Incidence and etiological analysis of abnormal cardiac uptake in patients underwent oncologic PET/CT imaging / 中华心血管病杂志

Objective: To investigate the incidence, location and etiology of abnormal cardiac uptake in patients underwent oncologic 18F-fluorodeoxyglucose positron emission tomography/computed tomography(PET/CT) imaging. Methods: The 18F-FDG PET/CT images of 2000 consecutive patients with suspected or diagnosed malignancy in Beijing Chaoyang Hospital from January 2014 to September 2016 were retrospectively analyzed. Fasting time was more than 12 hours before imaging, and fasting blood glucose level before 18F-FDG injection was less than 6.7 mmol/L. Focal uptake in the non-basal and non-papillary regions of the left ventricle, uptake in the right ventricle exceeding uptake in the left ventricle, and uptake in the atrium higher than that of the blood pool (when uptake of left ventricle was zero or low) were defined as abnormal, and all abnormal uptake was visually determined by experienced nuclear medicine physicians. General clinical data and the results of cardiac examination were collected to explore the incidence, location and etiology of cardiac diseases. Results: There were 138 patients with history of diabetes (6.90%), 228 patients with history of cardiovascular disease (11.40%) out of the 2 000 patients ((60.5±13.2) years, 1 117 male (55.85%)). The number and proportion of patients with malignancy, benign lesions, diseases of unknown etiology were 939 (46.95%), 484 (24.20%), 557 (28.85%), respectively. Abnormal cardiac uptake was detected in 145 patients (7.3%). The proportion of abnormal uptake in left ventricle, right ventricle and atrium was 52.4% (76 cases), 12.4% (18 cases), 35.2% (51 cases), respectively. Of the 76 individuals who had abnormal uptake in left ventricle, 25 cases (32.9%) were caused by coronary artery disease, and other causes included hypertrophic cardiomyopathy and myocardial damage caused by chemotherapy drugs, etc. Of the 18 cases who had abnormal uptake in right ventricle, 14 cases (14/18) were caused by pulmonary hypertension. In addition, 20 out of the 51 cases (39.3%) with abnormal uptake in atrium suffered from atrial fibrillation. Seventy-one patients with abnormal cardiac uptake (49.0%) had no clear manifestation and evidence of heart disease before imaging. Conclusions: The abnormal 18F-FDG uptake on oncologic PET/CT is not rare. The most common site of abnormal uptake is left ventricle, coronary artery disease, pulmonary hypertension and atrial fibrillation are common causes of abnormal 18F-FDG uptake.

Value of absolute quantification of myocardial perfusion by PET in detecting coronary microvascular disease in patients with non-obstructive coronaries / 中华心血管病杂志

Objective: To compare the incidence of coronary microvascular disease (CMVD) between patients with non-obstructive and obstructive coronary arteries. Methods: We retrospectively analyzed 97 patients with angina pectoris, who underwent the absolute quantitative PET examination of myocardial perfusion and coronary anatomy examination within 90 days. All patients were divided into two groups: non-obstructive group (72 cases, no stenosis ≥50% in all three coronary arteries) and obstructive group (25 cases, at least one coronary stenosis ≥50%; and at least one coronary stenosis<50%). Quantitative parameters derived from PET including rest myocardial blood flow (RMBF), stress myocardial blood flow (SMBF), coronary flow reserve (CFR) and cardiovascular risk factors were compared between the two groups. CMVD was defined as CFR<2.90 and SMBF<2.17 ml·min(-1)·g(-1). Results: Incidence of CMVD was significant higher in the non-obstructive coronary arteries of the obstructive group than in the non-obstructive coronary arteries of non-obstructive group (47.1% (16/34) vs. 25.5% (55/216), χ(2)=6.738, P=0.009) while incidence of CMVD was similar between non-obstructive and obstructive patients ((44% (11/25) vs. 33.3% (24/72), χ(2)=0.915, P=0.339). RMBF ((0.83±0.14) ml·min(-1)·g(-1) vs. (0.82±0.17) ml·min(-1)·g(-1)), SMBF ((2.13±0.60) ml·min(-1)·g(-1) vs. (1.91±0.50) ml·min(-1)·g(-1)) and CFR (2.59±0.66 vs. 2.36±0.47) were similar between the two groups (all P>0.05). Conclusions: CMVD can occur in non-obstructive coronary arteries in both patients with non-occlusive coronary arteries and patients with obstructive coronary arteries. Prevalence of CMVD is significantly higher in patients with obstructive coronary arteries than in patients with non-obstructive coronary arteries. The CMVD severity is similar between the two groups.

Differentiation of true from false left ventricular aneurysm with magnetic resonance imaging in patients after myocardial infarction / 中华心血管病杂志

<p><b>OBJECTIVE</b>To observe the value of cardiac magnetic resonance imaging (MRI) for differentiation of true from false left ventricular aneurysm in patients after myocardial infraction (MI).</p><p><b>METHODS</b>Twenty-six patients [22 males/4 females, mean age (59.3 ± 9.3) years] with left ventricular aneurysm after MI were imaged with MRI, echocardiography and coronary angiography. The respective findings were compared with surgical pathology results.</p><p><b>RESULTS</b>There were 24 patients with dyspnea and 15 patients with hypertension. LVEF measured by echocardiography was 36.9% ± 9.1% in this patient cohort. Cardiac MRI showed that the left ventricular end diastolic wall thickness was thinner than 5.5 mm in 24 cases, and between 5.5 to 8 mm in 2 cases. The dimension of left ventricle was (67.8 ± 9.3) mm. Dyskinesia presented in 24 cases, and akinesia in 2 cases. Delayed enhancement was shown in all cases by MRI. Cardiac MRI detected left ventricular true aneurysm in 23 cases, false aneurysm in 3 case and left ventricular thrombi in 7 cases. The diagnosis by magnetic resonance imaging corresponded well to pathological findings. Echocardiography misdiagnosed pseudoaneurysm in 1 patient, and failed to detected left ventricular thrombi in 2 cases.</p><p><b>CONCLUSION</b>Cardiac MRI could correctly differentiate true from false left ventricular aneurysm in patients after MI.</p>

Clinical characteristics of fat replacement of left ventricular myocardium / 中华心血管病杂志

<p><b>OBJECTIVE</b>To evaluate the clinical characteristics of left ventricular fat replacement.</p><p><b>METHODS</b>We identified 45 patients [28M/17F, mean age (51.9 ± 14.7) years] with left ventricular myocardial fat replacement (CT value ≤ -30 Hu) by cardiovascular CT.</p><p><b>RESULTS</b>Among 45 patients, 25 patients [20M/5F, mean age (61.2 ± 10.4) years]were diagnosed as coronary artery disease (CAD). There was 56%single-vessel disease, 20% double-vessel disease and 24%triple-vessel disease, true left ventricular aneurysm was detected in 3 patients and left ventricular thrombi in 1 patient, the dimension of left ventricle was (54.5 ± 9.4) mm and the LVEF was (51.8 ± 13)% in CAD group. In this group, fat replacement occurred in the region of myocardial infarction and presented as curvilinear band in subendocardial region. The left ventricular wall thickness was lower than 5 mm in 21 cases. The location of fat replacement in CAD group is as follows: apical region in 18 patients, distal septal in 15 patients, distal anterior in 11 patients, mid-septal in 7 patients, mid-anterior in 7 patients and basal in 1 patients. The age of remaining 20 patients (8M/12F) without CAD were (57.8 ± 13.3) years. In the group of non-CAD, dilated cardiomyopathy was diagnosed in 3 patients, atrial septal defect in 1 patient, rheumatic heart disease in 1 patient, there was no structural heart disease in the remaining 15 patients. The dimension of left ventricle was (51.1 ± 9.1) mm and the LVEF was (59.4 ± 13.9)%. In non-CAD group, fat replacement mainly occurred in septal region, presented as curvilinear band in 17 patients and patch in 3 patients. The location of fat replacement in this group is as follows: mid-septal region in 11 patients, distal-septal in 10 patients and apical in 9 patients. The intramural fat replacement was detected in 14 patients: subendocardial fat replacement in 10 patients and both intramural and subendocardial fat replacement in 4 patients.</p><p><b>CONCLUSIONS</b>Left ventricular fat replacement could be documented in CAD patients, non-CAD cardiomyopathy patients and in patients without structural heart disease. Left ventricular fat replacement often positioned in apical region in CAD patients as a consequence of infarct healing while mostly positioned in septal region in non-CAD patients, the definite clinical implication of left ventricular fat replacement in non-CAD patients remains to be clarified.</p>

Impact of myocardial perfusion imaging on in-hospital coronary angiography and revascularization of patients with suspected coronary artery disease / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>Noninvasive cardiac imaging is now central to the diagnosis and management of patients with moderate probability for coronary artery disease. The aim of this study was to assess the impact of stress myocardial perfusion single photon emission computerized tomography (SPECT) on in-hospital coronary angiography and revascularization for such patients.</p><p><b>METHODS</b>Between January 2005 and June 2007, 1053 consecutive in-hospital patients (423 women, the average age of (57.2 ± 11.2) years) with suspected coronary artery disease but without any prior interventional treatment were retrospectively analyzed. All patients underwent a 2-day stress/rest (99m)Tc-methoxyisobutylisonitrile (MIBI) myocardial perfusion SPECT, including 984 exercise test and 69 adenosine test.</p><p><b>RESULTS</b>Overall, stress/rest myocardial perfusion SPECT was normal in 973 patients (92.4%) and abnormal in 80 patients (7.6%). A total of 190 patients underwent coronary angiography, 46 underwent percutaneous coronary intervention and 10 coronary artery bypass grafting during hospitalization. From the whole perspective, only 14.7% of patients with normal SPECT underwent coronary angiography, so did 58.8% of patients with abnormal SPECT (χ(2) = 97.0, P < 0.001); furthermore, the rates of revascularization in patients with normal and abnormal SPECT were 2.8% and 36.3%, respectively (27 out of 973 vs. 29 out of 80, χ(2) = 157.9, P < 0.001). The extent and severity of ischemia did not add more predictive value for subsequent coronary angiography, but did have impact on revascularization. Multivariate analysis showed that reversible perfusion defect was the most predictive variable for referral rate to coronary angiography (odds ratio = 7.5, P < 0.001).</p><p><b>CONCLUSIONS</b>Abnormal myocardial perfusion SPECT is a powerful referral for in-hospital coronary angiography and revascularization during the same hospitalization. Thus, stress/rest SPECT is an effective gatekeeper for early coronary angiography and invasive treatment for patients with suspected coronary artery disease.</p>

Clinical application of stress/rest myocardial perfusion imaging in the patients with 50 ％-75 ％ coronary stenosis / 中华核医学杂志

Objective To evaluate the clinical application of stress/rest MPI in the patients with 50％-75％ coronary artery stenosis.Methods The criteria for patient selection were that the patients should have at least one main coronary artery with stenosis more than 50％,and the maximal stenosis should be less than 75％ according to CAG.The stress/rest MPI was performed in 2 weeks before or after CAG.A total of 244 patients (178 males,66 females) with mean age (57 ± 10) years were included in this study.Symptom restriction stress test was used and stress MPI was performed 1 - 1.5 h after 99Tcm-MIBI (925 MBq)injection at the exercise peak.Rest MPI was performed within 48 - 72 h after stress MPI.Myocardial ischemia was diagnosed when there was a reduced uptake or even a defect in 2 different tomographic sections or in the same part of a myocardium in the continuous 2 slices.When there was an irreversible reduced uptake or defect,myocardial infarction was given as the final diagnosis.No reduced uptake or defect in all slices was shown as normal.The impact of MPI images on the selection for optimal clinical therapy plans was also discussed.X2 test was used for statistical analysis.Results A total of 340 coronary arteries with stenosis 50％ - 75％ were found by CAG.According to stress/rest MPI results,207 patients (84.8％)presented normal,33 had myocardial ischemia,3 had myocardial infarction,and 1 had both myocardial infarction and ischemia.In abnormal MPI images,there were 61 ischemic segments and 9 infarct segments,which were associated with 43 stenotic arteries (23 LAD,10 LCX,and 10 RCA).Patients were divided into 2 groups according to the results of MPI:Group 1 with normal MPI (207/244,84.8％ ) and Group 2 with abnormal MPI (37/244,15.2％ ).In Group 1,9 patients underwent coronary artery revascularization (PTCA or CABG),and the others had medical treatment.Eight patients had PTCA and 29 patients had medical treatment in Group 2.There was a statistically significant difference between the two groups ( X 2 =11.9,P =0.001 ).Conclusion Stress-rest MPI may be an effective method to evaluate ischemia degree for patients with 50％ -75％ coronary artery stenosis and a useful indicator for the individual treatment.

Alterations in electrocardiogram of adenosine test for 99Tcm-MIBI myocardial perfusion imaging / 中华核医学杂志

Objective To analyze alterations in electrocardiogram (ECG) of adenosine test in 99Tcm-MIBI myocardial perfusion imaging(MPI)SPECT study. Methods A total of 641 patients were in cluded in the study. The patients each underwent 99Tcm-MIBI MPI with adenosine test. The ECGs were taken before, during, and after adenosine infusion. Results In all, abnormal ECGs were found in 205(32.0%) patients. During adenosine infusion, 20.6%(132/641) of patients suffered from arrhythmia,29.5%(39/132) had atrial premature beats, 34. 1% (45/132) had premature ventricular beats, and 6. 1% (8/132) had sinoatrial block. In addition, 5.3% (7/132) had first-, 24.2% (32/132) had second-, and 0.8%(1/132) had third-degree atrioventricular block (AVB). After adenosine infusion, 4.4%( 28/641) of patients suffered from arrhythmia, 57.1% (16/28) had atrial premature beats, 39.3%(11/28) had premature ventricular beats, and 3.6% (1/28) had sinoatrial block. The perfusion images showed ischemia in 36 patients and infarction in 8 patients. Adenosine infusion was terminated in 39 patients (6. 1%) because of poorly tolerated side effects. However, no death or acute myocardial infarction occurred in the study. Conclusions Adenosine pharmacologic test for 99TcmMIBI MPI may result in relatively high incidence of arrhythmia in ECG monitoring.

18F-FDG/99Tcm-MIBI SPECT myocardial imaging for the detection of myocardial viability in patients with acute myocardial infarction / 中华核医学杂志

Objective To investigate the value of 18F-FDG/99Tcm-MIBI SPECT myocardial imaging for the detection of myocardial viability and prognosis in patients with AMI. Methods 18F-FDG/99Tcm-MIBI SPECT myocardial imaging was performed in 98 consecutive patients [man 87, women 11; average age (58 ±11)y] with AMI. The myocardium was scored individually for nine segments: mildly decreased uptake = 1,significantly decreased uptake = 2, and no uptake = 3. Perfusion defect but preserved 18 F-FDG uptake was defined as perfusion-metabolism mismatch, indicating jeopardized but viable myocardium. Perfusion defect and decreased 18 F-FDG uptake were defined as match, indicating myocardial necrosis. Echocardiogram was performed before and after treatment for evaluating the LVEF. All patients were followed after treatment.The rate of cardiac events was calculated and compared between patients with medication and revascularization. Paired t test, Chi-square test and log-rank test were used for statistical analysis. Results In the group with viable myocardium, 27 patients received revascularization and 10 received medication. In the group with infarcted myocardium, 26 patients received medication and 35 received revascularization. Patients underwent revascularization and with medication had no significant difference in improvement of LVEF between both groups (viable myocardium group: χ2 = 0.509, P > 0. 05; infarcted myocardium group: χ2 =0.035, P > 0.05). In viable myocardium group, cardiac event rate was significantly higher in patients with medication than in those who had undergone revascularization (50.0% vs 14.8%, χ2 =4.91, P<0.05).In the infarcted myocardium group, cardiac event rate was also significantly higher in patients with medication (30.7% vs5.7% ,χ2 =6.83, P<0.05). Conclusions 18F-FDG/ -MIBI SPECT myocardial imaging may well be of value but limited for the detection of myocardial viability and prediction of improvement in cardiac function as well as prognosis. However, more prospective data are needed for final evaluation.

The 18F-FDG myocardial metabolic imaging in twenty seven pilots with regular aerobic training / 中华心血管病杂志

<p><b>OBJECTIVE</b>To evaluate the characteristics of myocardial (18)F-FDG imaging in pilots with regular aerobic exercise training.</p><p><b>METHODS</b>Twenty seven healthy male pilots with regular aerobic exercise training were included in this study. The subjects were divided into fasting (n = 17) or non-fasting group (n = 10). Fluorine-18-labeled deoxyglucose and Tc-99m-sestamibi dual-nuclide myocardial imaging were obtained at rest and at target heart rate during bicycle ergometer test. The exercise and rest myocardial perfusion imaging were analyzed for myocardial ischemia presence. The myocardial metabolism imaging was analyzed with the visual semi-quantitative analyses model of seventeen segments.</p><p><b>RESULTS</b>The secondary-extreme heart rate (195-age) was achieved in all subjects. There was no myocardial ischemia in all perfusion imaging. In the visual qualitative analyses, four myocardial metabolism imaging failed in the fasting group while one failed in the non-fasting group (P > 0.05). In the visual semi-quantitative analyses, myocardial metabolism imaging scores at rest or exercise in all segments were similar between two groups (P > 0.05). In the fasting group, the myocardial metabolism imaging scores during exercise were significantly higher than those at rest in 6 segments (P < 0.05). In the non-fasting group, the scores of 3 exercise myocardial metabolism imaging were significantly higher than those at rest (P < 0.05).</p><p><b>CONCLUSION</b>Satisfactory high-quality myocardial metabolism imaging could be obtained at fasting and exercise situations in subjects with regular aerobic exercise.</p>

Assessment of therapeutic effects of stem cell transplantation in heart failure patients with old myocardial infarction by magnetic resonance imaging / 中华心血管病杂志

<p><b>OBJECTIVE</b>To evaluate the therapeutic effects of stem cell transplantation in heart failure patients with old myocardial infarction (OMI) by MRI.</p><p><b>METHODS</b>Heart failure patients [NYHA 2.7 +/- 0.7, male = 18, mean age (59.5 +/- 10.1) y] with OMI were randomly divided into 2 groups (group A: CABG + stem cell transplantation, group B: CABG; n = 10 each). Left ventricular (LV) function was measured by MRI, viable myocardium was detected by (18)F-FDG myocardial metabolism imaging and late contrast-enhanced at baseline and 6 months post intervention.</p><p><b>RESULTS</b>LVEF and LVEDV at baseline for group A were (20.71 +/- 6.09)% and (172.73 +/- 32.74) ml, and for group B were (27.59 +/- 2.31)% and (155.13 +/- 28.36) ml, respectively (P > 0.05). The LVEF was equally improved in group A and B (mean 8.63% vs. 10.37%, P > 0.05) while DeltaLVEDV was significant higher in group A than that in group B [(9.91 +/- 39.50) ml vs. (-22.34 +/- 31.35) ml, P < 0.05]. Ventricular wall thickening ratio at 6 months post intervention was significantly higher in group A than that in group B [(11.40 +/- 11.53)% vs. (2.27 +/- 7.20)%, P < 0.05]. Late contrast-enhanced MRI results correlated with (18)F-FDG myocardial metabolism imaging SPECT well in assessment of myocardial viability (kappa value: 0.446, P < 0.001; sensitivity: 68.3% and specificity: 92.5%).</p><p><b>CONCLUSIONS</b>Stem cell therapy on top of CABG aggravated LV remodeling in heart failure patients with old myocardial infarction. The specificity of MRI is similar to (18)F-FDG SPECT while the sensitivity is inferior to (18)F-FDG SPECT on detecting viable myocardium.</p>

Noninvasive detection of experimental vulnerable atherosclerotic plaques with 99Tc m labeled C2A domain of synaptotagmin I / 中华心血管病杂志

<p><b>OBJECTIVE</b>(99)Tc(m) labeled C2A domain of synaptotagmin I ((99)Tc(m)-Syt I-C2A) is used for noninvasive detection of vulnerable atherosclerotic plaque.</p><p><b>METHODS</b>Recombinant C2A domain of synaptotagmin I, overexpressed in E. Coli, was thiolated with 2-iminothiolane (2-IT) and labeled with (99)Tc(m). Atherosclerotic plaques were produced in 5 rabbits by deendothelialization of the abdominal aorta and the rabbits were fed with cholesterol diet for 3 months. Three rabbits not manipulated served as normal controls. All animals were injected with (99)Tc(m)-Syt I-C2A and underwent in vivo imaging thereafter. Aortas were then explanted for ex vivo imaging and histological characterization.</p><p><b>RESULTS</b>In deendothelialized animals, intense radio-uptake in abdominal aorta, showed by gamma camera at 2 h after injection, was visualized and T/B was 3.25 +/- 0.51 by ROI measurement, quantitative uptake ratio of abdominal aortas with atherosclerotic lesions to thoracic aortas was 8.39 +/- 1.74 in ex vivo imaging. The mean uptake in specimens of abdominal aortas with lesions was 12.6-fold higher than in control abdominal aortas, and 10.2-fold higher than in thoracic aortas of deendothelialized animals by gamma-counter.</p><p><b>CONCLUSION</b>(99)Tc(m)-Syt I-C2A has a high affinity for vulnerable atherosclerotic plaque and is a suitable a gent for the noninvasive detection of vulnerable atherosclerotic plaque.</p>

Identification of viable myocardium delayed enhancement magnetic resonance imaging and 99Tcm-sestamibi or 18F-fluorodeoxyglucose single photon emission computed tomography / 中华心血管病杂志

<p><b>OBJECTIVE</b>The aim of this study was to investigate the feasibility and accuracy of delayed enhancement magnetic resonance imaging (DE-MRI) for the assessment of myocardial viability in patients with myocardial infarction in comparison with (99)Tc(m)-sestamibi (MIBI) single photon emission computed tomography (SPECT) and (18)F-fluorodeoxyglucose (FDG) SPECT. Scar was defined as regionally increased MRI signal intensity 15 minutes after injection of 0.2 mmol/kg gadolinium-diethylenetriamine pentaacetic acid or reduced perfusion and glucose metabolism defined by SPECT.</p><p><b>METHODS</b>A total of 34 patients with myocardial infarction (29 males, 58.0 +/- 9.8 years) were imaged with MRI and SPECT.</p><p><b>RESULTS</b>A total of 578 segments were analyzed. DE-MRI and SPECT identified 431 and 336 viable segments respectively and SPECT also identified 30 ischemic segments. Necrotic segments identified by DE-MRI and SPECT were 147 and 212 respectively. Sensitivity and specificity of DE-MRI in identifying segments with matched flow/metabolism defects (scar tissues) was 61.3% and 95.4%, respectively. Quantitatively assessed relative MRI infarct area correlated well with SPECT infarct size. The value of Kappa was 0.51.</p><p><b>CONCLUSION</b>DE-MRI provides a good tool for differentiating viable myocardium from scar tissues and the detection accuracy is comparable between DE-MRI and SPECT.</p>

Experimental study of pharmaceutic stress myocardial perfusion imaging with higenamine / 中华心血管病杂志

<p><b>OBJECTIVE</b>To evaluate the applicative value of higenamine used as a new agent for pharmaceutical stress test in detection of coronary artery disease by radionuclide myocardial perfusion imaging.</p><p><b>METHODS</b>Thirteen pigs with chronic coronary artery stenosis by placement of the Ameroid constrictor in the middle section of left anterior descending artery were included in this study. Rest, higenamine and dobutamine stress radionuclide myocardial perfusion imaging was performed with Tc-99m-sestamibi.</p><p><b>RESULTS</b>The sensitivity for detection of coronary artery disease by radionuclide myocardial perfusion imaging was 85% in both higenamine and dobutamine stress imaging. Imaging scores (9.9 +/- 8.5 vs. 9.4 +/- 8.6, P = NS) and defect severity (68% +/- 12% vs. 68% +/- 15%, P = NS) showed no significant difference between higenamine and dobutamine stress test. Agreement of imaging scores between higenamine and dobutamine stress imaging was good (kappa = 0.849, P < 0.0001).</p><p><b>CONCLUSION</b>Our study demonstrates that detection of coronary artery stenosis and ischemic myocardium by myocardial perfusion imaging with higenamine is highly sensitive.</p>

Comparison of pulmonary perfusion imaging with pulmonary angiography in diagnosis of pulmonary involvement in Takayasu's arteritis / 中华心血管病杂志

<p><b>OBJECTIVE</b>To compare the diagnostic value of (99)Tc(m)-MAA pulmonary perfusion imaging with that of pulmonary angiography for pulmonary involvement in Takayasu's arteritis.</p><p><b>METHODS</b>Twenty-one patients (19 women, 2 men), with diagnosed Takayasu's arteritis and underwent both (99)Tc(m)-MAA pulmonary perfusion imaging and pulmonary angiography, were retrospectively analyzed.</p><p><b>RESULTS</b>Out of the 21 patients, pulmonary angiography detected 11 patients with pulmonary artery involvement whereas 13 patients were revealed perfusion defects by pulmonary perfusion imaging. The agreement of diagnosis by pulmonary perfusion imaging with that by pulmonary angiography existed in 19 patients (90.5%, Kappa = 0.81, P < 0.0001). There were 331 pulmonary segments consistently diagnosed by pulmonary perfusion imaging and pulmonary angiography in 378 pulmonary segments (87.5%, Kappa = 0.74, P < 0.0001).</p><p><b>CONCLUSIONS</b>Pulmonary perfusion imaging is highly accorded with pulmonary angiography in detecting the pulmonary involvement in Takayasu's arteritis. Therefore, pulmonary perfusion imaging could be used as a non-invasive screening test for the pulmonary artery involvement in Takayasu's arteritis.</p>