Scoring model based on cardiac CT and clinical factors to predict early good mitral valve repair in rheumatic mitral disease

OBJECTIVE

We aimed to evaluate the mitral valve calcification and mitral structure detected by cardiac computed tomography (cardiac CT) and establish a scoring model based on cardiac CT and clinical factors to predict early good mitral valve repair (EGMR) and guide surgical strategy in rheumatic mitral disease (RMD).

MATERIALS AND METHODS

This is a retrospective bi-center cohort study. Based on cardiac CT, mitral valve calcification and mitral structure in RMD were quantified and evaluated. The primary outcome was EGMR. A logical regression algorithm was applied to the scoring model.

RESULTS

A total of 579 patients were enrolled in our study from January 1, 2019, to August 31, 2022. Of these, 443 had baseline cardiac CT scans of adequate quality. The calcification quality score, calcification and thinnest part of the anterior leaflet clean zone, and papillary muscle symmetry were the independent CT factors of EGMR. Coronary artery disease and pulmonary artery pressure were the independent clinical factors of EGMR. Based on the above six factors, a scoring model was established. Sensitivity = 95% and specificity = 95% were presented with a cutoff value of 0.85 and 0.30 respectively. The area under the receiver operating characteristic of external validation set was 0.84 (95% confidence interval [CI] 0.73-0.93).

CONCLUSIONS

Mitral valve repair is recommended when the scoring model value > 0.85 and mitral valve replacement is prior when the scoring model value < 0.30. This model could assist in guiding surgical strategies for RMD.

CLINICAL RELEVANCE STATEMENT

The model established in this study can serve as a reference indicator for surgical repair in rheumatic mitral valve disease.

KEY POINTS

• Cardiac CT can reflect the mitral structure in detail, especially for valve calcification. • A model based on cardiac CT and clinical factors for predicting early good mitral valve repair was established. • The developed model can help cardiac surgeons formulate appropriate surgical strategies.

Prognostic Value of Left Ventricular Longitudinal Function and Myocardial Fibrosis in Patients With Ischemic and Non-Ischemic Dilated Cardiomyopathy Concomitant With Type 2 Diabetes Mellitus: A 3.0 T Cardiac MR Study

BACKGROUND

Poorly controlled type 2 diabetes mellitus (T2DM) is known to result in left ventricular (LV) dysfunction, myocardial fibrosis, and ischemic/nonischemic dilated cardiomyopathy (ICM/NIDCM). However, less is known about the prognostic value of T2DM on LV longitudinal function and late gadolinium enhancement (LGE) assessed with cardiac MRI in ICM/NIDCM patients.

PURPOSE

To measure LV longitudinal function and myocardial scar in ICM/NIDCM patients with T2DM and to determine their prognostic values.

STUDY TYPE

Retrospective cohort.

POPULATION

Two hundred thirty-five ICM/NIDCM patients (158 with T2DM and 77 without T2DM).

FIELD STRENGTH/SEQUENCE

3T; steady-state free precession cine; phase-sensitive inversion recovery segmented gradient echo LGE sequences.

ASSESSMENT

Global peak longitudinal systolic strain rate (GLPSSR) was evaluated to LV longitudinal function with feature tracking. The predictive value of GLPSSR was determined with ROC curve. Glycated hemoglobin (HbA1c) was measured. The primary adverse cardiovascular endpoint was follow up every 3 months.

STATISTICAL TESTS

Mann-Whitney U test or student's t-test; Intra and inter-observer variabilities; Kaplan-Meier method; Cox proportional hazards analysis (threshold = 5%).

RESULTS

ICM/NIDCM patients with T2DM exhibited significantly lower absolute value of GLPSSR (0.39 ± 0.14 vs. 0.49 ± 0.18) and higher proportion of LGE positive (+) despite similar LV ejection fraction, compared to without T2DM. LV GLPSSR was able to predict primary endpoint (AUC 0.73) and optimal cutoff point was 0.4. ICM/NIDCM patients with T2DM (GLPSSR < 0.4) had more markedly impaired survival. Importantly, this group (GLPSSR < 0.4, HbA1c ≥ 7.8%, or LGE (+)) exhibited the worst survival. In multivariate analysis, GLPSSR, HbA1c, and LGE (+) significantly predicted primary adverse cardiovascular endpoint in overall ICM/NIDCM and ICM/NIDCM patients with T2DM.

CONCLUSIONS

T2DM has an additive deleterious effect on LV longitudinal function and myocardial fibrosis in ICM/NIDCM patients. Combining GLPSSR, HbA1c, and LGE could be promising markers in predicting outcomes in ICM/NIDCM patients with T2DM.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: 5.

Deep learning-based prognostic model using non-enhanced cardiac cine MRI for outcome prediction in patients with heart failure

OBJECTIVES

To evaluate the performance of a deep learning-based multi-source model for survival prediction and risk stratification in patients with heart failure.

METHODS

Patients with heart failure with reduced ejection fraction (HFrEF) who underwent cardiac magnetic resonance between January 2015 and April 2020 were retrospectively included in this study. Baseline electronic health record data, including clinical demographic information, laboratory data, and electrocardiographic information, were collected. Short-axis non-contrast cine images of the whole heart were acquired to estimate the cardiac function parameters and the motion features of the left ventricle. Model accuracy was evaluated using the Harrell's concordance index. All patients were followed up for major adverse cardiac events (MACEs), and survival prediction was assessed using Kaplan-Meier curves.

RESULTS

A total of 329 patients were evaluated (age 54 ± 14 years; men, 254) in this study. During a median follow-up period of 1041 days, 62 patients experienced MACEs and their median survival time was 495 days. When compared with conventional Cox hazard prediction models, deep learning models showed better survival prediction performance. Multi-data denoising autoencoder (DAE) model reached the concordance index of 0.8546 (95% CI: 0.7902-0.8883). Furthermore, when divided into phenogroups, the multi-data DAE model could significantly discriminate between the survival outcomes of the high-risk and low-risk groups compared with other models (p < 0.001).

CONCLUSIONS

The proposed deep learning (DL) model based on non-contrast cardiac cine magnetic resonance imaging could independently predict the outcome of patients with HFrEF and showed better prediction efficiency than conventional methods.

CLINICAL RELEVANCE STATEMENT

The proposed multi-source deep learning model based on cardiac magnetic resonance enables survival prediction in patients with heart failure.

KEY POINTS

• A multi-source deep learning model based on non-contrast cardiovascular magnetic resonance (CMR) cine images was built to make robust survival prediction in patients with heart failure. • The ground truth definition contains electronic health record data as well as DL-based motion data, and cardiac motion information is extracted by optical flow method from non-contrast CMR cine images. • The DL-based model exhibits better prognostic value and stratification performance when compared with conventional prediction models and could aid in the risk stratification in patients with HF.