RESUMO
Objective To investigate the clinical applicability and different characteristics of three commonly used diagnostic methods for drug-induced liver injury from the two aspects of liver injury induced by Western medicine and liver injury induced by traditional Chinese medicine. Methods A prospective cohort study was performed for 289 hospitalized patients with acute drug-induced liver injury who were admitted to The Fifth Medical Center of Chinese PLA General Hospital from January 2015 to December 2020 and did not receive integrated traditional Chinese and Western medicine therapy, among whom 187 patients had herb-induced liver injury and 102 had Western medicine-induced liver injury. The 289 patients were diagnosed by the integrated evidence chain (IEC), Roussel Uclaf Causality Assessment Method (RUCAM), and the Structured Expert Opinion Process (SEOP) method, and related data at acute onset were collected, including general information, latency period, detailed medication, and laboratory markers such as alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl transpeptidase, alkaline phosphatase, and total bilirubin. A statistical analysis was performed to investigate the consistency between IEC, RUCAM, and SEOP in the diagnosis of Western medicine-induced liver injury and herb-induced liver injury and their own applicability. The Kruskal-Wallis H test was used for comparison of non-normally distributed continuous data; the chi-square was used for comparison of categorical data. Results The hepatocellular type was the main type of clinical liver injury in both Western medicine-induced liver injury and herb-induced liver injury, accounting for 81.4% and 74.3%, respectively, and laboratory examination showed higher levels of ALT and AST. Western medicine-induced liver injury cases were diagnosed by IEC, RUCAM, and SEOP, with a clinical diagnosis rate of 65.7%, 100%, and 63.7%, respectively, and the constituent ratio of Western medicine-induced liver injury was 23.2%, 35.3%, and 22.5%, respectively. Herb-induced liver injury cases were diagnosed by these three methods, with a clinical diagnosis rate of 47.6%, 100%, and 29.9%, respectively, and the constituent ratio of herb-induced liver injury was 30.8%, 64.7%, and 19.4%, respectively. The consistency test of the three diagnostic methods showed that in the diagnosis of Western medicine-induced liver injury cases, there was good consistency between IEC and SEOP (Kappa=0.785, P 0.05) and between RUCAM and SEOP (Kappa=0.117, P > 0.05); in the diagnosis of herb-induced liver injury cases, there was poor consistency between RUCAM and SEOP (Kappa=0.066, P > 0.05), while there was good consistency between RUCAM and IEC (Kappa=0.026, P < 0.05) and between IEC and SEOP (Kappa=0.437, P < 0.05). Conclusion The IEC method shows good applicability for both Western medicine-induced liver injury and herb-induced liver injury, and there is good consistency between IEC and SEOP in the diagnosis of Western medicine-induced liver injury cases, while there is a relatively low level of consistency between IEC and SEOP in the diagnosis of herb-induced liver injury. There is poor consistency between RUCAM and the other two methods. In the clinical diagnosis of Western medicine-induced liver injury, IEC, RUCAM, and SEOP should be used in combination to accurately judge the causal relationship between drugs and liver injury.
RESUMO
Objective To investigate the influencing factors for the prognosis of adult patients with chronic drug-induced liver injury (DILI). Methods A total of 255 patients who were diagnosed with chronic DILI by liver biopsy in The Fifth Medical Center of Chinese PLA General Hospital from January 2014 to December 2018 were enrolled, and according to the liver function after 2 years, they were divided into non-recovery group and recovery group. The two groups were analyzed in terms of the clinical data including age, sex, body mass index, types of drugs used, type of DILI injury, severity of DILI injury, underlying diseases, laboratory markers, liver histology, and 2-year prognosis. The t -test was used for comparison of normally distributed continuous data between two groups, and the Mann-Whitney U test was used for comparison of non-normally distributed continuous data between two groups; the chi-square test was used for comparison of categorical data between two groups. Univariate and multivariate logistic regression analyses were used to investigate the independent risk factors for the prognosis of chronic DILI. Results After 2 years of follow-up, 195 patients (76.5%) achieved the recovery of liver function, while 60 patients (23.5%) did not achieve such recovery. There were significant differences between the two groups in the type of DILI injury ( P =0.028), the proportion of patients with diabetes ( P =0.048), and the degree of liver fibrosis ( P 2×upper limit of normal (ULN) ( OR =3.080, 95% CI : 1.331-7.127, P =0.009) were independent risk factors for the prognosis of chronic DILI. Conclusion When patients meet the diagnostic criteria for chronic DILI, the independent risk factors PLT 2×ULN may be used to screen out the patients who are more likely to have poor prognosis.
RESUMO
Objective To investigate the association between baseline IgM level and treatment response to ursodeoxycholic acid (UDCA) in patients with primary biliary cholangitis (PBC). Methods A retrospective analysis was performed for the clinical data of 637 PBC patients who were diagnosed and treated with UDCA for the first time in The Fifth Medical Center of Chinese PLA General Hospital from January 2010 to January 2020. The PBC patients were divided into UDCA complete response group with 436 patients and UDCA poor response group with 201 patients, and baseline clinical data were compared between the two groups. According to the optimal cut-off value of IgM determined by the area under the ROC curve (AUC) of baseline indices in predicting the risk of poor treatment response, the patients were divided into IgM ≥1.5×ULN group and IgM < 1.5×ULN group, and baseline parameters, treatment response, and prognostic model score were compared between groups. The t -test was used for comparison of normally distributed continuous data between two groups, and the Mann-Whitney U test was used for comparison of non-normally distributed continuous data between two groups; the chi-square test was used for comparison of categorical data between two groups. The Cochran-Mantel-Haenszel test was used for subgroup analysis, and forest plots were plotted for related risk values. Results Compared with the UDCA complete response group, the UDCA poor response group had significantly higher proportion of patients with liver cirrhosis, levels of total bilirubin, aspartate aminotransferase (AST), alkaline phosphatase (ALP), total bile acid, total cholesterol (TC), IgA, and IgM, and positive rate of anti-Gp210 antibody at baseline ( χ 2 =4.596, Z =-9.932, -8.931, -8.361, -7.836, -4.694, -3.242, and -2.115, χ 2 =15.931, all P < 0.05). The UDCA poor response group had significantly higher Mayo Risk Score, Globe score, and UK-PBC risk score than the UDCA complete response group ( t =4.092, Z =-10.910 and -11.646, all P < 0.001). Compared with the normal IgM group, the elevated IgM group had significantly higher levels of AST, ALP, TC, IgA, and IgG and a significantly higher positive rate of anti-Gp210 antibody ( Z =-3.774, -5.063, -4.344, -2.051, and -6.144, χ 2 =25.180, all P < 0.05). IgM had an AUC of 0.552 in predicting poor treatment response. Compared with the IgM < 1.5×ULN group, the IgM ≥1.5×ULN group had significantly higher levels of AST, ALP, TC, and IgG, a significantly higher positive rate of anti-Gp210 antibody, and a significantly higher poor UDCA response rate ( Z =-4.193, -5.044, -3.250, and -5.465, χ 2 =25.204 and 8.948, all P < 0.05). IgM ≥1.5×ULN had an odds ratio of 1.416 (95% confidence interval [ CI ]: 1.129-1.776, P =0.003) in predicting poor response. The subgroup analysis showed that for patients without liver cirrhosis, IgM ≥1.5×ULN had an odds ratio of 1.821 (95% CI : 1.224-2.711, P =0.003) in predicting poor response. Conclusion Baseline IgM level has an important value in predicting UDCA response. IgM level should be closely monitored during treatment in PBC patients with a high baseline IgM level, and second-line drugs should be given in time if the abnormality persists.