Table 3.
Outcome | Liver biopsy cohort (N = 1515) | Swedish Obese Subjects Study (N = 3329) | Dallas Heart Study (N = 4455)a | ||||||
---|---|---|---|---|---|---|---|---|---|
β | 95% c.i. | P value | β | 95% c.i. | P value | ||||
ALT | +0.57 | (0.42–0.77) | 2.8 × 10−11 | +0.63 | (0.490.78) | 1.4 × 10−17 | +0.33 | (0.18–0.49) | 2.8 × 10−5 |
AST | +0.57 | (0.38–0.70) | 8.0 × 10−10 | +0.59 | (0.44–0.74) | 1.9 × 10−14 | +0.25 | (0.09–0.41) | 0.0019 |
Necroinflammation | +0.70 | (0.48–0.81) | 6.6 × 10−12 | – | – | – | – | – | – |
Ballooning | +0.48 | (0.30–0.63) | 2.2 × 10−7 | – | – | – | – | – | – |
Fibrosis | +0.70 | (0.52–0.89) | 1.3 × 10−14 | – | – | – | – | – | – |
APRI scoreb | – | – | – | – | – | – | +0.17 | (−0.01 to 0.35) | 0.061 |
Hypertension | +0.04 | (−0.15 to 0.22) | 0.78 | +0.06 | (−0.01 to 0.21) | 0.47 | −0.48 | (−0.91 to −0.05) | 0.028 |
T2D | −0.07 | (−0.26 to 0.11) | 0.42 | +0.15 | (−0.01 to 0.31) | 0.057 | −0.08 | (−0.62 to 0.46) | 0.77 |
HOMA‐IR | +0.27 | (0.07–0.48) | 0.0089 | +0.21 | (0.06–0.35) | 0.006 | +0.08 | (−0.07 to 0.22) | 0.30 |
HDL | −0.15 | (−0.33 to 0.03) | 0.10 | −0.16 | (−0.31 to 0.01) | 0.039 | −0.13 | (−0.28 to 0.02) | 0.083 |
Adjusted standardized beta coefficients and (95% c.i.) are reported. Coefficients were adjusted for age, sex, BMI, use of statins, recruitment centre in the LBC or ethnicity in the DHS. The regression coefficients are expressed in unit of SD of feature per unit increment in GRS. For binary outcomes, the relationship was assessed using logistic regression, with beta coefficients representing the log of the odds of the outcome per a one‐unit increment in GRS. BMI, body mass index; T2D, type 2 diabetes; HOMA‐IR, homoeostasis model assessment‐insulin resistance index; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; ALT, alanine aminotransferases; AST, aspartate aminotransferases; HF, hepatic fat, that is normalized histological steatosis of hepatic triglycerides content. After correction for multiple comparisons, P < 0.0065, P < 0.008 and P < 0.0071 are considered statistically significant in the LBC, SOS and DHS, respectively. aAll individuals with available genotype, phenotype and covariate data (up to N = 4455) were included. bAvailable for N = 3133 in DHS.