A Clinical Study Of High-resolution Ultrasound Techniques For Detection Of Coronary Artery Disease And Ischemic Stroke

BackgroundCardiovascular disease has become the leading cause of death worldwide,especially in developing countries including China.According to a recent national report,the number of Chinese patients with cardiovascular disease has reached 0.29 billion and about 3.5 million people die of cardiovascular disease each year,equivalent to 1 cardiovascular death every 10 sec.Atherosclerosis is the fundamental pathology for most of these diseases.Atherosclerosis is commonly deemed as systematic vascular lesions involving mainly the intima of the aortic,coronary,carotid and peripheral aretries.It is also considered as the major underlying pathology for cardiovascular or cerebrovascular disease such as coronary artery disease or ischemic stroke.Until now,there are several methods we clinically adopted to image atherosclerotic plaque,including OCT(optical coherence tomography),intravenous ultrasound(IVUS),carotid or peripheral arterial ultrasonography,and photoacoustic imaging(PAI).However,their inherent defect as well as the invasive nature of these three techniques and their costy expense prevented them from routine clinical applications.Ultrasonography has been widely used due to its convenient,non-invasive and relatively cheap nature.As a paradigm,carotid intima-media thickness(CIMT),a marker of subclinical atherosclerosis derived by carotid ultrasound,has been widely used for risk stratification and prediction of adverse cardiac events in CAD patients.However,CIMT includes thickness of both intima and media and because carotid media is less affected by atherosclerosis than intima,an increased CIMT may be caused by medial hypertrophy in response to hypertension rather than intima hyperplasia induced by atherosclerosis.In recent years,a high frequency(55-MHz)ultrasound biomicroscopy system with a high spatial resolution(～30pm)has been used successfully in our laboratory to assess the carotid atherosclerotic lesions and hemodynamic changes in ApoE-/-mice.It is also reported that there is a positive correlation between pathological changes happened in radial artery and coronary artery,as well as carotid artery.In addition,in patients undergoing coronary artery bypass grafting,it has been reported that the radial artery was more likely to have atherosclerosis,intimal hyperplasia and medial calcification than the internal thoracic artery.So it is rational and feasible to examine radial artery for detection of coronarty atherosclerotic disease(CAD).Thus,it is conceivable that ultrasound biomicroscopy might provide a high resolution technique to measure the intima thickness of the radial arteries in patients with coronary artery disease so as to detect the existance of atherosclerosis and aid in diagnosis of CAD.Aims(1)To test whether RIT has an independent and additive value for diagnosis of CAD.(2)To test whether radial artery intima thickness(RIT)is closely associated with atherosclerotic risk factors.(3)To explore the best diagnostic model for CAD based on radial artery measurement.MethodsStudy populationOne hundred and sixteen consecutive patients with CAD(68 men,mean age 60.79± 9.03 years)and 79 age and gender matched healthy controls(55 men,mean age 58.75± 9.12 years)were enrolled from Shandong University Qilu Hospital with informed consent signed.Patients with CAD were recruited from the Department of Emergency,Department of Cardiology and Department of Cardiothoracic Surgery,all of whom had a history of percutaneous coronary intervention(PCI)or coronary artery bypass grafting(CABG).Demographic feature and biochemical informationData on gender,age,hypertension,body mass index(BMI),smoking status and DM were collected.Levels of blood pressure,fasting blood glucose(GLU),total cholesterol(TC),triglycerides(TG),high-density lipoprotein cholesterol(HDL-C),low-density lipoprotein cholesterol(LDL-C),homocysteine(HCY)were obtained at baseline.Administration of aspirin,statins and anti-hypertensive drugs at baseline was also recorded.Carotid artery ultrasonography and image analysisWith patients lying supinely,5 consecutive cine loops and frozen images were obtained from both sides of carotid arteries by use of a linear transducer(L11-3,iE33 xMatrix,Philips Ultrasound,Bothell WA,USA)with electrocardiogram simultaneously monitored.CIMT was measured with frame at the peak of R wave.Radial artery ultrasonography and image analysisA high resolution 55-MHz linear mechanical transducer(RMV708,Vevo 770,Visualsonics,Toronto,Canada)was used for radial ultrasound examination on both sides of radial arteries.Radial intima thickness(RIT),radial media thickness(RMT)and radial intima-media thickness(RIMT)were measured in the far wall with the largest luminar diameter.Ratios for RIT/RMT,RIT/RIMT were also calculated.Statistical analysisStatistical analysis involved use of SPSS 16.0(SPSS Inc.,Chicago,IL,USA)and Online MedCalc software.Continuous data were presented as mean± SD.Categorical data were presented as number(percentage).Kolmogorov-Smirnov was performed to test for normality.Independent t test was used to compare groups of normally distributed continuous data.Mann-Whitney U test was used to compare non-normally distributed continuous data.Chi-square test was used to compare categorical data.Both univariate linear regression analysis and multivariate linear regression analysis were performed in both groups of subjects to analyze the relation between ultrasonic parameters and traditional risk factors or biochemical biomarkers.Binary logistic regression was performed to assess the differential value of ultrasonic parameters,biochemical biomarkers or traditional risk factors on CAD.MedCalc software was used to perform Bland-Altman plots.A two-tailed p<0.05 was considered statistically significant.ResultsDemographic dataThe two groups did not differ significantly with respect to age,gender and serum levels of TC and LDL-C.In CAD group,patients suffered more risk factors such as smoking and overweight as well as histories of hypertension and DM.The majority of CAD patients received statins and aspirin treatment and most patients with hypertension received anti-hypertensives treatment.It is noticeable that serum level of HCY,TG and fasting glucose were higher,while that of HDL-C lower in the CAD group than the control group.Ultrasonic measurements of radial and carotid arteriesPatients with CAD exhibited significantly thicker RIT,RMT,RIMT,and CIMT compared with healthy controls.The relative difference in RIT between the twogroups was the largest(25.7%).Univariate linear regression analysis showed that RIT was associated with age,smoking,DM,hypertension,SBP,DBP and statin administration,RMT was associated with age,BMI,smoking,DM,hypertension and serum glucose level,and CIMT was associated with age,gender,smoking,DM,hypertension,SBP,statin administration and serum homocysteine level.By comparison,multivariate linear regression analysis revealed that RIT was associated with age,SBP,statin administration and hypertension,while RMT was associated with fasting serum glucose level only.CIMT was associated with age,SBP,statin administration and serum homocysteine and glucose levels.In addition,there was a significant correlation between CIMT and RIT,RMT or RIMT.Diagnostic value of radial and carotid ultrasonic parameters for CADAfter validation with binary logistic regression,smoking,TG,RIT and CIMT were found predictive for the existence of CAD.Compared to traditional risk factor(including smoking and TG),the add-on value of CIMT to traditional risk factors was not significant,whereas the add-on value of RIT to traditional risk factors was highly significant.In addition,the addition of RIT to CIMT and traditional risk factors increased significantly AUC from 0.724 to 0.867(p = 0.003).Reproducibility of RIT and CIMT measurementsThe intra-class and inter-class correlation coefficients of RIT were 0.899(p<0.001)and 0.858(p<0.001)respectively,with coefficient of variation(COV)for intra-observer and inter-observer measurements being 5.3%and 7.4%,respectively.The intra-class and inter-class correlation coefficients of CIMT were 0.987 and 0.980(p<0.001 for both)respectively,with COV for intra-observer and inter-observer measurements being 3.8%and 6.6%,respectively.Conculsions1.RIT measured by ultrasound biomicroscopy was associated with age,SBP,statinadministration and hypertension,and could detect CAD independently similarly to that of CIMT.2.The add-on value of RIT to traditional risk factors for detecting CAD was superior to that of CIMT.3.The addition of RIT and CIMT to traditional risk factors markedly increased the power to diagnose CAD.4.RIT measured by ultrasound biomicroscopy provided a novel approach to the non-invasive diagnosis of CAD.BackgroundCerebrovascular disease is known to be responsible for significant morbidity and mortality worldwide.The etiologies of ischemic stroke are heterogeneous,with atherosclerosis as the main underlying cause.Until now,carotid artery is currently most favorable artery to monitor atherosclerosis by measurement of carotid intima-media thickness(CIMT)which is a simple,inexpensive and noninvasive method to assess atherosclerosis and a predictor for ischemic cardio/cerebrovascular disease.However.CIMT includes thickness of both intima and media,intima hyperplasia induced by atherosclerosis might not be accurately and sensitively revealed by CIMT,ensuing reduced efficacy of this technique for accurate detection and intervention of atherosclerosis at an early stage.Besides,carotid artery is an elastic artery,which differs from the medium-sized cerebral arteries(such as middle cerebral artery and vertebral artery)as the latter are unique muscular arteries with the medium consisting primarily of smooth muscle.Therefore,an ideal imaging technique capable of differentiating vascular intima from media of muscular artery may be more helpful to predict cerebrovascular events in clinical practice.Recently,there are several studies involving measurement of peripheral artery intima thickness(including radial artery,tibial anterior artery and dorsalis pedis artery)by using a 55-MHz high frequency biomicroscopy(resolution-30pm),which was initially developed for investigation of small animals with respect to embryonic development of cardiac structure and function,and later spanned to cardiovascular disease in humans.The accuracy of measurments of intima thickness by this 55-MHz high frequency biomicroscopy has been validated against histology in rat and human subjects.However,measurements of carotid artery with this 55-MHz high frequency biomicroscopy is not applicable due to its insufficient penetration capability.However,due to its unique anatomical structure and hemodynamic character,carotid artery is one of the most susceptible vessels to atherosclerosis.And because of the pivotal role for cerebral blood supply which carotid artery played,it is of vital importance for ischemic heart disease or ischemic stroke patient to detect atherosclerosis in carotid artery.Recently,there is a report about measuring carotid intima thickness(CIT)in premenopausal women with systemic lupus erythematosus(SLE)with a 22-MHz frequency transducer.However,its accuracy has not been validated,nor its consistency and variability being compared to 55-MHz transducer for measurement of intima thickness with the same artery.Aims(1)To test the consistency and variability of measurements of radial intima thickness(RIT)with 55-MHz frequency transducer and 24-MHz frequency transducer.(2)To test whether CIT,RIT and PIT has an independent and additive value for predicting ischemic stroke.(3)To test whether CIT,RIT and PIT aided in differentiating ischemic stroke subtypes.MethodsParticipantsOne hundred and twenty-nine consecutive patients admitted to neurology department of Shandong university Qilu Hospital and the Second hospital of Shandong University with a diagnosis of acute ischemic stroke were enrolled in this study.Stroke subtypes in this study was categorized according to an etiology-based classification criteria(TOAST)into large-artery atherosclerosis(LAA)and small-vessel occlusion(SVO).Patients with other subtypes of ischemic stroke were excluded.One hundred and four age and gender matched controls hospitalized in these two hospitals for other reasons other than cardiovascular and cerebrovascular diseases in the same period were enrolled.Demographic featuresData on age,gender,body mass index(BMI),histories of smoking,hypertension,diabetes mellitus(DM),administrations of antihypertensives and statins were collected on enrollment.Levels of fasting blood glucose,serum triglycerides(TG),total cholesterol(TC),low-density lipoprotein cholesterol(LDL-C),high density lipoprotein cholesterol(HDL-C)and homocysteine(HCY)were also collected.Carotid artery ultrasonography and image analysisAll participants were lying a supine position for carotid examination.A 24-MHz frequency linear transducer(i24Lx8)connected to an ultrasound system(Aplio i900,Canon-Toshiba Ultrasound,Tochigi-ken,Japan)was used to scan bilateral carotid arteries with an electrocardiogram simultaneously recorded.Measurements of carotid intima thickness(CIT),carotid media thickness(CMT)and carotid intima-media thickness(CIMT)were performed at far walls of the bilateral carotid arteries 1 and 2 cm proximal to the carotid bifurcation in the common carotid artery(CCA)in the long-axis view at the peak of R wave.Each site was measured three times and averaged.The mean value of the 4 measurement sites in each participant was calculated.Radial artery ultrasonography and image analysisBilateral radial artery ultrasonography was performed immediately after carotid ultrasound scanning with the same ultrasound equipment.With the participants in a lying position,longitudinal view of both sides of radial arteries 1 to 2 cm proximal to the radial styloid process was obtained.Cine loops of at least 3 consecutive cardiac beats and frozen images were obtained analyzed afterward for radial intima thickness(RIT),radial media thickness(RMT)and radial intima-media thickness(RIMT)in the far wall of the radial artery at the peak of R wave.Dorsalis pedis artery ultrasonography and image analysisBilateral dorsalis pedis artery ultrasonography was performed immediately after radial artery examination with the same ultrasound equipment.Similarly,longitudinal view of both sides of dorsalis pedis arteries at the level of the neck of the ankle bone were scanned.Cine loops of at least 3 consecutive beats and frozen images were obtained and digitally stored.Podalic intima thickness(PIT),podiali media thickness(PMT)and podalic intima-media thickness(PIMT)were measured in the far wall of the dorsalis pedis artery at the peak of R wave.Each measurement was repeated three times and the mean values from bilateral dorsalis pedis arteries were averaged.ReproducibilityThirty random subjects were selected to evaluate the consistency of CIT and RIT measurements.These data were analyzed by two experienced sonographers in a blinded way,and analyzed again by one of the sonographer 1 week later.Intra-and inter-observer variability were calculated using coefficient of variation(COV)according to the formula:COV = SD(x-y)/mean(x,y)×100%.A Bland-Altman analysis was also performed to evaluate intra-observer and inter-observer variability for CIT and RIT.Correlation and variability of RIT measured by 24-MHz and 55-MHz transducerSixty-seven random subjects who accomplished radial artery examination with 24-MHz transducer were selected to perform radial artery examination of both sides with 55-MHz linear mechanical transducer(RMV708,Vevo 770,Visualsonics,Toronto,Canada)again.Longitudinal images of the radial artery 1 to 2 cm proximal to the radial styloid process were obtained.Radial intima thickness(RIT)was measured three times and averaged in the far wall with the largest luminar diameter and was defined as the distance from the leading edge of the intimal-luminal interface to intimal-medial interface.Paired t test was used to compare groups of normally distributed continuous data.Pearson correlation analysis was performed to explore the association between RIT values derived by different transducers.Coefficient of variant was calculated as follows:COV = SD(x-y)/mean(x,y)× 100%.Statistical analysisStatistical analysis involved the use of SPSS 23.0(SPSS Inc.,Chicago,IL,USA)and Online MedCalc software(https://www.medcalc.org/).Kolmogorov-Smirnov was performed to test for normality.Independent t test was used to compare groups of normally distributed continuous data.Chi-square test was used to compare categorical data.Multivariate linear regression analysis was performed to analyze the relationship between ultrasonic parameters and biochemical biomarkers or traditional risk factors.Binary logistic regression analysis was performed to select and validate potential diagnostic parameters for IS and differential parameters for LAA and SVO subtypes.Receiver operating characteristic(ROC)curve analysis was performed to explore diagnostic value of ultrasonic,biochemical and demographic parameters as well as different combination models.MedCalc software was used to plot Bland-Altman analysis.A two-tailed p<0.05 was considered statistically significant.ResultsCorrelation and variability of RIT measured by 24-MHz and 55-MHz transducerThere was no difference of RIT value derived by the 24-MHz and 55-MHz transducer(10.62 ± 2.79×10-2mm vs.10.52± 2.84×10-2mm,p=0.082).There was a good correlation between RIT measurements with 24-MHz transducer and 55-MHz transducer.The correlation coefficient was 0.986(p<0.001,95%CI:0.978-0.992).The coefficient of variation for the two measurement was only 3.69%.Demographic characteristicsThere was no difference between ischemic stroke group and control group regarding age,gender,smoking,as well as levels of TG,TC and LDL-C.In patients with ischemic stroke,prevalence of DM,hypertension,administrations of antihypertensives and statins were higher than in the control group.There was no difference for all demographic features among different subtypes of ischemic stroke.Ultrasonic measurementsComparing with controls,the measurements CIT,CMT,CIMT,RIT,RMT,RIMT,PIT,PMT and PIMT were higher in ischemic stroke patient,especially in LAA subtype.RIT,PIT,PMT and PIMT were significantly higher in LAA subtype than SVO subtype.Risk factors of intima thickness and media thicknessMultivariate linear regression analysis revealed that both age and hypertension were positively correlated with intima thickness and media thickness in all three arteries.DM was positively associated with CIT and CIMT,while smoking and HCY influence RMT and RIMT.GLU contributed to the thickness of PIT and PIMT.In addition,there was a significant correlation between CIT and RIT or PIT(Pearson correlation coefficient 0.458 and 0.400,respectively,p<0.01 for both).Likewise,there was a significant correlation between CMT and RMT or PMT(Pearson correlation coefficient 0.280 and 0.]70,respectively,p<0.01 for both).Diagnostic value of radial and carotid ultrasonic parameters for ischemic strokeAfter validation with both forward:LR selection and backward:LR selection.history of hypertension,level of GLU,HDL-C and HCY,CIT and RIT remained in the regression model for diagnosis of ischemic stroke.Compared with traditional risk factor,the add-on value of CIT or RIT to traditional risk factorswas not significant.However,the addition of both RIT and CIT to traditional risk factors significantly increased AUC significantly from 0.911 to 0.967(p = 0.014).Moreover,CIT had added-on value to traditional risk factors for diagnosis of LAA subtype.RIT also tend to have add-on value for traditional risk factors(p=0.075).The addition of both RIT and CIT to traditional risk factors yielded the largest AUC of 0.969(p=0.009).Differentiatial value of radial and carotid ultrasonic parameters for IS subtypesAfter validation with forward:LR selection,only RIT was left in the regression model.The AUC for RIT was 0.680 with differentiating power of RIT for LAA was 79.8%(x=9.82,p =0.002).Reproducibility of CIT and RIT measurementsThe intra-class and inter-class correlation coefficients of CIT were 0.979(p<0.001)and 0.975(p<0.001)respectively,with coefficient of variation(COV)for intra-observer and inter-observer measurements being 6.3%and 10%,respectively.The intra-class and inter-class correlation coefficients of RIT were 0.931(p<0.001)and 0.907(p<0.001)respectively,with coefficient of variation(COV)for intra-observer and inter-observer measurements being 8.6%and 14%,respectively.Conculsions1.There was a good correlation and consistency of RIT derived by 24-MHz transducer and 55-MHz transducer.24-MHz transducer could be used to measure arterial intima thickness with satisfactory accuracy.2.CIT and RIT were associated with traditional risk factors for atherosclerosis.3.CIT and RIT could help in predicting the existence of ischemic stroke.4.CIT and RIT could help in subtype classification for ischemic stroke.5.Combination of CIT,RIT and traditional risk factors may pave an avenue to accurate diagnosis of ischemic stroke.BackgroundPathological studies in the last decade have revealed that in the majority of patients with acute coronary syndrome(ACS),the catastrophic event is caused by atherosclerotic plaque rupture or erosion with subsequent intraluminal thrombosis,and vascular lesions prone to these events were deemed vulnerable plaque.Local inflammation plays a pivotal role in the pathogenesis of plaque formation,progression and disruption,and a number of plasma biomarkers detected in the peripheral or coronary circulation have been reported to be associated with plaque inflammation and might be useful in predicating cardiovascular events(CVE)in patients with ACS.However,as most of these studies on circulatory biomarkers were retrospective or cross sectional in nature without longitudinal follow-up results,the predictive values of these biomarkers remain inconclusive.Thus,exploration of novel circulatory biomarkers capable of predicting cardiovascular events induced by vulnerable plaque is still a major challenge.In addition,most vulnerable plaques are non-obstructive lesions evading detection by conventional coronary angiography.Although intravascular ultrasonography(IVUS)has been used to evaluate the geometrical properties of plaque,lumen and vascular wall,assessment of plaque inflammation and deformation are beyond its capability.Recently,we have developed a novel algorithm for calculating two-dimensional strain of atherosclerotic plaques from IVUS images,constructed a new IVUS elastography(IVUSE)system and validated this technique with quantitative histological analysis in a rabbit model.However,it remains elusive whether this new technique can predict future cardiovascular events in patients with ACS.Previous studies suggested that inflammation plays a key role in the initiation and development of atherosclerosis,and a number of inflammatory biomarkers such as secretory phospholipase A2(sPLA2),phosphatidylcholine-specific phospholipase C(PC-PLC),soluble CD40 ligand(sCD40L),interleukin-6(IL-6),myeloperoxidase(MPO),monocyte chemoattractant protein-1(MCP-1),tissue factor(TF),and YKL40 were investigated to reveal this procedure.Yet there is still room to further elucidate their additional value for different score system or models we adopted in daily clinical practice.The GRACE risk score was initially developed to predict in-hospital mortality of patients with ACS in 2003,and subsequently applied to predict 6-month mortality or myocardial infarction in a registry cohort of ACS in 2004.However,most patients with ACS today may receive percutaneous coronary intervention(PCI)and optimal medical therapy,with a consequence of very low in-hospital mortality,although a substantial proportion of patients continue to develop recurrent cardiovascular events on long-term follow-up.It is still an open question whether GRACE risk score can predict long-term multiple cardiovascular events such as cardiovascular death,non-fatal myocardial infarction,unstable angina pectoris and revascularization after ACS,and whether measurement of plasma inflammatory biomarkers and plaque morphology and strain may add incremental predictive value to GRACE score in patients with ACS.Aims(1)To evaluate the capability of morphological and strain parameters oderived by IVUS and IVUSE technique for prediction of long-term cardiovascular events in ACS patients.(2)To compare the predictive value of GRACE score with plasma biomarkers in peripheral and coronary circulation,coronary plaque morphology,and coronary plaque strain.(3)To examine the additive predictive value of plaque measurements and plasma biomarkers to GRACE score in ACS patients.MethodsStudy design and paritcipantsThis study was a multicenter,prospective and long-term follow-up trial.100 ACS patients with complete clinical,biochemical and IVUS information were enrolled from Shandong University Qilu Hospital and Shenyang Military Region General Hospital.The study protocol was approved by the Ethics Committee of Shandong University Qilu Hospital and informed consent was obtained from all patients.Biochemical AssayThe plasma levels of lipids,blood glucose,creatinine,uric acid,cardiac troponin,and circulating and intracoronary levels of biomarkers including sPLA2,PC-PLC,sCD40L,IL-6,MPO,MCP-1,TF and YKL40 were measured.Intracoronary samples were collected immediately after IVUS imaging,and were taken proximal and distal to plaque we intended to investigated.Biomarker gradient was calculated as distal level of biomarker minus proximal level of the same biomarker.IVUS imaging and analysisDuring the process of selective coronary angiography,all patients underwent IVUS imaging of coronary arteries and IVUS image analysis was conducted to derive a number of plaque parameters.IVUSE construction and analysisA validated two-dimensional IVUSE technique was used to measure maximal shear strain(SSmax)and maximal area strain(ASmax)of a plaque from IVUS images.GRACE score calculationThe GRACE score included eight clinical variables during hospitalization(age,heart rate,systolic blood pressure,plasma creatinine concentration,Killip class at presentation,cardiac arrest at admission,ST-segment deviation,and elevated cardiac enzymes/markers).Values for these variables at admission were entered into the GRACE risk calculator(available at http://www.outcomes-umassmed.org/grace)to obtain estimates of the cumulative risk of all-cause mortality and pre-specified composite endpoint of cardiovascular events.Follow-up and endpointAll patients were followed up clinically or by telephone every 6 months after enrollment for 3 years.Routine physical checkup,bioassay of blood glucose and plasma lipid profile and electrocardiography(ECG)were performed at each follow-up and cardiovascular event and event times were recorded.The clinical endpoints included sudden cardiac death,non-fatal myocardial infarction,unstable angina pectoris requiring hospitalization,and coronary revascularization,which were verified by telephone call and in-hospital case records.Statistical analysisStatistical analysis involved use of SPSS 16.0(SPSS Inc.,Chicago,IL,USA).Kolmogorov-Smirnov was performed to test for normality.Continuous data were expressed as mean± SD and were analyzed by independent t test,and categorical data were expressed as percentages and were analyzed by Chi-square test.For numerical variables of skew distribution,Mann-Whitney U test was applied to compare the between-group difference.The Cox proportion risk model,which takes into account the time to first cardiovascular event,was adopted to assess the predictive power of variables for cardiovascular events,and the hazard ratio(HR)and 95%confidence interval(95%CI)were calculated.Parameters with significant predictive power were then introduced into combination models to assess the additive predictive values of plasma biomarkers and plaque strain to GRACE score by applying C-statistics.The Kaplan-Meier survival plot was used to visualize cumulative cardiovascular events during follow-up categorized according to different parameter values divided into bottom/middle/top tertiles.The log-rank test for trends was used for tertiles.A p<0.05 was considered statistically significant.ResultsBaseline characteristics of study populationAmong 100 ACS patients,2 patients had ST segment elevation myocardial infarction(STEMI),8 non-ST segment elevation myocardial infarction(NSTEMI)and 90 unstable angina pectoris(UAP)who aged 40 to 75 years(mean age 57.24 ± 1.06 years)with the majority(64%)being males.Most of these patients had multiple risk factors for atherosclerosis such as hypertension,current smoking,diabetes and obesity.Cardiovascular eventsThe median follow-up time was 28 months(range 6-51 months).We identified 32 cardiovascular events comprising 1 cardiac death,29 recurrent unstable angina pectoris requiring hospitalization and 2 coronary revascularizations.The median duration of time from enrollment to cardiovascular events was 26 months(range 6-51 months).There was no significant difference between patients with and without cardiac events in terms of age,gender,body mass index,as well as most of the circulating and intracoronary biomarkers and plaque prameters.However,the plasma levels of hsCRP,SP-D and DKK1,intracoronary gradients of sPLA,and SP-D as well as plaque parameters of SSmax and ASmax were significantly higher in patients with than without cardiovascular events.Predictive values of GRACE score,plasma biomarkers and plaque strainGRACE score,plasma levels of SP-D and DKK1,intracoronary gradients of sPLA-2 and SP-D,as well as plaque parameters of SSmax and ASmax,may independently predict cardiovascular events(p<0.001～0.05).In addition,higher tertiles of SSmax and ASmax,as well as intracoronary gradient of SP-D and plasma levels of SP-D and TC were significantly associated with increased cumulative cardiovascular events.The GRACE score yielded a C-statistic of 0.457,and the C-statistic for plasma levels of SP-D and DKK1,intracoronary gradients of sPLA2 and SP-D,and plaque parameters of SSmax and ASmax were 0.492,0.505,0.500,0.504,0.528 and 0.531,respectively,none of which was significantly higher than the C-statistic of GRACE score alone。Additive predictive values of plasma biomarkers and plaque strain to GRACE scoreTo explore the incremental value of parameters with significant predictive power for combination,C-statistics were calculated for different combination models.Based on GRACE score,addition of plasma levels of SP-D and DKK1,intracoronary gradients of sPLA2 and SP-D,and plaque parameters of SSmax and ASmax yielded the largest C-statistics(0.457 vs.0.667,p=0.014),with sensitivity and specificity being 68%and 64%,respectively,as revealed by ROC analysis.Conculsions1.Plaque strain parameters SSmax and ASmax,as well as plasma levels of SP-D and DKK1,intracoronary gradients of sPLA2 and SP-D,and can independently predict long-term cardiovascular events in patients with ACS similar to GRACE score.2.The combination of plaque strain,circulating and intracoronary biomarkers and GRACE score provides a better prediction than GRACE score alone.3.This combinational approach may pave a new avenue to precise risk stratification in ACS patients.