Establishment and verification of prediction model on failure factors of radial arterial puncture in patients with heart failure

Article ID: 4671

Keywords: heart failure; radial arterial puncture; prediction model; regression

Abstract

Objective: This paper aims to establish a prediction model for the failure factors of radial arterial puncture in patients with heart failure and verify its effectiveness. Method: 969 patients from the internal medicine department of Dujiangyan Hospital of Traditional Chinese Medicine from July 2019 to July 2021 were divided into a training set (n=700, 72.5%) and a validation set (n= 269, 27.5%) using the leave-one-out (LOO) method. Logistic regression was used to select the risk factors of failure of the first needle of radial artery puncture in patients with heart failure, and the corresponding prediction model was established. The ROC curve was drawn and the effectiveness of the model was verified. Results: Logistic regression showed that increased heart rate, rapid respiration, decreased mean arterial pressure, and decreased plasma albumin were independent risk factors for radial artery puncture failure in patients with heart failure. The probability of radial artery puncture failure was calculated as 8.010+0.025×Heart rate+0.086×Respiratory rate -0.070×Means arterial pressure-0.232×Plasma albumin. The area under the ROC curve (AUC) was 0.823 in the training set and 0.765 in the verification set. The Hosmer-Lemeshow test showed that the prediction model had good stability. The maximum value of the Jordan index of the model was 0.524, the optimal critical value of the corresponding prediction probability was 0.658, the sensitivity was 84.7%, and the specificity was 67.8%. Conclusion: The prediction model based on regression analysis has a moderate predictive power for the radial arterial puncture in patients with heart failure and can assist clinical decisions, reduce patient pain, and save medical consumable expense