Normothermic machine perfusion versus static cold storage in donation after circulatory death kidney transplantation: a randomized controlled trial

Abstract

Kidney transplantation is the optimal treatment for end-stage renal disease, but it is still severely limited by a lack of suitable organ donors. Kidneys from donation after circulatory death (DCD) donors have been used to increase transplant rates, but these organs are susceptible to cold ischemic injury in the storage period before transplantation, the clinical consequence of which is high rates of delayed graft function (DGF). Normothermic machine perfusion (NMP) is an emerging technique that circulates a warmed, oxygenated red-cell-based perfusate through the kidney to maintain near-physiological conditions. We conducted a randomized controlled trial to compare the outcome of DCD kidney transplants after conventional static cold storage (SCS) alone or SCS plus 1-h NMP. A total of 338 kidneys were randomly allocated to SCS (n = 168) or NMP (n = 170), and 277 kidneys were included in the final intention-to-treat analysis. The primary endpoint was DGF, defined as the requirement for dialysis in the first 7 d after transplant. The rate of DGF was 82 of 135 (60.7%) in NMP kidneys versus 83 of 142 (58.5%) in SCS kidneys (adjusted odds ratio (95% confidence interval) 1.13 (0.69-1.84); P = 0.624). NMP was not associated with any increase in transplant thrombosis, infectious complications or any other adverse events. A 1-h period of NMP at the end of SCS did not reduce the rate of DGF in DCD kidneys. NMP was demonstrated to be feasible, safe and suitable for clinical application. Trial registration number: ISRCTN15821205 .

Fig. 1. CONSORT diagram.

Two eligibility periods are presented due to the COVID-19 pandemic: 4 September 2020 is the date the trial officially closed to recruitment, and 23 March 2020 was the start of the UK national lockdown. No participants were recruited between 23 March 2020 and 4 September 2020. Two participants did not have a cold ischemic time (CIT) reported, and two participants received dual transplants and, hence, do not have a left/right kidney variable populated; these four cases were excluded from all risk-adjusted modeling. **Two additional exclusion criteria (donors who underwent normothermic regional perfusion or one of a pair already randomized as a single kidney in the trial) were introduced on 13 October 2017, and, retrospectively, two participants fulfilled these exclusion criteria (both were randomized to NMP). These patients were included in the MITT analysis and excluded from the per-protocol analysis.

Fig. 2. Renal function in renal transplant recipients receiving donor kidneys randomized to preservation by SCS or NMP.

Serum creatinine levels (a) and eGFR levels (b) presented as unadjusted mean ± s.e.m. at baseline (timepoint 0) and adjusted mean ± s.e.m. for 1, 3, 6 and 12 months after transplant.

Fig. 3. Patient and allograft survival in renal transplant recipients receiving donor kidneys randomized to preservation by SCS or NMP.

Kaplan–Meier plot for 12-month patient survival (a) and graft survival (b).