Table 3.
Multiple Linear Regression Analyses with Backward Elimination to Identify Markers that Predict Carotid Arterial Wall inflammation in PAD patients with concomitant Diabetes
| Standardized coefficient β | 95 % confidence interval | p-value | |
|---|---|---|---|
| Glucose-Corrected TBRmean | |||
| Circulating Insulin | 0.541 | 0.514–1.458 | 0.001 |
| Hba1c | 0.537 | 1.014–2.903 | 0.001 |
| Glucose-Corrected TBRmax | |||
| Circulating Insulin | 0.552 | 0.550–2.049 | 0.002 |
| Hba1c | 0.477 | 0.748–3.748 | 0.006 |
| Glucose-Corrected TBRmds | |||
| Circulating Insulin | 0.536 | 0.723–4.007 | 0.008 |
| Hba1c | 0.480 | 0.501–2.142 | 0.004 |
Whole Carotid glucose Corrected Target-to-Background ratio (TBRmean), maximum uptake (TBRmax) and Most Diseased Segment (TBRmds) were the response variables. Explanatory factors comprised cardiovascular risk factors (Age, Gender, BMI, smoking, Systolic Blood Pressure) and factors associated with Diabetes regulation (Insulin and Hba1c, both log-transformed). Variables were retained in the model when P < 0.10. Data are standardized coefficient (β) with 95 % confidence intervals (CI)