| Objective:To evaluate the predictive value of differrent departments for pulmonary embolism(Wells score,modified Geneva score and Caprini score)in departments of high incidence of pulmonary embolism.compare the advantages of different predictive scales in different departments of high incidence of pulmonary embolism,and explore the selection of more valuable clinical possibility assessment forms by comparing the sensitivity,specificity and coincidence rate of different scales for detection of pulmonary embolism,so as to improve the detection rate of pulmonary embolism and reduce the mortality rate.Method:From February 2012 to February 2022,all patients who were admitted to the Department of Respiratory,Orthopedics,Oncology and Intensive Care of the First Affiliated Hospital of Dalian Medical University for diagnosis of pulmonary embolism were collected,and the cases were screened according to the inclusion criteria and exclusion criteria.The clinical data and computed pulmonary angiography(CTPA)data of the selected patients were complete,and the final clinical diagnosis was obtained through CTPA examination.Collect clinical data and conduct retrospective analysis,including high-risk factors(medical history),symptoms,signs,and auxiliary examinations(ECG,blood gas analysis,D-dimer,chest X-ray).The patients were divided into four groups according to different departments,namely,respiratory department,orthopedics department,oncology department and intensive care department.The patients in each group were given Wells score,modified Geneva score and Caprini score respectively,which were independently judged by two doctors.If they disagreed,the final result was decided by the doctor or department director with senior title.All patients were evaluated for the clinical possibility of pulmonary embolism.The Wells score,the modified Geneva score and the Caprini score were independently evaluated by two doctors.In case of disagreement,the final result was determined by the doctor with senior title or the department director.For patients in each department,three scoring methods were used to count the diagnosis of pulmonary embolism with different scoring scales.The patients were classified as true positive,false positive,false negative and true negative cases according to the gold standard(CTPA results),and the sensitivity,specificity and accuracy of various scales were calculated to evaluate pulmonary embolism.The receiver operating characteristic(ROC)curve was used to evaluate the predictive value of different scoring scales for pulmonary embolism in different departments.The area under the curve(AUC)and 95% confidence interval(CI)were calculated to evaluate the area under the curve.Z test was used to compare the area under the curve in pairs,and the predictive value of each scoring method in different departments with high incidence of pulmonary embolism: respiratory department,orthopedics department,oncology department,and intensive care department was analyzed and evaluated.Results:1.Finally,1683 effective cases were included,including 794 males and 889 women;The average age was(67.49 ± 3.68),and the age was 21-101 years.1108 of 1683 patients with suspected acute pulmonary embolism were diagnosed as pulmonary embolism by CTPA after admission,with a diagnosis rate of 65.8%.2.Comparison of predictive value of Wells score,modified Geneva score and Caprini score for acute pulmonary embolism in respiratory department: 786 of 1273 patients were diagnosed with pulmonary embolism in respiratory department.The clinical evaluation of the patients with suspected pulmonary embolism was 412 cases with low PE possibility and 861 cases with PE possibility by using the two classification method of Wells score.The sensitivity,specificity and accuracy of diagnosing pulmonary embolism were 78.63%,50.10% and 67.71% respectively;Using the modified simplified geneva score two-classification method,704 patients with suspected pulmonary embolism were clinically assessed as less likely to PE and 569 as possible to PE.The sensitivity,specificity and accuracy of diagnosing pulmonary embolism were 58.78%,77.82% and 66.06% respectively;Using the Caprini scoring method,low risk was determined as low possibility of PE,medium risk,high risk and extremely high risk were determined as PE possibility.The clinical evaluation of suspected pulmonary embolism patients was 549 cases with low possibility of PE and724 cases with PE possibility.The sensitivity,specificity and accuracy of diagnosis of pulmonary embolism were 80.15%,80.70% and 80.36% respectively.We can see that the sensitivity,specificity and accuracy of the Caprini score method in the diagnosis of pulmonary embolism patients in respiratory department are higher than Wells score and modified Geneva score.Comparison of the clinical predictive value of the simplified Wells score,the modified simplified Geneva score and the Caprini score for PE in respiratory patients.The simplified Wells score AUC is about 0.579,the revised Geneva score AUC is about0.624,and the Caprini score AUC is about 0.826.The Z test shows that the Caprini score has a better predictive value than the simplified Wells score(Z=3.052,P=0.013)and the revised Geneva score(Z=2.125,P=0.034).The difference is statistically significant.3.Comparison of predictive value of Wells score,modified Geneva score and Caprini score for acute pulmonary embolism in orthopedics: among 40 patients to be diagnosed in orthopedics,the simplified version of Wells score dichotomous classification method showed that 9 patients were less likely to have PE and 31 patients were likely to have PE.The sensitivity,specificity and accuracy of diagnosing pulmonary embolism were 87.50%,62.50% and 82.50% respectively;Using the modified simplified geneva score two-classification method,the clinical evaluation of suspected pulmonary embolism patients was that PE was less likely in 13 cases and PE was possible in 27 cases.The sensitivity,specificity and accuracy of diagnosis of pulmonary embolism were 75.00%,62.50% and 72.50% respectively;Using the Caprini scoring method,the low risk is determined as PE with little possibility,and the medium risk,high risk and extremely high risk are determined as PE with possibility.The clinical evaluation of suspected pulmonary embolism was that PE was less likely in 7cases and PE was possible in 33 cases.The sensitivity,specificity and accuracy of diagnosis of pulmonary embolism were 96.88%,75.00% and 92.50% respectively.We can see that the sensitivity,specificity and accuracy of the Caprini scoring method in the diagnosis of patients with orthopedic pulmonary embolism are higher than the Wells score and the modified Geneva score.Comparison of the clinical predictive value of the simplified Wells score,the modified simplified Geneva score and the Caprini score on the PE of orthopedic patients: the simplified Wells score AUC is about 0.604,the modified Geneva score AUC is about 0.642,and the Caprini score AUC is about 0.736.The Z test shows that the predictive value of the Caprini score is better than the simplified Wells score(Z=2.143,P=0.025)and the modified Geneva score(Z=1.982,P=0.042),with a statistically significant difference.4.Comparison of predictive value of Wells score,modified Geneva score and Caprini score on acute pulmonary embolism in oncology department: Among 295 patients proposed to be diagnosed in oncology department,the simplified version of Wells score dichotomous method showed that 85 patients were less likely to have PE and 210 patients were likely to have PE.The sensitivity,specificity and accuracy of diagnosing pulmonary embolism were 91.90%,80.00% and 88.47% respectively;Using the modified simplified geneva score dichotomy,73 patients with suspected pulmonary embolism were less likely to have PE and 222 patients with PE.The sensitivity,specificity and accuracy of diagnosing pulmonary embolism were 93.81%,82.35% and90.51% respectively;Using the Caprini scoring method,the low risk is determined as PE with little possibility,and the medium risk,high risk and extremely high risk are determined as PE with possibility.The clinical evaluation of suspected pulmonary embolism was that 71 cases were less likely to have PE and 224 cases were likely to have PE.The sensitivity,specificity and accuracy of diagnosis of pulmonary embolism were 89.52%,57.65% and 80.34% respectively.It can be found that the sensitivity,specificity and accuracy of the modified Geneva score in the diagnosis of pulmonary embolism patients in oncology department are higher than those of Wells score and Caprini score.Comparison of the clinical predictive value of the simplified Wells score,the modified simplified Geneva score and the Caprini score on PE of patients in oncology department: the simplified Wells score AUC is about 0.754,the modified Geneva score AUC is about 0.872,and the Caprini score AUC is about 0.678.Z test shows that the predictive value of the Geneva score is better than the simplified Wells score(Z=1.991,P=0.046)and the Caprini score(Z=2.431,P=0.019),with a statistically significant difference.5.Comparison of predictive value of Wells score,modified Geneva score and Caprini score for acute pulmonary embolism in intensive care department: among 75 patients proposed to be diagnosed in intensive care department,the simplified version of Wells score dichotomous method showed that 47 patients were less likely to have PE and 28 patients were likely to have PE.The sensitivity,specificity and accuracy of diagnosing pulmonary embolism were 56.25%,76.74% and 68.00% respectively;Using the modified and simplified geneva score two-classification method,48 patients with suspected pulmonary embolism were less likely to have PE and 27 patients with PE.The sensitivity,specificity and accuracy of diagnosing pulmonary embolism were43.75%,69.77% and 58.67% respectively;Using the Caprini scoring method,the low risk is determined as PE with little possibility,and the medium risk,high risk and extremely high risk are determined as PE with possibility.The clinical evaluation of suspected pulmonary embolism was that PE was less likely in 40 cases and PE was possible in 35 cases.The sensitivity,specificity and accuracy of diagnosis of pulmonary embolism were 50.00%,55.81% and 53.33% respectively.It can be found that the sensitivity,specificity and accuracy of using Wells score to diagnose patients with pulmonary embolism in ICU department are higher than the modified Geneva score and Caprini score.Comparison of the clinical predictive value of the simplified Wells score,the modified simplified Geneva score and the Caprini score for PE in patients with severe illness: the simplified Wells score AUC is about 0.801,the modified simplified Geneva score AUC is about 0.742,and the Caprini score AUC is about 0.721.Z test shows that the predictive value of the Wells score is better than the modified simplified Geneva score(Z=0.014,P=0.017)and the Caprini score(Z=1.998,P=0.039),The difference was statistically significant.Conclusion:1.Wells score,modified Geneva score and Caprini score have important value in predicting acute pulmonary embolism.Predicted sensitivities in orthopedics and oncology reached over 75%.The sensitivity and specificity of respiratory and intensive care departments also exceeded or approached 50%.2.Geneva score has better predictive value for patients with pulmonary embolism in oncology department;Wells score in intensive care department is more advantageous;and Caprini score is more suitable for the possibility assessment of pulmonary embolism in orthopedics and respiratory patients.3.The adoption of a more suitable possibility assessment scales in different departments can effectively improve the detection rate,avoid missed diagnosis,and provide reference for patient screening and follow-up treatment. |
PDF Full Text Request