P91 Prognostic value of tissue-tracking mitral annular displacement by speckle-tracking echocardiography in asymptomatic patients with aortic stenosis with preserved left ventricular ejection fraction

Background

  Management of asymptomatic severe aortic stenosis (AS) patients with preserved left ventricular (LV) ejection fraction (EF) remains controversial. Recent studies using have shown that decreased LV longitudinal deformation assessed by global longitudinal strain analysis can predict adverse cardiac events in AS patients with preserved EF. Tissue-tracking mitral annular displacement (TMAD) by speckle-tracking echocardiography provides rapid and simple assessment of LV longitudinal deformation even when the acoustic window is poor (Fig.1).

Purpose

 The purpose of this study was to examine the value of TMAD to predict occurrence of the cardiac events in asymptomatic severe AS patients with preserved EF.

Methods

  We studied 103 patients with severe AS and preserved EF [aortic velocity >4m/s or aortic valve area (AVA) <1.0 cm2, EF >50%]in whom TMAD was measured, and a total of 44 patients were included in the final data setaccording to the exclusion criteria. Using TMAD analysis software, the base-to-apex displacement of automatically defined mid-point of mitral annular line in four-chamber view was quickly assessed, and the percentage of its displacement to LV length at end-diastole (%TMAD) was calculated (Fig.1). We investigated the association between %TMAD and the cardiac events including implementation of hospitalization due to heart failure, decreased EF (< 50%), aortic valve replacement or transcatheter aortic valve implantation due to appearance of symptoms and cardiac death, 

Results

 In all the final study patients, %TMAD was successfully and quickly (within 10 seconds) evaluated. During a follow-up, the cardiac events developed in 16 (36%) of 44 patients. Tableshows echocardiographic parameters in patients with and without the cardiac events. %TMAD was significantly impaired in patients with the cardiac events compared with those without the cardiac events (9.6 ± 0.6 vs 12.1 ± 0.4%, p= 0.002). The other parameters were not involved in the event occurrence; age, LV mass index, EF, aortic velocity, AVA, tricuspid regurgitation pressure gradient (TR-PG), early diastolic /atrial filling velocity (E/A), early diastolic velocity of the mitral valve annulus (e’) and E/e’. In multiple variable analysis, %TMAD was an independentpredictor of the cardiac events (HR; 12.1, p= 0.001). ROC analysis revealed that the area under the curve of %TMAD was 0.81 for the cardiac events. Kaplan-Meier analysis showed %TMAD (cut-off: 11.9) provides a significant difference in the cardiac event (Fig. 2). Conclusions.  The present results suggests that TMAD easily and rapidly estimated by speckle-tracking echocardiography can be used as a simple method to predict occurrence of the cardiac events in asymptomatic severe AS patients with preserved EF.

Abstract P91 Figure 1,2 and Table

P1252Diagnostic performance of quantitative flow ratio from coronary angiography versus fractional flow reserve from computed tomography

Background

QFR and FFRCT are recently developed, less-invasive techniques for functional assessment of coronary artery disease.

Objectives

We compared the diagnostic performance between fractional flow reserve derived from computed tomography (FFRCT) and quantitative flow ratio (QFR) derived from coronary angiography, using FFR as the standard reference.

Methods

We measured FFRCT, QFR and FFR in 152 patients (233 vessels) with stable coronary artery disease.

Results

QFR was highly correlated with FFR (r=0.78, p<0.001), while FFRCT was moderately correlated with FFR (r=0.63, p<0.001). Both QFR and FFRCT showed good agreements with FFR, presenting small values of mean difference and root-mean-squared deviation (FFR -QFR: 0.02±0.09 and FFR -FFRCT: 0.03±0.11). The AUC of QFR was significantly greater than that of 3D-QCA-derived %DS (0.93 vs. 0.78; difference: 0.15; 95% CI: 0.09 to 0.20; p<0.001). The AUC of FFRCTwas significantly greater than that of CCTA-derived %DS (0.82 vs. 0.70; difference: 0.12; 95% CI: 0.05 to 0.19; p<0.001). The AUC of QFR was significantly greater than that of FFRCT (0.93 vs. 0.82; difference: 0.11; 95% CI: 0.05 to 0.16; p<0.001). The sensitivity, specificity, positive predictive value, and negative predictive valueof QFR ≤0.80 for predicting FFR ≤0.80 were 90%, 82%, 81%, and 90%, respectively. Those of FFRCT ≤0.80 for predicting FFR ≤0.80 were 82%, 70%, 70%, and 82%, respectively. The diagnostic accuracy of QFR ≤0.80 for predicting FFR ≤0.80 was 85% [95% confidence interval: 81% to 89%], while that of FFRCT≤0.80 for predicting FFR ≤0.80was 76% [95% confidence interval: 70% to 80%].

Figure 1. Comparison of FFR ≤0.80 predictors

Conclusions

Both QFR and FFRCTpossessed the ability to accurately evaluate the functional severity of coronary stenosis.

P2700Impact of instantaneous wave-free ratio on graft failure after coronary artery bypass graft surgery

Background

It has been reported that preoperative fractional flow reserve (FFR) is associated with graft patency after coronary artery bypass graft (CABG) and the patency is excellent when a bypass graft is anastomosed on a vessel with positive FFR. However, the association with graft patency has not yet been investigated in its novel counterpart, instantaneous wave-free ratio (iFR), and iFR sometimes contradicts FFR results.

Purpose

The purpose of this study is to assess an impact of preoperative iFR on a graft failure after CABG in patients with coronary arteries showing positive FFR (≤0.80).

Methods

We retrospectively identified patients who had undergone preoperative coronary angiography in conjunction with resting and hyperemic intra-coronary pressure measurements, CABG, and graft evaluation by coronary computed tomography angiography. After excluding vessels with negative FFR (>0.80), vessels were divided into two groups: negative iFR group (iFR >0.89) and positive iFR group (iFR ≤0.89). The rate of graft failure within 1 year after CABG was compared between the two groups.

Results

We analyzed 131 vessels in 89 patients (35 vessels in the negative iFR group and 96 vessels in the positive iFR group). The negative iFR group showed significantly higher iFR (0.92±0.02 vs. 0.74±0.13, P<0.0001) and FFR (0.72±0.06 vs. 0.63±0.09, P<0.0001) than the positive iFR group, although percent diameter stenosis (%DS) was comparable (57±10 vs. 56±9, P=0.47). The graft failure significantly often occurred in the negative iFR group than in the positive iFR group (28.6% vs. 8.3%, P=0.0029). In order to reduce the imbalance in the baseline characteristics except for iFR, 70 vessels were selected by using propensity score matching (n=35 in each group). The propensity score matched vessels also demonstrated significantly higher rate of graft failure in the negative iFR group than in the positive iFR group (28.6% vs. 5.7%, p=0.026) despite much more balanced FFR (0.72±0.06 vs. 0.69±0.07, p=0.02) and %DS (57±10 vs. 57±9, p=1.000).

Conclusions

Even when FFR is positive, the graft failure is likely to occur when a bypass graft is anastomosed on a vessel with negative iFR compared to a vessel with positive iFR.

Acknowledgement/Funding

None

P3387Cholesterol crystals in superficial plaque layer detected by optical coherence tomography as a new morphological feature for plaque rupture

Background

While plaque rupture (PR) is the leading cause of acute myocardial infarction (AMI), other etiologies are also involved in the onset of AMI. Cholesterol crystals (CCs) are usually present abundantly in atherosclerotic plaques, especially in the culprit site of AMI. However, the relationship between in vivo CCs and PR is unclear. Optical coherence tomography (OCT) is a high-resolution imaging technique that allows for the in vivo identification of various plaque characteristics including PR and CCs.

Purpose

The aim of this study was to investigate prevalence and distribution of CCs between patients with AMI with PR, AMI without PR, and SAP.

Method

This study consisted of 146 patients with coronary artery disease (AMI with PR; n=64, AMI without PR; n=41, and SAP; n=41) who underwent OCT prior to percutaneous coronary intervention. Plaque characteristics in OCT images were assessed according to the consensus document. We classified the distribution of CCs as follows; superficial type CCs were defined by any of the CCs invading the fibrous cap and remaining CCs as deep type CCs.

Result

There was no statistical difference in clinical characteristics among the three groups. The % diameter stenosis was significantly smaller in the SAP group than others (AMI with PR 91±12% vs. AMI without PR 86±13% vs. SAP 65±9%, p<0.001). The prevalence of CCs was significantly higher in the AMI with PR group than others (AMI with PR 78% vs. AMI without PR 41% vs. SAP 39%, p<0.001). The prevalence of superficial type CCs was significantly different among the groups (AMI with PR 72% vs. AMI without PR 24% vs. SAP 7%, p<0.001). Multivariable logistic analysis demonstrated that lipid plaque (OR 84.5, 95% CI [6.30–11332.33], p<0.001) and superficial type CC (OR 9.5, 95% CI [2.61–34.89], p<0.001) were independent predictors of PR.

Conclusion

Plaque with CCs invading the fibrous cap is frequently associated with PR in patients with AMI, suggesting. In vivo CC detection is a new morphological feature for plaque rupture.

Acknowledgement/Funding

This study was supported by a grant from JSPS KAKENHI (17K09557).

P3365Relationship between early diastolic intra-ventricular pressure gradient shortly after aortic valve closure estimated by vector flow mapping and left ventricular diastolic untwisting rate in humans

Background

Early diastolic suction is an important determinant of early diastolic function. Previous studies using color Doppler M-mode and speckle-tracking echocardiography have shown left ventricular (LV) early diastolic LV untwisting rate is directly related to LV intra-ventricular pressure gradient (IVPG) between LV base and apex during early diastole. Recent introduction of vector flow mapping (VFM) using combination of color Doppler and speckle-tracking echocardiography provides noninvasive and feasible assessment of early diastolic IVPG shortly after aortic valve closure including isovolmic relaxation period (ED-IVPG) in humans. However, relationship between VFM–derived ED-IVPG and early diastolic LV untwisting rate has not been well investigated.

Purpose

The purpose of this study was to examine relationship between ED-IVPG estimated by VFM and LV untwisting rate by speckle-tracking echocardiography.

Methods

The study population consists of 66 patients without segmental wall motion abnormality, significant valvular diseases, and atrial fibrillation who underwent echocardiography for evaluation of LV function (age: 60±15 years, LVEF: 49±16%). From the apical long-axis views by color Doppler echocardiography, we analyzed peak ED-IVPG between LV base and apex just after aortic valve closure (figure) using commercially available VFM analysis software (DAS-RS1, Hitachi). We assessed peak early diastolic LV untwisting rate and LV torsion from LV basal and apical short-axis view by speckle-tracking echocardiography. We evaluated correlation between ED-IVPG and LV untwisting rate. We also evaluated correlation between ED-IVPG and peak systolic LV torsion, LV end-diastolic (EDV) and end-systolic volumes (ESV), ejection fraction (EF), early diastolic velocity (E) of LV inflow, average early diastolic velocity (e') of mitral annulus, and average E/e'.

Results

In all the study patients, ED-IVPG was successfully and quickly evaluated. 1) ED-IVPG correlated well with peak LV untwisting rate (r=0.64, p<0.0001). 2) ED-IVPG significantly correlated with LV torsion, LVEDV, LVESV, and LVEF (r=0.47, r=−0.48, r=−0.46, and r=0.48, respectively, p<0.001). 3) There were no significant correlations between ED-IVPG and other indexes including E, average e', and average E/e'. According to receiver operating characteristic analysis, the best cut-off value of ED-IVPG for determining impaired LV untwisting rate (<80 degrees/s) was found at 0.42 mmHg (sensitivity 81%, specificity 76%, and area under the curve 0.86)

ED-IVPG measurement by VFM

Conclusions

The present results showed that noninvasive VFM-derived peak ED-IVPG shortly after aortic valve closure is related to early diastolic peak LV untwisting rate. ED-IVPG easily and quickly estimated by VFM may be used as an additional index for LV diastolic function.

P2453Value of pre-operative left atrial minimum volume as a surrogate for post-operative symptoms in patients with aortic stenosis who underwent aortic valve replacement

Background

Previous reports have shown that symptoms after aortic valve replacement (AVR) are not uncommon depending on severity of myocardial fibrosis in patients with severe aortic stenosis (AS). Pre-operative minimum left atrial volume (LAVmin) at end-diastole determined by direct exposure of left ventricular end-diastolic pressure may be used as a surrogate for post-operative symptoms in patients with severe AS undergoing AVR.

Purpose

The purpose of this study was to examine the value of pre-operative echocardiographic LAVmin index (LAVImin) to predict post-operative symptomatic status after AVR in patients with severe AS.

Methods

The study population consisted of 219 patients with severe AS who underwent AVR and were followed up for 1000 days after AVR. Pre-operative maximum LAV index (LAVImax), LAVImin, LA emptying fraction (LAEF), LV volume indexes, LV ejection fraction (LVEF) by biplane Simpson's method, aortic valve area index (AVAI), mean aortic valve pressure gradient (mAV-PG), E/A, mean E/e' from LV inflow and mitral annular velocity, and pulmonary artery systolic pressure (PASP) were evaluated by Doppler echocardiography.

Results

After exclusion of 136 patients who met the exclusion criteria (atrial fibrillation, significant coronary artery disease, significant mitral valve diseases, pacemaker rhythm, and inadequate echocardiographic images), the final study population consisted of 75 patients (75±7 years old, 46 female). During a follow-up, 19 patients (25%) complained post-operative symptoms. There were no significant differences in pre-operative serum hemoglobin, creatinine, BNP, chronic obstructive pulmonary disease, hypertension, diabetes, LV volume indexes, LVEF, AVA, mAV-PG between patients with and without post-operative symptoms. There were significant differences in pre-operative LAVImax, LAVImin, and LAEF between patients with and without post-operative symptoms. (60±15 vs 47±15 ml/m2, 45±15 vs 28±1 ml/m2, and 29±12 vs 42±11 ml/m2, respectively). E/A, mean E/e', and PASP in patients with symptoms were significantly greater compared with patients without symptoms (1.0±0.3 vs 0.7±0.2, 25±3 vs 18±2, 44±17 vs 32±9 mmHg, respectively). In the multivariate analysis, pre-operative LAVImin was the independent predictor of the post-operative symptomatic status after AVR (odds ratio: 1.11, 95% confidence interval: 1.04 - 1.18). Receiver operating characteristic analysis revealed that area under the curve (AUC) of LAVImin (cutoff: 30ml/m2) for post-operative symptoms was the largest (0.84) among the other echocardiographic parameters, and significantly larger than that of mean E/e' (0.67, *p<0.01) and LVEF (0.53, **p<0.05) (figure).

Figure 1. ROC analysis

Conclusions

The present results suggest that pre-operative echocardiographic LAVImin may be used as a surrogate for post-operative symptomatic status after AVR in patients with severe AS.

102Diagnosis of coronary plaque rupture, plaque erosion, and calcified nodule by using near-infrared spectroscopy intravascular ultrasound

Objectives

This study sought to investigate the ability of near-infrared spectroscopy intravascular ultrasound (NIRS-IVUS) to differentiate among plaque rupture (PR), plaque erosion (PE), and calcified nodule (CN) in acute myocardial infarction (AMI) using an optical coherence tomography (OCT) diagnosis as a reference standard.

Background

In vivo, precise differentiation among PR, PE and CN is a major challenge for intravascular imaging.

Methods

The study enrolled 156 AMI patients who had a de novo culprit lesion in a native coronary artery. The culprit lesions were assessed by both NIRS-IVUS and OCT.

Results

OCT identified 112 PR, 29 PE, and 15 CN. IVUS-detected plaque ulceration showed a high specificity (100%) to identify OCT-PR although the sensitivity (62%) was intermediate. IVUS-detected convex calcium showed a high sensitivity (93%) and specificity (100%) to identify OCT-CN. In NIRS, the maximum lipid core burden index in 4 mm (maxLCBI4mm) was greatest in OCT-PR (values are median [interquartile range]) (671 [530 to 853]), followed by OCT-CN (355 [303 to 432]) and OCT-PE (283 [89 to 357]) (p<0.001). MaxLCBI4mm of <422 was the best cut-off to discriminate OCT-PE from OCT-PR and OCT-CN. The NIRS-IVUS classification algorithm using plaque ulceration, convex calcium, and maxLCBI4mm <422 showed a sensitivity and specificity of 96% and 95% for identifying OCT-PR, 93% and 95% for OCT-PE, and 93% and 100% for OCT-CN, respectively.

NIRS-IVUS classification algorism

Conclusion

Lipid component assessed by NIRS-IVUS was different among OCT-PR, OCT-PE and OCT-CN. The NIRS-IVUS classification algorism was highly sensitive and specific for differentiating these unstable lesion types in AMI.

Acknowledgement/Funding

None

P4349Simple and rapid estimation of left ventricular longitudinal deformation by tissue-tracking mitral annular displacement in single apical view

Background

Noninvasive assessment of left ventricular (LV) deformation using global longitudinal strain (GLS) has prognostic value in patients with and without preserved ejection fraction (EF). Application of speckle-tracking technology to the mitral annulus provides rapid and easy assessment of displacement of septal and lateral mitral annulus and mid-point of mitral annular line in single apical view (TMAD) even in poor echo-image quality. TMAD may be used as a simple index of LV longitudinal deformation in patients with and without preserved EF (Figure).

Purpose

The purpose of this study was to examine whether TMAD can be used as a simple index of LV longitudinal deformation in patients with and without preserved EF.

Methods

The study population consists of 95 patients without segmental wall motion abnormality, significant valvular diseases, and atrial fibrillation in whom both TMAD and GLS measurements were applied by QLAB software (Philips). We estimated GLS from apical 4- and 2-chamber views and apical longitudinal views, and TMAD from apical 4-chamber view. TMAD was automatically and quickly evaluated as the base-to-apex displacement of septal (TMADsep), lateral (TMADlat), and mid-point of annular line (TMADmid) (Figure). The percentage of M-TMAD to LV length from the mid-point of mitral annuls to the apex at end-diastole (%TMADmid) was also calculated. We compared each TMAD values with GLS values by linear regression analysis, and evluated TMAD values by a receiver operating characteristic (ROC) analysis to detect impaired LV longitudinal deformation (|GLS|<12.0%).

Results

TMAD was successfully assessed in 94 of 95 patients (99%) while GLS was measured in 84 of 95 patients (87%, p=0.0082 vs TMAD). There were good correlations between each TMAD index and |GLS| (TMADsep:r=0.77, TMADlat:r=0.81, TMADmid:r=0.82, %TMADmid:r=0.87). According to ROC curve, the best cut-off values for TMADsep, TMADlat, TMADmid, and %TMADmid in determining LV longitudinal deformation were 6.8mm, 8.0mm, 7.8mm, and 9.5% respectively (Table).

Conclusions

The present results suggest that rapid and easy assessment of TMAD in single apical view may be used as a simple index of LV longitudinal deformation.