Background
Methods
Data sources
Laboratory methods
Mathematical methods
Gender | Serum creatinine (mg/dl) | Equation for estimating GFR |
---|---|---|
Female | ≤ 1.2 |
\(92 \times (SC \div 1.2)^{ - 0.534} \times 0.994^{Age}\)
|
Female | > 1.2 |
\(79 \times (SC \div 1.2)^{ - 0.516} \times 0.994^{Age}\)
|
Male | ≤ 1.0 |
\(98 \times (SC \div 1.0)^{ - 0.450} \times 0.996^{Age}\)
|
Male | > 1.0 |
\(105 \times (SC \div 1.0)^{ - 0.640} \times 0.993^{Age}\)
|
Statistical analysis
Results
Study population
Characteristic | Development (N = 1002) | External validation (N = 417) | P value#
|
---|---|---|---|
Age (years) | 55.7 ± 15.0 | 51.3 ± 16.0 | < 0.001 |
Male proportion | 570 (56.9) | 262 (62.8) | 0.044 |
Diabetes | 500 (49.9) | 97 (23.2) | < 0.001 |
Body mass index (kg/m2) | 24.0 ± 3.7 | 22.9 ± 3.6 | < 0.001 |
Weight (kg) | 63.7 ± 12.3 | 61.5 ± 11.8 | 0.002 |
Height (cm) | 162.5 ± 8.3 | 163.6 ± 7.6 | 0.020 |
Body-surface area (m2) | 1.7 ± 0.2 | 1.7 ± 0.2 | 0.107 |
Serum creatinine (mg/dl) | 1.7 ± 1.8 | 2.7 ± 2.5 | < 0.001 |
Measured GFR | |||
Mean (ml/min/1.73 m2) | 70.0 ± 29.6 | 53.4 ± 26.5 | < 0.001 |
< 15 (ml/min/1.73 m2) | 10 (1.0) | 9 (2.2) | < 0.001 |
≥ 15 and < 30 (ml/min/1.73 m2) | 99 (9.9) | 94 (22.5) | |
≥ 30 and < 60 (ml/min/1.73 m2) | 275 (27.4) | 149 (35.7) | |
≥ 60 and < 90 (ml/min/1.73 m2) | 345 (34.4) | 123 (29.5) | |
≥ 90 (ml/min/1.73 m2) | 273 (27.2) | 42 (10.1) |
Overall performance of GFR estimation models in the external validation data set
Variable | Measured GFR (ml/min/1.73 m2) | |||
---|---|---|---|---|
Overall | < 30 | ≥ 30 and < 60 | ≥ 60 | |
Bias = median difference (95% CI) | ||||
Regression model | 2.3 (1.0–3.4) | 4.4 (2.9–5.9) | 3.1 (1.5–6.5) | −1.9 (−4.5 to 0.9) |
ANN model | 3.2 (2.2–5.4) | 5.4 (3.1–7.4) | 5.4 (2.4–7.6) | 0.8 (−3.9 to 2.7) |
SVM model | 3.6 (2.6–4.9) | 6.8 (4.9–9.0)‡
| 4.0 (2.2–6.4) | −0.2 (−2.3 to 2.6) |
Ensemble model | 3.4 (2.3–4.4) | 5.6 (3.7–8.2) | 4.0 (2.1–6.7) | −0.5 (−3.9 to 2.3) |
Precision = IQR of the difference (95% CI) | ||||
Regression model | 14.0 (12.4–15.9) | 9.2 (7.3–11.8) | 13.5 (11.2–18.0) | 19.6 (16.8 to 23.5) |
ANN model | 15.1 (13.6–17.0)‡
| 11.1 (9.1–14.8)‡
| 14.9 (13.1–17.7)‡
| 20.5 (17.9 to 25.1)‡
|
SVM model | 14.2 (12.4–16.0)‡
| 9.5 (7.5–12.1)‡
| 12.9 (10.3–16.2)‡
| 18.5 (14.9 to 21.5)‡
|
Ensemble model | 13.5 (11.8–14.9)‡
| 8.9 (7.0–11.0)‡
| 12.7 (10.4–16.0)‡
| 17.9 (15.44 to 21.9)‡
|
Accuracy = 30% accuracy (95% CI) | ||||
Regression model | 75.1 (70.7–79.4) | 52.4 (42.7–61.2) | 75.2 (67.8–81.9) | 89.1 (83.6 to 93.3) |
ANN model | 73.4 (69.0–77.2) | 54.4 (44.7–64.1) | 70.5 (63.11–77.2) | 87.9 (82.4 to 92.1) |
SVM model | 73.1 (68.8–77.2) | 47.6 (37.9–57.3) | 71.1 (63.11–77.9) | 90.9 (86.1 to 94.5) |
Ensemble model | 75.5 (71.5–79.6) | 52.4 (42.7–62.1) | 73.8 (65.11–79.9) | 91.5 (86.7 to 95.2) |