| Objective:Pulmonary embolism(PE)is a general term for a group of diseases or clinical syndromes caused by various emboli blocking the pulmonary artery system,including pulmonary thromboembolism,fat embolism syndrome,amniotic fluid embolism,air embolism,etc.PE is a life-threatening disease with atypical clinical manifestations.Lung imaging suggests filling defects in lung vessels as the gold standard for diagnosis.However,excessive lung imaging examination may increase exposure to radiation and contrast agents.Multiple PE prediction criteria such as the Wells score,the simplified Wells score,the age-adjusted D-dimer assessment,the modified Geneva score,the YEARS score,and the pulmonary embolism rule out Criteria(PERC)have been developed in adults to assess PE and reduce overtesting.Current rules for predicting the assessment of PE in children are almost entirely based on adult studies and have poor adaptability in children.Also,children are more vulnerable to ionizing radiation than adults.Therefore,a more appropriate method for the predictive evaluation of PE in children is needed to reduce children’s exposure to radiation or contrast media.In this study,a retrospective analysis was performed on 192 patients under 18 years of age who were clinically suspected of PE and underwent lung enhanced CT examination or computed tomography pulmonary angiography(CTPA)examination.The previous history,clinical symptoms,signs and auxiliary examinations of PE positive patients and PE negative patients were analyzed and compared.Clinical characteristics of PE positive patients and PE negative patients were described,high risk factors of PE positive patients were found,and clinical probability scoring system related to PE in children was developed to provide reference for the diagnosis and evaluation of PE in children.Methods:A total of 192 cases of children under 18 years of age who were hospitalized in our hospital from January 2018 to December 2022 and received CTPA or enhanced lung CT examination due to suspected pulmonary embolism were continuously collected,and their general status,clinical manifestations,laboratory data and imaging data were sorted out.SPSS 25.0 software was used to analyze and plot the data.Results:1.Comparison of basic data192 subjects were included in the study.Among them,95 cases(49.5%)were male and 97(50.5%)were female.According to the presence or absence of PE,the patients were divided into PE positive group and PE negative group,with 16 PE positive patients(8.3%)and 172 PE negative patients(91.7%).In the PE positive group,there were 9 males(56.2%)and 7 females(43.8%),aged 9 years(6,15),respiratory rate 25.00 beats/min(20.00,37.50)and heart rate 109.50 beats/min(105.25,137.50).PE negative patients were 86(48.9%)males and 90(51.1%)females,aged 8(4-12)years,respiratory rate 24.00 beats/min(20.00,28.00),heart rate 110.00beats/min(100.00.125.00);There were no significant differences in gender,age,respiratory rate and heart rate between the two groups(P>0.05).After adjusting for age,there was a statistically significant difference between the two groups in the presence of hyperrespiration and tachycardia(P<0.01);There was no significant difference in tachycardia(P>0.05).The median D-dimer values of PE positive group and PE negative group were 9.515 ng/L(6.485,19.825)and 2.160 ng/L(0.585,6.188),respectively,and there was significant statistical significance in D-dimer values between the two groups(P<0.001).In addition,the ratio of CVC patients between the two groups was statistically significant(P<0.01).In terms of comorbidity,there were statistically significant differences in nephrotic syndrome and liver injury between the two groups(P<0.01).There was no significant difference in coagulopathy,myocardial injury,respiratory failure,severe pneumonia,pleural effusion,congenital heart disease,active cancer,asthma and other diseases(P>0.05).There was no statistically significant difference between PE positive and PE negative patients in previous PE or DVT,surgery or immobilization in the past four weeks,cough,chest pain,chest tightness,palpitation,hemoptysis,lower limb pain,or upper limb swelling(P>0.05).There were significant differences in dyspnea,upper limb pain,lower limb swelling DVT signs and DVT between the two groups(P<0.05).2.Evaluation value of D-dimer on PE:When D-dimer ≥0.5ng/L was positive for D-dimer,the sensitivity and specificity of D-dimer for diagnosis of pulmonary embolism were 1.0(95%CI,0.76-1.0),0.20(95%CI,0.14-0.27),and 0.10(95%CI,0.06-0.16).The negative predictive value was1.0(95%CI,0.88-1.0),the positive likelihood ratio was 1.25(1.16-1.34),and the negative likelihood ratio was 0(0).The ROC curve analysis of D-dimer alone on PE children suggested that the AUC of PE children was 0.822,95%CI was 0.745-0.900,and the cut-off value was 5.985ng/L according to the Jorden index.If D-dimer≥6.0ng/L was positive for D-dimer,the sensitivity and specificity of D-dimer in the diagnosis of PE were 0.81(0.54-0.95),0.75(0.68-0.81),positive predictive value was0.23(0.13-0.36),and negative predictive value was 0.98(0.93-1.0).The positive likelihood ratio was 3.25(2.30-4.60)and the negative likelihood ratio was 0.25(0.9-0.70).3.Evaluation value of Wells score on PE:When Wells score showed a high possibility of PE,Wells score was applied to diagnose PE.There were 161 patients(83.9%)with positive Wells score and 31patients(16.1%)with negative Wells score.The sensitivity of Wells score diagnostic criteria was 1.0(0.76-1.0).Specificity was 0.18(0.12-0.24),positive predictive value was 0.10(0.06-0.16),negative predictive value was 1.0(0.86-1.0),positive likelihood ratio was 1.21(1.13-1.30),negative likelihood ratio was 0(0).When the heart rate≥100 beats/min was adjusted for age,90 patients(46.9%)had positive Wells score and 102 patients(53.1%)had negative Wells score.The sensitivity and specificity of Wells score were 0.88(0.60-0.98)and 0.57(0.49-0.64)respectively.Positive predictive value was 0.16(0.09-0.25),negative predictive value was 0.98(0.92-1.0),positive likelihood ratio was 2.03(1.58-2.60),negative likelihood ratio was 0.22(0.06-0.81).When Wells score showed little possibility of PE,Wells score was applied to diagnose PE.There were 18 patients(9.4%)with positive Wells score and 174patients(90.6%)with negative Wells score.The sensitivity of Wells score diagnostic criteria was 0.50(0.25-0.75).Specificity was 0.94(0.90-0.97),positive predictive value was 0.44(0.22-0.69),negative predictive value was 0.95(0.91-0.98),positive likelihood ratio was 8.8(4.05-19.12),negative likelihood ratio was 0.53(0.32-0.87).When the heart rate ≥100 beats/min was adjusted for age,9 patients(4.7%)had positive Wells score and 183 patients(95.3%)had negative Wells score.The sensitivity and specificity of Wells score were 0.25(0.08-0.53)and 0.97(0.93-0.99)respectively.Positive predictive value was 0.44(0.15-0.77),negative predictive value was 0.93(0.89-0.96),positive likelihood ratio was 8.8(2.62-29.54),negative likelihood ratio was 0.77(0.88-1.02).4.Evaluation value of PERC score on PEWhen PE was excluded by PERC score,185 patients(96.4%)with high possibility of PE and 7 patients(3.6%)with low possibility of PE were diagnosed.The sensitivity and specificity of PERC score diagnostic criteria were 1.0(0.76-1.0)and 0.04(0.02-0.08).Positive predictive value was 0.09(0.05-0.14),negative predictive value was 1.0(0.56-1.0),positive likelihood ratio was 1.04(1.01-1.07),negative likelihood ratio was 0(0).When heart rate ≥100 beats/min was adjusted for age,there were 142 patients(74.0%)with high probability of PE and 50 patients(26.0%)with low probability of PE.The sensitivity and specificity of the diagnostic criteria of PERC score were 0.75(0.47-0.92)and 0.27(0.21-0.34),respectively.Positive predictive value was 0.08(0.05-0.15),negative predictive value was 0.92(0.85-0.95),positive likelihood ratio was 1.03(0.77-1.39),negative likelihood ratio was 0.91(0.38-2.20).5.Regression analysis of clinical manifestations,examinations and other related factors in children with PEIt was found that when DVT signs,hemoptysis,chest pain,unilateral lower limb edema,upper limb pain,lower limb pain,CVC,dyspnea,apnea(age adjusted),and D-dimer ≥6.0ng/L were included in stepwise logistic regression analysis,dyspnea,DVT signs,D-dimer ≥6.0ng/L were independently correlated with the diagnosis of pediatric PE.Based on the binary logistic regression analysis of the three variables,Hosmer-Lemeshaw test P=0.953,and the goodness of fit effect of the model is good.The clinical prediction rule consisting of dyspnea,DVT signs and D-dimer ≥6.0ng/L was named Model 1,which included 1 or more of the 3 variables,and PE was highly likely.Model 1 could identify 15 out of 16 patients with pulmonary embolism with sensitivity of 93.8%(95%CI,0.68-1.0),specificity of 61.9%(95%CI,0.54-0.69),positive predictive value of 0.18(95%CI,0.11-0.29),and negative predictive value of0.99(95%CI,0.68-1.0).0.94-1.0),positive likelihood ratio 2.46(95%CI,1.96-3.09),negative likelihood ratio 0.10(95%CI,0.02-0.68).The ROC curve analysis of Model1 on children with PE suggested that the AUC of children with PE was 0.895,and the95%CI was 0.802-0.987,suggesting that Model 1 had good efficacy in evaluating the diagnosis of PE.When DVT,hemoptysis,chest pain,unilateral lower limb edema,upper limb pain,lower limb pain,CVC,dyspnea,apnea(age adjusted),and D-dimer ≥6.0ng/L were included in the analysis,upper limb pain,DVT,and D-dimer ≥6.0ng/L were independently associated with the diagnosis of childhood PE when included in stepwise backward logistic regression analysis.By binary logistic regression analysis of the three variables,Hosmer-Lemeshaw test P=0.710,and the goodness of fit effect of the model is good.The clinical prediction rule consisting of upper limb pain,DVT and D-dimer ≥6.0ng/L was named Model 2,which included 1 or more of the 3variables,and PE was highly likely.Model 2 could identify 14 of 16 PE patients with sensitivity of 87.5%(95%CI,0.60-0.98),specificity of 71.0%(95%CI,0.64-0.77),positive predictive value of 0.21(95%CI,0.12-0.34),and negative predictive value of0.98(95%CI,0.95).0.94-1.0),positive likelihood ratio 3.02(95%CI,2.25-4.06),negative likelihood ratio 0.18(95%CI,0.05-0.64).The ROC curve analysis of Model2 on children with PE suggested that the AUC of children with PE was 0.877,and the95%CI was 0.766-0.987,suggesting that model 2 was effective in evaluating the diagnosis of PE.Conclusion:1.When the cut-off value of D-dimer was 6.0ng/L,the specificity was significantly improved compared with that of 0.5ng/L,and the sensitivity was more than 80%.2.The specificity of Wells score increased significantly after adjusting for age.The specificity of PERC score was low in children with PE regardless of age adjustment.3.In Model 1 and Model 2,more than 50% of children with suspected pulmonary embolism could stop further diagnostic imaging,far greater than the PERC score,which could significantly reduce the radiation and contrast agent exposure of children with suspected PE. |
PDF Full Text Request