Biochemical outcomes and safety profiles of compound glycyrrhizin versus magnesium isoglycyrrhizinate in Kawasaki disease: a retrospective propensity score‑matched cohort study

Introduction

Kawasaki disease (KD) is an acute systemic vasculitis that predominantly affects young children and represents the leading cause of acquired heart disease in developed countries (1,2). While high-dose intravenous immunoglobulin (IVIG) combined with aspirin significantly lowers the risk of coronary complications, increasing attention has been drawn to extra-cardiac manifestations, particularly hepatic dysfunction (3-5). Notably, over 90% of KD patients exhibit at least one abnormal liver function test result (6). Among these abnormalities, elevated alanine aminotransferase (ALT)—a biomarker primarily reflecting hepatic injury—is observed in around 40% of patients (7,8), suggesting systemic inflammation and liver involvement in the disease process (9,10). The pathogenesis of liver enzyme abnormality in KD, however, remains incompletely elucidated. Proposed mechanisms include immune-mediated inflammation, vascular injury, and drug-induced hepatotoxicity (11,12). Regardless of whether liver injury originates from the disease itself or its treatment, the presence of hepatic abnormalities in children with KD may adversely affect clinical prognosis, limit therapeutic options, contribute to IVIG resistance, and complicate overall disease management (13,14). Therefore, early detection and appropriate hepatoprotective interventions are essential in the comprehensive care of these patients.

Despite the high prevalence of hepatic involvement in KD, no standardized guidelines exist for its management. Current clinical practice largely relies on general pediatric hepatology principles or case-based experience. Supportive care—including withdrawal or adjustment of hepatotoxic medications—forms the basis of management (15). In cases of persistent transaminitis despite control of systemic inflammation, hepatoprotective agents may be considered (16). In China, two drugs commonly used in pediatric patients are compound glycyrrhizin (CG) (17,18) and magnesium isoglycyrrhizinate (MgIG). CG, derived from licorice root (Glycyrrhiza glabra), contains glycyrrhizic acid and its derivatives, often formulated with amino acids such as methionine. Its hepatoprotective mechanisms include anti-inflammatory, antioxidant, and immunomodulatory actions, such as cytokine suppression and hepatocyte membrane stabilization (19,20). It has shown efficacy in the treatment of viral hepatitis, drug-induced liver injury, and non-alcoholic fatty liver disease (21). Although generally well tolerated, adverse effects such as edema, hypertension, and gastrointestinal discomfort have been reported (22). MgIG, a fourth-generation glycyrrhizic acid product, is the magnesium salt of a stereoisomer of glycyrrhizic acid (23). Compared to CG, MgIG exhibits enhanced solubility, improved hepatoprotective efficacy, and a more favorable safety profile (24,25). It has demonstrated benefit in managing nonalcoholic fatty liver disease (26), liver fibrosis (27), and drug-related liver injury (28). Nevertheless, its use in children remains relatively understudied, with limited large-scale safety and efficacy data available. Although both agents are widely used in adult patients across China and other Asian countries, robust evidence supporting their application in children with KD-related liver enzyme abnormality is lacking.

Given the frequent hepatic involvement in KD and the absence of standardized management protocols for liver enzyme abnormalities in this context, systematic evaluation of hepatoprotective therapies is urgently needed. CG and MgIG are among the most frequently used hepatoprotective agents in Chinese pediatric practice, yet no direct comparative studies have assessed their outcomes specifically in KD patients. This retrospective, observational study aims to address this gap by comparing the biochemical outcomes and safety profiles of CG and MgIG in pediatric KD patients with liver enzyme abnormalities. We identified children diagnosed with KD who developed abnormal liver enzymes during the acute phase and were treated with either CG or MgIG as monotherapy. The primary outcomes assessed were the complete normalization, partial biochemical response, and the incidence of drug-related adverse events associated with the two hepatoprotective regimens. By evaluating these two agents in a real-world clinical setting, this study aims to provide evidence that may inform hepatoprotective strategies in the management of KD-associated liver dysfunction. We present this article in accordance with the STROBE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0279/rc).

Methods

Study population

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Children’s Hospital of Soochow University (approval No. 2024CS137). As this is a retrospective study, the requirement for informed consent was waived. This 3-year [2021–2023] retrospective cohort study enrolled inpatients from the Children’s Hospital of Soochow University. Eligible participants were pediatric patients diagnosed with KD who received monotherapy with either CG or MgIG. CG was administered intravenously at daily doses of 10 mL (for body weight <10 kg) or 20 mL (≥10 kg), with each 20 mL containing 40 mg glycyrrhizin, 400 mg glycine, and 20 mg cysteine hydrochloride. MgIG was administered at 5 mL daily (<10 kg) or 10 mL daily (≥10 kg), corresponding to 25 and 50 mg of MgIG, respectively. Exclusion criteria comprised patients diagnosed with concurrent viral hepatitis or infections caused by hepatitis A–E, Epstein-Barr virus, cytomegalovirus, rotavirus, or norovirus as secondary diagnoses. Liver enzyme elevations were defined as follows: borderline, ALT and/or aspartate aminotransferase (AST) elevation as <2× upper limit of normal (ULN); mild, AST and/or ALT elevation as 2–5× ULN; moderate, AST and/or ALT elevation as 5–15× ULN; severe, AST and/or ALT elevation as >15× ULN; and massive, AST and/or ALT elevation as >10,000 IU/L (29). Liver enzyme abnormality was defined as ALT and/or AST levels >2× ULN. The following parameters were collected for all enrolled patients: (I) demographic information: gender, age, body weight, and duration of medication; (II) medical and disease history; (III) timing of hepatoprotective therapy relative to disease onset; (IV) IVIG resistance; (V) serum biochemical parameters before and during hepatic dysfunction, including ALT, AST, alkaline phosphatase (ALP), and total bilirubin (TBIL); and (VI) safety profiles of CG and MgIG, recorded by adverse events such as drug eruption, hypotension, and serum potassium changes. Only patients with complete data for all relevant variables at both baseline and the 1-week follow-up were included in the final analysis. The number of cases in the area during the study period determined the sample size. All evaluations were independently conducted by three investigators to ensure consistency and objectivity.

Study endpoint

Liver function was assessed at baseline and again 1 week after the initiation of hepatoprotective therapy, at which time liver function tests, including ALT and AST, were repeated in both groups. Treatment outcomes were evaluated based on changes in ALT and AST levels from baseline to 1 week. First, the mean absolute change in ALT and AST from baseline to 1 week was calculated, and the difference between the two treatment groups was analyzed. This continuous outcome was used as the primary analytical approach to preserve statistical power and to avoid the information loss inherent in categorical thresholds. Subsequently, for descriptive purposes only, biochemical responses were categorized as follows: Complete normalization: both ALT and AST returned to normal levels. Partial biochemical response: liver enzyme levels improved but did not fully normalize, defined as both ALT and AST <2× ULN, in accordance with the borderline definition of the American College of Gastroenterology Clinical Guideline (29). No response: liver enzyme levels did not meet the criteria for complete normalization or partial response, or showed no clinically meaningful improvement.

Statistical analysis

To address potential confounding factors, we performed 2:1 propensity score matching (PSM) between the CG and MgIG groups. Propensity scores were estimated using a multivariable logistic regression model that included the following covariates: gender, age, body weight, duration of medication, baseline levels of ALT, AST, ALP, TBIL, and the severity of liver enzyme elevation. These covariates were selected based on two primary considerations. First, baseline liver enzyme levels and the severity of enzyme elevation are well-established prognostic factors for liver enzyme recovery; their inclusion reduces confounding by indication and improves the precision of the treatment effect estimate (30). Second, demographic and clinical characteristics—including gender, age, body weight, and duration of medication—may influence a patient’s baseline risk profile and were therefore included to enhance exchangeability between the two groups (31). Matching was carried out without replacement using a caliper matching algorithm. All data processing and statistical analyses were conducted using R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net). Normality of continuous variables was first assessed using the Shapiro-Wilk test. If the data deviated from normality, non-parametric tests were applied; otherwise, parametric tests were used. Continuous variables are presented as medians with interquartile ranges, and categorical variables as percentages. Between-group comparisons for continuous variables were performed using the Mann-Whitney U test, with results reported as the Z value and P value. Non-ordinal categorical variables (gender) were analyzed using the Pearson Chi-squared test, and the χ² statistic and P value were reported. Ordinal categorical variables (disease severity: mild, moderate, severe) were compared using the Cochran-Armitage test for trend, with the corresponding P value reported. These comparisons were carried out both before and after PSM to assess baseline balance and to evaluate outcomes in the matched cohort. A two-sided P value <0.05 was considered statistically significant.

Results

Demographic and clinical characteristics before PSM

According to the predefined inclusion criteria (patients diagnosed with KD who received monotherapy with either CG or MgIG alone), a total of 387 pediatric KD patients were initially enrolled. This cohort consisted of 254 patients in the CG group and 133 patients in the MgIG group. After excluding cases without sufficient data, 178 patients remained in the CG group and 67 in the MgIG group. Subsequently, patients with suspected liver enzyme abnormalities, defined as ALT >2× ULN and/or AST >2× ULN, were further selected. This process yielded a final analytical cohort of 168 patients in the CG group and 60 patients in the MgIG group. The patient selection and exclusion process is illustrated in the flow diagram (Figure 1). Baseline demographic and clinical characteristics of the two treatment groups are summarized in Table 1.

Figure 1 Study flow diagram. Initially, 387 KD patients receiving CG or MgIG monotherapy were enrolled. After excluding cases with insufficient data, those without liver enzyme abnormalities and PSM, the final analytical cohort comprised 80 CG patients and 49 MgIG patients. ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CG, compound glycyrrhizin; KD, kawasaki disease; MgIG, magnesium isoglycyrrhizinate; PSM, propensity score matching; TBIL, total bilirubin; ULN, upper limit of normal.

Initial analysis revealed no significant difference in gender composition between the CG group and MgIG group (χ²=3.66, P=0.06). Comparative analyses of other baseline characteristics were subsequently conducted using the Mann-Whitney U test for continuous variables. Patients in the CG group were significantly older than those in the MgIG group (Z=−2.44, P=0.02). However, no statistically significant differences were observed in body weight (Z=−1.86, P=0.06) or the duration of medication (Z=−1.54, P=0.12) between the two groups. Comparisons of liver enzyme profiles upon admission showed that the CG group presented with a significantly higher TBIL level than the MgIG group (Z=−2.21, P=0.03). No significant differences were found for ALT, AST, or ALP. Finally, patients were stratified into three subgroups based on the severity of liver enzyme elevation (mild, moderate, severe). A significant difference was found in the distribution across these severity subgroups between the treatment groups (P=0.006).

PSM and balance diagnostics

To address baseline imbalances between the two groups, PSM was performed. Patients from the CG group were matched to those from the MgIG group at a 2:1 ratio using nearest-neighbor matching with a caliper width of 0.05. This caliper was chosen as a well-established default that ensures good balance without excessive loss of subjects (32), while the 2:1 ratio was chosen because the CG group was larger than the MgIG group, thereby preserving statistical power while maintaining balance. After matching, 88 of 168 CG patients (52.4%) and 11 of 60 MgIG patients (18.3%) were excluded because no suitable match could be found within the caliper. To assess whether this exclusion introduced selection bias, we compared the baseline characteristics of excluded versus retained patients separately for each group (Table 2). In the CG group, excluded patients (n=88) differed significantly from retained patients (n=80) in age, body weight, and duration of medication (all P<0.05), but no significant differences were observed for other baseline variables (gender, ALT, AST, ALP, TBIL, or severity). In the MgIG group, excluded patients (n=11) had significantly higher baseline ALT and AST levels, as well as more severe liver enzyme elevation, compared with retained patients (n=49) (all P<0.05); other baseline characteristics did not differ significantly between the two subgroups. These findings indicate that the matched cohort represents a subset of the original population with overlapping propensity scores, and caution is warranted when generalizing the results to patients at the extremes of the clinical spectrum.

The baseline characteristics of the matched cohorts are summarized in Table 3. Covariate balance after PSM was assessed using both graphical and quantitative methods. As shown in the bar chart in Figure 2, the distribution of key covariates was highly similar between the two matched groups in terms of bar height, shape, and central tendency, indicating that systematic differences were substantially reduced after matching. A more detailed covariate balance plot (Figure 3) displays the standardized mean differences (SMDs) for all variables before and after matching. After matching, the absolute SMD for all covariates fell below the conventional threshold of 0.1, with the exception of gender, which showed a post-matching absolute SMD of 0.146, indicating a minor residual imbalance in gender distribution. Consistent with these graphical assessments, quantitative between-group comparisons in the matched sample confirmed no statistically significant differences for any baseline variable (all P>0.05; Table 3), further supporting that the matching procedure achieved acceptable overall balance. Collectively, these diagnostics confirm that PSM effectively created well-balanced groups for comparative outcome analysis. The potential for residual confounding by gender was addressed by including it as a covariate in the final multivariable model.

Figure 2 Bar chart comparing covariate distribution between CG and MgIG groups after PSM. Distribution of propensity scores in the CG and MgIG groups after PSM. The substantial overlap in the heights and ranges of the bars between the two groups indicates a well-balanced distribution of baseline characteristics and successful reduction of systematic differences between the matched cohorts. CG, compound glycyrrhizin; MgIG, magnesium isoglycyrrhizinate; PSM, propensity score matching.

Figure 3 Covariate balance before and after PSM. SMDs for all covariates before (red) and after (blue) matching. The dashed lines indicate the commonly accepted balance threshold of absolute SMD =0.1. After matching, all covariates achieved an absolute SMD below 0.1 except for gender, which showed a minor residual imbalance (post-matching absolute SMD =0.146). ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; PSM, propensity score matching; SMD, standardized mean difference; TBIL, total bilirubin.

Biochemical responses of treatment outcome

We first analyzed the absolute changes in ALT and AST from baseline to 1 week after treatment. In both treatment groups, ALT and AST levels decreased significantly from baseline (all P<0.001). However, no statistically significant differences were observed between the CG and MgIG groups in the absolute changes of ALT (P=0.07) or AST (P=0.11). The pre- and post-treatment values of ALT and AST at 1 week are summarized in Table 4. We next examined categorical biochemical responses at week 1 for descriptive purposes. Complete normalization of both ALT and AST was achieved in 36/80 patients (45.00%) in the CG group and 15/49 patients (30.61%) in the MgIG group, with no significant difference between groups (χ²=2.631, P=0.11). In contrast, partial biochemical response was observed in 69/80 patients (86.25%) in the CG group and 34/49 patients (69.39%) in the MgIG group, and this difference was statistically significant (χ²=5.369, P=0.02, Table 5).

Given the potential influence of baseline gender imbalance (SMD =0.146), the timing of hepatoprotective therapy relative to disease onset, and IVIG resistance on liver enzyme outcomes (14,33), we performed a multivariable logistic regression analysis. This analysis aimed to evaluate the independent effect of treatment group on partial biochemical response, adjusting for gender, time from diagnosis to hepatoprotective therapy initiation (continuous, days), and IVIG resistance status. The multivariable logistic regression revealed that, compared with the CG group, the MgIG group did not show a statistically significant difference in the odds of achieving a partial biochemical response [odds ratio (OR) =0.542, 95% confidence interval (CI): 0.203–1.452, P=0.22]. Female sex was not significantly associated with treatment response compared with male sex (OR =1.094, 95% CI: 0.433–2.763, P=0.85). Similarly, IVIG resistance showed no significant independent association with treatment outcome (OR =2.450, 95% CI: 0.625–9.609, P=0.20). Notably, each 1-day delay in treatment initiation from diagnosis was associated with a 19% reduction in the odds of response (OR =0.810, 95% CI: 0.663–0.989, P=0.04), making it the only significant independent predictor in the model (Table 6).

Treatment safety profiles

In the CG group and MgIG group, adverse events were reported in 3 (3.75%) and 2 (4.08%) cases, respectively. All events were cases of hypokalemia. No serious adverse events occurred in either group, and there was no statistically significant difference in safety between the two groups (P>0.99). A detailed summary of adverse events during the treatment period is provided in Table 7.

Discussion

This retrospective, propensity score-matched cohort study provides the first comparative analysis of two hepatoprotective agents, CG and MgIG, specifically in pediatric patients with KD-associated liver enzyme abnormalities. After adjusting for baseline imbalances using PSM, both treatments were associated with significant reductions in ALT and AST levels from baseline to 1 week. However, no statistically significant difference was observed between the two groups in the absolute changes of ALT or AST. In the multivariable logistic regression adjusting for gender, timing of treatment initiation, and IVIG resistance, treatment group (MgIG vs. CG) was not independently associated with the odds of achieving a partial biochemical response. Notably, each 1-day delay in treatment initiation from diagnosis was associated with a 19% reduction in the odds of response, making it the only significant independent predictor in the model. These findings should be interpreted as short-term surrogate biochemical outcomes without direct clinical benefit implications.

Our findings contribute to a sparse but growing literature on the management of KD complications. Previous research has firmly established the high prevalence of hepatic dysfunction in KD and its association with more severe disease (6) and IVIG resistance (34). However, therapeutic guidelines remain absent. Most published studies on hepatoprotective agents in children are either case series or focus on conditions like drug-induced liver injury (35). By systematically comparing the two most commonly used agents in a real-world setting, we move beyond anecdotal evidence toward evidence-based selection. Although both drugs share a glycyrrhizin acid core, their distinct pharmacological profiles may explain the observed trend toward a higher partial response rate with CG at the 1-week assessment. CG is a complex preparation containing glycyrrhizin acid alongside amino acids such as glycine and cysteine (36). This multi-component formulation may confer a broader or more synergistic hepatoprotective action. Glycine and cysteine are precursors for glutathione, the body’s principal endogenous antioxidant (37). In the context of KD, where systemic inflammation generates significant oxidative stress in the liver (38), the direct supplementation of glutathione precursors in CG may provide a rapid, adjunctive cytoprotective effect, stabilizing hepatocyte membranes more swiftly than the isolated anti-inflammatory action of glycyrrhizin acid derivatives. In contrast, MgIG is a stereochemically purified and highly soluble derivative that exerts anti-inflammatory effects primarily via the inhibition of NF-κB, autophagy, and MAPK signaling pathways (39-41). Consequently, it is particularly advantageous in the management of acute or moderate-to-severe liver injuries, such as drug-induced liver injury (24,42). Given that KD-related hepatitis in our cohort was predominantly mild-to-moderate, the balanced anti-inflammatory and antioxidant effects of CG may offer a plausible explanation for its earlier biochemical response. However, this hypothesis requires confirmation in dedicated mechanistic studies.

Several limitations of this study must be acknowledged to properly interpret its findings. First, the retrospective, observational design inherently carries risks of unmeasured confounding. Despite rigorous PSM adjusting for available clinical and biochemical variables, factors not captured in medical records—such as details of standard KD treatment, overall disease severity and inflammatory burden, or the clinical rationale for treatment allocation—could influence outcomes. Second, the study was conducted at a single tertiary center, which may limit the generalizability of the results to other populations with genetic backgrounds. Third, our primary endpoint was biochemical normalization, a surrogate marker. While clinically relevant, it does not directly correlate with hard clinical outcomes such as time to full clinical recovery or prevention of adverse events. Fourth, after PSM, we observed significant differences in baseline characteristics between excluded and retained patients. Excluded CG patients were older, had higher body weight, and longer medication duration, while excluded MgIG patients had higher baseline ALT/AST levels and more severe disease. Consequently, our findings are primarily generalizable to patients with comparable clinical profiles who fall within the common support region. Caution is warranted when extrapolating these results to patients at the extremes of the propensity score distribution. Fifth, the safety analysis was limited by the small number of reported adverse events and the modest sample size, particularly in the MgIG group. No serious adverse events occurred, but the study was underpowered to detect clinically meaningful differences in safety outcomes. Sixth, formal sensitivity analyses for PSM were not performed due to the limited sample size of the MgIG group (n=49 after matching). Alternative matching parameters would have further reduced sample size and compromised statistical power.

Despite these limitations, this study offers preliminary, hypothesis-generating observations. All reported biochemical outcomes should be interpreted as short-term surrogate measures. No direct clinical benefit can be inferred from these findings. A multicenter, prospective, randomized controlled trial comparing CG and MgIG, with stratified randomization based on baseline liver injury severity, is the logical next step to confirm our findings. Future studies should also explore the potential of these agents to modulate not only hepatic inflammation but also the systemic inflammatory burden of KD, and investigate any potential impact on the rate of IVIG resistance.

Conclusions

In this retrospective, propensity score-matched cohort study, both CG and MgIG were associated with significant reductions in ALT and AST levels from baseline to 1 week. After adjusting for gender, timing of treatment initiation, and IVIG resistance, no independent association was found between treatment group and partial biochemical response. Earlier initiation of hepatoprotective therapy was the only factor significantly associated with better biochemical outcomes. The observation of a higher partial response rate with CG at the 1-week assessment is preliminary and hypothesis-generating, warranting further investigation in prospective, adequately powered studies. These findings should be interpreted as short-term surrogate biochemical outcomes, and no clinical preference for either agent can be recommended based on this study alone.

Acknowledgments

We are grateful to the patients and their families who have made this research possible.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0279/rc

Data Sharing Statement: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0279/dss

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0279/prf

Funding: This study was supported by the grants from the Gusu Health Talents Project of Suzhou Municipal Health Commission (No. GSWS2022062) and the Tianqing Project of Jiangsu Province Pharmaceutical Committee (No. Q202235). The funders had no role in study design, data collection, analysis or interpretation of the data, preparation of the manuscript or decision to publish the results.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0279/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Children’s Hospital of Soochow University (approval No. 2024CS137). As this is a retrospective study, the requirement for informed consent was waived. Confidentiality of the information was secured throughout the study process. Furthermore, the collected data were anonymous.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Day-Lewis M, Son MBF, Lo MS. Kawasaki disease: contemporary perspectives. Lancet Child Adolesc Health 2024;8:781-92. [Crossref] [PubMed]
2. Shimizu C, Pérez Pomares JM, Loy CJ, et al. Pathology of the Coronary Arteries and Myocardium in Kawasaki Disease. Annu Rev Pathol 2026;21:349-71. [Crossref] [PubMed]
3. Goel AR, Yalcindag A. An Update on Kawasaki Disease. Curr Rheumatol Rep 2024;27:4. [Crossref] [PubMed]
4. Zhao ZH, Yang XY, Zhang H, et al. Expression of key genes in IVIG-unresponsive Kawasaki disease and their immune correlation based on bioinformatics analysis. Immunol Res 2026;74:10. [Crossref] [PubMed]
5. Xu Q, Wu Y, Ding Y. The Autophagy-Inflammation Axis in Kawasaki Disease: Pathogenic Mechanisms and Translational Opportunities. J Clin Med 2026;15:3918. [Crossref] [PubMed]
6. Mammadov G, Liu HH, Chen WX, et al. Hepatic dysfunction secondary to Kawasaki disease: characteristics, etiology and predictive role in coronary artery abnormalities. Clin Exp Med 2020;20:21-30. [Crossref] [PubMed]
7. Eladawy M, Dominguez SR, Anderson MS, et al. Abnormal liver panel in acute kawasaki disease. Pediatr Infect Dis J 2011;30:141-4. [Crossref] [PubMed]
8. Tremoulet AH, Jain S, Chandrasekar D, et al. Evolution of laboratory values in patients with Kawasaki disease. Pediatr Infect Dis J 2011;30:1022-6. [Crossref] [PubMed]
9. Pan Y, Jiao FY. Relationship between Kawasaki disease and alanine aminotransferase levels in pediatric patients. World J Hepatol 2025;17:98840. [Crossref] [PubMed]
10. Yongxing Z, Zheng Z, Qi Z, et al. Infliximab for intravenous immunoglobulin-resistant Kawasaki disease complicated with cholestatic hepatitis: a case report and discussion on coronary artery aneurysm prevention. Front Pediatr 2026;14:1794686. [Crossref] [PubMed]
11. Tomita Y, Fukaya T, Yamaura Y, et al. Implications of hepatic dysfunction in Kawasaki disease: Time-related changes in aspartate aminotransferase, alanine aminotransferase, total bilirubin, and C-reactive protein levels. Pediatr Investig 2019;3:19-26. [Crossref] [PubMed]
12. Gao L, Wang W, Wang H, et al. The safety and effectiveness of clopidogrel versus aspirin in Kawasaki disease with mild-to-moderate liver injury. Sci Rep 2023;13:18324. [Crossref] [PubMed]
13. Ren Y, Zhang C, Xu X, et al. A case report of atypical Kawasaki disease presented with severe elevated transaminases and literature review. BMC Infect Dis 2021;21:415. [Crossref] [PubMed]
14. Masuda H, Ae R, Koshimizu TA, et al. Serum alanine aminotransferase level and intravenous immunoglobulin resistance in patients with kawasaki disease. Clin Rheumatol 2022;41:3125-33. [Crossref] [PubMed]
15. Amin MD, Harpavat S, Leung DH. Drug-induced liver injury in children. Curr Opin Pediatr 2015;27:625-33. [Crossref] [PubMed]
16. Służały P, Paśko P, Galanty A. Natural Products as Hepatoprotective Agents-A Comprehensive Review of Clinical Trials. Plants (Basel) 2024;13:1985. [Crossref] [PubMed]
17. Liang SB, Hou WB, Zheng RX, et al. Compound glycyrrhizin injection for improving liver function in children with acute icteric hepatitis: A systematic review and meta-analysis. Integr Med Res 2022;11:100772. [Crossref] [PubMed]
18. Qian Y, Wu Y, Li W, et al. Combination Therapy with 308-nm Excimer Laser, Compound Glycyrrhizin, and Tacrolimus for Pediatric Facial and Cervical Vitiligo. Clin Cosmet Investig Dermatol 2026;19:593313. [Crossref] [PubMed]
19. Sharifi-Rad J, Quispe C, Herrera-Bravo J, et al. Glycyrrhiza Genus: Enlightening Phytochemical Components for Pharmacological and Health-Promoting Abilities. Oxid Med Cell Longev 2021;2021:7571132. [Crossref] [PubMed]
20. Ma T, Hao H, Wang S, et al. Glycyrrhizin preparations in liver diseases: a narrative review of mechanisms and therapeutic potential. Front Pharmacol 2026;17:1795174. [Crossref] [PubMed]
21. Li JY, Cao HY, Liu P, et al. Glycyrrhizic acid in the treatment of liver diseases: literature review. Biomed Res Int 2014;2014:872139. [Crossref] [PubMed]
22. Li M, Xiang L, Li Y. Efficacy and safety of compound glycyrrhizin in patients with alopecia areata: a systematic review and meta-analysis. Ann Med 2025;57:2491659. [Crossref] [PubMed]
23. Zhang Y, Li H, Liu X, et al. Integrating Metabolomics and Network Pharmacology to Decipher the Hepatoprotective Effect Mechanisms of Magnesium Isoglycyrrhizinate Injection. Curr Issues Mol Biol 2023;46:279-98. [Crossref] [PubMed]
24. Gong JY, Ren H, Peng SY, et al. Comparative effectiveness of glycyrrhizic acid preparations aimed at preventing and treating anti-tuberculosis drug-induced liver injury: A network meta-analysis of 97 randomized controlled trials. Phytomedicine 2022;98:153942. [Crossref] [PubMed]
25. Chen H, Zheng F, Wang M, et al. Meta-Analysis Based on Clinical RCTs: The Effect of Molecular Epimerism on the Safety of Glycyrrhizic Acid. Evid Based Complement Alternat Med 2020;2020:3869698. [Crossref] [PubMed]
26. Lai X, Zhou H, Wan Y, et al. Magnesium isoglycyrrhizinate attenuates nonalcoholic fatty liver disease by strengthening intestinal mucosal barrier. Int Immunopharmacol 2024;128:111429. [Crossref] [PubMed]
27. Yang YZ, Zhao XJ, Xu HJ, et al. Magnesium isoglycyrrhizinate ameliorates high fructose-induced liver fibrosis in rat by increasing miR-375-3p to suppress JAK2/STAT3 pathway and TGF-β1/Smad signaling. Acta Pharmacol Sin 2019;40:879-94. [Crossref] [PubMed]
28. Xia S, Hu ZY, Cao R, et al. Magnesium isoglycyrrhizinate prevented the liver injury of acetaminophen by promoting mitochondrial biogenesis. Toxicol Res (Camb) 2025;14:tfaf024. [Crossref] [PubMed]
29. Kwo PY, Cohen SM, Lim JK. ACG Clinical Guideline: Evaluation of Abnormal Liver Chemistries. Am J Gastroenterol 2017;112:18-35. [Crossref] [PubMed]
30. Giannini EG, Testa R, Savarino V. Liver enzyme alteration: a guide for clinicians. CMAJ 2005;172:367-79. [Crossref] [PubMed]
31. Austin PC. An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies. Multivariate Behav Res 2011;46:399-424. [Crossref] [PubMed]
32. Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat 2011;10:150-61. [Crossref] [PubMed]
33. Peng Y, Jing F, Liu D, et al. IVIG Resistance in Kawasaki Disease: Clinical and Laboratory Risk Factors and the Potential Role of Administration Timing. J Inflamm Res 2025;18:5923-9. [Crossref] [PubMed]
34. Cheah CS, Lee WWL, Suhaini SA, et al. Kawasaki disease in Malaysia: Biochemical profile, characterization, diagnosis and treatment. Front Pediatr 2022;10:1090928. [Crossref] [PubMed]
35. Niu H, Atallah E, Alvarez-Alvarez I, et al. Therapeutic Management of Idiosyncratic Drug-Induced Liver Injury and Acetaminophen Hepatotoxicity in the Paediatric Population: A Systematic Review. Drug Saf 2022;45:1329-48. [Crossref] [PubMed]
36. Koga K, Kikuchi H. Comparative study for pharmaceutical quality among bland-name drug and generic drugs of compound glycyrrhizin injections in China. J Basic Clin Pharm 2012;3:346-51. [Crossref] [PubMed]
37. Wu G, Fang YZ, Yang S, et al. Glutathione metabolism and its implications for health. J Nutr 2004;134:489-92. [Crossref] [PubMed]
38. Li Y, Lan S, Zhang H. Expression of Oxidative Stress and Inflammatory Indicators for Coronary Artery Disease in Kawasaki Disease. Mediterr J Hematol Infect Dis 2024;16:e2024052. [Crossref] [PubMed]
39. Ma D, Zhang J, Zhang Y, et al. Inhibition of myocardial hypertrophy by magnesium isoglycyrrhizinate through the TLR4/NF-κB signaling pathway in mice. Int Immunopharmacol 2018;55:237-44. [Crossref] [PubMed]
40. Zhou X, Li X, Yi K, et al. Magnesium isoglycyrrhizinate ameliorates lipopolysaccharide-induced liver injury by upregulating autophagy and inhibiting inflammation via IL-22 expression. Bioorg Chem 2022;128:106034. [Crossref] [PubMed]
41. Gao Y, Tian Y, Zhang X, et al. Magnesium isoglycyrrhizinate ameliorates concanavalin A-induced liver injury via the p38 and JNK MAPK pathway. Immunopharmacol Immunotoxicol 2020;42:445-55. [Crossref] [PubMed]
42. Wang Y, Wang Z, Gao M, et al. Efficacy and safety of magnesium isoglycyrrhizinate injection in patients with acute drug-induced liver injury: A phase II trial. Liver Int 2019;39:2102-11. [Crossref] [PubMed]