Renal transplantation with venous drainage through the superior mesenteric vein in cases of thrombosis of the inferior vena cava

Abstract

Renal transplantion usually is performed by placing the graft in the iliac fossa, anastomosing the renal vein to the iliac vein or, when this is not possible, to the vena cava. When vascular complications occur, particularly on the venous side, the position of the graft may have to be changed. This report describes orthotopic renal grafts and positioning of the organ with anastomosis to the splenic vessels. Venous drainage was established directly into the mesenteric-portal territory, with two cases to the portal vein and one to the inferior mesenteric vein. A new technique for the venous drainage of the renal graft is shown. We have used this model in two cases of infrarenal inferior vena cava thrombosis. The kidney was located in a retroperitoneal position, with venous drainage to the superior mesenteric vein through an orifice in the posterior peritoneum.

INTRODUCTION

Renal transplantation is now a frequently used technique, worldwide. It is common that a patient may receive two or more grafts due to failure of a previous graft. When this occurs, venous drainage of the graft through the iliac veins may not be possible due to ileocaval thrombosis. Less frequently, this thrombosis is found during placement of the first graft. We present two patients with thrombosis of the infrahepatic inferior vena cava in which the superior mesenteric vein was used for venous drainage; no description of this technique in renal transplant literature has been found.

CASE REPORT

Patient One

Our first patient was a 24-year-old woman with tuberous sclerosis and terminal chronic renal failure secondary to polycystic kidneys who had been on peritoneal dialysis for 3 years. She had received two previous renal grafts. The first organ had been located in the right iliac fossa, anastomosing the renal vessels to the common iliac artery and the external iliac vein. Ureteric implantation was carried out using the technique described by Woodruff. The patient lost the graft caused by an arteriovenous thrombosis and rupture of the renal capsule with emphysematous pyelonephritis. Six months later, after an episode of peritonitis leading to relative immobility, the patient presented with bilateral deep venous thrombosis, with thrombosis of the superficial femoral system at the level of the proximal great saphenous vein. Anticoagulation with oral anticoagulants (acenocoumarol) was required for 3 months. The patient received a second graft 1 year after the first. The graft developed another venous thrombosis requiring removal of the graft; 18 months after the second graft, the patient received a third organ from a 47-year-old cadaveric donor.

During this third operation we observed that the entire ileocaval axis was collapsed and fibrotic, and there was a well-developed retroperitoneal collateral circulation although it was of fine caliber. The infrarenal vena cava (IVC) was fibrotic up to its retrohepatic segment. The McBurney incision was extended and the peritoneum was opened to dissect the superior mesenteric vein (SMV). Tightening the transverse colon, from the mesenteric root through the pelvis, we acceded to the SMV in the right edge of the mesenteric root—this dissection should not be carried out too close to the transverse mesocolon because, if an accidental lesion occurs, a greater distance from the portal vein eases repair. We dissected the SMV 2 cm from the transverse mesocolon, leaving a sufficient length. The first venous jejunal branch is usually included in the dissection and clamped with the SMV if necessary for performing the anastomosis. An extensive mobilization of the right colon is necessary to place the graft in a retroperitoneal position, and the renal vein was passed through an orifice in the posterior peritoneum to reach the peritoneal cavity near the SMV. An end-to-side anastomosis was made of the renal vein onto the SMV with a 6-0 nylon running suture (the procedure took 15 minutes) (Fig. 1).

Figure 1: Renal transplantation with venous drainage to the SMV. The kidney is placed in a retroperitoneal position, and the renal vein is passed through an orifice in the posterior peritoneum. (Courtesy of Dr. I. Rey Simó, Juan Canalejo Medical Centre, Las Jubias, Spain.)

The renal artery was anastomosed to the right common iliac artery with a 5-0 nylon running suture (performed in 14 minutes). Urinary reconstruction was achieved using the Woodruff technique with a double J stent. The right colon was then positioned over the graft and sutured to the right parietocolic. There was no problem with the space, although the native kidneys were polycystic. The graft did not function initially, and the isotope and ultrasonographic studies were compatible with tubular necrosis. After the second week, progressive functional improvement was observed and it was possible to withdraw hemodialysis. Hematological screening showed a partial deficit of protein S, and treatment was initiated with oral dicoumarins. Ten months after the transplant, the creatinine levels are 0.9 mg/dL.

Patient Two

Our second patient was 68 years old and had undergone right nephrectomy for clear cell carcinoma 13 years previously, and showed no evidence of metastasis to date. In the immediate postoperative period, the patient developed IVC thrombosis, which affected the left renal vein, leading to acute renal failure and required hemodialysis. Renal function recovered and stabilized at a functional level of 45% for 10 years, after which there was a progressive deterioration, and the patient started permanent hemodialysis.

In the pretransplantation study, ileocavography was performed in which obstruction of both iliac veins was observed, associated with the development of collateral circulation through the lumbar veins; the cava was not permeable at any level. The patient was included on the waiting list and received an organ from a 41-year-old cadaveric donor. A xiphoid-to-pubic midline laparotomy approach was used. The left kidney was implanted in the right iliac fossa, in a retroperitoneal position, after mobilization of the right colon. The renal vein was anastomosed to the SMV by an end-to-side anastomosis, which took 16 minutes. The arterial anastomosis required the interposition of an 8-cm segment of donor common carotid artery between the recipient’s external iliac and renal arteries, because of the short length of the donor artery. Both arterial anastomoses were performed with a 5-0 nylon running suture (14 and 15 minutes, respectively). A ureteroneocystostomy was performed according to the Woodruff technique, with no double J stent. The cold ischemia time was 25 hours. Initially, the graft did not function, and a biopsy on day 24 showed acute tubular necrosis. During the following days, there was a progressive functional improvement and an increase in diuresis. The patient’s present serum creatinine level is 1.7 mg/dL, 8 months after transplant.

The iliac vessels are used for anastomosis in most renal transplants. However, these vessels are sometimes unusable because of severe arteriosclerosis, previous renal transplant, thrombosis, fibrosis, or more rarely, anatomic abnormalities. When thrombosis of the IVC is found, with permeable iliac veins that drain through a collateral circulation, it is important to measure the pressure in these veins. If this pressure exceeds 25 mm Hg, either early graft failure or progressive chronic damage will occur from raised drainage pressure. In these cases new strategies must be considered for graft insertion. When infrarenal IVC thrombosis is found, the infrahepatic segment may be used as long as the remnants of the recipient’s renal vessels, as well as the gonadal and adrenal vessels, are patent to maintain a good flow; this flow should be confirmed by intraoperative venogram with pressure measurement (^1,2). When the IVC cannot be used, or is of a high pressure owing to poor collateral circulation, it is necessary to drain the graft to the portal territory. Few cases have been described: Gil-Vernet (³) was the first to perform an anastomosis of the renal vessels to the splenic vessels. These were end-to-end anastomoses, and the short gastric vessels maintained the viability of the spleen. Urinary drainage was achieved by a pyelopyelostomy. Shapira (⁴) performed renal transplant in a 6-year-old child in whom the vena cava had been resected together with the kidney, owing to Wilms’ tumor; the contralateral kidney had been lost. The renal vein was anastomosed to the splenic vein and the renal artery to the aorta. Rosenthal (⁵) described using the IMV to drain the graft. Because of dense adhesions in the area of the portal and splenic veins, it was not possible to use these. Venous revascularization was achieved by an end-to-side anastomosis of the renal vein to the interior mesenteric vein (IMV). The renal artery was anastomosed directly onto the aorta. The ureter was implanted using the Leadbetter-Politano technique. Wolfe (⁶) used the portal vein to achieve venous drainage in a 1-year-old child, using an end-to-side anastomosis of the renal vein to the portal vein. The artery was anastomosed directly onto the aorta. The ureter was sufficiently long enough to be implanted into the bladder.

When problems exist in the IVC territory that impede the use of this vein, the renal graft must be drained to the portal territory. This could have a beneficial immunological effect, as it has been postulated that the first pass of venous blood from the graft through the liver could inactivate the circulating antigens from the transplant. In studies using rats, it has been demonstrated that there is a degree of improvement in renal, cardiac, or intestinal grafts drained by way the portal system (^7,8). However, no data are available to confirm this possible advantage in humans, although there are a few cases, and most of these concern the pancreas.

With respect to the orthotopic positioning of the kidneys, as described by Gil-Vernet and Talbot-Wright (⁹), this is carried out using an extraperitoneal lumbar approach. In this way, direct access may be gained to the splenic vessels and aorta, or even the recipient’s renal vein may be used. This type of approach requires considerable experience. The dissection must be carried out by surgeons who are experts in surgery of the pancreas and portal hypertension due to the risk of lesion to the pancreas or spleen.

The use of the portal vein for drainage of the transplanted kidney is technically difficult because it requires dissection of the hepatic hilum and poses a risk of damage to the biliary tract or hepatic artery. If there is a right hepatic branch of the superior mesenteric artery, it usually runs on the right side of the portal vein, behind the biliary tract. Access to the portal vein is not easy and there is a risk of damaging vital structures.

Anastomosis of the renal vein to the IMV is technically simpler owing to the anatomic situation of this vein, which is accessible in the peritoneal cavity and with a relatively long course. However, the small caliber of this vein may compromise the safety of this type of anastomosis. Furthermore, because of its accessibility and long course, there is a higher risk of rotation, but despite this, there is no doubt that this may be a good option in certain cases.

We have studied the technique described by the Toronto group for pancreatic transplant (¹⁰). This consists of the intraperitoneal localization of the grafted organ, anastomosing the portal vein of the graft to the recipient’s SMV in an end-to-side anastomosis. The splenic and superior mesenteric arteries of the graft are joined in an arterial “Y” graft and this is extended, if necessary, with a segment of the donor innominate artery to anastomose with the common iliac artery. Using this technique the retroperitoneal positioning of the kidney with venous drainage to the SMV is described. We believe that this is the ideal vein for the drainage of the renal transplant when the vena cava cannot be used. The approach is at the root of the mesentery. Easy access is achieved to this vein with minimal dissection; this dissection should not be carried out too close to the transverse mesocolon because, if an accidental lesion occurs, a greater distance from the portal vein eases repair.

The SMV is sufficiently long to carry out the suture and it is of a larger diameter than the IMV, reducing the risk of torsion or thrombosis. The graft may locate in a retroperitoneal position, behind the ascending colon, which has to be mobilized previously. The renal vein is passed through a small orifice in the posterior peritoneum to reach the SMV. Depending on their characteristics the renal artery may be anastomosed to the common or external iliac artery. If there is an insufficient length of artery, direct suture on to the aorta may be carried out, or extension of the artery, using an arterial graft from the donor such as the innominate artery. This positioning of the graft prevents rotation. The ureter has not caused problems in either of our patients because the length of the ureter normally obtained at extraction was sufficient. In any case, if it is not possible to reach the bladder, the recipient’s urinary tract may be used by a ureteroureterostomy.

It is important to attempt to diagnose possible anomalies or thrombosis of the ileocaval axis before surgery because it is preferable to consider the possible solutions before undertaking the operation. Venograms and the measurement of pressures in the different territories are of fundamental importance. In the cases of absence or thrombosis of the IVC, the collateral circulation must be assessed and possible high pressure must be determined, because this would contraindicate venous drainage of the graft to this territory. If the cava is not useable, there is no other option than use of the portal territory.