Abstract
Purpose
This single-center retrospective study aimed to assess the correlation between triglyceride-glucose index (TyG) and prognosis in severe fever with thrombocytopenia syndrome (SFTS) and investigate its correlation with metabolic alterations.
Methods
A total of 754 patients diagnosed with SFTS at Wuhan Union Hospital between January 2022 and April 2024 were enrolled. 754 SFTS patients were stratified into four groups based on TyG index quartiles. The primary outcome was in-hospital mortality; secondary endpoints included disease severity and secondary infection. Untargeted metabolomics via liquid chromatography-tandem mass spectrometry was conducted to identify significant metabolites and pathways between low and high TyG groups based on the cutoff value of 9.28 among 37 SFTS individuals.
Results
SFTS patients with a higher TyG index had significantly worse survival than those with a lower TyG index (HR = 2.53, P< 0.001). The predictive performance of the TyG index, evaluated by ROC analysis, yielded an area under the curve (AUC) of 0.773. Logistic regression revealed that a high TyG index was an independent predictor of severe disease (OR = 2.67, P= 0.002; Pfor trend = 0.035) and secondary infection (OR = 2.31, P= 0.008; Pfor trend = 0.025) after full adjustment. Metabolomic analysis indicated that levels of asperagenin, Leu-Arg-Asn-Arg, PC(20:0/15:0) were significantly different between the low and high TyG groups. KEGG pathway analysis identified insulin resistance as one of the most relevant pathways differentiating the low and high TyG groups.
Conclusion
A high TyG index is significantly associated with increased mortality and risk of secondary infection in SFTS, suggesting a potential role of insulin resistance in the progression of SFTS. The TyG index may serve as a reliable prognostic biomarker for risk stratification in SFTS patients.
Keywords: triglyceride glucose index, severe fever with thrombocytopenia syndrome, Cox regression, biomarker
Introduction
Severe fever with thrombocytopenia syndrome (SFTS) is caused by Dabie bandavirus, an emerging pathogen in the bunyavirus family. Since its initial detection in ticks in rural China in 2009, SFTSV has been isolated from a growing number of host species, and the incidence of SFTS has been rising.1 SFTS causes a series of symptoms including fever, thrombocytopenia, and multi-organ dysfunction and even haemophagocytic lymphohistiocytosis.2 SFTS carries a high mortality rate of approximately 30%, constituting a serious public health challenge. The existing paucity of validated prognostic biomarkers and clinical risk-assessment tools makes it imperative to discover reliable prognostic indicators and implement tailored follow-up protocols for optimized clinical management.
Accumulating clinical evidence has confirmed that a variety of inflammatory biomarkers exert reliable prognostic value in predicting SFTS-related mortality. Elevated inflammatory burden index is closely linked to increased mortality risk and acts as an independent predictor of unfavorable survival outcomes;3 likewise, the C-reactive protein-to-albumin ratio4 and serum ferritin concentration5 have also been validated as reliable prognostic indicators for fatal clinical outcomes in SFTS patients. While inflammation is critically involved in the initiation and progression of SFTS, accumulating studies have indicated that metabolic dysregulation also contributes substantially to the pathogenesis of acute infectious diseases. Our previous study6 reported that high fasting plasma glucose level on admission was correlated with worse prognosis in SFTS individuals without preexisting diabetes. While levels of fasting plasma glucose tend to fluctuate greatly, which are easily affected by some factors, such as metabolic syndrome. The triglyceride-glucose index (TyG index) is derived from fasting triglyceride and fasting glucose, which was first proposed by Gisela Unger.7 TyG index has been viewed as a reliable and easy-to-use index of insulin resistance. TyG index has already outperformed the homeostasis model assessment (HOMA), a well-established assessment method for insulin resistance.8 The TyG index is easy to measure and imposes fewer time and cost constraints, rendering it suitable for wild application in various clinical settings. TyG index is a simple biomarker for evaluating the risk of various cardiovascular diseases, including heart failure, coronary heart disease, and ischemic cardiomyopathy among the general population.9 TyG index exhibited promising potential in predicting the occurrence of non-alcoholic fatty liver disease.10
A recent meta-analysis revealed that the TyG index is a novel synthetic index and should be widely used in the whole-course management of individuals with coronary artery disease.11 A recent observational study demonstrated that an elevated TyG index is associated with adverse outcomes in surgical intensive care individuals.12 Moreover, Huang et al,13 demonstrated that high levels of TyG index were significantly associated with less favorable prognosis in patients with acute decompensated heart failure. However, SFTS is an emerging acute infectious disease with high mortality, and whether TyG index could also reflect the survival outcome of SFTS is still unknown.
In this study, a retrospective cohort study was conducted to investigate the association between TyG index levels and in-hospital mortality in SFTS patients. Subsequently, we compared the metabolic profiles of patients with low versus high TyG indices to elucidate the metabolic basis underlying this prognostic association. The main outcome of this study is the in-hospital survival status, and secondary outcomes include disease severity and secondary infection. To our knowledge, this is the first study to explore the correlation of TyG with metabolic alterations in SFTS patients.
Materials and Methods
SFTS Individuals
A total of 901 patients diagnosed with SFTS from Wuhan Union Hospital between January 2022 to April 2024 were initially screened. The diagnostic criteria for SFTS were based on the guidelines issued by the Chinese Center for Disease Control and Prevention.14 The inclusion criteria were: (1) Epidemiological link (eg, tick exposure); (2) Clinical presentation of fever with thrombocytopenia and leukopenia; (3) Laboratory confirmation via detection of Dabie bandavirus RNA by Reverse Transcription-Polymerase Chain Reaction or serological evidence of infection. The exclusion criteria were: (1) co-infection with hemorrhagic fever with renal syndrome or COVID-19; and (2) incomplete clinical data. After screening, 754 patients with SFTS were included in the final analysis. The study protocol was approved by the Ethics Committee of Wuhan Union Hospital (Approval No.2024–0611) and was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants.
Data Collection
Data were obtained from the hospital’s electronic medical records. This included laboratory results, treatment information, and survival outcomes. Clinical variables encompassed demographic characteristics, Dabie bandavirus load, and key laboratory indices (eg, blood routine, coagulation function, liver and kidney function). We routinely collect fasting blood test results from SFTS patients on the morning of the second day after admission. Viral loads of SFTS individuals were stratified at a threshold of 1000 TCID50. For critical patients with self-discharge abandoning treatment, telephone follow-up was performed to confirm their final outcome.
TyG Definition
 |
Outcome Assessment
The main outcome of this study is the in-hospital survival status, and the other outcomes of this clinical study are disease severity and secondary infection. The disease type of the individuals with SFTS was categorized into mild, moderate, severe, and critical types according to the SFTS guideline issued by the Chinese Center for Disease Control and Prevention.14 Secondary infection was defined as the presence of bacteria or fungi, as determined by sputum or blood culture.
Metabolic Analysis
Given that the first phase was a retrospective clinical analysis with no available serum samples for experimental use, we further performed a prospective study during April 2024 to June 2024. Blood samples were collected from SFTS patients following the same inclusion criteria as the first cohort for metabolomic profiling. A total of 37 acute-phase samples were obtained upon admission to Wuhan Union Hospital, including 18 in the low TyG group and 19 in the high TyG group. After centrifuge, plasm samples were obtained from the fresh blood samples. LC-MS/MS analyses of plasm were conducted using a UHPLC system (Agilent Technologies, Santa Clara, California, USA), and the detailed procedures are reported in our previous study.15 Data preprocessing, including peak picking, retention time correction, normalization and multiple testing correction was executed. Metabolite identification between the low and high TyG groups was based on accurate mass matching and MS/MS spectral comparison against reference databases. Statistical analysis identified differential metabolites based on fold change, variable important in projection (VIP) values, and Student’s t-test. Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were subsequently applied to map the biological pathways between the low and high TyG groups.
Statistical Analysis
All statistical analyses were performed using SPSS (version 22.0) and R software (version 4.3.1). Continuous variables are presented as mean± standard deviation or median (interquartile range, IQR), based on their distribution. Statistical analysis of continuous variables among four groups: use analysis of variance (ANOVA) for normally distributed variables and non-parametric tests for skewed distributed variables. Categorical variables are expressed as numbers (counts). Survival differences across TyG index quartiles were assessed using Kaplan-Meier curves, with between-group comparisons made by the Log rank test. The hazard ratio (HR) and 95% confidence interval (CI) for the association between the TyG index and in-hospital mortality were calculated using Cox proportional hazards regression. Logistic regression models were further employed to evaluate the associations of TyG index levels with disease severity and secondary infection. The predictive performance of the TyG index was evaluated using receiver operating characteristic (ROC) curve analysis, with the area under the curve (AUC) reported. A two-sided P < 0.05 was considered statistically significant.
Results
Demographic Features of the SFTS Individuals
Following screening, 754 patients with diagnosed SFTS were included in the final analysis. Participants were subsequently stratified into four groups according to TyG index quartiles. Their baseline clinical features are detailed in Table 1. Overall, the cohort had a mean age of 61.7±10.5 years, with 42.2% being female. The average length of hospitalization was 9 days, and the secondary infection rate was 14.7%. Notably, individuals in the highest TyG quartile displayed significantly higher levels of aspartate Aminotransferase, blood urea nitrogen, creatinine, and lactate Dehydrogenase, but lower platelet counts (all P < 0.05). Furthermore, this group had a significantly greater proportion of fatalities, secondary infections, and severe/critical cases (all P < 0.001).
Table 1.
Comparisons of Baseline Characteristics for All Patients in Union Cohort
| Characteristics | Overall | Q1 (< 8.94) | Q2 (8.94–9.27) | Q3 (9.28–9.69) | Q4 (> 9.69) | P value |
|---|---|---|---|---|---|---|
| N | 754 | 193 | 190 | 188 | 183 | – |
| Age, years old | 61.7 ± 10.5 | 59.6 ± 11.6 | 61.6 ± 9.8 | 62.2 ± 10.4 | 63.3 ± 9.5 | 0.006 |
| Gender, male, n (%) | 436 (57.8) | 83 (43.0) | 81 (42.6) | 85 (45.2) | 69 (37.7) | 0.515 |
| Clinical symptoms, n (%) | ||||||
| Altered mental status | 177 (23.5) | 53 (27.5) | 42 (22.1) | 39 (20.7) | 43 (23.5) | 0.454 |
| Muscle soreness | 212 (28.1) | 56 (20.0) | 49 (25.8) | 55 (29.3) | 52 (28.4) | 0.871 |
| Muscle tremor | 78 (10.3) | 19 (9.8) | 17 (8.9) | 21 (11.2) | 21 (11.5) | 0.840 |
| Comorbidities, n (%) | ||||||
| Hypertension | 152 (20.2) | 39 (20.2) | 37 (19.5) | 44 (23.4) | 32 (17.5) | 0.559 |
| COPD | 49 (6.5) | 17 (8.8) | 7 (3.7) | 13 (6.9) | 12 (6.6) | 0.219 |
| Viral load, n (%) | 0.362 | |||||
| Low | 161 (21.4) | 47 (24.4) | 44 (23.2) | 38 (20.2) | 32 (17.5) | |
| High | 593 (78.6) | 146 (75.6) | 146 (76.8) | 150 (79.8) | 151 (82.5) | |
| Laboratory values | ||||||
| WBC, × 109/L | 3.7 ± 1.2 | 3.4 ± 1.3 | 4.1 ± 1.2 | 3.5 ± 1.1 | 3.9 ± 1.4 | 0.069 |
| Hemoglobin, g/L | 126.3 ± 18.2 | 125.8 ± 17.4 | 126.0 ± 19.5 | 127.6 ± 19.5 | 126.1 ± 17.6 | 0.747 |
| Platelet, × 109/L | 52.0 ± 17.4 | 59.5 ± 21.2 | 53.1 ± 15.8 | 48.2 ± 11.7 | 47.0 ± 10.2 | <0.001 |
| AST, U/L | 190.5 (103.0, 378.3) | 153.0 (79.0, 296.5) | 179.5 (99.3, 336.0) | 201.5 (113.5, 378.8) | 264.0 (145.0, 563.0) | 0.003 |
| ALT, U/L | 82.0 (48.0, 147.3) | 64.0 (36.0, 115.5) | 84.0 (51.5, 150.0) | 83.0(52.0, 126.8) | 104.0 (59.0, 181.0) | 0.073 |
| ALP, U/L | 66.5 (54.8. 92.3) | 64.0 (53.0, 86.0) | 64.5 (54.8, 90.3) | 69.0 (55.0, 93.0) | 76.0 (56.0, 99.0) | 0.166 |
| TBIL, umol/L | 11.5 ± 4.3 | 10.7 ± 3.5 | 11.4 ± 4.4 | 11.0 ± 4.8 | 12.9 ± 5.7 | 0.166 |
| TP, g/L | 57.4 ± 6.2 | 57.5 ± 5.8 | 57.2 ± 6.3 | 56.9 ± 6.6 | 57.8 ± 6.2 | 0.418 |
| Albumin, g/L | 31.7 ± 4.5 | 32.9 ± 4.4 | 31.7 ± 4.5 | 31.1 ± 4.6 | 31.2 ± 4.2 | <0.001 |
| BUN, umol/L | 5.6 ± 2.2 | 5.0 ± 2.1 | 4.8 ± 1.7 | 5.7 ± 2.3 | 6.9 ± 2.0 | <0.001 |
| Creatinine, umol/L | 79.7 ± 19.3 | 75.8 ± 17.5 | 71.6 ± 17.4 | 78.7 ± 16.7 | 93.4 ± 24.0 | <0.001 |
| Glucose, mmol/L | 7.2 ± 1.3 | 5.6 ± 1.4 | 6.2 ± 1.9 | 7.9 ± 1.9 | 9.3 ± 4.2 | <0.001 |
| TG | 2.5 ± 0.9 | 1.5 ± 0.5 | 2.1 ± 0.6 | 2.9 ± 0.7 | 3.7 ± 1.0 | <0.001 |
| LDH, U/L | 774.0 (499.5, 1257.3) | 644.0 (423.5, 1013.0) | 725.5 (478.0, 1115.3) | 821.0 (519.0, 1807.9) | 980.0 (588.0, 1612.0) | 0.008 |
| CK, U/L | 517.5 (234.5, 1289.0) | 503.0 (203.5, 1159.0) | 451.0 (235.3, 1196.3) | 508.5 (239.0, 1304.0) | 641.0 (270.0, 1667.0) | 0.831 |
| APTT, s | 54.6 ± 18.9 | 52.6 ± 16.3 | 52.1 ± 16.4 | 57.0 ± 20.0 | 56.6 ± 21.9 | 0.015 |
| PT, s | 10.9 ± 3.5 | 11.0 ± 3.8 | 11.0 ± 3.5 | 10.9 ± 3.8 | 10.6 ± 3.3 | 0.857 |
| D dimer | 4.5 ± 1.8 | 4.4 ± 1.9 | 4.2 ± 1.1 | 4.5 ± 1.2 | 5.1 ± 2.1 | 0.292 |
| PCT | 0.3 (0.1, 0.8) | 0.2 (0.1, 0.6) | 0.3 (0.1, 0.7) | 0.3 (0.1, 0.9) | 0.4 (0.2, 1.0) | 0.458 |
| Hs-CRP | 4.5 (3.1, 9.7) | 3.8 (3.1, 8.2) | 4.4 (3.1, 9.6) | 4.8 (3.2, 9.7) | 5.3 (3.3, 11.5) | 0.388 |
| Outcomes | ||||||
| Time, days | 9.0 (7.0, 12.0) | 9.0 (7.0, 12.0) | 9.0 (7.0, 12.0) | 9.0 (7.0, 12.0) | 9.0 (6.0, 12.0) | 0.713 |
| Death, n (%) | 101 (13.4) | 13 (6.7) | 17 (8.9) | 30 (16.0) | 41 (22.4) | <0.001 |
| Secondary infection | 111 (14.7) | 14 (7.3) | 24 (12.6) | 35 (18.6) | 38 (20.8) | <0.001 |
| Disease severity | <0.001 | |||||
| Mild/moderate | 334 (44.3) | 115 (59.6) | 94 (49.5) | 70 (37.2) | 55 (30.0) | |
| Severe/critical | 420 (55.7) | 78 (40.4) | 96 (50.5) | 118 (62.8) | 128 (69.9) |
Notes: Statistical analysis of continuous variables among four groups: use analysis of variance (ANOVA) for normally distributed variables and non-parametric tests for skewed distributed variables.
Abbreviations: COPD, chronic obstructive pulmonary disease, WBC, white blood cells, AST, aspartate aminotransferase, ALT, alanine aminotransferase, ALP, alkaline phosphatase, total bilirubin, TP, total protein, BUN, blood urea nitrogen, TG, triglyceride, TyG, triglyceride-to-glucose index, LDH, lactate dehydrogenase, CK, creatine kinase, APTT, activated partial thromboplastin time, PT, prothrombin time, PCT, procalcitonin, hs-CRP, high-sensitivity C-reactive protein.
Correlation Between Levels of TyG and Survival Outcome in SFTS Patients
To visualize the relationship between the TyG index and in-hospital mortality, restricted cubic spline (RCS) analysis was performed. As shown in Figure 1A, a linear association was observed (Pfor non-linearity=0.571). Kaplan-Meier survival analysis further demonstrated that SFTS patients in the highest TyG index quartile had significantly higher in-hospital mortality compared to those in the lowest quartile (P<0.001, Figure 1B). Consistently, a distribution plot of TyG index levels versus survival outcome indicated that non-survivors were more frequently observed among patients with higher TyG index values (Figure 2A). Subgroup analyses confirmed that the association between TyG index quartiles and in-hospital mortality remained consistent across all predefined subgroups, including age, sex, viral load, altered mental status, muscle soreness, muscle tremor, chronic obstructive pulmonary disease (COPD), and hypertension (Figure 2B).
Figure 1.
RCS analysis detected the linear trend between levels of TyG index and in-hospital mortality (A). The prognostic significance of TyG index revealed by Kaplan-Meier plot (B). RCS analysis detected the linear trend between levels of TyG and disease type (C), and secondary infection (D).
Figure 2.
Association between levels of TyG index and survival outcome in SFTS individuals (A). Subgroup analysis of TyG index based on significant clinical features (B). Relationships between levels of TyG index and disease severity (C), and secondary infection in SFTS individuals (D).
Cox regression analysis identified the TyG index as an independent risk factor for in-hospital mortality (unadjusted HR=3.56, 95% CI:1.91–6.65, P< 0.001; Pfor trend <0.001) (Table S1). This association remained significant after adjusting for age, sex, and viral load (HR = 3.14, 95% CI: 1.67–5.91, P< 0.001; Pfor trend<0.001), and after further adjustment for clinical symptoms and comorbidities (fully adjusted HR=2.65, 95% CI:1.34–5.28, P=0.005; P for trend=0.027). These results demonstrate the robust prognostic value of the TyG index in SFTS.
The TyG index was dichotomized into low and high groups using a cutoff value of 9.28 based on the median value of TyG (Figure 3A). Most non-survivors were in the high TyG group (Figure 3B). A heatmap illustrated the distributions of the TyG index, triglyceride (TG), and glucose (Glu) in the two groups (Figure 3C). Survival analysis confirmed that patients in the low TyG group had significantly better survival than those in the high TyG group (HR = 2.53, 95% CI: 1.65–3.87, P< 0.001; Figure 3D). Decision curve analysis (DCA) indicated the clinical utility of the TyG index for risk prediction (Figure 3E). Finally, the predictive performance of the TyG index was evaluated using ROC curve analysis, yielding an AUC of 0.773 (95% CI: 0.726–0.820; Figure 3F), outperforming the predictive ability of fasting plasma glucose (Figure S1).
Figure 3.
TyG index was divided into the low risk and high risk according to the cutoff value of 9.28 (A). Changes of survival status of SFTS individuals with hospitalization time (B). Heat map shows the distribution of TyG index, triglyceride and glucose levels(C). Kaplan-Meier curve shows the prognostic value of TyG index (D). DCA curve demonstrates the clinical utility of TyG index (E). ROC curve shows the good predictive ability of TyG index for survival outcome (F).
Relationship Between Levels of TyG and Disease Severity in SFTS Patients
RCS analysis revealed a linear relationship between the TyG index and disease severity (P for non-linearity = 0.529, Figure 1C). As shown in Figure 2C, the proportion of patients with severe or critical illness was significantly higher in the upper TyG index quartiles compared to the lower quartiles. Subgroup analysis confirmed that the association between the TyG index and disease severity remained consistent across all subgroups, including age, sex, viral load, altered mental status, muscle soreness, muscle tremor, COPD, and hypertension (Figure 2B).
Logistic regression was used to assess whether a high TyG index is a risk factor for severe/critical SFTS (Table S2). In the unadjusted model, the TyG index was identified as an independent risk factor (OR=3.43, 95% CI: 2.24–5.26, P< 0.001; Pfor trend < 0.001). This association remained significant after adjusting for age, sex, and viral load (OR=3.11, 95% CI:2.01–4.82, P= 0.001; Pfor trend < 0.001) and after further adjustment for clinical symptoms and comorbidities (fully adjusted OR = 2.67, 95% CI:1.18–3.22, P=0.002; Pfor trend = 0.035). When TyG index was dichotomized into low and high groups, it remained an independent predictor of severe disease after full adjustment (OR =1.59, 95% CI:1.05–2.41, P=0.028). In summary, a high TyG index is a reliable biomarker for predicting disease severity in SFTS patients.
Association Between Levels of TyG and Secondary Infection in SFTS
Secondary infection was analyzed as an additional clinical outcome, given that hyperglycemia is a recognized risk factor for infectious complications. We therefore examined whether a high TyG index also predicted secondary infection in SFTS patients. RCS analysis indicated a linear association between the TyG index and secondary infection (Pfor non-linearity=0.694, Figure 1D). Patients in the higher TyG quartiles had a greater proportion of secondary infections compared to those in the lower quartiles (Figure 2D). Subgroup analyses confirmed that the association between TyG index quartiles and secondary infection risk was consistent across all subgroups, including age, sex, viral load, altered mental status, muscle soreness, muscle tremor, COPD, and hypertension (Figure 2B).
Logistic regression was used to further evaluate this relationship (Table S2). In the unadjusted model, the TyG index was an independent risk factor for secondary infection (OR = 3.35, 95% CI: 1.75–6.42, P< 0.001; P for trend < 0.001). After adjusting for age, sex, and viral load, the risk estimate remained significant (OR=2.89, 95% CI:0.49–5.59, P= 0.001; Pfor trend= 0.005). Following full adjustment for clinical symptoms and comorbidities, the association persisted (OR=2.31, 95% CI:1.25–4.63, P=0.008; Pfor trend=0.025). When the TyG index was dichotomized into low and high groups, it remained a significant predictor of secondary infection after full adjustment (OR=1.81, 95% CI: 1.16–2.84, P= 0.009). In summary, a high TyG index is a reliable indicator for predicting secondary infection in patients with SFTS.
Metabolic Profile
Metabolomic profiling of plasma samples from SFTS patients was performed using untargeted LC-MS/MS.A total of 1928 and 1996 plasma metabolites were identified in the high and low TyG index groups, respectively (Figure 4A). Partial Least Squares Discriminant Analysis (PLS-DA) revealed a distinct separation trend between the two groups (Figure 4B). A volcano plot visualization highlighted the statistical significance and magnitude of changes in metabolite abundance, showing 264 upregulated and 126 downregulated metabolites in the high TyG group compared to the low TyG group (Figure 4C). Finally, variable importance in projection (VIP) analysis identified asperagenin, leu-Arg-Asn-Arg, PC(20:0/15:0), and gancaonin U as the most discriminating metabolites between the two groups based on VIP scores (Figure 4D).
Figure 4.
Analysis of plasmatic metabolites between the low TyG and high TyG groups. Venn plot showing the numbers of plasmatic metabolites between the low TyG and high TyG groups (A). PLS-DA analysis (B). The volcano plot showing a visual representation of the statistical magnitude of differences in serum metabolite abundance between the low TyG and high TyG groups (C). VIP analysis identified the significant metabolites (D).
Notes: *Stands for P<0.05, ** stands for P<0.01, *** stands for P<0.0001.
KEGG Analysis of Differential Metabolites Between the Low and High TyG Groups
KEGG pathway enrichment analysis was performed to explore the functional implications of the differentially abundant metabolites between the low and high TyG groups. As shown in Figure 5, the enriched pathways included insulin resistance, choline metabolism in cancer, axon regeneration, caffeine metabolism, phenylalanine metabolism, tyrosine metabolism, and glycerophospholipid metabolism. Among these, the most significant enrichment pathway was observed in the insulin resistance.
Figure 5.
KEGG enrichment analysis identified insulin resistance as the most significant pathway.
Discussion
While previous studies have examined the role of blood glucose and lipids in SFTS progression, the prognostic value of the TyG index in this population remains unexplored. The present study systematically evaluated the association between TyG index levels and clinical outcomes, including survival outcome, disease severity, and secondary infection, thereby addressing an important gap in the literature. Our results demonstrate that a higher TyG index is associated with an increased risk of severe disease and secondary infection in SFTS patients. To our knowledge, this is the first study to link the TyG index to these key clinical endpoints in SFTS population.
Analysis of metabolic changes behind the low and high TyG groups can help us gain insights into the regulatory mechanisms underlying the pathogenesis of SFTS. As we know, most researches related to the TyG index did not investigate the metabolic changes behind the low and high TyG groups.16–18 This study identified several important metabolites associated with the levels of TyG index, indicating that these metabolites might play a key role in the progression of SFTS. Importantly, KEGG analysis proved that these significant metabolites between the low and high TyG groups were mostly enriched in insulin resistance. The metabolic analysis proved that the TyG index might reflect insulin resistance, which also plays an important role in SFTS. However, given the relatively limited sample size, it must be acknowledged that the metabolomics findings are exploratory and hypothesis-generating in nature.
Several previous studies demonstrated that diabetes mellitus is an independent risk factor for poor prognosis among SFTS individuals.19,20 Our previous study showed that SFTS survivors had significantly lower admission fasting plasma glucose levels than non-survivors (6.583±0.103 mmol/L vs. 8.623±0.4215 mmol/L, P<0.0001), which could strongly predict in-hospital mortality in non-diabetic SFTS patients.6 In this study, we did not exclude SFTS patients with preexisting dysglycemia. ROC analysis revealed the TyG index had superior predictive capability relative to fasting plasma glucose. However, the incremental predictive value of TyG requires further validation in larger SFTS cohort studies.
Although our data indicate that higher TyG levels are associated with increased mortality compared to lower levels, it is important to note that both acute stress hyperglycemia and chronic insulin resistance likely contribute to adverse outcomes. Post-COVID-19 syndrome was reported to be correlated with increased insulin resistance, which was significantly associated with severe inflammation in COVID-19.21 High levels of plasma insulin impair the balance of mitogen-activated protein kinase-dependent pathway and lead to endothelial cell damage.22 He et al,23 deemed that insulin resistance is the main cause of hyperglycemia upon SARS-2 infection, which drives the progression of COVID-19. Insulin resistance and subsequent hyperglycemia contribute to viral proliferation in human monocytes and also boost glycolysis, which leads to reactive oxygen species production and cytokine release.24 For the first time, this study demonstrates that the TyG index participates in SFTS progression, and metabolic dysregulation is a plausible mediating mechanism.
This study identified TyG reflects a metabolic state consistent with insulin resistance, as a key metabolic risk factor for disease progression in SFTS patients. A recent study indicates that insulin resistance further prompts macrophages and other inflammatory cells to release a large number of pro-inflammatory cytokines, which in turn exacerbates the systemic inflammatory response and may eventually progress to multiple organ dysfunction and failure.25 Research on diabetes and its complications has investigated the association between TyG and direct measures of insulin resistance (such as HOMA-IR, insulin, HbA1c, and C-peptide), consistently finding a strong positive link between TyG and these direct indicators.26,27 Therefore, assessing insulin resistance in SFTS patients can help identify high-risk individuals early and facilitate timely intervention.
Limitations
Several limitations should be considered in the interpretation of our findings. First, this was a retrospective single-center study, and potential residual confounding factors, such as unmeasured metabolic comorbidities, treatments during hospitalization.might exist in this study. Second, we did not directly examine the relationship between TyG and direct insulin resistance markers (such as HOMA-IR, insulin, HbA1c, and C-peptide) in SFTS individuals. Finally, a small sample size (N=37) in the metabolomics section, relevant metabolic results should be interpreted with caution. Although we linked TyG levels to metabolic alterations, the precise mechanisms through which TyG influences SFTS progression require validation in mechanistic and interventional studies.
Conclusion
This study suggests a possible association between an elevated TyG index, as a surrogate marker of prognostic biomarker in SFTS. These findings underscore the clinical relevance of the TyG index for risk assessment, support TyG as a prognostic biomarker and as a basis for future mechanistic and interventional studies. This study warrant cautious interpretation owing to the inherent limitations of the single-center retrospective design.
Data Sharing Statement
The original data of this study was accessible on reasonable request to the corresponding author (Chunxia Guo).
Ethics Statement
The clinical research was reviewed and then approved by the Ethics Committee of Wuhan Union Hospital (Approval No.2024-0611), and this research was implemented in line with the main principles of Helsinki Declaration.
Disclosure
The authors report no conflicts of interest in this work.
References
- 1.Cui H, Shen S, Chen L, et al. Global epidemiology of severe fever with thrombocytopenia syndrome virus in human and animals: a systematic review and meta-analysis. Lancet Reg Health-w. 2024;48:101133 doi: 10.1016/j.lanwpc.2024.101133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wang G, Ge HH, Hu L, et al. Severe fever with thrombocytopenia syndrome complicated by haemophagocytic lymphohistiocytosis: a retrospective cohort study. Clin Microbiol Infect. 2024;30:558–12. doi: 10.1016/j.cmi.2024.01.009 [DOI] [PubMed] [Google Scholar]
- 3.Yang Z, Wang L, Hong B, et al. Inflammatory burden index as a predictor of in-hospital mortality in patients with severe fever with thrombocytopenia syndrome. J Med Virol. 2025;97:e70225. doi: 10.1002/jmv.70225 [DOI] [PubMed] [Google Scholar]
- 4.Yang X, Yin H, Xiao C, Li R, Liu Y. The prognostic significance of C-reactive protein to albumin ratio in patients with severe fever with thrombocytopenia syndrome. Front Med-Lausanne. 2022;9:879982. doi: 10.3389/fmed.2022.879982 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Xiao W, Zhang L, Zhang Y, et al. Association between serum ferritin and mortality in patients with severe fever with thrombocytopenia syndrome: a retrospective cohort study. Plos Neglect Tropical Dis. 2025;19:e13104. doi: 10.1371/journal.pntd.0013104 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Pan J, Yang Z, Xu W, et al. Effect of fasting plasma glucose level in severe fever and thrombocytopenia syndrome patients without diabetes. Plos Neglect Trop Dis. 2024;18:e12125. doi: 10.1371/journal.pntd.0012125 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Unger G, Benozzi SF, Perruzza F, Pennacchiotti GL. Triglycerides and glucose index: a useful indicator of insulin resistance. Endocrinol Nutr. 2014;61:533–540. doi: 10.1016/j.endonu.2014.06.009 [DOI] [PubMed] [Google Scholar]
- 8.Irace C, Carallo C, Scavelli FB, et al. Markers of insulin resistance and carotid atherosclerosis. A comparison of the homeostasis model assessment and triglyceride glucose index. Int J Clin Pract. 2013;67:665–672. doi: 10.1111/ijcp.12124 [DOI] [PubMed] [Google Scholar]
- 9.Tao LC, Xu JN, Wang TT, Hua F, Li JJ. Triglyceride-glucose index as a marker in cardiovascular diseases: landscape and limitations. Cardiovasc Diabetol. 2022;21:68. doi: 10.1186/s12933-022-01511-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Chen Q, Hu P, Hou X, et al. Association between triglyceride-glucose related indices and mortality among individuals with non-alcoholic fatty liver disease or metabolic dysfunction-associated steatotic liver disease. Cardiovasc Diabetol. 2024;23:232. doi: 10.1186/s12933-024-02343-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Liang S, Wang C, Zhang J, et al. Triglyceride-glucose index and coronary artery disease: a systematic review and meta-analysis of risk, severity, and prognosis. Cardiovasc Diabetol. 2023;22:170. doi: 10.1186/s12933-023-01906-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liu D, Ren B, Tian Y, Chang Z, Zou T. Association of the TyG index with prognosis in surgical intensive care patients: data from the MIMIC-IV. Cardiovasc Diabetol. 2024;23:193. doi: 10.1186/s12933-024-02293-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Huang R, Wang Z, Chen J, et al. Prognostic value of triglyceride glucose (TyG) index in patients with acute decompensated heart failure. Cardiovasc Diabetol. 2022;21:88. doi: 10.1186/s12933-022-01507-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Chinese Society of Infectious Diseases, Chinese Medical Association. Consensus on the diagnosis and treatment of severe fever with thrombocytopenia syndrome. Chin J Infect Dis. 2022;40(12):711–721. (in Chinese). [Google Scholar]
- 15.Guo C, Li R, Wang X, Peng X. Aspartate aminotransferase to platelet ratio correlates with poor prognosis and metabolic alterations in Dabie bandavirus infection. Front Immunol. 2024;15:1471511. doi: 10.3389/fimmu.2024.1471511 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Duan M, Zhao X, Li S, et al. Metabolic score for insulin resistance (METS-IR) predicts all-cause and cardiovascular mortality in the general population: evidence from NHANES 2001-2018. Cardiovasc Diabetol. 2024;23:243. doi: 10.1186/s12933-024-02334-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Avagimyan A, Pogosova N, Fogacci F, et al. Triglyceride-glucose index (TyG) as a novel biomarker in the era of cardiometabolic medicine. Int J Cardiol. 2025;418:132663. doi: 10.1016/j.ijcard.2024.132663 [DOI] [PubMed] [Google Scholar]
- 18.Yang J, Liu C, Zhao S, et al. The association between the triglyceride-glucose index and sarcopenia: data from the NHANES 2011-2018. Lipids Health Dis. 2024;23:219. doi: 10.1186/s12944-024-02201-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhang Y, Zhang Y, Xu Y, Huang Y. The associations between fasting blood glucose levels and mortality of SFTS in patients. BMC Infect Dis. 2021;21:761. doi: 10.1186/s12879-021-06463-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ge H, Zhao J, Zhang S, et al. Impact of glycemia and insulin treatment in fatal outcome of severe fever with thrombocytopenia syndrome. Int J Infect Dis. 2022;119:24–31. doi: 10.1016/j.ijid.2022.03.038 [DOI] [PubMed] [Google Scholar]
- 21.Fierro P, Martin D, Pariente-Rodrigo E, et al. Post-COVID-19 syndrome, inflammation and insulin resistance: a retrospective cohort study. Minerva Endocrinol. 2024. doi: 10.23736/S2724-6507.23.04108-8. [DOI] [PubMed] [Google Scholar]
- 22.Gangadharan C, Ahluwalia R, Sigamani A. Diabetes and COVID-19: role of insulin resistance as a risk factor for COVID-19 severity. World J Diabetes. 2021;12:1550–1562. doi: 10.4239/wjd.v12.i9.1550 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.He X, Liu C, Peng J, et al. COVID-19 induces new-onset insulin resistance and lipid metabolic dysregulation via regulation of secreted metabolic factors. Signal Transduct Target Ther. 2021;6:427. doi: 10.1038/s41392-021-00822-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Xi G, Shen X, Wai C, White MF, Clemmons DR. Hyperglycemia induces vascular smooth muscle cell dedifferentiation by suppressing insulin receptor substrate-1-mediated p53/KLF4 complex stabilization. J Biol Chem. 2019;294:2407–2421. doi: 10.1074/jbc.RA118.005398 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.de Lima EV, Nogueira CO, Christoff RR, et al. Central Zika virus infection causes hypothalamic inflammation and persistent insulin resistance in adult mice. Cell Death Dis. 2025;16:722. doi: 10.1038/s41419-025-08046-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Yoon JS, Lee HJ, Jeong HR, et al. Triglyceride glucose index is superior biomarker for predicting type 2 diabetes mellitus in children and adolescents. Endocr J. 2022;69:559–565. doi: 10.1507/endocrj.EJ21-0560 [DOI] [PubMed] [Google Scholar]
- 27.Zhang M, Zhou X, Ma W, et al. Predictive value of triglyceride-glucose index and sex hormone-binding globulin in diabetic retinopathy. Int J Ophthalmol. 2026;19:921–926. doi: 10.18240/ijo.2026.05.11 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The original data of this study was accessible on reasonable request to the corresponding author (Chunxia Guo).