Usefulness of risk assessment tools in predicting hemodynamic outcome after balloon pulmonary angioplasty: a comparative analysis.

Abstract

Objectives: Several parameters of widely used risk assessment tools for pulmonary arterial hypertension (PAH) have been linked to hemodynamic outcomes of balloon pulmonary angioplasty (BPA). Therefore, we aimed to determine whether these risk assessment tools could be used to predict hemodynamic outcomes following BPA.

Methods: In this retrospective study, we included 139 patients with chronic thromboembolic pulmonary hypertension who had undergone BPA at Center for Pulmonary Vascular Diseases, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College (Beijing, China). We compared the accuracies of seven well-validated risk assessment tools for predicting hemodynamic outcomes following BPA. A favorable hemodynamic outcome was defined as a mean pulmonary arterial pressure < 30 mmHg at follow-up.

Results: The baseline risk profiles varied significantly among the risk assessment tools. The US Registry to Evaluate Early and Long-Term PAH Disease Management risk scales and the French risk assessment tools rated most patients as high-risk, while the Comparative, Prospective Registry of Newly Initiated Therapies for Pulmonary Hypertension (COMPERA) series and laboratory examination-based risk scales categorized most patients as having intermediate-risk profile. COMPERA 2.0 (4-strata) exhibited the highest predictive power among all risk stratifications. Noninvasive risk stratification (COMPERA 2.0 [3-strata]) showed a comparable predictive ability to that of invasive risk stratification (COMPERA 1.0) (area under the curve 0.649 vs. 0.648). Moreover, incorporating diffusing capacity of the lungs for carbon monoxide and tricuspid regurgitation velocity into COMPERA 2.0 (4-strata) further enhanced its predictive power (net reclassification index 0.153, 95% confidence interval 0.009-0.298, p = 0.038). Additionally, this refined COMPERA version had a high calibration accuracy (slope 0.96).

Conclusion: Although the risk strata distribution varied among different risk assessment tools, the proportion of patients achieving favorable hemodynamics decreased with the escalation of risk stratification in most models. The well-validated risk assessment tools for PAH could also predict hemodynamic outcomes following BPA, and the refined COMPERA 2.0 model exhibited the highest predictive ability among these. Applying risk assessment tools before BPA can facilitate early identification of patients in need of closer monitoring and more intensive interventions, contributing to a better prognosis after BPA.