3T多参数前列腺MRI采用前列腺影像报告和数据系统(PIRADS)第2版与活检参考标准的相关性研究
研究纳入134例未治疗患者,行3T多参数前列腺MRI并以PIRADS v2评估,与MRI引导下前列腺活检结果对照。PIRADS≥3诊断临床显著前列腺癌的准确度89%、敏感度76.5%、特异度89.7%、阳性预测值31.7%、阴性预测值98.4%;PIRADS 3、4、5结节的临床显著前列腺癌检出率分别为6.1%、33.3%、64.4%。结论认为PIRADS≥3特异度和阴性预测值高,无对应PIRADS≥3结节时活检部位存在临床显著前列腺癌的可能性很低。
要点
- PIRADS≥3诊断临床显著前列腺癌的阴性预测值达98.4%,特异度89.7%。
- PIRADS 3、4、5结节的临床显著前列腺癌检出率分别为6.1%、33.3%、64.4%。
- 23/1676(1.4%)活检部位含临床显著前列腺癌但MRI未识别出结节。
核心结论
- 以活检为参考标准,PIRADS≥3对临床显著前列腺癌具有高特异度和高阴性预测值,无对应PIRADS≥3结节时活检部位存在临床显著前列腺癌的可能性很低。
研究设计与样本
- 研究纳入134例连续未治疗患者(平均年龄64岁,范围41-82岁),均行MRI引导下前列腺活检;采用MRI-TRUS融合活检77例(57.5%)、认知融合活检51例(38.0%),6例(4.5%)无目标结节仅行系统活检;共1676个活检部位中237个(14.1%)MRI显示PIRADS 3、4或5结节;58例(43.3%)患者有临床显著前列腺癌。
诊断标准与关键数值
- PIRADS≥3诊断临床显著前列腺癌的准确度89%、敏感度76.5%、特异度89.7%、阳性预测值31.7%、阴性预测值98.4%;PIRADS 3、4、5结节的临床显著前列腺癌检出率分别为6.1%(4/66)、33.3%(42/126)、64.4%(29/45);23/1676(1.4%)活检部位含临床显著前列腺癌但MRI未识别出结节。
临床应用与阅片要点
- MRI阅片者、活检操作者、融合活检方法和前列腺结节带区位置均未显著影响活检结果为临床显著前列腺癌的比值比。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
OBJECTIVE: To correlate the findings on 3T multiparametric prostate MRI using PIRADS version 2 with prostate biopsy results as the standard of reference.
MATERIALS AND METHODS: 134 consecutive treatment naive patients (mean age 64 years, range 41-82 years) underwent MRI-directed prostate biopsy. MRI-TRUS fusion biopsy was used for 77 (77/134 = 57.5%) patients, cognitive fusion for 51 (51/134 = 38.0%) patients, and 6 patients (6/134 = 4.5%) without a target nodule had systematic biopsy only. Out of the 1676 biopsy sites, 237 (237/1676 = 14.1%) were positive on MRI for a PIRADS 3, 4, or 5 nodule. Fifty-eight (58/134, 43.3%) patients had clinically significant prostate cancer (csPCa). The findings on MRI using PIRADS version 2 were correlated with the biopsy results.
RESULTS: The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of PIRADS ≥ 3 for csPCa were 89%, 76.5%, 89.7%, 31.7%, and 98.4%, respectively. The detection rates of csPCa for PIRADS 3, 4, and 5 nodules were 6.1% (4/66), 33.3% (42/126), and 64.4% (29/45), respectively. MRI did not identify a nodule in 23/1676 (1.4%) biopsy sites that contained csPCa. The MRI reader, biopsy operator, method of fusion biopsy, and zonal location of prostate nodule did not significantly affect the odds of having a biopsy result positive for csPCa.
CONCLUSION: PIRADS ≥ 3 had high specificity and high negative predictive value for csPCa using biopsy results as the standard of reference. The presence of csPCa from a biopsy site was highly unlikely in the absence of a corresponding PIRADS ≥ 3 nodule.
引用
- Mathur S, O'Malley ME, Ghai S, et al. Correlation of 3T multiparametric prostate MRI using prostate imaging reporting and data system (PIRADS) version 2 with biopsy as reference standard. Abdom Radiol (NY). 2019. PMID: 30032385. doi:10.1007/s00261-018-1696-8
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。