Chacon Guillermo E, Dawson Kenneth H, Myall Robert W T, Beirne O Ross
Department of Oral and Maxillofacial Surgery, University of Washington, Seattle, WA, USA.
J Oral Maxillofac Surg. 2003 Jun;61(6):668-72; discussion 673. doi: 10.1053/joms.2003.50134.
We sought to compare the sensitivity and specificity of panoramic radiographs with those of coronal computed tomography (CT) scans in the diagnosis of mandibular condylar fractures in the pediatric population.
Medical, dental, and radiographic records of patients who presented between 1995 and 2000 were evaluated for injuries involving the mandibular condyle. The sample included 22 males and 15 females with ages ranging from 2 to 15 years (mean, 8 years). Control subjects were added and matched by age and gender. The panoramic radiographs were blocked to allow separate evaluation of each condyle. Representative images from the CT scans were selected and individually photographed for projection. Both sets of images were evaluated by 4 groups of examiners: oral and maxillofacial (OMF) surgeons who regularly deal with pediatric trauma (n = 2), community OMF surgeons who had been out of training for at least 5 years (n = 6), OMF radiologists (n = 3), and OMF surgery residents (n = 6). Each image was shown for 20 seconds and the examiners were given 3 options to choose: 1) fracture, 2) no fracture, and 3) uncertain.
The overall diagnostic accuracy of CT scanning was 90% (sensitivity, 92%; specificity, 87%), and that of panoramic radiographs was 73% (sensitivity, 70%; specificity, 77%). Of interest, the most experienced clinicians were also those most likely to respond that panoramic films were not diagnostic. Statistical analysis of the results was performed using chi(2) analysis. The differences for sensitivity measurements using the CT scan were not statistically significant (P >.1). However, the differences in sensitivity measurements using the panoramic radiographs and the specificity measurements using both the CT and panoramic radiographs were statistically significant (P <.05).
CT scans provide consistently greater accuracy of diagnosis, sensitivity, and specificity than panoramic radiographs in the assessment of children suspected of having condylar fractures. In view of the high rate of false-negative and false-positive results associated with panoramic radiographs, coronal CT scans should be considered routine investigation in these patients.
我们试图比较全景X线片与冠状面计算机断层扫描（CT）在诊断儿童下颌髁突骨折方面的敏感性和特异性。
对1995年至2000年间就诊患者的医学、牙科和影像学记录进行评估，以确定是否存在涉及下颌髁突的损伤。样本包括22名男性和15名女性，年龄在2至15岁之间（平均8岁）。增加了对照受试者，并按年龄和性别进行匹配。全景X线片进行遮挡处理，以便对每个髁突进行单独评估。从CT扫描中选择代表性图像并单独拍照用于投影。两组图像由4组检查人员进行评估：经常处理儿童创伤的口腔颌面外科（OMF）医生（n = 2）、至少5年未接受培训的社区OMF外科医生（n = 6）、OMF放射科医生（n = 3）和OMF外科住院医师（n = 6）。每张图像展示20秒，检查人员有3种选择：1）骨折，2）无骨折，3）不确定。
CT扫描的总体诊断准确率为90%（敏感性92%，特异性87%），全景X线片的总体诊断准确率为73%（敏感性70%，特异性77%）。有趣的是，经验最丰富的临床医生也是最有可能认为全景片无法诊断的医生。使用卡方分析对结果进行统计学分析。使用CT扫描进行敏感性测量的差异无统计学意义（P>.1）。然而，使用全景X线片进行敏感性测量以及使用CT和全景X线片进行特异性测量的差异具有统计学意义（P<.05）。
在评估疑似髁突骨折的儿童时，CT扫描在诊断准确性、敏感性和特异性方面始终优于全景X线片。鉴于全景X线片存在较高的假阴性和假阳性结果率，对于这些患者，冠状面CT扫描应被视为常规检查。
