Mandell Jacob C, Weaver Michael J, Khurana Bharti
Department of Radiology, Division of Musculoskeletal Imaging and Intervention, Harvard Medical School, Brigham and Women's Hospital, 75 Francis St., Boston, MA, 02115, USA.
Department of Orthopedic Surgery, Harvard Medical School, Brigham and Women's Hospital, Boston, MA, USA.
Emerg Radiol. 2018 Jun;25(3):265-273. doi: 10.1007/s10140-018-1580-4. Epub 2018 Jan 11.
The purpose of this study was to evaluate the diagnostic performance of CT for assessment of occult fractures of the proximal femur, pelvis, and sacrum.
A retrospective review was performed on patients who received a CT of the hip or pelvis for suspected occult fracture after negative or equivocal radiographs performed within 24 h. The official radiology report was utilized for the determination of CT findings and calculation of sensitivity and specificity. Surgical reports, MRI reports, and clinical follow-up were used as the standard of reference. Sensitivity and specificity were calculated with 95% confidence intervals.
Seventy-four patients received CT of the hip or pelvis for clinical concern for occult fracture after negative or equivocal radiographs. By the reference standard, a total of 40 fractures were present in 25/74 (33.8%) patients, including 35 conservatively treated fractures of the greater trochanter, pelvis, and sacrum, and 5 operatively treated proximal femoral fractures. A total of 14/74 (18.9%) of patients had an MRI within 1 day of CT. MRI identified an operatively treated femoral neck fracture not seen on CT and an operatively treated intertrochanteric fracture, which CT described as a greater trochanteric fracture. There were two false negative conservatively treated pelvic fractures not seen on CT but diagnosed on MRI. On a per-patient basis, CT had an overall sensitivity of 88% (22/25; 95% confidence intervals 69-97%), specificity of 98% (48/49; 95% confidence intervals 89-100%), and negative predictive value of 94%. For the five operative proximal femoral fractures, the sensitivity of CT was 60% (3/5; 95% confidence intervals 15-95%), specificity was 99% (68/69; 95% confidence intervals 92-100%), and negative predictive value was 97%.
In the clinical setting of suspected occult fracture, the sensitivity of clinical CT reports for detection of any type of fracture of the proximal femur, pelvis, or sacrum was 88%. For the small number of operatively treated proximal femoral fractures seen in the study, sensitivity of CT was 60% (3/5) and negative predictive value was 97%, although the relatively few patients needing fixation precludes statistical analysis.
本研究旨在评估CT对股骨近端、骨盆和骶骨隐匿性骨折的诊断效能。
对在24小时内X线片为阴性或不明确后因疑似隐匿性骨折而接受髋部或骨盆CT检查的患者进行回顾性分析。利用官方放射学报告来确定CT检查结果，并计算敏感性和特异性。手术报告、MRI报告及临床随访结果作为参考标准。计算敏感性和特异性的95%置信区间。
74例患者在X线片为阴性或不明确后因临床怀疑隐匿性骨折接受了髋部或骨盆CT检查。根据参考标准，25/74（33.8%）例患者共存在40处骨折，其中包括35处经保守治疗的大转子、骨盆和骶骨骨折，以及5处经手术治疗的股骨近端骨折。74例患者中有14/74（18.9%）在CT检查后1天内进行了MRI检查。MRI发现了1处CT未显示的经手术治疗的股骨颈骨折和1处经手术治疗的转子间骨折，CT将后者描述为大转子骨折。有2处经保守治疗的骨盆骨折CT未显示但MRI诊断为骨折，为假阴性。以患者为基础计算，CT的总体敏感性为88%（22/25；95%置信区间69 - 97%），特异性为98%（48/49；95%置信区间89 - 100%），阴性预测值为94%。对于5处经手术治疗的股骨近端骨折，CT的敏感性为60%（3/5；95%置信区间15 - 95%），特异性为99%（68/69；95%置信区间92 - 100%），阴性预测值为97%。
在疑似隐匿性骨折的临床情况下，临床CT报告对检测股骨近端、骨盆或骶骨任何类型骨折的敏感性为88%。对于本研究中少数经手术治疗的股骨近端骨折，CT的敏感性为60%（3/5），阴性预测值为97%，尽管需要固定治疗的患者相对较少，无法进行统计学分析。
