| ObjectiveTuberculosis（TB）is one of the chronic infectious diseases that seriously endanger human health.TB burden is very high in China which is one of the 30 high TB burden countries around the world.Tuberculin pure protein derivative（PPD）skin test is commonly used for TB clinically auxiliary diagnosis and screening,but the specifically is poor.In recent years,on the basis of T cells,T-SPOT test has been rapidly developed due to its high sensitivity and specificity and is used for TB auxiliary diagnosis and screening for Mycobacterium tuberculosis（MTB）infection,but the reagents are expensive,the operation is difficult to automate.So,it is not suitable for TB auxiliary diagnosis in remote areas and TB screening in a large population.Therefore,the development of a skin test reagent with specificity,reasonable price and convenient use is a significant prerequisite for TB control.Based on the current status of tuberculosis in our country and the specific antigens,early secretory antigenic target 6（ESAT6）and culture filter protein 10（CFP10）of MTB,our country has independently developed a recombinant fusion protein ESAT6-CFP10（EC）skin test reagent for clinical trial research.The main purposes of this study were as follows:1.to evaluate the sensitivity and specificity of the EC skin test.2.to evaluate the association and consistency of the diagnosis results of EC skin test,PPD skin test and T-SPOT test in various subjects.3.to evaluate the clinical validity of EC skin test.5.to evaluate the safety of EC skin test.MethodsThe original hypothesis,alternative hypothesis,non-inferiority and superiority cutoff value for the sensitivity difference and specificity difference of EC and PPD skin test,EC skin test and T-SPOT test were set.Subsequently,the tests were repeated to obtain the power,iteratively increased to obtain the sample size.Under the condition of clinical large sample data,the standard deviation of the difference in sensitivity and specificity of the two tests were obtained,the confidence interval was calculated.From October 2015 to March 2018,1090 patients were enrolled.Patients clinical trial were jointly completed by Shanghai Public Health Clinical Center,Tianjin Haihe Hospital,Wuhan Tuberculosis Control Center,Beijing Chest Hospital,First Affiliated Hospital of Chongqing Medical University,Wuxi Fifth People’s Hospital,Fuzhou Pulmonary Hospital,Zhenjiang Third People’s Hospital,Anhui Provincial Hospital and Shenzhen Third People’s Hospital.From January 2016 to June 2016,1564 healthy volunteers were enrolled.The clinical trial of healthy volunteers was completed by Jiangsu Province Center for Disease Control and Prevention.The demographic and clinical characteristics were statistically described in every subjects.Statistical analysis was performed on the differences between age,height,weight,blood pressure and body temperature in the BCG group and placebo group by t-test.Statistical analysis was performed on the differences between gender,ethnicity,combined disease and medication in the two groups of the triple negative population（healthy volunteers with negative T-SPOT test,PPD skin test and EC skin test）by chi-square test/Fisher exact probability test.Calculated the sensitivity,specificity,consistency with clinical diagnosis results,Youden index and Kappa value of induration and erythema as the diagnostic criteria of EC skin test,respectively;calculated the positive detection rate of induration and erythema,and the positive coincidence rate,negative coincidence rate,consistency and Kappa value between two of them;calculated the sensitivity,specificity,and consistency with clinical diagnosis results among the four indicators.Comparison rates in groups were used by X2test.The comparison of EC skin test positive response rates between groups was tested byc2 test.Calculated the sensitivity,specificity,consistency with clinical diagnosis results,Youden index and Kappa value of EC skin test,PPD skin test and T-SPOT test results.Comparison of sensitivity,specificity,and consistency among the three detection tests were used byc2 test.The consistency between two of three tests in different subjects were evaluated.The difference between groups was compared with Mc Nemarc2 test.Subjects were divided into TB and non-TB groups;the paired four tables were constructed.The sensitivities of the two methods were calculated separately in TB group,and the specificity of the two methods were calculated separately in non-TB group.The sensitivity and specificity of the two tests were differentiated,and the standard deviation of the difference was obtained.The credible interval was used to determine the clinical utility of the EC skin test.Results1.Through the sample size simulation of the three-arm test,in terms of sensitivity:when S eB changed between[0.75,0.85],the sample size increases as the values of S eB and SeC increase.On the whole,the sample size required forSeA=0.80 was much larger than that forSeA=0.82 andSeA=0.85.WhenSeA=0.80,SeB=0.75,SeC=0.80,the test power was the largest,about 0.997,whenSeA=0.80,SeB=0.85,SeC=0.85,the test power was the smallest,about 0.695.In terms of specificity:the sample size increases with the increase of S pA and S pB.WhenSpA=0.90 andSpB=0.97,the required sample size was the largest,which was 1048;whenSpA=0.95andSpB=0.90,the required sample size was the smallest,which was 46.After obtaining the real data of large clinical samples,the credible interval method was relatively simple to evaluate the clinical effectiveness.If the lower limit of the confidence interval was greater than the non-inferiority threshold,it can be concluded that the new diagnostic test was not inferior to the traditional diagnostic test;the lower limit of the confidence interval was greater than the superiority threshold,then it can be obtained the new diagnostic test was superior to the traditional diagnostic test.2.The sensitivity of using erythema as the diagnostic criterion of EC skin test was higher than that of using induration as the diagnostic criterion（P<0.01）;in healthy people,24 h after skin test,the specificity of erythema as the diagnostic criterion of the skin test was lower than that of using induration as the diagnostic criterion（P<0.01）,48 h after skin test,there was no statistical difference between the specificity of erythema and induration as diagnostic criterion respectively（P=0.23）;in the BCG vaccination population,24 h（P=0.18）and 48 h（P=0.32）after skin test,there were no differences between the specificity of erythema and induration as diagnostic criterion respectively.In tuberculosis patients,the consistency of diagnosis results between induration and erythema at 48 h and 72 h after skin test was higher than that of 24 h（P<0.01）;in patients with non-tuberculous lung diseases（P=0.60）,healthy volunteers（P=0.11）and BCG vaccinated population（P=0.49）,there was no statistical difference in the consistency of diagnosis results between induration and erythema at each time point after skin test.In tuberculosis patients,the sensitivities of erythema used alone and parallel used of induration or erythema as diagnostic indicator were higher（P<0.01）;in patients with non-tuberculous lung diseases,the combined use of induration and erythema as diagnostic indicator had the highest specificity（P<0.01）,in healthy people（P=0.27）and BCG vaccination population（P=0.42）,there was no statistical difference in the specificity of the four diagnostic indicators;in tuberculosis patients and BCG vaccination population,erythema used alone and parallel use of induration or erythema as diagnostic indicator had a higher consistency with the clinical diagnosis results（P<0.01）.3.In the non-tuberculous population,the positive reaction rate of EC skin test was significantly lower than that of in tuberculosis patients（P<0.01）;in all types of tuberculosis patients,the positive reaction rate of EC skin test was significantly higher than that of in patients with non-tuberculous lung diseases（P<0.01）;in all types of tuberculosis patients,the positive reaction rate of EC skin test was significantly higher than that of in BCG vaccination population（P<0.01）;in patients with bacteria-negative tuberculosis,the positive reaction rate of EC skin test was lower than that of in patients with bacteria-positive tuberculosis（P<0.01）;there was no statistical difference in the positive reaction rate of EC skin test between patients with different treatment types（P=0.86）.4.In tuberculosis patients,the sensitivity of the three detection tests was very high,the sensitivity of the three tests was not statistically different（P>0.05）;in healthy people,especially in the BCG vaccination population,the specificity of the EC skin test was very high,the specificity of the PPD skin test was significantly lower than that of EC skin test and T-SPOT detection（P<0.01）;EC skin test,T-SPOT test with clinical diagnosis results had a higher consistency,respectively;the consistency of PPD skin test with clinical diagnosis results was lower than that of EC skin test,T-SPOT test with clinical diagnosis results respectively（P<0.01）.5.In all types of TB patients,the consistency was about 90.00%between two of the three tests,there were no statistically significant differences in consistency between two of the three tests（P>0.05）.In BCG vaccination populations,the consistency between EC skin test and T-SPOT test was 94.94%,and there was no statistical difference in consistency between the two detection tests（both P>0.05）.In non-TB lung other diseases patients,healthy volunteers and BCG vaccination population,there were a statistically significant differences in consistency between EC and PPD skin test,PPD skin test and T-SPOT test（P<0.01）.6.In terms of sensitivity,the sensitivity of the EC skin test was not inferior to the T-SPOT test and the PPD skin test.In terms of specificity,the specificity of the EC skin test was not inferior to the T-SPOT test,and was superior to the PPD skin test.7.In all subjects,the adverse reactions after EC skin test were mainly local reactions at the injection site,most of which were itching and pain,and the adverse reactions were mostly mild,and no serious adverse events occurred.Conclusions1.We used the three-arm test method to simulate the sample size of clinical trial and calculate the test power.According to reports in the literature,the method of determining the clinical validity of diagnostic tests was simplified under the condition of clinical large samples.The lower limit of the confidence interval of the difference between the sensitivity（or specificity）of the two diagnostic tests was greater than the non-inferiority threshold（or the superiority threshold）,then the clinical non-inferiority（or superiority）conclusion can be drawn.2.The sensitivity and specificity of the EC skin test were both high,and the consistency with the clinical diagnosis results was well;the consistency of the EC skin test and the T-SPOT test results was high;in terms of sensitivity,the EC skin test was not inferior to the PPD skin test and T-SPOT test,in terms of specificity,EC skin test was not inferior to T-SPOT test and superior to PPD skin test;the safety was good.EC skin test could replace T-SPOT and PPD skin test for screening of Mycobacterium tuberculosis infection,auxiliary diagnosis and differential diagnosis of tuberculosis. |
PDF Full Text Request