Development and Validation of a Nomogram-Based Clinical Prediction Model for Congestive Heart Failure in Patients with Parkinson's Disease
Development and Validation of a Nomogram-Based Clinical Prediction Model for Congestive Heart Failure in Patients with Parkinson's Disease
Where did the research take place?
The study site has not been established. Author addresses may differ from where the research occurred.
Publication status: preprint
A plain-language reading has not been prepared for this paper yet.
Original abstract
Abstract Background Patients with Parkinson's disease (PD) carry a significantly elevated risk of cardiovascular complications, including congestive heart failure (CHF), yet validated clinical prediction tools specifically targeting this population remain lacking. Early identification of CHF risk in PD patients may facilitate timely intervention and improve clinical outcomes. Objective To develop and validate a nomogram-based clinical prediction model for CHF in patients with Parkinson's disease using routinely available clinical variables from the MIMIC-IV database. Methods A total of 293 patients with Parkinson's disease were included using data extracted from the MIMIC-IV database, with CHF occurrence as the binary outcome. LASSO regression was first applied for candidate variable screening, followed by backward stepwise multivariable logistic regression guided by the Akaike Information Criterion (AIC) for final variable selection and model construction. A nomogram was subsequently developed based on the final model. Internal validation was performed using bootstrap resampling, and model performance was assessed by the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). Results The final model included 11 clinical variables: age, minimum creatinine, maximum respiratory rate, minimum heart rate, maximum CK-MB, maximum bicarbonate, minimum blood glucose, maximum hemoglobin, minimum INR, minimum PTT, and maximum LDH. The nomogram achieved an AUC of 0.832, demonstrating good discriminative ability. Calibration curves showed satisfactory agreement between predicted and observed probabilities, and DCA indicated favorable net clinical benefit across a clinically relevant threshold probability range. The DeLong test confirmed that the 11-variable model significantly outperformed the simplified 8-variable model (Z = 2.5305, P = 0.01139). Conclusion The developed nomogram demonstrated good discrimination, calibration, and clinical utility, providing a practical and accessible tool for early identification and individualized risk assessment of CHF in patients with Parkinson's disease. External validation using larger multicenter cohorts is warranted before broader clinical application.