Inferior Early Posttransplant Outcomes for Recipients of Right Versus Left Deceased Donor Kidneys: An ANZDATA Registry Analysis

Authors

  • H. Vacher-Coponat,

    1. AP-HM, Hôpital de la Conception, Centre de Néphrologie et Transplantation Rénale, Marseille, France
    2. Aix-Marseille University, Marseille, France
    Search for more papers by this author
  • S. McDonald,

    1. ANZDATA Registry, Royal Adelaide Hospital, Adelaide, South Australia, Australia
    Search for more papers by this author
  • P. Clayton,

    1. ANZDATA Registry, Royal Adelaide Hospital, Adelaide, South Australia, Australia
    2. Transplantation Services, Royal Prince Alfred Hospital and University of Sydney, NSW, Australia
    Search for more papers by this author
  • A. Loundou,

    1. Aix-Marseille Univ, Délégation à la Recherche Clinique, Faculté de Médecine, Marseille, France
    Search for more papers by this author
  • R. D. M. Allen,

    1. Transplantation Services, Royal Prince Alfred Hospital and University of Sydney, NSW, Australia
    Search for more papers by this author
  • S. J. Chadban

    Corresponding author
    1. ANZDATA Registry, Royal Adelaide Hospital, Adelaide, South Australia, Australia
    2. Transplantation Services, Royal Prince Alfred Hospital and University of Sydney, NSW, Australia
    • AP-HM, Hôpital de la Conception, Centre de Néphrologie et Transplantation Rénale, Marseille, France
    Search for more papers by this author

Corresponding author: Steve Chadban

steve.chadban@sswahs.nsw.gov.au

Abstract

Anatomical differences between right and left kidneys could influence transplant outcome. We compared graft function and survival for left and right kidney recipients transplanted from the same deceased organ donor. Adult recipients of 4900 single kidneys procured from 2450 heart beating deceased donors in Australia and New Zealand from 1995 to 2009 were included in a paired analysis. Right kidneys were associated with more delayed graft function (DGF) (25 vs. 21% for left kidneys, p < 0.001) and, if not affected by DGF, a slower fall in serum creatinine. One-year graft survival was lower for right kidneys (89.1 vs. 91.1% for left kidneys, p = 0.001), primarily attributed to surgical complications (66 versus 35 failures for left kidneys). Beyond the first posttransplant year, kidney side was not associated with eGFR, graft or patient survival. Receipt of a right kidney is a risk factor for inferior outcomes in the first year after transplantation. A higher incidence of surgical complications suggests the shorter right renal vein may be contributory. The higher susceptibility of right kidneys to injury should be considered in organ allocation.

Abbreviations
ANZDATA

Australian and New Zealand Dialysis and Transplant Registry

CIT

cold-ischemia time

DGF

delayed graft function

eGFR

estimated glomerular filtration rate

ESRD

end-stage renal disease

IRI

ischemia-reperfusion injury

MDRD

modification of diet in renal disease

PRA

panel reactive antibodies

RR

relative risk.

Introduction

Early posttransplant events are important determinants of graft survival and function. Acute rejection is an occasional cause of graft loss and a contributor to chronic allograft damage and dysfunction for a significant minority of recipients [1]. However, ischemia-reperfusion injury (IRI) is a far more common insult and initially manifest by slow function or, when more severe, delayed graft function (DGF) [2, 3]. Ischemia-reperfusion may injure the graft through several pathways: [1] severe IRI may cause acute tubular necrosis, clinically manifest as delayed graft function; [2] innate immune activation resulting from IRI may generate a milieu which promotes development of adaptive immunity and acute rejection [4]; [3] severe IRI may predispose the graft to tubulointerstitial fibrosis and atrophy. As deceased donor age and comorbid status has increased and particularly with increasing usage of deceased cardiac donors, the clinical manifestations of IRI may be increasing [5-7].

The impact of donor kidney side (right vs. left kidney) has been a subject of debate [8]. Anecdotally, transplant surgeons prefer to use the left kidney for live donor transplantation. The longer vein with the deceased donor left kidney enables greater ease and safety at implantation without any requirement to elongate or graft the vein [9, 10]. Equally, right-sided donor kidneys have longer arteries and appear to incur an increased risk of kinking which may compromise graft perfusion [10]. Multiples vessels, with higher thrombotic risk, are reported to occur more frequently on the right side [10, 11] and right kidneys may on average be smaller than left [12].

In deceased donor transplantation, the kidney side is generally not included in the allocation decision. Limited reports with conflicting results, derived from single-center studies of small cohorts, have examined the impact of kidney side on transplant outcome [13-15]. To more definitively address the hypothesis that use of right kidneys yields inferior outcomes to left, we obtained data from the Australia and New Zealand Dialysis and Transplant registry (ANZDATA) to determine the impact of kidney side on transplant outcomes using a paired analysis design.

Patients and Methods

We accessed the ANZDATA registry and retrospectively analyzed data for all adult recipients of a single kidney obtained from a heart-beating, brain-dead donor transplanted between January 1995 and December 2009, where both kidneys were implanted into different recipients. Follow-up was until December 2010. Analyses were stratified by left versus right kidney in order to eliminate all other donor factors as a source of bias.

For each transplanted kidney the following parameters were recorded:

  • - Donor parameters: gender, age, weight, height, BMI, diabetes, hypertension, smoking, cause of death, terminal serum creatinine, kidney side (left or right) and cold ischemia time.
  • - Recipient parameters: gender, age, weight, height, BMI, cause of end-stage renal disease, dialysis duration, cardiovascular disease, diabetes, chronic lung disease, panel reactive antibodies (PRA), previous transplant.
  • - Transplantation parameters: HLA mismatch, delayed graft function (DGF) defined as dialysis required within the first 72 hours, eGFR using the MDRD simplified formula [16], donor and graft survival, causes of graft failure.

Statistical Analysis

Continuous variables were compared using paired t-tests or Wilcoxon's signed rank test depending on distribution. Paired binary variables were compared using McNemar's test. The significance threshold was p lower than 0.05. Risk factors for delayed graft function (DGF) were analyzed by conditional logistic regression, which compares the risk factors of paired recipients in which one recipient suffered DGF and the other did not. Graft and patient survival were analyzed using Cox regression stratified by donor. The stratification fits a different survivor function for each kidney pair. Owing to nonproportional hazards, right versus left kidney was fit as a time-varying covariate—first year versus subsequent years. Multivariable models were constructed by screening variables for significance at the 0.25 level. Variables that were statistically or clinically significant (the latter including age, sex, right versus left kidney and ischemic time) were included in a base multivariable model. Backward selection was then used to remove variables not significant at the 0.05 level. Statistical analyses were conducted using the statistical package Stata/IC 12.1 (StataCorp, College Station TX, USA). As the study was a paired analysis, donor factors were the same for right and left kidneys. Therefore only kidney side, recipient and transplantation parameters were included in the analysis.

Results

Between January 1995 and December 2009, 6519 kidneys recovered from 3343 deceased donors were transplanted in Australia and New Zealand, having discarded 223 kidneys (3.3%): 118 right and 105 left (p = 0.37). We excluded 144 en bloc or double-kidney transplants, 221 donations after cardiac death, 437 kidneys transplanted along with another organ, 150 pediatric recipients (<16 years) and a further 667 kidneys in which only one kidney was transplanted into a patient meeting inclusion criteria. Thus, 4900 recipients were included in the analysis, enabling comparison of 2450 left versus 2450 right kidneys. Their clinical characteristics are summarized in Tables 1 and 2.

Table 1. Donor characteristics
Factor(n = 2450)
  1. eGFR, MDRD simplified.

  2. Clearance = Creatinine (umol/L)−1.154 × Age (years)−0.203 × 32 788. (Woman × 0.72).

Age (years), mean ± SD43.2 ± 16.9
Sex 
 Female1040 (42.4%)
 Male1410 (57.6%)
Height (cm), mean ± SD170.7 ± 12.2
Weight (cm), mean ± SD75.2 ± 17.4
BMI, mean ± SD26.2 ± 4.9
Cause of death 
 Stroke1355 (55.3%)
 Accident618 (25.2%)
 Other477 (19.5%)
Hypertension462 (19.4%)
Smoking 
 Current936 (38.2%)
 Former355 (14.5%)
 Never1087 (44.4%)
 Unknown72 (2.9%)
Diabetes26 (1.1%)
Terminal creatinine (μmol/L), median (IQR)80.0 (61.0, 100.0)
Terminal eGFR (mL/min/1.73 m2), median (IQR)89.5 (67.4, 116.9)
Table 2. Recipient and transplantation characteristics for right and left kidneys
FactorLeftRightp-Value
  1. eGFR was calculated using the simplified MDRD formula: Clearance = Creatinine (umol/L) −1.154 × Age (years)−0.203 × 32 788. (woman × 0.72). p-Value ≥ 0.05 was considered not significant (ns).

N24502450 
Age (years), median (IQR)48.6 (38.5, 57.2)48.9 (38.5, 57.4)0.87
Sex   
 Female939 (38.3%)936 (38.2%)0.93
 Male1511 (61.7%)1514 (61.8%) 
Race   
 Caucasian1978 (80.7%)1999 (81.6%)0.15
 ATSI115 (4.7%)99 (4.0%) 
 Asian204 (8.3%)227 (9.3%) 
 Other153 (6.2%)125 (5.1%) 
Height (cm), mean ± SD168.8 ± 11.1168.4 ± 11.80.19
Weight (kg), mean ± SD74.3 ± 16.572.9 ± 15.80.002
BMI, mean ± SD26.1 ± 4.725.7 ± 4.50.006
Primary renal disease   
  GN1194 (48.7%)1234 (50.4%)0.69
  Polycystic331 (13.5%)303 (12.4%) 
  Vascular105 (4.3%)98 (4.0%) 
  Diabetes200 (8.2%)204 (8.3%) 
  Other620 (25.3%)611 (24.9%) 
Peripheral vascular disease114 (4.7%)102 (4.2%)0.42
Coronary artery disease250 (10.2%)221 (9.1%)0.17
Cerebrovascular disease83 (%)71 (%)0.33
RRT duration (years), median (IQR)3.3 (1.7, 6.4)3.2 (1.6, 6.4)0.45
Previous grafts   
 02090 (85.3%)2091 (85.3%)0.55
 1286 (11.7%)296 (12.1%) 
 267 (2.7%)52 (2.1%) 
 36 (0.2%)10 (0.4%) 
 41 (<1%)1 (<1%) 
Peak PRA (%), median (IQR)5 (0, 29)5 (0, 31)0.22
HLA-AB mismatch   
 0179 (7.3%)180 (7.3%)0.98
 1473 (19.3%)477 (19.5%) 
 2722 (29.5%)705 (28.8%) 
 3642 (26.2%)658 (26.9%) 
 4434 (17.7%)430 (17.6%) 
HLA-DR mismatch   
 01004 (41.0%)1028 (42.0%)0.45
 1849 (34.7%)859 (35.1%) 
 2596 (24.3%)558 (22.8%) 
Ischaemia (h), median (IQR)14.0 (11.0, 18.0)14.0 (11.0, 18.0)0.78

Delayed Graft Function

DGF developed in 1128 (23.0%) recipients, with a higher rate for right kidneys (24.9 vs. 21.1% for the left, p = 0.0002). Among recipients who did not receive dialysis after transplantation, right kidney recipients more frequently exhibited slow onset of kidney function, as defined by spontaneous fall of serum creatinine <10% within the first 24 h (18.6% vs. 15.6% for left kidneys, p = 0.04). The higher incidence of DGF among right kidney recipients remained apparent after exclusion of all kidneys lost during the first week (21.5 vs. 19.2%, p = 0.01). By multivariable analysis, right kidney was confirmed as an independent risk factor for DGF (adjusted odds ratio 1.46 [95% CI: 1.20–1.77], p = 0.001; Table 3).

Table 3. Risk factors for delayed graft function. Multivariable analysis of delayed graft function risk factors with a conditional logistic regression model
FactorOdds ratio95% CIp-Value
  1. a

    Reference group.

  2. BMI = body mass index; HLA = human leukocyte antigen; PRA = panel reactive antibodies.

Right kidney1.461.20 – 1.770.0002
Age0.990.98 – 1.000.06
Male sex1.230.91 – 1.660.18
BMI (kg/m2)   
 <201a <0.0001
 20–24.92.191.21 – 3.95 
 25–29.94.062.19 – 7.52 
 ≥306.123.19 – 11.73 
Diabetes1.521.01 – 2.290.04
HLA mismatch1.351.18 – 1.55<0.0001
Peak PRA (%)   
 <51a 0.0004
 5–791.771.30 – 2.43 
 ≥802.171.27 – 3.72 
Ischemic time (h)   
 <81a 0.09
 8–112.461.15 – 5.25 
 12–172.531.23 – 5.19 
 ≥182.581.21 – 5.50 

Kidney Graft Function

Estimated GFR was similar for recipients of right kidneys as compared with left at all time points (month 1 eGFR right 47.0 ± 19.8 (mean ± sd) vs. left 46.7 ± 19.4 mL/min/1.73 m2, p = 0.45; month 3 eGFR 50.2 ± 17.8 vs. 49.9 ± 17.6 ml/min/1.73 m2, p = 0.33; year 1 51.0 ± 18.3 vs. 50.9 ± 18.9 mL/min/1.73 m2, p = 0.98; year 5 50.2 ± 20.7 vs. 49.5 ± 19.5 mL/min/1.73 m2, p = 0.25).

Patient and Graft Survival

Graft survival was significantly lower for right kidney recipients, significant from the first month (1 month: 93.7 vs. 95.8%; 1 year: 89.1 vs. 91.1%; 5 years: 76.6 vs. 78.5% for left kidneys; p = 0.006) (Figure 2). Survival curves differed early then became parallel. Indeed, when kidneys which had been lost during the first year were excluded from the analysis, survival beyond the first year was not affected by kidney side (5 year survival 85.9 vs. 86.2%; p = 0.17). By multivariable analysis, right side was confirmed as an independent risk factor for inferior graft survival during the first year after transplant (adjusted hazard ratio 1.62, 95% CI 1.22–2.15, p = 0.001) but not during subsequent years (adjusted hazard ratio 1.10, 95% CI 0.97–1.25, p = 0.14) (Table 4).

Table 4. Risk factors for first year graft failure. The multivariable Cox regression model for graft failure risk factors, stratified by donor
FactorHazard ratio95% CIp-Value
  1. a

    Reference group.

  2. RRT = renal replacement therapy; GN = glomerulonephritis; HLA = human leukocyte antigen; PRA = panel reactive antibodies.

Right kidney   
 First year1.721.25 – 2.350.0007
 Subsequent years1.060.92 – 1.220.44
Male sex1.170.96 – 1.430.11
Age (years)1.021.01 – 1.020.0002
BMI (kg/m2)   
 <201a 0.008
 20–24.90.970.68 – 1.39 
 25–29.90.800.55 – 1.15 
 ≥301.290.86 – 1.93 
RRT duration (years)   
 <1 year1a 0.008
 1 to <2 years1.120.80 – 1.56 
 2 to <3 years1.230.86 – 1.77 
 3 to <5 years1.511.06 – 2.16 
 ≥5 years1.841.27 – 2.66 
Primary disease   
 GN1a 0.0001
 Polycystic0.850.62 – 1.17 
 Vascular1.160.71 – 1.90 
 Diabetes1.951.40 – 2.70 
 Other1.341.06 – 1.68 
HLA mismatch1.101.00 – 1.200.04
Peak PRA (%)   
 <51a 0.05
 5–791.180.94 – 1.47 
 ≥801.591.09 – 2.34 
Ischemic time (h)   
 <81a 0.05
 8–111.570.96 – 2.55 
 12–171.380.86 – 2.21 
 ≥181.761.08 – 2.89 
Figure 1.

Study design: A paired kidney analysis.

Figure 2.

Graft survival: Kaplan–Meier estimates for kidney survival to 5 years posttransplantation. Graft survival was superior for recipients of left kidneys compared to right, both in terms of death Excluded censored (solid lines) and overall (dotted lines) survival (p<0.01). Survival curves separated within the first year, then became parallel, indicating higher rates of graft loss in the first year for right kidney recipients with no difference beyond that time.

To explore our hypothesis that right kidneys would incur a higher rate of surgical complications, we analyzed all causes of transplant failure during the first year. A higher number of surgical causes (vascular stenosis or thrombosis, local hemorrhage or urological complications) were reported as the cause of graft loss within the first year with right kidneys, 66 versus 35 for left kidneys, with similar results obtained when the analysis was restricted to cases of graft loss within the first week (47 vs. 23 for left kidneys) (Table 5).

Table 5. Causes of graft loss, censored for death, during the first year posttransplantation. During the first year posttransplantation, 349 transplants failed. Causes of graft failure are cumulative and expressed as a percentage of all grafts lost. Principal causes of transplant failure were surgical complications and rejection. Surgical complications included all kidney failures reported as being caused by arterial stenosis or thrombosis, hemorrhage or urological complications
 PrimarySurgicalGlomerular  
 nonfunctioncomplicationsdiseaseRejectionOthers
 DelayLeftRightLeftRightLeftRightLeftRightLeftRight
posttransplantationnnnnnnnnnn
1 week223823471016161122
3 months223833592031292235
6 months223834633132363243
12 months223835665336414855

Patient survival was not different for recipients of right or left kidneys (first-year survival for right kidney recipients: 95.8 vs. 96.0% for left; 5 years survival for right kidney recipients: 87.4 vs. 87.5% for the left, p = 0.77).

Discussion

In this paired analysis, we evaluated the outcome of right and left kidneys recovered from the same deceased donor. Recipients of right-sided kidneys were more likely to experience delayed graft function or slow graft function and were more likely to incur graft loss within the first year after transplantation. Graft losses were more frequently attributed to surgical causes for right kidney recipients. Beyond the first year after transplantation, donor-kidney side was not associated with graft function, graft or patient survival.

The influence of donor kidney side on transplant outcome has rarely been explored in the literature. Three single-center studies have analyzed its impact on DGF [13-15]. Johnson et al., analyzing 201 kidney pairs transplanted in the same center, reported a similar DGF rate for right and left kidneys [15]. However, the DGF rates of 6% and 4% for right and left kidneys respectively were low in comparison to published international rates, close to 20% in the ANZDATA and USRDS registries [17, 18], and the study was not adequately powered to detect a difference. A more recent study of 60 kidney pairs similarly found no difference, but once more with a low DGF rate and a mean cold ischemia time of 3 h, the study was underpowered [14]. Lechevallier et al., in a retrospective analysis that included 257 patients, suggested that right kidney recipients are more frequently affected by DGF with an odds ratio close to 3. However, left and right kidneys were not paired and differences in donor characteristics may have confounded their results [13].

Our present study addresses both key methodological flaws of previous studies by conducting a large, registry-based study with sufficient power. The paired analysis eliminates donor factors other than kidney anatomy as potential sources of bias. Although the reason for right-sided kidneys being more prone to DGF and early graft failure cannot be proven by our retrospective study, the data obtained strongly suggest that the anatomical differences between left and right predispose to a higher degree of surgical difficulty, and consequently, a higher rate of kidney ischemia and other mechanical complications. Anatomic differences between right and left renal vessels are the likely explanation. Indeed, the anatomical alignment of a short donor renal vein and long artery recipient will never be the “best fit” when the recipient iliac vein is always more deeply situated than the iliac artery [8]. The short vein that is either relatively stretched, or requires an extension graft, may incur a greater risk of anastomotic disruption causing hemorrhage, or, thrombosis [19] causing ischemia and venous infarction. The longer artery, unless shortened by excising the aortic patch, may also be at greater risk of kinking causing hemodynamic disturbances and impaired perfusion.

Although our registry data did not include anastomosis times, the greater technical challenges associated with transplantation of a right donor kidney are likely to contribute to longer warm ischemia times and subsequent higher degree of IRI. Right kidneys have also been reported to be smaller than left kidneys [12]. While this would be unlikely to contribute to risk of DGF, lesser kidney mass may be expected to generate inferior kidney function (Brenner theory). Contrary to this hypothesis, we found no difference in kidney function beyond 3 months after transplantation suggesting that either size does not matter, or more likely that any differences in size between left and right does not equate to a similar difference in either nephron number or function.

Our study has several limitations. As a retrospective, registry study, our analysis was restricted to data captured and the quality of that data. In cases of graft loss attributed to surgical causes, specific etiologies such as arterial thrombosis, venous infarction and ureteric obstruction are not captured by the registry. It is possible that uncaptured clinical details, such as side of transplantation, variations in the surgical technique, recipient hypotension and fluid management may have altered the outcomes reported. Our study was restricted to organs obtained from heart-beating, brain-dead organ donors. Extrapolation of our results to elective live donor kidney transplantation, necessarily involving the most experienced surgeons within the transplant team, should be done with caution.

Our significant findings may have implications for practice. Firstly, when two kidneys from a single donor are available for allocation, knowledge that the recipient of the right may be exposed to greater risk of vascular complications and DGF than the recipient of the left kidney could be factored into the allocation algorithm. For example, if two recipients are selected, there may be benefit in allocating the right kidney to the recipient deemed to be at the least risk of DGF or surgical complications. Secondly, there may be benefit in allocating the right kidney first in order to minimize cold ischemia time. Thirdly, and perhaps most importantly, may be the need to allocate the more experienced surgeons to the right-sided donor kidneys.

With surgical experience comes the ability and confidence to compensate for the short right-sided renal donor renal vein by either mobilizing the whole of the iliac vein or using adjacent donor inferior vena cava to extend the renal vein. The latter is the technically easier option, but is only possible if inferior vena cava (IVC) is made available by the donor surgeon. Often the IVC is inadequate because of the need preserve IVC for the donor liver. Equally, the donor renal artery may benefit from being shortened by sacrificing the donor aortic patch that would otherwise facilitate an easier arterial anastomosis.

In conclusion, our data suggest recipients of right-sided kidneys obtained from heart beating, brain dead donors are at greater risk of developing DGF, inferior graft function and greater risk of graft loss in the first year after transplantation. Importantly, there was no apparent impact on either graft survival or function beyond the first year after transplantation. This information may be important to kidney allocation services, to recipient surgeons who may be able to modify timing of implantation, and to researchers examining short-term outcomes after kidney transplantation.

Acknowledgments

H. Vacher-Coponat received a financial support from the Société Francophone de Transplantation fellowship.

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation. The manuscript was not prepared or funded by a commercial institution but was written in its entirety by the authors.

Ancillary