INTRODUCTION AND HYPOTHESIS: The spinal curvature irregularity index (SCII) is a quantitative measure of the irregularity of the spinal curvature. We evaluated the predictive ability of SCII to identify subjects with vertebral fractures (VF). METHODS: Vertebral heights were measured by quantitative vertebral morphometry in 461 Lebanese women 20-89 years of age and VFs were ascertained by the grade 1 Eastell method. SCII scores were log-transformed and expressed as Z-SCII, the number of standard deviations above or below the mean ln(SCII) of young patients without VF. Univariate and multivariate binary logistic regression models were used to identify clinical predictors of VF. RESULTS: Women with a higher SCII were more likely to have prevalent VF. A higher SCII was associated with a greater prevalence of VF within each category of femoral neck BMD (normal, osteopenia, osteoporosis). In univariate analysis, predictors of VF included Z-SCII (odds ratio, OR: 2.21, 95% CI: 1.80-2.71) and femoral neck T-score (OR: 1.35, 95% CI: 1.12-1.63). In multivariate analysis, predictors of VF were: Z-SCII (OR: 1.54, 95% CI: 1.02-2.32), femoral neck T-score (OR: 1.41, 95% CI: 1.11-1.78) and age(3) (OR: 1.40, 95% CI 1.10-1.82). At a cutoff SCII of 9.5%, the sensitivity and specificity of SCII for VF were 71 and 64% respectively, and higher SCII cutoffs identified VFs with greater specificity. CONCLUSION: The SCII is a robust, simple and independent indicator of the presence of VFs.
INTRODUCTION AND HYPOTHESIS: The spinal curvature irregularity index (SCII) is a quantitative measure of the irregularity of the spinal curvature. We evaluated the predictive ability of SCII to identify subjects with vertebral fractures (VF). METHODS: Vertebral heights were measured by quantitative vertebral morphometry in 461 Lebanese women 20-89 years of age and VFs were ascertained by the grade 1 Eastell method. SCII scores were log-transformed and expressed as Z-SCII, the number of standard deviations above or below the mean ln(SCII) of young patients without VF. Univariate and multivariate binary logistic regression models were used to identify clinical predictors of VF. RESULTS:Women with a higher SCII were more likely to have prevalent VF. A higher SCII was associated with a greater prevalence of VF within each category of femoral neck BMD (normal, osteopenia, osteoporosis). In univariate analysis, predictors of VF included Z-SCII (odds ratio, OR: 2.21, 95% CI: 1.80-2.71) and femoral neck T-score (OR: 1.35, 95% CI: 1.12-1.63). In multivariate analysis, predictors of VF were: Z-SCII (OR: 1.54, 95% CI: 1.02-2.32), femoral neck T-score (OR: 1.41, 95% CI: 1.11-1.78) and age(3) (OR: 1.40, 95% CI 1.10-1.82). At a cutoff SCII of 9.5%, the sensitivity and specificity of SCII for VF were 71 and 64% respectively, and higher SCII cutoffs identified VFs with greater specificity. CONCLUSION: The SCII is a robust, simple and independent indicator of the presence of VFs.
Authors: M Barake; R El Eid; S Ajjour; M Chakhtoura; L Meho; T Mahmoud; J Atieh; A M Sibai; G El-Hajj Fuleihan Journal: Osteoporos Int Date: 2021-04-07 Impact factor: 4.507