Guthrie David, Peacock Zachary S, Sadow Peter, Dodson Thomas B, August Meredith
Harvard School of Dental Medicine, Boston, MA, USA.
J Oral Maxillofac Surg. 2012 Nov;70(11):2566-72. doi: 10.1016/j.joms.2011.11.023. Epub 2012 Jan 28.
To compare the accuracy of intraoperative frozen section (FS) and preoperative incisional biopsy (IB) techniques to diagnose benign intraosseous jaw lesions.
This is a retrospective cohort study composed of subjects with benign intraosseous jaw lesions. The predictor variable was the technique for establishing a preliminary diagnosis of the lesion, preoperative IB or intraoperative FS. The outcome variable was the accuracy of the biopsy technique when compared with the final histologic diagnosis and was classified as concordant or discordant. The comparative diagnostic accuracy of the techniques was assessed with the χ(2) test.
A total of 71 subjects met inclusion criteria. The mean age was 39 years (range, 5 to 85 years), and 58% (41) were male patients. Of the subjects, 20 (28%) underwent IB. In 14 (70%) of these, the results of biopsy agreed with the final diagnosis. 51 (72%) underwent intraoperative FS and in 31 (62%) of these, the results of biopsy agreed with the final diagnosis. The difference in diagnostic accuracy between IB (70%) and FS (61%) was statistically insignificant (P = .48). Sources of biopsy error included sampling error (46%), insufficient epithelial tissue (15%), inflammation (15%), pathologist's experience (8%), and artifact (4%).
Preoperative IB and intraoperative FS provide comparable accuracy of diagnosis in patients with benign intraosseous jaw pathology. Sampling error was the most common reason for discordant results.
比较术中冰冻切片（FS）和术前切开活检（IB）技术诊断颌骨骨内良性病变的准确性。
这是一项回顾性队列研究，研究对象为颌骨骨内良性病变患者。预测变量是病变初步诊断的技术，即术前IB或术中FS。结果变量是活检技术与最终组织学诊断相比的准确性，分为一致或不一致。采用χ(2)检验评估技术的比较诊断准确性。
共有71名受试者符合纳入标准。平均年龄为39岁（范围5至85岁），58%（41名）为男性患者。其中，20名（28%）接受了IB。其中14名（70%）的活检结果与最终诊断一致。51名（72%）接受了术中FS，其中31名（62%）的活检结果与最终诊断一致。IB（70%）和FS（61%）之间诊断准确性的差异无统计学意义（P = 0.48）。活检误差来源包括抽样误差（46%）、上皮组织不足（15%）、炎症（15%）、病理学家经验（8%）和人为假象（4%）。
术前IB和术中FS在颌骨骨内良性病变患者中提供了相当的诊断准确性。抽样误差是结果不一致的最常见原因。
