不同注射剂量方案CZT-SPECT心肌灌注显像快速采集时间的优化

Study on optimizing rapid acquisition time of CZT-SPECT myocardial perfusion imaging using different injection doses

目的探讨行一日法静息/负荷碲锌镉心脏专用SPECT(CZT-SPECT)心肌灌注显像(MPI)时使用高、低剂量方案的最优快速采集时间和临床实用性。方法回顾性分析101例可疑或确诊冠状动脉粥样硬化性心脏病患者的CZT-SPECT一日法门控MPI图像资料,其中2017年11月至12月间的高剂量组51例[男22例,女29例,年龄(55.4±8.5)岁],2018年7月至10月间的低剂量组50例[男27例,女23例,年龄(59.1±12.8)岁]。高剂量组静息显像99Tc^m-甲氧基异丁基异腈(MIBI)注射剂量为296~370 MBq,低剂量组为111~222 MBq;负荷显像剂量为静息时的3倍。高剂量组静息/负荷采集总时间为6 min/4 min,低剂量组为8 min/6 min。使用lister软件调用患者MPI原始表(list)文件,分别重建采集时间为1 min、2 min……至最长采集时间的MPI数据,对不同采集时间的MPI结果进行图像质量评价、半定量分析及左心室射血分数(LVEF)测定。采用配对t检验或Wilcoxon符号秩检验、Pearson相关分析及Bland-Altman分析对不同采集时间定量参数进行统计学分析。结果高剂量组静息/负荷显像分别采集3 min/2 min及以上时,低剂量组分别采集4 min/3 min及以上时,图像质量评价均为好或优秀[100%(51/51)与100%(50/50)];高剂量组静息/负荷显像采集时间3 min/2 min与6 min/4 min,低剂量组静息/负荷显像采集时间4 min/3 min与8 min/6 min的灌注参数[静息总积分(SRS)或负荷总积分(SSS)、总灌注缺损(TPD)]及LVEF比较,差异均无统计学意义(t值:-1.196~1.597,z值:-1.963~1.945,均P>0.05),且具有强相关性(均r>0.700,均P<0.001),Bland-Altman分析也显示定量结果一致性较好。结论CZT-SPECT一日法静息/负荷MPI高剂量方案最优采集时间为3 min/2 min,低剂量为4 min/3 min,通过优化快速采集时间进一步缩短了显像时间,在保证图像质量的前提下获得可靠的灌注参数与LVEF,具有较好的临床实用性。

Objective To explore the optimal and rapid imaging acquisition time and clinical applicability when using high- and low-dose one day rest/stress myocardial perfusion imaging (MPI) with cadmium-zinc-telluride heart dedicated SPECT (CZT-SPECT). Methods The MPI data with high-dose rest/stress protocol in 51 patients (22 males, 29 females, age:(55.4±8.5) years) between November 2017 and December 2017 and those with low-dose protocol in 50 patients (27 males, 23 females, age:(59.1±12.8) years) between July 2018 and October 2018 were retrospectively analyzed. The MPI was performed with CZT-SPECT. Each patient received 296-370 MBq rest dose of 99Tcm-methoxyisobutylisonitrile (MIBI) for high-dose protocol, 111-222 MBq rest dose of 99Tcm-MIBI for low-dose protocol. Stress dose was 3 times of the corresponding rest dose. Rest and stress scans were acquired 6 min and 4 min for high-dose protocol, while 8 min and 6 min for low-dose protocol in total. All rest or stress imaging data were reconstructed from list-mode raw data to obtain scan durations of 1 min, 2 min, etc, up to the maximum of acquisition time. Image quality at different acquisition times were evaluated, and myocardial perfusion and function parameters were compared. Paired t test, Wilcoxon signed rank test, Pearson correlation and Bland-Altman analyses were used for data analysis. Results The image quality was rated as excellent/good in all patients (100%, 51/51) when acquisition time ≥3 min and ≥2 min respectively for rest and stress imaging with high-dose protocol, the similar results was obtained (100%, 50/50) when acquisition time ≥4 min and ≥3 min respectively for rest and stress imaging with low-dose protocol. The quantitative perfusion parameters(summed rest scores, summed stress scores, total perfusion deficit) and left ventricle ejection fraction (LVEF) at shorter acquisition times (3 min/2 min, 4 min/3 min) were not significantly different from the results at the longer acquisition times (6 min/4 min, 8 min/6 min;t values: from -1.196 to 1.597, z values: from -1.963 to 1.945, all P>0.05). Those parameters at shorter and longer acquisition times showed strong correlations (all r>0.700, all P<0.001), and Bland-Altman analysis revealed good agreement between them. Conclusion The optimal acquisition time is 3 min/2 min for high-dose one-day rest/stress CZT-SPECT MPI, and 4 min/3 min for low-dose protocol, which can significantly shorten the MPI acquisition time, then reliable perfusion parameters and LVEF can be obtained under the premise of ensuring image quality, making it the better clinical applicability.