| Osteoporotic vertebral compression fractures(OVCF)is the most common osteoporotic fracture,which severely affects the quality of life of the elderly and increases mortality.Although healing in most OVCF following conservative treatment,some progress into OVCF nonunion.On this basis,if the risks of fracture in osteoporotic vertebrae and the risks of nonunion in OVCF were quantitatively evaluated,we can block the progress of osteoporotic vertebrae into OVCF or OVCF into nonunion following early treatment.The risk of fracture in osteoporotic vertebrae and the risk of nonunion in OVCF was quantitatively evaluated by magnetic resonance(MR)imaging in this study.This study is divided into the following three parts:Part ⅠPurpose: To predict the fracture risk of osteoporotic vertebral body by quantitatively evaluating the bone quality of cancellous bone(BQCB)of the osteoporotic vertebral body on mid-sagittal T1-weighted MR imaging.Methods: Thirty-nine acute osteoporotic vertebral compression fracture(OVCF)patients(fracture group)and sixty osteoporosis patients(osteoporosis group)were collected in this study.The BQCB of the osteoporotic vertebral bodies between the two groups were quantitatively evaluated on the mid-sagittal T1-weighted MR imaging to compare the differences.Binary multiple logistic regression analysis was performed to determine the relative contribution of the BQCB and demographic data for differentiation of acute OVCF and osteoporosis to determine whether BQCB can be used as the only statistically significant factor to distinguish two groups.To evaluate the diagnostic utilities of BQCB,we performed receiver operator characteristic curve(ROC)analysis to determine sensitivities,specificities,and cutoff values.Results: The interclass correlation coefficient(ICC)for BQCB was 0.930 [95%confidence interval(CI),0.870-0.960,P<0.001] in fracture group and 0.912(95%CI,0.852-0.948,P<0.001)in osteoporosis group.The BQCB score in fracture group was significantly higher than that in osteoporosis group(P<0.001).The Binary multiple logistic regression analysis showed that there were no significant differences in gender(P = 0.513),age(P = 0.155),BMI(P = 0.364),smoking(P = 0.713)and drinking(P =0.934)between the two groups.The BQCB score was the only variable used to independently differentiate fractured vertebrae from osteoporotic vertebrae(P<0.001).The area under the curve(AUC)for ROC was 0.951(95%CI,0.896-0.982,P<0.0001).The BQCB score cutoff value of 3.13 was clinically applicable,affording excellent diagnostic accuracy(sensitivity,78.69%;specificity,100.0%).When the cutoff was2.29,no fractured risk for assessing osteoporotic vertebrae was 100%;that is,the osteoporotic vertebrae with no risk of fracture had BQCB score lower than 2.29.When the cutoff was 3.13,specificity was 100% for fractured risk for assessing osteoporotic vertebrae;all osteoporotic vertebrae with the risk of fracture had BQCB score were greater than 3.13.Conclusions: The BQCB of vertebral bodies quantitatively evaluated on T1-weighted MR imaging can be used to distinguish acute fractured vertebrae from osteoporotic vertebrae,and was a good indicator for assessing the fracture risk of osteoporotic vertebral bodies.Part ⅡPurpose: To predicted the risk of OVCF nonunion by quantifying the vertebral marrow fat fraction(VMFF)of the acute fractured vertebral body on mid-sagittal T2-weighted MR Dixon imaging.Methods: The fractured vertebrae from 39 patients with acute OVCF from the first part were divided into nonunion group and union group basing on the prognosis in this study.The VMFF of the fractured vertebrae were quantified on mid-sagittal T2-weighted Dixon imaging to compare the differences between the two groups.Binary multiple Logistic regression analysis was performed to determine the relative contribution of the VMFF and demographic data for differentiation of nonunion or union of OVCF to determine whether it can be used as the only statistically significant factor to distinguish two groups.To evaluate the diagnostic utilities of VMFF,we performed ROC analysis to determine sensitivities,specificities,and cutoff values.Results: The interclass correlation coefficient(ICC)for VMFF was 0.957(95%CI,0.896-0.983,P<0.001)in nonunion group and 0.946(95%CI,0.901-0.971,P<0.001)in union group.The VMFF for the nonunion group(16.40±15.65%)was significantly lower than the score for the union group(55.18±20.93%)(P<0.001).The Binary multiple Logistic regression analysis showed that there were no significant differences in gender(P = 0.201),age(P = 0.319),BMI(P = 0.635),smoking(P = 0.775),drinking(P = 0.521),time from injury to first visit(P = 0.251),trauma event(P = 0.911)and fracture site(P = 0.495)between the two groups.The VMFF from acute fracture vertebrae was the only variable to independently differentiate the prognosis of acute OVCF(P<0.001).The AUC for ROC was 0.928(95%CI,0.831-0.979,P<0.0001).The VMFF cutoff value of 24% was clinically applicable,affording excellent diagnostic accuracy(sensitivity,80.00%;specificity,95.00%).When the cutoff was 57.00%,the sensitivity of predicting union of OVCF was 100%;that is,all union of OVCF had VMFF higher than 57.00%.With a VMFF cutoff of 11.00%,specificity was 100% for OVCF nonunion for assessing fractured vertebral body;all nonunion OVCF had VMFF lower than 11.00%.Conclusions: The VMFF of fractured vertebral body quantitatively evaluated on midsagittal T2-weighted MR Dixon imaging can be used to distinguish the differences in fracture vertebrae in the acute phase,and can be used as a good indicator to assess the prognosis of acute OVCF.Part ⅢPurpose: Both acute OVCF patients and nonunion of OVCF patients were treated with teriparatide.The reliability of the BQCB score for assessing the risk of fractures in patients with osteoporosis was further verified by evaluating the therapeutic effects and the difference in BQCB of the vertebral bodies before and after treatment.Methods: Twenty patients with nonunion of OVCF(OVCF nonunion group)from the second part in our study and twenty patients with acute OVCF(OVCF group)were treated with teriparatide.At 12 months,CT or MR imaging was used to evaluate the bone union of fractured vertebrae.The vertebral collapse,referred to KA increment ＞10° and or loss rate of vertebral height ＞ 15%,was assessed by the changes in kyphosis angle(KA),anterior border height(ABH)and posterior border height(PBH).The BQCB before and after treatment were compared.The diagnostic value scale for BQCB score from the first part was used to evaluate the refracture risk of the union vertebrae.Results: By the end of follow-up,CT and MR imaging confirmed that bone union in all fractured vertebrae following treatment.From baseline to month 12,the KA increment was 1.70 ± 3.35° and the loss rates of ABH and PBH were 9.90 ± 4.88% and0.85 ± 3.17% respectively in OVCF group.From baseline to month 12,the KA increment was 3.35±1.81° and the loss rates of ABH and PBH were 10.85±4.75% and0.25±1.12% respectively in OVCF nonunion group.During the whole follow-up period,no new OVCF was found in both groups.In OVCF group,BQCB score decreased from3.54 ± 0.40 at baseline to 2.56 ± 0.24 at months 12(P < 0.001).In OVCF nonunion group,BQCB score decreased from 3.74 ± 0.41 at baseline to 2.60 ± 0.50 at months 12(P < 0.001).The BQCB score for both groups was less than the cutoff value of BQCB score for assessing osteoporotic vertebral fracture risk.According to the diagnostic value scale for BQCB score from the first part,the sensitivity of BQCB score for assessing fracture risk increased from 51.67-53.33% to 98.33% and the specificity decreased from 100% to 3.33-8.33 in fracture group.In nonunion group,the sensitivity for assessing fracture risk increased from 43.33-45.00% to 98.33%,and the specificity decreased from 100% to 3.33-8.33.Conclusions: Teriparatide treatment can promote fracture healing and prevent progressive vertebral collapse in patients with OVCF and OVCF nonunion.The changes in BQCB score before and after treatment verified the accuracy of score in evaluating the fracture risk of osteoporotic vertebrae. |
PDF Full Text Request