The Significance Of Early Diagnosis Of AMI With Dynamic Detection Of High-sensitivity Cardiac Troponin I And An Investigation On The Laboratory Diagnostic Mode Of AMI

Objective: To study the clinical value of hs-cTnI in the early diagnosis of acuteAMI as well as the mode of laboratory on AMI diagnosis.Methods: The data of0h and3h serum from207suspected cases of ACS withchest pain between August2011and November2011. The serum was separated in1hour and stored under-20℃for analysis, freezing and thawing once. The patients weredivided into AMI group and non-AMI group; according to the2007ESC and ACCstandard on AMI diagnosis, the patients receiving the definite diagnosis were AMIgroup, the rest were non-AMI group. The AMI group contained51cases, between theage of43and72, out of which33cases were male patients and18cases were female.The non-AMI group contained156cases, among which there were97males and59females between the age of46and73. The experimental method: the concentration ofhs-cTnI in all specimens was tested; the mean, the lowest median, the minimum valueand the maximum value of the two groups were calculated; the rate of hs-cTnIconcentration change within3h of hospital admission was tested. The rate ofconcentration change=(the hs-cTnI concentration value at3h of admission–thehs-cTnI concentration value at0h of admission)/the hs-cTnI concentration value at0hof admission. Through the SPSS16.0statistical analysis software, ROC curves weredrawn to analyze the characteristics of diagnosing AMI with hs-cTnI on hospitaladmission. According to Youden exponent, the best threshold and the suspected rangeof clinical judgment were identified. By comparison with the given reference by thevendor, the differences of the rate of change between the two groups were comparedusing Kruskal-Wallis method. Kappa was adopted to test and verify the clinicalconsistency between the values from the clinical judgment and the threshold from the vendor’s reference. AMI was diagnosed according to the ROC curve combined with thesuspected range resulted from the clinical judgment plus the best judgment point ofdynamic variation, and the clinical application of AMI diagnosis rules were compared:The hs-cTnI above the vendor’s normal reference value on0h hospital admission wasdiagnostic AMI rule A; the hs—cTnI above the vendor’s diagnostic threshold on0hhospital admission was rule B; the hs-cTnI above the vendor’s normal reference valueon3h hospital admission was rule C; the hs—cTnI above the vendor’s diagnosticthreshold on3h hospital admission was diagnostic AMI rule D, the diagnosis based onthe combination of the suspected range and the rate of change was rule E. The diagnosisof AMI was conducted towards the emergent patients with chest pain. The value ofsensitivity，specificity， accuracy，positive predicative, and negative predicative werecalculated to compare the diagnostic characteristics.Result: AMI group: the mean of0h hs—cTnI was1.616ng/ml; the median was0.48ng/ml, the maximum value was20.6ng/ml and the minimum value was0.008ng/ml.The mean of3h hs—cTnI was2.13ng/ml, the median was0.87ng/ml, the maximumvalue was26.3ng/ml and the minimum value was0.053ng/ml. The mean of thechanging rate between the two time points was138.5%(2.3%-1377.8%); the medianwas67.16%. Non-AMI group: the mean of0h hs—cTnI was0.031ng/ml; the medianwas0.015ng/ml; the maximum value was0.35ng/ml and the minimum value was0.005ng/ml. The mean of3h hs—cTnI was0.035ng/ml; the median was0.018ng/ml;the maximum value was0.34ng/ml and the minimum value was0.005ng/ml.The meanof changing rate between the two time points was21.8%(-28.6%-400.0%), the medianwas16.7%. There was significance in the rate of change between the two time points（P<0.01）. According to the analysis of the ROC curves of0h hs—cTnI, the area underthe curve (AUC) was0.916, the diagnosis value was high. According to the Youdenindex, the best diagnostic value was0.0905ng/ml, based on which, the sensitivitydegree was78.4%, the specific degree was92.9%, the accuracy was89.4%, the positivepredictive value was78.4%, the negative predictive value was92.9%. The result wasclose to the threshold0.1ng/ml given by the vendor. When the diagnostic boundaryvalue was0.1ng/ml, the sensitivity value was76.5%, the specific degree was92.9%,the accuracy was88.9%, the positive predictive value was78%, and the negativepredictive value was92.3%. The Kappa value was0.954, which was well above0.75and close to1. It suggests that the consistency of the two values was very high inclinical diagnosis. It can be seen from the curves that, taking the normal reference range 0.04ng/ml as the sensitivity value with diagnostic boundary value, the sensitivity valuewas82.4%, the specificity was78.2%, the accuracy was79.2%, the positive predictivevalue was55.2%, and the negative predictive value was93.1%. When based on thesingle diagnostic value there would be an overlapping area of the indicators betweenAMI patients and non-AMI patients, which would result in misdiagnosis. Therefore,according to ROC curve, the misdiagnosis rate might be decreased if two diagnosticpoints was set in the overlapping area, and the original two distributions, i.e. thepositive and the negative could be redivided into three distributions, i.e. the positive, thenegative, and the suspected. According to the ROC curves,0.04ng/ml was taken as theminimum level of the suspected range and0.355ng/ml the maximum, and the optimalrate for judgment was35%when the rate of concentration change was analyzed.According to the analysis, AMI was excluded if the3h value was below minimum level,whereas the patients could be diagnosed as having AMI if the value was above themaximum level. If it was within the suspected range, the concentration change rate of3h can be relied on. When the new judgment method was employed to diagnose AMI,the sensitivity level was98%, the specificity level was97.4%, the accuracy was97.6%,the positive predictive value was92.6%, and the negative predictive value was99.3%.The accuracy of the AMI diagnosis is higher than those based on the single value.According to the consistency analysis on the results obtained from rule E and fromclinical diagnosis, the Kappa indicator was0.937, which was far above0.75and closeto1. This suggested a high consistency and a high application value according to rule Efor screening diagnosis of AMI.Conclusion: As the patients with chest pain were on admission to the emergencydepartment, the continuous dynamic hs-cTnI value should be tested. The diagnosis ofAMI through the combination of setting the suspected range and referring to thehs-cTnI change rate is better than the AMI diagnosis with single point judgment value.The suspected range could be set using ROC curve. This would effectively enhance theearly diagnosis of AMI with hs-cTnI, and the number of suspected AMI patients wouldalso be decreased due to the improvement of hs-cTnI testing methods as well as theimprovement in accuracy. The number of suspected ACS cases without ECGabnormalities and indefinite diagnosis could also be more effectively confirmed.However, a great many clinical researches are in need for the assessment of the exactvalue of dangerous suspected range.