Development of heart failure risk prediction models based on a multi-marker approach using random forest algorithms / 中华医学杂志(英文版)

BACKGROUND@#The early identification of heart failure (HF) risk may favorably affect outcomes, and the combination of multiple biomarkers may provide a more comprehensive and valuable means for improving the risk of stratification. This study was conducted to assess the importance of individual cardiac biomarkers creatine kinase MB isoenzyme (CK-MB), B-type natriuretic peptide (BNP), galectin-3 (Gal-3) and soluble suppression of tumorigenicity-2 (sST2) for HF diagnosis, and the predictive performance of the combination of these four biomarkers was analyzed using random forest algorithms.@*METHODS@#A total of 193 participants (80 patients with HF and 113 age- and gender-matched healthy controls) were included from June 2017 to December 2017. The correlation and regression analysis were conducted between cardiac biomarkers and echocardiographic parameters. The accuracy and importance of these predictor variables were assessed using random forest algorithms.@*RESULTS@#Patients with HF exhibited significantly higher levels of CK-MB, BNP, Gal-3, and sST2. BNP exhibited a good independent predictive capacity for HF (AUC 0.956). However, CK-MB, sST2, and Gal-3 exhibited a modest diagnostic performance for HF, with an AUC of 0.709, 0.711, and 0.777, respectively. BNP was the most important variable, with a remarkably higher mean decrease accuracy and Gini. Furthermore, there was a general increase in predictive performance using the multi-marker model, and the sensitivity, specificity was 91.5% and 96.7%, respectively.@*CONCLUSION@#The random forest algorithm provides a robust method to assess the accuracy and importance of predictor variables. The combination of CK-MB, BNP, Gal-3, and sST2 achieves improvement in prediction accuracy for HF.

Usefulness of lumen area parameters determined by intravascular ultrasound to predict functional significance of intermediate coronary artery stenosis / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>Coronary artery disease is the leading cause of death in China. Percutaneous coronary intervention is a recent milestone technology for treatment coronary artery disease. However, clinical decision making for patients with intermediate coronary stenosis is still controversial. We designed this study to assess the optimal intravascular ultrasound (IVUS) criteria for predicting functional significance of intermediate coronary lesions.</p><p><b>METHODS</b>We enrolled 141 patients with 165 intermediate coronary lesions located in vessels with a diameter ≥ 2.50 mm. IVUS of intermediate coronary lesions were performed before intervention. Pressure-derived fractional flow reserve (FFR) was measured at maximal hyperemia induced by adenosine infusion. An FFR < 0.80 was considered as abnormal functional significance.</p><p><b>RESULTS</b>For the overall 165 lesions, the mean FFR value was 0.84 ± 0.09. The diameter of the stenosis by visual estimation on angiogram was (59.63 ± 11.29)%. Minimum lumen diameter (MLD), minimum lumen area (MLA) and plaque burden (PB) were (2.00 ± 0.36) mm, (3.88 ± 1.34) mm(2), (67.28 ± 9.89)% respectively by IVUS measurements. An FFR < 0.80 was seen in 43 lesions (30.5%). There was a moderate correlation between IVUS parameters and FFR, including MLD (r = 0.372, P < 0.001), MLA (r = 0.442, P < 0.001) and PB (r = -0.172, P < 0.05). MLA was a predictor for FFR as a continuous variable independent of possible confounding variables (P < 0.05), and MLA and PB, were predictors for FFR < 0.80 as binary variables (P < 0.05). The best cutoff value of MLA to predict FFR < 0.80 was < 3.15 mm(2), with a 73.6% diagnostic accuracy; sensitivity 71.4%, specificity 67.0%, AUC = 0.709, and P < 0.001. The cutoff value of the PB to predict FFR < 0.80 was 65.45%; sensitivity 82.6%, specificity 41.2%, AUC = 0.644, and P < 0.01. If both MLA and PB were taken into account, the negative predictive value and the positive predictive value were 88.7% and 64.8% respectively.</p><p><b>CONCLUSIONS</b>Anatomic measurements of intermediate coronary lesions obtained by IVUS showed a moderate correlation to FFR values. IVUS-derived MLA ≥ 3.15 mm(2) may be useful to exclude FFR < 0.80, but poor specificity limits its applicability for physiological assessment of lesions < 3.15 mm(2). MLA was one of many factors affecting coronary flow hemodynamics. Both MLA and PB should be taken into account when determining functional ischemia.</p>