Copyright: ©Author(s) 2026.
World J Transplant. Sep 18, 2026; 16(3): 121821
Published online Sep 18, 2026. doi: 10.5500/wjt.121821
Published online Sep 18, 2026. doi: 10.5500/wjt.121821
Table 1 Features selected for modelling and their missingness percentage before imputation
| Features | Missing data full cohort (n = 15904) (%) | Missing data MP subgroup (n = 544) (%) |
| Recipient BMI | 19.98 | 22.61 |
| Serum creatinine (umol/L) 12-month post-transplant | 19.54 | 13.42 |
| Serum creatinine (umol/L) 3-month post-transplant | 15.79 | 13.79 |
| Recipient deprivation group | 16.17 | 31.25 |
| Predniscolone/predniscone immunosuppression 3 months post-transplant | 16.10 | 13.24 |
| Type of dialysis at transplant | 15.17 | 16.36 |
| Donor allergy | 12.09 | 10.85 |
| Initial graft function | 8.53 | 9.93 |
| Donor cause of death | 5.29 | 4.60 |
| Recipient positive CMV status | 4.89 | 5.33 |
| Donor family history of diabetes | 4.52 | 4.41 |
| Recipient ethnicity | 4.44 | 4.04 |
| Donor positive EBV status | 4.21 | 2.76 |
| Donor creatinine at retrieval | 3.41 | 5.88 |
| Donor ethnicity | 3.40 | 2.21 |
| Donor history of drug abuse | 2.50 | 3.68 |
| Donor history of cardiac disease | 2.18 | 4.23 |
| Donor history of UTI | 1.88 | 2.57 |
| Donor hypertension | 1.62 | 2.02 |
| Donor positive toxo status | 1.37 | 1.10 |
| Perfusate used | 1.08 | 0.74 |
| Donor history of diabetes | 0.91 | 1.29 |
| Perfusion quality | 0.74 | 2.94 |
| Donor positive CMV status | 0.59 | 1.10 |
| Graft status | 0.57 | 0 |
| Follow-up time | 0.55 | 0 |
| Donor BMI | 0.53 | 0.37 |
| Donor smoker | 0.52 | 0.55 |
| Dialysis status at registration | 0.35 | 0.18 |
| Cold ischemia time | 0.22 | 0.37 |
| Allocation matchability score | 0.18 | 0 |
| Waiting list time | 0.17 | 0.18 |
| Dialysis status at transplant | 0.13 | 0 |
| Kidney used by local centre | 0.10 | 0 |
| Donor ABO blood group | 0.04 | 0 |
| Recipient sex | 0.04 | 0.18 |
| Highly sensitised status flag | 0.03 | 0 |
| Donor homozygous at DR locus | 0.01 | 0 |
| Donor homozygous at B locus | 0.01 | 0 |
| Donor homozygous at A locus | 0.01 | 0 |
| HLA mismatch group | 0.01 | 0 |
| Recipient homozygous at B locus | 0.01 | 0 |
| Recipient homozygous at DR locus | 0.01 | 0 |
| Recipient homozygous at A locus | 0.01 | 0 |
| Number of mismatches at DR locus | 0.01 | 0 |
| Number of mismatches at A locus | 0.01 | 0 |
| Number of mismatches at B locus | 0.01 | 0 |
| Recipient age | 0 | 0 |
| Donor sex | 0 | 0 |
| Donor type | 0 | 0 |
| Recipient diabetes | 0 | 0 |
| Donor age | 0 | 0 |
| Recipient ABO blood group | 0 | 0 |
| Donor organ | 0 | 0 |
| Transplant year | 0 | 0 |
| Machine perfusion type | 0 | 0 |
- Citation: Castellanos De Brigard J, Rangganata E, Papalois VE. Machine perfusion distribution across clinical phenotypes in kidney transplantation: A national cohort study using unsupervised clustering. World J Transplant 2026; 16(3): 121821
- URL: https://www.wjgnet.com/2220-3230/full/v16/i3/121821.htm
- DOI: https://dx.doi.org/10.5500/wjt.121821