Table 3.
Association of circulating PCSK9 with incident type 2 diabetes in females
| PCSK9 | p for trend | ||||
|---|---|---|---|---|---|
| Q1 n = 704 |
Q2 n = 705 |
Q3 n = 704 |
Q4 n = 704 |
||
| Incidence of type 2 diabetes % | 45 (6.4) | 96 (13.6) | 105 (14.9) | 115 (16.4) | < 0.001 |
| Model 1 HR (95% CI) | 1.00 | 2.20 (1.17–4.14) | 2.28 (1.21–4.26) | 2.57 (1.39–4.77) | < 0.001 |
| Model 2 HR (95% CI) | 1.00 | 2.01 (1.06–3.79) | 2.19 (1.17–4.10) | 2.31 (1.24–4.31) | < 0.001 |
| Model 3 HR (95% CI) | 1.00 | 1.96 (1.04–3.71) | 2.17 (1.16–4.07) | 2.23 (1.20–4.16) | < 0.001 |
| Model 4 HR (95% CI) | 1.00 | 1.96 (1.04–3.70) | 2.14 (1.14–4.01) | 2.22 (1.19–4.13) | 0.002 |
| Model 5 HR (95% CI) | 1.00 | 1.87 (0.99–3.53) | 2.13 (1.14–4.00) | 2.16 (1.16–4.04) | 0.006 |
Model 1 was adjusted for age. Model 2 was adjusted for the variable in model 1 plus smoking, drinking, physical activity, and educational attainment. Model 3 was adjusted for the variables in model 2 plus BMI. Model 4 was adjusted for the variables in model 3 plus waist circumference. Model 5 was adjusted for the variables in model 4 plus TG, LDL-C, SBP, ALT, AST, GGT, CRP, and HOMA-IR. Subjects with a baseline circulating PCSK9 level in the lowest quartile group served as the reference group. Cutoff values in the four groups were Q1 < 220.20 ng/mL, Q2 220.21–272.36 ng/mL, Q3 272.37–342.61 ng/mL, and Q4 > 342.62 ng/mL for females