| It has been widely used now to get the CT measurement data of the pedicle screw parameters in pre-operation or intra-operation in clinic in the posterior vertebral pedicle operation, especially to the severious adolescent idiopathic scoliosis (AIS) and others in which the vertebral rotational angle changed markable in 3D. Up to now no standard CT methods of pedicle screw measurements have been widely accepted, and to the author's knowledge, no articles were found about the accuracy and the repetitivity of the CT method in intranational literature, and no articles were found to validate the accuracy of all the pedicle screw CT parameters with basal experimentation in international literature. Part 1: A new CT measuring method of pedicle screw parametersObjectives: To present a new CT measuring method of the pedicle screw parameters.Material and Methods: The new method was showed with a thoracic vertebral specimen. The thoracic vertebral specimen settled with a rotational angle in three dimensions was scanned with the spiral CT. The data was reformatted with multi-planar reformation (MPR) technique, the following planes were reformatted ordinally with modification : the coronal, sagittal and axial planes of the vertebra, and the sagittal, axial planes of the pedicle screws.Results: Some certain parameters of the vertebra or the pedicle screw were measured accurately on the following corresponding planes: the vertebral rotational angle on the coronal, sagittal and the axial plane , the rotational angle of the pedicle screw on the sagittal and the axial plane of the vertebra itself, and the length and width of the pedicle screw.Conclusion: With this new CT method not only the exact axial plane of the vertebra and the pedicle screw can be found accurately, but also the parameters of the vertebra and the pedicle screw can be measured accurately, including the length, width of the pedicle screw and the corresponding rotational angle in three dimensions.Part 2: Evaluation of the accuracy of the CT measuring method of the vertebral pedicle screw parametersObjectives: To evaluate the accuracy of the CT measuring method of the vertebral pedicle screw parameters in different vertebral rotational angles.Material and Methods: A thoracic vertebral specimen was scanned 10 times with spiral CT in different rotational angles. 10 groups of CT data were reformatted and measured four times each by three observers separately with double blind method. The CT measuring method is to obtain ordinally the following parameters with multi-planar reformation(MPR) technique and modification: the coronal, sagittal and axial planes of the vertebra, the sagittal and axial planes of the pedicle screw, and the vertebral rotational angles on the coronal, sagittal and axial planes (angles: c, s, a) , the pedicle screw's rotational angles on the sagittal and axial planes of the vertebra itself (angles: ss, aa) , and the pedicle screw's length and width (Length, Width) . The measuring results of the parameters were analyzed with SPSS 10.0 (One-Way ANOVA, General linear model-repeated measures).Results : The measuring data of c, s, a. from intraobservers or interobservers of each of the 10 groups were not significant different statistically (P>0.05). The measuring data of aa, Length, Width from intraobservers or interobservers of the 10 groups were not significant different statistically (P>0. 05). The measuring data of aa, Length, Width from intergroups of the 10 groups were not significant different statistically (P>0. 05) ( the covariate variables had been considered into the statistical analysis of the data of Length and Width).Conclusion: The parameters of the vertebral the pedicle screw can be measured accurately with this CT method.Part 3: The analysis of the error in the spiral CT measurementObjectives: To study the causing elements of the error in the spiralCT measurement.Material and Methods: Six small round metal balls (the diameter is0.5 mm) were lined on a horizontal plane with each one of them arrangedat intervals of 20mm and scanned with spiral CT when angled with the CT position line of the coronal plane in 0 degree, 30 degree, 60 degree and 90 degree separately. The central distance between the adjacent two balls, and the distortion extent of the six balls were measured on CT images. The measurements were repeated when the central locations of reconstruction were changed.Results: The distance between every two adjacent balls' centers equaled to 20mm in the groups of either different angles or different central location of reconstruction on the CT image, and similar changes were found in the six balls: the balls turned into rice-shaped with the width of 2mm and the length of 5mm, the central line of the length was vertical to that of the width and paralleled to the line around which the CT bed moved in and out.Conclusion: The error in spiral CT measurement correlated with neither the location of the objects in the CT SFOV nor the central location of reconstruction but the distortion of the objects after had been scanned. The distortion was serious along the line around which the CT bed moved in and out, while slight in the direction that was vertical to the line above.Part 4: To evaluate the effect of CT slice thickness to the accuracy of the pedicle screw CT parametersObjectives: To evaluate the effect of CT slice thickness to the accuracy of the pedicle screw CT parameters.Material and Methods: A vertebral specimen in nine different rotational angles both in the sagittal and the coronal planes from 0°to 80°, were scanned twice with the 2.5mm and 0.625mm reconstruction slice thickness with computed tomography(CT). Two observers measured twice respectively the pedicle screw parameters including c, a, s, aaR, LenR, WidR with double blinded method. The measuring data c, a, s, aaR, LenR, WidR from the two slice thickness groups were analyzed with SPSS10. 0(Paired-Samples T Test).Results: The measuring data c, a, s, aaR, LenR from the 2. 5mm and 0. 625mm reconstruction slice thickness groups were not significant different statistically (P>0. 05). The measuring data WidR from the 2. 5mm and 0. 625mm reconstruction slice thickness groups were significant different statistically (P<0. 05).Conclusion: There was not more error increased in the measuring data c, a, s, aaR, LenR from the 2. 5mm reconstruction slice thickness groups than from the 0.625mm reconstruction slice thickness groups except the measuring data WidR.Part 5: The correlations of CT parameters of the pedicle screwObjectives: To analyze the correlations of CT parameters of the pedicle screw.Material and Methods: A vertebral specimen was scanned with spiral CT under ten different rotational angles. And the data were reformatted and measured by three doctors with double blind method separately. This CT measuring method contained such followed steps: Firstly, got the vertebral rotational angle on the coronal, axial and the sagittal plane, and the axial plane of the vertebra which passed through the pedicles; then got the pedicle screw' s rotational angle on the sagittal, axial plane of the vertebra itself, and the axial plane of the pedicle screw; lastly got the pedicle screw' s length and width. These parameters' data were analyzed statistically with SPSS10. 0. Results: 1. The measuring data of aaR, LengthR and WidthR of the ten groups of different rotational angles or of the three different doctors were not significant different statistically(P>0. 05). And those of c, a, s, aaR, LengthR and WidthR in each group of the three different doctors were not significant different statistically either(P>0. 05). ( One-Way ANOVA, General linear model-repeated measures) 2. In each group,①aaR correlates with va(va:in 5/10 groups) ;②LengthR correlates with aaR, va, c (c:in 3/10 groups) ;③WidthR correlates with aaR, |c—(c|-)|, (aaR:in 1/10 group;|c—(c|-)|:in 2/10 groups). (r>0. 5, P<0.05) (Bivariate) 3. Among these ten groups,①aaR correlates with |va-(va|──)|, s|-;②LengthR correlates with D, aaR, a|-, s|-, |s—(s|-)|; (3)WidthR correlates with D, |s—(s|-)|,s, c, c|-. (P<0. 05) (Linear Regression)Conclusion: 1. The pedicle screw parameters could be measured accuratelywith this CT method. 2. There are correlations among these parameters, and theerror of measurement of c, a, s and aaR could cause the error of aa, Length andWidth secondarily.
|
PDF Full Text Request