Table 2.
Comparison of serum activin A and follistatin levels according to glycemic status.
| Normal (n = 215) |
Prediabetes (n = 168) |
Diabetes (n = 87) |
p value Normal vs. Prediabetes |
p value Normal vs. Diabetes |
p value Prediabetes vs. Diabetes |
p value between groups |
Effect size | Observed power | |
|---|---|---|---|---|---|---|---|---|---|
| ANOVA | |||||||||
| Activin A, pg/ml | 491.2 ± 165.3 | 559.0 ± 178.5 | 572.7 ± 167.0 | <0.001 | <0.001 | 0,544 | <0.001 | 0.044 | 0.990 |
| Follistatin, pg/ml | 1513 ± 586.3 | 1577.4 ± 523.1 | 1803.7 ± 802.5 | 0.312 | <0.001 | 0.006 | 0.001 | 0.029 | 0.925 |
| Activin A / follistatin | 0.36 ± 0.17 | 0.39 ± 0.19 | 0.37 ± 0.18 | 0.130 | 0.680 | 0.433 | 0.314 | 0.005 | 0.255 |
| cIMT, mm | 0.71 ± 0.17 | 0.71 ± 0.14 | 0.76 ± 0.17 | 0.017 | 0.022 | 0.021 | 0.015 | 0.018 | 0.738 |
| ANCOVA 1 | |||||||||
| Activin A, pg/ml | 508.6 ± 11.2 | 542.6 ± 12.4 | 561.2 ± 17.7 | 0.045 | 0.015 | 0.386 | 0.027 | 0.016 | 0.671 |
| Follistatin, pg/ml | 1536.8 ± 42.7 | 1547.7 ± 47.0 | 1804.3 ± 66.8 | 0.870 | 0.001 | 0.002 | 0.002 | 0.027 | 0.892 |
| Activin A / follistatin | 0.37 ± 0.01 | 0.39 ± 0.01 | 0.37 ± 0.02 | 0.273 | 0.818 | 0.267 | 0.416 | 0.004 | 0.201 |
| cIMT, mm | 0.69 ± 0.01 | 0.73 ± 0.01 | 0.75 ± 0.02 | 0.040 | 0.003 | 0.198 | 0.008 | 0.021 | 0.799 |
| ANCOVA 2 | |||||||||
| Activin A, pg/ml | 508.8 ± 11.3 | 542.4 ± 12.5 | 561.1 ± 17.7 | 0.052 | 0.016 | 0.385 | 0.032 | 0.015 | 0.649 |
| Follistatin, pg/ml | 1536.4 ± 42.8 | 1547.2 ± 47.2 | 1805.6 ± 66.5 | 0.870 | 0.001 | 0.001 | 0.002 | 0.027 | 0.899 |
| Activin A/follistatin | 0.37 ± 0.01 | 0.39 ± 0.01 | 0.36 ± 0.02 | 0.406 | 0.721 | 0.302 | 0.528 | 0.003 | 0.157 |
| cIMT, mm | 0.69 ± 0.01 | 0.73 ± 0.01 | 0.75 ± 0.02 | 0.036 | 0.003 | 0.190 | 0.007 | 0.022 | 0.818 |
Data were assessed by one-way analysis of variance (ANOVA). Analysis of covariance (ANCOVA) was used to adjust covariates. ANCOVA 1 adjusted confounders: age, sex, waist circumference, coronary artery disease history, anti-hypertensive medication, lipid lowering, current smoker and current alcohol. ANCOVA 2 adjusted confounders the same as ANCOVA 1 setting except body mass index instead of waist circumference. Values are mean ± standard deviation in ANOVA analysis. Values are estimated marginal mean ± standard error in ANCOVA analysis. cIMT: carotid intima-media thickness.