Share this article on:

Oxygen Consumption During Oxygenated Hypothermic Perfusion as a Measure of Donor Organ Viability

Bunegin, Leonid*; Tolstykh, Gleb P.*; Gelineau, Jerry F.*; Cosimi, A. Benedict; Anderson, Lisa M.

doi: 10.1097/MAT.0b013e318292e865
Kidney/Dialysis/Vascular Access

Hypothermic machine perfusion (HMP) for the preservation of kidneys, recovered from extended criteria organ donors (ECDs), presents the opportunity for assessing ex vivo parameters that may have value in predicting postimplantation organ viability. Organ perfusion and vascular resistance are the parameters most frequently cited as the basis for the decision to use or discard a donor kidney. The limitation of these measures is emphasized by the observation that a significant percentage of ECD kidneys with poor perfusion parameters can provide life-sustaining function after transplantation. It has been suggested that whole organ oxygen consumption (OC) during oxygenated HMP may better reflect the proportion of viable tissue in the organ and more reliably predict posttransplant organ function. Our study correlates renal OC and renal vascular resistance (RVR) during oxygenated HMP with postpreservation glomerular filtration rates (GFRs) in rodent kidneys after 24 hours of oxygenated HMP. Kidneys from adult rodents were preserved for 24 hours using oxygenated HMP and static cold storage (SCS). During oxygenated HMP preservation, organ OC, renal organ flow rates, and RVR were serially measured. After the preservation period, organs were mounted onto a Langendorff device for warming to normal body temperature and measurement of GFR. Oxygen consumption and RVR during HMP were correlated with postpreservation GFR. Oxygen consumption during oxygenated HMP was significantly correlated (r2 =0.871; p < 0.05) with postpreservation GFR, suggesting that higher OC predicts better postpreservation GFR. In contrast, RVR was poorly correlated with postpreservation GFR (r2 =0.258; p = 0.199). Glomerular filtration rate in SCS kidneys was 0.002 ± 0.003 ml/min/g. We demonstrate that measurement of organ OC during oxygenated HMP may have significant value in predicting postpreservation organ function.

From the *Department of Anesthesiology, University of Texas Health Science Center, San Antonio, Texas; Department of Surgery, Transplantation Unit, Harvard Medical School at Massachusetts General Hospital, Boston, Massachusetts; Paragonix Technologies Inc., Cambridge, Massachusetts.

Submitted for consideration September 2012; accepted for publication in revised form March 2013.

Disclosure: The authors have no conflicts of interest to report.

Reprint Requests: Leonid Bunegin, Department of Anesthesiology, University of Texas Health Science Center, San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229-3900. Email:

With the increasing use of more marginal donor organs, hypothermic machine perfusion (HMP) has gained greater clinical acceptance for kidney preservation over the past decade. Currently, >35% of recovered kidneys are being perfused before transplantation. This increased utilization of HMP is based on definitive scientific and clinical evidence, showing that HMP is a superior preservation method compared with static storage on ice in most donor categories.1–4 Specifically, transplantation of machine-perfused organs results in superior clinical short-term and long-term outcomes.1–4 This improvement is especially evident in the ever-expanding pool of extended criteria donors (ECDs) and donation after cardiac death (DCD). Preservation methods based on HMP techniques have solidified the notion that the delivery of necessary nutrients by perfusion of deceased donor kidneys is superior to static cold storage (SCS).2,3,5–7

An important advantage of HMP is the opportunity for monitoring perfusion parameters that might predict posttransplant organ viability.8 It is a simple task to measure various perfusion parameters and obtain perfusate samples for biomarker assessment. Although several biomarkers indicative of kidney viability, such as lactate dehydrogenase, glutathione S-transferase, and aminotransferase, in the perfusate have been suggested,9–13 the perfusion parameters of flow rate and organ vascular resistance have been most frequently used to predict posttransplant function. Because the predictive value of renal vascular resistance (RVR) has been shown to be only moderate, the current discard rate of up to 40% of ECD and DCD kidneys likely results in the loss of many potentially transplantable organs.14 Although poor perfusion parameters have been shown to correlate with delayed renal graft function, Guarrera et al.15 have demonstrated that a significant percentage of ECD kidneys with poor perfusion parameters do indeed have the potential to provide life-sustaining function after transplantation. Sonnenday et al.16 caution that when organ flow and RVR are used without consideration of donor and recipient clinical factors that could predict primary nonfunction (PNF), the decision to discard an organ may be inappropriate.

We and others have previously suggested that oxygenated perfusion will provide superior preservation and yield higher quality organs and extended preservation times.17–22 Despite a much reduced metabolic rate during periods of hypothermia, approximately 10% of the organ’s metabolism continues to take place and oxygen utilization by organs at low temperature is measurable. Indeed, provision of oxygen, either by direct oxygenation or by oxygen extraction from air-equilibrated perfusate, is an important and attractive feature of HMP.23,24 Because irreversible structural damage of hypothermic, nonoxygenated kidneys mostly occurs during reperfusion, warming, and recovery of oxidative phosphorylation, the provision of oxygen, and hence maintenance of energetic homeostasis, during preservation is desirable.23 An additional benefit of active oxygenation is that much lower perfusion pressures can be used, while still providing a sufficient amount of oxygen. Perfusate equilibrated with air results in oxygen content of 0.009 ml O2/ml perfusate, as such, flow in excess of 200 ml/min is necessary for provision of oxygen for metabolic needs. Thus, in organs having a vascular resistance of 0.3, a perfusion pressure of 60 mm Hg is not uncommon. By oxygenating perfusate to a PaO2 of, as little as, 400 mm Hg, oxygen content of 0.024 ml O2/ml perfusate is possible. A threefold increase in perfusate oxygen content reduces by threefold, the flow needed to maintain metabolic needs. Thus, in organs having a vascular resistance of 0.3, a perfusion pressure of ≤20 mm Hg is possible. Porcine kidneys perfused for 20 hours using a system with pulsatile perfusion and oxygenation resulted in better kidney functional recovery after transplantation.24 Low-flow oxygenated HMP is possible and has been shown to be advantageous in predamaged kidneys, improving graft viability after transplantation.25 Short-term oxygenated HMP has been shown to recondition kidneys stored on ice and improve postischemic graft function upon warm reperfusion compared with ice-stored or nonoxygenated HMP-perfused kidneys.26

Indeed, Weegman et al.27 recently published data on the use of whole organ oxygen consumption (OC) during oxygenated hypothermic machine perfusion as a means to predict posttransplant organ viability. They hypothesized that healthy, viable tissue will extract oxygen from the inflow perfusate to maintain metabolism commensurate with the storage temperature. They further investigated whether OC is a reflection of tissue viability in the organ. As such, measuring organ OC during HMP may be more accurate in predicting which organs would become problematic grafts after transplantation and, most importantly, prevent organs from being discarded that in reality would become functional in the recipient.

This study investigates the correlations between OC and RVR during oxygenated hypothermic perfusion and postpreservation glomerular filtration in rodent kidneys.

Back to Top | Article Outline

Materials and Methods

Kidney Recover Protocol

The experimental protocol was reviewed and approved by the Institutional Animal Care and Usage Committee. Animal usage complied with the “Principles of Laboratory Animal Care” (National Institutes of Health Publication No. 86-23). Adult Sprague-Dawley albino rats weighing 350–500 g were anesthetized by intravenous injection with 25 ml/kg of sodium pentobarbital. Rodents were intubated and ventilated with 40% oxygen to maintain normal arterial blood oxygenation. In the supine position, kidneys were exposed via a midline incision in the lower abdominal cavity. The rodent was heparinized, followed by the insertion of a catheter into the descending aorta above the kidneys. A second catheter was inserted into the inferior vena cava just below the kidneys. After cross-clamping of the aorta and inferior vena cava above and below the catheters, an infusion of cold standard University of Wisconsin Solution (UWS) at 4°C was initiated. Infusion was terminated after all blood was cleared from the isolated organs (approximately 100 ml at a perfusion pressure not exceeding 20 mm Hg). Cold saline (4°C) was poured over the kidneys during infusion and removed by suction. The aorta and inferior vena cava were ligated at the cross-clamp and then cut, as were the ureters. The kidneys were quickly dissected free and placed on ice for catheterization of the renal arteries, veins, and ureters.

Back to Top | Article Outline

Preservation Protocol

Each kidney (n = 8) was attached to the organ preservation device (previously described by Tolstykh et al.20) via a renal arterial cannula and immersed into cold (4°C) freshly prepared standard UWS (Figure 1). The renal venous catheter was attached to a two way port that could be switched so that the outflow perfusate from the kidney could be collected either externally or recirculated back to the organ storage chamber. Perfusion was initiated at 70 pulses/min at approximately 0.6 ml/min/g. The preservation device was placed into Styrofoam case into which three cold packs (7.0″ × 1.5″ × 6.75″ cooling packs stored at −20°C for 12 hours) had been previously placed. During the subsequent 24 hour preservation period, temperature, renal arterial and venous pressures were monitored continuously. Samples of the preservation solution flowing into and out of the organs were taken at 60 minute intervals for oxygen partial pressure measurements. Renal perfusion was measured by collecting renal venous outflow from each kidney for 15 seconds and adjusting to flow/minute. Renal vascular resistance was calculated by dividing the mean renal perfusion pressure by the renal perfusion, then normalizing this result by the wet weight of the organ at the start of preservation.

where Rmpp is the mean renal perfusion pressure (mm Hg);

Rperf is the renal perfusion (ml/min);

Owt is the organ weight (g).

Oxygen consumption was calculated by: 1) multiplying the difference between the pO2 of the delivered UWS and the pO2 of the organ outflow by the oxygen solubility factor per milliliter of solution at 4°C,28 2) multiplying the result by the organ perfusion in ml/min, and 3) normalizing this result by the wet weight of the organ at the start of preservation.

where ΔpO2 is the difference between the pO2 of the delivered UWS and the pO2 of the organ outflow (mm Hg);

SO2 is 6 × 10–5 ml O2/ml/mm Hg, the oxygen solubility per ml of solution at 4°C28;

Rperf is the renal perfusion (ml/min);

Owt is the organ weight (g).

Static cold storage controls (n = 4) were prepared and stored identically as the perfused organs, except that perfusion was not initiated.

Back to Top | Article Outline

Organ Function Measurement Protocol

After the preservation period, kidneys were removed from the preservation device and connected to a Langendorff device via the renal artery catheter (Figure 2). While on this device, the rodent organs were perfused with warm (37°C) oxygenated (100% O2) Krebs-Henseleit (K-H) solution containing inulin (15 mg/100 ml). Perfusion was initiated slowly and incremented at 5 minute intervals until a mean arterial pressure of 100 mm Hg was achieved. Urine, arterial and venous samples were collected from each kidney at 7, 30, 60, 90, 120, and 150 minutes for inulin clearance and urine output measurement. Inulin was measured using the method of Walser as modified by Brown and Nolph.29 The PaO2 in the K-H entering the renal arteries and exiting the renal veins was measured on a Irma TruPoint blood gas machine (International Technidyne Corporation, Edison, NJ). Organ perfusion was measured by collecting the outflow from the renal veins during a 15 second time interval and adjusting to flow/minute. Oxygen consumption was calculated by Equation 2 using 3.0 × 10–5 ml O2/ml/mm Hg, the oxygen solubility factor per ml of solution at 37°C,28 and RVR was calculated as described above in Equation 1.

Back to Top | Article Outline

Measurement of Glomerular Filtration Rate

where Uin is the inulin concentration in the urine (mg/ml);

Vur is the urine flow (ml/min);

Pin is the inulin concentration in the plasma (mg/ml);

Owt is the organ weight (g).

Back to Top | Article Outline

Statistical Analysis

Regression analysis was performed using CurveExpert Professional 1.5.0 (Copyright © 2012 Daniel G. Hyams). Glomerular filtration rate (GFR) as a function of time during reperfusion on the Langendorff was analyzed by least squares nonlinear regression modeling with determination of the correlation coefficient.

Glomerular filtration rate and RVR as a function of OC during HMP were analyzed by least squares linear regression modeling with determination of correlation coefficients. Goodness of fit was estimated by an F statistic. p < 0.05 rejects the null hypothesis, and 95% confidence bands and 95% prediction bands were also determined. Physiologic and perfusion data are represented as means ± standard deviations.

Back to Top | Article Outline


The pO2 of the UWS entering the kidneys during HMP steadily increased from approximately 150 mm Hg to a mean of 361 ± 110 mm Hg by 24 hours. The pO2 of the perfusate exiting the organs was approximately 80 mm Hg and steadily rose to 291 ± 117 mm Hg during the 24 hour perfusion period. The arterial-venous (A-V)pO2 was relatively constant throughout the perfusion period at 70.3 ± 30.2 mm Hg. Mean oxygen delivery to the HMP kidneys was 0.107 ± 0.036 ml O2/min (582.4 nM O2/min/g). Kidney weight, temperature, perfusion pressure, perfusate flow, OC, and vascular resistance during oxygenated HMP are summarized in Table 1.

All kidneys reached stable GFR within 30 minutes of rewarming after attachment to the Langendorff device (Figure 3). During functional testing on the Langendorff, the pO2 of the K-H solution entering the renal artery was 392 ± 78 mm Hg and exiting the renal vein was 272.0 ± 81.0 mm Hg. The A-VpO2 was stable during the 180 minutes of testing at 120.0 ± 34.1 mm Hg. Kidney weight, temperature, perfusion pressure, perfusate flow, OC vascular resistance, and GFR during Langendorff testing are summarized in Table 2. We first analyzed the correlation between RVR during oxygenated perfusion preservation and postpreservation organ function as determined by GFR measured on the Langendorff device. The results showed that the correlation was poor (r2 = 0.258; p=0.199; Figure 4). We then analyzed the correlation between OC and postpreservation organ function. Oxygen consumption by the kidneys during oxygenated HMP correlated highly (r2 =0.817; p<0.05) with postpreservation function of the organs (Figure 5). Both the 95% confidence and predictive band for the RVR versus GFR correlation were significantly broader than for OC versus GFR correlation.

Glomerular filtration rate in SCS kidneys was 0.002 ± 0.003 ml/min/g. Additional observations are summarized in Tables 1 and 2.

Back to Top | Article Outline


The utilization of HMP for primary kidney preservation, and, more recently, as a potential approach to ex vivo resuscitation of “marginal” kidneys, has stimulated extensive ex vivo studies seeking to identify the perfusion parameters that most reliably predict posttransplant organ viability.

Here, we show a significant correlation between OC of rodent kidneys during oxygenated HMP and postpreservation function as measured by GFR using a Langendorff device. This correlation was much more consistent than that provided by measurements of vascular resistance in the same organs. Our findings suggest that 1) kidneys preserved by oxygenated HMP use the oxygen dissolved in the perfusion solution and 2) this OC is highly correlated with GFR.

Monitoring whole organ OC during oxygenated HMP preservation as a means to predict posttransplant organ viability has been recently suggested by Weegman et al.27 Their data similarly showed that oxygenated HMP porcine kidneys were extracting oxygen. After the infusion of formalin, the kidneys ceased to extract oxygen as evidenced by a drop in A-VpO2 to nearly zero. An equally important finding was the demonstration that in nonoxygenated HMP, the dissolved oxygen in the perfusate that is in equilibrium with air is extracted within minutes, leaving the perfused organ in a hypoxic condition.27 This finding may explain why RVR is a moderate-to-poor predictor of organ viability. It may also provide insight into why some HMP-perfused organs develop delayed graft function or PNF. Nonoxygenated HMP is similar to SCS, in that all available dissolved oxygen is rapidly consumed by the organs resulting in no further OC. The significantly lower postpreservation GFR in our control SCS group appears to reflect the lack of oxygen during the preservation period.

If a sufficient level of dissolved oxygen is present in the inflow perfusate, healthy, viable tissue will extract only the amount of oxygen needed to support a metabolism commensurate with the storage temperature. Because the effect of hypothermia on organ metabolism is known, it becomes relatively easy to estimate the expected OC for the entire organ during HMP. If the estimate and measured oxygen utilization are similar, the viability of the organ tissue can be presumed to be high. Oxygen utilization less than the calculated estimate may suggest reduced viability. The underlying hypothesis suggests that OC during oxygenated HMP reliably reflects the proportion of viable tissue in the organ. Further work will be needed to more precisely define this correlation.

In separate studies of canine heart preservation by oxygenated HMP, Calhoon et al.17 demonstrated in a transplant model that hearts which consumed oxygen at 97% of the calculated value expected for healthy hearts stored at 5°C for 12 hours had ventricular contractility that was statistically the same as freshly harvested controls. Bunegin et al.18 demonstrated in an isolated organ model that the left ventricular stroke work in hearts that had an OC rate of 100% of the value expected for healthy hearts stored at 7°C was also not different from that observed in freshly harvested hearts. The work of Tolstykh et al.20 showed that rodent kidneys consumed oxygen at a rate equal to the expected consumption of healthy kidneys at 7°C during 24 hours of oxygenated HMP. After rewarming, the GFR in these kidneys was not statistically different from freshly harvested organs as measured in an isolated organ preparation. Most recently, Tolstykh et al.22 demonstrated similar findings in canine kidneys preserved 24 hours using oxygenated HMP. These studies support the notion that maintenance of organ metabolism during perfusion preservation by oxygenated HMP is critical to protect postpreservation function and that monitoring of OC during the preservation period provides a reliable measure of organ viability.

The findings from this study corroborate the findings of others that show RVR is, at best, moderately correlated with postpreservation function.15,30,31 They also emphasize that, even during oxygenated HMP, RVR may not be a better predictor of postpreservation function than during nonoxygenated HMP (Figure 4). On the contrary, OC during oxygenated HMP is highly correlated with postpreservation organ function, is linearly related, and appears to have a tighter 95% confidence and predictive band than RVR (Figure 5). Although the sample size in this study is too small for broad generalization, the correlation is strong and may indeed reflect the relationship for organ systems in general. Other organ systems will probably have their own unique correlation between tissue viability and OC during oxygenated HMP and will be a function of their basal metabolic rate.

We propose that a true marker of viability should be consistent throughout the preservation time. Decline in or low O2 consumption after stabilization of the organ during oxygenated HMP may indicate tissue damage or lack of viability. The concept that viable organ tissue will extract oxygen from the perfusate during HMP is logical. Further studies in human organs will be needed to investigate the correlation between OC and viability and posttransplant function.

Back to Top | Article Outline


1. Wszola M, Kwiatkowski A, Latek M, et al. Long term medical and economical benefit of machine perfusion (MP) kidney storage in comparison to cold storage (CS). Ann Transplant. 2009;14:24–29
2. Wight J, Chilcott J, Holmes M, Brewer N. The clinical and cost-effectiveness of pulsatile machine perfusion versus cold storage of kidneys for transplantation retrieved from heart-beating and non-heart-beating donors. Health Technol Assess. 2003;7:1–94
3. Moers C, Smits JM, Maathuis MH, et al. Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med. 2009;360:7–19
4. Sung RS, Christensen LL, Leichtman AB, et al. Determinants of discard of expanded criteria donor kidneys: Impact of biopsy and machine perfusion. Am J Transplant. 2008;8:783–792
5. Hoffmann RM, Stratta RJ, D’Alessandro AM, et al. Combined cold storage-perfusion preservation with a new synthetic perfusate. Transplantation. 1989;47:32–37
6. Wight JP, Chilcott JB, Holmes MW, Brewer N. Pulsatile machine perfusion vs. cold storage of kidneys for transplantation: A rapid and systematic review. Clin Transplant. 2003;17:293–307
7. Schold JD, Kaplan B, Howard RJ, Reed AI, Foley DP, Meier-Kriesche HU. Are we frozen in time? Analysis of the utilization and efficacy of pulsatile perfusion in renal transplantation. Am J Transplant. 2005;5:1681–1688
8. Bhangoo RS, Hall IE, Reese PP, Parikh CR. Deceased-donor kidney perfusate and urine biomarkers for kidney allograft outcomes: A systematic review. Nephrol Dial Transplant. 2012;27:3305–3314
9. Daemen JW, Oomen AP, Janssen MA, et al. Glutathione S-transferase as predictor of functional outcome in transplantation of machine-preserved non-heart-beating donor kidneys. Transplantation. 1997;63:89–93
10. Kievit JK, Nederstigt AP, Oomen AP, Janssen MA, Schoot L, Kootstra G. Release of alpha-glutathione S-transferase (alpha GST) and pi-glutathione S-transferase (pi GST) from ischemic damaged kidneys into the machine perfusate—Relevance to viability assessment. Transplant Proc. 1997;29:3591–3593
11. Gok MA, Pelsers M, Glatz JF, et al. Comparison of perfusate activities of glutathione S-transferase, alanine aminopeptidase and fatty acid binding protein in the assessment of non-heart-beating donor kidneys. Ann Clin Biochem. 2003;40(pt 3):252–258
12. Mishra J, Ma Q, Prada A, et al. Identification of neutrophil gelatinase-associated lipocalin as a novel early urinary biomarker for ischemic renal injury. J Am Soc Nephrol. 2003;14:2534–2543
13. Kohn M, Ross H. Lactate dehydrogenase output of the excised kidney as an index of acute ischaemic renal damage. Transplantation. 1971;11:461–464
14. Mozes MF, Skolek RB, Korf BC. Use of perfusion parameters in predicting outcomes of machine-preserved kidneys. Transplant Proc. 2005;37:350–351
15. Guarrera JV, Goldstein MJ, Samstein B, et al. ‘When good kidneys pump badly’: Outcomes of deceased donor renal allografts with poor pulsatile perfusion characteristics. Transpl Int. 2010;23:444–446
16. Sonnenday CJ, Cooper M, Kraus E, Gage F, Handley C, Montgomery RA. The hazards of basing acceptance of cadaveric renal allografts on pulsatile perfusion parameters alone. Transplantation. 2003;75:2029–2033
17. Calhoon JH, Bunegin L, Gelineau JF, et al. Twelve-hour canine heart preservation with a simple, portable hypothermic organ perfusion device. Ann Thorac Surg. 1996;62:91–93
18. Bunegin L, Sadzius L, Naples J, et al. Interstitial pO2 and high energy phosphates in the canine heart during hypothermic preservation in a new, portable pulsatile perfusion device. Cardiovasc Eng. 1998;3:179–184
19. Bunegin L, Gelineau JF. The application of fluidics based technology for perfusion preservation of adult, human sized, canine hearts. Cardiovasc Eng. 2002;8:73–78
20. Tolstykh GP, Gelineau JF, Bunegin L. Perfusion preservation of rodent kidneys in a portable preservation device based on fluidics technology. Transplantation. 2002;73:1508–1510
21. Bunegin L, Gelineau JF. The application of fluidics technology for organ preservation. Biomed Instrum Technol. 2004;38:155–164
22. Tolstykh GP, Gelineau JF, Maier LM, Bunegin L. Novel portable hypothermic pulsatile perfusion preservation technology: Improved viability and function of rodent and canine kidneys. Ann Transplant. 2010;15:35–43
23. Fuller BJ, Lee CY. Hypothermic perfusion preservation: The future of organ preservation revisited? Cryobiology. 2007;54:129–145
24. Maathuis MH, Manekeller S, van der Plaats A, et al. Improved kidney graft function after preservation using a novel hypothermic machine perfusion device. Ann Surg. 2007;246:982–988; discussion 989
25. Manekeller S, Leuvenink H, Sitzia M, Minor T. Oxygenated machine perfusion preservation of predamaged kidneys with HTK and Belzer machine perfusion solution: An experimental study in pigs. Transplant Proc. 2005;37:3274–3275
26. Koetting M, Frotscher C, Minor T. Hypothermic reconditioning after cold storage improves postischemic graft function in isolated porcine kidneys. Transpl Int. 2010;23:538–542
27. Weegman BP, Kirchner VA, Scott WE 3rd, et al. Continuous real-time viability assessment of kidneys based on oxygen consumption. Transplant Proc. 2010;42:2020–2023
28. Graham M. The solubility of oxygen in physiological salines. Fish Physiol Biochem. 1987;4:1–4
29. Brown P, Nolph KD. Chemical measurements of inulin concentrations in peritoneal dialysis solution. Clin Chim Acta. 1977;76:103–112
30. de Vries EE, Hoogland ER, Winkens B, Snoeijs MG, van Heurn LW. Renovascular resistance of machine-perfused DCD kidneys is associated with primary nonfunction. Am J Transplant. 2011;11:2685–2691
31. Jochmans I, Moers C, Smits JM, et al. The prognostic value of renal resistance during hypothermic machine perfusion of deceased donor kidneys. Am J Transplant. 2011;11:2214–2220

donor kidney; transplantation; organ preservation; hypothermic machine perfusion; oxygen consumption

Copyright © 2013 by the American Society for Artificial Internal Organs