全自由呼吸心脏MRI与屏气技术的临床可行性、效率及诊断一致性比较
这项前瞻性研究在3.0-T扫描仪上比较了全自由呼吸与屏气心脏MRI方案,纳入605名参与者(平均年龄48岁±17,422名男性)。自由呼吸方案将扫描时间从34.4分钟±7.5缩短至20.9分钟±3.1(P<.001),各序列图像质量保持诊断水平,左心室功能参数与组织特征与屏气法一致性优良(组内相关系数0.98-0.99)。
要点
- 自由呼吸心脏MRI方案扫描时间20.9分钟±3.1,较屏气法34.4分钟±7.5缩短(P<.001)。
- 自由呼吸电影、T1 mapping和LGE采集时间分别减少54%、28%和33%。
- 自由呼吸与屏气法左心室功能参数组内相关系数0.98-0.99,观察者间一致性κ=0.80-0.93。
核心结论
- 全自由呼吸心脏MRI在减少扫描时间的同时,保持诊断图像质量,且各序列与屏气技术高度一致。
研究设计与样本
- 前瞻性研究,2024年6月至2024年8月,使用3.0-T扫描仪,纳入605名参与者(平均年龄48岁±17,422名男性),比较自由呼吸与屏气MRI采集。
诊断标准与关键数值
- 自由呼吸电影图像质量评分4(IQR 4-4) vs 屏气5(IQR 4-5),P<.001;T1/T2 mapping和血流序列评分均为5(P>.05);LGE评分自由呼吸5(IQR 5-5) vs 屏气5(IQR 4-5),P<.001。
- 观察者间一致性κ=0.80-0.93。
- 自由呼吸左心室舒张末期容积189.94 mL±84.74 vs 屏气192.92 mL±86.32(P<.001);收缩末期容积104.99 mL±83.14 vs 108.52 mL±84.45(P<.001);射血分数50.01%±17.93 vs 48.92%±17.66(P<.001)。
- 自由呼吸与屏气法左心室功能参数组内相关系数0.98-0.99。
- 自由呼吸原生T1值和LGE质量略高(P≤.001),但与屏气法一致性优良。
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
Purpose To evaluate the feasibility, efficiency, and diagnostic concordance of a full comprehensive free-breathing cardiac MRI protocol. Materials and Methods In this prospective study conducted between June 2024 and August 2024, free-breathing and breath-holding MRI acquisitions were compared across all sequences using a 3.0-T scanner. Outcomes included scanning time, image quality, left ventricular (LV) functional parameters, and tissue characterization. Paired comparisons between free-breathing and breath-holding acquisitions were performed using matched-pairs Wilcoxon tests or t tests as appropriate. Results In this analysis of 605 participants (mean age ± SD, 48 years ± 17; 422 male), the full free-breathing cardiac MRI protocol reduced scanning duration compared with breath-holding techniques (20.9 minutes ± 3.1 vs 34.4 minutes ± 7.5; P < .001); free-breathing cine, T1 mapping, and late gadolinium enhancement (LGE) acquisitions achieved 54%, 28%, and 33% time reductions, respectively. Free-breathing cardiac MRI consistently achieved diagnostic image quality: Cine had a slightly lower score (4 [IQR, 4-4] vs 5 [IQR, 4-5]; P < .001), whereas T1/T2 mapping and flow showed similar scores of 5 (P > .05). LGE scores were higher with free-breathing (5 [IQR, 5-5] vs 5 [IQR, 4-5]; P < .001). Interobserver agreement was strong (κ = 0.80-0.93). Compared with the breath-holding measurement, free-breathing LV end-diastolic volume and LV end-systolic volume were higher (189.94 mL ± 84.74 vs 192.92 mL ± 86.32, P < .001; 104.99 mL ± 83.14 vs 108.52 mL ± 84.45, P < .001, respectively), whereas LV ejection fraction was lower (50.01% ± 17.93 vs 48.92% ± 17.66, P < .001), with excellent agreement between methods (intraclass correlation coefficient, 0.98-0.99). Native T1 values and LGE mass were slightly higher with free-breathing (P ≤ .001) but remained in excellent agreement with breath-holding methods. Conclusion Full free-breathing cardiac MRI reduced scanning duration while maintaining diagnostic image quality and strong agreement with breath-holding techniques across all sequences. Keywords: Cardiac MRI, Free-breathing, Breath-holding, Cardiac Imaging Supplemental material is available for this article. © RSNA, 2026.
引用
- Zhou W, Yin G, An J, et al. Clinical Feasibility, Efficiency, and Diagnostic Concordance of Full Free-breathing Cardiac MRI Compared with Breath-holding Techniques. Radiol Cardiothorac Imaging. 2026. PMID: 41989284. doi:10.1148/ryct.250258
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。