# Construction and validation of a prediction model for complications of femoral artery access

## TL;DR

This study identifies risk factors for femoral artery access complications and builds a highly accurate prediction model to help prevent them.

## Contribution

A novel nomogram prediction model for femoral artery access complications with high accuracy and validated performance.

## Key findings

- Four independent risk factors were identified: vascular calcification, diabetes, platelet count, and puncture point position.
- The nomogram model achieved 94.7% accuracy, 81.8% sensitivity, and 95.2% specificity in predicting complications.
- Calibration and decision curve analyses confirmed the model's strong predictive performance and clinical utility.

## Abstract

To analyze the risk factors for the complications of access and to construct and validate a nomogram prediction model for their occurrence.

Patients undergoing endovascular intervention via femoral artery access between January 2020 and April 2025 were enrolled in the study. Related clinical data were retrospectively collected and analyzed. Patients were divided into complication (n = 19) and non-complication (n = 488) groups based on the occurrence of postoperative complications associated with femoral artery puncture site. The general cohort characteristics were compared between the two groups, and the risk factors for the postoperative complications were identified based on univariate and multivariate logistic regression analyses. A nomogram prediction model was constructed and its performance was evaluated using the area under the receiver operating characteristic (ROC) curve, the Hosmer-Lemeshow test, calibration curve, and decision curve analyses.

Four potential predictors were identified based on the multivariate logistic regression analysis results: vascular calcification [odds ratio (OR) = 7.952, 95% confidence interval (CI): 1.653–38.254], history of diabetes (OR = 18.793, 95% CI: 3.670–96.225), platelet count (OR = 0.980, 95% CI: 0.967–0.994), and positional relationship between the puncture point and femoral head (OR = 6.125, 95% CI: 1.048–35.800). The nomogram model incorporating these factors demonstrated strong performance, with an area under the ROC curve of 0.924 (95% confidence interval: 0.839–1.000), sensitivity of 81.80%, specificity of 95.20%, and overall accuracy of 94.70%.The Hosmer-Lemeshow test yielded χ2 = 12.535 and P = 0.8184, indicating a good model fit. Calibration curves showed strong agreement between the nomogram predictions and observed outcomes. Both the ROC and decision curve analysis confirmed the nomogram's robust predictive performance.

Platelet count, history of diabetes, vascular calcification, and positional relationship between the puncture point and the femoral head are independent risk factors for the complications of femoral artery access. The nomogram model established based on these indicators demonstrated a high accuracy in predicting the risk of complications.