Establishment of risk prediction model for acute pulmonary embolism complicated with pulmonary hypertension
Objective To establish a prediction model for the patients with acute pulmonary embolism(APE)complicated with pulmonary hypertension(PH).Methods The clinical data of 155 APE patients were retrospectively analyzed.The patients were divided into two groups of PH(with PH,77 cases)and NPH(without PH,78 cases).The clinical features of the patients were analyzed statistically.A nomogram prediction model was constructed to predict the risk for the concurrent PH.Results Single-factor analysis was performed on the clinical features of the two groups,and multi-factor logistic regression analysis was performed after Lasso screening variables.The results showed that NT-proBNP[OR=1.000 4,95％CI(1.000 1-1.000 7)],NEUT[OR=1.396 4,95％CI(1.108 5-1.801 0)]and deep venous thrombosis[OR=3.706 2,95％CI(1.481 5-10.150 8)]were the independent influencing factors for the concurrent PH(P＜0.01).A nomogram prediction model was constructed with a goodness-of-fit R2 of 0.372 and C-index of 0.820.The areas under the ROC curves of the nomogram prediction model and each independent influencing factor were 0.820,0.778,0.705,and 0.634,respectively.The nomogram prediction model had good prediction efficiency.Conclusion The established nomogram prediction model can conveniently and early identify the concurrent PH in APE patients.
Acute pulmonary embolismPulmonary hypertensionNomogramPrediction model
NETL
NSTL
万方数据
