TABLE 5

Multivariate logistic regression analysis of markers of liver injury with risk of incident NCEP metabolic syndrome in IRAS subjects

Independent variableUnitModel AModel BModel C
ALTSD of log1.43 (1.15–1.77)*1.29 (1.03–1.62)1.31 (1.04–1.66)
Q4 vs. Q12.50 (1.38–4.51)*2.04 (1.08–3.84)2.12 (1.10–4.09)
AST-to-ALT ratioSD of log0.72 (0.57–0.90)*0.79 (0.62–0.99)0.76 (0.60–0.98)
Q4 vs. Q10.40 (0.22–0.74)*0.49 (0.26–0.95)0.48 (0.25–0.95)
ALKSD of log1.31 (1.06–1.61)1.22 (0.97–1.52)1.22 (0.97–1.53)
Q4 vs. Q12.28 (1.24–4.20)*1.74 (0.92–3.31)1.84 (0.95–3.60)
CRPSD of log1.33 (1.09–1.63)*1.03 (0.81–1.29)0.97 (0.76–1.23)
Q4 vs. Q12.01 (1.12–3.60)0.90 (0.47–1.73)0.78 (0.39–1.54)
  • Data are OR (95% CI) from separate models. Model A adjusted for age, sex, ethnicity, clinical center, and alcohol intake; model B adjusted for age, sex, ethnicity, clinical center, alcohol intake, waist circumference, and IGT; model C adjusted for age, sex, ethnicity, clinical center, alcohol intake, waist circumference, IGT, Si, and AIR. ORs in rows indicated by SD refer to risk of incident metabolic syndrome associated with a 1-SD difference in the independent variable (see Table 3); ORs in rows indicated by Q4 vs. Q1 refer to risk of incident metabolic syndrome among those in the 4th quartile of the independent variable of interest compared with those in the lowest quartile.

  • * P < 0.01,

  • P < 0.05.