PI-QUAL 临床应用系统综述:可重复性、诊断影响与教育培训
结果显示,背景和经验相近的阅片者间一致性多为中度至几乎完美,结构化培训可提高评分表现并缩小专业组间差异。较高的 PI-QUAL 评分与临床显著性前列腺癌检出敏感度提高、PI-RADS 3 模棱两可评估减少及更可靠的局部分期相关,但特异度、阳性预测值和阴性预测值的改善不一致。
要点
- PI-QUAL 评分在背景相近的阅片者间一致性多为中度至几乎完美,可用于前列腺 MRI 质控。
- 结构化培训能改善 PI-QUAL 评分表现,减少不同专业组间的变异。
- PI-QUAL 评分较高与临床显著性前列腺癌检出敏感度提高及 PI-RADS 3 评估减少相关。
核心结论
- 现有证据支持将 PI-QUAL 作为前列腺 MRI 可重复且具有临床相关性的质量保证框架,在常规实践中更广泛地整合应用。
研究设计与样本
- 系统综述,依据 PRISMA 指南检索 PubMed、Medline Ovid 和 Embase,检索时间截至 2026 年 6 月。
- 采用 QUADAS-2 工具评估偏倚风险。
关键影像表现/征象
- 结构化教育可一致地改善 PI-QUAL 评分表现,并降低不同专业组之间的评分变异。
- 较高的 PI-QUAL 评分与临床显著性前列腺癌检出敏感度提高相关。
- 较高的 PI-QUAL 评分与模棱两可的 PI-RADS 3 评估数量减少相关。
- 较高的 PI-QUAL 评分与更可靠的局部肿瘤分期相关。
- PI-QUAL 评分改善与特异度、阳性预测值和阴性预测值的关联不一致。
诊断标准与关键数值
- 阅片者间一致性:背景和经验相近的阅片组间 PI-QUAL 一致性一般处于中度至几乎完美的范围。
- 方法学异质性是导致不同研究间 PI-QUAL 一致性存在差异的影响因素之一。
临床应用与阅片要点
- PI-QUAL 可作为前列腺 MRI 日常质控的客观工具,建议阅片者掌握评分标准并用于评估图像质量。
- 对阅片者进行结构化 PI-QUAL 培训有助于提高评分一致性,不同专业背景的医师均可从中获益。
- 图像质量评分较低时需警惕临床显著性病变漏检风险增加及 PI-RADS 3 评估增多,报告时应予以关注。
局限性
- 纳入研究间存在方法学异质性,影响 PI-QUAL 一致性结果的横向比较。
- 仍需前瞻性多中心研究评估实施策略及患者层面的临床结局。
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
INTRODUCTION: Variation in prostate MRI image quality persists despite established acquisition guidelines, highlighting the need for objective and reproducible methods of quality assessment. The Prostate Imaging Quality (PI-QUAL) score has been proposed to address this. This systematic review evaluates the inter-reader reproducibility of PI-QUAL scoring, its impact on prostate MRI interpretation, and factors influencing its implementation.
METHODS: A systematic review was conducted in accordance with PRISMA guidelines, with searches of PubMed, Medline Ovid and Embase from inception to June 2026. Studies were included if PI-QUAL was evaluated as a primary or defined secondary outcome. Risk of bias was assessed using QUADAS-2.
RESULTS: Thirty-one studies were included. Inter-reader agreement for PI-QUAL generally ranged from moderate to near-perfect among reader groups of similar background and experience, although methodological heterogeneity contributed to variability across studies. Structured education consistently improved PI-QUAL scoring performance and reduced variability between professional groups. Higher PI-QUAL scores were associated with improved sensitivity for clinically significant prostate cancer detection, fewer equivocal PI-RADS 3 assessments, and more reliable local tumour staging, whereas improvements in specificity, positive predictive value, and negative predictive value were less consistent.
CONCLUSION: Current evidence supports broader implementation of PI-QUAL as a reproducible and clinically relevant quality assurance framework for prostate MRI. Its reproducibility across different reader groups and association with improved MRI interpretation support integration into routine prostate MRI practice, although prospective multicentre studies evaluating implementation strategies and patient-level clinical outcomes remain required.
引用
- Ng, Cabbabe, Li, et al. Implementing PI-QUAL in Clinical Practice: A Systematic Review of Reproducibility, Diagnostic Impact, and the Role of Education and Automation. Eur J Radiol. 2026. PMID: 42574769. doi:10.1016/j.ejrad.2026.113115
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。