Table 3.
Associations between serum albumin and renal outcomes.
| Hazard ratios (95% confidence interval) & p value | |||||
|---|---|---|---|---|---|
| Serum albumin, median (range) (g/L) | Unadjusted | Model 1a | Model 2b | Model 3c | |
| Per 1 SD serum albumin | 34.15 (14.00-48.80) | 0.35 (0.27-0.46) p < 0.001 |
0.18 (0.06-0.58) p = 0.004 |
0.20 (0.06-0.66) p = 0.008 |
0.21 (0.06-0.67) p = 0.009 |
| Normal group | 40.20 (35.10-48.80) | Reference | Reference | Reference | Reference |
| Mild group (35-30 g/L) | 33.25 (30.30-34.90) | 2.99 (1.36-6.61) p = 0.007 |
1.45 (0.49-4.31) p = 0.501 |
2.24 (0.72-6.90) p = 0.162 |
2.09 (0.67-6.56) p = 0.205 |
| Moderate group (25-30 g/L) | 27.50 (25.20-29.90) | 6.03 (3.05-11.95) p < 0.001 |
4.67 (1.59-13.74) p = 0.005 |
6.56 (2.05-20.94) p = 0.001 |
6.20 (1.95-19.76) p = 0.002 |
| Severe group (<25 g/L) | 22.00 (14.00-24.80) | 13.74 (6.63-28.44) p < 0.001 |
6.14 (1.27-29.66) p = 0.024 |
10.61 (1.87-60.07) p = 0.008 |
7.37 (1.24-43.83) p = 0.028 |
Serum albumin was analyzed as a continuous variable with hazard ratios (HRs) calculated per SD increment. SD: standard deviation. aModel 1 adjusted for the baseline age, gender, duration of diabetes, DR (yes or no), hypertension (yes or no), hematuria, total cholesterol, hemoglobin, and log proteinuria and e-GFR. bModel 2 adjusted for covariates in model 1 plus renal pathological findings (the glomerular class, IFTA, interstitial inflammation scores, and arteriolar hyalinosis). cModel 3 adjusted for covariates in model 2 plus RAAS inhibitor use (yes or no).