Exploring the prognostic impact of triglyceride-glucose index in critically ill patients with first-ever stroke: insights from traditional methods and machine learning-based mortality prediction
Main finding · ICU all-cause mortalityResult — go to the resultHazard ratio 1.92 95% CI 1.38–2.66Authors' conclusion
TyG (optimal cut-off: 9.265) was identified as an independent risk factor for ICU and hospital all-cause mortality in first-ever strokes. The ML model incorporating TyG demonstrated strong predictive performance. This emphasises the importance of insulin resistance (with TyG as a surrogate measure) in the prognostic assessment and early risk stratification of first-time stroke patients.
Study context
- Population
- Critically ill patients with first-ever stroke, eICU Collaborative Research Database 2014–2015 (median age 68.0 years; 63.0% male) and local hospital intensive care unit (ICUs) as external validation (n=201)primary analysis cohort
- Participants
- 3,173 patients
- Exposure
- High TyG vs low TyG
- Outcome
- ICU all-cause mortality
Methodological detailseICU Collaborative Research Database primary cohort, training and internal validation cohorts, external validation cohort (local hospital ICUs), multivariate Cox proportional hazards models, restricted cubic spline analyses
- eICU Collaborative Research Database primary cohort
- Training and internal validation cohorts
- External validation cohort (local hospital ICUs)
- Multivariate Cox proportional hazards models
- Restricted cubic spline analyses
Main finding
As reported in the abstract
Hazard ratio
1.92
95% CI 1.38–2.66
High TyGvs Low TyG
HR 1.92 (95% CI 1.38–2.66)
Observational data: associations, not effects.
Further findings
| Outcome | HR (95% CI) | |
|---|---|---|
| Hospital all-cause mortalityMultivariate Cox | 1.69 (1.32–2.16) | |
| 0.512Hazard ratio (log scale) |
Correlation of TyG with LOS and Glasgow Coma Score
- ICU LOS, r
- 0.077
- Hospital LOS, r
- 0.042
- Glasgow Coma Score, r
- −0.132
Machine learning (ML) prediction of ICU all-cause mortality (random forest, TyG + 6 features)
- Internal validation AUC
- 0.900
- Internal validation G-mean
- 0.443
- External validation AUC
- 0.776
- External validation G-mean
- 0.399
- Random forest performed best
Original publication
Exploring the prognostic impact of triglyceride-glucose index in critically ill patients with first-ever stroke: insights from traditional methods and machine learning-based mortality prediction. Cardiovasc Diabetol. Published 18 December 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →