A Short Period of Normothermic Machine Perfusion May Not Be Able to Predict Primary Nonfunction in Uncontrolled Circulatory Death Kidneys.

Abstract

The application of normothermic machine perfusion (NMP) machine technology as an assessment device may help to increase the utility of kidneys from uncontrolled donation after circulatory death (uDCD) donors. Here, we describe a case in which NMP perfusion characteristics were misleading and failed to predict primary nonfunction in 2 kidneys from a uDCD. The donor was a 37-y-old man who died of an intracerebral hemorrhage. Before withdrawal of life supporting treatment the donor had a cardiac arrest. He was rapidly transferred to the operating room and the abdominal organs cold perfused in situ with University of Wisconsin solution at 4°C. The first warm ischemic time was 27 min. Both kidneys appeared very poorly perfused and were further flushed on the back table with 4 L of hyperosmolar citrate solution over a period of 90 min. The appearance of both kidneys improved but they still remained patchy. The kidneys underwent a 1-h period of NMP as previously described.1,2 During NMP, the macroscopic appearance, renal blood flow, and urine output of both kidneys met our published criteria for transplantation2 (Table 1). The kidneys were transplanted into 2 separate nonsensitized recipients (Table 1). Immunological cross-matching was negative for both transplants. Both kidneys had a single renal artery and vein and these were anastomosed to the external iliac vessels. There was no intraoperative hypotension. A baseline biopsy taken from the left kidney 30 min after transplantation showed severe ischemic/reperfusion injury with almost all glomerular and peritubular capillaries filled with sludged red cells. Percutaneous needle core biopsies of both kidneys at day 8 showed infarction of the entire specimen. Computerized tomography angiography demonstrated patency of the main renal arteries and veins but diffuse poor parenchymal enhancement. In the absence of any improvement in renal function, primary nonfunction (PNF) was diagnosed in both kidneys at 3 mo. TABLE 1. - Recipient characteristics, preservation times, and normothermic machine perfusion parameters of the left and right kidney Left kidney Right kidney Recipient characteristics Gender/age Male 44 y Male 43 y Cause of ESRF APCKD Membranous GN Dialysis Predialysis Predialysis eGFR (mL/min) 9 10 HLA mismatch 1-0-1 1-0-1 Preservation times CIT (first) min 177 291 NMP min 67 60 CIT (second) min 158 433 Anastomosis min 47 55 Total CIT min 335 724 NMP parameters Macroscopic appearance 2 1 Mean RBF (mL/min/100g) 65.6 112.8 Total urine output (mL) 45 65 Oxygen consumption (mL/min/g) 36.6 48.0 Pre-NMP Post-NMP Pre-NMP Post-NMP pH 7.42 7.43 7.51 7.53 Potassium (mmol/L) 10.8 15.4 8.4 13.2 Lactate (mmol/L) 10.4 8.2 7.3 7.5 Glucose (mmol/L) 10.6 7.2 10.4 5.2 Recipient characteristics; gender/age, cause of ESRF, dialysis, eGFR, and HLA mismatch. Preservation times; CIT, NMP, second CIT, and anastomosis time. NMP parameters; macroscopic appearance (1, excellent; 2, patchy; 3, poor), mean RBF, total urine output, oxygen consumption after 60 min, prearterial and postarterial NMP levels of pH, potassium, lactate, and glucose. APCKD and membranous GN.APCKD, autosomal dominant polycystic kidney disease; CIT, cold ischemic time; eGFR, estimated glomeruli filtration rate; ESRF, end-stage renal failure; membranous GN, membranous glomerulonephritis; NMP, normothermic machine perfusion; RBF, renal blood flow. Extensive red cell plugging at revascularization led to cortical necrosis and allograft quality assessment by NMP failed to predict PNF. The young age of the donor and the short ischemic times were favorable factors. However, inadequate clearance of the microcirculation after in situ and back-table flushing is a well-described complication of prolonged warm-ischemic injury in uDCD organs.3,4 NMP theoretically offers a more accurate pretransplant viability assessment, because it restores cellular metabolism.2 We have used NMP to successfully assess and salvage kidneys from controlled donation after circulatory death donors because of inadequate clearance of the microcirculation after in situ cooling.2 This is the first report of applying the same principle to uDCD kidneys, and the adverse outcome is admonitory. The failure of NMP to predict PNF in these cases may be related to the duration and nature of the perfusion protocol. A 1-h period of NMP has been the standard in our practice and its effect on delayed graft function is currently being trialed in donation after circulatory death kidneys.5 However, cortical necrosis takes some hours to evolve and a more extended period of NMP might have been more predictive of the outcome. Furthermore, NMP conditions are designed to be protective and could mask the actual level of ischemic injury within a short timeframe. In conclusion, NMP strategies need to be further developed to reliably salvage and assess kidneys with severe ischemic injury complicated by poor clearance of the microvasculature.