手工测量特发性脊柱侧弯Cobb角的可靠性评价

目的 应用无标记的X线片评价手工测量特发性脊柱侧弯Cobb角的可信度和可重复性.方法 4名骨科医师独立对53例患儿的术前站立位全脊柱正位及侧位X线片进行2次冠状面及矢状面Cobb角测量,2次测量间隔2周.冠状面主弯和Cobb角大于10°的代偿弯以及与Lenke和/或PUMC分型相关的矢状面T2～5,T5～12、T10～L2、T12～L1角度纳入分析.应用组内相关系数判定可靠性和可重复性.结果 36例上胸弯、53例胸弯、41例胸腰弯/腰弯及矢状面T2～5、T5～12、T10～L2及T12～L1后弯Cobb角各53例纳入分析,主弯Cobb角22°～167°,平均54.4°.结果 显示冠状面各角度测量中,上胸弯Cobb角的可信度(γ=0.592～0.714)和可重复性(γ=0.538～0.735)最差,有中等到好的一致性.胸弯Cobb角测量达到好到极佳的可信度(γ=0.881～0.942)和极佳的可重复性(γ=0.919～0.964).胸腰弯/腰弯角度测量有中等到极佳的可信度(γ=0.649～0.915)和好到极佳可重复性(γ=0.875～0.930).矢状面测量中,T2～5后弯Cobb角测量的可靠性最差,仅有低到中等的可信度(γ=0.373～0.662)和可重复性(γ=0.433～0.527).T5～12、T10～L2、T12～L1后弯cobb角测量均有好到极佳的可信度(分别为γ=0.787～0.909;0.857～0.964;0.889～0.983).T5～12Cobb 角测量有好到极佳的可重复性(γ=0.873～0.943),T10～L2、T12～L1 Cobb角测量均有极佳的可重复性(分别为γ=0.915～0.968;0.937～0.981).结论 大多数的角度测量指标有好到极佳的可信度和可重复性,可以应用于临床工作和科研中.总体上冠状面角度测量一致性较矢状面好.由于受到周围组织阴影干扰,冠状面上胸弯和矢状面T2～5 Cobb角测量的一致性较差.

Reliability of manual measurement of Cobb angle for patients with idiopathic scoliosis

Objective To analyze the reliability of manual measurement of Cobb angle of the patients with idiopathic scoliosis.Methods Fifty three patients with idiopathic scoliosis had full-length standing posteroanterior and lateral radiographs were measured by four orthopedic surgeons independently.Coronal measures included Cobb angles of the three main curves.Sagittal parameters included T2-T5 and T5-T12 regional thoracic kyphosis,T10-L2 thoracolumbar junction sagittal curvature,T12-L2lumbar lordosis.The measurements were redone 2 weeks later.Inter-observer and intra-observer reliability for each parameter were assessed using intraclass correlation coefficient(ICC)method.Results Of the 53 patients,36 patients'proximal thoracic Cobb angles(PT),53 patients'main thoracic Cobb angles(MT),41 patients'thoracolumbar/lumbar Cobb angles,and all patients' T2-5.T5-12,T10-L2 and T12-L1 Cobb angles were measured.Main curves Cobb angles were 54.4(range,22 to 167).In the coronal Cobb angle measurements,the proximal thoracic curves measurements showed fair to good interobserver reliability(γ=0.592-0.714)and fair to good intraobserver reproducibility(γ=0.538-0.735);the main thoracic curves measurements showed good to excellent interobserver reliability(γ=0.881-0.942)and excellent intraobserver reproducibility(γ=0.919-0.964),while thoracolumbar/lumbar curves demonstrated fair to excellent interobserver reliability(γ=0.649-0.915)and good to excellent intraobserver reproducibility(γ=0.875-0.930).In the sagittal Cobb angle measurements,T2-5 curves demonstrated worst reliability and validity,low to fair interobserver reliability(γ=0.373-0.662)and intraobserver reproducibility(γ=0.433-0.527).T5-12,T10-L2 and T12-L1 curves demonstrated good to excellent interobserver reliability(γ=0.787-0.909;0.857-0.964;0.889-0.983).T5-12 curves demonstrated good to excellent intraobserver reproducibility(γ=0.873-0.943).T10-L2 and T12-L1 curves demonstrated excellent intraobserver reproducibility(γ=0.915-0.968;0.937-0.981).Conclusions The majority of the radiographic parameters for idiopathic scoliosis assessment demonstrate good or excellent reliability for manual measurement.