The relationship between triglyceride-glucose index and prospective key clinical outcomes in patients hospitalised for coronary artery disease

Table 3 Threshold effect analysis of TyG index on all-cause mortality and MACEs

	Crude HR (95% CI)	Adjusted HR^a (95% CI)
All-cause mortality
Total	0.93 (0.83–1.03)	1.14 (1.01–1.29)
Fitting by two-piecewise Cox regression model
Inflection point	8.93	8.77
TyG index < inflection point (per unit)	0.76 (0.60–0.97)	1.01 (0.73–1.38)
TyG index > inflection point (per unit)	1.37 (1.09–1.72)	1.53 (1.19–1.96)
p for Log-likelihood ratio	0.001	0.045
MACEs
Total	1.02 (0.94–1.11)	1.13 (1.03–1.25)
Fitting by two-piecewise Cox regression model
Inflection point	8.95	8.62
TyG index < inflection point (per unit)	0.79 (0.65–0.95)	0.71 (0.50–0.99)
TyG index > inflection point (per unit)	1.36 (1.14–1.62)	1.28 (1.10–1.48)
p for Log-likelihood ratio	< 0.001	0.009

HR hazard ratio, CI confidence interval, MACEs major adverse cardiovascular events, TyG triglyceride-glucose, ACS acute coronary syndrome, ACE-I angiotensin-converting enzyme inhibitor, ARB angiotensin II receptor blocker, BMI body mass index, eGFR estimated glomerular filtration rate, LVEF left ventricular ejection fraction, TC total cholesterol, LDL-C low-density lipoprotein cholesterol
^aAdjusted for age, gender, BMI, smoking, ACS, previous myocardial infarction, stroke, hypertension, diabetes mellitus, statin, beta blocker, ACE-I/ARB, LVEF, eGFR, TC, LDL-C

ISSN: 1475-2840