Construction and Validation of a Predictive Model for Chronic Obstructive Pulmonary Disease with Pulmonary...

Chronic obstructive pulmonary disease (COPD) is often complicated by pulmonary hypertension (PH), leading to symptom exacerbation, increased mortality, and poor prognosis. Early identification of high-risk patients is crucial for timely intervention. This study aims to construct and validate a machine learning-based predictive model for COPD patients with PH.

A retrospective analysis was conducted on 537 COPD patients admitted to Anhui NO.2 Provincial People's Hospital from December 2023 to November 2025. The patients were randomly divided into a model construction group (n=376) and an internal validation group (n=161) at a ratio of 7:3. Clinical and laboratory data were collected. LASSO regression, a method that effectively selects the most relevant predictors by shrinking less important variables to zero, was used to screen predictive variables. This was followed by the construction of a nomogram model through multivariate logistic regression. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curve, and clinical impact curve.

LASSO regression identified six independent predictors: Duration of COPD, Long-Term Domiciliary Oxygen Therapy, systemic inflammatory response index, red blood cell distribution width, arterial oxygen partial pressure, and B-type natriuretic peptide. The nomogram model demonstrated excellent discriminative power, with an AUC of 0.812 (95% CI: 0.762-0.863) in the model group and 0.826 (95% CI: 0.754-0.897) in the validation group. Calibration curves and clinical impact curves showed high consistency between predicted and observed values.

Our study constructed and internally validated a nomogram model based on LASSO regression, incorporating six easily accessible clinical variables, which can be used to predict the risk of PH in COPD patients. The model facilitates early clinical identification of high-risk patients, risk stratification, and implementation of individualized management.