文献速递泌尿生殖腹部MR

前列腺MRI质量评估:PI-QUAL v2在1.5T与3T扫描仪间的比较及其与T分期一致性的关联

Prostate MRI quality assessment using PI-QUAL version 2: Comparative analysis between 1.5 T and 3 T scanners and its association with T-staging agreement.
PubMed 文献 · Eur J Radiol · 2026-03-25

本回顾性研究纳入416例术前行前列腺MRI并接受根治性前列腺切除术的患者,使用PI-QUAL v2评估图像质量并比较1.5T与3T扫描仪的质量分布,同时分析图像质量与MRI-based T分期和病理T分期一致性的关联。结果显示3T组可接受图像质量比例显著高于1.5T组(85.8% vs. 75.2%,p=0.004),且PI-QUAL评分越高,T分期一致性越好(OR 1.59,95%CI 1.17-2.16,p=0.003)。

要点

  • 3T扫描仪的前列腺MRI可接受图像质量比例显著高于1.5T(85.8% vs. 75.2%,p=0.004)。
  • PI-QUAL v2评分越高,MRI与病理T分期一致性越好(OR 1.59,95%CI 1.17-2.16,p=0.003)。
  • PI-QUAL v2可作为前列腺MRI临床质量保证的实用工具。

核心结论

  • 更高的前列腺MRI图像质量(PI-QUAL v2评估)与MRI-based和病理T分期一致性的改善显著相关。
  • 研究结果支持标准化图像质量评估的临床相关性,并强调PI-QUAL v2作为前列腺MRI实用质量保证工具的作用。

研究设计与样本

  • MRI检查使用1台1.5T扫描仪和2台3T扫描仪;图像质量由两名放射科医生独立使用PI-QUAL v2评估。
  • MRI-based T分期(mrT2、mrT3a或mrT3b)由共识确定,并与病理T分期比较。
  • 采用 logistic 回归分析PI-QUAL评分与T分期不匹配的关联;采用百分比一致性和Gwet一致性系数评估观察者间一致性。

诊断标准与关键数值

  • 可接受图像质量定义为PI-QUAL v2 ≥ 2;3T组可接受质量比例为85.8%,1.5T组为75.2%(p=0.004)。
  • 三台扫描仪间可接受图像质量比例差异显著(1.5T:75.2%,MR1:82.6%,MR2:89.1%;p=0.026)。
  • PI-QUAL评分在MRI-based与病理T分期一致的病例中显著高于不匹配病例(p=0.010)。
  • 更高的PI-QUAL评分独立关联于改善的T分期一致性(OR 1.59;95%CI 1.17-2.16;p=0.003),尤其在3T亚组中。
  • PI-QUAL v2参数的观察者间一致性总体为中等至几乎完全一致,扩散加权成像参数存在变异性。

临床应用与阅片要点

  • 日常前列腺MRI阅片和报告应关注图像质量,使用PI-QUAL v2进行标准化评估有助于识别可能影响T分期判断的低质量扫描。
  • 对于PI-QUAL v2评分较低的检查(尤其1.5T扫描仪),需警惕T分期低估或高估风险,必要时建议复查或补充序列。
  • 扩散加权成像参数的观察者间一致性存在变异,报告时需注意该序列质量对分期判断的影响。

局限性

  • 回顾性研究设计。
  • 单中心研究。
  • 样本仅纳入接受根治性前列腺切除术的患者,可能存在选择偏倚。

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

PURPOSE: To assess prostate MRI image quality using Prostate Imaging Quality (PI-QUAL) version 2, to compare quality distributions between 1.5 T and 3 T MRI scanners, and to evaluate the association between image quality and concordance of MRI-based and pathological T-staging.

MATERIALS AND METHODS: This retrospective study included 416 patients who underwent preoperative prostate MRI followed by radical prostatectomy between December 2021 and August 2024. MRI examinations were performed using one 1.5 T scanner and two 3 T scanners. Image quality was independently evaluated by two radiologists using PI-QUAL v2. MRI-based T-staging (mrT2, mrT3a, or mrT3b) was determined by consensus and compared with pathological T-staging. Differences in PI-QUAL scores were analyzed across scanners, and the association between PI-QUAL score and T-stage mismatch was evaluated using logistic regression. Inter-observer agreement was assessed using percent agreement and Gwet's agreement coefficients.

RESULTS: The proportion of examinations with acceptable image quality (PI-QUAL v2 ≥ 2) was significantly higher in 3 T group than in 1.5 T group (85.8% vs. 75.2%, p = 0.004). Among the three MRI scanners, the proportion of examinations with acceptable image quality differed significantly across scanners, with the highest proportion observed in MR2 (1.5 T: 75.2%, MR1: 82.6%, MR2: 89.1%; p = 0.026). PI-QUAL scores were significantly higher in cases with concordant MRI-based and pathological T-staging than in mismatched cases (p = 0.010). Higher PI-QUAL scores were independently associated with improved T-staging agreement (odds ratio, 1.59; 95% confidence interval, 1.17-2.16; p = 0.003), particularly in the 3 T subgroup. Inter-observer agreement for PI-QUAL v2 parameters was generally substantial to almost perfect, with variability observed for diffusion-weighted imaging parameters.

CONCLUSION: Higher prostate MRI image quality, as assessed by PI-QUAL v2, is significantly associated with improved concordance between MRI-based and pathological T-staging. These findings support the clinical relevance of standardized image quality assessment and highlight the role of PI-QUAL v2 as a practical quality assurance tool in prostate MRI.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Song JS, Jang W. Prostate MRI quality assessment using PI-QUAL version 2: Comparative analysis between 1.5 T and 3 T scanners and its association with T-staging agreement. Eur J Radiol. 2026. PMID: 41916137. doi:10.1016/j.ejrad.2026.112821

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。