| ObjectiveIn order to explore the risk factors of pulmonary embolism after surgical treatment,a new predictive model was established.Objective and easily detectable clinical indicators are monitored to assess the risks of all patients,and anticoagulant therapy is given to specific inpatients or outpatients,thereby reducing the incidence and mortality of pulmonary thromboembolism in highrisk patients,and reducing the hospitalization costs of low-risk patients.At the same time,it inherits and develops the existing Wells model and the revised Geneva model,and updates the knowledge of the two prediction models.Data and methods1.This study systematically reviewed the patients who were admitted to the Second Hospital of Jilin University from January 2018 to October 2021,and were clinically considered for diagnosis of pulmonary embolism.Through strict inclusion and exclusion criteria,143 patients finally met the inclusion criteria,and were divided into PE group and non-PE group according to the CTPA results.Age,sex,body mass index,D-dimer,heart rate,surgical grade,hypertension,anesthesia category,tumor history,and VTE history were 10 variables.Single factor logistic regression and multifactor logistic regression were used to identify the high risk factors of pulmonary embolism,and then the associated risk factors were assigned.The patients were divided into high risk group and low risk group by using Yorden index to determine the cut off value of the prediction model.The ROC curve was used to evaluate the prediction accuracy of the model and determine the reliability of clinical application.In order to reduce the waste of medical resources,targeted anticoagulant therapy should be carried out in advance for high risk patients to reduce mortality,whereas conservative treatment for low risk patients.2.The D-dimer test was performed on all patients using the ELASA method by the same equipment,and the test results were taken to two decimal places.For patients with initial diagnosis of DVT,double lower extremity deep vein color doppler ultrasonography should be selected as the gold standard for diagnosis.The examination report should be independently issued by two physicians with clinical work experience,and the final diagnosis should be determined by a specific superior physician with rich clinical experience when there is any divergence in the report.The previous diagnosis of VTE must be supported by examination results.The classification of surgical grade should be determined by two clinicians through consultation and checked again by the researchers in the medical record department.When there are differences in the classification of the two surgical grades,the standards for classification of surgical grade by the researchers in the medical record department should be prevailed.For the patients receiving multiple operations,priority should be given to the last operation before CTPA examination.Higher grade operations should be prevailed when patients received multiple operation at the same time.The postoperative time was calculated in days,and less than one day should be counted as one day.General anesthesia mainly includes intubation anesthesia and intravenous anesthesia,while local anesthesia includes surface infiltration anesthesia and subarachnoid anesthesia.If the admission department is inconsistent with the discharge department,the Department which patients suffered pulmonary embolism is the main department.The tumor history mainly includes previously diagnosed as maligant tumor,which not included inflammatory pseudotumor,lipoma,hamartoma and other benign tumor or carcinoma in situ,microinvasive carcinoma.In this paper,the Chi-square test or Fisher test is used to compare the counting data.The t-test or the homogeneity test is used when the normal distribution is satisfied;the rank sum test is used when the data is skewed.Result1.Of the 143 patients who met the inclusion and exclusion criteria,116 were in the PE group,accounting for 81.1%,and 27 in the non-PE group,accounting for 18.9%.The high risk factors of PE included Age（X~2= 10.732,P = 0.001）,BMI index（X~2= 14.984,P = 0.001）,tumor history（X~2=18.674,P < 0.001）,VTE history（X~2= 21.401,P < 0.001）,surgical factors（X~2= 21.558,P < 0.001）,D dimer（X~2= 10.091,P = 0.001）et.al.The incidence of PE was not significant by sex（X~2 = 0.226,P = 0.634）,hypertension（X~2 = 3.229,P = 0.072）and narcotic factors（X~2= 0.935,P = 0.334）.PE is more common in people ≥ 65 years of age,overweight or obese,undergoing a four grade operation,and having a history of cancer or VTE.The distribution of suspected cases of pulmonary embolism is obviously tendentious;respiratory medicine（32 cases）,gastrointestinal surgery（17 cases）,emergency medicine（13 cases）,gynecology（11 cases）,nephrology（11 cases）,trauma surgery（10cases）,cardiovascular surgery（7 cases）,spine surgery（5 cases）and thoracic surgery（5 cases）are more common;22 cases of pulmonary embolism are confirmed in the remaining 13 departments.2.The results of single-factor logistic regression analysis showed that: age ≥ 65（OR = 5.052,95% CI（1.790 ～ 14.254））,overweight or obesity（OR = 4.508,95% CI（1.764 ～ 11.520））,tumor history（OR = 6.891,95% CI（2.667 ～ 17.800））,VTE history（OR = 7.834,95% CI（3.018 ～ 20.336））,four grade surgery（OR = 6.650,95% CI（2.484 ～ 17.803））,heart rate（OR = 1.062,95%CI（1.028 ～1.098））and D dimer（OR = 3.968,95% CI（1.635 ～ 1.69.25））were significantly positively related to the risk of pulmonary embolism（P < 0.05）.Patients’ gender（OR = 0.816,95% CI（0.352 ～ 1.889））,hypertension（OR = 2.321,95% CI（0.911 ～ 5.915））,anesthesia factors（OR = 1.528,95% CI（0.645～ 3.620））were not significantly related to the risk of pulmonary embolism（P > 0.05）.3.The results of multivariate logistic regression analysis with relevant study variables showed that only overweight or obese（OR = 15.156,95% CI（2.578 ～ 89.117））,four grade surgery（OR =33.977,95% CI（4.722 ～ 244.463））,tumor history（OR = 13.532,（2.380 ～ 76.934））,VTE history（OR = 11.863,95% CI（2.213 ～ 63.600））and heart rate（OR = 1.062,95% CI（1.009 ～ 1.118））were independent risk factors for pulmonary embolism（P < 0.05）,and positive results or higher values increased the risk of pulmonary embolism.Multivariate logistic regression analysis excluded mutual interference between variables,and showed that no variables were protective factors for pulmonary embolism.4.According to the results of multivariate logistic regression,the risk factors of pulmonary embolism such as age ≥ 65,overweight or obesity,tumor history,VTE history and heart rate were assigned,and the total score of pulmonary embolism risk was obtained by adding the multiple variables assigned.The maximum value of Yorden index was 0.756,and the sensitivity and specificity were 0.793 and 0.963,respectively.All the patients were divided into two groups: low risk group and high risk group.The risk of PE was significant between high risk group and low risk group（X~2= 55.059,P < 0.001）,and the probability of PE was significantly higher in high risk group than in low risk group.5.The area under the ROC curve of this model is 0.835,which shows that the new model for predicting pulmonary embolism in surgical patients presented in this paper is of medium accuracy and has certain clinical value.However,this model lacks rigorous multicenter external validations.Conclusion1.In this paper,the diagnostic rate of suspected PE patients was 81.1%;The negative predictive value of D dimer alone in the diagnosis of PE was lower than that of D dimer combined prediction model,which was 84.6%.2.After univariate and multivariate logistic regression analysis,it is finally determined that BMI index,surgical factors,VTE history,tumor history and heart rate belong to the independent risk factors for patients experienced surgical treatment for PE.Among the five variables,surgical factors have the greatest influence in occurrence of PE,while heart rate have less influence in occurrence of PE.3.Using BMI index,surgical factors,VTE history,tumor history and heart rate,the clinical prediction model for PE patients has high accuracy,AUC = 0.835;It can provide great help for the preventive diagnosis of PE patients.Whether it can be used in clinical work still needs further external validations. |
PDF Full Text Request