The Impact of Meeting Donor Management Goals on the Development of Delayed Graft Function in Kidney Transplant Recipients

Authors


Corresponding author: Darren J. Malinoski, malinosk@ohsu.edu

Abstract

Many organ procurement organizations (OPOs) utilize preset critical care endpoints as donor management goals (DMGs) in order to standardize care and improve outcomes. The objective of this study was to determine the impact of meeting DMGs on delayed graft function (DGF) in renal transplant recipients. All eight OPOs of the United Network for Organ Sharing Region 5 prospectively implemented nine DMGs in every donor after neurologic determination of death (DNDD). “DMGs met” was defined a priori as achieving any seven of the nine DMGs and this was recorded at the time of consent for donation to reflect donor hospital ICU management, 12–18 h later, and prior to organ recovery. Multivariable analyses were performed to identify independent predictors of DGF (dialysis in the first week after transplantation) with a p < 0.05. A total of 722 transplanted kidneys from 492 DNDDs were included. A total of 28% developed DGF. DMGs were met at consent in 14%, 12–18 h in 32% and prior to recovery in 38%. DGF was less common when DMGs were met at consent (17% vs. 30%, p = 0.007). Independent predictors of DGF were age, Cr and cold ischemia time, while meeting DMGs at consent was significantly protective. The management of potential organ donors prior to consent affects outcomes and should remain a priority in the intensive care unit.

Abbreviations
CIT

cold ischemia time

Introduction

As of November 2012, there were over 116 000 patients on the Organ Procurement and Transplantation Network (OPTN)/United Network for Organ Sharing (UNOS) waiting list with nearly 80% of those patients waiting for kidney transplants (1). In 2010, approximately 14 100 renal transplants were performed and over 3 600 patients died while on the kidney waiting list[1]. Considering that transplantation is often the treatment of choice for patients with end-stage organ failure, optimization of donor management and organ procurement is crucial in order to yield an increased quantity and quality of organs transplanted.

In an effort to address the profound shortage of transplantable organs, the US Department of Health and Human Services, Health Resources and Services Administration (HRSA) has set goals for organ donation and transplantation through the Donation and Transplantation Community of Practice (DTCP). The DTCP encourages the utilization of preset critical care endpoints as donor management goals (DMG) in order to increase both the number of organs transplanted per donor (OTPD) as well as overall conversion rates. Meeting predefined DMGs has demonstrated beneficial effects on OTPD[2-5]. However, there are often reasons for organ acceptance/declination that are unrelated to donor management or graft quality, making it difficult to truly identify best practices by merely focusing on organ utilization/transplantation rates. Additionally, of recent, the psychological and social benefits of organ donation for families of organ donors are being increasingly recognized[6]. While maintaining critical care practices in these patients through the use of catastrophic brain injury guidelines (CBIG's) is encouraged by the DTCP, the direct effect that this donor hospital management has on organ donation outcomes is unknown.

Given this, a potentially more informative study would seek to determine the impact of meeting DMGs on graft function. Currently, methods to decrease the occurrence of delayed graft function (DGF) in deceased donor kidney transplant recipients include the use of low-dose dopamine, minimizing cold ischemia time (CIT), pulsatile perfusion, donor selection, recipient selection and immunosupression[7-11]. Interestingly, with the exception of dopamine, none of these modalities involve management of the donor prior to organ recovery.

In efforts to further investigate the impact of meeting DMGs on organ donation outcomes and to obtain a general assessment of the critical care provided to patients with catastrophic brain injuries who progress to become donors after neurologic determination of death (DNDD), the eight OPOs in UNOS Region 5 prospectively implemented a checklist of nine DMGs and gathered data on all DNDD's. Having recently published a report on DMGs looking only at the impact that meeting DMGs has on the number of organs transplanted per donor[5], the purpose of this study was to expand our understanding by evaluating the impact that meeting predefined DMGs has on graft function through prospectively studying deceased donor kidney transplant recipients. It was hypothesized that meeting DMGs in DNDDs would be associated with less DGF in renal transplant recipients.

Methods

Aspects of the methodology discussed below have been previously described in a recent original report[5] and are represented for completeness.

Donor management goals

In June 2008, a checklist of nine DMGs intended to represent critical care endpoints that aim to optimize the function of organs and guide resuscitation efforts was formed (Table 1). This revised checklist was created on the basis of recommendations from the Canadian Council for Donation and Transplantation[12], previous regional DMG data[4], and the experiences and expert opinions of representatives from the eight OPOs in UNOS Region 5 (see the Acknowledgements section for the list of OPOs). Similar to as described in a previous report[5], the administration of thyroid hormone and measurement of serum creatinine were not considered as DMGs, but their impact on graft function was also analyzed. Specifically, thyroid hormone was not uniformly used by all OPOs in the study due to differing opinions on its utility in donor management. Additionally, serum creatinine was not considered as a DMG as it was thought in many cases to represent chronic donor medical characteristics as well as the general inflammatory response of brain death rather than a modifiable factor.

Table 1. UNOS Region 5 donor management goals
  % of Renal grafts meeting specific DMG at
  Time of consent12–18 h laterPrior to organ recovery
Donor management goalsParameters(n = 722)(n = 722)(n = 722)
  1. aLow dose of vasopressors was defined as dopamine ≤ 10 mcg/kg/min, neosynephrine ≤ 60 mcg/kg/min and norepinephrine ≤ 10 mcg/kg/min.
1. Mean arterial pressure60–100 mmHg78%82%86%
2. Central venous pressure4–10 mmHg32%62%64%
3. Ejection fraction≥50%16%49%52%
4. Vasopressors≤1 and low dosea59%71%81%
5. Arterial blood gas pH7.3–7.4569%79%71%
6. PaO2:FiO2≥30039%43%43%
7. Serum sodium135–155 mEq/L69%74%76%
8. Blood glucose≤150 mg/dL53%41%46%
9. Urine output0.5–3 cc/kg/h over 4 h64%63%69%

Study design and data collection

A prospective interventional study of all DNDDs, both standard criteria donors (SCD) and extended criteria donors (ECD), managed by the eight OPOs in UNOS Region 5 was conducted from July 2008 to January 2009. ECDs were declared legally brain dead by hospital criteria for neurologic determination of death and were aged either older than 59 years or 50–59 years with at least two of the following: chronic hypertension, stroke as the cause of death, or serum creatinine >1.5 mg/dL. SCDs were also declared legally brain dead and were aged either younger than 50 years or 50–59 years with less than two of the aforementioned criteria.

As described in a previous study by our group which evaluated the impact of meeting DMGs on the number of OTPD[5], the nine critical care endpoints (Table 1) were prospectively implemented in the management of every DNDD. Although implementation by each OPO was customized, it included either a paper or online checklist that was used at the bedside during donor management. “DMGs met” was defined a priori as meeting any seven or more of the nine DMGs and this was recorded at the time of consent for donation to reflect donor hospital critical care management, 12–18 h later to represent the general time that organ offers are being made, and prior to organ recovery to illustrate the final product of donor management. All DNDDs in the study were managed by their respective OPO based on their local donor management guidelines. Even though guidelines between OPOs may have differed slightly, they are all based on the UNOS clinical pathway[13] and thus donors who met DMGs and those who did not were managed with the same vasopressors, ventilator setting, etc.

Data regarding donor demographics, donor type, thyroid hormone use, serum creatinine, CIT, organs transplanted and DGF of renal grafts were also collected. DGF in these patients was defined as the requirement for dialysis in the first week after transplantation. Each kidney and its donor's associated data were included as individual cases due to different CIT times and different graft function results.

DMG data for each donor was uploaded by all participating OPOs to a regional donor database contained within the Region 5 UNOS SharePoint (Microsoft, Redmond, Washington) website. The graft function data were supplied by UNOS in a deidentified fashion, as it was attached to the UNOS identification number of each deceased donor. All eight OPOs agreed to share their data for pooled analysis and dissemination. This study was approved internally by the research oversight bodies of each OPO and was determined to represent non-human subjects research by the Cedars-Sinai Medical Center IRB.

Statistical analysis

The primary outcome measure of this study was the presence of DGF in deceased donor kidney transplant recipients. Univariable analyses comparing the following factors at the time of consent, 12–18 h later, and prior to recovery in patients who developed DGF to those who did not were conducted: [1] achieving “DMG's met”, [2] achieving individual DMGs, [3] the number of DMG's met, [4] thyroid hormone use, [5] serum creatinine and [6] change in serum creatinine. Additionally, donor age, donor type and CIT were compared. Univariable analyses of categorical variables were conducted using either Pearson's chi-square or Fisher's exact test. Univariable analyses of continuous variables were performed using Student's t-test.

To determine which factors had an independent impact on DGF, a multivariable analysis using binary logistic regression was performed and clinically relevant variables with a p < 0.1 on univariable analysis were included in the multivariable analysis model. Individual DMGs being achieved (categorical data) as well as the exact number of individual DMGs achieved (continuous data) were not included in the multivariable analysis, since the main intervention was implementation of the bundle as a checklist and inherently some of the DMGs were interrelated (e.g. central venous pressure [CVP] and urine output, etc.). As previously defined[5], the measure of success in terms of implementing the bundle was achieving any seven or more of the nine DMGs and this was termed “DMGs met” for the purposes of the analyses. All variables with p < 0.05 on multivariable analysis were considered significant and related variables were evaluated using separate models.

Additionally, as part of a separate analysis, the impact of other organs being transplanted from multi organ donors, on DGF of the renal grafts from those donors, was evaluated. Statistical analysis was performed with SPSS version 18.0 for Windows (SPSS, Chicago, Illinois). Lastly, it should be noted that values in the text and tables are reported as mean ± standard deviation (SD) or percent (%) frequency.

Results

Over the 7-month-study period, 492 donors were pro-spectively studied, of which 402 donated at least one kidney. From these donors, data were complete for 722 kidneys. Of the patients who donated these 722 kidneys, the mean donor age was 37.3 ± 16.5 years. Additionally, 85% came from SCD donors and 15% came from ECD donors. The average serum creatinine prior to organ recovery was 1.18 ± 0.92 mg/dL and the average CIT was 16.4 ± 8.1 h. DMGs were met at consent in 14%, at 12–18 h in 32%, and prior to recovery in 38%. Of the 722 kidneys transplanted, 28% had DGF.

With regard to individual DMGs, Table 1 displays how consistent each of the DMGs were met across the time points studied. The CVP and ejection fraction parameters were the most likely to be achieved after consent, often due to the fact that donors did not have central lines or echocardiograms prior to the OPO assuming responsibility for medical management. The P:F ratio was the least often met parameter, which reflects the most challenging part of donor management as well as the difficulty in successfully placing lungs for transplantation.

The result of the univariable analysis of categorical and continuous variables are presented in Table 2 and Table 3, respectively. From these results it can be seen that DGF was less common in patients who met DMGs at the time of consent (17.3% vs. 30.1%, p = 0.007), and more common in those who were ECD (35.8% vs. 26.9%, p = 0.058). Thyroid hormone treatment did not appear to have an impact on DGF. Additionally, kidneys with DGF came from older donors (41 ± 16 vs. 36 ±16 years, p < 0.001), had higher terminal serum creatinine values (1.4 ± 1.1 vs. 1.1 ± 0.8 mg/dL, p = 0.001), and had longer CIT (18 ± 9 vs. 16 ± 8 h, p = 0.001). When changes in serum creatinine were evaluated, it was found that kidney grafts that did not develop DGF had a statistically significant relative decrease in serum creatinine from the time of consent to 12–18 h later and from the time of consent to prior to organ recovery (Table 3).

Table 2. Univariable analysis of categorical data associated with delayed graft function (DGF) after deceased donor renal transplantation
 % of Renal grafts% of Renal grafts 
 with DGF when thewith DGF when the 
Variablevariable is NOT metvariable is metp-Valuea
  1. ap-Values calculated using Pearson's chi-square.
  2. bDMGs met was defined a priori as achieving any seven of the nine DMGs.
  3. cLow dose of vasopressors was defined as dopamine ≤ 10 mcg/kg/min, neosynephrine ≤ 60 mcg/kg/min and norepinephrine ≤ 10 mcg/kg/min.
DMGs at time of consentb30.1%17.3%0.007
DMGs 12–18 h laterb28.9%26.8%0.543
DMGs prior to organ recoveryb26.5%31.2%0.173
Specific DMGs at the time of consent   
 Mean arterial pressure 60–100 mmHg27.2%28.6%0.725
 Central venous pressure 4–10 mmHg29.7%25.2%0.209
 Ejection fraction > 50%28.5%26.9%0.718
 Vasopressors ≤ 1 and low dosec26.7%29.3%0.436
 Arterial blood gas pH 7.30–7.4531.3%26.9%0.222
 PaO2:FiO2 > 30028.4%28.1%0.929
 Serum sodium 135–155 mEq/L27.8%28.5%0.839
 Blood glucose < 150 mg/dL29.2%27.5%0.612
 Urine output 0.5–3 cc/kg/h over 4 h29.9%27.3%0.464
Specific DMGs 12–18 h later   
 Mean arterial pressure 60–100 mmHg29.2%28.0%0.785
 Central venous pressure 4–10 mmHg28.9%27.8%0.751
 Ejection fraction > 50%27.2%29.4%0.511
 Vasopressors ≤ 1 and low dosec28.3%28.1%0.959
 Arterial blood gas pH 7.30–7.4526.7%28.7%0.627
 PaO2:FiO2 > 30029.3%26.8%0.459
 Serum sodium 135–155 mEq/L26.6%28.8%0.557
 Blood glucose < 150 mg/dL28.8%27.5%0.712
 Urine output 0.5–3 cc/kg/h over 4 h26.6%29.2%0.447
Specific DMGs prior to organ recovery   
 Mean arterial pressure 60–100 mmHg31.6%27.7%0.424
 Central venous pressure 4–10 mmHg27.8%28.5%0.822
 Ejection fraction > 50%27.1%29.3%0.517
 Vasopressors ≤ 1 and low dosec28.1%28.3%0.954
 Arterial blood gas pH 7.30–7.4523.7%30.1%0.083
 PaO2:FiO2 > 30029.8%26.3%0.304
 Serum sodium 135–155 mEq/L21.0%30.6%0.014
 Blood glucose < 150 mg/dL28.9%27.5%0.675
 Urine output 0.5–3 cc/kg/h over 4 h21.0%31.5%0.004
Donor type   
 Extended criteria donor26.9%35.8%0.058
Treatment   
 Thyroid hormone use at time of consent29.2%25.2%0.376
 Thyroid hormone use 12–18 h later27.9%28.4%0.898
 Thyroid hormone use prior to organ recovery29.1%28.0%0.741
 % of Renal grafts% of Renal grafts 
 with DGF when otherwith DGF when other 
Other organs transplantedother organs were declinedother organs were transplantedp-Valuea
 Heart33.5%20.0%<0.001
 Left lung29.6%23.3%0.119
 Right lung29.4%27.1%0.595
 Liver28.7%28.0%0.882
 Pancreas31.0%18.8%0.003
Table 3. Univariable analysis of continuous data associated with delayed graft function (DGF) after deceased donor renal transplantation
VariableNo DGFDGFp-Valuea
  1. ap-Values calculated using Pearson's chi-square.
  2. bLow dose of vasopressors was defined as dopamine ≤ 10 mcg/kg/min, neosynephrine ≤ 60 mcg/kg/min, and norepinephrine ≤ 10 mcg/kg/min.
Donor age (years)36.0 ± 16.340.8 ± 16.4<0.001
Number of DMGs met at time of consent4.82 ± 1.64.72 ± 1.40.401
Number of DMGs met 12–18 h later5.65 ± 1.75.67 ± 1.60.866
Number of DMGs met prior to organ recovery5.82 ± 1.76.04 ± 1.60.118
Other known predictors of graft function   
 Creatinine at time of consent1.14 ± 0.971.24 ± 0.840.191
 Creatinine 12–18 h later1.05 ± 0.781.29 ± 0.870.001
 Creatinine prior to organ recovery1.10 ± 0.811.39 ± 1.120.001
 Cold ischemia time (h)15.8 ± 7.818.0 ± 8.70.001
Relative changes in variables   
 Change in serum creatinine from consent to 12–18 h later−0.09 ± 0.730.05 ± 0.460.017
 Change in serum creatinine from 12 h to 18 h after consent to prior to organ recovery0.04 ± 0.410.10 ± 0.760.307
 Change in serum creatinine from consent to prior to organ recovery−0.04 ± 0.770.15 ± 0.810.004

Of note, also upon univariable analysis, it was found that a larger proportion of donors meeting the DMGs for arterial blood gas pH, serum sodium, and urine output prior to organ recovery, had DGF. Consequently, the appropriateness of the ranges initially chosen for the DMGs were further examined and it was found that kidney grafts from donors who had values outside of the goal range for pH, sodium, urine output and CVP prior to recovery had lower rates of DGF (Table 4).

Table 4. Impact of being out of range for specific DMGs prior to organ recovery on renal delayed graft function (DGF) after deceased donor transplantation
 % of Renal grafts with DGF 
DMGBelow rangeIn rangeAbove rangep-Valuea
  1. ap-Values calculated using Pearson's chi-square.
Central venous pressure<4 mmHg4–10 mmHg>10 mmHg0.079
 6.3%28.5%23.1% 
Arterial blood gas pH<7.307.30–7.45>7.450.185
 24.2%30.1%23.1% 
Serum sodium<135 mEq/L135–155 mEq/L>155 mEq/L0.034
 19.0%30.6%21.9% 
Urine output (over 4 h)<0.5 cc/kg/h0.5–3 cc/kg/h>3 cc/kg/h0.002
 30.3%31.2%14.8% 

In regard to multivariable analysis, creatinine after 12–18 h, creatinine prior to organ recovery, change in serum creatinine from consent to 12–18 h later and change in serum creatinine from consent to prior to organ recovery all had a p < 0.1 on univariable analysis and are inherently related. Therefore, separate models were used to analyze the impact of each of these variables in relation to age, CIT, ECD status and meeting DMGs at the time of consent. Upon multivariable analyses (Table 5), independent predictors of DGF were age, CIT and any of the aforementioned variables involving creatinine, while meeting DMGs at consent was significantly protective in all four models. The odds ratios for these variables were all similar in the different models. Additionally, ECD status was not noted to be an independent predictor of DGF in any model once age and creatinine were taken into consideration.

Table 5. Multivariable analyses: independent predictors of delayed graft function (DGF) after deceased donor renal transplantation
VariableaOR95% CI for ORp-Valueb
  1. aSince creatinine 12–18 h later, creatinine prior to organ recovery, change in serum creatinine from consent to 12–18 h later and change in serum creatinine from consent to prior to organ recovery all had a p < 0.1 on univariable analysis and are inherently related, separate models were used to analyze the impact of each of these variables.
  2. bp-Value calculated using a binary logistic regression model.
Model 1   
 Donor age1.0191.006–1.0320.004
 Cold ischemia time1.0281.007–1.0490.008
 Extended criteria donor0.8920.523–1.5230.676
 DMGs “met” at time of consent0.5070.294–0.8760.015
 Creatinine 12–18 h later1.3651.120–1.6640.002
Model 2   
 Donor age1.0201.007–1.0330.003
 Cold ischemia time1.0271.006–1.0470.011
 Extended criteria donor0.8880.520–1.5170.664
 DMGs “met” at time of consent0.5200.301–0.8970.019
 Creatinine prior to organ recovery1.3621.144–1.6210.001
Model 3   
 Donor age1.0181.005–1.0310.005
 Cold ischemia time1.0301.010–1.0510.004
 Extended criteria donor0.9570.559–1.6380.873
 DMGs “met” at time of consent0.4780.278–0.8230.008
 Change in serum creatinine from consent to 12–18 h later2.0281.277–3.2200.003
Model 4   
 Donor age1.0201.007–1.0320.002
 Cold ischemia time1.0281.008–1.0490.007
 Extended criteria donor0.9350.547–1.5970.805
 DMGs “met” at time of consent0.4980.289–0.8570.012
 Change in serum creatinine from consent to prior to organ recovery1.6271.206–2.1950.001

Lastly, the impact of having other organs transplanted (from multiorgan donors) on DGF of kidney transplants can be seen as part of the categorical data at the bottom of Table 3. Notably, renal grafts from donors who also had their heart or pancreas transplanted, were found to have significantly less DGF whereas no correlation was noted with lung or liver transplants.

Discussion

With the demand for organs that are available for transplantation substantially outweighing the supply, a significant effort to increase the quantity as well as the quality of donated organs is underway. Although a number of previous reports have tried to assess the impact of donor management on recipient kidney graft function (11,14–24), none of these studies have attempted to standardize donor management on a regional level or evaluate the impact of the prospective implementation of a DMG checklist on DGF. In the current investigation, the impact of meeting such a checklist of DMGs on graft outcomes in deceased donor kidney recipients was prospectively investigated. We found that although DGF was fairly common, occurring in 28% of cases, meeting DMGs (defined as having met seven out of nine critical care endpoints) prior to consent was associated with a significant decrease in its development. This finding illustrates the significance of the critical care provided to patients with catastrophic brain injuries who progress to brain death until the desire to donate can be appropriately delineated.

Due to the different criteria used in various studies to determine acceptable donors, as well as differing definitions of DGF, the range of reported DGF in deceased donor renal transplant is large, with a review article by Perico et al. stating it to be from 2% to 50%[8]. The current study used the most common definition of DGF in published reports, requiring dialysis in the first week after transplantation, and found the rate to be in the middle of this range at 28%. A recent article by Premasathian et al. that analyzed risk factors and outcomes of DGF after deceased donor kidney transplantation found that recipients of deceased donor kidneys who developed DGF were found to have both lower graft survival and higher mortality at 1 and 5 years after transplantation[25]. Other studies have found that while DGF may be a risk factor for acute rejection episodes, it is not a risk factor for increased graft loss at 1 year[26]. Either way, the occurrence of DGF leads to increased utilization of medical resources and emotional stress for physicians, recipients and their families[27].

Traditional factors that are known to affect DGF in kidney transplants have been classified as being related to the following[9]: donor tissue quality, brain death and related stress, preservation variables, immune factors and recipient variables. The results from the current study corroborate these traditional factors, specifically, donor age, CIT and terminal creatinine as being independent predictors of DGF in renal allografts. Although some of the aforementioned factors pertain to optimizing donor hemodynamics, none of these factors specifically relate to the standardization of donor management, the implementation of a bundle of critical care endpoints as DMGs, or the impact of the care provided by donor hospital intensive care teams prior to consent for donation. The current study found meeting DMGs prior to consent to be significantly protective of DGF in deceased donor kidney transplants. The finding that neither meeting DMGs 12–18 h later nor prior to organ recovery had this same protective effect suggests that the critical care management of potential organ donors prior to consent (not just during the interval from brain death to time of consent) may be of most significant importance in determining renal function in the recipient. While members of the DTCP cannot and should not directly influence the care of patients prior to the declaration of neurologic death and obtaining consent for donation, many hospitals have adopted catastrophic brain injury guidelines (CBIGs) to help in the management of patients with severe neurologic injuries. CBIGs (which come in forms such as clinical pathways, treatment algorithms, etc.) contain standard critical care practices that aim to improve perfusion of the brain, which in effect, also improve perfusion to other organs. These guidelines often reflect and are similar to the donor management protocols used by many OPOs for managing donors after consent and it has been shown that hospitals with CBIGs have a significantly higher median number of donors as well as a significantly higher conversion rate. Using data collected in the current study, however, we are unable to ascertain the relationship between the achievement of DMGs, or the rates of DGF, and the use of CBIGs given that the donor data were deidentified and we do not know which hospitals have such guidelines in place.

Despite the fact that individual DMGs were not evaluated as risk factors for DGF, it was interesting to note that patients achieving the specific DMGs of CVP, arterial blood gas pH, serum sodium and urine output prior to organ recovery had higher rates of DGF. This observation indicates that the ranges for these DMGs are probably inappropriate and may be one reason why meeting DMGs 12–18 h after consent or prior to organ recovery did not have the same effect as meeting DMGs prior to consent. As stated previously, the ranges for these DMGs were originally chosen based on published recommendations[12], previous regional DMG data, as well as by consensus based on the experiences and expert opinions of the representatives of the eight OPOs in Region 5. After analyzing the results of this study and further consideration, the range for arterial blood gas pH, has been changed to 7.3–7.5. In regard to serum sodium, the lower sodium limit was eliminated after it was discovered that kidneys from donors with sodium levels <135 meq/dL prior to recovery had less DGF. Although those who had sodium levels > 155 mEq/dL also had less DGF, it was decided that this limit should not be altered, as doing so could potentially impact liver graft utilization. In terms of urine output, the upper limit was initially chosen to provide another checklist item that could control for DI. However, kidneys from donors with urine output levels above the initially chosen range had less DGF. Consequently, the consensus of the region was to rely on the sodium DMG to control for DI and to eliminate the upper limit of the urine output goal. Of note, the CVP DMG was set at 4–10 mmHg based on the consensus of the eight OPOs; while some had previously targeted CVP levels up to 12 mmHg, others thought levels > 10 mmHg could potentially impact lung function. Upon further analysis of our results which showed a trend toward less DGF in patients with CVPs above 10 mmHg, and noting that some OPOs outside the region use a range of 4–12 mmHg, this DMG has recently been changed to 4–12 mmHg. A potential explanation for why meeting the original DMG ranges was statistically protective prior to consent, but not 12–18 h later or prior to recovery, is that the common OPO practice of diuresing donors after consent has the potential to lower CVP, raise urine output, induce a metabolic alkalosis (raise arterial pH) and lower serum sodium levels. With continued prospective examination, we aim to determine the impact of these changes on both organ utilization and function in order to optimize the bundle of DMGs that should be implemented.

In regard to renal grafts from multiple organ donors compared to kidney only donors, a previous report by Koning et al. did not note a difference in DGF rates between these groups[7]. Results from the current study indicated, however, that donors with heart or pancreas transplants had a significantly lower incidence of DGF. Since hemodynamic stability is critical in determining the acceptability of heart or pancreas grafts for transplantation, it is possible that this protective effect is due to the associated hemodynamic stability of the donor.

Limitations of the current study include that data on other known predictors of graft function, such as panel reactive antibody status and pulsatile perfusion parameters, were not available. Similarly, it is important to acknowledge that some recipient variables such as human leukocyte antigen (HLA) mismatches, pretransplant transfusions, immunosupression protocols, machine perfusion parameters and duration of dialysis prior to transplantation as well as some donor variables such as warm ischemia time and time of brain death were not factored into the current analysis as data on these factors were not collected. Additionally, the duration of DGF as well as follow-up data pertaining to the development of acute rejection episodes, graft survival and patient survival were not recorded. This in turn limits what can be said about the outcomes of patients with renal DGF.

The current study continues to identify traditional predictors of DGF but is distinct in that it is the first to evaluate the impact of meeting DMGs in the brain dead donor on renal DGF in the recipient. The novel finding that meeting DMGs at consent was associated with an approximately 50% decrease in the risk of DGF highlights the importance of maintaining normal critical care end points in patients with devastating brain injuries in order to preserve the option of donation and maximize the gift of those who go on to donate their organs.

Acknowledgments

The authors would like to thank all of the members of the United Network for Organ Sharing (UNOS) Region 5 Donor Management Goals (DMG) Workgroup as well as the donors, recipients, staff and leadership of the eight organ procurement organizations in Region 5 for their support of this study and for aid in the collection and management of data: Jennifer Treece (Donor Network of Arizona); Ricardo Elizondo (California Transplant Donor Network); Christopher Good and Tamra Grote (Golden State Donor Services); Scott Bunting and Alisa Suarez (OneLegacy); Jill Stinebring (Lifesharing); Ann Roberson (New Mexico Donor Services); Galyn Schoenstein (Nevada Donor Network); and Craig Myrick and Mike Ingraham (Intermountain Donor Services).

Additionally, this work was supported in part by Health Resources and Services Administration contract 234-2005-37011C. The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government.

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

Ancillary