Ho Jin Lee1, Soon Tae You, Jung Hee Kim, Il Sup Kim, Jae Hoon Sung, Jae Taek Hong. 1. Department of Neurosurgery, St. Vincent Hospital, College of Medicine, The Catholic University of Korea, Suwon Department of Neurosurgery, Seoul Medical Center Department of Neurosurgery, Eunpyeong St. Mary's Hospital, College of Medicine, The Catholic University of Korea, College of Medicine, Seoul, Korea.
Abstract
STUDY DESIGN: This study was a retrospective review. OBJECTIVE: The purpose of this study was to investigate the correlation analysis between Hounsfield units (HU) and dual x-ray absorptiometry (DXA) based on the clinical results of patients who underwent anterior cervical discectomy and fusion (ACDF) surgery. SUMMARY OF BACKGROUND DATA: There is no technique to directly measure bone mineral density (BMD) in the cervical spine. As computed tomography is a very popular preoperative planning modality, using the HU value from that analysis to predict osteoporosis is important for patient outcomes and applications in the clinical field. MATERIALS AND METHODS: We reviewed the records for 235 patients who underwent 1-level (n=120) or 2-level (n=115) ACDF surgery. In the 1-level ACDF group, the HU was measured from C3 to C6 vertebra, while that for the 2-level ACDF group was measured from 3 surgical index vertebrae. The correlation patterns were analyzed with the corresponding DXA (T-score) for each patient. Subsidence of fusion segment was defined as change in distance between plate-tip and upper (lower) margin of the vertebra (index level) after 4 months of follow-up. In addition, to determine the relevant factors that influence fusion segment subsidence, other preoperative (C2 slope, C7 slope, C2-C7 angle, and C2-C7 sagittal vertical axis) and postoperative parameters (coronal angle and segmental angle change) were measured. RESULTS: The correlation coefficient between HU and DXA ranged from 0.57 to 0.71 in the 1-level ACDF group and from 0.59 to 0.66 in the 2-level ACDF group. The correlation between HU and DXA was statistically significant regardless of the degree of anterior osteophyte (r=-0.65 to 0.78). Total subsidence height was 3.8 mm after ACDF, and both HU and DXA were statistically correlated with total subsidence (r=0.26-0.28). In multivariate analysis, HU (middle vertebra) value was statistically associated with the degree of total subsidence. The high-subsidence group (≥4.5 mm) showed smaller HU values (284.1 vs. 316.0) and T-scores (-0.5 vs. 0.1) compared to the low-subsidence group (<4.5 mm). The discrepancy group, defined as cases with excess plate shift on 1 side, also showed smaller HU values (260.4 vs. 312.4) and higher degrees of total subsidence than the matched group. CONCLUSIONS: The correlation between HU in cervical computed tomography and lumbar DXA (T-score) was statistically significant in both 1-level and 2-level ACDF. The level of BMD (HU or DXA) is a very important factor for clinically determining the amount and regularity of subsidence after ACDF. Therefore, HU can be a good alternative assessment to accurately reflect as much of the BMD degree as DXA in the cervical spine.
STUDY DESIGN: This study was a retrospective review. OBJECTIVE: The purpose of this study was to investigate the correlation analysis between Hounsfield units (HU) and dual x-ray absorptiometry (DXA) based on the clinical results of patients who underwent anterior cervical discectomy and fusion (ACDF) surgery. SUMMARY OF BACKGROUND DATA: There is no technique to directly measure bone mineral density (BMD) in the cervical spine. As computed tomography is a very popular preoperative planning modality, using the HU value from that analysis to predict osteoporosis is important for patient outcomes and applications in the clinical field. MATERIALS AND METHODS: We reviewed the records for 235 patients who underwent 1-level (n=120) or 2-level (n=115) ACDF surgery. In the 1-level ACDF group, the HU was measured from C3 to C6 vertebra, while that for the 2-level ACDF group was measured from 3 surgical index vertebrae. The correlation patterns were analyzed with the corresponding DXA (T-score) for each patient. Subsidence of fusion segment was defined as change in distance between plate-tip and upper (lower) margin of the vertebra (index level) after 4 months of follow-up. In addition, to determine the relevant factors that influence fusion segment subsidence, other preoperative (C2 slope, C7 slope, C2-C7 angle, and C2-C7 sagittal vertical axis) and postoperative parameters (coronal angle and segmental angle change) were measured. RESULTS: The correlation coefficient between HU and DXA ranged from 0.57 to 0.71 in the 1-level ACDF group and from 0.59 to 0.66 in the 2-level ACDF group. The correlation between HU and DXA was statistically significant regardless of the degree of anterior osteophyte (r=-0.65 to 0.78). Total subsidence height was 3.8 mm after ACDF, and both HU and DXA were statistically correlated with total subsidence (r=0.26-0.28). In multivariate analysis, HU (middle vertebra) value was statistically associated with the degree of total subsidence. The high-subsidence group (≥4.5 mm) showed smaller HU values (284.1 vs. 316.0) and T-scores (-0.5 vs. 0.1) compared to the low-subsidence group (<4.5 mm). The discrepancy group, defined as cases with excess plate shift on 1 side, also showed smaller HU values (260.4 vs. 312.4) and higher degrees of total subsidence than the matched group. CONCLUSIONS: The correlation between HU in cervical computed tomography and lumbar DXA (T-score) was statistically significant in both 1-level and 2-level ACDF. The level of BMD (HU or DXA) is a very important factor for clinically determining the amount and regularity of subsidence after ACDF. Therefore, HU can be a good alternative assessment to accurately reflect as much of the BMD degree as DXA in the cervical spine.