Vara Julio, Pagliuca Mariachiara, Springer Serena, Gonzalez de Canales Juan, Brotons Isabel, Yakcich Javiera, Ajossa Silvia, Pascual Maria Angela, Guerriero Stefano, Alcazar Juan Luis
Department of Obstetrics and Gynecology, Clinica Universidad de Navarra, University of Navarra, 31008 Pamplona, Spain.
Centro Integrato di Procreazione Medicalmente Assistita (PMA) e Diagnostica Ostetrico-Ginecologica, Azienda Ospedaliero Universitaria-Policlinico Duilio Casula Monserrato, University of Cagliari, 09124 Cagliari, Italy.
Diagnostics (Basel). 2023 Feb 10;13(4):673. doi: 10.3390/diagnostics13040673.
The O-RADS system is a new proposal for establishing the risk of malignancy of adnexal masses using ultrasound. The objective of this study is to assess the agreement and diagnostic performance of O-RADS when using the IOTA lexicon or ADNEX model for assigning the O-RADS risk group.
Retrospective analysis of prospectively collected data. All women diagnosed as having an adnexal mass underwent transvaginal/transabdominal ultrasound. Adnexal masses were classified according to the O-RADS classification, using the criterion of the IOTA lexicon and according to the risk of malignancy determined by the ADNEX model. The agreement between both methods for assigning the O-RADS group was estimated using weighted Kappa and the percentage of agreement. The sensitivity and specificity of both approaches were calculated.
454 adnexal masses in 412 women were evaluated during the study period. There were 64 malignant masses. The agreement between the two approaches was moderate (Kappa: 0.47), and the percentage of agreement was 46%. Most disagreements occurred for the groups O-RADS 2 and 3 and for groups O-RADS 3 and 4. The sensitivity and specificity for O-RADS using the IOTA lexicon and O-RADS using the ADNEX model were 92.2% and 86.1%, and 85.9% and 87.4%, respectively.
The diagnostic performance of O-RADS classification using the IOTA lexicon as opposed to the IOTA ADNEX model is similar. However, O-RADS group assignment varies significantly, depending on the use of the IOTA lexicon or the risk estimation using the ADNEX model. This fact might be clinically relevant and deserves further research.
O-RADS系统是一项利用超声评估附件包块恶性风险的新提议。本研究的目的是评估在使用IOTA术语表或ADNEX模型来指定O-RADS风险组时，O-RADS的一致性和诊断性能。
对前瞻性收集的数据进行回顾性分析。所有被诊断为附件包块的女性均接受经阴道/经腹超声检查。根据O-RADS分类、使用IOTA术语表的标准以及ADNEX模型确定的恶性风险对附件包块进行分类。使用加权Kappa和一致性百分比来估计两种指定O-RADS组方法之间的一致性。计算两种方法的敏感性和特异性。
在研究期间评估了412名女性的454个附件包块。其中有64个恶性包块。两种方法之间的一致性为中等（Kappa：0.47），一致性百分比为46%。大多数分歧发生在O-RADS 2组和3组以及O-RADS 3组和4组之间。使用IOTA术语表的O-RADS和使用ADNEX模型的O-RADS的敏感性和特异性分别为92.2%和86.1%，以及85.9%和87.4%。
与IOTA ADNEX模型相比，使用IOTA术语表的O-RADS分类的诊断性能相似。然而，O-RADS组的指定存在显著差异，这取决于IOTA术语表的使用或使用ADNEX模型进行的风险估计。这一事实可能具有临床相关性，值得进一步研究。
