Search

The purpose of the current study was to evaluate the utility of using machine-learning (ML) algorithm and their ability to integrate data inputs to make predictions of future development of cardiovascular disease (CVD) in a nationwide sample of middle-aged Korean adults. Study population was randomly sampled from the Korea National Health Insurance (NHI) database. We identified 143,453 people (79,584 men and 63869 women) who were aged 40 years or older, free from CVD or cancer, and completed health screening tests between 2002 and 2003. CVD was defined as composite of cardiovascular death, myocardial infarction, stroke, and coronary artery disease requiring coronary artery intervention or bypass surgery until 2013, and were identified from the NHI claim database. We employed multiple supervised ML algorithms to build a CVD prediction models using clinical and laboratory data during the follow-up period. Prediction performance of the ML algorithm was compared to those of traditional scoring system or logistic prediction models. ML algorithm [area under curve (AUC) 0.895, 95% confidential interval (CI) 0.889-0.902] outperformed a traditional scoring system (AUC 0.721, 95% CI 0.713-0.730), logistic regression model with baseline characteristics (AUC 0.731, 95% CI 0.723-0.739) and logistic model with time-series analysis (AUC 0.804, 95% CI 0.797-0.811) in prediction of future CVD in men. In women, ML algorithm (AUC 0.908, 95% CI 0.901-0.915) also exhibited the most accurate predictive power, compared to the traditional scoring system (AUC 0.749, 95% CI 0.740-0.758), logistic regression model with baseline characteristics (AUC 0.775, 95% CI 0.767-0.784) and logistic model with time-series analysis (AUC 0.842, 95% CI 0.834-0.849). Our findings suggest that ML algorithm with time-series data can improve the performance of CVD risk prediction.

Introduction: Unprovoked venous thromboembolism (VTE) is related higher incidence of occult cancer. However, routine screening for cancer in this population is still controversial. D-dimer is a biomarker reflecting the activation of hemostasis and fibrinolysis. D-dimer is clinically used to exclude VTE, also related in patients with malignancy. Hypothesis: We aimed to explore the predictive value of d-dimer level for detecting occult cancer in patients with unprovoked VTE. Methods: We retrospectively analyzed 840 patients who diagnosed deep vein thrombosis or pulmonary thromboembolism. Among them, 175 (20.8%) patients who diagnosed with unprovoked VTE selected to participate in this study. Odds ratios (OR) with 95% confidence intervals (95% CI) were calculated for incidence of occult cancer as well as d-dimer level reference to patient without occult cancer. Results: During follow-up for 5.0±1.9 years, 30 of 840 patients diagnosed cancer in this population. The primary origin of tumor was 6 (20%) of stomach, 6 (20%) of lung, 4 (13%) of colon and others. Number of patients with metastatic cancer was 17 (57%). Log d-dimer was significantly higher in occult cancer group than unprovoked VTE without cancer (3.5 ± 0.5 vs. 3.2 ± 0.5, p=0.009). When the patients with unprovoked VTE were divided by d-dimer according to Youden index, high d-dimer (≥2432) was associated with higher prevalence of occult cancer (26.9% vs. 9.2%, p=0.002). In logistic regression analysis, d-dimer is independently associated with occult cancer (OR: 3.93, 95% CI: 1.46-10.54, p=0.007) after adjusted gender, age, type of VTE (i.e., deep vein thrombosis or pulmonary thromboembolism) and tumor markers. In addition, d-dimer is more significantly associated with patients who diagnosed cancer with distant metastasis (OR: 4.31, 95% CI: 1.16-15.99, p=0.029) Conclusions: High level of d-dimer concentrations are indicator of an increased risk of occult cancer in patients with unprovoked VTE.

Introduction: Zero coronary artery calcium (CAC) is a reliable predictor of absent atherosclerosis and serves as a useful adjunct for identifying those at low risk. Despite this, the “warranty period” that displays the protective value associated with the absence of CAC towards experiencing a cardiovascular event among Asians is not well established. Hypothesis: To examine whether the absence of CAC displays a more favorable warranty period for all-cause death compared with the presence of CAC in a Korean cohort. Methods: A total of 48,215 asymptomatic Koreans (mean age: 54±8.8 years; 25% women) were enrolled and stratified by the absence or presence of CAC. Time to exceeding 1% of cumulative all-cause death was estimated in order to identify low-risk individuals. Hazard ratios (HR) with 95% confidence intervals (95% CI) for all-cause death were estimated according to prespecified cardiac risk factors and the presence of CAC. Results: In total, 30,605 (63.5%) individuals presented with a zero CAC. Across a median follow-up of 4.4 years (Interquartile range: 2.7-6.6 years), 415 (0.9%) individuals experienced the endpoint of all-cause death. For those with a zero CAC, the time to exceeding 1% risk was found to be 9 years, indicating a substantially longer warranty period compared with participants with a CAC>0 (e.g., 5 year warranty period). The time to exceeding 1% risk tended to decline for individuals on the background of increasing CAC scores. For each of the other prespecified risk factor groups, a zero CAC provided a longer cumulative event free period than in the presence of any CAC. Cox regression analyses also revealed that the absence of CAC was independently associated with a lower risk of all-cause death in each of the respective risk factor groups when compared with CAC>0. Conclusions: In a large cohort of asymptomatic Korean individuals, the absence of CAC evokes a strong protective effect against all-cause death as demonstrated by a longer warranty period.

Background: We aimed to characterize progression of aortic stenosis (AS) with end stage renal disease (ESRD), and to determine metabolic and hemodynamic factors attributing to AS progression. Method:A total of 73 ESRD (50 males, age 72±11 years) on dialysis with mild (n=56) to moderate AS (n=17) were enrolled. Age, sex and AS severity matched control group with eGFR above 60 mL/min/1.73m2 was selected (n=79). Transthoracic echocardiography was performed at baseline and during follow up (48±23 months). Severity of AS was accessed by aortic valve area (AVA) calculated by continuity equations. Acceleration of AS was assessed by annual AVA reduction rate. Accelerated group in ESRD patients was defined as those with the highest quartile of AVA reduction rate. Clinical outcomes included aortic valve intervention, hospital admission due to heart failure, and death. Results: Annual AVA reduction rate was significantly higher in ESRD compared to control group (0.36±1.9 vs. 0.05±0.14 cm2/yr, p Conclusion: AS in ESRD progress in more accelerated manner compared to those with normal kidney function, even in mild to moderate AS. Therefore, more frequent follow up with proper risk stratification by assessing hemodynamic and metabolic characteristics, such as LAVI and PTH, may benefit ESRD patients to reduce clinical events.

Introduction: While coronary computed tomographic angiography (CCTA) demonstrates high diagnostic performance for identification and exclusion of high-grade anatomic stenosis, it is unable to effectively discriminate coronary lesions that causes ischemia. Hypothesis: To study whether quantitative assessment of CCTA offer incremental information for discrimination of lesion ischemia beyond stenosis measures. Methods: 252 consecutive patients with suspected or known coronary artery disease (CAD) from 17 centers in 5 countries were enrolled (mean age 63±9 years, 71% male). Patients underwent CCTA and invasive coronary angiography (ICA), with 407 coronary lesions interrogated by invasive fractional flow reserve (FFR) at the time of maximum hyperemia. For these coronary lesions, we evaluated % diameter stenosis (%DS), % area stenosis (%AS), minimal luminal diameter (MLD, mm), minimal luminal area (MLA, mm 2 ), and plaque burden at MLA by CCTA. Plaque burden (PB, %) was defined as (vessel area–lumen area)/vessel areaх100. Lesion-specific ischemia by FFR was defined as a value ≤0.8. Results: In quantitative analysis area under the receiver operating characteristic curves (AUC) value of %DS, %AS, MLD, MLA, and PB for prediction of ischemia were 0.72 (95% confidence interval (CI) 0.68-0.77, p Conclusions: Quantitative plaque assessment using CCTA could predict lesion-specific ischemia with good discrimination. Plaque burden showed incremental value over % area stenosis for ischemia prediction.

Background: Beyond stenosis severity, coronary computed tomographic angiography (CCTA) enables morphologic evaluation of atherosclerotic plaque characteristics (APCs), including aggregate plaque volume, positive arterial remodeling (PR), low attenuation plaque (LAP) and spotty calcification (SC), features associated with global myocardial ischemia and acute coronary syndromes. We determined whether the combination of stenosis severity plus APCs would improve the discrimination of coronary lesions that cause ischemia. Method: 252 patients from 17 centers in 5 countries [mean age 63 years, 71% males] underwent CCTA and invasive angiography, with invasive fractional flow reserve (FFR) performed for 407 coronary lesions. Stenosis severity was determined by luminal diameter reduction (%DS). Aggregate plaque volume (APV) was defined as the plaque volume from ostium to distal end of the lesion, with APV percent (%APV) defined as APV divided by total vessel volume. Other APCs by CCTA were defined as: (1) PR, lesion diameter/reference diameter >1.10; (2) LAP, any voxel Results: Coronary stenosis severity and APCs demonstrated good discrimination for lesion ischemia: %DS, (Area under the receiver operating characteristics curve [AUC] of 0.72 (95% CI 0.68-0.77), APV (AUC 0.69, 95% CI 0.64-0.73), and %APV (0.75, 98% CI 0.70-0.79). Over %DS alone, the addition of %APV improved the discrimination of lesion-specific ischemia (0.79, 95% CI 0.75-0.83, p Conclusion: In this prospective multicenter international study, the combination of quantitative measures of stenosis severity with morphologic features of atherosclerotic plaque by CCTA resulted in improved discrimination of coronary lesion-specific ischemia.

Introduction: Data describing the prevalence, characteristics, and management of chronic total occlusions (CTO) in patients undergoing coronary computed tomography angiography (CCTA) have not been reported. Objectives: The purpose of this study was to determine the prevalence, characteristics, and treatment strategies of CTO identified by CCTA. Methods: We identified 23,745 patients who underwent CCTA for suspected coronary artery disease (CAD) from the CONFIRM (COronary CT Angiography EvaluatioN For Clinical Outcomes: An InteRnational Multicenter) registry. Baseline clinical data were collected, and allocation to early coronary revascularization performed within 90 days of CCTA was determined. Multivariable logistic regression reporting odds ratios (OR) with 95% confidence intervals (95%CI) was performed. Results: Patients with CTO were older (64.7 ± 10.8 vs 57.2 ± 12.6 years, p Conclusions: CTOs are not uncommon in a contemporary CCTA population and are associated with age, gender, angina status and CAD risk factors. Most individuals with CTO undergoing CCTA are managed medically with higher rates of surgical revascularization in patients with versus without CTO.

Introduction: Coronary CT angiography (CCTA) demonstrates improved performance for diagnosis of high-grade coronary stenoses, but may be affected by artifacts and overestimation of stenosis severity. Whether the addition of resting myocardial perfusion attenuation patterns subtended by stenosis seen on CCTA improves diagnostic performance has not been examined to date. Methods: We evaluated 127 patients (mean age 53.0, 54.3% male) who underwent CCTA and ICA. Percentage of coronary stenosis was assessed by quantitative coronary angiography (QCA), which served as the reference comparator to CCTA. CCTA stenosis was categorized as 0%, 1-24%, 25-49%, 50-69%, 70-99%, and 100% luminal diameter reduction. Automated software (SmartHeart, Redwood City, CA) was used to measure resting CT perfusion attenuation patterns in myocardial segments by AHA 17-segment model. Segmental CT attenuation values were assigned to territories subtended by left anterior descending (LAD), left circumflex (LCX), and right coronary arteries (RCA). Per-patient and per-vessel analyses were based on highest severity (maximal stenosis, minimal attenuation). On both per-patient and per-vessel basis, logistic regression was devised for CCTA stenosis alone and for CCTA plus resting myocardial attenuation. Diagnostic accuracy and area under the receiver operating characteristics curve (AUC) were determined. Results: Diagnostic accuracy of CCTA alone was 84.0%, 85.5%, 90.4%, and 88.6%, at per-patient, per-LAD, per-LCX and per-RCA level, respectively. In comparison, the accuracy of CCTA plus myocardial attenuation were 89.6%, 91.9%, 95.2%, and 92.7%. The AUCs using CCTA alone to discriminate QCA-confirmed coronary stenoses >70% were 0.823 (95% CI: 0.737-0.909), 0.782 (95% CI: 0.667-0.898), 0.690 (95% CI: 0.503-0.878), and 0.793 (95% CI: 0.640-0.945) for per-patient, per-LAD, per-LCX, and per-RCA analysis, respectively. The AUCs using CCTA plus myocardial attenuation improved to 0.864 (95% CI: 0.765-0.962), 0.881 (95% CI: 0.793-0.968), 0.772 (95% CI: 0.535-1.000), and 0.820 (95% CI: 0.685-0.954). Conclusions: The addition of resting CT myocardial perfusion attenuation patterns improves identification and discrimination of high-grade coronary stenosis by CCTA.

Background: Apical hypertrophic cardiomyopathy (ApHCM) might have different morphological and clinical features according to the age of disease onset, probably result from the distinct mechanism of pathogenesis. The aim of this study was to evaluated the morphological and clinical differences according to the age of disease onset in patients with ApHCM. Methods: This was a retrospective, observational study of 56 ApHCM patients (31 males, mean 65±9 years). Among 426 patients with diagnosed ApHCM from 2000 until 2013, we selected 56 patients who met the inclusion criteria: [[Unable to Display Character: ⑴]] patiens who diagnosed with ApHCM before 60 years old and current age >60 years old, (group A, early onset, n=16); and [[Unable to Display Character: ⑵]] who diagnosed with ApHCM after 60 years old who had no evidence of apical hypertrophy confirmed by previous echocardiography (group B, late onset, n=40). Morphological LV feature was assessed by transthoracic echocardiography, which were divided into pure and mixed type. Furthermore, hemodynamic parameters and clinical events were also evaluated. Results: There were no differences in the presence of hypertension, LV systolic function and left atrial volume index between group A and B. However, early onset group A had more increased maximal wall thickness of LV apex (17.08±2.02 vs 15.43±1.68mm, p=0.005), and more incidence of pure type (100% (16/16) vs 56.8% (21/40), p=0.001) compare than late onset group B. Group B had older age (65±5 vs 72±4 years, p=0.001), increased LV end-diastolic dimension (51.25±2.32 vs 48.58±3.37mm, p=0.006), increased E/E’ (11.24±2.94 vs 14.60±6.37, p=0.049) with decreased E’ (5.61±1.34 vs 4.66±1.38cm/s, p=0.024) respectively.There were no significant differences in the risk of cardiovascular complication such as atrial fibrillation and stoke between group A and B. Conclusions: There was obvious different clinical characteristics according to the age of disease onset. Early onset ApHCM has typical morphological feature and relatively preserved LV diastolic function, compared with late onset ApHCM. Our result supports that late onset ApHCM was related to decreased LV diastolic function and increased LV wall thickness, which reflects the potential myocardial remodeling with aging.

An 86-year-old woman without any history of heart disease or trauma to the chest experienced an acute onset of dyspnea associated with a productive cough for 2 hours. Her medical history revealed that she was hypertensive for the past 20 years. Her blood pressure was 90 mmHg in systole and 50 mmHg in diastole. Immediately after admission, mechanical ventilation was initiated because there was progressive respiratory depression. A chest x-ray showed marked pulmonary congestion (Figure 1A). On physical examination, a grade III systolic murmur was noted at the apex, and a coarse breathing sound was detected within the whole lung field. ECG showed sinus tachycardia without any evidence of myocardial infarction (Figure 1B). Creatine kinase and its MB isoenzyme were not elevated, but high-sensitivity troponin T was 0.035 ng/mL (normal, 0–0.014 ng/mL). Transthoracic echocardiography revealed left ventricular hypertrophy without regional wall motion abnormalities. The left ventricular ejection fraction was >70%, and E/E′ was 34, suggesting …

Background— The predictive value of coronary computed tomographic angiography (cCTA) in subjects without chest pain syndrome (CPS) has not been established. We investigated the prognostic value of coronary artery disease detection by cCTA and determined the incremental risk stratification benefit of cCTA findings compared with clinical risk factor scoring and coronary artery calcium scoring (CACS) for individuals without CPS. Methods and Results— An open-label, 12-center, 6-country observational registry of 27 125 consecutive patients undergoing cCTA and CACS was queried, and 7590 individuals without CPS or history of coronary artery disease met the inclusion criteria. All-cause mortality and the composite of all-cause mortality and nonfatal myocardial infarction were measured. During a median follow-up of 24 months (interquartile range, 18–35 months), all-cause mortality occurred in 136 individuals. After risk adjustment, compared with individuals without evidence of coronary artery disease by cCTA, individuals with obstructive 2- and 3-vessel disease or left main coronary artery disease experienced higher rates of death and composite outcome ( P P Conclusions— Although the prognosis for individuals without CPS is stratified by cCTA, the additional risk-predictive advantage by cCTA is not clinically meaningful compared with a risk model based on CACS. Therefore, at present, the application of cCTA for risk assessment of individuals without CPS should not be justified.

BACKGROUND The prevalence of coronary artery disease (CAD) increased in patients with moderate to severe renal insufficiency. The impact of renal insufficiency on the development of CAD may increase as renal function continues to deteriorate, however, there is no data available in mild renal insufficiency. The purpose of this study was to explore the association between mild renal insufficiency and coronary atherosclerosis, using coronary CT angiography (CCTA). METHODS We consecutively enrolled 4984 asymptomatic subjects (mean age 49 ± 10 years, 61% male), who underwent CCTA (64-slice MDCT) as part of a general health evaluation. We categorized subjects based on their estimate of renal function (estimated glomerular filtration rate (eGFR), using MDRD equation: ≥ 60 vs. 45 to 59 ml/min/1.73 m 2 ). Patients with eGFR less than 45 ml/min/1.73 m 2 were excluded. Coronary artery calcium score (CACS), presence of plaque, and significant stenosis (≥ 50% lumen diameter) were compared between 2 groups. Regression analysis was performed to assess the correlation between CAD and mild renal insufficiency. RESULTS Three hundreds forty-two subjects (7%) had mild renal insufficiency (eGFR: 45 to 59) and 4642 (93%) had normal renal function (eGFR: ≥ 60 ml/min/1.73 m 2 ). Subjects with mild renal insufficiency showed higher incidence of CAD (124 (36%) vs. 1039 (22%), p CONCLUSIONS The higher incidence of CAD and significant stenosis in subjects with mild renal insufficiency was observed, however, which are primarily attributed to the effect of other conventional risk factors. Mild renal insufficiency may not act as an independent risk factor for coronary atherosclerosis.

Introduction: Full volumetric cardiac imaging over a single heart beat using a 256 row multidetector computed tomography scanning system provides distinct advantages over the current generation of MDCT scanners. 256x0.5mm MDCT allows for a protocol that combines MDCT angiography (MDCTA) with rest and stress CT perfusion (CTP) imaging. The purpose of this study was to test the accuracy of combined 256x0.5mm MDCTA and CTP in patients with suspected coronary artery disease. Methods: Nineteen patients with a positive single photon emission computed tomography (SPECT) perfusion study were imaged by 256 row MDCT with 0.5mm slice thickness (12.8 cm coverage) during one heart beat in a non-helical mode. Stress CTP (120kV, 100mAs, 3 gantry rotations at 0.5 secs, 7.7mSv dose) during adenosine infusion (140μg/kg/min) was followed by rest CTP and CTA (120kV, 175mAs, 3 gantry rotations at 0.5 secs, 13.5mSv dose). Using a territory based analysis; rest and stress CTP images were visually assessed for subendocardial perfusion deficits, CTA was evaluated for stenoses ≥50%, and SPECT images were evaluated for reversible and fixed perfusion deficits. In a second analysis the gold standard for ischemia secondary to epicardial atherosclerosis was defined as a SPECT perfusion deficit in a territory supplied by a coronary artery with a stenosis ≥50%. Results: Compared to ≥50% stenosis by CTA, the sensitivity and specificity of CTP was 85% and 77% compared to 69% and 74% by SPECT. More importantly, when considering the combination of ≥50% stenosis by CTA and a corresponding perfusion deficit on SPECT as the gold standard for ischemic CAD, the sensitivity and specificity for combined CTA and CTP imaging was 78% and 90%. The area under the ROC curve for the combination of CTA and CTP was 0.84, p Conclusions: In patients with suspected coronary artery disease, 256x0.5mm MDCT enables the combination of non-invasive CT angiography and CT perfusion imaging with high accuracy and at radiation levels similar to currently used dual isotope nuclear imaging techniques.

Introduction: The advent of 256x0.5 mm high resolution multidetector computed tomography (MDCT) created the potential to quantify subendocardial myocardial blood flow (MBF) reduction, the hallmark of ischemia. The purpose of this study was to test the hypothesis that adenosine stress / rest 256x0.5mm MDCT can detect ischemia by measuring differences in subendo-cardial versus subepicardial MBF. Methods: Nineteen patients with abnormal SPECT perfusion studies underwent adenosine (140μg/kg/min) stress 256x0.5mm MDCT perfusion imaging (CTP): 120kV, 100mAs, 3 gantry rotations at 0.5 secs, followed by rest 256x0.5mm MDCT angiography (CTA): 120kV, 175mAs, 3 gantry rotations at 0.5 secs. CT perfusion images were reconstructed in the short axis with a 3 mm slice thickness. Endocardial and epicardial borders were defined, myocardium was sectored according to the standard 17 segment model, and divided circumferentially into endocardial and epicardial layers. The transmural perfusion ratio (TPR) was calculated by dividing the endocardial attenuation density (AD) by the epicardial AD in each sector. Ischemia was defined as TPR Results: Mean TPR in abnormal and normal sectors was 0.71±0.05 and 1.01±0.06, respectively (p Conclusions: The transmural perfusion ratio measured from 256x0.5mm MDCT perfusion images can detect the presence of obstructive atherosclerosis with similar accuracy compared with SPECT imaging. The combination of 256x0.5mm MDCT angiography and perfusion imaging may avoid the high false positive rate seen with SPECT imaging alone as well as improve its prediction of territorial ischemia.

Background— Guidelines for the management of patients with suspected coronary artery disease (CAD) rely on the age, sex, and angina typicality–based pretest probabilities of angiographically significant CAD derived from invasive coronary angiography (guideline probabilities). Reliability of guideline probabilities has not been investigated in patients referred to noninvasive CAD testing. Methods and Results— We identified 14048 consecutive patients with suspected CAD who underwent coronary computed tomographic angiography. Angina typicality was recorded with the use of accepted criteria. Pretest likelihoods of CAD with ≥50 diameter stenosis (CAD50) and ≥70 diameter stenosis (CAD70) were calculated from guideline probabilities. Computed tomographic angiography images were evaluated by ≥1 expert reader to determine the presence of CAD50 and CAD70. Typical angina was associated with the highest prevalence of CAD50 (40 in men, 19 in women) and CAD70 (27 men, 11 women) compared with other symptom categories ( P P Conclusion— In this multinational study of patients referred for coronary computed tomographic angiography, determination of pretest likelihood of angiographically significant CAD by the invasive angiography-based guideline probabilities greatly overestimates the actual prevalence of disease.