Ex vivo renal artery reconstructions: Indications and techniques

ORIGINAL ARTICLES

Ex vivo renal artery reconstructions: Indications and techniques Richard H . Dean, M.D., Patrick W. Meacham, M.D., and Fred A. Weaver, M.D.,

Nashville, Tenn. Ex vivo renal artery surgery has been reported by several investigators and has extended the role of revascularization in the treatment of lesions previously managed by nephrectomy alone. Several techniques are available for use, and selection of the most appropriate method can be tailored to the specific anatomy being managed. Our total experience included 27 kidneys that have been managed by ex vivo renal artery reconstruction. Lesions managed in this manner include two kidneys with renal artery stenosis and renal tumors, one kidney with a congenital branch arteriovenous malformation, and 24 kidneys with branch occlusive or aneurysmal disease from fibromuscular dysplasia. Postoperative angiography was performed in 22 cases and defined successful revascularization without technical error in 20 cases. One operative death occurred as a result of myocardial infarction. One patient required reoperation to control bleeding, and two patients had temporary acute tubular necrosis during the postoperative period. Techniques employed included ex vivo repair with autotransplantation to the iliac system (six kidneys), mobilization and perfusion without transection of the renal vein (10 kidneys), and ex vivo perfusion and repair with replacement into the original renal fossa (11 kidneys). The authors believe this latter technique of reconstruction to be preferable to autotransplantation for the usual patient undergoing ex vivo repair of complex renovascular lesions. (J VAsc SURG 1986; 4:546-52.)

Since the case report by Ota et al. I of the first clinical application of ex situ reconstruction in the treatment of complex renal artery occlusions in 1967, numerous authors have reported similar successes with techniques that allow renal parenchymal protection during prolonged renovascular reconstructive procedures. Through the use of these techniques, lesions that were previously amenable only to nephrectomy have been corrected and the kidneys saved. However, during the past 19 years, several modifications and improvements in the techniques available have provided a heterogeneous selection of methods for management of such lesions. Clinical experience with renal parenchymal preservation techniques and ex vivo reconstruction in the From the Division of Vascular Surgery, Vanderbilt University Medical Center. Presented at the Tenth AnnualMeeting of the Southern Association fbr Vascular Surgery,Cerromar Beach, Puerto Rico, Jan. 30-Feb. 1, 1986. Supported in part by National Institutes of Health Grant No. 14192-16, SpecializedCenter of Research in Hypertension, Vandberbilt UniversityMedicalCenter. Reprint requests: Richard H. Dean, M.D., Division of Vascular Surgery,T-2104 MedicalCenterNorth, VanderbiltUniversiw, Nashville, TN 37232. 546

treatment of renovascular lesions began in our institution in 1972. 2 Within a cumulative experience of more than 700 renovascular procedures during the past 25 years, we have used a variation of ex vivo reconstruction in 27 instances. This article summarizes that experience and reviews the indications and methods of the respective techniques used in our center for management of these more complex renovascular lesions. PATIENTS Ex vivo renal artery reconstruction of 27 kidneys has been performed in our center since 1972. There were 11 men and 16 women, ranging in age from 21 to 66 years (mean, 47 years). Three patients underwent operation on a solitary kidney. Renal function was significantly reduced preoperatively in three patients with serum creatinine concentrations of 2.9, 3.3, and 3.6 mg/dl, respectively. Pathologic features necessitating ex vivo management (Table I) included fibromuscular dysplasia with aneurysms involving renal artery branches (11 kidneys), multiple branch renal artery stenoses resulting from fibromuscular dysplasia (seven kidneys), fibromuscular dysplasia with renal artery dissection

Volume4 Number6 1)eccmbcr1986

Table II. Operative procedures

Table I. Lesions requiring ex vivo management No. of kidneys

Fibromuscular dysplasia with: Branch stenoses and aneurysms Dissection into branches Multiple branch stenoses alone Congenital arteriovenous fistula Renal cell carcinoma and renal artery,stenosis

547

Ex vivo renal arte~ reconstructions

11 5 7 1 2

and branch occlusion (five kidneys), congenital arteriovenous fistula of renal artery branches (one kidney), renal cell carcinoma with renal artery stenosis (two kidneys), and aneurysmal degeneration of a saphenous vein aortorenal bypass (one kidney). OPERATIVE MANAGEMENT Renal artery reconstruction was achieved by use of autogenous saphenous vein in 25 instances and with autogenous hypogastric artery in two cases. Partial nephrectomy was combined with renal artery reconstruction in four cases to manage an intrarenal component of a congenital arteriovenous fistula, segmental renal artery thrombosis with an infarcted lower pole, and in both cases of renal cell carcinoma. One patient underwent simultaneous replacement of an abdominal aortic aneurysm with an aorto-bilateral femoral graft. Techniques of renal preservation and ex vivo reconstruction used in these patients are summarized in Table II. The single operative death occurred intraoperatively as a result of a massive myocardial infarction during the combined ex vivo renal artery reconstruction and replacement of abdominal aortic aneurysm. Early postoperative complications included temporary acute tubular necrosis in two patients and retroperitoneal bleeding that required reoperation in one patient. Postoperative angiography was performed in 22 instances. These studies demonstrated successful renal artery reconstruction in all but two cases. One patient had a thrombosed reattached renal artery branch, and one patient had a stenosis of a branch-saphenous vein anastomosis. Follow-up angiography at 1 year identified one instance of vein graft stenosis resulting from subendothelial fibroblastic proliferation that required repeat bypass and one recurrent branch stenosis that required nephrectomy. Blood pressure response to operation was categorized in survivors according to the criteria previously reported, s Ten patients were cured of hypertension, 13 patients had significant improvement in

Operative procedures

Mobilization without renal vein transection Branch repair with saphcnous vein graft Ex vivo repair with autotransplantation Branch repair with saphenous vcin graft Branch repair, partial nephrectomy with saphenous vein graft Branch repair with hypogastric arteq, graft Ex vivo repair with replacement to renal fossa Branch repair with saphenous vein graft Branch repair, partial ncphrectomy with saphenous vein graft Branch repair with saphenous vein graft and replacement of abdominal aorta aneurysm

No. of kidneys

i0 1 3 2

9 1 1~

~Operative death. severity of hypertension, and one patient had no beneficial blood pressure response. TECHNIQUES OF RENAL FUNCTION PROTECTION Renal preservation solutions Most extensive laboratory studies evaluating perfusates for use in protection of renal parenchymal function during periods of total renal ischemia were begun to facilitate homologous renal transplantation. Several perfusates have been studied by Sacks, Petritsch, and Kaufman, 4 Collins et al.,5,6 and others. 7-9 Through these studies, solutions approximating intracellular electrolyte concentrations have been shown to be superior to solutions such as Ringer's lactate or other solutions that mimic extracellular fluid concentrations. Furthermore, those perfusates that are hyperosmolar, such as Collin's solution, are also superior because of their ability to minimize the cellular swelling during and after the period of perfusion. For these reasons, we employ a modified Collin's solution, the composition of which is listed in Table III. Methods of renal perfusion vary among centers in which ex vivo renal artery reconstructions are performed. Complex perfusion pump systems have been developed by Belzer, Ashby, and Dunphy 1° and others H to allow continuous perfusion during the period of total renal ischemia. Although continuous perfusion may have comparative superiority for prolonged renal preservation during storage periods, simple intermittent flushing with the chilled preservative solution provides equal protection during the shorter periods (2 to 3 hours) required for ex vivo

548

Journal of" VASCULAR SURGERY

Dean, Meacham, and Weaver

Table III. Electrolyte solution Composition (gm/L) Component K2HPO4 KH2PO~ KCI NaHCOa

Ionic concentration (mEq/L)

Amount

Electrolyte

Concentration

7.4 2.04 1.12 0.84

Potassium Sodium Phosphate (HPO4-) Phosphate (H2PO4-) Chloride Bicarbonate

115 10 85 15

Additives at time of use to 930 ml of solution 50% dextrose 70 rnl Sodium heparin 2000 units

15 10

NOTE: Solution (electrolyte solution for kidney preservation) supplied by Travenol Labs, Inc., Deerfield, I11.

dissection and complex renal artery reconstructions. This procedure is performed in our center by refrigerating the Collin's solution overnight. Immediately before use, the additional components are added and the liter of chilled (5 ° to 10 ° C) solution is hung on an intravenous stand to provide at least 2 m gravitational perfusion pressure. Five hundred milliliters of solution is flushed through the kidney immediately after its removal from the renal fossa. At the completion of each anastomosis during ex vivo repair, an additional 150 to 200 ml of solution is flushed through the kidney, which also provides an examination of the suture line for leaks.

Hypothermia preservation Potentially of greatest importance in renal function preservation techniques is the maintenance of profound renal parenchymal hypothermia during the period of total ischemia. Canine studies have shown that warm ischemia times in excess of 30 minutes are associated with increasing degrees of loss ofexcreto~ function. ~2Furthermore, more than 1 hour of warm ischemia is associated with increasing amounts of permanent loss of function. For these reasons, ex vivo perfusion of the kidney is performed with chilled preservative solution. Although lowering the renal core temperature dramatically slows energy-dependent metabolism of the cortical cells, it also inactivates the cellular membrane sodium-potassium pump with subsequent cellular swelling occurring as a result of salt and water entrance into the cell. ~3 However, the use of flushing solutions with intracellular electrolyte composition minimizes this latter adverse effect of hypothermia. Since intermittent perfusion of the chilled preservative is used, surface hypothermia provides improved maintenance of constant hypothermia during ex vivo renal artery reconstruction. Many techniques for producing surface hypothermia have been described. ~,as Our method consists of placing 2 liter bottles of normal saline solution in an ice

slush overnight. The removed kidney is placed in a watertight plastic sheet from which excess saline solution can be suctioned away. Laparotomy pads are placed over the kidney and kept cool and moist by a constant drip of the chilled saline solution. By use of intermittent perfusion with chilled preservative solution and application of continuous surface hypothermia, renal core temperatures of 10 ° to 15 ° C are maintained throughout the period of ischemia. Operative techniques for ex vivo reconstruction General aspects of patient management and operative techniques are outlined elsewhere.16 In all instances, we use ×2.5 to × 3.5 magnification and 6-0, 7-0, and even 8-0 monofilament polypropylene suture material is used for the anastomoses. Three techniques for management of the kidney during ex vivo reconstruction have been used in our center; these include (1) total renal mobilization without renal vein transection, (2) ex vivo repair with autotransplantation, and (3) ex vivo repair with replacement back to the renal fossa. A midline xiphoid to pubic incision is used for most renovascular procedures and is preferred when autotransplantation of the reconstructed kidney or combined aortic reconstruction is to be performed. An extended flank incision parallel to the lower rib margin and carried to the posterior axillary line (Fig. 1) currently is used when complex branch renal artery repairs alone arc required. This is our preferred approach when ex vivo reconstruction without autotransplantation is performed. In all cases of ex vivo reconstructions the ureter is mobilized but left intact. However, an elastic sling or noncrushing clamp is placed around the ureter to prevent continued collateral perfusion and inadvertent rewarming as well as continued blood loss through this route during the cx vivo repair. Complete mobilization without renal vein transection. This technique has been used in 10 patients to facilitate renal artc~ reconstruction in our

Volume 4 Number 6 December 1986

Ex vivo renal artery reconstructions 549

RENAL STUMP

/c OLD

COLD FLUSH

•

,~

I

"~..~..

/~

RENALARTERY

URETER Fig. 1. Demonstration of patient positioning and site of incision for flank approach to right kidney and renal vasculamre.

Fig. 2. Demonstration of use of renal vein occlusion and proximal cannulation. Note that kidney is completely mobilized for ex vivo reconstruction and is connected only by occluded renal vein and ureter.

center. This technique is seldom used at this time, fbr the authors believe that most instances requiring perfusion and ex vivo repair are best managed by bringing the kidney onto the abdominal wall. However, occasionally this technique will provide the necessary increased exposure for multiple branch repairs and allow the same capacity for surface cooling and perfusion with chilled preservative solution without the necessity for reattachment of the renal vein. The flank incision most easily facilitates this exposure and technique. Total mobilization of the kidney is accomplished after the extrarenal portion of the renal artery and the renal vein have been completely mobilized from surrounding tissue. When the kidney is mobilized, Gerota's capsule is entered medially and totally lifted from around the kidney, leaving a pocket for subsequent replacement of the kidney. After aortic mobilization and graft harvesting 'were completed, the mobilized ureter is looped by a :silicone rubber (Silastic) sling to occlude inflow to the kidney. Then, the kidney is covered with moistened laparotomy pads and placed in the plastic envelope within the wound yet elevated from the renal fossa. After the renal artery is transected, the renal vein is cross-clamped at its junction with the vena cava and cannulated on the renal side of the clamp with a 14-gauge catheter (Fig. 2). The kidney is then flushed with 500 ml of the preservative solution with the effluent escaping the renal vein via the cannula.

Similarly, as with other ex vivo techniques, surface hypothermia is maintained by a constant drip of chilled saline solution onto the surface 0fthe kidney. After all renal artery branches are attached to the graft, the kidney is replaced into the renal fossa and the aortic attachment of the graft is performed in standard fashion. At the moment of reperfusion of the kidney, the renal vein clamp and renal vein cannula are removed and the renal vein puncture site is closed with sutures. Finally, after suture line and retroperitoneal hemostasis is assured, Gerota's capsule is reapproximated to stabilize the kidney in its original position. Ex vivo reconstruction with autotransplantation. Autotransplantation of the reconstructed renal vasculature has been used in six cases in our center. Although initially used as the preferred technique, we now believe that autotransplantation should be limited to cases in which replacement of the kidney to the renal fossa is contraindicated. With this bias, we now limit autotransplantation to patients in whom previous dissection of the juxtarenal aorta has led to intense scarring of the renal vein, when excision of the renal vein at its junction with the vena cava is technically ill-advised, and when reanastomosis will leave a foreshortened vein. The performance of this technique is most simply performed through a midline xiphoid to pubis incision. However, the kidney and renal vasculature are

Journal of VASCUI,AR SURGERY

550 Dean, Meacham, and Weaver

COLD SOLUTION

[VC RENAL VEIN

t :IC

URETER

Fig. 3. Drawing of autotransplanted kidney after ex vivo renal artery reconstruction. Renal vein anastomosis is performed first. Note that in this example the hypogastric artery has been mobilized and divided at branch level for attachment to renal artery branches.

exposed from a lateral approach and the ascending or descending colon is mobilized medially. After complete mobilization as described in the preceding section, the renal artery and vein are transected and the kidney transferred to the abdominal wall. Mobilization of the ureter to a point below the pelvic brim facilitates transfer of the kidney out of the abdomen without the necessity for ureteral transection. However, again an elastic sling or noncrushing clamp is placed across the ureter to occlude ureteral collateral vessels to the kidney. With the kidney placed into the plastic envelope, perfused and cooled as previously described, the renal artery reconstruction is performed. When autotransplantation is planned, the ipsilateral hypogastric artery is exposed and examined. When this vessel is normal and matches the renal artery in size, no graft material is required. In this instance, attaching renal artery branches to the distally transected hypogastric artery branches facilitates revascularization without the need for a proximal graft anastomosis. In this technique, the spatulated renal vein is anastomosed to the iliac vein before renal artery or graft anastomosis to the hypogastric or common iliac artery (Fig. 3).

Fig. 4. Demonstration of placement of partially occluding clamp on the vena cava and excision of ellipse ofvena cava wall with renal vein origin; this allows reattachment without concern for renal vein anastomotic stenosis.

Ex vivo reconstruction without autotransplantation. We believe this technique to be the superior method for most ex vivo renal artery reconstructions. Although autotransplantation to the iliac fossa has become the accepted method for reattachment of the ex vivo reconstructed renal artery, this latter technique was initially borrowed from the renal transplant surgeon without thought given to the significant differences in the two patient populations. Reduction in the magnitude of the operative exposure, manual palpation of the transplanted kidney, potential use of irradiation for episodes of rejection, and ease of removal when treatment in the case of rejection has failed are all historic and practical reasons to place the transplanted kidney into the recipient's iliac fossa. None of these advantages applies to the patient with complex renal artery disease that requires ex vivo reconstruction. In contrast, factors most important in ex vivo renal artery reconstruction relate to improving the predictability of permanent success of revascularization. Since many ex vivo procedures are performed in relatively young patients, durability of operation should be measured in terms of decades. For this reason, we believe that attachment of the kidney to the iliac arterial system within or below sites commonly susceptible to significant atherosclerotic occlusive disease can subject the autotransplanted kidney to disease that may reduce the permanency of success of the renovascular surgery. In

Volume 4 Number 6 December 1986

Ex vivo renal arte~ reconstructions 551

GEROTAS F A C t A - -

RENAL VEIN ARTERYGRAFT

Fig. 5. Drawing of replaced kidney after completion of ex vivo repair and repositioning of kidney into renal fossa. Note that Gerota's capsule is reattached to provide stabili~ to replaced kidney.

Fig. 6. Preoperative (A) and postoperative (B) arteriogram in a 25-year-old white woman who had a congenital solitary right kidney with multiple aneurysms of the right renal artery separately involving each of the branches. Ex vivo repair and replacement into the renal fossa were performed. addition, we believe that subsequent management of peripheral vascular disease may be complicated by the presence of the autotransplanted kidney. Furthermore, after the kidney has been replaced into the original renal fossa, the attachment of the renal artery graft to the aorta at a proximal infrarenal site should mimic standard aortorenal bypasses with a high probability of technical success (greater than 97%) and long-term durability (more than 90%). 17

To replace the kidney into its original site, Gerota's capsule is removed from the kidney during mobilization, as described earlier. At the point of renal vein transection, a large vascular clamp is placed to partially occlude the vena cava at the site of renal vein entrance. An ellipse ofvena cava containing the renal vein origin is then excised and the kidney is removed for ex vivo perfusion and reconstruction (Fig. 4). When the microvascular renal artery/graft anasto-

552

Dean, Meacham, and Weaver

moses are completed, the kidney is replaced into its original bed and the ellipse ofvena cava is reattached (Fig. 5). This technique ensures that there is no chance for renal vein anastomotic stenosis from technical error. After this, the renal artery graft is attached to the aorta in the standard manner. Fig. 6 demonstrates the angiographic appearance of the pre- and postoperative anatomy of a 25-year-old white woman with multiple renal artery aneurysms in a solitary right kidney that was managed in this manner.

DISCUSSION Ex vivo renal artery reconstruction with the attendant capacity to allow extended periods of safe ischemic time for prolonged microvascular reconstructions has expanded the role of renal revascularization and salvage in the treatment of renovascular disease and renovascular hypertension. In a recent review of their experience, Dubernard et al.~8 even found that autotransplantation was associated with superior results when compared with in situ aortorenal bypass. However, this uncontrolled, sequential, and highly selective comparison fails to point out that similar improvements in the success of aortorenal bypass have occurred during the past two decades. The rate of technical failures was as high as 22% in our early experience of the 1960s; however, more recently standard in situ aortorenal bypass has been associated with a 98% technical success rate. Therefore, we do not believe that ex vivo reconstruction and/or autotransplantation is a substitute for aortorenal bypass but instead is an addition to our surgical armamentarium to be used in highly selected cases. Through improvements in in situ techniques for management of most branch renal artery lesions, ex vivo reconstruction is seldom required and its utilization rate is less than 4% in our center. Nevertheless, it is a valuable adjunct for use in the management of more complex reconstructive procedures when prolonged renal ischemia is anticipated (in excess of 45 minutes) or when dissection and exposure for small branch repairs are impaired when attempted in the in situ position. Under these circumstances, the use of surface hypothermia and renal preservative solutions for cooling and protection of renal tubular cell function provides the capacity for precision in branch renovascular reconstruction. Finally, since ex vivo reconstruction is usually performed for these advantages alone, we do not believe autotransplantation should be a standard component to the op-

Journalof VASCULAR SURGERY

erative technique. Instead, replacement of the kidney to its original position with use of a standard aortic attachment site provides the best probability of longterm success and can be performed without accepting the theoretic disadvantages of autotransplantation. REFERENCES 1. Ota K, Mori S, Awane Y, Ueno A. Ex-situ repair of renal artery for renal vascular hypertension. Arch Surg 1967; 94:370-3. 2. Orcutt TE. Bilateral ex vivo renal artery, reconstruction with autotransplantation. Rev Surg 1973; 30:374-6. 3. Dean RH, Krueger TC, Whiteneck JM, Dupont B, Foster JH, Smith BM, Hollifield JW, Oates JA. Operative management of renovaseular hypertension. Results after a follow-up of fifteen to twenty-three years. J Vase SURG 1984; 1: 234-42. 4. Sachs SA, Petritsch PH, Kaufman JJ. Canine kidney perfusion using a new perfusate. Lancet 1973; 1:1024-8. 5. Collins GM, Bravo-Sugarman M, Terasaki PI. Kidney preservation for transportation. Lancet 1969; 2:1219-22. 6. Collins GM, Halsz NA. Forty-eight hour ice storage of kidneys: Importance of cation content. Surgery 1976; 79: 432-5. 7. Feduska NJ, Collins GM, Amend WI, Vincenti F, Duca RM, Stieper KW, Mitchell JW, Cochrum KC, Salvatierra O. Comparative study of albumin solution and cryoprecipitatcd plasma for renal preservation: A preliminary report. Transplant Proc 1979; 11:472-7. 8. Marshall VF, Whitsell J, McGovern JH, Miscall BG. The practicality of renal autotransplantation in humans. JAMA 1966; 196:1154-6. 9. Clunie GJA, Murphy KJ, Lubin L, Nicoll P, Marsden RTH. Autotransplantation of the kidney in the treatment of renovascular hypertension. Surgery 1971; 69:326-31. 10. Belzer FO, Ashby BS, Dunphy JE. 24-hour and 72-hour preservation of canine kidneys. Lancet 1967; 2:536-9. 11. Milsten R, Neifield J, Koontz WW. Extracorporeal renal surgery. J Urol 1974; 112:425-7. 12. Ward JP. Determination of the optimum temperature tbr regional renal hypothermia during temporary renal ischemia. Br J Urol 1975; 47:17-24. 13. Cort JH, Kleinzeller A. The effect of temperature on the transport of sodium and potassium by kidney cortex sliccs. J Physiol 1958; 142:208-18. 14. Gibbons RP, Correa RJ, Cummings KB, Mason IT. Surgical management of renal lesions using in situ hypothermia and ischemia. J Urol 1976; 115:12-7. 15. Metzner PI, Boyce WH. Simplified renal hypothermia: An adjunct to conservative renal surgery,. Br J Urol 1972; 44: 76-85. 16. Dean RI-I, Foster JH. Surgery of the renal artery. In: Haimovici H, ed. Vascular surgery: Principles and techniques. Norwalk, Conn: Appleton-Century-Crofts, 1984:827-40. 17. Dean RH. Renovascular hypertension. In: Moore WS, cd. Vascular surgery: A comprehensive review. New York: Grune & Strarton, Inc, 1984. 18. Dubernard JM, Martin X, Gelet A, Mongin D, Canton F, Tabib A. Renal autotransplantation versus bypass techniques for renovascular hypertension. Surgery 1985; 97:529-34.