Construction and validation of risk prediction models for pulmonary embolism in hospitalized patients based on different machine learning methods.

Objective: This study aims to apply different machine learning (ML) methods to construct risk prediction models for pulmonary embolism (PE) in hospitalized patients, and to evaluate and compare the predictive efficacy and clinical benefit of each model. Methods: We conducted a retrospective study involving 332 participants (172 PE positive cases and 160 PE negative cases) recruited from Guangdong Medical University. Participants were randomly divided into a training group (70%) and a validation group (30%). Baseline data were analyzed using univariate analysis, and potential independent risk factors associated with PE were further identified through univariate and multivariate logistic regression analysis. Six ML models, namely Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Naive Bayes (NB), Support Vector Machine (SVM), and AdaBoost were developed. The predictive efficacy of each model was compared using the receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC). Clinical benefit was assessed using decision curve analysis (DCA). Results: Logistic regression analysis identified lower extremity deep venous thrombosis, elevated D-dimer, shortened activated partial prothrombin time, and increased red blood cell distribution width as potential independent risk factors for PE. Among the six ML models, the RF model achieved the highest AUC of 0.778. Additionally, DCA consistently indicated that the RF model offered the greatest clinical benefit. Conclusion: This study developed six ML models, with the RF model exhibiting the highest predictive efficacy and clinical benefit in the identification and prediction of PE occurrence in hospitalized patients.

The Association between Obstructive Sleep Apnea and Venous Thromboembolism: A Bidirectional Two-Sample Mendelian Randomization Study.

BACKGROUND: Despite previous observational studies linking obstructive sleep apnea (OSA) to venous thromboembolism (VTE), these findings remain controversial. This study aimed to explore the association between OSA and VTE, including pulmonary embolism (PE) and deep vein thrombosis (DVT), at a genetic level using a bidirectional two-sample Mendelian randomization (MR) analysis. METHODS: Utilizing summary-level data from large-scale genome-wide association studies in European individuals, we designed a bidirectional two-sample MR analysis to comprehensively assess the genetic association between OSA and VTE. The inverse variance weighted was used as the primary method for MR analysis. In addition, MR-Egger, weighted median, and MR pleiotropy residual sum and outlier (MR-PRESSO) were used for complementary analyses. Furthermore, a series of sensitivity analyses were performed to ensure the validity and robustness of the results. RESULTS: The initial and validation MR analyses indicated that genetically predicted OSA had no effects on the risk of VTE (including PE and DVT). Likewise, the reverse MR analysis did not find substantial support for a significant association between VTE (including PE and DVT) and OSA. Supplementary MR methods and sensitivity analyses provided additional confirmation of the reliability of the MR results. CONCLUSION: Our bidirectional two-sample MR analysis did not find genetic evidence supporting a significant association between OSA and VTE in either direction.

A Nomogram for Predicting 3-Year Mortality in Patients with Pulmonary Hypertension due to Left Heart Failure: A Retrospective Analysis of a Prospective Registry Study.

INTRODUCTION: Pulmonary hypertension due to left heart failure (PH-LHF) is a disease with high prevalence and 3-year mortality rates. Consequently, timely identification of patients with high mortality risk is critical. This study aimed to build a nomogram for predicting 3-year mortality and screening high-risk PH-LHF patients. METHODS: This nomogram was developed on a training cohort of 175 patients with PH-LHF diagnosed by right heart catheterization. Multivariate Cox regression was used to identify independent predictors and develop this nomogram. The median total points obtained from the nomogram were used as a cutoff point, and patients were classified into low- and high-risk groups. The concordance index (C-index) and calibration curve were utilized to ascertain the predictive accuracy and discriminative ability of the nomogram. External validation was performed using a validation cohort of 77 PH-LHF patients from other centers. RESULTS: Multivariate Cox regression showed that the New York Heart Association Functional classification (NYHA FC), uric acid level, and mean pulmonary arterial pressure were all independent predictors and incorporated into the nomogram. The nomogram showed good discrimination (C-index of 0.756; 95% CI: 0.688-0.854) and good calibration. The Kaplan-Meier survival analysis showed that patients in the high-risk group had worse survival (p < 0.001). In the external validation, the nomogram showed both good discrimination (C-index of 0.738; 95% CI: 0.591-0.846) and calibration. CONCLUSION: The nomogram had a good performance in predicting 3-year mortality and can effectively identify high-risk patients. The nomogram may help to reduce the mortality of PH-LHF.

Prevalence, risk factors, and survival associated with pulmonary hypertension and heart failure among patients with underlying coronary artery disease: a national prospective, multicenter registry study in China.

BACKGROUND: Coronary artery disease (CAD) is the commonest cause of heart failure (HF), whereas pulmonary hypertension (PH) has not been established or reported in this patient population. Therefore, we assessed the prevalence, risk factors, and survival in CAD-associated HF (CAD-HF) complicated with PH. METHODS: Symptomatic CAD-HF patients were continuously enrolled in this prospective, multicenter registry study. Echocardiography, coronary arteriography, left and right heart catheterization (RHC), and other baseline clinical data were recorded. Patients were followed up and their survival was recorded. RESULTS: One hundred and eighty-two CAD-HF patients were enrolled, including 142 with HF with a preserved ejection fraction (heart failure with preserved ejection fraction [HFpEF]; left ventricular ejection fraction [LVEF] ≥50%) and 40 with a reduced ejection fraction (heart failure with reduced ejection fraction [HFrEF]; LVEF < 50%). PH was diagnosed with RHC in 77.5% of patients. Patients with PH showed worse hemodynamic parameters and higher mortality. HFrEF-PH patients had worse survival than HFpEF-PH patients. CAD-HF patients with an enlarged left ventricular end-diastolic diameter and reduced hemoglobin were at higher risk of PH. Nitrate treatment reduced the risk of PH. Elevated creatinine and mean pulmonary arterial pressure (mPAP), diastolic pressure gradient (DPG) ≥7 mmHg, and previous myocardial infarction (MI) entailed a higher risk of mortality in CAD-HF patients with PH. CONCLUSIONS: PH is common in CAD-HF and worsens the hemodynamics and survival in these patients. Left ventricle enlargement and anemia increase the risk of PH in CAD-HF. Patients may benefit from nitrate medications. Renal impairment, elevated mPAP, DPG ≥7 mmHg, and previous MI are strong predictors of mortality in CAD-HF-PH patients. TRIAL REGISTRATION: ClinicalTrials.gov, NCT02164526.

The prevalence and survival of pulmonary hypertension due to left heart failure: A retrospective analysis of a multicenter prospective cohort study.

Background: Pulmonary hypertension due to left heart failure (PH-LHF) is currently the most common form of pulmonary hypertension (PH) encountered in clinical practice. Despite significant advances that have improved our understanding of PH-LHF over the past two decades, the mortality is still high in recent decades. This study aimed to describe the prevalence and survival of patients with PH-LHF, and explored the potential risk factors which may predict the prognosis of PH-LHF. Methods: A retrospective analysis of a prospective cohort study of left heart failure (LHF) patients who underwent right heart catheterization (RHC) between January 2013 and November 2016 was performed. The endpoint was all-cause mortality. Follow-ups were performed every 6 months ± 2 weeks. Results: A total of 480 patients with LHF were enrolled, with 215 (44.8%) having PH-LHF. The proportion of PH-LHF was significantly lower in coronary artery disease (CAD) group than without CAD (41.3 vs. 57.8%, p = 0.003). However, multivariable logistic regression analysis revealed that CAD was not associated with PH-LHF (Adjusted OR: 1.055, 95% CI: 0.576 - 1.935, p = 0.862). 75 of 215 (34.9%) patients with PH-LHF died during a median follow-up period of 84.6 months. The 1-, 3-, 5-, and 8-year survival rates of all PH-LHF patients were 94.3, 76.9, 65.8, and 60.2%, respectively. New York Heart Association Functional Class (NYHA FC), hemoglobin, and systolic pulmonary artery pressure (sPAP) were associated with mortality of PH-LHF in multivariate Cox analysis. Conclusion: PH is commonly identified in patients with LHF, with a prevalence of approximately 45%. The mortality is still high in patients with PH-LHF. NYHA FC, hemoglobin, and sPAP are independent risk predictors of mortality for PH-LHF. These findings may be useful for risk stratification in future clinical trial enrollment.

Impact of borderline pulmonary hypertension due to left heart failure on mortality in a multicenter registry study: A 3-year survivorship analysis.

Background: Patients with left heart failure (LHF) are often associated with the development of pulmonary hypertension (PH) which leads to an increased risk of death. Recently, the diagnostic standard for PH has changed from mean pulmonary arterial pressure (mPAP) ≥25 mmHg to >20 mmHg. Nonetheless, the effect of borderline PH (mPAP: 21-24 mmHg) on the prognosis of LHF patients is unclear. This study aimed to investigate the relationship between borderline PH and 3-year clinical outcomes in LHF patients. Methods: A retrospective analysis of a prospective cohort study was done for LHF patients who underwent right heart catheterization (RHC) between January 2013 and November 2016. The primary outcome was all-cause mortality; the secondary outcome was rehospitalization. Results: Among 344 patients, 62.5% were identified with a proportion of PH (mPAP ≥ 25), 10.8% with borderline PH (21-24), and 26.7% with non-PH (≤20), respectively. Multivariable Cox analysis revealed that borderline PH patients had a higher adjusted mortality risk (HR = 3.822; 95% CI: 1.043-13.999; p = 0.043) than non-PH patients. When mPAP was treated as a continuous variable, the hazard ratio for death increased progressively with increasing mPAP starting at 20 mmHg (HR = 1.006; 95% CI: 1.001-1.012). There was no statistically significant difference in adjusted rehospitalization between borderline PH and non-PH patients (HR = 1.599; 95% CI: 0.833-3.067; p = 0.158). Conclusions: Borderline PH is independently related to increased 3-year mortality in LHF patients. Future research is needed to evaluate whether more close monitoring, and managing with an intensifier improves clinical outcomes in borderline PH caused by LHF. Clinical trials registration: www.clinicaltrials.gov NCT02164526.

Risk prediction in pulmonary hypertension due to chronic heart failure: incremental prognostic value of pulmonary hemodynamics.

BACKGROUND: There is no generally accepted comprehensive risk prediction model cooperating risk factors associated with heart failure and pulmonary hemodynamics for patients with pulmonary hypertension due to left heart disease (PH-LHD). We aimed to explore outcome correlates and evaluate incremental prognostic value of pulmonary hemodynamics for risk prediction in PH-LHD. METHODS: Consecutive patients with chronic heart failure undergoing right heart catheterization were prospectively enrolled. The primary endpoint was all-cause mortality. Individual variable selection was performed by machine learning methods. Cox proportional hazards models were conducted to identify the association between variables and mortality. Incremental value of hemodynamics was evaluated based on the Seattle heart failure model (SHFM) and Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) scores. RESULTS: A total of 276 PH-LHD patients were enrolled, with a median follow-up time of 34.7 months. By L1-penalized regression model and random forest approach, diastolic pressure gradient (DPG) and mixed venous oxygen saturation (SvO2) were the hemodynamic predictors most strongly associated with mortality (coefficient: 0.0255 and -0.0176, respectively), with consistent significance after adjusted for SHFM [DPG: HR 1.067, 95% CI 1.024-1.113, P = 0.022; SvO2: HR 0.969, 95% CI 0.953-0.985, P = 0.002] or MAGGIC (DPG: HR 1.069, 95% CI 1.026-1.114, P = 0.011; SvO2: HR 0.970, 95% CI 0.954-0.986, P = 0.004) scores. The inclusion of DPG and SvO2 improved risk prediction compared with using SHFM [net classification improvement (NRI): 0.468 (0.161-0.752); integrated discriminatory index (IDI): 0.092 (0.035-0.171); likelihood ratio test: P < 0.001] or MAGGIC [NRI: 0.298 (0.106-0.615); IDI: 0.084 (0.033-0.151); likelihood ratio: P < 0.001] scores alone. CONCLUSION: In PH-LHD, pulmonary hemodynamics can provide incremental prognostic value for risk prediction. CLINICAL TRIAL REGISTRATION: NCT02164526 at https://clinicaltrials.gov .

Acute conduction recurrence of mitral isthmus: Incidence, clinical characteristics, and implications.

BACKGROUND: Data on the incidence, clinical characteristics, and implications of acute conduction recurrence during mitral isthmus (MI) ablation are scarce. METHODS: MI ablation was performed in patients with atrial fibrillation. After confirming bidirectional conduction block, the acute conduction recurrence of MI was systematically evaluated. Clinical and electrophysiological characteristics were analyzed. RESULTS: A total of 66 consecutive patients in whom bidirectional conduction block of MI was achieved were prospectively enrolled in a single center. Acute conduction recurrence of MI developed in 12 (18.2%) patients within 14.2 ± 11.5 minutes after the confirmation of bidirectional conduction block. There were two recurrent conduction breakthrough sites of MI along the course of the great cardiac vein (4.5 ± 3.5 min) in two patients and 11 along the course of the ligament of Marshall (LOM) (16.0 ± 11.6 min, P = .035) in 11 patients. LOM accounted for most (84.6%, 11/13) acute MI conduction recurrence. MI length, total ablation time, and procedure time for MI were greater in patients with acute conduction recurrence than in those without acute conduction recurrence. During follow-up, arrhythmia recurrences were less observed in patients with acute conduction when compared to patients without acute conduction recurrence (0% vs 26.4%, P = .055). CONCLUSION: Acute conduction recurrence, predominantly due to recurrent LOM conduction, was a common phenomenon during MI ablation, and its evaluation should therefore be the focus to improve MI ablation efficacy and durability.

A Practical Method for Ablation Catheter Reintroduction into the Left Atrium via Prior Transseptal Puncture, without Radiation.

BACKGROUND: We evaluated the feasibility and safety of reintroducing an ablation catheter (ABL) into the left atrium (LA) through a previously punctured interatrial septum under guidance of the show-catheter image-track function of the CARTO 3 3-dimensional (3D) electroanatomic mapping system. METHODS: One hundred consecutive paroxysmal or persistent drug-refractory atrial fibrillation (AF) patients (men: 55; mean age, 64.7 ± 12.1 years) who had undergone 2 fluoroscopy-guided transseptal punctures and anatomical LA reconstruction under CARTO 3-guidance, and required ABL reinsertion into the LA during mapping or ablation, were included. They were randomized 1:1 to the show-catheter (reintroduction under the CARTO 3 show-catheter image-track function) or fluoroscopy group (reintroduction under conventional fluoroscopy). RESULTS: Although the reconstructed 3D anatomy map was displaced in 21/100 patients (21.0%), the ABL was successfully reintroduced in all patients. In the show-catheter and fluoroscopy groups, model displacement incidence (18% versus 24%), tachyarrhythmias (46.0% versus 52.0%), complications (2% versus 4%), and number of ABLs reintroduced into the LA (3.3 ± 0.8 versus 3.1 ± 0.9) were similar (all P > .05). The show-catheter group displayed shorter ABL reintroduction time (9.5 ± 5.5 s versus 156.4 ± 35.5 s, P < .01), ABL reintroduction X-ray exposure time (0 s versus 39.3 ± 13.8 s, P < .01), and total X-ray exposure time (4.1 ± 1.4 min versus 4.7 ± 0.8, P < .05). CONCLUSION: During AF ablation, the catheter can be safely reintroduced into the LA, without additional fluoroscopy, under guidance of the CARTO 3 show-catheter image track function.

ITE inhibits growth of human pulmonary artery endothelial cells.

AIM: Pulmonary arterial hypertension (PAH), a deadly disorder is associated with excessive growth of human pulmonary artery endothelial (HPAECs) and smooth muscle (HPASMCs) cells. Current therapies primarily aim at promoting vasodilation, which only ameliorates clinical symptoms without a cure. 2-(1'H-indole-3'-carbonyl)-thiazole-4-carboxylic acid methyl ester (ITE) is an endogenous aryl hydrocarbon receptor (AhR) ligand, and mediates many cellular function including cell growth. However, the roles of ITE in human lung endothelial cells remain elusive. Herein, we tested a hypothesis that ITE inhibits growth of human pulmonary artery endothelial cells via AhR. MATERIALS AND METHODS: Immunohistochemistry was performed to localize AhR expression in human lung tissues. The crystal violet method and MTT assay were used to determine ITE's effects on growth of HPAECs. The AhR activation in HPAECs was confirmed using Western blotting and RT-qPCR. The role of AhR in ITE-affected proliferation of HPAECs was assessed using siRNA knockdown method followed by the crystal violet method. RESULTS: Immunohistochemistry revealed that AhR was present in human lung tissues, primarily in endothelial and smooth muscle cells of pulmonary veins and arteries, as well as in bronchial and alveolar sac epithelia. We also found that ITE dose- and time-dependently inhibited proliferation of HPAECs with a maximum inhibition of 83% at 20 µM after 6 days of treatment. ITE rapidly decreased AhR protein levels, while it increased mRNA levels of cytochrome P450 (CYP), family 1, member A1 (CYP1A1) and B1 (CYP1B1), indicating activation of the AhR/CYP1A1 and AhR/CYP1B1 pathways in HPAECs. The AhR siRNA significantly suppressed AhR protein expression, whereas it did not significantly alter ITE-inhibited growth of HPAECs. CONCLUSIONS: ITE suppresses growth of HPAECs independent of AhR, suggesting that ITE may play an important role in preventing excessive growth of lung endothelial cells.

Quercetin Inhibits Pulmonary Arterial Endothelial Cell Transdifferentiation Possibly by Akt and Erk1/2 Pathways.

This study aimed to investigate the effects and mechanisms of quercetin on pulmonary arterial endothelial cell (PAEC) transdifferentiation into smooth muscle-like cells. TGF-ß1-induced PAEC transdifferentiation models were applied to evaluate the pharmacological actions of quercetin. PAEC proliferation was detected with CCK8 method and BurdU immunocytochemistry. Meanwhile, the identification and transdifferentiation of PAECs were determined by FVIII immunofluorescence staining and α-SMA protein expression. The related mechanism was elucidated based on the levels of Akt and Erk1/2 signal pathways. As a result, quercetin effectively inhibited the TGF-ß1-induced proliferation and transdifferentiation of the PAECs and activation of Akt/Erk1/2 cascade in the cells. In conclusion, quercetin is demonstrated to be effective for pulmonary arterial hypertension (PAH) probably by inhibiting endothelial transdifferentiation possibly via modulating Akt and Erk1/2 expressions.

Effects of emodin and irbesartan on ventricular fibrosis in Goldblatt hypertensive rats.

Left ventricular (LV) fibrosis is one of the most prominent pathophysiological results of hypertension. We initiated this study to investigate the effects and mechanisms of emodin and its combination with irbesartan on LV fibrosis in Goldblatt (2K1C) hypertensive rats. Goldblatt hypertension rats were prepared by two kidney one clip (2K1C) operations and then treated with either emodin, irbesartan or their combination. As a result, the systolic blood pressure (SBP) and the left ventricular mass index (LVMI) increased significantly (P < or = 0.05) in all 2K1C rats. After drugs treatment, irbesartan and the drug combination remarkably decreased SBP, LVMI, contents of angiotensinII (AngII), hydroxyproline and collagen, the mRNA and protein expression levels of matrix metalloproteinase-2 (MMP-2) and tissue inhibitor of metalloproteinase-2 (TIMP-2) (P < or = 0.05). As for the emodin, LVMI, contents of hydroxyproline and collagen, and MMP-2 and TIMP-2 expression were found to decrease significantly; however, the SBP and AngII contents stayed stable within certain extent. Therefore, emodin, irbesartan or two drugs together can potentially inhibit the ventricular fibrosis in Goldblatt hypertensive rats by reducing MMP-2 and TIMP-2 expression. Furthermore, the combination of these two drugs may provide a better anti-fibrosis effect than the single application.