The Prognostic Value of a Validated and Automated Intravascular Ultrasound-Derived Calcium Score

Artificial Intelligence for Cardiovascular Care-Part 1: Advances: JACC Review Topic of the Week

Recent artificial intelligence (AI) advancements in cardiovascular care offer potential enhancements in diagnosis, treatment, and outcomes. Innovations to date focus on automating measurements, enhancing image quality, and detecting diseases using novel methods. Applications span wearables, electrocardiograms, echocardiography, angiography, genetics, and more. AI models detect diseases from electrocardiograms at accuracy not previously achieved by technology or human experts, including reduced ejection fraction, valvular heart disease, and other cardiomyopathies. However, AI's unique characteristics necessitate rigorous validation by addressing training methods, real-world efficacy, equity concerns, and long-term reliability. Despite an exponentially growing number of studies in cardiovascular AI, trials showing improvement in outcomes remain lacking. A number are currently underway. Embracing this rapidly evolving technology while setting a high evaluation benchmark will be crucial for cardiology to leverage AI to enhance patient care and the provider experience.

Impact of calcium on the procedural techniques and outcomes of chronic total occlusion percutaneous coronary intervention

BACKGROUND

Coronary calcification is common and increases the difficulty of chronic total occlusion (CTO) percutaneous coronary intervention (PCI).

METHODS

We examined the impact of calcium on procedural outcomes of 13,079 CTO PCIs performed in 12,799 patients at 46 US and non-US centers between 2012 and 2023.

RESULTS

Moderate or severe calcification was present in 46.6% of CTO lesions. Patients whose lesions were calcified were older and more likely to have had prior coronary artery bypass graft surgery. Calcified lesions were more complex with higher J-CTO score (3.0 ± 1.1 vs. 1.9 ± 1.2; p < 0.001) and lower technical (83.0% vs. 89.9%; p < 0.001) and procedural (81.0% vs. 89.1%; p < 0.001) success rates compared with mildly calcified or non-calcified CTO lesions. The retrograde approach was more commonly used among cases with moderate/severe calcification (40.3% vs. 23.5%; p < 0.001). Balloon angioplasty (76.6%) was the most common lesion preparation technique for calcified lesions, followed by rotational atherectomy (7.3%), laser atherectomy (3.4%) and, intravascular lithotripsy (3.4%). The incidence of major adverse cardiovascular events (MACE) was higher in cases with moderate or severe calcification (3.0% vs. 1.2%; p < 0.001), as was the incidence of perforation (6.5% vs. 3.4%; p < 0.001). On multivariable analysis, the presence of moderate/severe calcification was independently associated with lower technical success (odds ratio, OR = 0.73, 95% CI: 0.63-0.84) and higher MACE (OR = 2.33, 95% CI: 1.66-3.27).

CONCLUSIONS

Moderate/severe calcification was present in nearly half of CTO lesions, and was associated with higher utilization of the retrograde approach, lower technical and procedural success rates, and higher incidence of in-hospital MACE.

Optical coherence tomography-derived predictors of stent expansion in calcified lesions

BACKGROUND

Severe coronary artery calcification is associated with stent underexpansion and subsequent stent failure.

AIMS

We aimed to identify optical coherence tomography (OCT)-derived predictors of absolute (minimal stent area [MSA]) and relative stent expansion in calcified lesions.

METHODS

This retrospective cohort study included patients who underwent percutaneous coronary intervention (PCI) with OCT assessment before and after stent implantation between May 2008 and April 2022. Pre-PCI OCT was used to assess calcium burden and post-PCI OCT was used to assess absolute and relative stent expansion.

RESULTS

A total of 361 lesions in 336 patients were analyzed. Target lesion calcification (defined as OCT-detected maximum calcium angle ≥ 30°) was present in 242 (67.0%) lesions. Following PCI, median MSA was 5.37 mm² in calcified lesions and 6.24 mm² in noncalcified lesions (p < 0.001). Median stent expansion was 78% in calcified lesions and 83% in noncalcified lesions (p = 0.325). In the subset of calcified lesions, average stent diameter, preprocedural minimal lumen area, and total calcium length were independent predictors of MSA in multivariable analysis (mean difference 2.69 mm² /mm² , 0.52 mm² /mm, and -0.28 mm² /5 mm, respectively, all p < 0.001). Total stent length was the only independent predictor of relative stent expansion (mean difference -0.465% per mm, p < 0.001). Calcium angle, thickness, and the presence of nodular calcification were not significantly associated with MSA or stent expansion in multivariable analyses.

CONCLUSION

Calcium length appeared to be the most important OCT-derived predictor of MSA, whereas stent expansion was mainly determined by total stent length.

Contribution of the Optical Coherence Tomography in Calcified Lesions

Coronary artery calcification is a complex process found predominantly in the elderly population. Coronary angiography frequently lacks sensitivity to detect, evaluate and quantify these lesions. Yet calcified lesions are considered stable, it remains associated with a higher rate of peri procedural complications during percutaneous coronary intervention (PCI) including an increased risk of stent under expansion and struts mal apposition leading to poor clinical outcome. Intracoronary imaging (Intravascular Ultra Sound (IVUS) and Optical Coherence Tomography (OCT)) allows better calcified lesions identification, localization within the coronary artery wall (superficial or deep calcifications), quantification. This lesions characterization allows a better choice of dedicated plaque-preparation tools (modified balloons, rotational or orbital atherectomy, intravascular lithotripsy) that are crucial to achieve optimal PCI results. OCT could also assess the impact of these tools on the calcified plaque morphology (plaque fracture, burring effects…). An OCT-guided tailored PCI strategy for calcified lesions still requires validation by clinical studies which are currently underway.

Advances in Diagnosis, Therapy, and Prognosis of Coronary Artery Disease Powered by Deep Learning Algorithms

Percutaneous coronary intervention has been a standard treatment strategy for patients with coronary artery disease with continuous ebullient progress in technology and techniques. The application of artificial intelligence and deep learning in particular is currently boosting the development of interventional solutions, improving the efficiency and objectivity of diagnosis and treatment. The ever-growing amount of data and computing power together with cutting-edge algorithms pave the way for the integration of deep learning into clinical practice, which has revolutionized the interventional workflow in imaging processing, interpretation, and navigation. This review discusses the development of deep learning algorithms and their corresponding evaluation metrics together with their clinical applications. Advanced deep learning algorithms create new opportunities for precise diagnosis and tailored treatment with a high degree of automation, reduced radiation, and enhanced risk stratification. Generalization, interpretability, and regulatory issues are remaining challenges that need to be addressed through joint efforts from multidisciplinary community.

Automated analysis of fibrous cap in intravascular optical coherence tomography images of coronary arteries

Thin-cap fibroatheroma (TCFA) and plaque rupture have been recognized as the most frequent risk factor for thrombosis and acute coronary syndrome. Intravascular optical coherence tomography (IVOCT) can identify TCFA and assess cap thickness, which provides an opportunity to assess plaque vulnerability. We developed an automated method that can detect lipidous plaque and assess fibrous cap thickness in IVOCT images. This study analyzed a total of 4360 IVOCT image frames of 77 lesions among 41 patients. Expert cardiologists manually labeled lipidous plaque based on established criteria. To improve segmentation performance, preprocessing included lumen segmentation, pixel-shifting, and noise filtering on the raw polar (r, θ) IVOCT images. We used the DeepLab-v3 plus deep learning model to classify lipidous plaque pixels. After lipid detection, we automatically detected the outer border of the fibrous cap using a special dynamic programming algorithm and assessed the cap thickness. Our method provided excellent discriminability of lipid plaque with a sensitivity of 85.8% and A-line Dice coefficient of 0.837. By comparing lipid angle measurements between two analysts following editing of our automated software, we found good agreement by Bland-Altman analysis (difference 6.7° ± 17°; mean ~ 196°). Our method accurately detected the fibrous cap from the detected lipid plaque. Automated analysis required a significant modification for only 5.5% frames. Furthermore, our method showed a good agreement of fibrous cap thickness between two analysts with Bland-Altman analysis (4.2 ± 14.6 µm; mean ~ 175 µm), indicating little bias between users and good reproducibility of the measurement. We developed a fully automated method for fibrous cap quantification in IVOCT images, resulting in good agreement with determinations by analysts. The method has great potential to enable highly automated, repeatable, and comprehensive evaluations of TCFAs.

[Invasive and non-invasive imaging analysis for calcified coronary artery lesions]

Coronary calcifications are frequently identified within coronary lesions as their incidence increases with age and cardiovascular risk factors. Their location can be superficial or deep, according to different pathological process. In all cases, the presence of calcifications within the vascular wall predicts poor clinical prognosis and unfavorable evolution after percutaneous revascularization. Coronary calcifications can be analyzed by angiography, CT or intracoronary imaging (IVUS or OCT) with variable accuracies. Angiography is the most frequently used method but is not very sensitive (sensitivity close to 50%) and insufficient for their precise quantification. The CT scan is a more effective non-invasive method leading to an accurate analysis of the lesion before coronary angiography. IVUS and OCT have an excellent spatial resolution and are the most sensitive methods for the identification (present in nearly 75-80% of lesions) and quantification of calcifications. These intracoronary imaging techniques offer interesting perspectives for identification of the highest-risk lesions, PCI procedures planning (including the choice of an optimal dedicated plaque preparation devices), the monitoring of their execution and the evaluation of the immediate post-stenting results.