© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Clin Transplant 2014: 28: 669–674 DOI: 10.1111/ctr.12363
Clinical Transplantation
Renal retransplantation after kidney and pancreas transplantation using the renal vessels of the failed allograft: pitfalls and pearls Chedid MF, Moreno Gonzales M, Raghavaiah S, Chauhan A, Taner T, Nedredal GI, Park WD, Stegall MD. Renal retransplantation after kidney and pancreas transplantation using the renal vessels of the failed allograft: pitfalls and pearls. Abstract: Renal retransplantation after a failed prior kidney and pancreas transplant is being increasingly performed. In these complex cases, both iliac fossae have been used for prior transplants, and the placement of the new allograft can be problematic. We describe our experience with an alternative technique for renal retransplantation (RRTx) in the setting of severe bilateral aortoiliac atherosclerosis or scarring and fibrosis on the iliac vessels. Nephrectomy of the failed allograft is performed, and the renal vessels of the failed allograft (RVFA) are preserved. The new kidney is implanted on RVFA at the same operative time. This technique was attempted and successfully accomplished in a total of six patients (mean operative time = 240 63 min). One postoperative complication occurred: poor arterial inflow to the allograft, being corrected reoperatively. Hospitalizations ranged from five to eight d. Five of the six patients were alive with a functioning allograft at last follow-up (a single graft failure occurred 21 months postoperatively in the setting of post-transplant lymphoproliferative disease that also led to patient death). Renal vessels of the failed allograft seem to be suitable alternative vascular conduits for renal retransplantation after prior kidney and pancreas transplants.
Marcio F. Chedid, Manuel Moreno Gonzales, Suresh Raghavaiah, Ashutosh Chauhan, Timucin Taner, Geir I. Nedredal, Walter D. Park and Mark D. Stegall Division of Transplantation Surgery, Department of Surgery and von Liebig Transplant Center, Mayo Clinic, Rochester, MN, USA Key words: allograft nephrectomy – operative technique – renal retransplantation – renal vessels of the failed allograft – simultaneous pancreas and kidney transplantation Corresponding author: Mark D. Stegall, MD, Division of Transplantation Surgery, Department of Surgery, Mayo Clinic, 200, 1st St. SW, Rochester, MN 55905, USA. Tel.: +1 507 266 2812; fax: +1 507 266 2810; e-mail: [email protected] Conflict of interest: None Accepted for publication 10 March 2014
The need for renal retransplantation (RRTx) after allograft failure is one of the most common indications for renal transplantation, comprising approximately 15% of the US kidney waiting list (1). Several reports have shown good short- and longterm outcomes after RRTx (2–4). For a second renal transplant, the surgical approach usually involves use of the iliac vessels contralateral to the prior transplant in order to avoid dealing with scar tissue. However, in complex cases, when both iliac fossae have been used for prior kidney and pancreas transplants, the placement of the new allograft can be problematic. Severe diffuse calcified aortoiliac disease and intense fibrotic pelvic scarring on the iliac vessels may complicate the approach and the use of the iliac vessels on RRTx. Here, we report our
experience with an alternative technique for RRTx after pancreas and kidney transplantation in the setting of severe bilateral aortoiliac atherosclerosis or scarring and fibrosis on the iliac vessels. Patients and methods
This IRB-approved study was a retrospective review of all six consecutive RRTx recipients who underwent RRTx after prior kidney and pancreas transplant using the renal vessels of the failed allograft (RVFA). This study was performed in accordance with the ethical standards laid down in an appropriate version of the 2000 Declaration of Helsinki as well as the Declaration of Istanbul 2008, and all persons gave their informed consent prior to their inclusion in the study. Nephrectomy
669
Chedid et al.
of the failed allograft and RRTx using RVFA was performed at the same operative time. All operations were performed at our institution from March 2007 to February 2012. Operative technique
Indications for using the RVFA included diffuse calcified atherosclerotic disease of the aorta and iliac system and scarring and fibrosis on the iliac vessels. Allograft nephrectomy was performed through an extracapsular approach, preferably by the use of a previous transplant incision. Failed kidney allograft was identified, detached from the surrounding fibrous tissue, and its posterior face was exposed. The allograft was then retracted laterally, and its superomedial aspect was gently released from the iliac artery by careful blunt dissection. Inferomedial aspect of the kidney was accessed, and the ureter was identified, doubly ligated, and divided. RVFA are then identified and dissected toward the midline. All remaining attachments to the iliac artery were released. Palpation of the arterial pulse helped individualizing the renal artery and its detachment from the vein. Iliac vessels were routinely identified by palpation, but were not fully dissected or exposed. Whenever possible, vascular clamps were placed on the RVFA to avoid manipulation of the recipient iliac vessels. RVFA were then individually ligated and severed next to the kidney hilum. The failed allograft was removed, and the preserved RVFA were prepared for vascular anastomoses. Vascular anastomoses were performed preferentially in an end-to-end fashion without the use of growth factor (6-0 Prolene for arterial anastomoses and 5-0 Prolene for venous anastomoses). Neither arterial patch nor branch patch were utilized routinely. Urinary reconstruction was accomplished by standard Lich ureteroneocystostomy over a double-J stent. Intra-operative Doppler ultrasound was not performed routinely, but was used selectively in cases where allograft positioning was problematic. We have followed all kidney transplants at our institution with immediate posttransplant ultrasound performed in the recovery room (5). Postoperatively, all patients were assessed for complications including postoperative allograft vessels obstruction, bleeding, reoperations, infections, and kidney allograft function. Pancreas allograft function was determined before and after RRTx. We have compared the outcomes of these six patients who underwent RRTx by the use of RVFA to all other patients whose status were post-kidney and pancreas transplant and under-
670
went RRTx by other operative techniques during the same study period (n = 12). Estimated GFR (eGFR) was calculated from serum creatinine using the Modification of Diet in Renal Disease (MDRD) Study equation. The primary study endpoints were feasibility and safety of the technique, intra-operative bleeding, and immediate and short-term postoperative outcomes such as patency of the vascular anastomoses, kidney allograft function without a need for dialysis. Secondary outcomes were renal function at one yr post-transplant. Data were expressed as mean or median with ranges throughout the manuscript. Comparisons of categorical variables were performed using the chi-square and or two-tailed Fisher’s exact test as appropriate. Comparisons of numerical variables were performed with ANOVA. Results
From March 1st, 2007 to January 31st, 2012, 1077 renal transplants were performed at our institution. Eighteen of those patients had undergone prior kidney and pancreas transplants. Our preferential technique for RRTx after pancreas and kidney transplantation was accomplished by implantation of the new allograft on the native iliac vessels (n = 12). The aorta and/or IVC were used whenever necessary. Alternatively, nephrectomy of the failed allograft and RRTx using RVFA was attempted and accomplished in six patients in whom there was severe bilateral aortoiliac atherosclerosis and/or scarring and fibrosis on the iliac vessels (Table 1). Surgical approach was accomplished via a lower lateral incision in three cases, and a lower midline abdominal incision was used in three cases. One new kidney allograft had an additional upper pole artery that was ligated, and all back-table preparations were standard with no artery reconstructions. Case #1 Indication: diffuse, calcified atherosclerotic disease of the aorta/iliac system. The first case in our series involved a living donor RRTx in a patient who had previously undergone a simultaneous pancreas and kidney transplant (SPKTx). This patient had failure of both pancreas and kidney allografts. He also had an additional subsequent solitary pancreas transplant which had failed as well. Preoperative imaging had revealed diffuse calcified atherosclerotic disease involving the aortoiliac arterial system bilaterally. At the time of RRTx, exploration of these vessels via the prior midline incision confirmed the presence of diffuse and hard calcified vessels bilaterally, precluding the creation of an arterial anastomosis. In contrast, palpation
Prior graft vessels for retransplantation Table 1. Demographics, pre-transplant characteristics, transplant and post-transplant outcomes of the six patients N, Patients Mean age (range, yr) Sex, male Donor, living Mean living-donor cold ischemic time (min) Mean deceased donor cold ischemic time (min) Mean operative time (range) (min) Mean blood loss (range) (mL) N, Immediate postoperative complications Reoperations Mean postoperative stay (range) (d) Mean dismissal serum creatinine (range) (mg/dL) Mean 30th postoperative day serum creatinine (range) (mg/dL) Mean post-transplant follow-up (length) (months) Mean 1-yr serum creatinine (range) (mg/dL) Graft losses
6 54 (51–56) 4 4 55 740 240 (180–345) 379 (50–1100) 1 (delayed graft function) 1 6 (5–8) 1.3 (0.6–2.1) 1.2 (0.8–1.7) 20 (13–25) 1.2 (0.9–1.7) 1 (21 months post-transplant)
of the RVFA revealed a soft artery with a strong pulse. Allograft nephrectomy was performed without difficulty, with the RVFA dissected individually and divided near the renal hilum. The new kidney allograft had a single renal artery and a single renal vein which facilitated end-to-end anastomoses to the RVFA. Postoperative course was unremarkable. The transplant allograft had a good function up to 21 months postoperatively, when the patient developed post-transplant lymphoproliferative disorder and multifactorial acute tubular necrosis. The patient ultimately died 23 months after RRTx. Case #2 Indication: allograft nephrectomy at the time of RRTx for significant proteinuria from the allograft and fibrotic scarring on the native iliac vessels. The second patient in this series also had a complex surgical history. This patient had undergone a prior SPKTx via a midline incision. The pancreas allograft had been converted from bladder drainage to enteric drainage years before and was still functioning. The first kidney allograft had failed, and the patient had required hemodialysis for seven months prior to the RRTx. The failed renal allograft continued to make urine and releasing 2541 mg of protein per 24 h. The decision was made to perform an allograft nephrectomy at the time of the RRTx in order to reduce protein loss. The failed allograft was approached via a left lower quadrant incision. Identification and dissection of RVFA were accomplished with no major
difficulties in this patient who was still on immunosuppressive medications. RVFA were long, easily accessible and had soft consistency. The native iliac vessels, in contrast, were involved with significant scar tissue due to prior dissection. RVFA were dissected individually, divided near the hilum, and prepared as new conduits (Fig. 1A). The new allograft had a single renal artery and single renal vein, allowing for end-to-end anastomoses to the RVFA (Fig. 1B). The new kidney allograft functioned well immediately, and the patient had an uneventful postoperative course up to the last follow-up at 25 months post-transplant. Subsequent experience and possible pitfalls
We utilized the RVFA in three other cases in which local conditions (fibrotic scarring) complicated dissection and isolation of the iliac vessels. In a yet additional case, severe iliac atherosclerosis was only detected intra-operatively despite a negative preoperative ultrasound. These were three patients with prior SPKTx, and one patient with a prior pancreas after kidney transplant. The only postoperative complication revealed a possible pitfall in our approach. In this case, the A
B
Fig. 1. (A) Following allograft nephrectomy, renal vessels of the failed allograft (RVFA) being prepared for RRTx. (B) New kidney allograft implanted using RVFA.
671
Chedid et al.
recipient’s prior renal allograft was not functioning and this patient had been on hemodialysis for a prolonged period prior to retransplantation (2.6 yr). We thus proceeded with nephrectomy of the failed allograft. As the arterial blood and venous blood flow through RVFA appeared to be adequate, the new kidney allograft was sewn to these vessels. However, a postoperative ultrasound demonstrated poor arterial inflow and increased serum creatinine. The patient was reexplored and the renal artery anastomosis was taken down, and a new anastomosis was made onto the native common iliac artery proximal to the site where the failed allograft had been connected. In this series of six patients, all allografts functioned well after transplantation (nadir creatinine in the first two wk 1.06 0.3 mg/dL). The mean length of the post-transplant follow-up period was 20 months (table 1). The only kidney allograft loss occurred at 21 months postoperatively (see case 1 above). Five of these six patients had a functioning pancreas allograft prior to RRTx. At one yr, four of them had a functioning allograft and the remaining patient had a failing pancreas allograft with only partial pancreatic function. Conventional technique
During this study period, 12 additional patients who had prior kidney and pancreas transplants underwent RRTx not using both the vein and the artery of the failed allograft for implantation of the RRTx graft. A midline incision was used in ten of these 12 patients. Two of these 12 patients had allograft nephrectomy performed along with RRTx, being the artery of the new allograft sewn to the iliac vessels, and only the vein of the RRTx graft connected to the vein of the failed allograft. Nine of these 12 patients underwent RRTx by connecting both the artery and the vein to iliac vessels (including two patients in whom the vein was sewn to IVC). In yet one additional patient, review of the RRTx operative report did not provide enough information to determine the reconstruction technique used on RRTx. Eleven of these 12 patients had a functioning kidney allograft one yr after RRTx (one patient died with a functioning kidney allograft). With respect to pancreas allograft, nine of 12 patients had a functioning pancreas allograft at the time of RRTX and pancreatic function was preserved in eight of those patients one yr after RRTx (the remaining patient died of operative complications secondary to RRTx with functioning kidney and pancreas allografts). Kidney RRTx allograft was sewn contralateral to the pancreas allograft in six
672
cases and ipsilateral to the pancreas in three cases. Comparing approaches
We have performed comparisons among the six patients who underwent RRTx using RFVA, the two patients in whom only the renal vein of the failed allograft was used and the nine patients retransplanted using the iliac vessels. There were no statistical differences on blood loss (p = 0.20), operative time (p = 0.73), and eGFR at one yr (p = 0.48). Discussion
This series suggests that RRTx using RVFA is a feasible alternative to conventional techniques and may be an option for complex cases. This study supports a prior series showing a high success rate (6). A recent series have included two cases of RRTx after SPKTx in which only the renal vein of the failed allograft was used suggesting that the vein may be used selectively for RRTx (7). In the control group of the present series, we also include two cases of RRTx combining the use of the iliac artery and renal vein of the failed allograft, confirming the feasibility of this hybrid technique. Using RVFA is not the ideal procedure for all RRTx. For recipients with one prior renal transplant and no prior pancreas transplants, and those who do not have diffuse atherosclerotic disease, we support the common practice of performing RRTx via a new incision contralateral to the prior transplant. One of the most common situations in which the RVFA might be considered is the presence of dense bilateral aortoiliac atherosclerosis, which is not an uncommon finding on RRTx after a failed kidney allograft on patients who underwent prior pancreas and kidney transplants. In this setting, options for placement of a new kidney allograft in a non-scarred area are limited. Historically, we explored such recipients via a midline incision and inspected both iliac arteries for an acceptably healthy segment of native iliac artery. This could be on the same side as the prior renal transplant. We also encountered cases in which the only “soft spot” was superior to the pancreas allograft. While we have used this location successfully, we contend that this approach may be risky given the potential for damage to the pancreas allograft. The use of a midline incision has the advantage of providing full access to the abdominal cavity, enabling bilateral access to the iliac vessels. Alter-
Prior graft vessels for retransplantation natively, whenever allograft nephrectomy is planned, an incision ipsilateral to the failed renal allograft provides direct access to the failed allograft. If preoperative Doppler ultrasound confirms satisfactory anatomy of the iliac vessels ipsilateral to the failed allograft, we have used a lateral abdominal incision. Therefore, a midline incision is reserved for exceptional cases in which the iliac system ipsilateral to the failed allograft has considerably more atherosclerosis than the side that contains the pancreas. Other alternative approaches to using the RVFA in patients with severe atherosclerosis might include endarterectomy or creation of an arterial bypass (8,9). In this setting, two studies have reported a 10–14.3% patient mortality and a 16.6– 28.5% early postoperative morbidity following simultaneous aortoiliac vascular procedures and kidney transplantation (8). Additional concern with this approach is the potential risk of prosthesis infection with or without the need for prosthesis removal and allograft nephrectomy. Accessing the RVFA and preparing those vessels for allograft implantation appeared to require less dissection than would be necessary for approaching other vessels (native renal, internal iliac, or mesenteric vessels). Our experience highlights the need for adequate preoperative assessment of the recipient’s arterial system. Our current protocol for RRTx after previous pancreas and kidney transplants or after two or more previous renal transplants for patients who did not undergo prior nephrectomy
of the failed allograft is summarized in Figure 2. We routinely perform preoperative ultrasound to document adequate arterial flow into both the iliac vessels and the failed allograft. Whether the routine use of preoperative magnetic resonance angiography or other approaches would be superior to ultrasound evaluation is unclear. In the absence of dense atherosclerosis, we still prefer the recipient’s native external or common iliac vessels as conduits for RRTx. In contrast, RVFA are always considered as an option if there is dense atherosclerosis of the native arteries and the failed allograft still has residual function suggesting good inflow. Although we commonly remove the failed allograft to provide room for the new one, we do not always do this if there is adequate space. In most cases, the final decision of allograft placement is made in the operating room based on the assessment of blood flow to the failed allograft and ease of approaching the native vessels. The only arterial inflow complication in the present series (poor arterial inflow and increased serum creatinine) was reoperatively corrected by reimplanting the artery of the new kidney allograft on the external iliac artery. This complication highlights the fact that RVFA should be used with caution and good surgical judgment will always need to be exercised in those complex RRTx cases. We have used the RVFA for RRTx successfully on few patients undergoing a third or subsequent kidney transplant whenever atherosclerosis and/or
Abdominal Imaging (US / MRA)
Fig. 2. Algorithm for RRTx after prior pancreas and kidney transplantation or for a third or subsequent renal transplant. If preoperative abdominal imaging examination verifies patency of renal vessels of the failed allograft (RVFA), the prior renal transplant incision is used and the failed allograft is removed. If RVFA are patent and demonstrate good arterial inflow at the time of nephrectomy, they may be used as the vascular conduits for the new allograft. *Aortoiliac reconstruction should be considered in the case of extensive bilateral aortoiliac atherosclerosis and occluded RVFA. US, ultrasound; MRA, magnetic resonance angiography.
Patent RVFA
Severe Atherosclerosis with Occluded RVFA
Proceed to Prior Abdominal Incision and Perform Allograft Nephrectomy
Iliac Atherosclerosis
No
Yes
Favorable Iliac Vessels Anatomy Yes
Use Iliac Vessels
No
Check RVFA
Patent RVFA Use RVFA
Occluded RVFA
Midline Incision and Bilateral Exploration
673
Chedid et al.
fibrosis of the iliac vessels is a concern (data not shown). A potential limitation for the use of the RVFA would be the presence of a failed allograft with multiple arteries, especially in the setting of two prior arterial anastomoses to the iliac vessels. The kidney is responsible for insulin clearance, and dialytic renal failure may masquerade a marginally functioning pancreatic allograft (10). In patients with a prior SPKTx, RRTx may be followed by early loss of the pancreatic allograft in up to 25% of the instances (10). In our series, there was one pancreas allograft failure in the group where RVFA were used for RRTx. There were no pancreatic failures in the control group. Although our experience with RRTx using RVFA is limited by small sample size and short follow-up, this technique does not seem to be more harmful to pancreas allograft function than the use of the iliac vessels ipsilateral to the prior allograft. In conclusion, the use of RVFA appears to be a safe alternative technique that might be considered for some patients undergoing RRTx. This case series suggests that RVFA may be a useful alternative in complex cases whenever atherosclerosis or scarring complicates the use of the native iliac vessels, especially when nephrectomy of the failed graft is warranted (e.g., massive proteinuria). The best surgical approach for RRTx in complex settings such as that of a prior SPKTx or a third renal transplant is still open to debate. Further experience is needed to evaluate the long-term outcomes of the surgical technique presented here. Authors’ contributions
Marcio F. Chedid: participated in research design, collection of research data, data analysis, and writing of the manuscript; Manuel Moreno Gonzales and Walter D. Park: participated in collection of research data, data analysis and
674
critical analysis of the manuscript, Suresh Raghavaiah, Ashutosh Chauhan, Timucin Taner, Geir I. Nedredal, and Mark D. Stegall: participated in research design, data analysis, and writing of the manuscript. Funding
This study was not supported by any funding. References 1. www.unos.org, accessed October, 2011. 2. OJO A, WOLFE RA, AGODOA LY et al. Prognosis after primary renal transplant failure and the beneficial effects of repeat transplantation: multivariate analyses from the United States Renal Data System. Transplantation 1998: 66: 651. 3. HAGAN C, HICKEY DP, LITTLE DM. A single-center study of the technical aspects and outcome of third and subsequent renal transplants. Transplantation 2003: 75: 1687. 4. RAO PS, SCHAUBEL DE, WEI G, FENTON SS. Evaluating the survival benefit of kidney retransplantation. Transplantation 2006: 82: 669. 5. CHEDID MF, MUTHU C, NYBERG SL et al. Living donor kidney transplantation using laparoscopically procured multiple renal artery kidneys and right kidneys. J Am Coll Surg 2013: 217: 144. 6. NGHIEM DD. Orthotopic kidney retransplantation in simultaneous pancreas kidney transplant patients with renal failure. Transplant Proc 2008: 40: 3609. 7. LAMATTINA JC, SOLLINGER HW, BECKER YT, MEZRICH JD, PIRSCH JD, ODORICO JS. Simultaneous pancreas and kidney (SPK) retransplantation in prior SPK recipients. Clin Transplant 2012: 26: 495. 8. TSIVIAN M, NERI F, NARDO B et al. Aortoiliac surgery concomitant with kidney transplantation: a single center experience. Clin Transplant 2009: 23: 164. 9. PITTALUGA P, HASSEN-KHODJA R, CASSUTO-VIGUIER E et al. Aortoiliac reconstruction and kidney transplantation: a multicenter study. Ann Vasc Surg 1998: 12: 529. 10. LAMATTINA J, SOLLINGER H, BECKER Y, MEZRICH J, PIRSCH J, ODORICO J. Long-term pancreatic allograft survival after renal retransplantation in prior simultaneous pancreaskidney recipients. Am J Transplant 2012: 12: 937.
Clin Transplant 2014: 28: 669–674 DOI: 10.1111/ctr.12363
Clinical Transplantation
Renal retransplantation after kidney and pancreas transplantation using the renal vessels of the failed allograft: pitfalls and pearls Chedid MF, Moreno Gonzales M, Raghavaiah S, Chauhan A, Taner T, Nedredal GI, Park WD, Stegall MD. Renal retransplantation after kidney and pancreas transplantation using the renal vessels of the failed allograft: pitfalls and pearls. Abstract: Renal retransplantation after a failed prior kidney and pancreas transplant is being increasingly performed. In these complex cases, both iliac fossae have been used for prior transplants, and the placement of the new allograft can be problematic. We describe our experience with an alternative technique for renal retransplantation (RRTx) in the setting of severe bilateral aortoiliac atherosclerosis or scarring and fibrosis on the iliac vessels. Nephrectomy of the failed allograft is performed, and the renal vessels of the failed allograft (RVFA) are preserved. The new kidney is implanted on RVFA at the same operative time. This technique was attempted and successfully accomplished in a total of six patients (mean operative time = 240 63 min). One postoperative complication occurred: poor arterial inflow to the allograft, being corrected reoperatively. Hospitalizations ranged from five to eight d. Five of the six patients were alive with a functioning allograft at last follow-up (a single graft failure occurred 21 months postoperatively in the setting of post-transplant lymphoproliferative disease that also led to patient death). Renal vessels of the failed allograft seem to be suitable alternative vascular conduits for renal retransplantation after prior kidney and pancreas transplants.
Marcio F. Chedid, Manuel Moreno Gonzales, Suresh Raghavaiah, Ashutosh Chauhan, Timucin Taner, Geir I. Nedredal, Walter D. Park and Mark D. Stegall Division of Transplantation Surgery, Department of Surgery and von Liebig Transplant Center, Mayo Clinic, Rochester, MN, USA Key words: allograft nephrectomy – operative technique – renal retransplantation – renal vessels of the failed allograft – simultaneous pancreas and kidney transplantation Corresponding author: Mark D. Stegall, MD, Division of Transplantation Surgery, Department of Surgery, Mayo Clinic, 200, 1st St. SW, Rochester, MN 55905, USA. Tel.: +1 507 266 2812; fax: +1 507 266 2810; e-mail: [email protected] Conflict of interest: None Accepted for publication 10 March 2014
The need for renal retransplantation (RRTx) after allograft failure is one of the most common indications for renal transplantation, comprising approximately 15% of the US kidney waiting list (1). Several reports have shown good short- and longterm outcomes after RRTx (2–4). For a second renal transplant, the surgical approach usually involves use of the iliac vessels contralateral to the prior transplant in order to avoid dealing with scar tissue. However, in complex cases, when both iliac fossae have been used for prior kidney and pancreas transplants, the placement of the new allograft can be problematic. Severe diffuse calcified aortoiliac disease and intense fibrotic pelvic scarring on the iliac vessels may complicate the approach and the use of the iliac vessels on RRTx. Here, we report our
experience with an alternative technique for RRTx after pancreas and kidney transplantation in the setting of severe bilateral aortoiliac atherosclerosis or scarring and fibrosis on the iliac vessels. Patients and methods
This IRB-approved study was a retrospective review of all six consecutive RRTx recipients who underwent RRTx after prior kidney and pancreas transplant using the renal vessels of the failed allograft (RVFA). This study was performed in accordance with the ethical standards laid down in an appropriate version of the 2000 Declaration of Helsinki as well as the Declaration of Istanbul 2008, and all persons gave their informed consent prior to their inclusion in the study. Nephrectomy
669
Chedid et al.
of the failed allograft and RRTx using RVFA was performed at the same operative time. All operations were performed at our institution from March 2007 to February 2012. Operative technique
Indications for using the RVFA included diffuse calcified atherosclerotic disease of the aorta and iliac system and scarring and fibrosis on the iliac vessels. Allograft nephrectomy was performed through an extracapsular approach, preferably by the use of a previous transplant incision. Failed kidney allograft was identified, detached from the surrounding fibrous tissue, and its posterior face was exposed. The allograft was then retracted laterally, and its superomedial aspect was gently released from the iliac artery by careful blunt dissection. Inferomedial aspect of the kidney was accessed, and the ureter was identified, doubly ligated, and divided. RVFA are then identified and dissected toward the midline. All remaining attachments to the iliac artery were released. Palpation of the arterial pulse helped individualizing the renal artery and its detachment from the vein. Iliac vessels were routinely identified by palpation, but were not fully dissected or exposed. Whenever possible, vascular clamps were placed on the RVFA to avoid manipulation of the recipient iliac vessels. RVFA were then individually ligated and severed next to the kidney hilum. The failed allograft was removed, and the preserved RVFA were prepared for vascular anastomoses. Vascular anastomoses were performed preferentially in an end-to-end fashion without the use of growth factor (6-0 Prolene for arterial anastomoses and 5-0 Prolene for venous anastomoses). Neither arterial patch nor branch patch were utilized routinely. Urinary reconstruction was accomplished by standard Lich ureteroneocystostomy over a double-J stent. Intra-operative Doppler ultrasound was not performed routinely, but was used selectively in cases where allograft positioning was problematic. We have followed all kidney transplants at our institution with immediate posttransplant ultrasound performed in the recovery room (5). Postoperatively, all patients were assessed for complications including postoperative allograft vessels obstruction, bleeding, reoperations, infections, and kidney allograft function. Pancreas allograft function was determined before and after RRTx. We have compared the outcomes of these six patients who underwent RRTx by the use of RVFA to all other patients whose status were post-kidney and pancreas transplant and under-
670
went RRTx by other operative techniques during the same study period (n = 12). Estimated GFR (eGFR) was calculated from serum creatinine using the Modification of Diet in Renal Disease (MDRD) Study equation. The primary study endpoints were feasibility and safety of the technique, intra-operative bleeding, and immediate and short-term postoperative outcomes such as patency of the vascular anastomoses, kidney allograft function without a need for dialysis. Secondary outcomes were renal function at one yr post-transplant. Data were expressed as mean or median with ranges throughout the manuscript. Comparisons of categorical variables were performed using the chi-square and or two-tailed Fisher’s exact test as appropriate. Comparisons of numerical variables were performed with ANOVA. Results
From March 1st, 2007 to January 31st, 2012, 1077 renal transplants were performed at our institution. Eighteen of those patients had undergone prior kidney and pancreas transplants. Our preferential technique for RRTx after pancreas and kidney transplantation was accomplished by implantation of the new allograft on the native iliac vessels (n = 12). The aorta and/or IVC were used whenever necessary. Alternatively, nephrectomy of the failed allograft and RRTx using RVFA was attempted and accomplished in six patients in whom there was severe bilateral aortoiliac atherosclerosis and/or scarring and fibrosis on the iliac vessels (Table 1). Surgical approach was accomplished via a lower lateral incision in three cases, and a lower midline abdominal incision was used in three cases. One new kidney allograft had an additional upper pole artery that was ligated, and all back-table preparations were standard with no artery reconstructions. Case #1 Indication: diffuse, calcified atherosclerotic disease of the aorta/iliac system. The first case in our series involved a living donor RRTx in a patient who had previously undergone a simultaneous pancreas and kidney transplant (SPKTx). This patient had failure of both pancreas and kidney allografts. He also had an additional subsequent solitary pancreas transplant which had failed as well. Preoperative imaging had revealed diffuse calcified atherosclerotic disease involving the aortoiliac arterial system bilaterally. At the time of RRTx, exploration of these vessels via the prior midline incision confirmed the presence of diffuse and hard calcified vessels bilaterally, precluding the creation of an arterial anastomosis. In contrast, palpation
Prior graft vessels for retransplantation Table 1. Demographics, pre-transplant characteristics, transplant and post-transplant outcomes of the six patients N, Patients Mean age (range, yr) Sex, male Donor, living Mean living-donor cold ischemic time (min) Mean deceased donor cold ischemic time (min) Mean operative time (range) (min) Mean blood loss (range) (mL) N, Immediate postoperative complications Reoperations Mean postoperative stay (range) (d) Mean dismissal serum creatinine (range) (mg/dL) Mean 30th postoperative day serum creatinine (range) (mg/dL) Mean post-transplant follow-up (length) (months) Mean 1-yr serum creatinine (range) (mg/dL) Graft losses
6 54 (51–56) 4 4 55 740 240 (180–345) 379 (50–1100) 1 (delayed graft function) 1 6 (5–8) 1.3 (0.6–2.1) 1.2 (0.8–1.7) 20 (13–25) 1.2 (0.9–1.7) 1 (21 months post-transplant)
of the RVFA revealed a soft artery with a strong pulse. Allograft nephrectomy was performed without difficulty, with the RVFA dissected individually and divided near the renal hilum. The new kidney allograft had a single renal artery and a single renal vein which facilitated end-to-end anastomoses to the RVFA. Postoperative course was unremarkable. The transplant allograft had a good function up to 21 months postoperatively, when the patient developed post-transplant lymphoproliferative disorder and multifactorial acute tubular necrosis. The patient ultimately died 23 months after RRTx. Case #2 Indication: allograft nephrectomy at the time of RRTx for significant proteinuria from the allograft and fibrotic scarring on the native iliac vessels. The second patient in this series also had a complex surgical history. This patient had undergone a prior SPKTx via a midline incision. The pancreas allograft had been converted from bladder drainage to enteric drainage years before and was still functioning. The first kidney allograft had failed, and the patient had required hemodialysis for seven months prior to the RRTx. The failed renal allograft continued to make urine and releasing 2541 mg of protein per 24 h. The decision was made to perform an allograft nephrectomy at the time of the RRTx in order to reduce protein loss. The failed allograft was approached via a left lower quadrant incision. Identification and dissection of RVFA were accomplished with no major
difficulties in this patient who was still on immunosuppressive medications. RVFA were long, easily accessible and had soft consistency. The native iliac vessels, in contrast, were involved with significant scar tissue due to prior dissection. RVFA were dissected individually, divided near the hilum, and prepared as new conduits (Fig. 1A). The new allograft had a single renal artery and single renal vein, allowing for end-to-end anastomoses to the RVFA (Fig. 1B). The new kidney allograft functioned well immediately, and the patient had an uneventful postoperative course up to the last follow-up at 25 months post-transplant. Subsequent experience and possible pitfalls
We utilized the RVFA in three other cases in which local conditions (fibrotic scarring) complicated dissection and isolation of the iliac vessels. In a yet additional case, severe iliac atherosclerosis was only detected intra-operatively despite a negative preoperative ultrasound. These were three patients with prior SPKTx, and one patient with a prior pancreas after kidney transplant. The only postoperative complication revealed a possible pitfall in our approach. In this case, the A
B
Fig. 1. (A) Following allograft nephrectomy, renal vessels of the failed allograft (RVFA) being prepared for RRTx. (B) New kidney allograft implanted using RVFA.
671
Chedid et al.
recipient’s prior renal allograft was not functioning and this patient had been on hemodialysis for a prolonged period prior to retransplantation (2.6 yr). We thus proceeded with nephrectomy of the failed allograft. As the arterial blood and venous blood flow through RVFA appeared to be adequate, the new kidney allograft was sewn to these vessels. However, a postoperative ultrasound demonstrated poor arterial inflow and increased serum creatinine. The patient was reexplored and the renal artery anastomosis was taken down, and a new anastomosis was made onto the native common iliac artery proximal to the site where the failed allograft had been connected. In this series of six patients, all allografts functioned well after transplantation (nadir creatinine in the first two wk 1.06 0.3 mg/dL). The mean length of the post-transplant follow-up period was 20 months (table 1). The only kidney allograft loss occurred at 21 months postoperatively (see case 1 above). Five of these six patients had a functioning pancreas allograft prior to RRTx. At one yr, four of them had a functioning allograft and the remaining patient had a failing pancreas allograft with only partial pancreatic function. Conventional technique
During this study period, 12 additional patients who had prior kidney and pancreas transplants underwent RRTx not using both the vein and the artery of the failed allograft for implantation of the RRTx graft. A midline incision was used in ten of these 12 patients. Two of these 12 patients had allograft nephrectomy performed along with RRTx, being the artery of the new allograft sewn to the iliac vessels, and only the vein of the RRTx graft connected to the vein of the failed allograft. Nine of these 12 patients underwent RRTx by connecting both the artery and the vein to iliac vessels (including two patients in whom the vein was sewn to IVC). In yet one additional patient, review of the RRTx operative report did not provide enough information to determine the reconstruction technique used on RRTx. Eleven of these 12 patients had a functioning kidney allograft one yr after RRTx (one patient died with a functioning kidney allograft). With respect to pancreas allograft, nine of 12 patients had a functioning pancreas allograft at the time of RRTX and pancreatic function was preserved in eight of those patients one yr after RRTx (the remaining patient died of operative complications secondary to RRTx with functioning kidney and pancreas allografts). Kidney RRTx allograft was sewn contralateral to the pancreas allograft in six
672
cases and ipsilateral to the pancreas in three cases. Comparing approaches
We have performed comparisons among the six patients who underwent RRTx using RFVA, the two patients in whom only the renal vein of the failed allograft was used and the nine patients retransplanted using the iliac vessels. There were no statistical differences on blood loss (p = 0.20), operative time (p = 0.73), and eGFR at one yr (p = 0.48). Discussion
This series suggests that RRTx using RVFA is a feasible alternative to conventional techniques and may be an option for complex cases. This study supports a prior series showing a high success rate (6). A recent series have included two cases of RRTx after SPKTx in which only the renal vein of the failed allograft was used suggesting that the vein may be used selectively for RRTx (7). In the control group of the present series, we also include two cases of RRTx combining the use of the iliac artery and renal vein of the failed allograft, confirming the feasibility of this hybrid technique. Using RVFA is not the ideal procedure for all RRTx. For recipients with one prior renal transplant and no prior pancreas transplants, and those who do not have diffuse atherosclerotic disease, we support the common practice of performing RRTx via a new incision contralateral to the prior transplant. One of the most common situations in which the RVFA might be considered is the presence of dense bilateral aortoiliac atherosclerosis, which is not an uncommon finding on RRTx after a failed kidney allograft on patients who underwent prior pancreas and kidney transplants. In this setting, options for placement of a new kidney allograft in a non-scarred area are limited. Historically, we explored such recipients via a midline incision and inspected both iliac arteries for an acceptably healthy segment of native iliac artery. This could be on the same side as the prior renal transplant. We also encountered cases in which the only “soft spot” was superior to the pancreas allograft. While we have used this location successfully, we contend that this approach may be risky given the potential for damage to the pancreas allograft. The use of a midline incision has the advantage of providing full access to the abdominal cavity, enabling bilateral access to the iliac vessels. Alter-
Prior graft vessels for retransplantation natively, whenever allograft nephrectomy is planned, an incision ipsilateral to the failed renal allograft provides direct access to the failed allograft. If preoperative Doppler ultrasound confirms satisfactory anatomy of the iliac vessels ipsilateral to the failed allograft, we have used a lateral abdominal incision. Therefore, a midline incision is reserved for exceptional cases in which the iliac system ipsilateral to the failed allograft has considerably more atherosclerosis than the side that contains the pancreas. Other alternative approaches to using the RVFA in patients with severe atherosclerosis might include endarterectomy or creation of an arterial bypass (8,9). In this setting, two studies have reported a 10–14.3% patient mortality and a 16.6– 28.5% early postoperative morbidity following simultaneous aortoiliac vascular procedures and kidney transplantation (8). Additional concern with this approach is the potential risk of prosthesis infection with or without the need for prosthesis removal and allograft nephrectomy. Accessing the RVFA and preparing those vessels for allograft implantation appeared to require less dissection than would be necessary for approaching other vessels (native renal, internal iliac, or mesenteric vessels). Our experience highlights the need for adequate preoperative assessment of the recipient’s arterial system. Our current protocol for RRTx after previous pancreas and kidney transplants or after two or more previous renal transplants for patients who did not undergo prior nephrectomy
of the failed allograft is summarized in Figure 2. We routinely perform preoperative ultrasound to document adequate arterial flow into both the iliac vessels and the failed allograft. Whether the routine use of preoperative magnetic resonance angiography or other approaches would be superior to ultrasound evaluation is unclear. In the absence of dense atherosclerosis, we still prefer the recipient’s native external or common iliac vessels as conduits for RRTx. In contrast, RVFA are always considered as an option if there is dense atherosclerosis of the native arteries and the failed allograft still has residual function suggesting good inflow. Although we commonly remove the failed allograft to provide room for the new one, we do not always do this if there is adequate space. In most cases, the final decision of allograft placement is made in the operating room based on the assessment of blood flow to the failed allograft and ease of approaching the native vessels. The only arterial inflow complication in the present series (poor arterial inflow and increased serum creatinine) was reoperatively corrected by reimplanting the artery of the new kidney allograft on the external iliac artery. This complication highlights the fact that RVFA should be used with caution and good surgical judgment will always need to be exercised in those complex RRTx cases. We have used the RVFA for RRTx successfully on few patients undergoing a third or subsequent kidney transplant whenever atherosclerosis and/or
Abdominal Imaging (US / MRA)
Fig. 2. Algorithm for RRTx after prior pancreas and kidney transplantation or for a third or subsequent renal transplant. If preoperative abdominal imaging examination verifies patency of renal vessels of the failed allograft (RVFA), the prior renal transplant incision is used and the failed allograft is removed. If RVFA are patent and demonstrate good arterial inflow at the time of nephrectomy, they may be used as the vascular conduits for the new allograft. *Aortoiliac reconstruction should be considered in the case of extensive bilateral aortoiliac atherosclerosis and occluded RVFA. US, ultrasound; MRA, magnetic resonance angiography.
Patent RVFA
Severe Atherosclerosis with Occluded RVFA
Proceed to Prior Abdominal Incision and Perform Allograft Nephrectomy
Iliac Atherosclerosis
No
Yes
Favorable Iliac Vessels Anatomy Yes
Use Iliac Vessels
No
Check RVFA
Patent RVFA Use RVFA
Occluded RVFA
Midline Incision and Bilateral Exploration
673
Chedid et al.
fibrosis of the iliac vessels is a concern (data not shown). A potential limitation for the use of the RVFA would be the presence of a failed allograft with multiple arteries, especially in the setting of two prior arterial anastomoses to the iliac vessels. The kidney is responsible for insulin clearance, and dialytic renal failure may masquerade a marginally functioning pancreatic allograft (10). In patients with a prior SPKTx, RRTx may be followed by early loss of the pancreatic allograft in up to 25% of the instances (10). In our series, there was one pancreas allograft failure in the group where RVFA were used for RRTx. There were no pancreatic failures in the control group. Although our experience with RRTx using RVFA is limited by small sample size and short follow-up, this technique does not seem to be more harmful to pancreas allograft function than the use of the iliac vessels ipsilateral to the prior allograft. In conclusion, the use of RVFA appears to be a safe alternative technique that might be considered for some patients undergoing RRTx. This case series suggests that RVFA may be a useful alternative in complex cases whenever atherosclerosis or scarring complicates the use of the native iliac vessels, especially when nephrectomy of the failed graft is warranted (e.g., massive proteinuria). The best surgical approach for RRTx in complex settings such as that of a prior SPKTx or a third renal transplant is still open to debate. Further experience is needed to evaluate the long-term outcomes of the surgical technique presented here. Authors’ contributions
Marcio F. Chedid: participated in research design, collection of research data, data analysis, and writing of the manuscript; Manuel Moreno Gonzales and Walter D. Park: participated in collection of research data, data analysis and
674
critical analysis of the manuscript, Suresh Raghavaiah, Ashutosh Chauhan, Timucin Taner, Geir I. Nedredal, and Mark D. Stegall: participated in research design, data analysis, and writing of the manuscript. Funding
This study was not supported by any funding. References 1. www.unos.org, accessed October, 2011. 2. OJO A, WOLFE RA, AGODOA LY et al. Prognosis after primary renal transplant failure and the beneficial effects of repeat transplantation: multivariate analyses from the United States Renal Data System. Transplantation 1998: 66: 651. 3. HAGAN C, HICKEY DP, LITTLE DM. A single-center study of the technical aspects and outcome of third and subsequent renal transplants. Transplantation 2003: 75: 1687. 4. RAO PS, SCHAUBEL DE, WEI G, FENTON SS. Evaluating the survival benefit of kidney retransplantation. Transplantation 2006: 82: 669. 5. CHEDID MF, MUTHU C, NYBERG SL et al. Living donor kidney transplantation using laparoscopically procured multiple renal artery kidneys and right kidneys. J Am Coll Surg 2013: 217: 144. 6. NGHIEM DD. Orthotopic kidney retransplantation in simultaneous pancreas kidney transplant patients with renal failure. Transplant Proc 2008: 40: 3609. 7. LAMATTINA JC, SOLLINGER HW, BECKER YT, MEZRICH JD, PIRSCH JD, ODORICO JS. Simultaneous pancreas and kidney (SPK) retransplantation in prior SPK recipients. Clin Transplant 2012: 26: 495. 8. TSIVIAN M, NERI F, NARDO B et al. Aortoiliac surgery concomitant with kidney transplantation: a single center experience. Clin Transplant 2009: 23: 164. 9. PITTALUGA P, HASSEN-KHODJA R, CASSUTO-VIGUIER E et al. Aortoiliac reconstruction and kidney transplantation: a multicenter study. Ann Vasc Surg 1998: 12: 529. 10. LAMATTINA J, SOLLINGER H, BECKER Y, MEZRICH J, PIRSCH J, ODORICO J. Long-term pancreatic allograft survival after renal retransplantation in prior simultaneous pancreaskidney recipients. Am J Transplant 2012: 12: 937.
