Table 3.
Univariable and multivariable linear regression analysis of global upslope in all DM patients (n = 120)
| Univariable | Multivariablea | ||||
|---|---|---|---|---|---|
| β | P value | β | P value | R2 | |
| Age (y)∆ | − 0.048 | 0.501 | – | – | 0.567 |
| Sex (female) | 0.351 | 0.000 | 0.372 | 0.000 | |
| BMI (kg/m2) | − 0.581 | 0.000 | − 0.516 | 0.000 | |
| Diabetes duration (year) | − 0.347 | 0.000 | − 0.169 | 0.022 | |
| HbA1c (%) | − 0.314 | 0.000 | − 0.184 | 0.010 | |
| TG (mmol/L) | − 0.153 | 0.031 | – | – | |
| SBP (mm Hg)∆ | 0.036 | 0.727 | – | – | |
| eGFR (mL/min) ∆ | 0.093 | 0.313 | – | – | |
| LV remodeling index | − 0.270 | 0.000 | – | – | |
BMI body mass index, eGFR estimated glomerular filtration rate, HDL high-density lipoprotein, SBP systolic blood pressure, TG plasma triglycerides
aCandidate variables for multivariable model were selected on clinical grounds, guided by univariable correlation with P value < 0.10 and the absence of collinearity. ∆Age, SBP, and eGFR was included in the multivariable analysis model based on clinical grounds. Medication history was included in the multivariable model but not listed because there was no significant correlation in univariable analysis