Application Evaluation Of Hepatic Function Test Model In Initial Onset Hyperthyroidism

Purpose: Establishing a screening model of hepatic function tests for newly diagnosed patients with hyperthyroidism so as to not only provide a reference base for early diagnosis of hyperthyroidism but also guiding suggestions for the clinical diagnosis of hyperthyroid liver damage.Methods:(1) Evaluation of routine hepatic function tests taken from patients with initial onset hyperthyroidism and the establishment and validation of screening model. Thirteen test items of total bilirubin(TB), direct bilirubin(DB), indirect bilirubin(FB), total protein(TP), albumin(propagated), globulin(GLOB), the ratio of albumin and globulin(A/G), glutamic acid transaminase(ALT), aspartate aminotransferase(AST), γ-glutamyl transferase(GGT), total bile acid(TBA), lactate dehydrogenase(LDH), alkaline phosphatase(ALP) were obtained after collecting 352 medical records of hyperthyroidism and 434 physical examination results from normal people,. Then serum of the 120 patients and the 146 normal people was collected to test all 13 indicators above. By comparing the areas under the ROC curve(AUC), evaluation was made of each indicator’s value to the first diagnosed patients with hyperthyroidism. The 352 patients were made as exercise group to establish the routine hepatic function test model of the patients first diagnosed with hyperthyroidism, using logistic regression statistical method. People whose serum was retained were made as the validation group to validate the model in order to evaluate its diagnostic value.(2)Evaluation of the hepatic function tests which have not yet been universally adopted. Each of the retained serum of the 120 patients first diagnosed with hyperthyroidism and the 146 normal people who had taken physical examination was tested the 6 hepatic function indicators [5- nucleotidase(5- NT), α- fucosidase(AFU), monoamine oxidase(MAO), leucine aminopeptidase(LAP), aspartate amino transferase isozymes(m- AST), glutamate dehydrogenase(GLDH)]which have not yet been universally adopted. By comparing the areas under the ROC curve(AUC), evaluation was made of the 6 indicator’s value to the first diagnosis of patients. Establish a combined diagnostic model using the routine hepatic function tests and the 6 non-universally adopted tests to have the model’s AUC tested, also using the logistic regression statistical method. Then make a comparison between the combined model and the routine model and explore the 6 tests’ correlation with the routine hepatic function tests- thus providing reference for choosing hepatic function tests for clinical use.(3) Discussion of Leucine aminopeptidase(LAP) as an independent diagnostic indicator of hyperthyroidism. Validation group with 120 hyperthyroidism cases were divided into two groups-one with liver damage and one without. Compare the mean value of LAP and the variance in those groups, and study the correlation of the LAP with thyroid hormone.Results:(1) Among the routine hepatic function tests, the top three ROC curve(AUC) in screening the patients with initial onset of hyperthyroidism in order are ALP(0.884), ALT(0.780) and GGT(0.702).The auxiliary diagnostic model of the hepatic function of the patients with initial onset of hyperthyroidism was constructed using the combined factors of DB, TP, GLOB, ALT, LDH and ALP. AUC of the exercise group was 0.937 and AUC of the validation group was 0.958, with the overall difference being 2.3%; when positioning diagnosis boundary value at 0.4, diagnostic accuracy of the exercise group was 90.2% and 90.5% for the validation group. The κ coefficient as analyzing the consistency of model diagnosis and clinical diagnosis reached 0.807.(2) Among the not-yet universalized hepatic function tests, the largest AUC in diagnosing initial onset hyperthyroidism is LAP, reaching 0.903. The five other indicators are relatively low in diagnostic value. The combined hepatic function diagnostic model consists of such eight items as TB, FB, TP, ALT, AST, ALP and LAP and 5 – NT with the model’s AUC standing at 0.982. The pairs of indicators with the highest correlation coefficient between the routine hepatic function tests and the 6 not yet universalized tests are: GLDH and AST(r = 0.436, p < 0.01); M- AST and AST(r = 0.742, p < 0.01); LAP and ALP(r = 0.497, p < 0.01); MAO and GGT(r = 0.245, p < 0.01); AFU and ALP(r = 0.221, p < 0.01); 5,-NT and GGT(r = 0.537, p < 0.01).(3) The LAP value showed no variance in the validation group of hyperthyroidism, the group of hyperthyroidism with liver damage and the group of hyperthyroidism without liver damage. But the statistical difference(p > 0.05) all turn up when comparing them with the LAP value of the routine group. The correlation coefficient between LAP and FT3, FT4 and TSH reaches 0.628, 0.640, and 0.743 respectively.Conclusion:(1) In the diagnosis of the initial onset hyperthyroidism, routine hepatic function tests have certain diagnostic value, of which ALP as the highest; while good application value is found in establishing the regression model consisting indicators of DB, TP, GLOB, ALT, LDH and ALP.(2) In liver function tests which have not yet being universally adopted, only LAP is found to possess high diagnostic value. Therefore when judging initial onset hyperthyroidism, five other tests can be left out to save cost and time. Although the AUC of the combined model shows moderate rise compared to the diagnostic model based on the routine tests, it doesn’t have a big advantage under comprehensive evaluation. So the latter can satisfy completely the requirements of clinical screening of hyperthyroidism.(3) The rise of LAP in the case of initial onset hyperthyroidism is caused by the disease itself, without anything to do with the liver’s damage. So the LAP is suggested to be taken as an independent risk indicator in the diagnosis of initial onset hyperthyroidism in the hope of catching clinical attention.