A comprehensive study on machine learning models combining with oversampling for bronchopulmonary dysplasia-associated pulmonary hypertension in very preterm infants.

BACKGROUND: Bronchopulmonary dysplasia-associated pulmonary hypertension (BPD-PH) remains a devastating clinical complication seriously affecting the therapeutic outcome of preterm infants. Hence, early prevention and timely diagnosis prior to pathological change is the key to reducing morbidity and improving prognosis. Our primary objective is to utilize machine learning techniques to build predictive models that could accurately identify BPD infants at risk of developing PH. METHODS: The data utilized in this study were collected from neonatology departments of four tertiary-level hospitals in China. To address the issue of imbalanced data, oversampling algorithms synthetic minority over-sampling technique (SMOTE) was applied to improve the model. RESULTS: Seven hundred sixty one clinical records were collected in our study. Following data pre-processing and feature selection, 5 of the 46 features were used to build models, including duration of invasive respiratory support (day), the severity of BPD, ventilator-associated pneumonia, pulmonary hemorrhage, and early-onset PH. Four machine learning models were applied to predictive learning, and after comprehensive selection a model was ultimately selected. The model achieved 93.8% sensitivity, 85.0% accuracy, and 0.933 AUC. A score of the logistic regression formula greater than 0 was identified as a warning sign of BPD-PH. CONCLUSIONS: We comprehensively compared different machine learning models and ultimately obtained a good prognosis model which was sufficient to support pediatric clinicians to make early diagnosis and formulate a better treatment plan for pediatric patients with BPD-PH.

Continuous heart monitoring to evaluate treatment effects in pulmonary hypertension.

BACKGROUND: The treatment of pulmonary hypertension (PH) has improved rapidly in recent decades. There is increasing evidence to support the role of early intervention and treatment in affecting clinical outcomes in PH. OBJECTIVES: To assess treatment effects before and after the escalation of specific PH treatments using continuous heart monitoring with a Reveal LINQ loop recorder. METHODS: Patients were compared before and after treatment escalation. Treatment escalation was defined as an additional pulmonary arterial hypertension (PAH) drug, pulmonary endarterectomy, percutaneous balloon angioplasty or bilateral lung transplantation. Specifically, changes in heart rate variability (HRV), heart rate (HR) and physical activity were assessed. RESULTS: In this prospective study, 41 patients (27 with PAH and 14 with chronic thromboembolic pulmonary hypertension (CTEPH)) were enrolled. Among them, 15 (36.6%) patients underwent PH treatment escalation. Prior to escalation, patients were monitored for a median of 100 (range: 68-100) days and after therapy escalation for a median duration of 165 (range: 89-308) days. In the escalation group, there was a significant increase in HRV, physical activity indexed by daytime HR and a significant decrease in nighttime HR assessed at baseline and after treatment escalation in both the PAH and CTEPH groups. This was paralleled by significant improvements in WHO functional class, 6-min walking distance and N-terminal pro-b-type natriuretic peptide. CONCLUSIONS: This is the first study to demonstrate an association between specific PH therapies and changes in HRV, HR nighttime and physical activity. This indicates the potential of continuous monitoring in the evaluation of treatment effects in PH.

Obstructive Sleep Apnea, Obesity Hypoventilation Syndrome, and Pulmonary Hypertension: A State-of-the-Art Review.

The pathophysiological interplay between sleep-disordered breathing (SDB) and pulmonary hypertension (PH) is complex and can involve a variety of mechanisms by which SDB can worsen PH. These mechanistic pathways include wide swings in intrathoracic pressure while breathing against an occluded upper airway, intermittent and/or sustained hypoxemia, acute and/or chronic hypercapnia, and obesity. In this review, we discuss how the downstream consequences of SDB can adversely impact PH, the challenges in accurately diagnosing and classifying PH in the severely obese, and review the limited literature assessing the effect of treating obesity, obstructive sleep apnea, and obesity hypoventilation syndrome on PH.

Identification of Serum Interleukin-22 as Novel Biomarker in Pulmonary Hypertension: A Translational Study.

Growing evidence suggests the crucial involvement of inflammation in the pathogenesis of pulmonary hypertension (PH). The current study analyzed the expression of interleukin (IL)-17a and IL-22 as potential biomarkers for PH in a preclinical rat model of PH as well as the serum levels in a PH patient collective. PH was induced by monocrotalin (60 mg/kg body weight s.c.) in 10 Sprague Dawley rats (PH) and compared to 6 sham-treated controls (CON) as well as 10 monocrotalin-induced, macitentan-treated rats (PH_MAC). Lung and cardiac tissues were subjected to histological and immunohistochemical analysis for the ILs, and their serum levels were quantified using ELISA. Serum IL levels were also measured in a PH patient cohort. IL-22 expression was significantly increased in the lungs of the PH and PH_MAC groups (p = 0.002), whereas increased IL17a expression was demonstrated only in the lungs and RV of the PH (p < 0.05) but not the PH_MAC group (p = n.s.). The PH group showed elevated serum concentrations for IL-22 (p = 0.04) and IL-17a (p = 0.008). Compared to the PH group, the PH_MAC group demonstrated a decrease in IL-22 (p = 0.021) but not IL17a (p = n.s.). In the PH patient collective (n = 92), increased serum levels of IL-22 but not IL-17a could be shown (p < 0.0001). This elevation remained significant across the different etiological groups (p < 0.05). Correlation analysis revealed multiple significant relations between IL-22 and various clinical, laboratory, functional and hemodynamic parameters. IL-22 could serve as a promising inflammatory biomarker of PH with potential value for initial diagnosis, functional classification or even prognosis estimation. Its validation in larger patients' cohorts regarding outcome and survival data, as well as the probability of promising therapeutic target structures, remains the object of further studies.

Impression life with pulmonary hypertension: clinically relevant characteristics and quality of life among patients in Gansu, China.

BACKGROUND: The adverse effects of pulmonary arterial hypertension (PAH) on physical, emotional, and health-related quality of life (HRQoL) remain primarily unrecognized, especially in resource-limited settings. OBJECTIVES: This study aims to characterize the HRQoL of patients with PAH in this area and also identify the potential role of clinically relevant characteristics, including the 6-min walk distance test (6MWD), WHO-Functional Classification (WHO-FC), and mental health in the occurrence of lowering quality of life. DESIGN: This was a cross-sectional observational study. METHODS: Inpatients with PAH were chosen from a tertiary hospital located in Gansu province, China. All participants were interviewed face-by-face by using questionnaires, including items from the 36-Item Short Form Health Survey (SF-36), the self-rating anxiety scale, and the self-rating depression scale. Data on demographic and clinically relevant characteristics, including WHO-FC and 6MWD, were also collected by tracing medical recorders. Multiple linear regression analysis was used to determine the association between demographic, clinically relevant characteristics data, and physical component summary (PCS) or mental component summary (MCS) in SF-36. RESULTS: Of the 152 participants, SF-36 differed significantly from Chinese norms in all eight domains, with role-physical (21.55 ± 9.87) less than one-third of the norm (88.79 ± 28.49). Multiple linear regression results showed that the factors with the greatest impact on PCS were anxiety scores (ß = -0.22, p = 0.001), followed by WHO-FC (ß = -0.16, p = 0.014) and 6MWD (ß = 0.15, p = 0.036). The factors with the greatest impact on MCS were WHO-FC (ß = -0.30, p < 0.001), followed by anxiety (ß = -0.23, p = 0.001) and depression scores (ß = -0.16, p = 0.013). CONCLUSION: HRQoL was substantially reduced among PAH patients in the resource-limited area, mainly the physiological functions. WHO-FC and anxiety scores were independently associated with both PCS and MCS in SF-36. Clinicians should make reasonable rehabilitation programs and plans for patients according to their cardiac function grade and the severity of clinical symptoms. In addition, psychological interventions should also be taken, especially for those with anxiety symptoms, so as to improve their HRQoL.

Development and Validation of a Nomogram of Persistent Pulmonary Hypertension in Adult Pretricuspid Shunts After Correction.

BACKGROUND: Pretricuspid shunts have been associated with poorer survival rates in patients with Eisenmenger syndrome compared with postricuspid shunts and complex lesions. However, the risk stratification for persistent pulmonary hypertension (PH) in this population remains uncertain. METHODS AND RESULTS: We retrospectively enrolled 103 patients with pretricuspid shunts with high total pulmonary resistance >4.5 Wood units (estimated pulmonary vascular resistance ≥3 Wood units). During a mean±SD follow-up of 20.95±24.84 months, 32 patients developed postoperative persistent PH after shunt correction. We identified 3 significant predictors of postoperative persistent PH, including mean pulmonary artery pressure after inhaled oxygen ≥40.5 mm Hg (odds ratio [OR], 7.78 [95% CI, 2.02-30.03]; P<0.01), total pulmonary resistance after inhaled oxygen ≥6.5 Wood units (estimated pulmonary vascular resistance ≥5 Wood units; OR, 12.23 [95% CI, 2.12-70.46]; P<0.01), and artery oxygen saturation at rest <95% (OR, 3.34 [95% CI, 1.07-10.44]; P=0.04). We established the prediction model with the C-statistics of 0.85 (95% CI, 0.77-0.93; P<0.01), and the C-statistic was 0.83 (95% CI, 0.80-0.86) after bootstrapping 10 000 times with a good performance of the nomogram calibration curve for predicting persistent PH. CONCLUSIONS: Our study presents a multivariable risk stratification model for persistent PH after shunt correction in adults with pretricuspid shunts. This model, based on 3 hemodynamic predictors after inhaled oxygen, may assist in identifying individuals at higher risk of persistent PH after shunt correction.

Fractional exhaled nitric oxide in idiopathic pulmonary arterial hypertension and mixed connective tissue disease complicating pulmonary hypertension.

BACKGROUND: Fractional exhaled nitric oxide (FeNO) has been extensively studied in various causes of pulmonary hypertension (PH), but its utility as a noninvasive marker remains highly debated. The objective of our study was to assess FeNO levels in patients with idiopathic pulmonary arterial hypertension (IPAH) and mixed connective tissue disease complicating pulmonary hypertension (MCTD-PH), and to correlate them with respiratory functional data, disease severity, and cardiopulmonary function. METHODS: We collected data from 54 patients diagnosed with IPAH and 78 patients diagnosed with MCTD-PH at the Shanghai Pulmonary Hospital Affiliated to Tongji University. Our data collection included measurements of brain natriuretic peptide (pro-BNP), cardiopulmonary exercise test (CPET), pulmonary function test (PFT), impulse oscillometry (IOS), and FeNO levels. Additionally, we assessed World Health Organization functional class (WHO-FC) of each patient. RESULTS: (1) The fractional exhaled concentration of nitric oxide was notably higher in patients with IPAH compared to those with MCTD-PH. Furthermore, within the IPAH group, FeNO levels were found to be lower in cases of severe IPAH compared to mild IPAH (P = 0.024); (2) In severe pulmonary hypertension as per the WHO-FC classification, FeNO levels in IPAH exhibited negative correlations with FEV1/FVC (Forced Expiratory Velocity at one second /Forced Vital Capacity), MEF50% (Maximum Expiratory Flow at 50%), MEF25%, and MMEF75/25% (Maximum Mid-expiratory Flow between 75% and 25%), while in severe MCTD-PH, FeNO levels were negatively correlated with R20% (Resistance at 20 Hz); (3) ROC (Receiving operator characteristic curve) analysis indicated that the optimal cutoff value of FeNO for diagnosing severe IPAH was 23ppb; (4) While FeNO levels tend to be negatively correlated with peakPETO2(peak end-tidal partial pressure for oxygen) in severe IPAH, in mild IPAH they had a positive correlation to peakO2/Heart rate (HR). An interesting find was observed in cases of severe MCTD-PH, where FeNO levels were negatively correlated with HR and respiratory exchange ratio (RER), while positively correlated with O2/HR throughout the cardiopulmonary exercise test. CONCLUSION: FeNO levels serve as a non-invasive measure of IPAH severity. Although FeNO levels may not assess the severity of MCTD-PH, their significant makes them a valuable tool when assessing severe MCTD-PH.

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.

Invasive Cardiac Hemodynamics in Apical Hypertrophic Cardiomyopathy.

BACKGROUND: Symptomatic limitations in apical hypertrophic cardiomyopathy may occur because of diastolic dysfunction with resultant elevated left ventricular filling pressures, cardiac output limitation to exercise, pulmonary hypertension (PH), valvular abnormalities, and/or arrhythmias. In this study, the authors aimed to describe invasive cardiac hemodynamics in a cohort of patients with apical hypertrophic cardiomyopathy. METHODS AND RESULTS: Patients presenting to a comprehensive hypertrophic cardiomyopathy center with apical hypertrophic cardiomyopathy were identified (n=542) and those who underwent invasive hemodynamic catheterization (n=47) were included in the study. Of these, 10 were excluded due to postmyectomy status or incomplete hemodynamic data. The mean age was 56±18 years, 16 (43%) were women, and ejection fraction was preserved (≥50%) in 32 (91%) patients. The most common indication for catheterization was dyspnea (48%) followed by suspected PH (13%), and preheart transplant evaluation (10%). Elevated left ventricular filling pressures at rest or exercise were present in 32 (86%) patients. PH was present in 30 (81%) patients, with 6 (20%) also having right-sided heart failure. Cardiac index was available in 25 (86%) patients with elevated resting filling pressures. Of these, 19 (76%) had reduced cardiac index and all 6 with right-sided heart failure had reduced cardiac index. Resting hemodynamics were normal in 8 of 37 (22%) patients, with 5 during exercise; 3 of 5 (60%) patients had exercise-induced elevation in left ventricular filling pressures. CONCLUSIONS: In patients with apical hypertrophic cardiomyopathy undergoing invasive hemodynamic cardiac catheterization, 86% had elevated left ventricular filling pressures at rest or with exercise, 81% had PH, and 20% of those with PH had concomitant right-sided heart failure.

Post-Pulmonary Embolism Syndrome: An Update Based on the Revised AWMF-S2k Guideline.

In survivors of acute pulmonary embolism (PE), the post-PE syndrome (PPES) may occur. In PPES, patients typically present with persisting or progressive dyspnea on exertion despite 3 months of therapeutic anticoagulation. Therefore, a structured follow-up is warranted to identify patients with chronic thromboembolic pulmonary disease (CTEPD) with normal pulmonary pressure or chronic thromboembolic pulmonary hypertension (CTEPH). Both are currently understood as a dual vasculopathy, that is, secondary arterio- and arteriolopathy, affecting the large and medium-sized pulmonary arteries as well as the peripheral vessels (diameter < 50 µm). The follow-up algorithm after acute PE commences with identification of clinical symptoms and risk factors for CTEPH. If indicated, a stepwise performance of echocardiography, ventilation-perfusion scan (or alternative imaging), N-terminal prohormone of brain natriuretic peptide (NT-proBNP) level, cardiopulmonary exercise testing, and pulmonary artery catheterization with angiography should follow. CTEPH patients should be treated in a multidisciplinary center with adequate experience in the complex therapeutic options, comprising pulmonary endarterectomy, balloon pulmonary angioplasty, and pharmacological interventions.

Vitamin C deficiency can lead to pulmonary hypertension: a systematic review of case reports.

BACKGROUND: In the early literature, unintentional vitamin C deficiency in humans was associated with heart failure. Experimental vitamin C deficiency in guinea pigs caused enlargement of the heart. The purpose of this study was to collect and analyze case reports on vitamin C and pulmonary hypertension. METHODS: We searched Pubmed and Scopus for case studies in which vitamin C deficiency was considered to be the cause of pulmonary hypertension. We selected reports in which pulmonary hypertension was diagnosed by echocardiography or catheterization, for any age, sex, or dosage of vitamin C. We extracted quantitative data for our analysis. We used the mean pulmonary artery pressure (mPAP) as the outcome of primary interest. RESULTS: We identified 32 case reports, 21 of which were published in the last 5 years. Dyspnea was reported in 69%, edema in 53% and fatigue in 28% of the patients. Vitamin C plasma levels, measured in 27 cases, were undetectable in 24 and very low in 3 cases. Diet was poor in 30 cases and 17 cases had neuropsychiatric disorders. Right ventricular enlargement was reported in 24 cases. During periods of vitamin C deficiency, the median mPAP was 48 mmHg (range 29-77 mmHg; N = 28). After the start of vitamin C administration, the median mPAP was 20 mmHg (range 12-33 mmHg; N = 18). For the latter 18 cases, mPAP was 2.4-fold (median) higher during vitamin C deficiency. Pulmonary vascular resistance (PVR) during vitamin C deficiency was reported for 9 cases, ranging from 4.1 to 41 Wood units. PVR was 9-fold (median; N = 5) higher during vitamin C deficiency than during vitamin C administration. In 8 cases, there was direct evidence that the cases were pulmonary artery hypertension (PAH). Probably the majority of the remaining cases were also PAH. CONCLUSIONS: The cases analyzed in our study indicate that pulmonary hypertension can be one explanation for the reported heart failure of scurvy patients in the early literature. It would seem sensible to measure plasma vitamin C levels of patients with PH and examine the effects of vitamin C administration.

Sendaway capillary NT-proBNP in pulmonary hypertension.

BACKGROUND: N-terminal pro-B-type natriuretic peptide (NT-proBNP) is a biomarker of cardiac ventricular wall stress that is incorporated into pulmonary hypertension (PH) risk stratification models. Sendaway sampling may enable patients to perform NT-proBNP tests remotely. This UK-wide study aimed to assess the agreement of sendaway NT-proBNP with standard venous NT-proBNP and to assess the effect of delayed processing. METHODS: Reference venous NT-proBNP was collected from PH patients. Samples for capillary and venous sendaway tests were collected contemporaneously, mailed to a reference laboratory and processed at 3 and 7 days using a Roche Cobas e411 device. Differences in paired measurements were analysed with Passing-Bablok regression, percentage difference plots and the % difference in risk strata. RESULTS: 113 patients were included in the study. 13% of day 3 capillary samples were insufficient. Day 3 capillary samples were not equivalent to reference samples (Passing Bablok analysis slope of 0.91 (95% CI 0.88 to 0.93) and intercept of 6.0 (95% CI 0.2 to 15.9)). The relative median difference was -7% and there were acceptable limits of agreement. Day 3 capillary NT-proBNP accurately risk stratified patients in 93.5% of cases. By comparison, day 3 venous results accurately risk stratified patients in 90.1% of cases and were equivalent by Passing-Bablok regression. Delayed sampling of sendaway tests led to an unacceptable level of agreement and systematically underestimated NT-proBNP. CONCLUSIONS: Sendaway NT-proBNP sampling may provide an objective measure of right ventricular strain for virtual PH clinics. Results must be interpreted with caution in cases of delayed sampling.

A clinician's guide to pulmonary hypertension.

ABSTRACT: Despite advances in diagnosis and treatment, pulmonary hypertension has high morbidity and mortality. The presenting symptoms often are vague and may mimic other more common diseases, so patients can be misdiagnosed or missed early in the disease process. Early detection of pulmonary hypertension by primary care providers can play an important role in patient outcomes and survival. Identifying signs and symptoms, understanding the causes and classifications, and knowing the systematic approach to evaluating and diagnosing patients with suspected pulmonary hypertension are key to preventing premature patient decline.

Pulmonary Hypertension in Women.

Pulmonary arterial hypertension (PAH) is a rare devastating disease characterized by elevated pulmonary artery pressure and increased pulmonary vascular resistance. Females have a higher incidence of PAH, which is reflected globally across registries in the United States, Europe, and Asia. However, despite female predominance, women had better outcomes compared with male patients, a finding that has been labeled the "estrogen paradox." Special considerations should be given to women with PAH regarding sexual health, contraception, family planning, and treatment before, during, and after pregnancy. Pregnant women with PAH should be referred to a pulmonary hypertension care center; a multidisciplinary team approach is recommended, and Cesarean section is the preferred mode of delivery. While pregnancy outcomes have improved over the years with PAH-specific therapy, pregnancy portends a high-risk for those with PAH. Continued research is needed to tailor PAH treatment for women.

Neonatal persistent pulmonary hypertension related to a novel TBX4 mutation: case report and review of the literature.

TBX4 gene, located on human chromosome 17q23.2, encodes for T-Box Transcription Factor 4, a transcription factor that belongs to the T-box gene family and it is involved in the regulation of some embryonic developmental processes, with a significant impact on respiratory and skeletal illnesses. Herein, we present the case of a female neonate with persistent pulmonary hypertension (PH) who underwent extracorporeal membrane oxygenation (ECMO) on the first day of life and then resulted to have a novel variant of the TBX4 gene identified by Next-Generation Sequencing. We review the available literature about the association between PH with neonatal onset or emerging during the first months of life and mutations of the TBX4 gene, and compare our case to previously reported cases. Of 24 cases described from 2010 to 2023 sixteen (66.7%) presented with PH soon after birth. Skeletal abnormalities have been described in 5 cases (20%). Eleven cases (46%) were due to de novo mutations. Three patients (12%) required ECMO. Identification of this variant in affected individuals has implications for perinatal and postnatal management and genetic counselling. We suggest including TBX4 in genetic studies of neonates with pulmonary hypertension, even in the absence of skeletal abnormalities.

Comment on: Severe pulmonary hypertension in pulmonary alveolar microlithiasis: A comprehensive literature review.

This letter commends the study "Severe pulmonary hypertension in pulmonary alveolar microlithiasis: A comprehensive literature review" for its thorough exploration of Pulmonary Alveolar Microlithiasis (PAM) and its association with pulmonary hypertension (PH). The study offers insights into PAM's genetics, clinical manifestations, diagnostic approaches, and treatment modalities. It highlights the importance of early diagnosis and management while discussing limitations such as its retrospective nature and small sample size. Despite these limitations, the study contributes significantly to understanding PAM and PH, emphasizing the need for larger prospective studies to validate findings and explore novel therapeutic avenues.