The use of the T1 sagittal angle in predicting overall sagittal balance of the spine

BACKGROUND CONTEXT: A balanced sagittal alignment of the spine has been shown to strongly correlate with less pain, less disability, and greater health status scores. To restore proper sagittal balance, one must assess the position of the occiput relative to the sacrum. The assessment of spinal balance preoperatively can be challenging, whereas predicting postoperative balance is even more difficult. PURPOSE: This study was designed to evaluate and quantify multiple factors that influence sagittal balance. STUDY DESIGN: Retrospective analysis of existing spinal radiographs. METHODS: A retrospective review of 52 adult spine patient records was performed. All patients had full-column digital radiographs that showed all the important skeletal landmarks necessary for accurate measurement. The average age of the patient was 53 years. Both genders were equally represented. The radiographs were measured using standard techniques to obtain the following parameters: scoliosis in the coronal plane; lordosis or kyphosis of the cervical, thoracic, and lumbar spine; the T1 sagittal angle (angle between a horizontal line and the superior end plate of T1); the angle of the dens in the sagittal plane; the angle of the dens in relation to the occiput; the sacral slope; the pelvic incidence; the femoral-sacral angle; and finally, the sagittal vertical axis (SVA) measured from both the dens of C2 and from C7. RESULTS: It was found that the SVA when measured from the dens was on average 16 mm farther forward than the SVA measured from C7 (p<.0001). The dens plumb line (SVA(dens)) was then used in the study. An analysis was done to examine the relationship between SVA(dens) and each of the other measurements. The T1 sagittal angle was found to have a moderate positive correlation (r=0.65) with SVA(dens), p<.0001, indicating that the amount of sagittal T1 tilt can be used as a good predictor of overall sagittal balance. When examining the other variables, it was found that cervical lordosis had a weak correlation (r=0.37) with SVA(dens) that was unexpected, given that cervical lordosis determines head position. Thoracic kyphosis also had a weak correlation (r=0.26) with SVA(C1), which was equally surprising. Lumbar lordosis had a slightly higher correlation (r=0.38), p=.006, than the cervical or thoracic spine. A multiple regression was run on the data to examine the relationship that all these independent variables have on SVA(dens) SPSS (SPSS, Inc., Chicago, IL, USA) was used to create a regression equation using the independent variables of T1 sagittal angle, cervical lordosis, thoracic kyphosis, lumbar lordosis, sacral slope, pelvic incidence, and femoral-sacral angle and the dependent variable of SVA(dens). The model had a strong correlation (r=0.80, r(2)=0.64) and was statistically significant (p<.0001). The T1 sagittal angle was the variable that had the strongest correlation with the SVA(dens) Spearman r=0.65, p<.0001, followed by pelvic incidence, p=.002, and lumbar lordosis, p=.006. We also observed that when the T1 tilt was higher than 25 degrees, all patients had at least 10 cm of positive sagittal imbalance. In addition, patients with negative sagittal balance had mostly low T1 tilt values, usually lower than 13 degrees. The other variables were not shown to have a statically significant influence on SVA. CONCLUSIONS: This analysis shows that many factors influence the overall sagittal balance of the patient, but it may be the position of the pelvis and lower spine that have a stronger influence than the position of the upper back and neck. Unfortunately, to our knowledge, there are no studies to date that have established a normal sagittal T1 tilt angle. However, our analysis has shown that when the T1 tilt was higher than 25 degrees, all patients had at least 10 cm of positive sagittal imbalance. It also showed that patients with negative sagittal balance had mostly low T1 tilt values, usually below 13 degrees of angulation. The T1 sagittal angle is a measurement that may be very useful in evaluating sagittal balance, as it was the measure that most strongly correlated with SVA(dens). It has its great utility where long films cannot be obtained. Patients whose T1 tilt falls outside the range 13 degrees to 25 degrees should be sent for full-column radiographs for a complete evaluation of their sagittal balance. On the other hand, a T1 tilt within the above range does not guarantee a normal sagittal balance, and further investigation should be performed at the surgeon's discretion. (C) 2010 Elsevier Inc. All rights reserved.