Predicting the prognosis of patients with cardiogenic acute stroke based on the nomogram model
10.3760/cma.j.cn115396-20210728-00284
- VernacularTitle:基于列线图模型预测机械取栓治疗心源性急性脑卒中患者的预后情况
- Author:
Haihua WANG
1
;
Xiaocheng HUANG
;
Ye QIAN
;
Jun CHENG
Author Information
1. 江阴市中医院神经外科，江阴 214400
- Keywords:
Stroke;
Prognosis;
Nomograms;
Models;
Cardiogenic stroke;
Mechanical thrombus removal;
Risk of death
- From:
International Journal of Surgery
2022;49(4):248-255
- CountryChina
- Language:Chinese
-
Abstract:
Objective:A nomogram model was constructed to predict poor prognosis and death risk of mechanical thrombectomy in patients with cardiogenic acute stroke.Methods:Selected 276 patients with cardiogenic acute stroke who were treated by Jiangyin Hospital of Traditional Chinese Medicine from January 2016 to June 2020 who underwent mechanical thrombectomy as the research objects, and recorded their general information and laboratory test results. On the 90th day, the subjects were divided into a good prognosis group ( n=122) and a poor prognosis group ( n=154) according to whether the prognosis was poor or not; according to whether they died, the subjects were divided into the survival group ( n=208) and the death group ( n=68). The differences in patient related data were compared, Logistic regression analysis was used to screen for risk factors for poor prognosis and death, the line chart prediction model was established, and the ability of the column chart model to predict poor prognosis and death was evaluated by using the subject work characteristic (ROC) curve. The independent factors selected by multivariate regression analysis were used as predictors to construct a nomogram model to predict the prognosis of mechanical thrombectomy surgery in patients with cardiogenic acute stroke. The degree of calibration and validity of the nomogram model established in this study Make an evaluation. The measurement data that obey the normal distribution were represented by the Mean ± standard deviation ( ± s), and the two independent sample t test was used for the comparison between groups; The comparison of enumeration data between groups adopted chi-square test. Results:Multivariate logistic regression analysis showed age ( OR=1.165; 95% CI: 1.046-1.284; P=0.001), diabetes ( OR=1.123; 95% CI: 1.021-1.225; P<0.001), hemorrhage transformation ( OR= 2.394; 95% CI: 1.857-2.931; P=0.001), recanalization ( OR=0.418; 95% CI: 0.410-0.552; P=0.001), NIHSS score ( OR=1.502; 95% CI: 1.373-1.631); P=0.001), neutrophil count (NEUT) ( OR=1.024; 95% CI: 1.009-1.139; P=0.001), NEUT/lymphocyte count (NLR) ( OR=1.235; 95% CI: 1.112-1.358; P=0.001), D-dimer ( OR=1.939; 95% CI: 1.328-2.551; P=0.001) was an independent risk factor for poor prognosis in patients with cardiogenic acute stroke; age ( OR=1.153; 95% CI: 1.080-1.226; P<0.001), hemorrhage transformation ( OR=6.330; 95% CI: 4.904-7.754; P=0.001), recanalization ( OR=0.418; 95% CI: 0.323-0.514; P=0.001), NIHSS score ( OR=2.051; 95% CI: 1.784-2.338; P=0.001), NEUT ( OR=1.399; 95% CI: 1.275-1.523; P=0.001), NLR ( OR=1.528; 95% CI: 1.414-1.642; P=0.001), D-dimer ( OR=2.391; 95% CI: 1.948-2.834; P=0.001) was an independent predictor of death in patients with cardiogenic acute stroke. The established nomogram model predicted poor prognosis and the area under the ROC curve of death were 0.814 (95% CI: 0.800-0.828) and 0.842 (95% CI: 0.828-0.857). Conclusions:Age, hemorrhage transformation, recanalization, NIHSS score, NEUT, NLR, and D-dimer are all important for the prognosis of patients with cardiogenic acute stroke by mechanical thrombectomy. Diabetes only has a suggestive effect on poor prognosis. The nomogram model established based on these factors can effectively help clinicians evaluate the prognosis of patients, formulate reasonable treatment plans for them, and improve the prognosis.
