The Clinical Challenge of Early Pulmonary Vascular Disease

Exercise-induced pulmonary hypertension (EiPH) marks a critical early stage in the progression of pulmonary vascular disease. Historically, clinicians have struggled to identify this condition non-invasively, especially in patients who present with borderline resting hemodynamics but clear exertional symptoms. While resting pulmonary hypertension has established diagnostic pathways, EiPH remains elusive because it only manifests when the pulmonary vasculature is stressed during physical activity. Identifying these patients early is vital because late-stage intervention often yields poorer outcomes. Consequently, researchers have sought a robust Exercise-Induced Pulmonary Hypertension Model that can bridge the gap between resting assessments and invasive gold standards. This transition from resting to exercise-based diagnosis represents a paradigm shift in how we manage patients with chronic pulmonary artery thrombosis and other pre-capillary conditions. Current diagnostic protocols frequently miss patients whose mean pulmonary artery pressure only rises abnormally during exertion. Therefore, developing a reliable, non-invasive predictive framework is not just a scientific goal but a clinical necessity for improved patient stratification and timely therapeutic intervention.

Methodology: Integrating Clinical and Stress-Based Data

In this prospective cohort study, investigators enrolled adult patients presenting with exercise limitations or increased tricuspid regurgitation velocity. To create a highly accurate predictive tool, the team integrated clinical characteristics with advanced stress-testing modalities. Specifically, every participant underwent simultaneous stress echocardiography, cardiopulmonary exercise testing (CPET), and invasive exercise right heart catheterization. This simultaneous approach is significant because it eliminates the physiological variability that often occurs when tests are performed on different days. The researchers utilized a sequential feature selection process involving Spearman correlation, statistical testing, and LASSO regression. This rigorous statistical pipeline identified the most impactful variables while minimizing noise from redundant data. Ultimately, they applied a logistic regression model with elastic net regularization to refine the predictive accuracy. By employing five-fold cross-validation and grid search, the study ensured that the model remained robust across different patient profiles. This methodology highlights the importance of a multimodal approach, as individual tests often lack the sensitivity required to confirm EiPH without invasive confirmation.

Core Features of the Exercise-Induced Pulmonary Hypertension Model

The final Exercise-Induced Pulmonary Hypertension Model identified three specific variables as the most potent predictors of the condition. These core features include patient age, pulmonary vascular resistance (PVR) at the 25-watt exercise stage, and the VE/VCO2 slope. Age emerged as a significant factor, likely reflecting the natural decline in pulmonary vascular compliance over time. Furthermore, the model found that PVR during low-level exercise provided a more accurate snapshot of vascular health than resting measurements alone. Notably, the VE/VCO2 slope, which measures ventilatory efficiency, proved to be an indispensable component of the prediction. This parameter reflects the mismatch between ventilation and perfusion, a hallmark of pulmonary vascular dysfunction. By combining these three distinct data points, the model provides a comprehensive view of the patient’s hemodynamic status. This integration is superior to using any single metric, as it captures clinical, structural, and functional aspects of the disease simultaneously. For clinicians, these findings emphasize that looking at a single echocardiographic or CPET value may be insufficient for a definitive non-invasive diagnosis.

Validation Results and Diagnostic Performance

The study’s results demonstrated exceptional diagnostic performance, with the multimodal model achieving an area under the curve (AUC) of 0.943. This performance was significantly higher than any single modality used in isolation. For instance, the AUC for the multimodal model surpassed the results of clinical characteristics alone, stress echocardiography alone, and CPET alone. Specifically, the sensitivity and specificity reached levels that make it a viable alternative for screening patients before they undergo invasive right heart catheterization. In subgroup analyses, the model showed remarkable robustness across various patient demographics and disease severities. The DeLong test further confirmed that the differences in AUC between the multimodal model and individual testing methods were statistically significant. These findings suggest that the model effectively captures the complex physiological interactions that define EiPH. By reaching such high levels of accuracy, the model provides clinicians with a powerful tool to confidently identify or rule out early-stage pulmonary vascular disease in a non-invasive setting. This level of validation against the invasive gold standard is a major milestone in respiratory medicine.

Clinical Significance and Implementation in Practice

Implementing this non-invasive model in clinical practice could revolutionize the management of patients with suspected pulmonary vascular disease. In regions like India, where access to specialized invasive hemodynamic suites may be limited, a multimodal non-invasive screening tool is invaluable. It allows for the early detection of EiPH, which is often associated with reduced exercise capacity and a higher risk of progressing to manifest pulmonary hypertension. By using age, low-level exercise PVR, and VE/VCO2 slope, physicians can better prioritize which patients truly require invasive confirmation. Furthermore, this approach reduces the healthcare burden by avoiding unnecessary invasive procedures in low-risk individuals. The study highlights that symptomatic patients with borderline resting hemodynamics should no longer be ignored if their resting tests are normal. Instead, a multimodal stress-based assessment should be the new standard of care. Ultimately, this research paves the way for more personalized and proactive treatment strategies, potentially improving the long-term survival and quality of life for patients at risk of pulmonary hypertension.

Frequently Asked Questions

How does the multimodal model compare to traditional resting echocardiography?

The multimodal model significantly outperforms traditional resting echocardiography because it captures the dynamic changes in pulmonary circulation during physical stress. Resting tests often miss the early stages of pulmonary vascular remodeling that only manifest as hypertension when cardiac output increases during exercise. By integrating clinical data, stress echo, and CPET, the model provides a comprehensive AUC of 0.943, offering far greater diagnostic reliability than any single resting measurement could provide.

Why is the VE/VCO2 slope such a critical predictor in this model?

The VE/VCO2 slope is essential because it serves as a highly sensitive marker of ventilatory efficiency and ventilation-perfusion mismatch. In patients with early pulmonary vascular disease, the pulmonary capillary bed cannot adequately handle increased blood flow, leading to increased dead space ventilation. This physiological abnormality is captured by the VE/VCO2 slope during exercise testing, making it a powerful non-invasive indicator of underlying hemodynamic dysfunction that other tests might overlook.

Can this model replace right heart catheterization for all patients?

While the multimodal model is highly accurate for screening and predicting EiPH, it does not yet replace right heart catheterization as the definitive diagnostic gold standard. Instead, it acts as a robust gatekeeper, helping clinicians identify high-risk patients who truly need invasive confirmation. By accurately ruling out EiPH in low-risk individuals, the model reduces the number of unnecessary invasive procedures while ensuring that those with latent disease receive the advanced diagnostic attention they require.

Disclaimer: This content is for informational and educational purposes only and does not constitute medical advice. It is not a substitute for professional medical judgment, diagnosis, or treatment. Always seek the advice of your physician or other qualified healthcare provider with any questions you may have regarding a medical condition. Refer to the latest local and national guidelines for clinical practice.

References

Dai X et al. Multimodal model for predicting exercise-induced pulmonary hypertension validated by invasive exercise hemodynamics: a prospective study. Respir Res. 2026 Jul 11. doi: 10.1186/s12931-026-03814-z. PMID: 42436574.

Santoro C, Sorrentino R, Esposito R, et al. Cardiopulmonary exercise testing and echocardiographic exam: An useful interaction. Cardiovascular Ultrasound. 2019 Dec;17(1):27.

Heresi GA, et al. Chronic Thromboembolic Pulmonary Disease With Exercise Pulmonary Hypertension: A Noninvasive Model to Predict Exercise Hemodynamics. JACC: Cardiovascular Imaging. 2025 Nov 17. doi: 10.1016/j.jcmg.2025.09.012.