Table 4.
C-statistics, integrated discrimination improvement, and net reclassification improvement of restricted cubic spline “renal function” formulas for cardiovascular mortality discrimination and reclassification in the studied cohorts
| Cohorts | Model comparison (A vs. B) | C-index A% | C-index B% | ΔC-index% | P value | IDI% (95% CI) | P value | NRI% (95% CI) | P value |
|---|---|---|---|---|---|---|---|---|---|
| Population | Model vs. MDRD4 | 72.8 | 75.7 | 2.9 (−0.3 to 6.1) | 0.072 | 0.5 (−0.1 to 4.3) | 0.16 | 16.7 (−2.0 to 44.1) | 0.072 |
| Model vs. CKD-EPI | 72.8 | 75.2 | 2.4 (−1.6 to 6.4) | 0.23 | 0.6 (0.1 to 4.8) | 0.016 | 23.4 (−0.9 to 44.4) | 0.054 | |
| Model vs. CG-BSA | 72.8 | 77.3 | 4.5 (−0.7 to 9.7) | 0.087 | 1.0 (0.2 to 5.7) | 0.020 | 39.1 (4.6 to 51.2) | 0.020 | |
| MDRD4 vs. CKD-EPI | 75.7 | 75.2 | 0.5 (−2.4 to 3.4) | 0.74 | 0.2 (0.0 to 0.8) | 0.026 | 23.2 (−10.7 to 36.8) | 0.20 | |
| MDRD4 vs. CG-BSA | 75.7 | 77.3 | 1.6 (−2.8 to 6.1) | 0.48 | 0.5 (0.0 to 2.1) | 0.045 | 27.4 (−6.8 to 47.9) | 0.13 | |
| CKD-EPI vs. CG-BSA | 75.2 | 77.3 | 2.1 (−0.1 to 4.4) | 0.067 | 0.4 (−0.2 to 1.6) | 0.14 | 31.3 (−13.9 to 48.1) | 0.17 | |
| CV risk | Model vs. MDRD4 | 58.9 | 66.2 | 7.3 (5.2 to 9.5) | <0.001 | 0.9 (0.6 to 1.6) | <0.001 | 18.5 (13.0 to 24.3) | <0.001 |
| Model vs. CKD-EPI | 58.9 | 67.5 | 8.6 (6.4 to 10.9) | <0.001 | 1.1 (0.7 to 1.7) | <0.001 | 20.3 (15.3 to 25.8) | <0.001 | |
| Model vs. CG-BSA | 58.9 | 70.2 | 11.3 (8.9 to 13.7) | <0.001 | 1.5 (1.0 to 2.4) | <0.001 | 27.3 (21.4 to 32.6) | <0.001 | |
| MDRD4 vs. CKD-EPI | 66.2 | 67.5 | 1.3 (0.8 to 1.8) | <0.001 | 0.1 (0.1 to 0.2) | 0.006 | 23.0 (16.7 to 29.5) | <0.001 | |
| MDRD4 vs. CG-BSA | 66.2 | 70.2 | 4.0 (2.7 to 5.3) | <0.001 | 0.6 (0.3 to 0.9) | <0.001 | 22.8 (16.8 to 29.1) | <0.001 | |
| CKD-EPI vs. CG-BSA | 67.5 | 70.2 | 2.7 (1.7 to 3.7) | <0.001 | 0.4 (0.2 to 0.7) | <0.001 | 21.1 (14.6 to 27.6) | <0.001 | |
| Heart failure | Model vs. MDRD4 | 70.7 | 75.0 | 4.2 (1.9 to 6.5) | <0.001 | 2.7 (1.2 to 4.8) | <0.001 | 24.6 (15.9 to 34.1) | <0.001 |
| Model vs. CKD-EPI | 70.7 | 75.5 | 4.7 (2.3 to 7.1) | <0.001 | 3.0 (1.5 to 5.2) | <0.001 | 25.5 (15.5 to 34.1) | <0.001 | |
| Model vs. CG-BSA | 70.7 | 77.0 | 6.3 (3.6 to 8.9) | <0.001 | 4.4 (2.6 to 7.2) | <0.001 | 29.6 (20.1 to 39.9) | <0.001 | |
| MDRD4 vs. CKD-EPI | 75.0 | 75.5 | 0.5 (0.2 to 0.8) | 0.003 | 0.3 (0.1 to 0.5) | 0.006 | 12.7 (1.6 to 23.8) | 0.024 | |
| MDRD4 vs. CG-BSA | 75.0 | 77.0 | 2.1 (1.1 to 3.0) | <0.001 | 1.7 (0.8 to 2.7) | <0.001 | 24.9 (14.7 to 34.7) | <0.001 | |
| CKD-EPI vs. CG-BSA | 75.5 | 77.0 | 1.6 (0.8 to 2.4) | <0.001 | 1.4 (0.6 to 2.3) | <0.001 | 26.5 (14.4 to 34.5) | <0.001 | |
| Post-MI | Model vs. MDRD4 | 61.7 | 66.5 | 4.9 (4.2 to 5.6) | <0.001 | 3.4 (2.9 to 3.9) | <0.001 | 21.1 (19.4 to 22.8) | <0.001 |
| Model vs. CKD-EPI | 61.7 | 67.3 | 5.6 (4.9 to 6.4) | <0.001 | 3.8 (3.4 to 4.4) | <0.001 | 22.4 (20.6 to 24.0) | <0.001 | |
| Model vs. CG-BSA | 61.7 | 68.8 | 7.1 (6.3 to 7.9) | <0.001 | 4.9 (4.3 to 5.6) | <0.001 | 24.9 (23.0 to 26.5) | <0.001 | |
| MDRD4 vs. CKD-EPI | 66.5 | 67.3 | 0.8 (0.6 to 0.9) | <0.001 | 0.5 (0.4 to 0.6) | <0.001 | 16.4 (13.6 to 20.0) | <0.001 | |
| MDRD4 vs. CG-BSA | 66.6 | 68.8 | 2.3 (1.9 to 2.6) | <0.001 | 1.6 (1.3 to 1.9) | <0.001 | 20.9 (18.9 to 23.1) | <0.001 | |
| CKD-EPI vs. CG-BSA | 67.3 | 68.8 | 1.5 (1.3 to 1.8) | <0.001 | 1.1 (0.9 to 1.4) | <0.001 | 16.6 (13.8 to 20.4) | <0.001 |
CG-BSA Cockcroft-Gault formula adjusted for body surface area, MDRD4 modification of diet in renal disease-4 formula, CKD-EPI Chronic Kidney Disease Epidemiology Collaboration equation, CV cardiovascular, MI myocardial infarction, IDI integrated discrimination improvement, NRI net reclassification improvement, Δ delta (change)
All models adjusted for gender, smoking status, history of hypertension, diagnosis of diabetes, heart rate and systolic blood pressure (not adjusted for hemoglobin or left ventricular ejection fraction due to > 75% of missing values in all datasets)
All “renal function” formulas are analyzed using restricted cubic splines