Risk prediction model for deterioration of heart failure in patients with chronic obstructive pulmonary disease

Objective To develop a prediction model for assessing the risk of worsening heart failure (HF) in patients with chronic obstructive pulmonary disease (COPD) in order to provide logical prediction assessment and timely systematic intervention for high-risk individuals. Methods A retrospective study was conducted on 481 COPD patients who were hospitalized for the first time due to congestive HF in our hospital from January 2016 to May 2022. Lasso regression was utilized to reduce the dimensionality of the variables, followed by fitting the Cox proportional hazards model. The accuracy and clinical utility of the model were evaluated using C-index, calibration curve, receiver operating characteristic (ROC) curve, and decision curve analysis (DCA). Based on the cutoff value determined by the nomogram model, these patients were divided into high-risk and low-risk groups, and Kaplan-Meier analysis and Log-rank test were used to analyze the difference in the probability of composite endpoints between the 2 groups. Results During the observation period, 279 patients (58.0%) experienced the outcome events, with 16 (3.3%) of them dead. The model retained 4 predictive variables: diuretic use, diagnosis of pulmonary infection, moderate or severe anemia, and left ventricular ejection fraction (LVEF). The AUC value of the established model for predicting at 0.5, 1 and 2 years were 0.699, 0.742, and 0.662, respectively, indicating good predictive accuracy. The calibration curve and DCA demonstrated the model had sound accuracy and clinical applicability. There was a significant difference in survival probability between the high-risk and low-risk groups (P < 0.001). Conclusion Our developed prediction model has good predictive performance for the probability of HF exacerbation in COPD patients. When it is used within a clinically acceptable range of risk thresholds, beneficial outcomes could be obtained.