Comparative study of living donor kidney transplants: Right vs left

Abstract

BACKGROUND

Transplant teams often hesitate to use the right kidney (RK) in living donor (LD) transplants due to the complexities of anastomosing the short, thin-walled right renal veins, which can potentially lead to graft loss or graft dysfunction. Nevertheless, circumstances may arise where selecting the RK over the left kidney (LK) is unavoidable. Consequently, it is crucial to thoroughly examine the implications of such a choice on the overall transplant outcome.

AIM

To compare transplant outcomes between recipients of RK and LK while examining the factors that influence these outcomes.

METHODS

We retrospectively analyzed data from adult patients who received LD kidney transplants involving meticulous patient selection and surgical techniques at our center from January 2020 to December 2023. We included all kidney donors who were over 18, fit to donate, and had undergone diethylenetriamine pentaacetic acid split function and/or computed tomography based volumetry. The variables examined comprised donor and recipient demographics, and outcome measures included technical graft loss (TGL), delayed or slow graft function (SGF), and post-transplant serum creatinine (SC) trends. We used a logistic regression model to assess the likelihood of adverse outcomes considering the donor kidney side.

RESULTS

Of the 250 transplants performed during the period, 56 (22%) were RKs. The recipient demographics and transplant factors were comparable for the right and LKs, except that the donor warm and cold ischemia time were shorter for RKs. TGL and SGF each occurred in 2% (n = 1) of RKs and 0.5% (n = 1) of LKs, the difference being insignificant. These complications, however, were not related to the venous anastomosis. One RK (2%) developed delayed graft function after 48 hours, which was attributable to postoperative hypoxia rather than the surgical technique. The post-transplant SC trend and mean SC at the last follow-up were similar across both kidney sides.

CONCLUSION

The donor kidney side has little impact on post-transplant adverse events and graft function in LD transplants, provided that careful patient selection and precise surgical techniques are employed.

Key Words: Living donor; Right kidney; Right renal vein; Venous anastomosis; Technical graft loss; Early graft function

Core Tip: Donor right kidneys (RKs) are associated with technical difficulty, and have a higher incidence of delayed graft function and peri-operative graft loss. Short length and thin wall of the renal vein makes the venous anastomosis challenging. We focus on the selection criteria for RKs and how to manage short veins during bench work, analyze our experience with RKs, and compare outcomes with left kidneys in living donor transplantation.

INTRODUCTION

Kidney transplantation is widely regarded as the best treatment option for end-stage renal disease due to its positive impact on patient survival and quality of life[1]. However, the procedure’s success depends on multiple factors, the most important being the vascular anastomoses between the donor kidney and the recipient vessels.

Anastomosing short and thin-walled veins of right kidneys (RKs) to the recipient’s veins presents more challenges than left kidneys (LKs), which typically have more favorable characteristics; consequently, RKs are more prone to complications[2-5]. Such a risk is increased if the kidney recovery is laparoscopic, where the application of clips and staples shortens the vein, and the option of acquiring a caval cuff to make up for the loss in length is not available[4,6,7]. Grafts with renal veins shorter than 2.5 cm are particularly vulnerable—more so in obese recipients or if the donor kidney is large. The complexities associated with RKs have led to the preference for LKs and their preponderance in living donor (LD) transplants[6]. We analyzed the short-term outcomes of RK and LK transplants in our LD transplant series, which involved meticulous patient selection and surgical technique. We omitted long-term results because complications related to technical issues manifest in the early post-transplant period.

MATERIALS AND METHODS

Study design

We gathered data from hospital records of LD transplants performed at the Kidney Transplant Unit of Rehman Medical Institute in Peshawar, Pakistan, between January 2020 and December 2023. The data were divided into two groups based on the donor kidney side to determine its impact on transplant outcomes. It included donor and recipient demographics with information about the donor kidney’s side, split function, number of arteries, donor nephrectomy types, open [open donor nephrectomy (ODN)] or laparoscopic [laparoscopic donor nephrectomy (LDN)], ischemia times, and immunosuppressant protocols. Outcome measures included rates of immediate graft function (IGF), slow graft function (SGF), delayed graft function (DGF), technical graft loss (TGL), and for graft function, we recorded number of hours it took for the pre-transplant serum creatinine (SC) to decrease by 50% (SC50), SC on day 7 after surgery (SCD7), days taken to normalize SC, and SC at last follow-up (LSC).

We included all kidney donors who were over 18, fit to donate, and had undergone diethylenetriamine pentaacetic acid (DTPA) split function and/or computed tomography (CT) based volumetry. Our sample size was 250, sampling was convenience type, and multiple colleagues were involved in data collection after being trained on the forms and coding used for various variables before data collection. All the information collected were rechecked to minimize errors and avoid bias.

Our choice of donor kidney depended on its vascular anatomy, DTPA-based split function, and CT-based volumetry; we opted for kidneys with a lower function and smaller volume (difference > 15 cc); however, if the right renal vein (RRV) was less than 2.5 cm, we opted for LKs. When both kidneys were approximately the same size, we preferred the RK if the LK had multiple arteries, provided the RRV had the minimum threshold length. In thin females, we even accepted a 2 cm length because their iliac veins are more superficial, and valuable length could be gained by careful circumferential dissection of the RRV in the renal hilum (Figure 1). The right renal artery, generally longer than the vein, was shortened to match the length of the vein, as described elsewhere[2,6].

Figure 1 Right renal vein circumferential mobilization. Lymphatic tissue (yellow arrow) between the artery (white arrow) and the vein (blue arrow), needs to be excised to mobilize the vein.

We used the following operational definitions. Donor warm ischemia (DWI) was defined as time from clamping the renal artery to cold flushing of the graft. Cold ischemia time (CIT) was defined as duration that the graft was stored in a cold solution. Recipient warm ischemia was defined as time from starting vascular anastomosis to graft reperfusion. TGL was defined as graft removal within 7 days of transplantation. DGF was defined as the need for dialysis in the 1^st week after transplant. SGF was defined as SC > 3 mg/dL on day 5 after transplant.

All information gathered from the patient file reviews and electronic medical records was entered into a proforma and then transferred to an excel sheet. All recorded variables were compared between left or right-side nephrectomy to observe the differences in outcomes.

Surgical technique

All grafts were placed in the retroperitoneum with end-to-side anastomosis between the graft vessels and the recipient external iliac vessels or to more proximal vessels. Ice-filled swabs and cold saline irrigation kept the graft cool during the entire implantation procedure. Multiple arteries were implanted individually or placed on a common ostium. The vesico-ureteric anastomosis was extravesical, and fashioned over a stent.

Statistical analysis

Data collection and analysis utilized Microsoft Excel 2013 and Statistical Package for the Social Sciences version 25 (IBM Corp., Armonk, NY, United States). Qualitative data are described as frequencies and percentages, while quantitative data are expressed as mean with standard deviation. The independent sample t-test was employed to assess the statistical significance of differences between groups. Additionally, a linear regression model was utilized to evaluate the influence of confounding variables on outcomes. A post hoc subgroup analysis was performed based on the donor nephrectomy technique, (open or laparoscopic), donor relation (related or unrelated), and immunosuppression protocol (steroid based or steroid sparing). Missing data were addressed through exclusion, and patients lost to follow-up or who died were censored as graft loss. All P values were based on two-sided tests and significance was set at a P value < 0.05.

RESULTS

A total of 250 LD kidney recipients and donors who satisfied the eligibility criteria were included in the study. Table 1 provides details about donor and recipient demographics and Table 2 shows transplant-related ischemia times. Among donors, 126 (50.4%) were male and 124 (49.6%) were female. The mean age of RK donors was 35.1 ± 10.5 years and 33.6 ± 10.1 years for LKs. Two hundred and thirteen (85.2%) recipients were male and thirty-seven (14.7%) were female, with seven females who received RKs (13%) and 30 (16%) who received LKs. The mean age of recipients who received RKs was 37 ± 11.6 years and 38.7 ± 11.3 years for LKs.

Table 1 Donor and recipient demographics and transplant factors.

Parameters	Living donor kidneys (n = 250)
	Right kidney	Left kidney	P value
Donor age in years	35.1 ± 10.5	33.6 ± 10.1	0.36
Donor sex	Males = 28 (50), females = 28 (50)	Males = 98 (51), females = 96 (49)	0.9
Mean donor weight in kg	65.9 ± 8.8	66.2 ± 11	0.9
Mean recipient age in years	37 ± 11.6	38.7 ± 11.3	0.32
Recipient sex	Males = 49 (87), females = 7 (13)	Males = 164 (84), females = 30 (16)	0.6
Mean recipient body mass index	23.6 ± 3.5	24.4 ± 3.8	0.1
Mean human leukocyte antigen mismatches, n ± SD	5 ± 2.9	5.5 ± 2.9	0.32
Donor kidney split function	47 ± 2	50 ± 2	0.000
Donor arteries multiplicity	Single = 51 (91), multiple = 5 (9)	Single = 161 (84), multiple = 31 (16)	0.17
Donor type	LRD = 42 (75), LURD = 14 (25)	LRD = 155 (80), LURD = 39 (20)	0.43
Donor nephrectomy type	ODN = 53 (95), LDN = 3 (5)	ODN = 182 (73), LDN = 68 (27)	0.000
Immunosuppressant, steroid sparing	6 (10.7)	20 (10.3)	0.9
Mean cold ischemia time in minutes	40.1 ± 16.6	48.6 ± 23.6	0.01
Mean donor warm ischemia in minutes	2.1 ± 0.9	2.7 ± 1.7	0.008
Mean recipient warm ischemia in minutes	44.2 ± 5.2	45.3 ± 8.3	0.36
Mean donor warm ischemia of cohort in minutes	ODN = 1.8 ± 0.9, LDN = 4.7 ± 1.2		0.000
Mean cold ischemia time of cohort in minutes	ODN = 41.5 ± 18.3, LDN = 60.4 ± 26.8		0.00

Data are n (%) or mean ± standard deviation (SD), unless otherwise indicated. LDN: Laparoscopic donor nephrectomy; LRD: Living related donor; LURD: Living unrelated donor; ODN: Open donor nephrectomy.

Table 2 Transplant outcomes and post-transplant serum creatinine trends.

Parameter	Right kidney	Left kidney	P value
Immediate graft function	53 (96)	192 (99)	0.15
Slow graft function	1 (1.7)	1 (0.5)
Delayed graft function	1 (1.7)	0
Technical graft loss	1 (1.7)	1 (0.5)
Mean pre-transplant SC in mg/dL	5.90 ± 2.05	5.78 ± 1.57	0.7
Mean time in hours to 50% fall in pre-transplant SC	17.36 ± 7.38	17.82 ± 10.15	0.8
Mean SC on post-transplant day 7 in mg/dL	1.107 ± 0.61	1.073 ± 0.42	0.6
Mean post-transplant time in days to normal SC	4.13 ± 2.848	4.11 ± 2.774	0.9
Mean SC on last follow-up in mg/dL	1.033 ± 0.217	1.038 ± 0.443	0.9

Data are n (%). SC: Serum creatinine.

The total number of transplants was 250, comprising 194 recipients (78%) who received LKs and 56 who received (22%) RKs. One hundred and eighty-two (73%) of the kidneys were recovered by ODN and sixty-eight (27%) by LDN. Of the latter, 65 (95%) were LKs and 3 (5%) were RKs. Compared to LKs, RKs had significantly smaller mean volumes (47 ± 2 vs 50 ± 2). Six (11%) RKs and thirty (15%) LKs had multiple arteries. Mean donor age, sex, weight, and artery multiplicity, along with mean human leukocyte antigen mismatches with recipients were not statistically different between the two groups. The RK DWI and CIT were significantly shorter than those of LKs (2.1 ± 0.9 vs 2.7 ± 1.7 and 40.1 ± 16.6 vs 48.6 ± 23.6, respectively), primarily because RKs were recovered by ODN, which also require less bench work. Significantly shorter DWI and CIT were recorded for ODN compared to LDN (1.81 ± 0.9 vs 4.7 ± 1.2 and 41.5 ± 18.2 vs 60.4 ± 26.8, respectively). The difference in rewarming ischemia (RWI) between right and LKs, or in ODN and LDN was not significant (Table 2).

As shown in Table 2, 96% of recipients of RKs and 99% of LKs had IGF; 1.7% of RKs and none of the LKs had SGF, while 1.7% of RKs and 0.5% of LKs had DGF; the difference was not statistically significant in either case. For both kidneys, SC50, SCD7, and LSC were comparable (Table 3).

Table 3 Cohort post-transplant serum creatinine trend.

Parameter	Frequency		SC to decrease by 50% in hours	P value	Normalize SC	P value	SC at last follow-up	P value	SC on day 7 after surgery in days	P value
Open donor nephrectomy	RK	53 (95)	17.9 ± 10.2	0.6	4 ± 2.7	0.4	0.9 ± 0.2	0.02	1 ± 0.5	0.9
	LK	129 (67)
	Total	182 (73)
Laparoscopic donor nephrectomy	RK	3 (5)	17.2 ± 7.5		4.4 ± 2.8		1.1 ± 0.6		1 ± 0.3
	LK	65 (33)
	Total	68 (27)
Living related donor	RK	42 (75)	17.3 ± 10	0.2	4 ± 2.7	0.2	0.99	0.000	1.0 ± 0.4	0.005
	LK	155 (80)
	Total	197 (79)
Living unrelated donor	RK	14 (25)	19.1 ± 7		4.6 ± 2.9		1.2		1.2 ± 0.8
	LK	39 (20)
	Total	53 (21)
Steroid sparing immunosuppression	RK	6 (11)	15 ± 0.7	0.13	3.5 ± 2.4	0.23	0.97 ± 0.2	0.36	0.9 ± 0.4	0.07
	LK	20 (10)
	Total	26 (10)
Steroid based immunosuppression	RK	50 (89)	18 ± 0.9		4.2 ± 2.8		1 ± 0.4		1.1 ± 0.5
	LK	174 (90)
	Total	224 (90)

Data are n (%).LK: Left kidney; RK: Right kidney; SC: Serum creatinine.

There were two TGLs, one LK and one RK, both caused by arterial bleeding; one was an anastomotic blowout in an LK because of an infected graft recovered by LDN, and the other was a cautery burn away from the anastomosis in an RK.

DISCUSSION

This study is the first such report from Pakistan that addresses the impact of right vs left donor kidney on TGL, delayed and SGF, and post-transplant SC trends. There is substantial evidence that RK recipients may have worse outcomes than LK recipients in LD transplantation[2,4,8-10]. The main reason is the variable length of the RRV, which is usually short with thin, fragile walls (Figure 2), making venous anastomosis difficult, especially in obese recipients with limited exposure. Lifting the graft to access the anastomotic site risks tearing a stretched, fragile vein, and attempts at repair may further extend the tear. Dealing with this can prolong warm ischemia and risks DGF, but rushing the anastomosis to save time may result in partial or complete venous occlusion and thrombosis. If bleeding ensues because the anastomosis is not perfect, the perfused, swollen graft hampers access and control of bleeding becomes challenging. A clear preference among transplant teams for LKs is thus quite understandable[2,6], and the surge in LDN, more suitable for LKs than RKs, has further boosted this preference.

Figure 2 Thin-walled, fragile right renal vein on full stretch.

Nevertheless, under certain circumstances, RKs are a better choice for transplantation. This is commonly the case when the LK vascular anatomy is unfavorable or when there are concerns about compromising the donor residual renal function by removing a larger LK. Thus, it is prudent for transplant surgeons to be aware of the complexities of RK transplantation and the strategies to cope with them. The choice of RKs for transplantation must be a deliberate and a calculated one, based on the verifiable length of the RRV on CT imaging. A tip is to note the angle at which the RRV joins the inferior vena cava (IVC): A horizontal entry makes the length shorter than when it enters at an obtuse angle (Figure 3). We advocate a minimum vein length of 2.5 cm to perform a safe venous anastomosis; severe technical problems can arise with a shorter length, which cannot be mitigated with renal vein augmentation from the adjacent IVC in LDs[11]. A caval cuff can be taken in LDs, but only during ODN, because claims of acquiring a caval cuff in LDN with an Endo-TA stapler remain unsubstantiated[12]. A useful technique to gain a valuable centimeter of length during bench work is to mobilize the renal vein circumferentially and separate it from the artery by ligating and dividing all tributaries and lymphatic tissue between the two (Figure 1). In cases with short donor RRVs, fully mobilizing the recipient external iliac vein helps to lift the vein up, making anastomosis easier as it decreases the distance between the two venotomies and reduces tension on the anastomosis. Another critical point about RRV is identifying a thin oval area in its ventral wall that looks like a structural defect (Figure 4). Suturing through it requires the surgeon and the assistant to be cautious and gentle; any inadvertent tear here is exceedingly difficult to repair and could end up with narrowing, thrombosis, and graft loss[6]. The Achilles heel of an RK is its vein and not the artery, which can be shortened to match vein length even when disproportionately longer. Other considerations in selecting RKs for transplantation include the recipient’s body mass index and the donor kidney size. In obese patients, the challenge of anastomosing a short donor RRV is exacerbated by the deep-seated iliac veins, whereas a large kidney restricts access to the anastomosis site following reperfusion.

Figure 3 Computed tomography images showing right renal vein entry into the inferior vena cava. A: Short vein in horizontal entry because perpendicular is the shortest distance; B: Long vein in obtuse angle entry.

Figure 4 Lengthened right renal vein. The weak oval area is clearly visible (green arrow) and silk ligatures on tissue around the vein and artery (white arrow) are the result of circumferential dissection to mobilize the vein. The caval cuff (blue arrow) was used for suturing to avoid weak area.

In our series, we conducted a comprehensive preoperative assessment of the donor RRV to be sure of its length, and if short, we adopted the techniques described above: First, to increase the vein length during bench work; and second, to fully mobilize the recipient external iliac vein. We used a fine Prolene 6/0 suture with a C-1 needle for the anastomosis, paired with a lightweight Castroviejo needle holder. This choice results in smaller needle holes, reduces venous wall consumption, and ensures precise and smooth suturing. In contrast, a conventional needle holder is bulkier and prone to uncontrolled movements during unlocking, with the potential to tear the vein along the needle tract, and troublesome bleeding after clamp removal. Attention to detail explains why the venous anastomoses in our series were uneventful, as evidenced by there being no differences in RWI between the RKs and LKs. It also accounted for the low incidence of adverse post-transplant events and similar trends in post-transplant SC in both kidneys. DWI and CIT were shorter for RKs than LKs, as the former was mostly recovered by ODN (96% vs 50%), which is a faster method than LDN, and kidneys thus recovered require less bench preparation. Due to the limited cases for analysis, we could not confirm whether these factors influenced the transplant results.

Studies comparing outcomes of right and LK transplants have shown varying results. Analyses of the Organ Procurement and Transplant Network and Australian and New Zealand Dialysis and Transplantation data also supported the association between RK transplants and increased risks of DGF and graft loss[3,5]. Another large registry-based study found that recipients of RKs had a higher risk of DGF and graft loss than those receiving LKs; 36% of transplants with RKs experienced primary non-function, compared to only 9% with LKs[10]. On the contrary, two meta-analyses failed to identify a significant association between the incidence of DGF and graft loss in RK grafts, despite the exclusive use of LDN in the studies analyzed[13,14]. This was unexpected and difficult to explain, because application of staples in laparoscopic recovery consumes length, and makes the anastomosis difficult. The contradictory findings of the reported studies and the findings in our series suggest that there may be other factors beyond vein length that play a role in determining graft outcome in RK transplantation. Identifying these factors would necessitate a more extensive and inclusive study involving a comprehensive dataset encompassing all relevant variables.

The studies that showed RKs resulted in more technical losses and poorer early graft function than LKs did not differ in longer-term results[3,6,8,10]. This is because technical complications manifest in the early postoperative period, and after grafts exhibit immediate function, the side of the kidney does not have a role in longer-term function. Our study further substantiates this observation by demonstrating that the right and LKs perform comparably despite the former having a lesser volume. It demonstrates that the differences in right and LKs are not in their nephron characteristics but in their vascular anatomy. In other words, the longer-term outcome depends on factors other than laterality, such as graft quality, rejection episodes, drug toxicity, drug compliance and disease recurrence. However, early setbacks always have consequences, and grafts that recover from DGF have poorer outcomes than grafts with immediate function, irrespective of whether they are right or left[2,15].

We imagine that the observed inferior outcomes associated with RK transplants may stem from inaccuracies in preoperative donor assessment of RRV length, limited surgical proficiency, and unfamiliarity with the techniques required to manage challenging operative situations. When faced with the dilemma of choosing between an RK with borderline vein length and an LK with multiple arteries, opting for LK may be safer and less risky for a surgeon not well-versed in dealing with fragile, short veins; studies suggest that transplants involving multiple arteries yield better outcomes than those involving kidneys with borderline-length veins[16]. Last, SCs in our recipients who received kidneys recovered by LDN were higher than those receiving kidneys recovered by ODN, suggesting that because RKs were recovered predominantly by ODN, this may have contributed positively to our results.

The main strength of our study is that it represents the largest cohort of LD transplants from Pakistan and is the first to report the impact of selecting right donor kidneys on graft outcomes and provide an insight for transplant surgeons when selecting RKs for transplantation. However, it is also important to recognize limitations of our study: It is retrospective in nature and carried out at a single center. A larger prospective multi-centric study would better identify factors other than donor vein length for disparity in outcomes between groups.

CONCLUSION

Contrary to common perception, our study shows that the donor kidney side has little impact on post-transplant adverse events and graft function in LD transplantation, provided careful patient selection and precise surgical techniques are employed.