|
| 用定量形态学测量法检测再发性脊柱骨折 |
| The application of quantitative morphometry for assessment of incident vertebral fractures：comparison with semiquantitative technique |
| |
| DOI： |
| 中文关键词: 骨质疏松 骨折 闭合性 形态学测量 |
| 英文关键词:Osteoporosis Fractures Closed Morphometry |
| 基金项目: |
|
| 摘要点击次数: 1305 |
| 全文下载次数: 1157 |
| 中文摘要: |
| 应用定量形态学测量法(quantitative morphometric assessment，QM)和半定量分析法(semi-
Quantitative assessment)诊断335名绝经期后妇女的再发性脊椎骨折，以比较这两种方法在诊断上的差异。由三名受过训练的放射科医师分别用SQ法对比每位受检者的脊椎平片的基础片和随访片作出初诊，再由该三名医师共同复阅所有平片，结合各自诊断和 QM的结果得出会诊阅片结果(Consensus Reading，CR)。并使用QM法的六点数字式技术测量平片，采用四种形态学定义作为诊断标准。结合每一标准所选用的四个常用划分骨折的阈值作出再发性骨折的诊断，即(1)随访片内椎高减少百分比(PHr)；(2)椎高减少绝对值(AHr)；(3)PHr加AHr 及(4)椎高比率减少绝对值(ARt)。比较定量法与半定量法的诊断结果显示,CR诊断再发性骨折率为0.955% 。三名医师之间以及各自与CR之间的诊断均有很高的符合率，Kappa值(κ值)达0.86~0.96；敏感性为84%~93% ；特异性为99.9% 。QM与CR的κ值为0.27~0.63；敏感性为 30%~80%，特异性为 95%~99.9% 。其中最高的κ值为AHr的0.63, 相应的敏感性为51% ，特异性为99.98%。 结果显示，定量形态测量法与半定量分析法对诊断再发性脊椎骨折的符合率仅为轻至中度。对于流行病学和临床药效的研究，单独使用定量形态测量法并非是一可靠的手段。因此该法应结合受训过的放射学医师或有相当经验的临床医师的阅片结果一并使用。 |
| 英文摘要: |
| This study compared semiquantitative reading and quantitative morphometry for the assessment of incident vertebral fractures．Radiographs of the thoracolumbar spine taken one year apart for 335 postmenopausal women with suspected vertebral fractures were evaluated．Quantitative morphometry(QM) and visual semiquantitative grading by 3 trained radiologists independently and in a joint consensus reading(CR) sesson were used to determine incident vertebral fractures. QM was performed using a six-point-digitization technique, and incident deformities using QM were defined using a range of cutoffs with four different morphometric criteria．By the CR 9.55% of the women and 1.33% of the vertebrae had incident fractures after 1 year. Excellent agreement was found among the radiologists readings as well as between the individual readings and the CR for the assessment of incident fractures, with kappa scores between 0.86 and 0. 96. Sensitivities ranged from 84% to 93% with specificities of 99.9% relative to the CR. Using selected cutoff thresholds with the four QM criteria, kappa scores between QM and the CR ranged from 0.27to 0.63，with corresponding sensitivities from 30% to 80% and speciflcities from 95% to 99.9% relative to the CR. The assessment of incident vertebral fractures using QM showed only fair to moderate agreement with the conseusus reading or with the individual readings．The assessment of incident fractures using QM alone may not be sufficiently reliable for detection of incident vertebral fractures in epidemilological studies and inclinical trials, and it should be performed principally in conjunction with the reading of a trained radiologist or a highly experienced clinician． |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |