Radiomics-based analysis by machine learning techniques improves characterization of functionally significant coronary lesions

Background

Computed Tomography Coronary Angiography (CTCA) is an effective non-invasive imaging modality for anatomo-functional assessment of coronary artery disease (CAD). Radiomics features have been used for diagnosis or outcome prediction, however, their potential value for characterizing flow limiting coronary lesions has not been explored.

Purpose

To assess whether application of novel radiomics and machine learning (ML) techniques on CTCA derived datasets improves characterization of functionally significant coronary lesions.

Methods

Consecutive patients with stable chest pain and intermediate pre-test likelihood for CAD, who underwent CTCA and PET-or SPECT-Myocardial Perfusion Imaging (MPI) respectively, were prospectively evaluated and included in the analysis. PET-MPI was considered abnormal when >1 contiguous segments showed both stress Myocardial Blood Flow ≤2.3mL/g/min and Myocardial Flow Reserve (MFR) ≤2.5 for 15O-water or <1.79 mL/g/min and ≤2.0 for 13N-ammonia respectively. Defect reversibility (DR) was defined as a summed difference score (SDS) between stress and rest images ≥2. CTCA and functional images were fused to assign each myocardial segment to the pertinent coronary territory. Stenosis severity, plaque characteristics and radiomic plaque features were assessed in the total length of the 3 main coronary vessels. In total, 1765 features were extracted from each vessel and a feature reduction and model creation pipeline was constructed [Figure 1]. Two separate datasets: a) coronary stenosis (≥50%) + plaque characteristics and b) coronary stenosis (≥50%) + plaque characteristics + radiomics were formulated and compared in terms of AUCs accordingly.

Results

A total of 292 coronary vessels (140 with corresponding PET-MPI data and 152 with SPECT MPI data) were analysed. Plaque burden and stenosis severity were the only independent predictors of impaired myocardial perfusion on PET-MPI, with an AUC = 0.749, (95% CI: 0.658–0.826). Stenosis severity, kurtosis, contrast, interquartile range and entropy were predictors of an abnormal PET-MPI result and their combination resulted in an AUC = 0.854, (95% CI: 0.775–0.914). The difference between the 2 models was statistically significant (p-diff: 0.02, 95% CI: 0.0165–0.194). Stenosis severity was the only predictor of a DR on SPECT-MPI, AUC = 0.624 (95% CI: 0.542–0.702). Small Dependence High Gray Level Emphasis, Cluster Prominence, Region Length, wavelet Median and square Median were predictors of a positive SPECT result, with AUC = 0.816, (95% CI: 0.745–0.875). The difference between the two models was statistically significant (p-diff: 0.006, 95% CI: 0.152–0.329)

Conclusion

Radiomic futures can be combined with anatomical and morphological characteristics of coronary lesions in CTCA imaging and provide valuable complementary information for characterizing functionally significant coronary lesions.

Funding Acknowledgement

Type of funding sources: Public grant(s) – EU funding. Main funding source(s): This work was supported from European Regional Development Fund, Operational Programme “Competitiveness, Entrepreneurship and Innovation 2014-2022 (EPAnEK)”, titled: The Greek Research Infrastructure for Personalized Medicine (pMED-GR)

Arterial stiffness and valvular calcifications in aortic stenosis: caught between a rock and a hard place

Background/Introduction

Arterial stiffness and aortic hemodynamics are independent predictors of adverse cardiovascular events. Indications for transcatheter aortic valve implantation (TAVI) are expanding and aortic valve calcifications (AVC) are an important prognostic factor of the success of TAVI.

Purpose

We sought to investigate the associations between AVC and aortic vascular function/hemodynamics.

Methods

Fifty-two high-risk patients (mean age 80.4±8.5 years, 27 male) with severe symptomatic aortic stenosis undergoing TAVI were included. Arterial stiffness was estimated through carotid-femoral pulse wave velocity (cfPWV) and brachial-ankle pulse wave velocity (baPWV). Aortic hemodynamics (aortic pressures, aortic augmentation index corrected for heart rate [AIx@75]) were also measured. Measurements were conducted prior to the implantation and at discharge. In all patients, a native and contrast-enhanced multislice cardiac computed tomography were performed pre-interventionally. AVC were then graded semi-quantitatively as follows: grade 1 – no calcification; grade 2 – mildly calcified (small isolated spots); grade 3 – moderately calcified (multiple larger spots); grade 4 – severely calcified (extensive calcification of all cusps).

Results

Group 1 (subjects with none/mild AVC, n=29) did not significantly differ in age, gender and body-mass index compared to group 2 (subjects with moderate/severe AVC, n=23). As far as the traditional cardiovascular risk factors were concerned, only hypertension (p=0.008), coronary artery disease (p=0.016), atrial fibrillation (p=0.075) and insulin-dependent diabetes mellitus (p=0.068) were found to be more prevalent in group 2. Group 2 had significantly higher both cfPWV and baPWV (8.3±1.7 vs 7.2±1.2 m/s and 1750±484 cm/s vs. 2101±590 cm/s with p=0.008 and p=0.022 respectively) compared to Group 1 (Figure 1). Even after adjustment for age, gender and systolic blood pressure, aortic stiffness indices were higher in Group 2 compared to Group 1 (p=0.038 and p=0.048, respectively). There was no statistically significant difference in peripheral or aortic pressures as well as in wave reflections indices between the two groups.

Conclusion

Our study shows that in patients with aortic valve stenosis there is a correlation between increased aortic stiffness and a greater extent of damage of aortic valvular leaflets as well as calcifications.

Funding Acknowledgement

Type of funding sources: None. Figure 1. PWV and aortic valve calcifications

Novel computed-tomography derived prognostic markers in patients undergoing TAVI with a self-expanding valve

Introduction

Transcatheter aortic valve implantation is the treatment of choice in a consistently expanding group of patients with severe aortic valve stenosis. Tricuspid and mitral annular dilatation with consequent valvular regurgitation are associated with adverse outcome. Computed tomography angiography (CTA) is routinely performed for preprocedural evaluation of vascular access and prosthesis sizing.

Purpose

To evaluate the impact of mitral and tricuspid annular dimensions in preprocedural CTA on prognosis of patients undergoing TAVI with a self-expanding valve.

Methods

CTAs of consecutive patients undergoing TAVI in a single high-volume center between 2016 and 2019 were retrospectively evaluated. Maximal septolateral tricuspid annular diameters (TAD) and mitral annular diameters (MAD) were obtained and measured from properly angulated three dimensional CTA datasets. Moreover, maximal pulmonary artery diameter perpendicular to the long axis was measured in every patient. Patients were followed up by clinical visits or telephone contacts. As clinical events were defined all-cause mortality, stroke and heart failure hospitalization.

Results

In total 123 patients were included in the study. The mean follow-up duration was 875±383 days and 21 clinical events were recorded. There was a moderate but statistical significant correlation between TAD and both pulmonary artery diameter (r=0.39, p<0.001) and pulmonary artery systolic pressure by echocardiography (r=0.23, p=0.015). In univariate logistic regression analysis pulmonary artery diameter and TAD were both associated with heart failure hospitalization (p=0.03 and 0.02 respectively). In addition, MAD was associated with total events (OR: 0.43, 95% CI 0.19–0.99, p=0.048). The relationship of MAD with events remained significant after adjustment for sex, age and tricuspid annular dimensions (OR: 0.28, 95% CI 0.1–0.79, p=0.02).

Conclusions

TAD and MAD were associated with heart failure rehospitalization and clinical events respectively in patients undergoing TAVI with a self-expanding valve. Further larger prospective studies are warranted to evaluate the prognostic value of these CTA markers.

Funding Acknowledgement

Type of funding sources: None.

Percutaneous access versus surgical cutdown in TAVI: vascular and bleeding complications

Background

Access options for transcatheter aortic valve implantation (TAVI) are vital, since use of large sheaths may lead to access-related complications and bleeding.

Purpose

To determine the access-related vascular and bleeding complications of patients undergoing transfemoral TAVI.

Methods

Consecutive patients scheduled for transfemoral TAVI were retrospectively grouped according to vascular access [percutaneous access (p-TAVI) and surgical cutdown (sc-TAVI)]. Primary end points were vascular and bleeding complications, based on the VARC-II criteria.

Results

Totally, 187 patients were included in the analysis (p-TAVI: 124 patients; sc-TAVI: 63 patients). Mean procedure time was shorter in the p-TAVI group compared to the sc-TAVI group (45.65±6.17 min versus 64.05±15.73 min, p<0.001). Contrast use was lower in the p-TAVI group compared to the sc-TAVI group (81.18±15.96 ml versus 106.75±25.67 ml, p<0.001), which resulted in higher rates of acute kidney injury in the sc-TAVI group (13% versus 1%, p=0.01). Vascular access complications occurred numerically but not statistically more often in the p-TAVI group compared to the sc-TAVI group (11% versus 5% for minor complications and 6% versus 3% for major complications respectively, p=0.10). Patients in the p-TAVI group had the same minor and major bleeding complications compared to the sc-TAVI group (11% versus 8% for minor, 10% versus 6% for major bleeding complications respectively, p=0.49), but no life-threatening bleeding (0% versus 1.5%).

Conclusions

Transfemoral access options in TAVI (surgical cutdown or percutaneous) have similar efficacy and should be offered in TAVI patients if and when appropriate.

Funding Acknowledgement

Type of funding sources: None.

Impact of severe mitral annular calcification on paravalvular leak after transcatheter aortic valve implantation

Aims

Paravalvular leak (PVL) remains a frequent complication after transcatheter aortic valve implantation (TAVI) and seems to affect short- and long-term survival.

Purpose

The aim of this study was: 1) to identify anatomical predictors of PVL after TAVI and 2) assess the impact of PVL on cumulative survival.

Methods and results

Patients with severe and symptomatic aortic stenosis [effective orifice area (EOA) ≤1cm2] referred for TAVI at our institution were consecutively enrolled. Prospectively collected demographic, laboratory and echocardiographic data were retrospectively analyzed. Patients were stratified into two groups according to the presence of PVL after TAVI and were followed up postoperatively with clinical and echocardiographic assessment. Primary clinical endpoint was all-cause mortality, as defined by the criteria proposed by the Valve Academic Research Consortium 2. In total, 291 patients were included (male: 50.2%, mean age: 80±7.6 years) in our study. Of these, 165 (56,8%) presented at least mild PVL after TAVI (mild: 85,5%, moderate: 13.3% and severe: 1.2%). The median follow-up period was 27.3 [min. 0, max 113 months. Two patients with severe PVL were excluded from the analysis. In the follow up period, there was no significant difference regarding all-cause mortality between patients with and those without PVL after TAVI, independently from the degree of PVL (log rank: 0.991 - Figure 1). Severe aortic annulus calcification, the presence of a bicuspid aortic valve and aortic root angulation, as assessed by computed tomography (CT), were found to associate with PVL after TAVI in univariate analysis. In the multivariate analysis, severe aortic annulus calcification was found to be the only independent predictor of mild or moderate PVL after TAVI [Exp(B): 1.540, 95% Confidence Interval: 1.067–2.224, B=0.432, p=0.021].

Conclusion

The presence of mild or moderate PVL after TAVI was not found to affect cumulative survival in the 27 months of follow up period. Severe annulus calcification assessed by CT-scan, was found to be the only independent predictor of PVL after TAVI.

Funding Acknowledgement

Type of funding sources: None.

One-year echocardiographic outcomes of transcatheter aortic valve implantation with or without predilatation of the aortic valve: insights from the multicenter, randomized DIRECT trial

Introduction

Transcatheter aortic valve implantation (TAVI) has become the standard of care for high-risk and inoperable surgical patients and a valid alternative in intermediate-risk patients with severe aortic stenosis.The DIRECT trial (Predilatation in Transcatheter Aortic Valve Implantation Trial) was a multicenter, randomized, clinical trial designed to evaluate the safety and efficacy of TAVI with or without balloon aortic valvuloplasty (BAV) in patients with symptomatic, severe aortic valve stenosis.

Purpose

To compare the one year echocardiographic findings among patients, who underwent TAVI using a self-expanding valve with or without BAV.

Methods

A total of 171 patients with severe aortic stenosis were randomly assigned at 4 tertiary centers to undergo TAVI with the use of self-expanding prostheses with (pre-BAV) or without pre-dilatation (no-BAV). Follow up transthoracic echocardiography was performed 1 year after TAVI.

Results

Of 171 patients, 86 patients were randomized to pre-BAV group and 85 to no-BAV group. One year echocardiographic follow up was available in 146 patients. In one year follow up there was no significant difference between pre-BAV and no-BAV group in aortic valve area (1.84±0.39cm2 vs. 1.85±0.44cm2, p=0.79), peak aortic valve gradient (15.95±9.97 mmHg vs. 14.51±6.60 mmHg, p=0.35), mean aortic valve gradient (8.37±5.01 mmHg vs. 7.99±4.04 mmHg, p=0.64), aortic valve peak velocity (1.90±0.51 m/s vs. 1.80±0.42m/s, p=0.24), ejection fraction (54.19±8.36% vs. 53.19±9.58%, p=0.52) and pulmonary artery systolic pressure (41.86±14.34 mmHg vs. 40.71±12.40 mmHg, p=0.64). The incidence of moderate or severe paravalvular regurgitation (PVL) in 1 year follow up was 6.2% without significant difference between the 2 study groups (5.7% in the no-BAV group vs. 6.6% in the pre-BAV group, p=0.83).

Conclusions

Direct transcatheter aortic valve implantation has no impact on one-year prosthesis function and PVL in patients undergoing TAVI with self-expanding valve

Funding Acknowledgement

Type of funding source: Private company. Main funding source(s): Medtronic

Splenic switch off as a novel marker for adenosine response in 13N-ammonia PET myocardial perfusion imaging – a pilot study

Background

Positron emission tomography myocardial perfusion imaging (PET MPI) is a robust and excellent tool for assessing ischemia. So far, however, no methodology has been established to distinguish truly reduced MFR due to microvascular dysfunction or three-vessel coronary disease (CAD) from seemingly impaired MFR due to inadequate adenosine response. Conversely, for cardiac stress magnetic resonance (CMR) the adenosine induced splenic switch-off (SSO) sign has been proposed as a potential marker for adequate adenosine response.

Purpose

We assessed the feasibility of detecting SSO in adenosine stress 13N-ammonia PET MPI using SSO in CMR as the standard of reference.

Methods

50 patients underwent simultaneous PET MPI and CMR on a hybrid PET/MR device with co-injection of 13N-ammonia and a gadolinium-based contrast agent during rest and adenosine-induced stress. In CMR, SSO was assessed qualitatively and quantitatively by calculating the ratio of the peak signal intensity of the spleen during stress over rest (SIR). Similarly, in PET MPI the splenic signal activity ratio (SAR) was calculated as the proportion of the maximal standard uptake value of the spleen in stress and rest. Additionally, MFR was quantified in PET MPI.

Results

Visual SSO in CMR was present in 37 (74%) patients, whereas 13 patients had no SSO. The median SIR in CMR was significantly lower in patients with visual SSO compared to patients without visual SSO (0.57 [IQR 0.49–0.62] vs. 0.89 [IQR 0.76–0.98]; p<0.001). Similarly, median SAR in PET was significantly lower in patients with visual SSO in CMR compared to patients without visual SSO (0.4 [IQR 0.32–0.45] vs. 0.8 [IQR 0.47–0.98]; p<0.001). SIR correlated significantly with SAR (r=0.4, p<0.05). Mean MFR was significantly higher in patients with visual SSO compared to patients without visual SSO (3.38±0.86 vs. 2.53±0.84; p<0.05).

Conclusion

Similarly to CMR, SSO can be detected in 13N-ammonia PET MPI. This might help distinguish adenosine non-responders from patients with truly impaired MFR due to microvascular dysfunction or multivessel CAD.

Figure 1. Splenic switch off (*) illustrated on transaxial 13N-ammonia PET MPI stress (A) compared to rest perfusion images (B) and similarly in stress (C) and rest (D) short axis CMR (**) in the same patient during adenosine induced stress and co-injection of 13N-ammonia and a gadolinium based contrast agent, acquired on a hybrid PET/MR device.

Funding Acknowledgement

Type of funding source: Public grant(s) – National budget only. Main funding source(s): Swiss National Science Foundation (SNSF)

Echocardiographic assessment of functional changes of prosthetic valve after transcatheter aortic valve implantation in one year follow up: insights from the multicenter, randomized DIRECT trial

Introduction

The DIRECT trial (Predilatation in Transcatheter Aortic Valve Implantation (TAVI) Trial) was a multicenter, randomized, clinical trial designed to evaluate the safety and efficacy of TAVI with or without balloon aortic valvuloplasty (BAV) in patients with symptomatic, severe aortic valve stenosis.

Purpose

To investigate by echocardiography the functional changes of self-expanding prosthetic valves during the first year after TAVI with or without BAV.

Methods

One hundred seventy one consecutive patients with severe aortic stenosis were enrolled at 4 centers and randomized to TAVI using self-expanding prostheses with (pre-BAV) or without pre-dilatation (no-BAV). Transthoracic echocardiography was obtained at baseline, 30 days and 1 year after TAVI.

Results

Of 171 patients, 86 patients were randomized to pre-BAV group and 85 to no-BAV group. Over the one year, 7 (4%) patients died and in 18 (10%) there was no available paired 30 day/1 year echo. At baseline echocardiography the peak and mean aortic valve gradient and the aortic valve area (AVA) in no-BAV group were 77.31±22.56 mmHg, 47.23±14.98 mmHg and 0.69±0.16cm2 and in pre-BAV group 81.97±23.17 mmHg, 49.39±14.78 mmHg and 0.65±0.15cm2 respectively. One year after TAVI, patients in no-BAV and pre-BAV group showed stable peak and mean aortic valve gradients similar to those at 30 days (from 16.36±7.88 to 14.51±6.6 mmHg vs. 17.17±8.88 to 15.95±9.97 mmHg and from 8.87±4.23 to 7.99±4.04 mmHg vs. 9.39±4.79 to 8.38±5.02 mmHg respectively, P<0.001 vs. baseline). The AVA was similarly stable in one year follow up in no-BAV group (from 1.85±0.43cm2 to 1.85±0.44cm2, P<0.001 vs. baseline) and in pre-BAV group (from 1.86±0.49cm2 to 1.84±0.39cm2, P<0.001 vs. baseline). The incidence of moderate or severe paravalvular regurgitation remained unchanged in both groups (from 4.7% to 5.7% in no-BAV group and from 5.8% to 6.6% in pre-BAV group).

Conclusions

In both pre-BAV and no-BAV groups the improvement in hemodynamics of self-expanding prosthetic valves remained durable during the one year echocardiographic follow up assessment.

Funding Acknowledgement

Type of funding source: Private company. Main funding source(s): MEDTRONIC

No impact of direct implantation of a self-expanding valve on one-year clinical outcomes. Insights from the multicenter, randomized DIRECT trial

Introduction

The DIRECT trial (Predilatation in Transcatheter Aortic Valve Implantation Trial) evaluated in a randomized fashion the safety and efficacy of direct (without balloon pre-dilatation) implantation of a self-expanding valve in all comers undergoing TAVI.

Purpose

To investigate the impact of direct implantation of a self-expanding valve on one-year clinical outcomes.

Methods

DIRECT trial randomized consecutive patients with severe aortic stenosis at 4 tertiary centers to undergo TAVI with the use of self-expanding prostheses with (pre-BAV) or without pre-dilatation (no-BAV). The primary endpoint was device success according to the VARC-2 criteria. Secondary endpoints included periprocedural mortality and stroke, new permanent pacemaker implantation and vascular complications.

All cause death, cardiac death, stroke and heart failure hospitalizations were recorded at one year and compared between the two groups using Kaplan-Meier plots.

Results

In total 171 patients were randomized in 4 centers. In the intention to treat analysis 86 patients were randomized to the pre-BAV group and 85 patients to the no-BAV TAVI group.

The device success according to the VARC-2 criteria was non-inferior in the no-BAV group compared to the pre-BAV group (65/85 - 76.5% for no-BAV versus 64/86 – 74.4% for pre-BAV, mean difference = 2.1%, 90% CI: −8.9 to 13). In the no-BAV group 25 (29.4%) patients underwent post balloon dilatation and in the pre-BAV group 13 patients (15.1%) (p=0.03).

At one year 4 deaths were recorded in pre-BAV group (4.7%) and 3 deaths in no-BAV group (3.5%). There was no difference in Kaplan-Meier plots between the two groups in all-cause mortality (log-rank p=0.72, figure). Similarly, there was no difference in one-year incidence of stroke (1 in pre-BAV and 2 in no-BAV group, log-rank p=0.55), cardiac death (log-rank p=0.66), non-cardiac death (log-rank p=0.98) and heart failure hospitalizations (1 in pre-BAV versus 3 in no-BAV group, log-rank p=0.31). Lastly, there was no difference in the incidence of permanent pacemaker implantation between the two groups at one year (27/67 in no-BAV group versus 20/69 in pre-BAV group, log-rank p=0.24)

Conclusions

Direct transcatheter aortic valve implantation is non-inferior to the procedure with pre-dilatation in self-expanding valve. Despite the overall low rate of events, direct procedure has no impact on clinical outcomes at one year.

Funding Acknowledgement

Type of funding source: Private company. Main funding source(s): Medtronic

Carotid artery temperature reduction with statin therapy in patients with familial hyperlipidemia syndromes

Background

Familial hypercholesterolemia (FH) syndromes constitute an important risk factor for premature atherosclerosis. Microwave radiometry (MWR) assess non-invasively carotid artery temperatures reflecting inflammation. Recent data support that statin therapy, that constitutes the cornerstone for the treatment of FH, reduces systemic inflammation.

Purpose

To investigate the impact of statin therapy either with simvastatin or with combination simvastatin plus ezetimibe on carotid artery temperatures.

Methods

Consecutive patients with diagnosis of either heterozygous hypercholesterolemia (hFH) or combined hyperlipidemia (FCH) not under statin therapy for at least 6 months were included in the study. FH pts were assigned to either simvastatin 40 mg or simvastatin 40 mg plus ezetimibe 10mg according to the discretion of the physician. FH patients who refused statin therapy were used as control group for the assessment of statins effect. In all subjects, common carotid intima-media thickness (ccIMT) was measured in the last 2 cm of the far wall of both common carotids close to the carotid bifurcation and ΔT (maximum-minimum) temperature measurements were performed across each carotid during MWR evaluation. Examinations were performed at baseline and after 6 months. Blood's lipid profile was also obtained in all patients.

Results

In total 115 patients were included in the study. Of them 40 patients received simvastatin (19 hFH and 11 FCH), 41 simvastatin + ezetimibe (31hFH and 10 FCH) and 34 (21 hFH and 13 FCH) no statin. There was no difference at baseline in ccIMT and ΔT measurements between hFH and FCH patients (0.10±0.03 vs 0.10±0.02, p=0.74 and 0.88±0.38 vs 0.84±0.32, p=0.52, respectively). Patients who refused statin therapy did not show any reduction in ccIMT and ΔT measurements between baseline and follow up (ccIMT: 0.10±0.02 vs 0.09±0.02, p=0.06 and ΔT: 0.72±0.26 vs 0.70±0.26). In contrast, there was a significant reduction in ccIMT and ΔT for patients under both simvastatin (0.10±0.03 vs 0.09±0.01, p=0.004 for ccIMT and 0.83±0.34 vs 0.63±0.24, p=0.04 for ΔT) and simvastatin + ezetimibe therapy (0.11±0.03 vs 0.09±0.02, p<0.001 and 1.00±0.38 vs 0.69±0.23, p<0.001 for ΔT). Patients under combination therapy reduced more significantly their carotid artery temperatures compared to patients under simvastatin monotherapy or patients without statin (−0.31±0.46 vs −0.2±0.40 vs −0.01±0.37, ANOVA p=0.04, Figure 1).

Conclusions

Both simvastatin and simvastatin + ezetimibe therapy among the beneficial effect on IMT, reduced carotid wall inflammation in FH pts.

Figure 1

Funding Acknowledgement

Type of funding source: None

P6157Indexed coronary volume - A potential novel low-dose CCTA derived predictor for cardiovascular events

Background

Coronary computed tomography angiography (CCTA) provides incremental prognostic information over traditional risk factors in patients with suspected coronary artery disease. However, little is known about the long-term predictive performance of CCTA-derived coronary volumes and mid-diastolic left ventricular (LV) mass.

Purpose

To assess long-term prognostic value of coronary volumes and mid-diastolic LV mass as novel potential imaging predictors derived from low-dose prospectively ECG-triggered CCTA.

Methods

Consecutive patients with suspected or known coronary artery disease, referred for low-dose CCTA, were included. Patients with previous revascularization were excluded. The following parameters were evaluated: calcium score, segment involvement score (SIS: 1 point for each coronary segment with presence of plaque), coronary volume, mid-diastolic LV mass and coronary volume indexed to LV mass. Major adverse cardiovascular events (MACE) were defined as all-cause death, non-fatal myocardial infarction and revascularization (PCI or CABG). The association between CCTA measures and the occurrence of events was quantified using cox regression hazard and Kaplan Meier analysis.

Results

A total of 147 consecutive patients were included in the study. Of them, 93 (63.3%) were male and 79 (53.7%) hat one or more traditional cardiovascular risk factors. There was a weak but statistical significant inverse correlation between indexed coronary volume and both calcium score (R=-0.3, p=0.01) and SIS (R=-0.24, p=0.005). After a median follow-up of 5.8 years 30 MACE occurred in 25 patients, including 3 deaths, 26 revascularizations and 1 non-fatal myocardial infarction. In univariate cox regression hazard analysis calcium score (HR=12.69, 95% CI 2.99–53.83, p<0.001), SIS (HR=1.66, 95% CI 1.43–1.94, p<0.001), LV mass (HR=1.02, 95% CI 1.01–1.03, p=0.007) and indexed coronary volume (HR=0.89, 95% 0.82–0.96, p=0.004) were associated with outcome. In multivariate analysis, indexed coronary volume, remained an independent predictor for MACE when adjusted for traditional risk factors and SIS (HR=0.93, 95% CI 0.87–1.00, p=0.05), while LV mass did not reach statistical significance (p=0.46). By ROC curve analysis, a value of 21.85 mm3/gr was defined as optimal cutoff for indexed coronary volume. In Kaplan Meier plots, patients with low indexed coronary volume (<21.85 mm3/gr) showed higher event rates (log rank p<0.001) compared to high indexed coronary volume (≥21.85 mm3/gr).

Conclusions

Indexed coronary volume, derived from low-dose CCTA, independently predicts cardiovascular events. Larger studies are mandated to confirm the predictive value of this potential new biomarker.

P3716Increased two-year cerebrovascular event rate in patients with bilateral high carotid tempratures

Introduction

The association of carotid plaque inflammation with cerebrovascular events is a matter of rigorous research. Microwave Radiometry (MWR) allows in vivo noninvasive measurement of the internal temperatures of tissues, reflecting inflammation.

Purpose

To investigate whether increased carotid temperatures in patients with documented coronary artery disease (CAD) are associated with cerebrovascular events.

Methods

Consecutive patients with significant CAD from three tertiary centers were included in the study. Maximum carotid plaque thickness was assessed in all carotids by ultrasound. ΔT by MWR was assigned as the temperature difference (maximal minus minimum) along the carotid artery. ΔT ≥0.90°C was assigned as high ΔT. All patients were followed-up clinically for two years and all strokes were adjudicated by an independent committee. Transient ischemic attacks were excluded.

Results

In total 300 patients were included in the study. High ΔT temperatures bilaterally were measured in 47 patients (15.7%). Three patients (1.0%) suffered a stroke, including one fatal. Stoke rate was 4.3% in the group with bilateral high ΔT and 0.4% in non-high ΔT group (p=0.02). In Kaplan-Meier plot patients with bilateral high ΔT showed higher stroke rate (log-rank p=0.004, figure)

Conclusions

Bilateral high carotid temperatures are associated with increased two-year stroke rate. The potential value of the present finding in risk stratification of intermediate carotid stenosis mandates further investigation.

P5302Significant correlation of visceral adiposity and adipocytokines with arterial inflammation in genetic dyslipidaemias

Background

The adipose tissue is now established as a major regulator of cardiovascular status, mediated by the secretion of several bioactive molecules, including adipocytokines. Individuals with genetic dyslipidaemias of either familial combined hyperlipidemia (FCH) or heterozygous familial hypercholesterolemia (heFH) subtype are characterized by accelerated atherosclerosis. Nonetheless, limited data exists on the relationship between adiposity and arterial inflammation, a marker of cardiovascular risk, in this setting.

Purpose

To investigate the relationship between adiposity indices and arterial inflammation evaluated by 18F fluorodeoxyglycose positron emission tomography (PET/CT) in patients with hereditary lipid metabolism disorders.

Methods

Consecutive patients with either FCH or heFH, free of statin therapy, and normolipidaemic individuals underwent PET/CT imaging. Arterial FDG uptake was estimated as the average value of target-to-background ratio (TBR) within aortic and carotid wall. Volumes of subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) were measured from CT images between the proximal (cephalic) end of the L1 and distal (caudal) end of the L3 vertebrae by selecting all voxels with attenuation between −190 and −30 Hounsfield Units (HU). Serum adiponectin and leptin levels were measured with ELISA by available commercial kits. For comparisons with arterial TBR, adiponectin and leptin concentrations above and below the 25th percentile, were stratified as high and low, respectively.

Results

In total, 60 individuals (20 FCH, 20 heFH and 20 controls) were included. A modest but significant correlation between SAT volume and arterial TBR (R=0.386, p=0.004) was detected. This relationship did not remain significant after controlling for VAT volume (p=0.303). A strong correlation between VAT volume and arterial TBR (R=0.621, p=0.001) was observed. This relationship remained significant after controlling for SAT volume (R=0.541, p<0.001). Arterial TBR values were higher in individuals with low plasma adiponectin levels (p=0.010). In addition, patients with higher leptin levels exhibited increased arterial FDG uptake compared to subjects with low serum leptin concentrations (p=0.05)

Conclusions

Abdominal adipose tissue imaging markers and serum adipocytokines correlate with arterial inflammation as assessed by PET/CT in patients with familial dyslipidaemias highlighting the role of abdominal adipose tissue for atherosclerosis progression in this population.

P730Decreased adiponectin levels and FDG uptake in visceral adipose tissue in familial combined hyperlipidemia compared to heterozygous familial hypercholesterolemia and normolipidemics

Background

Adipose tissue regulates energy balance and glucose homeostasis via the secretion of circulating molecules, termed adipokines, such as leptin and adiponectin. Excess adiposity and adipose tissue dysfunction have been involved in the pathogenesis of dyslipidemias. Positron emission tomography/computed tomography (PET/CT) with F-18-Fluorodeoxyglycose (FDG) has been used for the assessment of adiposity.

Purpose

To compare abdominal adipose tissue function assessed by FDG uptake with serum indices, such as plasma adipokines' levels in individuals with different subtypes of dyslipidemia and normolipidemics.

Methods

Seventy individuals (mean age 44±13 years, range 21–75, 43 men) with a clinical diagnosis of either heterozygous familial hypercholesterolemia (heFH) (n=38) or familial combined hyperlipidemia (FCH) (n=32), not under statins for at least one year, and 20 age and sex matched controls, were enrolled. Visceral (VAT) and subcutaneous adipose tissue metabolic activity (SAT) was assessed with FDG-PET/CT imaging and was quantified by calculating the target-to-background ratios (TBR) in consecutive axial fat images between the proximal (cephalic) end of the L1 and distal (caudal) end of the L3 vertebrae by dividing the average of the mean standard uptake value (SUV) to the mean SUV of the vena cava. Leptin and adiponectin were measured in all the subjects.

Results

There was no significant difference of plasma leptin values between FCH, heFH and non dyslipidemics subjects (p=0.204). FCH had reduced adiponectin values compared to heFH patients and controls [median 5.7 IQR (3.9–7.6) vs. 13.1 (9.2–23.3) vs. 10.9 (6.1–19.1) μg/mL, respectively, p<0.001]. There was no difference in FDG uptake in subcutaneous adipocytes (SATTBR) between FCH, heFH and controls (p=0.161). In contrast, patients with FCH had reduced VATTBR values compared to heFH patients and controls (0.63±0.14 versus 0.81±0.17 versus 0.86±0.28, p=0.005). This difference remained significant even after adjustment for age, sex and cardiovascular risk factors (b=-0.428, p=0.001, adjusted R2=0.219). SATTBR was inversely correlated to leptin levels (r=−0.484, p<0.001), while no significant association was observed with adiponectin values (p=0.167). No significant associations were observed between VATTBR and either serum leptin (p=0.066) or adiponectin levels (p=0.254).

Conclusions

Visceral adipose tissue FDG uptake is reduced in patients with FCH compared to those with heFH and normolipidemics. In addition, serum adiponectin levels are lower in patients with FCH. These findings highlight the different pathophysiological role of visceral fat function in the two most common types of familial dyslipidemia and suggest that visceral fat could be an attractive target for the treatment of FCH.