Artificial intelligence (AI) has transformed key aspects of human life. Machine learning (ML), which is a subset of AI wherein machines autonomously acquire information by extracting patterns from ...large databases, has been increasingly used within the medical community, and specifically within the domain of cardiovascular diseases. In this review, we present a brief overview of ML methodologies that are used for the construction of inferential and predictive data-driven models. We highlight several domains of ML application such as echocardiography, electrocardiography, and recently developed non-invasive imaging modalities such as coronary artery calcium scoring and coronary computed tomography angiography. We conclude by reviewing the limitations associated with contemporary application of ML algorithms within the cardiovascular disease field.
Hypertension is associated with more severe disease and adverse outcomes in COVID-19 patients. Recent investigations have indicated that hypertension might be an independent predictor of outcomes in ...COVID-19 patients regardless of other cardiovascular and noncardiovascular comorbidities. We explored the significance of coronary calcifications in 694 hypertensive patients in the Score-COVID registry, an Italian multicenter study conducted during the first pandemic wave in the Western world (March-April 2020). A total of 1565 patients admitted with RNA-PCR-positive nasopharyngeal swabs and chest computed tomography (CT) at hospital admission were included in the study. Clinical outcomes and cardiovascular calcifications were analyzed independently by a research core lab. Hypertensive patients had a different risk profile than nonhypertensive patients, with more cardiovascular comorbidities. The deceased hypertensive patients had a greater coronary calcification burden at the level of the anterior descending coronary artery. Hypertension status and the severity cutoffs of coronary calcifications were used to stratify the clinical outcomes. For every 100-mm
increase in coronary calcium volume, hospital mortality in hypertensive patients increased by 8%, regardless of sex, age, diabetes, creatinine, and lung interstitial involvement. The coronary calcium score contributes to stratifying the risk of complications in COVID-19 patients. Cardiovascular calcifications appear to be a promising imaging marker for providing pathophysiological insight into cardiovascular risk factors and COVID-19 outcomes.
Physiological assessment of coronary artery disease (CAD) has become one of the cornerstones of decision making for myocardial revascularization, with a large body of evidence supporting the benefits ...of using fractional flow reserve and other pressure-based indexes for functional assessment of coronary stenoses. Furthermore, physiology allows the identification of specific vascular dysfunction mechanisms in patients without obstructive CAD. Currently, more than 10 modalities of functional coronary assessment are available, although the overall adoption of these physiological tools, of either intracoronary or image-based nature, is still low. In this paper the authors review these modalities of functional coronary assessment according to their timing of use: outside the catheterization laboratory, in the catheterization laboratory prior to the percutaneous coronary intervention (PCI), and in the catheterization laboratory during or after PCI. The authors discuss how the information obtained can be used in setting the indication for PCI, in planning and guiding the procedure, and in documenting the final functional result of the intervention. The advantages and limitations of each modality in each setting are discussed. Furthermore, the key value of intracoronary physiology in diagnosing mechanisms of microcirculatory dysfunction, which account for the presence of ischemia in many patients without obstructive CAD, is revisited. On the basis of the opportunities generated by the multiplicity of diagnostic tools described, the authors propose an algorithmic approach to physiological coronary investigations in clinical practice, with the key aims of: 1) avoiding unneeded revascularization procedures; 2) improving procedural PCI and long-term outcomes in patients with obstructive CAD; and 3) diagnosing vascular dysfunction mechanisms that can be effectively treated in patients with NOCAD. The authors believe that such structured approach may also contribute to the wider adoption of available technologies for functional assessment of patients with CAD.
Display omitted
•Since the introduction of FFR, 10 more novel modalities for the assessment of coronary physiology have emerged.•Physiological modalities should be appropriately selected according to their timing of use: outside the catheterization laboratory prior to the treatment decision and in the catheterization laboratory before and after PCI.•The next challenge is to integrate the evaluation of microvascular circulation as part of daily practice.•In the near future, invasive and noninvasive modalities of assessing coronary physiology may be integrated to stratify patients with history of anginal symptoms.
Patients requiring diagnostic testing for coronavirus disease 2019 (COVID-19) are routinely assessed by reverse-transcription quantitative polymerase chain reaction (RT-qPCR) amplification of ...Sars-CoV-2 virus RNA extracted from oro/nasopharyngeal swabs. Despite the good specificity of the assays certified for SARS-CoV-2 molecular detection, and a theoretical sensitivity of few viral gene copies per reaction, a relatively high rate of false negatives continues to be reported. This is an important challenge in the management of patients on hospital admission and for correct monitoring of the infectivity after the acute phase. In the present report, we show that the use of digital PCR, a high sensitivity method to detect low amplicon numbers, allowed us to correctly detecting infection in swab material in a significant number of false negatives. We show that the implementation of digital PCR methods in the diagnostic assessment of COVID-19 could resolve, at least in part, this timely issue.
Objectives The aim of this study was to assess the long-term prognostic role of multidetector computed tomography coronary angiography (CTA) in patients with suspected coronary artery disease (CAD). ...Background Use of CTA is increasing in patients with suspected CAD. Although there is a large body of data supporting the prognostic role of CTA for major adverse cardiac events in the intermediate term, its long-term prognostic role in patients with suspected CAD is not well studied. Methods Between February 2005 and March 2008, 1,304 consecutive patients were prospectively studied with CTA for detecting the presence and assessing extent of CAD (disease extension and coronary plaque scores). Patients were classified according to the presence of normal coronaries and nonobstructive (<50%) and obstructive (>50%) coronary lesions. The composite rates of hard cardiac events (cardiac deaths and nonfatal myocardial infarctions) and all cardiac events (including late revascularization) were the endpoints of the study. Results Seventy patients were excluded because their CTA data were uninterpretable. Of the remaining 1,234 patients, clinical follow-up (mean 52 ± 22 months) was obtained for 1,196 (97%). A total of 475 events were recorded, with 136 hard events (18 cardiac deaths and 118 nonfatal myocardial infarctions) and 123 late revascularizations. A total of 216 patients with early elective revascularizations were excluded from the survival analysis. Significant independent predictors of events in multivariate analysis were multivessel disease and left main CAD. Cumulative event-free survival was 100% for hard and all events in patients with normal coronary arteries, 88% for hard events and 72% for all events in patients with nonobstructive CAD, and 54% for hard events and 31% for all events in patients with obstructive CAD. Multivessel CAD was associated with a higher rate of hard cardiac events. Conclusions CTA provides prognostic information in patients with suspected CAD and unknown cardiac disease, showing excellent long-term prognosis when there is no evidence of atherosclerosis and allowing risk stratification when CAD is present.
Coronary artery disease (CAD) is a major challenge in patients with type 2 diabetes (T2D). Coronary computed tomography angiography (CCTA) provides a detailed anatomic map of the coronary ...circulation. Proteomics are increasingly used to improve diagnostic and therapeutic algorithms. We hypothesized that the protein panel is differentially associated with T2D and CAD.
In CAPIRE (Coronary Atherosclerosis in Outlier Subjects: Protective and Novel Individual Risk Factors Evaluation-a cohort of 528 individuals with no previous cardiovascular event undergoing CCTA), participants were grouped into CAD
(clean coronaries) and CAD
(diffuse lumen narrowing or plaques). Plasma proteins were screened by aptamer analysis. Two-way partial least squares was used to simultaneously rank proteins by diabetes status and CAD.
Though CAD
was more prevalent among participants with T2D (HbA
6.7 ± 1.1%) than those without diabetes (56 vs. 30%,
< 0.0001), CCTA-based atherosclerosis burden did not differ. Of the 20 top-ranking proteins, 15 were associated with both T2D and CAD, and 3 (osteomodulin, cartilage intermediate-layer protein 15, and HTRA1) were selectively associated with T2D only and 2 (epidermal growth factor receptor and contactin-1) with CAD only. Elevated renin and GDF15, and lower adiponectin, were independently associated with both T2D and CAD. In multivariate analysis adjusting for the Framingham risk panel, patients with T2D were "protected" from CAD if female (
= 0.007), younger (
= 0.021), and with lower renin levels (
= 0.02).
We concluded that
) CAD severity and quality do not differ between participants with T2D and without diabetes;
) renin, GDF15, and adiponectin are shared markers by T2D and CAD;
) several proteins are specifically associated with T2D or CAD; and
) in T2D, lower renin levels may protect against CAD.
To determine diagnostic performance of non-invasive tests using invasive fractional flow reserve (FFR) as reference standard for coronary artery disease (CAD).
Medline, Embase, and citations of ...articles, guidelines, and reviews for studies were used to compare non-invasive tests with invasive FFR for suspected CAD published through March 2017.
Seventy-seven studies met inclusion criteria. The diagnostic test with the highest sensitivity to detect a functionally significant coronary lesion was coronary computed tomography (CT) angiography 88%(85%–90%), followed by FFR derived from coronary CT angiography (FFRCT) 85%(81%–88%), positron emission tomography (PET) 85%(82%–88%), stress cardiac magnetic resonance (stress CMR) 81%(79%–84%), stress myocardial CT perfusion combined with coronary CT angiography 79%(74%–83%), stress myocardial CT perfusion 77%(73%–80%), stress echocardiography (Echo) 72%(64%–78%) and stress single-photon emission computed tomography (SPECT) 64%(60%–68%). Specificity to rule out CAD was highest for stress myocardial CT perfusion added to coronary CT angiography 91%(88%–93%), stress CMR 91%(90%–93%), and PET 87%(86%–89%).
A negative coronary CT angiography has a higher test performance than other index tests to exclude clinically-important CAD. A positive stress myocardial CT perfusion added to coronary CT angiography, stress cardiac MR, and PET have a higher test performance to identify patients requiring invasive coronary artery evaluation.
•CTCA is an excellent tool to rule out obstructive CAD.•FFRCT or stress CTP improve the post-test likelihood of significant CAD.•PET, CMR, and CTCA combined with CTP showed the highest positive likelihood ratio.
Background Coronary artery fractional flow reserve (FFR) derived from CT angiography (FFT
) enables functional assessment of coronary stenosis. Prior clinical trials showed 13%-33% of coronary CT ...angiography studies had insufficient quality for quantitative analysis with FFR
Purpose To determine the rejection rate of FFR
analysis and to determine factors associated with technically unsuccessful calculation of FFR
Materials and Methods Prospectively acquired coronary CT angiography scans submitted as part of the Assessing Diagnostic Value of Noninvasive FFR
in Coronary Care (ADVANCE) registry (
: NCT02499679) and coronary CT angiography series submitted for clinical analysis were included. The primary outcome was the FFR
rejection rate (defined as an inability to perform quantitative analysis with FFR
). Factors that were associated with FFR
rejection rate were assessed with multiple linear regression. Results In the ADVANCE registry, FFR
rejection rate due to inadequate image quality was 2.9% (80 of 2778 patients; 95% confidence interval CI: 2.1%, 3.2%). In the 10 621 consecutive patients who underwent clinical analysis, the FFR
rejection rate was 8.4% (
= 892; 95% CI: 6.2%, 7.2%;
< .001 vs the ADVANCE cohort). The main reason for the inability to perform FFR
analysis was the presence of motion artifacts (63 of 80 78% and 729 of 892 64% in the ADVANCE and clinical cohorts, respectively). At multivariable analysis, section thickness in the ADVANCE (odds ratio OR, 1.04; 95% CI: 1.001, 1.09;
= .045) and clinical (OR, 1.03; 95% CI: 1.02, 1.04;
< .001) cohorts and heart rate in the ADVANCE (OR, 1.05; 95% CI: 1.02, 1.08;
< .001) and clinical (OR, 1.06; 95% CI: 1.05, 1.07;
< .001) cohorts were independent predictors of rejection. Conclusion The rates for technically unsuccessful CT-derived fractional flow reserve in the ADVANCE registry and in a large clinical cohort were 2.9% and 8.4%, respectively. Thinner CT section thickness and lower patient heart rate may increase rates of completion of CT fractional flow reserve analysis. Published under a CC BY 4.0 license.
See also the editorial by Sakuma in this issue.
Objectives
T1 mapping (T1-map) and cardiac magnetic resonance feature tracking (CMR-FT) techniques have been introduced for the early detection of interstitial myocardial fibrosis and deformation ...abnormalities. We sought to demonstrate that T1-map and CMR-FT may identify the presence of subclinical myocardial structural changes in patients with mitral valve prolapse (MVP).
Methods
Consecutive MVP patients with moderate-to-severe mitral regurgitation and comparative matched healthy subjects were prospectively enrolled and underwent CMR-FT analysis to calculate 2D global and segmental circumferential (CS) and radial strain (RS) and T1-map to determine global and segmental native T1 (nT1) values.
Results
Seventy-three MVP patients (mean age, 57 ± 13 years old; male, 76%; regurgitant volume, 57 ± 21 mL) and 42 matched control subjects (mean age, 56 ± 18 years; male, 74%) were included. MVP patients showed a lower global CS (− 16.3 ± 3.4% vs. − 17.8 ± 1.9%,
p
= 0.020) and longer global nT1 (1124.9 ± 97.7 ms vs. 1007.4 ± 26.1 ms,
p
< 0.001) as compared to controls. Moreover, MVP patients showed lower RS and CS in basal (21.6 ± 12.3% vs. 27.6 ± 8.9%,
p
= 0.008, and − 13.0 ± 6.7% vs. − 14.9 ± 4.1%,
p
= 0.013) and mid-inferolateral (20.6 ± 10.7% vs. 28.4 ± 8.7%,
p
< 0.001, and − 12.8 ± 6.3% vs. − 16.5 ± 4.0%,
p
< 0.001) walls as compared to other myocardial segments. Similarly, MVP patients showed longer nT1 values in basal (1080 ± 68 ms vs. 1043 ± 43 ms,
p
< 0.001) and mid-inferolateral (1080 ± 77 ms vs. 1034 ± 37 ms,
p
< 0.001) walls as compared to other myocardial segments. Of note, nT1 values were significantly correlated with CS (
r
, 0.36;
p
< 0.001) and RS (
r
, 0.37;
p
< 0.001) but not with regurgitant volume.
Conclusions
T1-map and CMR-FT identify subclinical left ventricle tissue changes in patients with MVP. Further studies are required to correlate these subclinical tissue changes with the outcome.
Key Points
•
T1 mapping (T1-map) and cardiac magnetic resonance feature tracking (CMR-FT) techniques have been introduced for the early detection of interstitial myocardial fibrosis and deformation abnormalities.
•
In MVP patients, we demonstrated a longer global nT1 with associated reduced global circumferential (CS) and radial strain (RS) as compared to control subjects.
•
Among MVP patients, the mid-basal left ventricle inferolateral wall showed longer nT1 with reduced CS and RS as compared to other myocardial segments. Further studies are required to correlate these subclinical tissue changes with the outcome.