Skip to main content

Table 2 Relationship of TyG with sICAS and ICASB

From: Triglyceride-glucose index, symptomatic intracranial artery stenosis and recurrence risk in minor stroke patients with hypertension

 

TyG quantiles*

OR/β (95% CI) (TyG as quantiles)

P value

OR/β (95% CI) (TyG as continuous)

P value

sICASa

 Unadjusted OR (95%CI)

Q1 (< 8.4)

1 (Ref.)

 

1.19 (1–1.41)

0.048

Q2 (8.4–8.9)

1.16 (0.81–1.66)

0.412

Q3 (8.9–9.4)

1.02 (0.71–1.46)

0.926

Q4 (≥ 9.4)

1.45 (1.02–2.06)

0.036

P for trend

0.074

 Adjusted model I OR (95%CI)

Q1 (< 8.4)

1 (Ref.)

 

1.23 (1.03–1.46)

0.022

Q2 (8.4–8.9)

1.16 (0.81–1.67)

0.411

Q3 (8.9–9.4)

1.05 (0.73–1.52)

0.799

Q4 (≥ 9.4)

1.54 (1.08–2.2)

0.018

P for trend

0.036

 Adjusted model II OR (95%CI)

Q1 (< 8.4)

1 (Ref.)

 

1.25 (1.01–1.55)

0.037

Q2 (8.4–8.9)

1.16 (0.79–1.71)

0.445

Q3 (8.9–9.4)

1.02 (0.68–1.53)

0.917

Q4 (≥ 9.4)

1.59 (1.04–2.43)

0.033

P for trend

0.074

ICASBb

 Unadjusted β (95%CI)

Q1 (< 8.4)

0 (Ref.)

 

0.29 (0.03, 0.55)

0.03

Q2 (8.4–8.9)

0.63 (0.1, 1.16)

0.02

Q3 (8.9–9.4)

0.52 (− 0.01, 1.05)

0.055

Q4 (≥ 9.4)

0.81 (0.27, 1.34)

0.003

P for trend

0.007

 Adjusted model I β (95%CI)

Q1 (< 8.4)

0 (Ref.)

 

0.46 (0.2–0.72)

0.001

Q2 (8.4–8.9)

0.72 (0.2–1.25)

0.007

Q3 (8.9–9.4)

0.76 (0.23–1.28)

0.005

Q4 (≥ 9.4)

1.15 (0.62–1.68)

 < 0.001

P for trend

 < 0.001

 Adjusted model II β (95%CI)

Q1 (< 8.4)

0 (Ref.)

 

0.43 (0.13–0.74)

0.006

Q2 (8.4–8.9)

0.74 (0.19–1.28)

0.008

Q3 (8.9–9.4)

0.63 (0.07–1.19)

0.028

Q4 (≥ 9.4)

1.08 (0.47–1.69)

0.001

P for trend

0.002

  1. *The first quantile of TyG was set as the reference group. aThe relationship between TyG and sICAS was examined by univariate and multivariate logistic regression models. bThe relationship between TyG and ICASB (as continuous) was examined by univariate and multivariate linear regression models. Model I was adjusted for age and sex, and model II was adjusted for BMI, SBP, DBP, smoking status, drinking, previous stroke, and previous diabetes mellitus. Confounders were determined based on univariate analysis and previous literature reports