The relationship between perfusion defects on myocardial SPECT and stenotic severity on CT coronary angiography / 中华核医学杂志

ABSTRACT

ObjectiveTo evaluate the diagnostic performance of CTCA in predicting myocardial perfusion defects through comparative analysis between MPI defects and severity of coronary stenosis on CTCA.MethodsFour hundred and seventy-eight patients who underwent CTCA and 99Tcm-MIBI MPI in the same period were analyzed retrospectively.According to the severity of coronary stenosis judged by visual evaluation of the vessel diameter,the patients were divided into five groupsno stenosis,mild stenosis,moderate stenosis,severe stenosis and total occlusion.MPI results were classified as negative or positive for perfusion defects,and the prevalence of perfusion defects in every group was calculated per-patient and per-vessel basis.The cut-off of stenotic severity for predicting myocardial perfusion defects was designated as 50％ or 75％,with MPI as standard reference.True positive,true negative,false positive and false negative statistics were thus determined separately on patient and vessel basis.The diagnostic performance for CTCA were calculated and compared.Pearson Chi-square and its partition tests or Fisher exact test were used to compare ordinal variables.ResultsFifty-eight patients showed myocardial perfusion defects.Either by patientbased or vessel-based analysis,the prevalence of myocardial perfusion defects showed an increasing trend with greater coronary artery stenosis in each group,and there were statistical differences among them (x2 =116.62 and 483.83,both P ＜ 0.05).On patient-based analysis,sensitivity ( SN),specificity ( SP),positive predictive value( PPV),negative predictive value(NPV) and accuracy (AC) for CTCA predicting myocardial perfusion defects were 62.1 ％ ( 36/58 ) and 34.5％ ( 20/58 ) (x2 =8.84,P ＜ 0.05 ),84.5％(355/420) and 97.1％ (408/420) (x2 =40.16,P ＜0.05),35.6％ (30/101) and 62.5％ (20/32) (x2 =7.19,P＜0.05),94.2％ (355/377) and 91.5％ (408/446) (x2 =2.18,P ＞0.05),81.8％ (391/478)and 89.5 ％ (428/478) (x2 =11.66,P ＜ 0.05 ) when the cutoff was set to 50％ and 75％,respectively.On vessel-based analysis,the SN,SP,PPV,NPV and AC for CTCA predicting myocardial perfusion defects were 58.8％ (40/68) and 30.9％ (21/68) (x2 =10.73,P ＜ 0.05),95.9％ (1768/1844) and 99.0％ (1826/1844) (x2 =36.72,P ＜ 0.05 ),34.5％ (40/116) and 53.8％ (21/39) (x2 =4.59,P ＜0.05 ),98.4％ (1768/1796) and 97.5％ ( 1826/1873 ) (x2 =4.14,P ＜ 0.05 ),94.6％ ( 1808/1912 ) and 96.6％ ( 1847/1912 ) (x2 =10.31,P ＜ 0.05 ),respectively.ConclusionsThe prevalence of myocardial perfusion defects correlates positively with the severity of coronary stenosis seen on CTCA.CTCA may predict perfusion defects with high SP and NPV.However,the PPV of CTCA in predicting myocardial perfusion defects is poor when the stenosis cut-off is set at 50％.It is significantly improved when the cutoff value is set at 75 ％.