threshold density measure has yet to be established where the probability of endplate fracture outweighs the benefit of anterior cervical procedure.

Fig. 1

Stratec XCT-2003 pQCT scan of a cervical vertebral body demonstrating the cross-sectional density of the specimen on the left and the area of the each vertebral body averaged for the trabecular bone density. The cortical edge is not included in the measurement

Fig. 2

The superior endplate of the vertebral body demonstrating the 2 × 3 indentation points

Fig. 3

Example of load–displacement curve from an indentation test

Fig. 4

Regression analysis of vertebral bone density and endplate yield load demonstrated a significant correlation (P = 0.0002, ES = 1.5)

Fig. 5

Regression analysis of vertebral bone density and endplate stiffness demonstrated a significant correlation (P = 0.0001, ES = 1.7)