相位分析SPECT心肌灌注显像检测心肌瘢痕不同步并提高特异性
本研究前瞻性纳入340例推导队列和138例验证队列,评估相位分析(PA)能否提高SPECT心肌灌注显像(MPI)诊断心肌瘢痕的特异性。结果显示熵值(≥59%)在两个队列中均与瘢痕独立相关(OR>5),加入熵值后AUC从0.59提升至0.67(p=0.04),验证队列中总体NRI为18%。
要点
- 相位分析熵值≥59%与心肌瘢痕独立相关,两队列OR均大于5。
- 熵值加入临床/影像变量后AUC显著改善(0.67 vs 0.59,p=0.04)。
- 熵值将24%无瘢痕患者正确重分类至较低风险组,总体NRI为18%。
核心结论
- 相位分析熵值显著提高SPECT MPI诊断心肌瘢痕的特异性,可能影响诊断报告、预后判断和治疗决策。
研究设计与样本
- 前瞻性研究,推导队列340例因SPECT-MPI转诊患者,验证队列138例独立患者。
- 以静息MPI灌注缺损作为瘢痕参考标准,按预检概率分为≤10%、11-74%、≥75%三组。
- 使用净重分类指数(NRI)评估相位分析变量对瘢痕和无瘢痕患者的重分类能力。
主要发现
- 熵值(≥59%)在两个队列中均与瘢痕独立相关,OR大于5。
- 熵值加入多临床/影像变量后,评估瘢痕的AUC从0.59提高至0.67(p=0.04)。
- 验证队列中熵值将24%经临床模型判定无瘢痕的患者正确重分类至较低风险类别,总体NRI为18%。
关键数值与推荐
- 熵值阈值≥59%用于判定与瘢痕的独立关联。
- AUC比较:0.67 vs 0.59,p=0.04。
- 验证队列总体NRI为18%,24%无瘢痕患者被正确降级。
临床怎么用
- 相位分析无需额外成像时间或辐射,可作为SPECT MPI解读的辅助工具。
- 熵值≥59%提示心肌瘢痕可能性增高,有助于提高MPI特异性并减少假阳性。
- 相位分析结果可辅助将无瘢痕患者降级,减少不必要的后续检查或治疗。
局限性
- 研究为单中心前瞻性设计,推导与验证队列样本量有限。
- 以静息MPI灌注缺损作为瘢痕参考标准,未与其它影像或临床金标准对比。
- 相位分析参数的普适性需更大规模多中心研究验证。
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)。
原文
BACKGROUND: Myocardial perfusion imaging (MPI) with single-photon emission computed tomography (SPECT) is commonly used to assess patients with cardiovascular disease. However, in certain scenarios, it may have limited specificity in the identification of hemodynamically significant coronary artery disease (e.g., false positive), potentially resulting in additional unnecessary testing and treatment. Phase analysis (PA) is an emerging, highly reproducible quantitative technology that can differentiate normal myocardial activation (synchrony) from myocardial scar (dyssynchrony). The objective of this study is to determine if PA can improve the specificity SPECT MPI. METHODS: An initial cohort of 340 patients (derivation cohort), referred for SPECT-MPI, was prospectively enrolled. Resting MPI studies were assessed for resting perfusion defects (scar). These were utilized as the reference standard for scar. Subsequently, we collected a second independent validation cohort of 138 patients and tested the potential of PA to reclassify patients for the diagnosis of "scar" or "no scar." Patients were assigned to three categories depending upon their pre-test probability of scar based on multiple clinical and imaging parameters: ≤ 10% (no scar), 11-74% (indeterminate), and ≥ 75% (scar). The ability of PA variables to reclassify patients with scar to a higher group and those without scar to a lower group was then determined using the net reclassification index (NRI). RESULTS: Entropy (≥ 59%) was independently associated with scar in both patient cohorts with an odds ratio greater than five. Furthermore, when added to multiple clinical/imaging variables, the use of entropy significantly improved the area under the curve for assessment of scar (0.67 vs. 0.59, p = 0.04). The use of entropy correctly reclassified 24% of patients without scar, by clinical model, to a lower risk category (as determined by pre-test probability) with an overall NRI of 18% in this validation cohort. DISCUSSION: The use of PA entropy can improve the specificity of SPECT MPI and may serve as a useful adjunctive tool to the interpreting physician. The current study determined the optimal PA parameters to detect scar (derivation cohort) and applied these parameters to a second, independent, patient group and noted that entropy (≥ 59%) was independently associated with scar in both patient cohorts. Therefore, PA, which requires no additional imaging time or radiation, enhances the diagnostic capabilities of SPECT MPI. CONCLUSION: The use of PA entropy significantly improved the specificity of SPECT MPI and could influence the labeling of a patient as having or not having myocardial scar and thereby may influence not only diagnostic reporting but also potentially prognostic determination and therapeutic decision-making.
引用
- Bois JP, Scott C, Chareonthaitawee P, et al. Phase analysis single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) detects dyssynchrony in myocardial scar and increases specificity of MPI. EJNMMI Res. 2019. PMID: 30706258. doi:10.1186/s13550-019-0476-y
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。