Table 3.
Associations between circulating levels of activin B and demographic and biochemical indices within subject groups.
| Total subjects | Normal | IFG/IGT | T2D | |
|---|---|---|---|---|
| Age | 0.100 | −0.206 | 0.267 | 0.159 |
| BMI | 0.170 | 0.092 | 0.321 | 0.201 |
| Smoking | 0.160 | 0.029 | 0.356 | 0.200 |
| Fasting glucose | 0.298∗∗,† | 0.072 | 0.080 | 0.406* |
| HbA1c | 0.331∗∗,† | −0.193 | 0.292 | 0.529** |
| Fasting insulin | 0.254* | 0.051 | 0.510* | 0.293 |
| HOMA-IR | 0.332∗∗,† | 0.056 | 0.514* | 0.426∗,† |
| AUCg | −0.100 | 0.066 | −0.155 | −0.274 |
| Crcl | 0.009 | 0.098 | −0.121 | 0.004 |
| HsCRP | 0.120 | 0.021 | −0.243 | −0.119 |
| Follistatin | 0.026 | 0.008 | −0.010 | 0.033 |
BMI: body mass index; HbA1c: glycated hemoglobin; HOMA-IR: homeostasis model assessment of insulin resistance; AUCg: area under the curve for OGTT glucose levels; Crcl: creatinine clearance; HsCRP: human sensitive C-reactive protein. Data presented as univariate Pearson correlation coefficients (R) value: *P < 0.05, **P < 0.01. Correlations with no superscript are not significant (P > 0.05). †After adjustment for confounding factors (including age, BMI, history of smoking, history of hypertension and medication use, blood pressure parameters) by multivariate analysis, activin B retained a positive association among total subjects with fasting glucose, HbA1c, and HOMA-IR but lost its correlation with fasting insulin. In the T2D group, only HOMA-IR retained a positive correlation with activin A after adjustment.