From: Application of deep learning to predict the low serum albumin in new hemodialysis patients
Variables | Total | Mean albumin ≥ 3.5 g/dL | Mean albumin < 3.5 g/dL | P |
---|---|---|---|---|
Case no. (row%) | 1567 (100.0%) | 1283 (81.9%) | 284 (18.1%) | Â |
Age (years) | 67.0 ± 13.9 | 65.4 ± 13.5 | 74.2 ± 13.2 | < 0.001 |
Gender | Â | Â | Â | 0.002 |
 Female | 718 (45.8%) | 564 (44%) | 154 (54.2%) |  |
 Male | 849 (54.2%) | 719 (56%) | 130 (45.8%) |  |
Comorbidity | Â | Â | Â | Â |
 Diabetes mellitus |  |  |  | 0.015 |
 | 822 (52.5%) | 654 (51%) | 168 (59.2%) |  |
 Hypertension |  |  |  | 0.198 |
 | 1262 (80.5%) | 1025 (79.9%) | 237 (83.5%) |  |
 Heart Failure |  |  |  | 0.030 |
 | 326 (20.8%) | 253 (19.7%) | 73 (25.7%) |  |
 Caner |  |  |  | 0.003 |
 | 545 (34.8%) | 424 (33%) | 121 (42.6%) |  |
Mortality |  |  |  | < 0.001 |
 Alive | 1497 (95.5%) | 1254 (97.7%) | 243 (85.6%) |  |
 Dead | 70 (4.5%) | 29 (2.3%) | 41 (14.4%) |  |
Laboratory measurements | Â | Â | Â | Â |
 Hb, g/dL | 10.5 ± 1.3 | 10.6 ± 1.2 | 9.9 ± 1.5 | < 0.001 |
 Albumin, g/dL | 3.8 ± 0.4 | 3.9 ± 0.3 | 3.2 ± 0.4 | < 0.001 |
 Fe, ug/dL | 68.8 ± 29.5 | 70.1 ± 27 | 63.1 ± 38.5 | 0.004 |
 Ferritin, ug/dL | 479.7 ± 523.3 | 426.9 ± 395.3 | 718.3 ± 858.9 | < 0.001 |
 Na, mEq/L | 136.3 ± 3.3 | 136.5 ± 3.1 | 135.4 ± 3.8 | < 0.001 |
 K, mEq/L | 4.5 ± 0.7 | 4.6 ± 0.7 | 4.3 ± 0.7 | < 0.001 |
 Ca, mg/dL | 9.2 ± 0.8 | 9.3 ± 0.8 | 9.0 ± 0.9 | < 0.001 |
 Phosphate, mg/dL | 5.0 ± 1.4 | 5.1 ± 1.4 | 4.5 ± 1.5 | < 0.001 |
 BUN, mg/dL | 66.7 ± 19.9 | 67.5 ± 18.7 | 62.7 ± 24.5 | 0.002 |
 Cr, mg/dL | 10.1 ± 2.7 | 10.5 ± 2.6 | 8.0 ± 2.6 | < 0.001 |
 Alkaline phosphatase, U/L | 83.4 ± 52.4 | 77.6 ± 42.1 | 109.3 ± 79.6 | < 0.001 |
 iPTH, pg/dL | 207.6 (97.7–408.3) | 219.8 (105.9–423.6) | 163.7 (67.7–323.5) | 0.004 |
 Cholesterol, mg/dL | 163.6 ± 38.8 | 165.7 ± 38.0 | 153.8 ± 40.9 | < 0.001 |
 Triglyceride, mg/dL | 156.5 ± 121.2 | 160.3 ± 124.6 | 139.5 ± 103.1 | 0.003 |
 Fasting glucose (AC), mg/dL | 142.5 ± 77.3 | 139.4 ± 74.6 | 156.2 ± 87.3 | 0.003 |