The importance of preoperative oxygen saturation as a predictor of pulmonary arterial hypertension after surgery of atrial septal defects

Development and web deployment of prediction model for pulmonary arterial pressure in chronic thromboembolic pulmonary hypertension using machine learning

BACKGROUND AND PURPOSE

Mean pulmonary artery pressure (mPAP) is a key index for chronic thromboembolic pulmonary hypertension (CTEPH). Using machine learning, we attempted to construct an accurate prediction model for mPAP in patients with CTEPH.

METHODS

A total of 136 patients diagnosed with CTEPH were included, for whom mPAP was measured. The following patient data were used as explanatory variables in the model: basic patient information (age and sex), blood tests (brain natriuretic peptide (BNP)), echocardiography (tricuspid valve pressure gradient (TRPG)), and chest radiography (cardiothoracic ratio (CTR), right second arc ratio, and presence of avascular area). Seven machine learning methods including linear regression were used for the multivariable prediction models. Additionally, prediction models were constructed using the AutoML software. Among the 136 patients, 2/3 and 1/3 were used as training and validation sets, respectively. The average of R squared was obtained from 10 different data splittings of the training and validation sets.

RESULTS

The optimal machine learning model was linear regression (averaged R squared, 0.360). The optimal combination of explanatory variables with linear regression was age, BNP level, TRPG level, and CTR (averaged R squared, 0.388). The R squared of the optimal multivariable linear regression model was higher than that of the univariable linear regression model with only TRPG.

CONCLUSION

We constructed a more accurate prediction model for mPAP in patients with CTEPH than a model of TRPG only. The prediction performance of our model was improved by selecting the optimal machine learning method and combination of explanatory variables.

A Numerical Simulation Comparing a Cavopulmonary Assist Device and VA ECMO for Failing Fontan Support

A cavopulmonary assist device (CPAD) has been developed for failing Fontan support. This CPAD pumps blood from superior/inferior vena cavae (SVC/IVC) to pulmonary artery. In this study, we compared failing Fontan support with CPAD versus veno-arterial extracorporeal membrane oxygenation (VA ECMO) in silico. A numerical lumped-parameter circulatory system model was used to simulate Fontan circulation. Failing Fontan was simulated by increased pulmonary resistance. Central venous pressure (CVP), mean pulmonary arterial pressure (mPAP), left atrial pressure (LAP), and univentricular outflow (CO) were simulated and compared with published clinical data. The CPAD and VA ECMO were simulated using 1-5 L/min pump flows. In agreement with published clinical data, the simulated failing Fontan condition had increased CVP (19 mmHg) and mPAP (18 mmHg) with decreased LAP (7 mmHg) and cardiac output (3 L/min) compared with functional Fontan condition. The CPAD achieved total Fontan assistance with pump flows higher than original CO. Veno-arterial extracorporeal membrane oxygenation provided partial Fontan assistance with low pump flows. Blood went through pulmonary circulation with CPAD whereas VA ECMO bypassed pulmonary circulation and diminished univentricular blood flow (0.8 L/min). This in silico study demonstrated that CPAD preserved heart/lung function whereas VA ECMO had very low univentricular flow, potentially leading to thrombosis or univentricular atrophy.