Renal and retroperitoneal abscesses are uncommon clinical entities that often pose a significant diagnostic challenge. Nonspecific signs and symptoms frequently lead to a delay in diagnosis and treatment. Consequently, they are associated with significant morbidity, and mortality rates approaching 50% have been reported. An understanding of the anatomy of the retroperitoneal space is essential for classification, diagnosis, and management of renal and retroperitoneal abscesses.

CLASSIFICATION

The retroperitoneal space is bounded by the posterior parietal peritoneum and transversalis fascia.  . It is divided into the perirenal space and the pararenal space.
The perirenal space surrounds the kidney and is bounded by the renal (Gerota’s) fascia. It contains a lemon-yellow layer of fat, which is thickest posteriorly and laterally. The anterior and posterior leaves of the renal fascia fuse above the adrenal gland, becoming continuous with the diaphragmatic fascia.1 A thinner, more variable layer meets between the adrenal gland and the kidney. Laterally, the fascial layers join to form the lateroconal fascia, which becomes continuous with the posterior parietal peritoneum. Medially, the posterior layer fuses with the psoas muscle fascia, and the anterior layer fuses with the connective tissue surrounding the great vessels and organs of the anterior retroperitoneum (i.e., the pancreas, duodenum, and colon). Because the perirenal space rarely crosses the midline, perirenal abscesses usually remain unilateral.16 Inferiorly, the renal fascial layers do not fuse but, rather, become continuous with the psoas and ureteral coverings.1,11 This opening inferiorly allows spread of perirenal infections to the pararenal space, to the pelvis, to the psoas muscle, and, in some cases, to the contralateral retroperitoneum.
The pararenal space is divided into two compartments: the anterior compartment, which is bounded by the posterior parietal peritoneum and the anterior renal fascia; and the posterior compartment, which is bounded by the posterior renal fascia and transversalis fascia. The pararenal space contains pale adipose tissue, which fills much of the retroperitoneal space. Because the anterior pararenal space extends across the midline, infection arising in one space may become bilateral. The posterior pararenal space does not cross the midline, and infection within it remains unilateral.16
The retrofascial compartment lies posterior to the transversalis fascia. It is important only in development of the rare retrofascial abscess from abscesses of the psoas, iliacus, and quadratus muscles.

PATHOGENESIS

Before the advent of antimicrobial therapy, most renal abscesses occurred as a result of hematogenous spread of gram-positive organisms, usually Staphylococcus aureus. These abscesses, which were called renal carbuncles, may still be seen in intravenous drug users and in patients with dermatologic disorders. They may resolve with aggressive antimicrobial therapy if treated before frank suppuration. Presently, most renal and retroperitoneal abscesses are caused by retrograde ascent of gram-negative bacteria from the bladder. The most common organisms include Escherichia coli, Proteus, Klebsiella, and Pseudomonas. Anaerobes may be isolated in abscesses associated with gastrointestinal and respiratory infections. Abscesses caused by opportunistic organisms such as Candida and Aspergillus may occur in immunosuppressed patients. Other uncommon pathogens include Mycobacterium tuberculosis and Echinococcus (see below).
A renal abscess is generally preceded by pyelonephritis, which progresses to abscess formation in the presence of a virulent uropathogen, a damaged or obstructed urinary tract, or a compromised host. Renal abscesses have a predilection for the cortical medullary region and may drain spontaneously through the renal collecting system. When renal infection is complicated by obstruction, a purulent exudate collects in the renal collecting system. Pyonephrosis refers to infected hydronephrosis with suppurative destruction of the parenchyma of the kidney, with total or near total loss of renal function. The most frequent cause of obstruction is calculous disease. A previous history of urinary tract infection or surgery is also common.
Perirenal abscesses usually occur by erosion of abscesses or pyonephrosis into the perirenal space. Because of gravity, the resulting perirenal suppuration tends to localize dorsolaterally to the lower pole of the kidney. Posterior pararenal abscesses may arise from perirenal abscesses or from anterior pararenal abscesses tracking into the pelvis, where the anterior and posterior pararenal spaces communicate. Occasionally they result from hematogenous spread. Anterior pararenal abscesses are rarely urologic in origin. They arise from infection involving the organs within the anterior pararenal space, namely the ascending and descending colon, appendix, duodenal loop, and the pancreas. Abscesses arising from the gastrointestinal tract usually harbor a mixture of microorganisms, of which E. coli is the most prevalent. Extension of anterior pararenal abscesses into the perirenal space is uncommon.

DIAGNOSIS

The diagnosis of renal and retroperitoneal abscesses requires a high index of suspicion, as they typically present with insidious, nonspecific signs and symptoms. Presenting symptoms may include fever, chills, abdominal or flank pain, irritative voiding symptoms, nausea, vomiting, lethargy, or weight loss. Symptoms have been present for more than 5 days in the majority of patients with renal and retroperitoneal abscesses, compared with 10% of patients with pyelonephritis. Over one-third of patients may be afebrile. The majority of patients diagnosed with renal and retroperitoneal abscesses have underlying, predisposing medical conditions. These include diabetes mellitus, urinary tract calculi, previous urologic surgery, urinary tract obstruction, polycystic kidney disease, and immunosuppression.
A palpable flank or abdominal mass is present in about half of the cases. The mass may be better appreciated by examination of the patient in the knee–chest position. There may also be signs of psoas muscle irritation with flexion of the thigh.
Laboratory tests are helpful but nondiagnostic. Leukocytosis, elevated serum creatinine, and pyuria are common. Blood and urine cultures are frequently negative; when positive, they usually correlate with culture results from the abscess.
Excretory urography may aid in the diagnosis of renal or retroperitoneal abscesses by showing diminished mobility on inspiratory–expiratory films. A renal abscess causes a decrease in function and enlargement of the nephrogram during the acute phase. Retroperitoneal abscesses may cause displacement of the kidneys or ureters by a mass, scoliosis of the spine, and free air or fluid in the retroperitoneal space. Computed tomography (CT) is highly sensitive for the diagnosis of renal and retroperitoneal abscesses. It precisely localizes and assesses the size of an abscess so that the type of intervention and its anatomic approach can be determined. The presence of gas within a lesion is pathognomonic for an abscess. Additional CT findings characteristic of an abscess include a mass with low attenuation, rim enhancement of the abscess wall after contrast, obliteration of tissue planes, and displacement of surrounding structures. Ultrasonography is less sensitive than CT but useful for monitoring response to therapy. Arteriography and radioisotope scanning rarely add significant information.

INDICATIONS FOR SURGERY

Renal and retroperitoneal abscesses are generally lethal if untreated. Therapeutic options include antimicrobial therapy, percutaneous catheter drainage, and open surgical drainage.

ALTERNATIVE THERAPY

Antimicrobial therapy as the sole treatment is an option, yet most abscesses cannot be cured without drainage. Small renal abscesses may resolve, however, if they are treated early with aggressive antimicrobial therapy. Prolonged antimicrobial therapy without drainage is indicated only if favorable clinical response and radiologic confirmation of abscess resolution indicate that the therapy is effective. If antimicrobial therapy is not effective, prompt percutaneous or open surgical drainage of the pus is mandatory. Progression of a renal abscess leads to perinephric abscess or perforation into the collecting system and results in signs and symptoms of urinary tract infection.
Antimicrobial therapy should be instituted after the urine has been Gram-stained and urine and blood cultures have been obtained. Broad-spectrum coverage should be guided by the presumptive diagnosis and the presumed pathogen. An aminoglycoside for gram-negative rods and ampicillin for gram-positive cocci are preferred. Anaerobic coverage with a drug such as clindamycin is warranted when Gram stain reveals a polymicrobial flora or when a gastrointestinal source is suspected. If the abscess may be of staphylococcal origin, a penicillinase-resistant penicillin, such as nafcillin, should be added. Antimicrobial therapy should be reevaluated when the results of culture and sensitivity tests are available. Unfortunately, urine and blood cultures are frequently sterile, and empirical therapy must be modified on the basis of clinical response and changes in imaging studies.

SURGICAL TECHNIQUE

Percutaneous Drainage

Most renal and retroperitoneal abscesses are treated with empirical antimicrobial therapy and immediate percutaneous drainage. When successful, minimally invasive therapy minimizes operative morbidity and allows for preservation of renal tissue. The abscess must be confirmed by CT-guided or ultrasonography-guided needle aspiration and must be drainable without injury to other organs. Immediate surgical drainage must be instituted if the procedure fails. After a multiport drainage catheter (8 to 12 Fr) is positioned, the abscess should be drained, and adequate evacuation should be confirmed by CT or ultrasonography. The catheter should then be connected to low intermittent suction, and drainage outputs should be monitored daily. If drainage stops abruptly, occlusion of the catheter should be suspected, and it should be irrigated gently with small amounts of normal saline. Computed tomography or ultrasonography should be performed periodically to monitor catheter position and size of the abscess. Direct instillation of contrast through the drainage tube may be helpful to confirm the catheter position or to rule out a fistula. To avoid bacteremia, prophylactic antimicrobial coverage should be given, and the contrast should be instilled under gravity or by gentle injection. Instillation of 2,500 units of urokinase in 50 ml of normal saline on a daily basis may be successful in evacuating an organizing infected hematoma. Routine abscess irrigation with antimicrobials is of questionable benefit and may promote overgrowth of resistant bacteria. The catheter should be withdrawn gradually as the abscess cavity shrinks and the drainage decreases. The usual duration of drainage is 1 to 3 weeks. The catheter is removed when drainage stops and CT and ultrasonography show complete resolution.

Open Surgical Drainage

The incision should be smaller than that used for routine nephrectomy, and usually a posterior flank muscle-splitting incision below the 12th rib is sufficient. When the retroperitoneal abscess is entered, the pus should be cultured, and the space gently but thoroughly explored to ensure that all loculated cavities are drained. Thorough irrigation of the cavity is essential. Multiple Penrose drains should be inserted into the space through separate stab wounds, and the ends of the drains should be sutured to the skin and tagged with safety pins. Fascial and muscular closure may be performed with chromic catgut suture, but skin and subcutaneous tissue should be left open to prevent the formation of a secondary body wall abscess. The wound can be left to heal from within, or skin sutures may be placed and left untied for dermal approximation 5 to 7 days postoperatively after drainage has ceased. The wound should be packed with gauze, and the packs should be changed daily. The drains should be left in place until purulent drainage has decreased, and then they can be removed slowly over several days.

ANCILLARY PROCEDURES

If a perinephric abscess is due to long-standing obstruction and there is no functioning renal tissue, a nephrectomy at the time of drainage is theoretically attractive. Drainage of a perinephric abscess should usually be performed as a primary procedure, however, with nephrectomy performed at a later date if necessary. Patients are frequently too ill for prolonged general anesthesia and surgical manipulation. Furthermore, nephrectomy is usually difficult to accomplish technically, and preoperative information is usually not sufficient to determine accurately the amount of functioning of salvageable renal tissue. After drainage of the abscess, removal of obstruction, and appropriate antimicrobial therapy, many kidneys may regain sufficient function to obviate future nephrectomy. Nephrectomy, if indicated, can be performed using a standard nephrectomy approach or a subcapsular nephrectomy technique outlined later.
A small renal abscess confined to one pole of the kidney may be managed by partial nephrectomy. If the infection extends beyond the apparent line of cleavage, however, it is essential to remove all infection, and the line of excision should extend through healthy tissue. If multiple abscesses are present, internal drainage is difficult, and nephrectomy may be required.

Subcapsular Nephrectomy

When a kidney is so adherent to surrounding tissues that dissection is difficult and hazardous, a subcapsular nephrectomy is indicated. These conditions are usually seen after multiple or chronic infections or previous operations have caused scarring to adjacent organs. Blunt dissection results in tearing of structures such as bowel wall. Sharp dissection when there is no definable tissue plane often results in lacerations of the vena cava, aorta, duodenum, spleen, and other structures. In subcapsular nephrectomy, dissection beneath the renal capsule enables one to avoid these vital structures. Subcapsular nephrectomy should not be performed for malignant disease and is undesirable in tuberculosis.
The main difficulty with subcapsular nephrectomy is that the capsule is adherent to the vessels in the hilum, and one usually must go outside the capsule to ligate the renal pedicle. In this setting, the renal hilum usually is involved in the inflammatory reaction, and separate identification of the vessels is difficult.
Kidney exposure is accomplished through the flank using a 12th rib incision. For low-lying kidneys, a subcostal incision may be satisfactory. When the kidney is reached, the capsule is incised and is freed from the underlying cortex. The capsule is stripped from the surface of the kidney, and an incision is made carefully in the capsule where it is attached to the hilum. The vessels may be protected by placing a finger in front of the pedicle when cutting the capsule. The dense apron of capsule can usually be incised best on the anterior aspect. Control of bleeding can be difficult in this procedure. Frequently all landmarks are obscured, and the renal artery and vein cannot be identified. Sharp dissection is usually required, and major vessels may be entered before they are recognized. Fortunately, the dense fibrous tissue tends to prevent their retraction. Frequently, several chromic suture ligatures can be placed through the pedicle between a proximally placed pedicle clamp and the kidney. To avoid damage to the duodenum or major vessels, pieces of capsule may be left behind. However, prolonged drainage can ensue, and as much of the infected tissue should be removed as possible. After ligation and cutting of the pedicle, the ureter is identified and cut, and the distal end is ligated. If distal ureteral obstruction has caused pyonephrosis, a small, 8 to 10 Fr red Robinson catheter may be placed in the distal ureter to allow postoperative antimicrobial irrigation. Multiple drains should be placed and brought through separate stab wounds.

OUTCOMES

Complications

Complications associated with percutaneous drainage include the formation of additional abscesses that communicate with the renal collecting system and may require temporary urinary diversion via percutaneous nephrostomy drainage to affect a cure. Sepsis, the most frequent complication of percutaneous drainage, occurs in fewer than 10% of patients. Other complications, such as transpleural puncture, vascular or enteric injury, and cutaneous fistula, are rare.
Additional complications to open or percutaneous drainage include prolonged purulent drainage, which may indicate a retained foreign body, calculus, or fistula.

Results

Cure rates for percutaneous drainage of renal and retroperitoneal abscesses range from 60% to 90%.8,15 Multiloculated, viscous abscesses and abscesses in immunocompromised hosts are associated with lower cure rates. Large abscesses may require more than one percutaneous access procedure to completely drain them.
In the past, mortality rates were reported to be as high as 50% in patients with retroperitoneal or perinephric abscesses. More recent reports indicate a significant improvement in mortality (approximately 10%), in large part because of more accurate diagnosis from improved imaging techniques, more effective antimicrobial therapy, and better supportive care.

SPECIAL CONSIDERATIONS

Renal Tuberculosis

Renal tuberculosis is caused by hematogenous dissemination from an infected source somewhere else in the body. Both kidneys are seeded with tuberculosis bacilli in 90% of cases. Clinically apparent renal tuberculosis is usually unilateral, however. The initial lesion involves the renal cortex, with multiple small granulomas in the glomeruli and in the juxtaglomerular regions. In untreated patients who fail to heal spontaneously, the lesions may progress slowly and remain asymptomatic for variable periods, usually 10 to 40 years. As the lesions progress, they produce areas of caseous necrosis and parenchymal cavitation. Large tumor-like parenchymal lesions or tuberculomas frequently have fibrous walls and resemble solid mass lesions. Once cavities form, spontaneous healing is rare, and destructive lesions result, with spread of the infection to the renal pelvis and development of a parenchymal or peri-nephric abscess.

Indications for Surgery

Surgery was once commonly used in the treatment of renal tuberculosis, but since the advent of effective antituberculosis chemotherapy, it is reserved primarily for management of local complications, such as ureteral strictures, or for treatment of nonfunctioning kidneys. If surgery is warranted, it is wise to precede the operation with at least 3 weeks and preferably 3 months of triple-drug chemotherapy. Use of isoniazid, 300 mg/day; pyrazinamide, 25 mg/kg to a maximum of 2 g, once daily; and rifampicin, 450 mg/day is recommended. If segmental renal damage is obvious and salvage of the kidney is possible, a drainage procedure or cavernostomy can be performed.7 Removal of a nonfunctioning kidney is usually indicated for advanced unilateral disease complicated by sepsis, hemorrhage, intractable pain, newly developed severe hypertension, suspicion of malignancy, inability to sterilize the urine with drugs alone, abscess formation with development of fistula or inability to have appropriate follow-up.

Alternative Therapy

Prophylactic removal of a nonfunctioning kidney to prevent complications, remove a potential source of viable organisms, and shorten the duration of convalescence and requirement for chemotherapy is advocated by some authors.5,19 Others, who followed a large series of patients treated with medical therapy alone, concluded that, because the frequency of late complications is only 6%, routine nephrectomy should not be performed for every nonfunctioning kidney.9 These authors, however, treated patients for at least 2 years. The merits of short-term therapy and prophylactic nephrectomy versus long-term 2-year chemotherapy and selective nephrectomy warrant further study. Modern percutaneous drainage techniques have largely replaced open cavernostomy for treatment of closed pyocalyx.

Surgical Technique

Cavernostomy

Renal tuberculosis sometimes results in caliceal infundibular scarring, causing a closed pyocalix. Unroofing of a pyocalix is called cavernostomy. If the calix still communicates with the renal pelvis, or if it is connected to significant functioning parenchyma, a cavernostomy should not be done because a urinary fistula or urinoma may result. To minimize wound contamination and tuberculous spread, thorough needle aspiration of purulent material and saline irrigation of the abscess cavity should be performed using a large-bore needle and syringe . The abscess cavity is then unroofed, and the edge is sutured with a running suture for hemostasis. Any unsuspected connection with the renal pelvis by an open infundibulum must be closed using 5-0 chromic catgut suture to prevent fistula or urinoma formation. After thorough wound irrigation, multiple drains are placed, and closure is undertaken. Drains are managed as previously described for perinephric abscess.

Nephrectomy

When unilateral tuberculosis causes more extensive parenchymal destruction or nonfunction, a partial or total nephrectomy, respectively, should be performed. For partial nephrectomy, a guillotine incision is made 1 cm beyond the abscess. If the renal pedicle can be freed and polar vessel located and occluded, the incision can be made at the line of demarcation of the ischemia. In partial nephrectomy, it is important to try to save the capsule (if it is not involved with the infection) to cover the raw surface for hemostasis. Alternatively, fat can be used for hemostasis. The amputated calyx is carefully ligated with a 4-0 chromic catgut suture to prevent urinary fistula or urinoma formation.
After nephrectomy, the distal ureter can be ligated and in most cases does not need to be brought to the skin because tuberculosis of the ureter generally heals with chemotherapy after nephrectomy. If renal tuberculosis is associated with severe tuberculosis cystitis, ureteral catheterization for 7 days postoperatively to minimize subsequent ureteral stump abscess formation should be considered.18

Renal Echinococcosis

Echinococcosis is a parasitic infection caused by the canine tapeworm E. granulosus. Echinococcal or hydatid cysts occur in the kidney in some 3% of patients with this disease. The hydatid cyst gradually develops at a rate of about 1 cm/year and is usually single and located in the cortex.

Diagnosis

The symptoms are those of a slowly growing tumor; most patients are asymptomatic or have a dull flank pain or hematuria. Excretory urography typically shows a thick-walled cystic mass, which is occasionally calcified. Ultrasonography and CT usually show a multicystic or multiloculated mass. Confirmation of the diagnosis is most reliably made by diagnostic tests using partially purified hydatid antigens in a double diffusion test.13 Complement fixation and hemagglutination are less reliable. Diagnostic needle puncture is associated with significant risk of anaphylaxis as a result of leakage of toxic cyst contents.

Indications for Surgery

Cyst removal is indicated when an enlarging cyst threatens renal function or produces obstruction.

Surgical Technique

The cyst should be removed without rupture to reduce the chance of seeding and recurrence. In cases where cyst removal is impossible because of its size or involvement of adjacent organs, marsupialization is required. The contents of the cyst initially should be aspirated, and the cyst should be filled with a scolecidal agent such as 30% sodium chloride, 2% formalin, or 1% iodide for about 5 minutes to kill the germinal portions. Complete evacuation of all hydatid tissue and thorough postmarsupialization irrigation are critical to preventing systemic effects. Penrose drains are left in the cystic cavity until drainage ceases. If large amounts of renal tissue have been damaged, partial or simple nephrectomy may be required.

Renal autotransplantation is a safe and effective procedure to reconstruct the urinary tract. The first successful surgery was performed by Hardy in 1963 in a patient with severe ureteral injury following aortic surgery. The advent of microvascular techniques and renal preservation extends the scope of the procedure, allowing for successful extracorporeal (bench) surgery and subsequent autotransplantation. Current indications of autotransplantation include renal-vascular disease, severe ureteral damage, tumors of the kidney and ureter, complex nephrolithiasis, and retroperitoneal fibrosis. The procedure is technically demanding and is contraindicated in the setting of severe occlusive atherosclerosis of the iliac arteries. The advantages of autotransplantation include optimal surgical exposure, bloodless surgical field, and hypothermic protection of the kidney from ischemia. In cases of malignancy, there is less risk of tumor spillage and better assessment of tumor margins than by in vivo renal reconstruction. It is possible that this procedure may be underutilized because a good proportion of urologists are unfamiliar with the principles of renal homotransplantation.

DIAGNOSIS

Preoperative renal and pelvic arteriography should be performed to define the renal artery anatomy and ensure disease-free iliac vessels. In cases where autotransplantation is performed for the management of ureteral disease, ureteral involvement can be assessed by intravenous or retrograde pyelography. A CT scan of the pelvis may be beneficial in cases of retroperitoneal fibrosis to assess pelvic extension of disease.

INDICATIONS FOR SURGERY

Renal autotransplantation is particularly attractive for a variety of vascular lesions affecting the aorta and renal artery. These include traumatic arterial injuries, renal artery stenosis with extension into the segmental branches (fibromuscular disease), large aneurysms, or arteriovenous fistulas. Other vascular indications include aortic aneurysms involving the renal arteries (Marfan’s syndrome) and occlusive aortic disease.

In patients with central, intrarenal tumors or multiple tumors in a solitary kidney, renal autotransplantation with extracorporeal surgery is a useful technique. Following radical nephrectomy and exterior hypothermic renal perfusion, the kidney is dissected beginning in the hilum. The vasculature to the tumor is ligated. After tumor-free margins are achieved, autotransplantation is carried out.

Renal autotransplantation allows for a direct anastomosis of the renal pelvis to the bladder. Therefore, it can be used in cases of ureteral damage or long ureteral lesions such as iatrogenic ureteral injuries, ureteral strictures, ureteral tumors, ureteral tuberculosis, failed urinary diversions, and retroperitoneal fibrosis. The procedure can also be used to facilitate stone passage in patients with complex nephrolithiasis. Renal autotransplantation has been effective in controlling the symptoms related to loin-pain hematuria syndrome.

ALTERNATIVE THERAPY

Replacement of the ureter for reconstruction of the urinary tract may be performed with a segment of ileum. The advantages of an ileal ureter over autotransplantation are threefold: (a) the procedure is technically less demanding; (b) vascular anastomosis is not necessary; and (c) bladder argumentation with bowel can be done simultaneously. Disadvantages include mucus production, metabolic and electrolyte imbalance, propensity for bacteriuria, and the need for indefinite radiologic surveillance of the ileal segment. Contraindications include intestinal diseases, hepatic dysfunction, and renal insufficiency (serum creatinine greater than 2.0 mg/dl).

For severe renovascular disease, the first surgical options typically include in situ reconstruction. This may involve endarterectomy or aortic–renal or splenorenal bypass grafting. When these techniques are not possible and microvascular reconstruction is required, autotransplantation becomes the procedure of choice.

SURGICAL TECHNIQUE

Perioperatively, a brisk diuresis should be induced by IV hydration and 12.5 g mannitol given 1 hour before surgery. This will minimize ischemic injury to the kidney and hasten restoration of renal function. A broad-spectrum antibiotic is also administered 1 hour before surgery. During the operation, an adequate central venous pressure should be maintained with fluid boluses as needed. Renal autotransplantation is a two-step procedure: first, the kidney is removed; then, it is transplanted. The surgical approach to removing the kidney is similar to that of living-donor nephrectomy. However, the operation may be complicated by the particular disease process necessitating the surgery. Typically, two incisions are needed: the first, either a subcostal transperitoneal or extrapleural–extraperitoneal flank to remove the kidney; and the second, a lower-quadrant curvilinear or midline incision to access the iliac fossa. In thin patients, an alternative approach is a single midline incision from the xiphoid to the symphysis pubis, although the exposure to the kidney is not optimal.

Nephrectomy

After the peritoneum and colon are reflected medially, Gerota’s facia is incised. The perinephric fat is sharply dissected off the renal capsule with minimal spreading using Metzenbaum scissors. Excessive retraction of the kidney should be avoided because that may result in subcapsular hematomas or capsular tears. The adrenal gland should be carefully separated from the upper pole of the kidney. In cases where ureteral continuity is not preserved, the distal ureter is isolated and transected before the dissection of the renal hilum. The ureteral stump can be tied with #0 chromic catgut suture. The ureter should be maintained with its vascularity and the gonadal vein. To maximize ureteral viability, the tissue between the lower pole of the kidney and the ureter should be kept intact. Urinary output can then be assessed before the renal pedicle is approached.

On the right side, the vena cava is carefully isolated from the surrounding tissue. The gonadal vein can be ligated at its insertion on the vena cava. The renal vein should be identified anterior to the renal artery. Accessory renal veins can be ligated, but accessory renal arteries must be maintained. Careful dissection of the renal artery is performed toward the aorta by slightly retracting the vena cava with a closed forceps or vein retractor. After a further 12.5 g of mannitol has been given, a right-angle clamp is placed on the renal artery, and it is transected. A Satinski vascular clamp is then placed proximal to the renal vein ostium, and the renal vein is transected. The renal artery can be tied with #0 silk ties or, alternatively, one #0 silk tie and a #0 silk ligature. We have found that the renal vein retracts after transection and that placement of a second larger Satinski clamp behind the first allows for a less stressful closure of the renal ostium. A 5-0 Proline suture is tied at one end, run down to the other end, back to the first, and retied. On the left side, the artery is divided at the aorta; the vein is divided anterior to the aorta and tied with two #0 silk sutures.

Renal Preservation

Once the kidney is removed, it is immersed in ice-cold slush at 4°C. The renal artery is flushed with either a Collins intracellular electrolyte solution or a lactated renal solution with 10,000 units/liter of heparin and 50 mEq/liter of sodium bicarbonate. Flushing is continued until the effluent from the renal vein is clear. An adaptor is used to hold a good seal during flushing. Adequate flushing allows about 4 to 6 hours of renal preservation. If extracorporeal renal surgery is required, a second team can close the flank incision. In closure of the flank, a strong monofilament absorbable suture such #1 PDS or Maxon is used. Anteriorly, the transversus abdominis, internal oblique, and external oblique muscles are closed separately. Often the transversus and internal oblique are closed together. Closure should begin after flexion is removed from the operating room table. Posteriorly, the intercostal muscles and latissimus dorsi are closed as separate layers.

Autotransplantation

The patient is placed in the supine position and draped. A Foley catheter carries 150 cc of 1% neomycin sulfate solution into the bladder, and the catheter is clamped. A curvilinear incision is made, extending from one fingerbreadth above the pubis to two fingerbreadths above the anterior superior iliac spine. The incision is carried down to the rectus facia. The rectus facia, the external oblique, internal oblique, and transversus abdominis muscles are opened along the line of the incision. The lateral edge of the rectus muscle is transected off the pubis to get better exposure to the pelvis. The epigastric vessels are transected beneath the transversus abdominis muscle and tied with two 2-0 silk ties. The round ligament or the spermatic cord is doubly ligated. In young men, the spermatic cord is preserved and displaced inferomedially. The peritoneum is reflected medially to expose the iliac vessels and bladder. A Buckwalter retractor is placed into the wound to provide optimal exposure.

The next part of the procedure is similar to that used for renal homotransplantation. The iliac vessels are evaluated for potential size for anastomosis with the renal artery. If the caliber of the internal iliac artery is sufficient, and there is not significant plaque formation, then this vessel is selected. It is mobilized from the common iliac to the first branch, the superior gluteal artery. A bulldog vascular clamp is placed just beyond the origin of the internal iliac artery, and a right-angle clamp is placed distally (Fig. 16-2). After transection of the vessel, the distal portion is tied with #0 silk tie. The bulldog vascular clamp is opened to test for flow. The proximal portion of the vessel is flushed with 2,000 units of heparin mixed as follows: 10,000 units per 100 cc normal saline. If a plaque is discovered, it can be trimmed back, an endarterectomy can be performed, or it can be tacked down with 6-0 silk. Further dissection of the common iliac and external iliac artery is not required. The external iliac vein is mobilized for 5 to 7 cm with special care to ligate any lymphatic vessels with 4-0 silk to prevent lymphocele formation.

When the internal iliac artery is unavailable, the external iliac artery is selected. After the external iliac artery is mobilized for 4 to 5 cm, an end-to-side anastomosis is performed between it and the renal artery. Wide mobilization of the external iliac artery may result in kinking of the vessel. Vascular clamps are placed proximally and distally, and an arteriotomy is performed. Typically only a slit is needed, and an ellipse of the anterior wall need not be removed. The distal artery is flushed with 2,000 units of dilute heparin with a red rubber catheter. A Satinski vascular clamp is placed distally on the external iliac vein, and a bulldog is placed proximal to the venotomy. The iliac vein is then carefully incised with a #11 blade scalpel to accommodate the renal vein. Four 5-0 Proline sutures are placed on the external iliac vein in an outside-to-in fashion, one at each apex and one at the midpoint on each side of the venotomy.

The kidney is placed in the operative field. An assistant holds the kidney with a surgical sponge in an anatomic position with the ureter inferiorly. To minimize warm ischemia while the anastomosis is accomplished, the kidney is irrigated with cold saline. The four Proline sutures in the external iliac vein are then brought through the renal vein in an inside-to-out fashion. The kidney is lowered into the wound, and each suture is tied with the knots on the outside. The two apex sutures are used to close the venotomy. The midpoint sutures are placed with mild traction to keep the back wall from being incorporated in the running suture. The internal iliac artery is then anastomosed to the renal artery end to end with 6-0 siliconized silk suture. The artery should be placed posterior to the renal vein to preserve anatomic relationships. The first two sutures of the arterial anastomosis are placed at either apex with double-armed needles such that the knots are on the outside. The remainder are placed with single-armed needles. We prefer an anastomosis with interrupted sutures when the internal iliac artery is used. Sutures should be placed close enough to avoid any gaps, especially at the apex. After one side is complete, the apical sutures are rotated to give exposure to the back wall. If the external iliac artery is used, the arteriotomy should be staggered with the venotomy to avoid kinking of the vessels. The anastomosis is performed with two to four continuous 6-0 Proline sutures. After completion of the anastomosis, oxidized cellulose is wrapped in small pieces around the arteriotomy and venotomy. The venous clamps are then removed, followed by the arterial clamps. It is important to maintain adequate intravascular volume with colloid or blood, especially when the clamps are removed, so that the kidney is well perfused. If this produces an excessively elevated central venous pressure, intravenous furosemide (Lasix) should be administered.

Occasionally, renal autotransplantation can be performed with the ureter left intact. Although it will follow a redundant course to the bladder, normal peristalsis will provide effective drainage from the kidney. Care must be taken to avoid positioning the kidney so as to produce an obstruction of the ureter.

If the ureter is transected, the urinary system can be reconstructed by a ureteroneocystostomy, ureteroureterostomy, pyeloureterostomy, or a pyelovesicostomy. We prefer an extravesical ureteroneocystostomy when there is adequate length of nondiseased ureter for a tension-free anastomosis. The Buckwalter retractor is repositioned to provide better exposure to the lateral wall of the bladder. A 2- to 3-cm tunnel is made in the bladder wall by incising the posterior lateral serosa and detrusor muscle. After the margins of the detrusor are retracted with 3-0 chromic stay sutures, the mucosa is mobilized and allowed to bulge. An ellipse of the mucosa is removed from the apex of the tunnel, and the spatulated ureter is anastomosed with the bladder mucosa using a continuous 50 chromic catgut suture. Two sutures are used, each of which incorporates 180 degrees of the anastomosis, and are tied without tension. The anastomosis is performed over a 4.8 Fr double-J ureteric stent, which is positioned into the bladder after the bladder mucosa has been opened. The stent will be removed in the postoperative period.

The detrusor is closed over the ureter with interrupted 3-0 chromic catgut suture. The tunnel should allow passage of a right-angle clamp between the ureter and overlying muscle. The wound is irrigated with a 1% neomycin solution. No external drains are required if the ureteral reimplantation is watertight.

In cases where the upper ureter is diseased, the area is removed, and the proximal ureter or renal pelvis is anastomosed to the normal lower ureter. The ureteroureterostomy or pyeloureterostomy is performed over a ureteric double-J stent by end-to-end anastomosis of the spatulated lower ureter to either the spatulated upper ureter or the renal pelvis. If the entire ureter is not viable, or for recurrent stone disease, a pyelovesicostomy is performed. This technique can be performed with a Boari flap and end-to-end anastomosis of the renal pelvis to tubularized bladder. The Boari flap should be secured to the psoas muscle to avoid tension on the anastomosis.

The wound is then closed in layers. The rectus muscle is approximated back to the tendinous insertion at the pubic bone with a #0 Proline suture. The internal oblique and transversus abdominis are closed with a #0 Proline suture. The external oblique is closed with continuous #0 Proline. The subcutaneous layer is closed with 3-0 Dexon and the skin is closed with 3-0 nonabsorbable suture or clips.

For optimal renal perfusion during the immediate postoperative period, the central venous pressure should be maintained adequately, and the diastolic blood pressure kept at 85 mm Hg or higher. Mild hypertension is preferred over normotension or mild hypotension. Aspirin can be started postoperatively to reduce the risks of graft thrombosis. A renal scan is obtained on the first postoperative day to document renal perfusion and again about postoperative day 7. Broad-spectrum antibiotics are administered during the immediate postoperative period to maintain sterile urine and help prevent infection of the vascular grafts. The ureteral stent is left in place for 2 to 3 weeks and is removed during outpatient cystoscopy. The Foley catheter is removed on postoperative day 5. It may be removed sooner if a ureteroureterostomy or pyeloureterostomy is performed, but it should be kept in place for 5 days following a pyelovesicostomy or ureteroneocystostomy. An intravenous pyelogram or a cystogram is obtained 1 to 2 weeks after surgery to evaluate ureteral integrity.

OUTCOMES

Complications

Early postoperative complications include bleeding from the vascular anastomosis, renal artery or vein thrombosis, distal extremity embolization, or urinary extravasation. Bleeding from a disrupted anastomosis is a rare event but requires immediate exploration. It is usually associated with anastomosis to diseased vessels or errors in surgical technique. Peripheral collateral vessels from the renal hilum can attain significant size if there is stenosis of the renal artery or vein and can be a source of postoperative bleeding. Renal artery or vein thrombosis occurs in fewer than 2% of cases and should be ruled out in cases of oliguria following autotransplant of a solitary kidney. The diagnosis is made by renal scan; if it is made without delay, salvage of the autotransplant should be attempted. Any significant hypotension or hypovolemic event in the postoperative period or error in surgical technique can predispose to this threat.

Distal extremity embolization as a result of dislodging of plaque during aortic clamping or unclamping can occur, especially with diseased blood vessels. Heparinization at the time the vessels are prepared aids in preventing this problem, but the distal pulses and color of the legs should be assessed after arterial clamps are opened. Deep venous thrombosis can result in propagation of clot from the renal vein. Intimal injury, low-flow states, and venous obstruction can predispose to this condition.

Urinary extravasation is the most common complication from autotransplantation. Placement of a ureteric double-J stent diminishes this risk. If a leak occurs, it should be treated by a percutaneous nephrostomy. In circumstances when these conservative measures fail, such as when the distal ureter is ischemic, operative repair is required.

The most common late complications include renal artery stenosis, ureteral stricture, and ureterovesical reflux. Renal artery stenosis may be manifested by hypertension or impaired renal function. Diagnosis is made by renal scan and digital subtraction angiography. Initial management should be percutaneous angioplasty. Obstruction of the urinary system demonstrated by pain or impaired renal function can be managed by dilation and stenting.

Results

Bodie et al. reported on 24 autotransplanted kidneys in 23 patients in whom the primary indication was to replace all or a major portion of the ureter. There were no operative deaths reported. Of the 24 autografts, three were ultimately lost (12%). The function of the remaining grafts was stable or improved postoperatively.1 Novick reported successful outcomes in 29 of 30 patients who underwent autotransplantation for the management of intrarenal branch arterial lesions.

Van der Valden reported on six cases of renal carcinoma treated by extracorporeal surgery and autotransplantation. Dialysis was not required, and the patients’ blood pressure improved or remained within normal limits. Mean follow-up time was 54 months, with three patients dying during this period. Zincke and Sen performed extracorporeal surgery and autotransplantation in 15 kidneys. Of these, 11 had renal cell carcinoma, and four had transitional cell carcinoma. Three autografts were lost because of venous and arterial thrombosis in two and necrosis of the renal pelvis and ureter in one. The remaining patients were dialysis-free with stable creatinine values. Other complications cited included a caliceal fistula requiring closure in one patient and an intimal injury requiring partial replacement of the external iliac artery with a Gore-Tex graft.10 Novick et al. observed an increased incidence of temporary and permanent renal failure for extracorporeal compared to in situ partial nephrectomy for renal cell carcinoma.8 Postoperative initial nonfunction occurred in five of 14 patients (36%) undergoing autotransplantation but in only two of 86 patients (2.3%) who underwent an in situ procedure. Permanent renal failure occurred in two of 14 (14.3%) autotransplanted patients and in one of 86 managed in situ(1.2%).

Renal autotransplantation is a rare procedure that is technically demanding with several potentially serious complications. However, in a variety of instances, it may be of great utility for organ salvage and should be included in the armamentarium of the urologist.

Historically, the incidence of urologic complications following kidney transplantation, manifested primarily as ureteral leaks or obstruction, was as high as 10%.1,5 The complications often resulted in significant morbidity, graft loss, and occasional patient death. Improvements in surgical techniques, immunosuppression, and methods for diagnosing and treating the complications have led to a significant decline in the rate of urologic complications to the current reported incidence of 2% to 2½%.This has resulted in lower morbidity and rare loss of a kidney or patient to urologic complications. However, despite these changes, the need for diligence in diagnosing these complications and quickly addressing them remains as true today as in the past.

The most common cause for ureteral complications following kidney transplantation is technical error. Damage to the ureteral blood supply during graft harvest or transplantation can result in ureteral ischemia and subsequent leak or obstruction. Additional technical errors such as excessive tension at the ureteroneocystostomy site or hematoma development within the tunnel may also cause problems.With careful attention to detail, most of these problems can be minimized, especially in the early postoperative setting. Long-term or delayed ureteral obstruction may be the result of ischemic changes secondary to chronic rejection or a continuation of the spectrum of damage associated with the organ harvest and transplantation, and although not all are preventable, the incidence can be markedly reduced with good surgical technique.

URINARY LEAKS

In current practice most urinary leaks are the result of ureteral problems, s a majority of surgeons now employ an extravesical ureteroneocystostomy technique for implantation of the ureter. This results in a shorter ureter, decreased likelihood of ischemia, and a limited cystostomy that rarely leads to leakage from the bladder. The majority of leaks occur early after transplantation and are manifested by either drainage from the wound, unexplained graft dysfunction, or a pelvic fluid collection. Signs and symptoms can also include fever, graft tenderness, and lower extremity edema.

Early urinary leaks can be divided into two types according to the timing of presentation. The first usually occurs within the first 1 to 4 days and is almost always related to technical problems with the implantation. In this case, the ureter has usually pulled out of a tunnel. This is likely caused by excessive tension at the anastomosis. This complication appears to be more common with the extravesical ureteroneocystostomies.8 Some investigators have recommended use of a ureteral stent to lessen the likelihood of this complication.

The second type of early ureteral leak is associated with distal ureteral ischemia, which may be a consequence of injury during the donor recovery, technical causes such as tunnel hematoma, or distal stripping of the blood supply. This type usually presents between 5 and 10 days posttransplant.

To correct the early leak caused by excessive tension, it is often possible to do a repeat ureteroneocystostomy. In most other cases, especially with the current techniques of extravesical reimplantation, a different operative procedure is often more suitable.

URETERAL OBSTRUCTION

Ureteral obstruction can also be the result of ureteral ischemia but occurs later than ureteral leaks and usually presents as graft dysfunction. It can occur years after the transplant and in this situation may represent vascular injury associated not only with the technical complications but also with chronic rejection. The spectrum of ureteral ischemic injury extends from early necrosis and urinary leakage to delayed ureteral obstruction, presenting months to years after the actual transplantation.

DIAGNOSIS

Urinary leaks are often suspected because of increased drainage from the wound. The fluid should be tested for BUN/creatinine to see if it is urine. Radiographic tests of help include an abdominal ultrasound and nuclear renal scan. A renal scan demonstrating extravasation is the most sensitive method to differentiate a urine leak from other fluid collections such as lymphoceles or hematomas. A cystogram should be performed if a bladder leak is suspected.

Ureteral obstruction, usually manifested by graft dysfunction, requires evaluation, and again an ultrasound and nuclear renal scan are the most common screening studies. Additional radiographic studies such as a CT scan may be of assistance in some cases. With both ureteral leaks and obstruction, endourologic techniques can be both diagnostic and therapeutic.

INDICATIONS FOR SURGERY

Anything that causes graft dysfunction or results in disruption of the urinary tract in a renal transplant patient is of utmost concern and requires rapid diagnosis and treatment. In the case of ureteral leakage or obstruction, the goals of treatment include careful and accurate diagnosis of the exact cause and site. If the problem has a physical cause such as a leak or an obstruction and is not associated with an acute rejection episode, then treatment is directed at stabilization of the renal function, minimization of morbidity, and a restoration of the continuity and function of the urinary tract. If there is concomitant rejection, then definitive operative therapy is withheld pending the treatment of rejection.

ALTERNATIVE THERAPY

The need for immediate open operative surgical intervention has been replaced, to a large extent, by early endourologic intervention. The placement of a percutaneous nephrostomy can divert a leak or relieve obstruction and allow more definitive diagnosis. As described by Streem et al., endourologic management algorithms can select patients for whom the likelihood of successful nonoperative management is good. Depending on the selection criteria, the results of management of distal ureteral leaks with stenting and a nephrostomy tube show that approximately one-third of patients do well long term and require no additional treatment. For ureteral strictures or stenoses, approximately 45% of patients, carefully selected, will avoid an open operative repair. For the other patients, percutaneous access can allow stabilization of renal function and a more critical assessment before open surgical repair is carried out. In a few cases, percutaneous access can offer long-term treatment with chronic stent management. This choice, in my opinion, is of limited application in most patients with a well-functioning graft because of the long-term risks (i.e., stone formation, infection, etc.) and inherent costs. However, in patients who are not operative candidates and for some patients with marginal graft function, chronic endourologic treatment can be an alternative to definitive repair.

DESCRIPTION OF THE PROCEDURE

There are many procedures available to restore the continuity of the urinary tract. In our experience dealing with a difficult ureteral stenosis or a leak from significant ureteral ischemic necrosis, we favor the use of the native ureter to replace the transplant ureter. Advantages of this repair include: the native ureter is usually nonrefluxing, the results are reliable, there is a low likelihood of recurrence of the primary problem, and a tension-free anastomosis with good blood supply is easily attained. The focus of this operative description is on that surgical choice.

Surgical access to the transplanted kidney and ureters (transplant and native) is usually achieved by reopening the old incision. Occasionally, if extensive mobilization of transplanted kidney is anticipated or access to the contralateral native ureter is planned, a midline incision is an option. Surgical access to repair an early ureteral leak is usually simplified because dense fibrosis has not yet occurred, the fascial layers are easily opened, the peritoneum and its contents are freely mobilized medially and cephalad, and the kidney and ureter are identified without much difficulty. A primary repair can often be performed, and in most cases, a repeat ureteroneocystostomy at a new site in the bladder is the best choice. Use of a mechanical retractor greatly simplifies exposure and allows excellent access to the pelvis.

If the repair has been delayed because of attempted endourologic management or because of delay in presentation or diagnosis, then access to the ureter and kidney can be much more challenging and hazardous. In these cases, mandatory preoperative preparation includes a review of the operative note, especially if the operation was performed by someone else. It is important to know whether the kidney to be operated on was the donor’s right or left kidney. It is critical to know the position of the ureter and renal pelvis in relation to the renal vessels (below or above), and this depends on which kidney was used and into which side of the recipient’s pelvis it was transplanted. Additional information to be sought includes the type of vascular anastomosis performed (end-to-end versus end-to-side, etc.) and whether or not the iliac vessels (especially the iliac vein) were mobilized. All of this information can help to determine the likely position of the kidney in relation to the transplanted and native ureter and the anticipated ease in gaining access to these structures.. Note that this depicts a donor right kidney on the right side, as the renal pelvis is posterior to the renal vessels.

In terms of the recipient, it is critical to know the status of the recipient’s urinary tract. This is especially true if the recipient had a history of ureteral reflux or had undergone nephroureterectomy and might not have a suitable native ureter available to use for repair. Finally, the status of the recipient’s urinary bladder in terms of capacity, compliance, and function can be important in determining which other repair options are available.

Additional preoperative preparation involves stabilization of the patient and function of the graft. It is important to delay any open operative repair until concurrent rejection episodes have been adequately treated and renal function stabilized. All patients should be treated with preoperative antibiotics based on anticipated contaminants or cultures obtained from the urine. If there is a likelihood that bowel might be needed (a very unusual circumstance) to repair the urinary tract, then a full bowel prep is indicated.

The goals of surgery are to repair the ureteral defect, reestablish continuity of the urinary system, get rid of all foreign bodies as quickly as possible, and avoid graft or patient loss. With a well-planned and executed procedure, these goals should be easily obtained in essentially all cases.

Delayed surgical repair because of attempted endourologic management, delayed diagnosis, or late presentation of obstruction makes surgical exposure of the kidney and ureter very challenging. As noted earlier, access is almost always achieved through the old transplant incision, and cephalad extension of the incision is often needed in these cases because of perinephric fibrosis, the increased size of the kidney posttransplant, and to achieve access to the iliac vessels and native ureter. It is usually possible to extend the incision several centimeters cephalad. Additional exposure, if needed, can also be obtained by extending the inferior aspect of the incision across the midline, though this is rarely needed and should be delayed until the need is present.

With delayed repair, the normal tissue planes are obliterated, and a dense fibrosis has occurred around the graft. This makes it very easy to violate the “renal capsule” and get into significant bleeding. As a routine, it is preferable to operate from a position of “known to unknown” with good exposure. The surgeon should also plan to gain vascular control proximally and distally if it appears that the kidney may need to be mobilized in order to permit access to the renal pelvis. A three-way Foley catheter should always be placed into the bladder before the start of the surgery to allow for irrigation and filling with an antibiotic solution.

In order to assure a safe and adequate exposure, I usually open the peritoneum early in cases where there is dense fibrosis. This allows better cephalad exposure, protects the bowel, and gives good access to the bladder.

Because the transplant ureter usually crosses the external iliac vessels below the renal vessels, one should take care to avoid these structures while gaining access to the ureter. This is a critical feature of this operative procedure because exact visualization of the renal vascular structures is often difficult, and many times one is operating based on the expected, not visualized, location of these structures. In some cases a percutaneous nephrostomy tube will be placed as well as a ureteral stent. If present, the nephrostomy tube should be accessible during a procedure as injection of saline or methylene blue may aid in identifying the ureter and renal pelvis. In some cases, because of the dense fibrosis, the ureter is identified only when it is actually cut. The routine placement of a ureteral stent is of limited value in most cases because the fibrosis is so dense, it is hard to discern the presence of the catheter. If the ureter is not in dense fibrosis, then access is usually easy.

Once access to the bony pelvis is obtained, careful dissection along the lateral wall of the bladder usually leads to the ureter. Once it is identified, care must be used in mobilizing the ureter to avoid any further vascular injury. When the site of leakage and/or obstruction has been identified, the most commonly used repairs include (a) a repeat ureteroneocystostomy, (b) use of the bladder (Boari flap or bladder hitch) to help bridge the gap, or (c) use of a native ureter to perform a ureteroureterostomy or ureteropyelostomy. Repeat ureteroneocystostomies are indicated only to repair early leaks when the problem was from tension at the anastomosis or distal ureteral ischemia and a well-vascularized minimally fibrosed ureter is present. In most circumstances, especially late, with a lot of periureteral reaction or ischemia, the preferred option is the use of the ipsilateral native ureter if it is present and of adequate caliber. If not, then a Boari flap is an excellent choice.

Access to the native ureter is obtained by identifying it as it crosses the common iliac vessels. Care must be used in mobilizing the ureter down into the pelvis to the level of the superior vesical artery to avoid injury to the ureter blood supply. The ureter is divided well above the iliac vessels, and the proximal end of the ureter is doubly ligated. In our experience of over 30 cases, this has not resulted in problems with the native kidney or ureter requiring any further intervention. The operative positioning of the native ureter depends on access to the transplant ureter and/or pelvis. In addition, whether a side-to-side ureteral anastomosis or a ureteropyelostomy is to be performed may make a difference to the exact positioning of the native ureter. All of these factors relate to the extent of fibrosis and the appearance of the transplant ureter. To prevent any additional future problem, a tension-free, widely spatulated anastomosis of well-vascularized ureter to either transplant ureter or renal pelvis is critical. The anastomosis is performed using 5-0 Maxon (Davis and Geck, Danbury, CT) or Polydioxanone (PDS, Ethicon, Somerville, NJ) in a watertight single layer. The critical aspect is to obtain a mucosa-to-mucosa approximation avoiding tension, devascularization, and urinary leak. A 12-cm 4.7 double-J stent is routinely used on all anastomoses. The anastomosis may be additionally wrapped in omentum or peritoneal flap, if available, to decrease further the risk of leak. The wound is well irrigated with antibiotic solution, and if no preoperative infection was present, we close the wound without a drain. If there is concern about urinary leak, lymphatic leak, or possible infection, one or two Jackson Pratt drains are indicated. The fascia is closed in layers with a 0 or #1 permanent monofilament suture. The subcutaneous tissue is not closed. The skin is usually closed with staples. A nephrostomy tube, if present, is removed at day 5 to 7 after an antegrade nephrostogram has been obtained to be sure that there is no leak. The ureteral catheter is left in for 4 to 6 weeks.

OUTCOMES

Complications

Complications that can occur postprocedure include infection, urinary leak, bleeding, recurrence of the stricture, and possible loss of graft. In all series, these are very uncommon complications.

Results

We have performed over 30 native-to-transplant ureteroureterostomies or ureteropyelostomies to treat ureteral obstruction or ureteral leaks or to deal with damaged ureters at the time of the transplant. In our experience, all kidneys involved have been “salvaged,” and none lost to urologic complications. There have been no significant postoperative complications and no patient deaths. We have not had to repeat any procedures in any of the patients we have treated and have not had any recurrence of either leak or stricture. As noted earlier, we routinely tie off the proximal native ureter, do not do a nephrectomy, and have not had any problems related to the native kidney. We feel that routine native nephrectomy is not indicated, and if one is ever subsequently indicated, a laparoscopic nephrectomy would be our choice.

Transplantation of a kidney allograft and subsequent immunosuppression in patients with renal failure demand surgical precision and zero tolerance for errors of judgment or technique. The devastating consequences of vascular, urologic, and infectious wound complications in renal transplantation, with their associated morbidity, mortality, and graft loss, are well documented. Fortunately, strict adherence to techniques and principles outlined in this chapter can reduce the incidence of these problems to very low levels.

INDICATIONS FOR SURGERY

Indications for surgery include patients with chronic renal failure. Contraindications to renal allograft transplantation include a history of cancer (especially hematopoietic, renal cell carcinoma, or melanoma), active infections, and patients who are a poor operative risk. Relative contraindications include oxalosis and other metabolic disorders, psychological instability, and focal glomerulosclerosis.

ALTERNATIVE THERAPY

The alternative to renal allotransplantation is chronic dialysis.

SURGICAL TECHNIQUE

Preparation of the Patient

The prospective transplantation recipient should be in metabolic, fluid, and electrolyte balance to avoid perioperative hyperkalemia, unstable blood pressure, pulmonary edema, dehydration, or difficult operative hemostasis associated with inadequate dialysis. When dialysis can be scheduled in advance, as with living related donor transplantation, it should be performed on the day before surgery. The patient’s cardiopulmonary status needs to be well documented, and central venous pressure monitoring is routine. Swan–Ganz monitoring is often useful.

The entire abdomen is shaved and prepped after the induction of anesthesia and insertion of an indwelling 16 or 18 Fr Foley catheter. Any urine present in the bladder is submitted for culture. Then the bladder is distended with 150 ml or more of a saline solution containing Neosporin GU Irrigant. This greatly facilitates the anterior cystotomy later in the procedure and, in addition, protects against possible wound contamination when the bladder is opened. After instillation of the antibiotic solution, the catheter is clamped. The clamp is removed only after cystotomy closure is completed.

Incision and Iliac Fossa Dissection

A lower quadrant curvilinear incision is extended from the symphysis pubis passing 2 cm medial to the anterior superior iliac spine and up to about 4 to 5 cm below the lower costal margin. The upper half of the incision is extended through the external oblique, internal oblique, and transversus abdominis muscles; in the lower half of the incision, the anterior rectus fascia is incised. The rectus muscle can then be dissected inferiorly to its tendinous insertion on the symphysis pubis and retracted medially. In thin patients we prefer to keep the cephalad portion of the incision also within the lateral border of the rectus muscle, thereby obviating any transection of muscle and simplifying the closure. The inferior epigastric vessels are identified as they pass across the incision and are preserved for possible use later. Next, an anterolateral retroperitoneal fascial plane is developed, permitting extraperitoneal entry into the iliac fossa.

With medial retraction of the peritoneum, the spermatic cord in the male patient or round ligament in the female patient is easily identified. In men, some of the connective tissue around the cord is freed to permit easier retraction. Usually, cord ligation should be avoided to prevent hydrocele formation, testicular atrophy, or infertility. In women, the round ligament is divided and ligated. Further development of the extraperitoneal space in the iliac fossa is accomplished with exposure of the distal common and external iliac artery. The insertion of a self-retaining retractor at this point assures adequate exposure for the subsequent iliac vessel dissection and vascular anastomoses.

The dissection and skeletonization of the iliac vessels must be performed in a manner that allows secure ligation of the divided lymphatics passing along and across these vessels. Usually, this process is best approached on the medial aspect of the external iliac vein, working cephalad with a right-angle clamp toward the internal iliac artery, which crosses the vein. In some cases, especially when the donor kidney is large or has a short vein, the internal iliac artery must be sacrificed in order to achieve sufficient mobilization of the underlying vein. The iliac vein can be skeletonized as far cephalad as the vena cava if necessary. Posterior venous tributaries must be divided to permit maximum anterior mobility of the iliac vein. It is best to ligate all tributaries doubly with 2-0 or 3-0 silk in continuity before division because a double-clamping maneuver may sometimes result in injury or avulsion of a poorly accessible stump during ligation. Hemostasis then can be achieved only with difficulty and with risk of obturator nerve injury. Unless the internal iliac artery already has been selected for an end-to-end allograft anastomosis, right-angle clamp dissection is used to partially skeletonize the common and external iliac arteries. The tissue overlying the arteries and containing the lymphatics is sequentially separated, doubly ligated with 3-0 silk, and divided, a strategy that greatly reduces the incidence of lymphocele. Again, this tissue should be doubly ligated before it is incised, in contrast with double clamping and division of the tissue. Just as with the vein, the anterior separation of tissue over the iliac artery is more easily performed in a cephalad direction.

At this point, palpation of the common iliac bifurcation and internal iliac artery determines the suitability of the internal iliac artery for an end-to-end anastomosis with the renal artery and the need for endarterectomy. If there is moderate or severe atherosclerosis extending into the bifurcation, or great size disparity, the internal iliac artery is usually not used. If an endarterectomy can be performed safely, or if there is little evidence of atheroma in the internal iliac vessel, skeletonization of this vessel prepares it for end-to-end anastomosis. Before skeletonization is begun, the lymphatics on the medial aspect of the iliac bifurcation should be doubly ligated and divided. If the internal iliac artery is to be used, it may be clamped proximally with a Fogarty clamp and divided distal to its bifurcation with appropriate ligation of the distal stumps deep in the pelvis. The mobilized internal iliac artery is irrigated with heparinized saline solution.

Allograft Positioning and Vascular Anastomoses

Before recipient vessel anastomotic sites are selected, visualization of the ultimate resting place for the allograft lateral or anterior to the iliac vessels should be considered, with all anatomic factors taken into account. The iliac vein is prepared for the end-to-side renal vein anastomosis by placement of clamps proximal and distal to the proposed venotomy. Fogarty clamps usually serve this purpose well. Excision of a thin ellipse of vein produces an ideal venotomy. The isolated segment of the iliac vein is irrigated with heparinized saline. After this, four 6-0 cardiovascular sutures are placed at the superior and inferior apices and at the midpoints of the medial and lateral margins of the venotomy. These sutures later are passed through corresponding points on the donor renal vein or vena cava patch for a four-quadrant end-to-side anastomosis.

If a cadaveric kidney is used, the allograft is removed from cold storage or perfusion preservation at this point. With living related transplantation, the flushed and cooled graft is obtained from the live donor in an adjacent operating room.

The kidney is secured in a sling or a 3-inch stockinette4 containing ice slush and held in position for the vascular anastomosis by the assistant. A clamp is used to secure the sling to relieve the assistant from holding the kidney in position with the hands, which might accelerate warming of the kidney during the performance of the vascular anastomoses.

The previously placed four sutures through the iliac vein are passed through the corresponding points of the donor renal vein, Carrel patch, or vena cava conduit and secured, bringing the renal vein into juxtaposition with the iliac vein. The medial and lateral sutures are retracted to separate the venotomy opening and facilitate rapid anastomosis without inadvertent suturing of the back wall. With the table rotated laterally, the superior suture is used as a running suture down the medial side of the renal vein to meet the inferior suture running up. The lateral suture line is then run in similar fashion after the table has been rotated medially. The clamps on the iliac vein may be left in place until completion of the arterial anastomosis, but application of a finger Fogarty or a bulldog clamp across the renal vein at this time allows for removal of the iliac vein clamps and earlier restoration of venous return from the lower extremity.

If the internal iliac artery is to be used for the arterial anastomosis, an end-to-end anastomosis is then performed with the renal artery. The two vessels are positioned to allow a gentle upward curve from the iliac bifurcation to the kidney by fixating the superior and inferior arterial apices with interrupted 6-0 cardiovascular suture. The anastomosis is completed with continuous or interrupted sutures. With the kidney resting in the iliac fossa or suspended in a sling, the initial interrupted suture may be placed midway between the apical sutures on the anterior vessel walls facing the operator, thus allowing better approximation of the opposing arterial margins, particularly when a discrepancy exists in the size of the vessels. Subsequently, the remaining sutures are placed to approximate each anterior quadrant. Next, the previously placed apical sutures are used to rotate the arteries so that the posterior vessel walls are now in the anterior position for subsequent interrupted or running suture placement. Just as before, a suture placed midway between the apical sutures again divides the rotated posterior vessel walls into quadrants for subsequent suture placement. A preference for interrupted sutures instead of a running suture in this end-to-end anastomosis prevails when one needs to avoid absolutely any pursestring effect that might occur from a running suture or to achieve optimal accommodation of the two vessels to each other when a size or thickness discrepancy exists.

In most cases, the internal iliac artery is left intact to preserve potency in men as well as gluteal and pelvic blood supply in the elderly. Therefore, end-to-side anastomosis of the renal artery to the external or common iliac artery is chosen more commonly than the end-to-end procedure just described. This anastomosis usually is placed cephalad to the level of the venous anastomosis. The location of clamp placement must be carefully selected so as not to disrupt existing arteriosclerotic plaques and precipitate embolization or thrombosis. A longitudinal incision is made on the anterior or anterolateral portion of the iliac artery segment with a #11 blade knife, and a 4.8- or 5.6-mm aortic punch is used to prepare an ideal oval arteriotomy. After the incision is made, regional heparinization of the lower extremity may be accomplished by instilling about 80 to 100 ml of heparinized saline (1,000 units/100 ml) into the distal iliac limb. Systemic heparinization is usually not necessary. This anastomosis is also performed with 6-0 cardiovascular continuous or interrupted sutures after initial fixation of the end of the renal artery to an apex of the arteriotomy with suture cinched down by parachute technique.

The previously placed sling around the kidney is removed. It is desirable to have obtained preoperative assessment of the recipient for existing cold agglutinins, because moderate to high titers of these agglutinins require warming of the kidney before the circulation is reestablished. The vascular clamps are released after IV infusion of mannitol and methylprednisolone, venous clamps before arterial. At this point, the patient should be judiciously overhydrated with saline and albumin, and a dopamine drip should be ready to optimize renal blood flow if needed.

Multiple Renal Vessels

Although the Carrel patch may frequently be used with single arteries and veins, a cadaveric kidney with multiple renal arteries perfused through the aorta is especially well suited to an end-to-side anastomosis of a Carrel patch encompassing the multiple arteries. If the vessels are close to each other, a single Carrel patch is sufficient. If the vessels are more than 2 cm apart, we prefer two Carrel patches. The Carrel patch of donor aorta is fashioned to accommodate the multiple vessels, and its anastomosis to the common or external iliac artery is performed with continuous 5-0 or 6-0 cardiovascular sutures after an arteriotomy that accommodates the width and length of the Carrel patch. This anastomosis is best performed by fixating the patch at the superior and inferior apices of the arteriotomy or by parachute technique. Each suture limb runs away from the apex.

The presence of multiple arteries in related donor transplantation is known in advance because all living related donors have preoperative arteriograms. Most donors have at least one kidney with a single artery, but, at times, a donor kidney with double arteries or triple arteries must be used. These arteries cannot be taken with a Carrel patch because of the risk to the donor. In these instances, several strategies for arterial anastomoses are possible: double end-to-side renal arteries to iliac artery, end-to-end superior renal artery to internal iliac artery with end-to-side inferior renal artery to external iliac artery, and implantation of an accessory artery end-to-side into the larger main renal artery, with the larger renal artery anastomosed to the internal, external, or common iliac artery. If two renal arteries are of similar diameter, the spatulation edges of the renal arteries can be joined with a running 6-0 or 7-0 cardiovascular suture to create a single bifurcating artery. An accessory artery to main renal artery anastomosis should be performed with ex vivo bench technique in cold ice slush before the renal vein anastomosis is done. Finally, some recipients have a deep inferior epigastric artery that is suitable for end-to-end 7-0 suture interrupted anastomosis of a small lower-pole artery, which may be essential for ureteral viability.8 Our experience in more than 30 cases with this technique has been excellent; no ureteral ischemia or necrosis has occurred.

Ureteroneocystostomy

Some patients are prepared for kidney transplantation by creation of an ileal loop or isolated ileal stoma to divert urine from a dysfunctional or absent bladder. These techniques are beyond the scope of this discussion. In addition, when the donor ureter is absent or damaged, the recipient ureter may be used for ureteroureterostomy or ureteropyelostomy to the allograft.

Various modifications of the Politano-Leadbetter, Paquin, and Lich techniques are used for allograft ureteral implantation into the bladder. In our experience, when the bladder is very small or the donor ureter is very short, an extravesical technique is best.6 Otherwise, we prefer the ease and reliability of a transvesical approach without a formal submucosal tunnel.7 In either case, previous filling of the bladder facilitates a longitudinal anterior cystotomy with minimal trauma to the bladder wall. In the transvesical approach, the bladder dome is packed and retracted cephalad, exposing the bas fond. An oblique tunnel is created in the bladder floor using a tonsil clamp directed toward the trigone from outside the bladder. This maneuver prevents subsequent angulation of the ureter when the bladder is distended. An 8 Fr Robinson catheter or heavy silk is passed through the tunnel in retrograde fashion and secured to the donor ureter.

The ureter is pulled down and brought into position in the bladder by gentle traction. This maneuver avoids any handling of the ureter, which is important because the ureter of the transplanted kidney receives its blood supply exclusively from the renal vessel branches that course in its adventitia. In male patients, it is important to pass the ureter beneath the spermatic cord. Intravesically, the ureter is hemitransected about 4 cm from its entrance site into the bladder and spatulated about 1 cm.

Four sutures of 4-0 chromic catgut are usually sufficient for an anastomosis incorporating bladder mucosa and muscularis as the ureteral transection is completed. When the apical stitch also catches ureteral adventitia 1 to 2 cm above the apex, a nice everted ureteral nipple may be produced. This eversion is especially desirable with patulous ureters. The ureter is not stented routinely. A no-touch technique is essential to avoid producing vascular insufficiency, ureteral necrosis, and urinary extravasation from injury to the adventitial vascular network of the ureter.

The oblique bladder tunnel and muscle hiatus must accommodate the ureter comfortably to avoid postoperative obstruction from edema, and a gentle oblique course of the ureter must be ensured so that no kinks, twists, or obstructions occur. This attention is important because the ureter of a transplanted kidney crosses the iliac vessels in a much more caudal position than the native ureter. A little redundancy of the ureter is established outside the bladder to ensure that the ureteroneocystostomy is done without tension and that postoperative allograft swelling will not unduly stretch or angulate the ureter. Patency of the ureteroneocystostomy is confirmed by gently passing a 5 Fr feeding tube or an 8 Fr or smaller soft Robinson catheter toward the renal pelvis.

Kidneys with a double ureter can also be transplanted successfully. These ureters should be dissected en bloc within their common adventitial sheath and periureteral fat so that the ureteral blood supply is protected. The technique of ureteroneocystostomy is essentially the same as with a single ureter, except that the ureters are brought through together side by side in a nonconstricting tunnel. The distal end of each ureter is spatulated, and the adjacent margins are approximated with 5-0 chromic catgut.

To ensure a watertight closure, the cystotomy incision is closed in three layers. The first 3-0 chromic running suture secures the full thickness of the bladder near the bladder neck and closes the mucosal layer. The second 2-0 chromic running suture is an inverting layer of muscularis. The third 2-0 chromic layer inverts the adventitia. Each layer should overlap the immediately underlying layer about 0.5 cm at each end of the cystotomy closure to avoid urinary extravasation at these two points.

Pediatric Kidneys

Although en bloc transplantation of kidneys from very young children is often desirable,5 it is not necessary to transplant both kidneys from young children en bloc each kidney can be used for a different recipient, as is the case with adult cadaveric donors, using Carrel patches of donor aorta and vena cava. A Carrel patch is mandatory in these cases because direct implantation of a small vessel into a much larger or diseased vessel may result in thrombosis or produce functional stenosis as the kidney grows. When the en bloc technique is used, the two ureters are implanted separately and stented. Pediatric kidneys have proven to be excellent donor grafts for carefully selected adults and children. Avoidance of older recipients or diabetics with advanced arteriosclerosis will minimize the potential for thrombosis. Rapid growth and hypertrophy occur in the immediate posttransplantation period. If early rejection can be avoided, these allografts achieve adult size and function in adult recipients within several weeks.

Pediatric Transplantation

In small children, the iliac fossa is not large enough to accommodate a kidney from an adult donor, and the pelvic vessels in a small child are so small that the disparity between the donor renal vessels and the recipient vessels precludes the technique described for adults. In these small children, graft implantation must use the recipient aorta and vena cava, which is best accomplished through a right-sided retroperitoneal or transperitoneal midline abdominal incision that provides ready access to the great vessels as well as the urinary bladder. After the right colon is reflected medially, the right kidney is usually removed to make room for the allograft. The vena cava is then freed from the level of the right renal vein inferiorly to its bifurcation or beyond. Posterior lumbar veins are doubly ligated with 5-0 silk and divided. Mobilization of the vena cava is important to facilitate the end-to-side anastomosis of the renal vein, which is performed with running 6-0 ardiovascular sutures, as described for the adult . Performing the venous anastomosis superiorly allows room for an end-to-side anastomosis of the renal artery to the inferior abdominal aorta. Aortic mobilization should be limited to its distal portion, from the level of the inferior mesenteric artery, and including both common iliac arteries. The segment of the aorta to be used for the end-to-side renal artery anastomosis can be isolated by a superior pediatric vascular clamp and by two inferior clamps or silastic loops on the common iliac arteries. The end-to-side anastomosis is performed with interrupted 6-0 cardiovascular sutures.

Important to the revascularization of an adult kidney in small children is the need to anticipate the impending consumption of several hundred milliliters of effective blood volume by the renal allograft. Initiation of blood transfusion before beginning the vascular anastomoses will avoid hypotension after release of the vascular clamps. When the vascular anastomoses are completed, the superior aortic clamp must be kept loosely in place until it is clear that hypotension is not a problem. Immediately after establishing circulation in the graft, the anesthesiologist must obtain blood pressures at 30-second or 1-minute intervals until stabilization is assured. The ureteral implantation is carried out as described except that the ureter must be passed retroperitoneally behind the bladder near the midline.

Wound Closure

Except in unusual cases, the allograft ureter is not stented, and the space of Retzius and iliac fossa are not drained. Jackson–Pratt suction may be employed, but Penrose drains are never used. If good hemostasis has been obtained, and if the principles of implantation as outlined in this chapter have been followed, there is no need for postoperative drainage other than a urethral catheter. The optimal period of Foley catheter drainage is debatable. We prefer to remove the catheter within 48 hours unless the patient has worrisome hematuria, large diuresis, or poor bladder function.

Before wound closure, the wound is thoroughly irrigated with saline. The wound is then closed using a 1 Maxon running suture to approximate transversus abdominis and internal oblique muscles in a single-layer closure; the adjacent fascia is included inferiorly at the tendinous insertion of the rectus muscle. Next, the rectus fascia anteriorly and the fascia of the external oblique are approximated with 1 Prolene running suture.

The subcutaneous tissue is thoroughly irrigated with saline and then may be approximated with interrupted 2-0 or 3-0 sutures. These sutures are placed about 2 to 3 cm apart and include both Scarpa’s fascia and the underlying fascia superficially. In this manner, one can obliterate dead space in the subcutaneous area in which a seroma in an immunosuppressed patient might become secondarily infected. The skin is approximated with interrupted fine nylon sutures or staples.

Renal injuries can be some of the most complex and challenging cases a urologist or trauma surgeon may face. The vast majority of renal injuries occur as a result of blunt trauma, and most of these are amenable to nonoperative management. Penetrating renal trauma usually occurs in conjunction with injuries to associated abdominal organs, which require urgent laparotomy. Systematic renal reconstruction at the time of laparotomy provides excellent functional results in the majority of cases.

DIAGNOSIS

Signs, symptoms, and laboratory findings that suggest renal injury should prompt immediate radiologic evaluation in stable patients. Gross hematuria after blunt trauma should warrant renal imaging in all cases. Adults with microhematuria in the presence of shock, deceleration injuries, or signs of significant abdominal, flank, or chest injuries after blunt trauma should also be imaged. Pediatric patients with significant microhematuria or signs of multiple injuries after blunt trauma should be radiographically evaluated. Penetrating wounds of the abdomen or flank with any degree of hematuria also warrant urgent renal imaging.

The best study for assessing the injured kidney in a stable patient is a renal CT scan. Renal images can be obtained in conjunction with an abdominal CT when trauma surgeons need this study to evaluate the extent of associated intra-abdominal injuries. When unstable patients are taken emergently for laparotomy and renal injuries are suspected, a one-shot intraoperative IVP is extremely useful. The intraoperative IVP consists of a high-dose (2 cc/kg) intravenous bolus injection of radiographic contrast; a single film is taken at 10 minutes. No scout film is necessary. This technique provides important information regarding the degree of injury of the kidney in question and the status of the contralateral kidney without delaying resuscitation.

INDICATIONS FOR SURGERY

The decision to surgically repair the traumatized kidney is based on consideration of the patient’s mechanism of injury, hemodynamic stability, associated injuries, and accurate radiographic staging of the injury. The vast majority of blunt traumatic renal injuries are clinically insignificant. At San Francisco General Hospital, fewer than 3% of patients with blunt renal trauma require renal exploration. Penetrating renal injuries, on the other hand, should usually be explored. Approximately 70% of patients with penetrating renal trauma are treated surgically at our trauma center. Only when radiographic staging clearly defines a penetrating injury as minor can a nonoperative approach be used successfully.

Persistent renal bleeding is an absolute indication for renal exploration. Relative indications for renal surgery include extensive urinary extravasation, nonviable renal tissue in association with a parenchymal laceration, incomplete clinical or radiographic staging, and arterial thrombosis. Also, if a trauma surgeon elects to perform an exploratory laparotomy to manage an associated abdominal injury, we will usually repair significant renal injuries at that time in order to prevent late complications. Nearly all renal lacerations occurring from gunshot wounds require immediate repair. In the absence of severe vascular injury or hemodynamic instability, renal reconstruction may safely be attempted. Successful reconstruction can be undertaken despite spillage from bowel injury, pancreatic injury, or other associated injuries.

ALTERNATIVE THERAPY

Nephrectomy, when required after renal trauma, usually occurs when an injury is deemed irreparable or in the setting of hemodynamic instability. Although nephrectomy is clearly a life-saving maneuver in these instances, it is only necessary in about 10% of cases. In general, patients requiring nephrectomy are much more seriously injured, are frequently in shock, and cannot be managed conservatively.

Renal stab wounds are successfully managed nonoperatively in about 50% of cases at San Francisco General Hospital. The types of stab wounds most amenable to an observational approach are those occurring posteriorly or in the flank, where intra-abdominal organs are unlikely to be involved. For those stab-wound patients in whom nonoperative management is being contemplated, renal CT provides excellent information regarding the depth of laceration, extent of urinary extravasation, and size of perirenal hematoma.

SURGICAL TECHNIQUE

Renal exploration in the trauma setting should be carried out through a standard midline abdominal incision. This approach provides complete access to the intra-abdominal viscera and vasculature, and it also gives the greatest flexibility to assess and repair a variety of genitourinary injuries. Major bleeding noted on opening the abdominal cavity should be controlled immediately with laparotomy packs followed by surgical control and repair. Associated injuries to other abdominal organs are usually addressed before examination of the kidneys if the patient is stable. The bowel, liver, spleen, pancreas, and other organs should be inspected systematically and carefully.

The renal vasculature is routinely isolated before a retroperitoneal hematoma surrounding an injured kidney is entered. This creates a safety net for reconstruction and reduces the risk of uncontrolled renal bleeding and subsequent nephrectomy. To facilitate access to the retroperitoneum, the transverse colon is lifted out of the abdomen superiorly and placed on moist laparotomy packs. The small bowel is placed in a bowel bag and lifted anteriorly to the right. An incision is made in the retroperitoneum over the aorta from the level of the inferior mesenteric artery to the ligament of Treitz, which can be divided for additional exposure. If hemorrhage obscures the aorta, the inferior mesenteric vein is identified, and the retroperitoneal incision is placed just medial to this important landmark.

Once the aorta is identified in the lower part of the incision, it is followed superiorly to the left renal vein, which reliably crosses anteriorly. The renal arteries can be found just posterior to the left renal vein on either side of the aorta. If the right renal vein is difficult to isolate through this approach, an alternative method of exposure is to mobilize the second portion of the duodenum off the vena cava. With lateral retraction on the vena cava, the right renal artery can then be isolated in its interaortocaval location.

The ipsilateral renal artery and vein are individually isolated with vessel loops. These vessels are not occluded initially unless bleeding is heavy, which occurs in approximately 12% of cases in our experience.1 Because the vessels are not routinely clamped, renal perfusion is continuous, and warm ischemia is avoided. Patients most likely to require temporary vascular occlusion are those in shock from active, uncontrolled renal bleeding. Vascular occlusion, when necessary for reconstruction, does not increase the incidence of postoperative complications when the warm ischemia time is kept around 30 minutes.

After vascular control, the kidney is exposed by incising the retroperitoneum just lateral to the colon. The colon is reflected medially, and dissection through the hematoma allows renal exposure. After the kidney has been bluntly and sharply mobilized, the entire renal surface, renal vasculature, and upper ureter are routinely inspected for the presence of exit wounds or multiple injured areas. If heavy bleeding ensues, Rummel tourniquets can be applied to the vessel loops for vascular occlusion. First, the renal artery alone is occluded. If bleeding persists, the renal vein is then occluded to eliminate back bleeding.

For major polar injuries, partial nephrectomy offers the best management. Nonviable tissue is sharply debrided from the injured area. Manual compression of the adjoining normal renal parenchyma, rather than formal vascular occlusion, is extremely useful during partial nephrectomy as an adjunct during control of moderate renal hemorrhage. Arcuate arteries are individually suture-ligated with 4-0 chromic suture to control hemorrhage. The collecting system is then closed watertight with a running 4-0 Vicryl suture. Methylene blue may be injected into the renal pelvis with simultaneous compression of the ureter to elucidate any leaks in the collecting system, which may then be oversewn.

The renal parenchymal defect should be covered with thrombin-soaked Gelfoam to enhance hemostasis and then covered with renal capsule, if possible. Typically, after partial nephrectomy for polar injuries, the remaining renal capsule is insufficient to allow for primary closure. In this case, an omental pedicle flap can be brought around the colon or through a window in the colon mesentery and attached with interrupted suture to the existing renal capsule for coverage of the defect. Its excellent vascular supply and lymphatic drainage make omentum an excellent tissue choice for coverage of renal injuries, especially in the setting of concomitant bowel or pancreatic injury. A retroperitoneal drain is placed routinely.

Major injuries to the midportion of the kidney are more difficult to repair than polar injuries, but the same surgical principles apply. Nonviable tissue is removed sharply. Sites of bleeding are individually ligated with fine absorbable sutures, and the collecting system is closed watertight. Interrupted 3-0 chromic sutures placed superficially are ideal for renal capsule approximation. Capsular sutures are best placed without incorporating the underlying parenchyma, as that tissue is extremely friable. Thrombin-soaked Gelfoam bolsters in the defect enhance hemostasis, prevent urinary leakage, and stabilize capsular closure. Again, omentum should be used if primary capsular closure cannot be achieved. We frequently place a row of small titanium staples in the renal capsule near the closure to visualize the operative site on subsequent imaging studies. A retroperitoneal Penrose drain is brought out through a separate incision in most cases. Suction-type drains may initiate or prolong urinary leakage.

Renal stab wounds may be repaired using the same methods detailed above. As discussed, many may be amenable to nonoperative management. If laparotomy is performed for associated injuries, renal reconstruction should be done concomitantly. Tissue destruction is frequently much less than that seen with gunshot injuries. Frequently, entrance and exit wounds may be simply oversewn.

Renal vascular injuries are a major cause of renal loss and may coexist with parenchymal lacerations. Main renal artery or complex renal vein injuries frequently lead to total nephrectomy. Venous injuries may occur along the main renal vein or in segmental branches. In either case, the first step is to temporarily occlude the main renal artery. Vascular clamps are then placed proximal and distal to the venous laceration. A running suture of 5-0 vascular silk is then used to close the venous defect. Segmental arterial injuries are best repaired in a similar fashion. Smaller segmental veins can safely be ligated because of the internal collateral circulation of the venous system. Also, the left main renal vein may be ligated proximally because there is extensive collateral flow through the adrenal, lumbar, and gonadal branches.

Gross blood in the urine usually clears within 24 hours, and patients should be observed at bed rest during this time. Ambulation is resumed once the urine is clear. Serial hematocrits should be monitored because delayed bleeding is possible. Renal angiography and selective embolization may be considered in the event of continued hemorrhage. Retroperitoneal drains are normally removed within 48 to 72 hours. If drainage is excessive, an aliquot may be checked for creatinine; a level similar to that of serum suggests peritoneal fluid rather than urine. Blood pressure is checked before discharge. A radionuclide study is usually obtained around the time of discharge to assess function, and a renal imaging study is again obtained at about 3 months.

OUTCOMES

Complications

Small amounts of urinary extravasation are usually not clinically significant as long as they do not become infected. Large urinomas are best treated with percutaneous drainage. Delayed renal hemorrhage is most likely within the first 2 weeks, and this complication is best treated initially with percutaneous embolization and supportive therapy. Hypertension occurs rarely after renal injuries, and it is usually easily controlled by medical therapy alone. Delayed urinary bleeding may be a sign of a vascular fistula to the collecting system: this complication is frequently difficult to reconstruct and may often be best treated with nephrectomy.

Results

Renal reconstruction has achieved adequate preservation of function in 83% of patients at our institution. We have found renal salvage to be safe in the presence of concomitant bowel or pancreatic injuries.

Renal and retroperitoneal abscesses are uncommon clinical entities that often pose a significant diagnostic challenge. Nonspecific signs and symptoms frequently lead to a delay in diagnosis and treatment. Consequently, they are associated with significant morbidity, and mortality rates approaching 50% have been reported. An understanding of the anatomy of the retroperitoneal space is essential for classification, diagnosis, and management of renal and retroperitoneal abscesses.

CLASSIFICATION

The retroperitoneal space is bounded by the posterior parietal peritoneum and transversalis fascia. It is divided into the perirenal space and the pararenal space.

The perirenal space surrounds the kidney and is bounded by the renal (Gerota’s) fascia. It contains a lemon-yellow layer of fat, which is thickest posteriorly and laterally. The anterior and posterior leaves of the renal fascia fuse above the adrenal gland, becoming continuous with the diaphragmatic fascia.1 A thinner, more variable layer meets between the adrenal gland and the kidney. Laterally, the fascial layers join to form the lateroconal fascia, which becomes continuous with the posterior parietal peritoneum. Medially, the posterior layer fuses with the psoas muscle fascia, and the anterior layer fuses with the connective tissue surrounding the great vessels and organs of the anterior retroperitoneum (i.e., the pancreas, duodenum, and colon). Because the perirenal space rarely crosses the midline, perirenal abscesses usually remain unilateral.16 Inferiorly, the renal fascial layers do not fuse but, rather, become continuous with the psoas and ureteral coverings. This opening inferiorly allows spread of perirenal infections to the pararenal space, to the pelvis, to the psoas muscle, and, in some cases, to the contralateral retroperitoneum.

The pararenal space is divided into two compartments: the anterior compartment, which is bounded by the posterior parietal peritoneum and the anterior renal fascia; and the posterior compartment, which is bounded by the posterior renal fascia and transversalis fascia. The pararenal space contains pale adipose tissue, which fills much of the retroperitoneal space. Because the anterior pararenal space extends across the midline, infection arising in one space may become bilateral. The posterior pararenal space does not cross the midline, and infection within it remains unilateral.

The retrofascial compartment lies posterior to the transversalis fascia. It is important only in development of the rare retrofascial abscess from abscesses of the psoas, iliacus, and quadratus muscles.

PATHOGENESIS

Before the advent of antimicrobial therapy, most renal abscesses occurred as a result of hematogenous spread of gram-positive organisms, usually Staphylococcus aureus. These abscesses, which were called renal carbuncles, may still be seen in intravenous drug users and in patients with dermatologic disorders. They may resolve with aggressive antimicrobial therapy if treated before frank suppuration. Presently, most renal and retroperitoneal abscesses are caused by retrograde ascent of gram-negative bacteria from the bladder. The most common organisms include Escherichia coli, Proteus, Klebsiella, and Pseudomonas. Anaerobes may be isolated in abscesses associated with gastrointestinal and respiratory infections. Abscesses caused by opportunistic organisms such as Candida and Aspergillus may occur in immunosuppressed patients. Other uncommon pathogens include Mycobacterium tuberculosis and Echinococcus (see below).

A renal abscess is generally preceded by pyelonephritis, which progresses to abscess formation in the presence of a virulent uropathogen, a damaged or obstructed urinary tract, or a compromised host. Renal abscesses have a predilection for the cortical medullary region and may drain spontaneously through the renal collecting system. When renal infection is complicated by obstruction, a purulent exudate collects in the renal collecting system. Pyonephrosis refers to infected hydronephrosis with suppurative destruction of the parenchyma of the kidney, with total or near total loss of renal function. The most frequent cause of obstruction is calculous disease. A previous history of urinary tract infection or surgery is also common.

Perirenal abscesses usually occur by erosion of abscesses or pyonephrosis into the perirenal space. Because of gravity, the resulting perirenal suppuration tends to localize dorsolaterally to the lower pole of the kidney. Posterior pararenal abscesses may arise from perirenal abscesses or from anterior pararenal abscesses tracking into the pelvis, where the anterior and posterior pararenal spaces communicate. Occasionally they result from hematogenous spread. Anterior pararenal abscesses are rarely urologic in origin. They arise from infection involving the organs within the anterior pararenal space, namely the ascending and descending colon, appendix, duodenal loop, and the pancreas. Abscesses arising from the gastrointestinal tract usually harbor a mixture of microorganisms, of which E. coli is the most prevalent. Extension of anterior pararenal abscesses into the perirenal space is uncommon.

DIAGNOSIS

The diagnosis of renal and retroperitoneal abscesses requires a high index of suspicion, as they typically present with insidious, nonspecific signs and symptoms. Presenting symptoms may include fever, chills, abdominal or flank pain, irritative voiding symptoms, nausea, vomiting, lethargy, or weight loss. Symptoms have been present for more than 5 days in the majority of patients with renal and retroperitoneal abscesses, compared with 10% of patients with pyelonephritis. Over one-third of patients may be afebrile. The majority of patients diagnosed with renal and retroperitoneal abscesses have underlying, predisposing medical conditions. These include diabetes mellitus, urinary tract calculi, previous urologic surgery, urinary tract obstruction, polycystic kidney disease, and immunosuppression.

A palpable flank or abdominal mass is present in about half of the cases. The mass may be better appreciated by examination of the patient in the knee–chest position. There may also be signs of psoas muscle irritation with flexion of the thigh.

Laboratory tests are helpful but nondiagnostic. Leukocytosis, elevated serum creatinine, and pyuria are common. Blood and urine cultures are frequently negative; when positive, they usually correlate with culture results from the abscess.

Excretory urography may aid in the diagnosis of renal or retroperitoneal abscesses by showing diminished mobility on inspiratory–expiratory films. A renal abscess causes a decrease in function and enlargement of the nephrogram during the acute phase. Retroperitoneal abscesses may cause displacement of the kidneys or ureters by a mass, scoliosis of the spine, and free air or fluid in the retroperitoneal space. Computed tomography (CT) is highly sensitive for the diagnosis of renal and retroperitoneal abscesses. It precisely localizes and assesses the size of an abscess so that the type of intervention and its anatomic approach can be determined. The presence of gas within a lesion is pathognomonic for an abscess. Additional CT findings characteristic of an abscess include a mass with low attenuation, rim enhancement of the abscess wall after contrast, obliteration of tissue planes, and displacement of surrounding structures. Ultrasonography is less sensitive than CT but useful for monitoring response to therapy. Arteriography and radioisotope scanning rarely add significant information.

INDICATIONS FOR SURGERY

Renal and retroperitoneal abscesses are generally lethal if untreated. Therapeutic options include antimicrobial therapy, percutaneous catheter drainage, and open surgical drainage.

ALTERNATIVE THERAPY

Antimicrobial therapy as the sole treatment is an option, yet most abscesses cannot be cured without drainage. Small renal abscesses may resolve, however, if they are treated early with aggressive antimicrobial therapy. Prolonged antimicrobial therapy without drainage is indicated only if favorable clinical response and radiologic confirmation of abscess resolution indicate that the therapy is effective. If antimicrobial therapy is not effective, prompt percutaneous or open surgical drainage of the pus is mandatory. Progression of a renal abscess leads to perinephric abscess or perforation into the collecting system and results in signs and symptoms of urinary tract infection.

Antimicrobial therapy should be instituted after the urine has been Gram-stained and urine and blood cultures have been obtained. Broad-spectrum coverage should be guided by the presumptive diagnosis and the presumed pathogen. An aminoglycoside for gram-negative rods and ampicillin for gram-positive cocci are preferred. Anaerobic coverage with a drug such as clindamycin is warranted when Gram stain reveals a polymicrobial flora or when a gastrointestinal source is suspected. If the abscess may be of staphylococcal origin, a penicillinase-resistant penicillin, such as nafcillin, should be added. Antimicrobial therapy should be reevaluated when the results of culture and sensitivity tests are available. Unfortunately, urine and blood cultures are frequently sterile, and empirical therapy must be modified on the basis of clinical response and changes in imaging studies.

SURGICAL TECHNIQUE

Percutaneous Drainage

Most renal and retroperitoneal abscesses are treated with empirical antimicrobial therapy and immediate percutaneous drainage. When successful, minimally invasive therapy minimizes operative morbidity and allows for preservation of renal tissue. The abscess must be confirmed by CT-guided or ultrasonography-guided needle aspiration and must be drainable without injury to other organs. Immediate surgical drainage must be instituted if the procedure fails. After a multiport drainage catheter (8 to 12 Fr) is positioned, the abscess should be drained, and adequate evacuation should be confirmed by CT or ultrasonography. The catheter should then be connected to low intermittent suction, and drainage outputs should be monitored daily. If drainage stops abruptly, occlusion of the catheter should be suspected, and it should be irrigated gently with small amounts of normal saline. Computed tomography or ultrasonography should be performed periodically to monitor catheter position and size of the abscess. Direct instillation of contrast through the drainage tube may be helpful to confirm the catheter position or to rule out a fistula. To avoid bacteremia, prophylactic antimicrobial coverage should be given, and the contrast should be instilled under gravity or by gentle injection. Instillation of 2,500 units of urokinase in 50 ml of normal saline on a daily basis may be successful in evacuating an organizing infected hematoma. Routine abscess irrigation with antimicrobials is of questionable benefit and may promote overgrowth of resistant bacteria. The catheter should be withdrawn gradually as the abscess cavity shrinks and the drainage decreases. The usual duration of drainage is 1 to 3 weeks. The catheter is removed when drainage stops and CT and ultrasonography show complete resolution.

Open Surgical Drainage

The incision should be smaller than that used for routine nephrectomy, and usually a posterior flank muscle-splitting incision below the 12th rib is sufficient. When the retroperitoneal abscess is entered, the pus should be cultured, and the space gently but thoroughly explored to ensure that all loculated cavities are drained. Thorough irrigation of the cavity is essential. Multiple Penrose drains should be inserted into the space through separate stab wounds, and the ends of the drains should be sutured to the skin and tagged with safety pins. Fascial and muscular closure may be performed with chromic catgut suture, but skin and subcutaneous tissue should be left open to prevent the formation of a secondary body wall abscess. The wound can be left to heal from within, or skin sutures may be placed and left untied for dermal approximation 5 to 7 days postoperatively after drainage has ceased. The wound should be packed with gauze, and the packs should be changed daily. The drains should be left in place until purulent drainage has decreased, and then they can be removed slowly over several days.

ANCILLARY PROCEDURES

If a perinephric abscess is due to long-standing obstruction and there is no functioning renal tissue, a nephrectomy at the time of drainage is theoretically attractive. Drainage of a perinephric abscess should usually be performed as a primary procedure, however, with nephrectomy performed at a later date if necessary. Patients are frequently too ill for prolonged general anesthesia and surgical manipulation. Furthermore, nephrectomy is usually difficult to accomplish technically, and preoperative information is usually not sufficient to determine accurately the amount of functioning of salvageable renal tissue. After drainage of the abscess, removal of obstruction, and appropriate antimicrobial therapy, many kidneys may regain sufficient function to obviate future nephrectomy. Nephrectomy, if indicated, can be performed using a standard nephrectomy approach or a subcapsular nephrectomy technique outlined later.

A small renal abscess confined to one pole of the kidney may be managed by partial nephrectomy. If the infection extends beyond the apparent line of cleavage, however, it is essential to remove all infection, and the line of excision should extend through healthy tissue. If multiple abscesses are present, internal drainage is difficult, and nephrectomy may be required.

Subcapsular Nephrectomy

When a kidney is so adherent to surrounding tissues that dissection is difficult and hazardous, a subcapsular nephrectomy is indicated. These conditions are usually seen after multiple or chronic infections or previous operations have caused scarring to adjacent organs. Blunt dissection results in tearing of structures such as bowel wall. Sharp dissection when there is no definable tissue plane often results in lacerations of the vena cava, aorta, duodenum, spleen, and other structures. In subcapsular nephrectomy, dissection beneath the renal capsule enables one to avoid these vital structures. Subcapsular nephrectomy should not be performed for malignant disease and is undesirable in tuberculosis.

The main difficulty with subcapsular nephrectomy is that the capsule is adherent to the vessels in the hilum, and one usually must go outside the capsule to ligate the renal pedicle. In this setting, the renal hilum usually is involved in the inflammatory reaction, and separate identification of the vessels is difficult.

Kidney exposure is accomplished through the flank using a 12th rib incision. For low-lying kidneys, a subcostal incision may be satisfactory. When the kidney is reached, the capsule is incised and is freed from the underlying cortex. The capsule is stripped from the surface of the kidney, and an incision is made carefully in the capsule where it is attached to the hilum. The vessels may be protected by placing a finger in front of the pedicle when cutting the capsule. The dense apron of capsule can usually be incised best on the anterior aspect. Control of bleeding can be difficult in this procedure. Frequently all landmarks are obscured, and the renal artery and vein cannot be identified. Sharp dissection is usually required, and major vessels may be entered before they are recognized. Fortunately, the dense fibrous tissue tends to prevent their retraction. Frequently, several chromic suture ligatures can be placed through the pedicle between a proximally placed pedicle clamp and the kidney. To avoid damage to the duodenum or major vessels, pieces of capsule may be left behind. However, prolonged drainage can ensue, and as much of the infected tissue should be removed as possible. After ligation and cutting of the pedicle, the ureter is identified and cut, and the distal end is ligated. If distal ureteral obstruction has caused pyonephrosis, a small, 8 to 10 Fr red Robinson catheter may be placed in the distal ureter to allow postoperative antimicrobial irrigation. Multiple drains should be placed and brought through separate stab wounds.

OUTCOMES

Complications

Complications associated with percutaneous drainage include the formation of additional abscesses that communicate with the renal collecting system and may require temporary urinary diversion via percutaneous nephrostomy drainage to affect a cure. Sepsis, the most frequent complication of percutaneous drainage, occurs in fewer than 10% of patients. Other complications, such as transpleural puncture, vascular or enteric injury, and cutaneous fistula, are rare.

Additional complications to open or percutaneous drainage include prolonged purulent drainage, which may indicate a retained foreign body, calculus, or fistula.

Results

Cure rates for percutaneous drainage of renal and retroperitoneal abscesses range from 60% to 90%.8,15 Multiloculated, viscous abscesses and abscesses in immunocompromised hosts are associated with lower cure rates. Large abscesses may require more than one percutaneous access procedure to completely drain them.

In the past, mortality rates were reported to be as high as 50% in patients with retroperitoneal or perinephric abscesses. More recent reports indicate a significant improvement in mortality (approximately 10%), in large part because of more accurate diagnosis from improved imaging techniques, more effective antimicrobial therapy, and better supportive care.

SPECIAL CONSIDERATIONS

Renal Tuberculosis

Renal tuberculosis is caused by hematogenous dissemination from an infected source somewhere else in the body. Both kidneys are seeded with tuberculosis bacilli in 90% of cases. Clinically apparent renal tuberculosis is usually unilateral, however. The initial lesion involves the renal cortex, with multiple small granulomas in the glomeruli and in the juxtaglomerular regions. In untreated patients who fail to heal spontaneously, the lesions may progress slowly and remain asymptomatic for variable periods, usually 10 to 40 years. As the lesions progress, they produce areas of caseous necrosis and parenchymal cavitation. Large tumor-like parenchymal lesions or tuberculomas frequently have fibrous walls and resemble solid mass lesions. Once cavities form, spontaneous healing is rare, and destructive lesions result, with spread of the infection to the renal pelvis and development of a parenchymal or peri-nephric abscess.

Indications for Surgery

Surgery was once commonly used in the treatment of renal tuberculosis, but since the advent of effective antituberculosis chemotherapy, it is reserved primarily for management of local complications, such as ureteral strictures, or for treatment of nonfunctioning kidneys. If surgery is warranted, it is wise to precede the operation with at least 3 weeks and preferably 3 months of triple-drug chemotherapy. Use of isoniazid, 300 mg/day; pyrazinamide, 25 mg/kg to a maximum of 2 g, once daily; and rifampicin, 450 mg/day is recommended. If segmental renal damage is obvious and salvage of the kidney is possible, a drainage procedure or cavernostomy can be performed.7 Removal of a nonfunctioning kidney is usually indicated for advanced unilateral disease complicated by sepsis, hemorrhage, intractable pain, newly developed severe hypertension, suspicion of malignancy, inability to sterilize the urine with drugs alone, abscess formation with development of fistula or inability to have appropriate follow-up.

Alternative Therapy

Prophylactic removal of a nonfunctioning kidney to prevent complications, remove a potential source of viable organisms, and shorten the duration of convalescence and requirement for chemotherapy is advocated by some authors. Others, who followed a large series of patients treated with medical therapy alone, concluded that, because the frequency of late complications is only 6%, routine nephrectomy should not be performed for every nonfunctioning kidney. These authors, however, treated patients for at least 2 years. The merits of short-term therapy and prophylactic nephrectomy versus long-term 2-year chemotherapy and selective nephrectomy warrant further study. Modern percutaneous drainage techniques have largely replaced open cavernostomy for treatment of closed pyocalyx.

Surgical Technique

Cavernostomy

Renal tuberculosis sometimes results in caliceal infundibular scarring, causing a closed pyocalix. Unroofing of a pyocalix is called cavernostomy. If the calix still communicates with the renal pelvis, or if it is connected to significant functioning parenchyma, a cavernostomy should not be done because a urinary fistula or urinoma may result. To minimize wound contamination and tuberculous spread, thorough needle aspiration of purulent material and saline irrigation of the abscess cavity should be performed using a large-bore needle and syringe. The abscess cavity is then unroofed, and the edge is sutured with a running suture for hemostasis. Any unsuspected connection with the renal pelvis by an open infundibulum must be closed using 5-0 chromic catgut suture to prevent fistula or urinoma formation. After thorough wound irrigation, multiple drains are placed, and closure is undertaken. Drains are managed as previously described for perinephric abscess.

Nephrectomy

When unilateral tuberculosis causes more extensive parenchymal destruction or nonfunction, a partial or total nephrectomy, respectively, should be performed. For partial nephrectomy, a guillotine incision is made 1 cm beyond the abscess. If the renal pedicle can be freed and polar vessel located and occluded, the incision can be made at the line of demarcation of the ischemia. In partial nephrectomy, it is important to try to save the capsule (if it is not involved with the infection) to cover the raw surface for hemostasis. Alternatively, fat can be used for hemostasis. The amputated calyx is carefully ligated with a 4-0 chromic catgut suture to prevent urinary fistula or urinoma formation.

After nephrectomy, the distal ureter can be ligated and in most cases does not need to be brought to the skin because tuberculosis of the ureter generally heals with chemotherapy after nephrectomy. If renal tuberculosis is associated with severe tuberculosis cystitis, ureteral catheterization for 7 days postoperatively to minimize subsequent ureteral stump abscess formation should be considered.18

Renal Echinococcosis

Echinococcosis is a parasitic infection caused by the canine tapeworm E. granulosus. Echinococcal or hydatid cysts occur in the kidney in some 3% of patients with this disease. The hydatid cyst gradually develops at a rate of about 1 cm/year and is usually single and located in the cortex.

Diagnosis

The symptoms are those of a slowly growing tumor; most patients are asymptomatic or have a dull flank pain or hematuria. Excretory urography typically shows a thick-walled cystic mass, which is occasionally calcified. Ultrasonography and CT usually show a multicystic or multiloculated mass. Confirmation of the diagnosis is most reliably made by diagnostic tests using partially purified hydatid antigens in a double diffusion test. Complement fixation and hemagglutination are less reliable. Diagnostic needle puncture is associated with significant risk of anaphylaxis as a result of leakage of toxic cyst contents.

Indications for Surgery

Cyst removal is indicated when an enlarging cyst threatens renal function or produces obstruction.

Surgical Technique

The cyst should be removed without rupture to reduce the chance of seeding and recurrence. In cases where cyst removal is impossible because of its size or involvement of adjacent organs, marsupialization is required. The contents of the cyst initially should be aspirated, and the cyst should be filled with a scolecidal agent such as 30% sodium chloride, 2% formalin, or 1% iodide for about 5 minutes to kill the germinal portions. Complete evacuation of all hydatid tissue and thorough postmarsupialization irrigation are critical to preventing systemic effects. Penrose drains are left in the cystic cavity until drainage ceases. If large amounts of renal tissue have been damaged, partial or simple nephrectomy may be required.

Anatrophic nephrolithotomy is a procedure that has been used by urologists for nearly 30 years in the removal of large renal calculi, specifically branched or staghorn calculi. These stones are often associated with urinary tract infections, and the coexistence of these two conditions makes it difficult to eradicate either. Definitive treatment of these stones is generally advocated because of the significant morbidity and mortality associated with untreated staghorn calculi. Blandy and Singh found that patient survival is reduced with untreated staghorn calculi, with a mortality rate of 28% at 10 years.
Since the early 1980s, with the development of less invasive approaches such as extracorporeal shock wave lithotripsy and percutaneous nephrolithotomy, the role of anatrophic nephrolithotomy and other open stone operations has certainly diminished.2 However, anatrophic nephrolithotomy remains the gold standard for the treatment of staghorn calculi and thus maintains a role in the treatment of these large complex stones.
The original description of anatrophic nephrolithotomy was by Smith and Boyce in 1968. The operation they described was based on the principle of placing the nephrotomy incision through a plane of the kidney that was relatively avascular. This approach would avoid damage to the renal vasculature with resulting atrophy of the renal parenchyma, hence the term anatrophic. The operation also involves reconstruction of the intrarenal collection system to eliminate anatomic obstruction, thus improving urinary drainage, reducing the likelihood of urinary tract infection, and preventing recurrent stone formation.
DIAGNOSIS
The diagnosis of staghorn calculi is usually established in a similar fashion as other forms of urolithiasis. Patients may have the typical symptoms of flank pain, fever, and hematuria, or they may be asymptomatic. The diagnosis of chronic urinary tract infection is common in patients with these types of stones. Urine culture is often positive, and typical organisms include urea-splitting organisms such as Proteus, Klebsiella, Providencia, and Pseudomonas.
Useful radiographic studies traditionally include plain abdominal radiographs, nephrotomograms, and excretory urograms to identify the stones, the collecting system, and, if present, to define the degree of obstruction. Computed tomography can be helpful for detection of radiolucent or poorly calcified stones. Retrograde pyelography is usually performed in cases of equivocal findings on excretory urography. Nuclear renal scans can help to determine differential renal function when such information might affect the surgical approach. Renal arteriography is usually not indicated unless there is suspicion of anomalous arterial anatomy such as in renal fusion anomalies.
Before elective surgery, a metabolic evaluation is recommended to attempt to determine an etiology for stone formation and to aid in preventing a recurrence. For instance, it is important to determine the presence of hypercalciuria, hyperuricosuria, hyperoxaluria, cystinuria, hyperparathyroidism, and renal tubular acidosis. The measurement of serum and urine calcium, phosphorus, creatinine, uric acid, and electrolytes should be routine. A 24-hour urine collection for creatinine clearance as well as urinary calcium, phosphorus, oxalate, citrate, cystine, and uric acid is also an integral part of the workup.
INDICATIONS FOR SURGERY
Anatrophic nephrolithotomy should be performed for the removal of branched or staghorn calculi, usually complete staghorn stones, or those associated with infundibular stenosis or other intrarenal anatomic obstruction, for the combined goals of removing all calculi and open surgical correction of the anatomical obstruction. This procedure may also be preferred in the treatment of a staghorn calculus in a kidney with a small intrarenal pelvis, making access to the renal pelvis difficult, or in a patient who has undergone prior renal surgery to avoid a more risky renal sinus dissection. This operation is also indicated for the treatment of staghorn calculi in patients who would benefit from or prefer a single therapeutic procedure versus multiple, less invasive, procedures such as extracorporeal shock wave lithotripsy and/or percutaneous nephrolithotomy.
The goals of the procedure should be to remove all calculi and fragments, to improve urinary drainage of any obstructed intrarenal collecting system, to eradicate infection, to preserve and improve renal function, and to prevent stone recurrence.10
ALTERNATIVE TREATMENTS
Most staghorn calculi can now be preferentially treated with percutaneous nephrolithotomy, with or without extracorporeal shock wave lithotripsy. The stone-free rates reported are approaching comparability with traditional anatrophic nephrolithotomy, and there is probably an advantage to be gained in shorter convalescent periods following the less invasive methods. These alternative odalities can sometimes require multiple different procedures to accomplish a stone-free state. The American Urological Association Nephrolithiasis Clinical Guidelines Panel recommended as a guideline that initial percutaneous nephrolithotomy followed by extracorporeal shock-wave lithotripsy and/or further percutaneous procedures should be the treatment for most standard patients with staghorn calculi. Open surgery is recommended as an appropriate option in unusual cases when a stone is not expected to be removed with a reasonable number of the less-invasive procedures.8 Despite impressive advances with the less-invasive techniques, anatrophic nephrolithotomy remains a treatment option for large complete staghorn calculi or staghorn stones associated with anatomic obstruction and requiring open surgical correction.
SURGICAL TECHNIQUE
After administration of general anesthesia and placement of a Foley catheter, the patient is placed in the standard flank position with elevation of the kidney rest and flexion of the operating table to achieve adequate spacing between the lower costal margin and the iliac crest. Three-inch-wide adhesive tape applied at the shoulders and hips can be used to secure the patient to the table. Adequate padding should be used to protect pressure points.
A standard flank approach is used. The incision can be placed through the bed of either the 11th or 12th rib, depending on the estimated position of the kidney. If a previous flank incision has been made for renal surgery, it is preferable to place the incision above the old scar, ensuring that access to the kidney can be achieved through unscarred tissue. After rib resection, when access has been gained into the retroperitoneal space, Gerota’s fascia is identified overlying the kidney. Gerota’s fascia is incised in a cephalad–caudal direction, which facilitates returning the kidney to its fatty pouch at the end of the operation. The kidney is then fully mobilized, and the perinephric fat is carefully dissected off the renal capsule with care taken not to disrupt the renal capsule. Should the capsule become inadvertently incised, it can be closed at that time with chromic catgut sutures. The kidney is now free to be suspended in the operative field by utilizing two 1-inch umbilical tapes as slings. At this point a preliminary portable plain radiograph can be obtained.
The renal hilar dissection is the next step. The main renal artery and its branches are carefully dissected and identified. The avascular plane, or Brodel’s line, can be identified by temporarily clamping the posterior segmental artery and injecting 20 ml of methylene blue intravenously, which results in the blanching of the posterior renal segment while the anterior portion turns blue5 . This allows identification of this avascular plane. Placing the nephrotomy incision through this plane will achieve maximal renal parenchymal preservation and minimize blood loss. The avascular plane can also be identified with the use of a Doppler stethoscope to localize the area of the kidney with minimal blood flow.
Extensive renal hilar dissection can be avoided by utilizing a modification of the original procedure described by Smith and Boyce. Redman and associates relied on the relatively constant segmental renal vascular supply in advocating placing the incision at the expected location of the avascular line after clamping the renal pedicle with a Satinsky clamp, seeking to prevent vasospasm of the renal artery and warm ischemia. This modification can be time-saving and spare extensive dissection of the renal hilum. However, we continue to advocate precise identification of the avascular plane to minimize parenchymal loss.
At this point, 25 g of intravenous mannitol is administered. This promotes a postischemic diuresis and prevents the formation of intratubular ice crystals by increasing the osmolarity of the glomerular filtrate. The main renal artery can now be occluded with a noncrushing bulldog vascular clamp. A bowel bag or barrier drape is placed around the kidney, and hypothermia is initiated with the placement of iced saline slush within the barrier surrounding the kidney. Dry laparotomy sponges are used to pack away the peritoneal contents and to protect and insulate them from hypothermia. The kidney should be cooled for 10 to 20 minutes before the nephrotomy incision is made. This should allow achievement of a core temperature in the 5° to 20°C range, which will allow safe ischemic times from 60 to 75 minutes and minimize renal parenchymal damage.

The renal capsule is then incised sharply over the previously identified avascular plane, and the renal parenchyma can be bluntly dissected with the back of the scalpel handle. Blunt dissection minimizes injury to the intrarenal arteries that are traversed. Small bleeding vessels can be controlled with 5-0 or 6-0 chromic catgut suture ligature. If renal back bleeding continues to be a problem despite these measures, the main renal vein can be occluded. As the incision is angled toward the midportion of the renal hilum, the nephrotomy should theoretically remain close to the avascular plane.
As the nephrotomy incision proceeds toward the renal hilum, the ideal location to enter the collecting system is at the base of the posterior infundibula. Occasionally, with large posterior calyceal calculi, a dilated posterior calyx will be entered initially. The remainder of the collecting system can then be identified and opened with a probe or stone forceps. If a posterior infundibulum is entered first, the incision is then carried toward the renal pelvis. The stone is palpated, and the remainder of the infundibula are incised in a similar fashion. Attention is then turned towards the anterior infundibula. All of the calyceal extensions should be identified and incised. In order to minimize stone fragment formation and retained calculi, the stone should not be manipulated or removed until all of the calyceal and infundibular extensions are appropriately identified and incised, allowing for complete visualization and mobilization of the collecting system and calculi. Ideally, the stone or stones should be removed without fragmentation; however, often it is inevitable that there will be some piecemeal extraction. After removal of all stone fragments, the renal pelvis and calyces are copiously irrigated with cold saline and carefully inspected for retained fragments. A plain radiograph is obtained at this point to rule out residual calculi or fragments.
At this time, a “double-J” stent is passed from the renal pelvis into the bladder. The routine use of internal ureteral catheters is encouraged. They provide good urinary drainage and protect the freshly reapproximated collecting system and minimize postoperative urinary extravasation. The stents also prevent intraoperative migration of smaller calculi into the ureter.
The next step in the procedure is the reconstruction of the intrarenal collecting system. Infundibular stenosis or stricture, which results in obstruction promoting urinary stasis and recurrent stone formation, should be corrected with calyorrhaphy or calycoplasty. The former is the repair of a single narrowed calyx, achieved by incising the calyx along its appropriate margin (anterior margin for posterior calyces and posterior margin for anterior calyces) and suturing those margins to the renal pelvis, resulting in a shorter, wider calyx. The infundibulum can also be incised longitudinally and then closed transversely in a Heinecke–Mickulicz fashion. Calycoplasty is the repair of adjacent stenotic calyces by suturing the adjacent walls of the neighboring calyces, thus forming a single structure. All intrarenal reconstructive suturing should be accomplished with 5-0 or 6-0 chromic catgut sutures. When suturing the mucosal edges, it is important to avoid incorporation of underlying interlobular arteries, thus preventing ischemia.
The renal pelvis is then closed with a running 6-0 chromic catgut suture. The renal capsule is closed with a running 4-0 chromic suture. The use of mattress-type parenchymal sutures can lead to tissue ischemia and should be avoided if possible. One should inspect closely for further parenchymal bleeding points and ensure good hemostasis before closing the renal capsule. After the capsule is closed and hemostasis has been achieved, the slush surrounding the kidney is removed, and the renal artery unclamped. The kidney is observed for good hemostasis and return of pink color and good turgor after unclamping. The kidney is then returned into Gerota’s fascia, and the kidney and proximal ureter are covered with some perirenal fat to minimize the postoperative scar formation. If Gerota’s fascia is unavailable because of prior surgery, omentum can be mobilized through a peritoneal opening and wrapped around these structures. The peritoneal opening should be sutured to the omentum to prevent herniation of the abdominal viscera.
A Penrose or suction-type drain is placed within Gerota’s fascia and brought out through a separate stab incision. This drain is left in place until minimal drainage occurs, usually by the third or fourth postoperative day. Nephrostomy tubes are generally avoided because of their potential for causing infection or further renal damage. The flank musculature and skin are closed in the standard fashion.
Postoperative management after anatrophic nephrolithotomy should follow the same principles that guide management after other major operations. Intravenous fluids are maintained to achieve brisk urine output and until the patient is able to tolerate a clear liquid diet. Broad-spectrum intravenous antibiotics are administered perioperatively and continued postoperatively. Antibiotic coverage is guided by preoperative urine culture and sensitivity results. The patient is usually converted to appropriate oral antibiotics and maintained on a 14-day course. The ureteral stent is removed cystoscopically at approximately 6 weeks after the operation in uncomplicated cases.
OUTCOMES
Complications
Pulmonary complications are perhaps the most common following anatrophic nephrolithotomy, especially atelectasis. Patients with a history of pulmonary disease should probably undergo preoperative evaluation with pulmonary function testing and initiation of vigorous pulmonary toilet prior to surgery. Postoperatively, patients should be encouraged to breathe deeply, and use of an incentive spirometer should be routine. Early ambulation will also be beneficial.
Pneumothorax should occur in fewer than 5% of patients.10 Inadvertent opening of the pleura, usually during incision and resection of a rib, should be readily identified intraoperatively. The defect should be closed immediately with a running chromic catgut suture. The lung is hyperinflated just before the final suture is placed to ensure reexpansion of the lung. Chest tubes are not routinely used but may be necessary if any question remains regarding the reliability of the pleural closure. A chest radiograph should be obtained in the recovery room for any patient who undergoes repair of a pleural defect.
Pulmonary embolism remains a potential complication of any major surgery. Routine use of elastic support hose and sequential-compression stockings can lower the risk of deep venous thrombosis. Encouragement of early ambulation is also an important preventative measure.
Significant postoperative renal hemorrhage should occur in fewer than 10% of patients. Assimos and associates reported an incidence of 6.4%.1 Bleeding usually occurs immediately or about a week postoperatively. Extensive intrarenal reconstruction, older age, worse renal function, and presence of blood dyscrasias were found to be significant risk factors. Slow bleeding will usually resolve on its own; management includes correction of any bleeding abnormalities and replacement with blood products as necessary. Oral e-aminocaproic acid can be successful in certain cases. Bleeding that is brisk or cannot be adequately treated conservatively will require a more aggressive approach. A renal arteriogram can help identify the lesion, and an attempt at arteriographic embolization can be considered. Reexploration may be required in the remainder of the cases, with reinstitution of hypothermia and suture ligation of the bleeding vessel(s). Persistent hematuria 1 to 4 weeks postoperatively should alert the clinician to the possibility of renal arteriovenous fistula formation.1
Stone recurrence rates following anatrophic nephrolithotomy have been reported from 5% to 30%.10 Inspection, intraoperative plain radiographs, intraoperative ultrasound, and nephroscopy can all aid in the identification and treatment of retained calculi. Recurrent calculi usually form in those with persistent urinary tract infections, persistent urinary drainage impairment, and those with previously unidentified or refractory metabolic disturbances.
Urinary drainage or extravasation should occur infrequently with the routine use of perinephric drains and internal ureteral catheter drainage. Should drainage recur or persist following removal of the drain and/or ureteral stent, replacement of the ureteral stent should be considered to decompress the system and relieve any obstruction.
Results
When performed for appropriate indications and with meticulous technique, anatrophic nephrolithotomy can achieve successful removal of all calculi, preservation of renal function, improved urinary drainage, and eradication of infection. Stone-free rates greater than 90% should be achieved. We believe that for large complex staghorn calculi and those associated with some anatomic abnormality leading to impaired urinary drainage, anatrophic nephrolithotomy remains superior to percutaneous nephrolithotomy or combination therapy with respect to both stone-free rates and the achievement of a stone-free state with a single operative procedure. In the long term, treatment of these staghorn calculi with anatrophic nephrolithotomy should preserve renal function in the involved kidney and, in a majority of patients, eradicate stone disease and chronic urinary infection.

Although the true incidence of renovascular hypertension is unknown, it is estimated that between 5% and 10% of all hypertensive patients suffer from renovascular hypertension.4 During the past two decades, there have been dramatic changes in the diagnosis and treatment of renovascular hypertension. There is clearly a better understanding of the renin–angiotensin system, and newer, more potent antihypertensive medications are available. In addition, newer diagnostic radiologic procedures, such as digital subtraction angiography, captopril renal scans, balloon angioplasty, and newer surgical techniques, have dramatically changed the ways in which we diagnose and treat renovascular hypertension today.
DIAGNOSIS
In the past, the clinician’s task of identifying potentially curable patients in a safe, cost-effective, and reliable manner was difficult. Recently we have been given the means to reliably identify patients with a physiologically significant renal artery stenosis that, in the past, might have eluded the physician. A single-dose captopril test is reported by some investigators to be a reliable screening test and is well suited for outpatient use. Although it is less reliable in patients with a degree of renal insufficiency, the peripheral plasma renin response to a single dose of oral captopril has proved to be a simple and sensitive test.7 In patients with a functional renal artery stenosis, ACE inhibitors lead to a disproportionate increase in peripheral plasma renin activity as a result of the disappearance of the inhibitory effect of angiotensin II on renin secretion. This phenomenon can be used diagnostically to detect unilateral renal artery stenosis. In cases of bilateral stenosis, however, the test cannot be used reliably as an indicator of renal artery stenosis.1 However, Postma et al.8 have reported that the captopril test is not a reliable screening test, and as a consequence, the value of this procedure as a screening test in detection of renal artery disease is at present unsettled.
Another test that has recently been analyzed for the detection of renal artery stenosis is the renal scintigram with isotopic nephrography following the administration of captopril.9 In a functional renal artery stenosis when ACE is inhibited, the glomerular infiltration rate decreases as a consequence of the decreased inhibition of angiotensin II on the vasa efferens. This can be demonstrated by technetium DTPA scintography. The sensitivity of this means of identifying renal artery stenosis varies from 71% to 92% with a specificity of 72% to 97%. In the hipuran scintigram, the patient with renal artery stenosis showed a continual enrichment of the isotope in the renal cortex, probably related to a decreased excretion of hipuran because of the ACE inhibitor.
Digital venous subtraction angiography (DSA), in my opinion, remains the most definitive and reliable screening test available. It has a sensitivity and specificity of nearly 90%. Intravenous DSA is susceptible to artifacts such as crossing vessels and from intestinal motility. In these patients, digital arterial subtraction angiography achieves an excellent view of the renal circulation using less contrast dye than conventional angiography. Because this method uses a comparatively small 4- to 5-Fr catheter, groin hematoma is rare, and the technique can be used in an outpatient setting.
Several other new modalities have appeared recently. Duplex Doppler ultrasound scanning is now a recognized way of demonstrating and locating focal renal artery stenosis. This noninvasive test is gaining acceptance in the diagnosis of both atherosclerosis and fi-bromuscular hyperplasia. In a recent study, Ferdinandi et al.3 report the ability to detect renal artery stenosis with a success rate of over 90%. Further studies will be necessary to determine the ultimate role of duplex sonography and color Doppler evaluation of renal ar-tery stenosis.
At present, digital subtraction angiography in conjunction with divided renal vein renin assays, which demonstrate contralateral suppression, remains the best screening test and predictor of treatment outcome.
In those patients who have azotemia, in whom contrast is contraindicated, MRI angiography and CO2 digital subtraction angiography are useful12 These studies avoid the use of contrast and obviate the occurrence of contrast toxicity.
INDICATIONS FOR SURGERY
The surgical management of renovascular hypertension has changed dramatically in the last two decades. In the early 1970s, we demonstrated that renal function could be preserved or restored by renal revascularization of nonfunctioning kidneys with totally occluded renal arteries. This contribution led to the notion that if patients who had totally occluded renal arteries and nonfunctioning kidneys could have restoration of renal function, then we could treat patients with renal artery stenosis who had azotemia with the expectation that they could also have preservation or improvement of renal function.5 Progressive azotemia in the elderly atherosclerotic patient population is now one of the indications for renal revascularization. Secondly, based on our observations, the use of alternative bypass procedures has significantly reduced the morbidity and mortality of high-risk patients undergoing renovascular surgery. The use of the hepatic artery, the gastroduodenal artery, and other alternative procedures instead of the aortorenal saphenous vein bypass graft has not only reduced the morbidity and mortality of surgery but, in doing so, has dramatically changed the nature of our patient population.
ALTERNATIVE THERAPY
Converting enzyme inhibitors prevent the conversion of angiotensin I to angiotensin II. These drugs in conjunction with calcium channel blockers have greatly improved the medical management of patients who suffer from renovascular hypertension. Unfortunately, even if adequate blood pressure control is maintained by pharmacologic means, progression of renal artery disease is not prevented, and renal ischemia and renal damage may clearly progress.11
When medical management fails or azotemia progresses, then balloon angioplasty and surgical treatment must be considered. The choice between angioplasty and surgery relies on a well-defined set of criteria established by published results. Angioplasty is indicated in the treatment of fibrous dysplasia and atherosclerosis of the mid–main renal artery. Surgery is indicated for the treatment of osteal atherosclerosis and for branch lesions of the renal artery. Renal artery aneurysms are a different problem. Surgery is indicated when they are the cause of hypertension. Also aneurysms larger than 2 cm in diameter that are noncalcified, especially in gestational women, should be repaired, as they are prone to rupture during pregnancy.
It is our feeling that patients who develop recurrent disease following balloon angioplasty are probably best subjected to surgical management, as repeat balloon angioplasty is associated with a significant complication rate. Use of the thoracic aorta may also be a viable alternative on the left side because the thoracic aorta is usually less atherosclerotic than the abdominal aorta.6
SURGICAL TECHNIQUES
Aortorenal Bypass Graft
The widespread popularity of the bypass graft for renal artery disease was attained by virtue of its technical ease of insertion and the favorable short- and long-term patency rates achieved. Bypass grafts are applicable to almost any disease process involving the main renal artery or its branches. This procedure also eliminates the more hazardous and tedious dissection of the juxtrarenal portion of the aorta required in endarterectomy. Bypass grafts are particularly suitable for fibrous lesions that affect long and multiple segments of the renal artery and its branches (Fig. 10-1). Dacron, autogenous artery (hypogastric and splenic), and autogenous saphenous vein may be chosen as aortorenal bypass grafts in properly selected patients.
Dacron has been applied extensively in renal artery reconstruction but has been associated with a relatively high rate of early thrombosis. Excellent long-term patency rates have been reported with a segment of autogenous hypogastric artery. Such a graft matches the size of the renal artery and is sutured more simply than the Dacron prosthesis.
Autogenous hypogastric artery is the most favorable graft material for children with renal artery disease because the saphenous vein is usually too small and is more prone to aneurysmal dilation than in adults. The major disadvantage is that the hypogastric artery is often the first to be involved with generalized atherosclerosis and therefore is not suitable graft material in older patients. It is also a short vessel and occasionally is technically more difficult to insert between the renal arteries and aorta.
During the past two decades the autologous saphenous vein has emerged as our preferred graft material and is the most common source for restoration of renal blood flow at our hospital. Saphenous vein is readily available and closer in size to the lumen of the renal artery than other vascular conduits. Its intima is less thrombogenic than prosthetic material and accommodates the creation of a precise contoured anastomosis with a delicate thin-walled distal renal artery. Patent anastomoses can be achieved with the most challenging 2- to 3-mm-lumen branches beyond the major bifurcation. Because of its inherent properties and the favorable surgical results obtained, saphenous vein has become the conduit of choice for aortorenal bypass at most major renovascular centers. If the saphenous veins are not available, we use cephalic vein and Gore-Tex graft, in that order, as substitutes.
Procurement of Saphenous Vein
The procurement of an adequate segment of the long saphenous vein is critical to the success of the graft procedure. Meticulous technique in exposure and excision of the vein is essential to prevent mural trauma and ischemia. Improper harvesting of the vein may result in the delayed complications of stenosing intimal hyperplasia and aneurysmal dilation. Removal of the saphenous vein should be performed by an experienced surgeon.
The saphenous vein is usually obtained from the thigh opposite the renal lesion so that two surgeons may simultaneously expose the renal vessels and mobilize the graft, shortening the operative time. The vein is mobilized through a single long incision in the upper thigh (Fig. 10-2), which begins parallel to and below the groin crease over the palpable femoral pulses and is extended toward the knee after the junction of the saphenous and femoral veins has been exposed. The incision should be made directly over the vein to avoid producing devascularized skin flaps that can result in necrotic edges and wound sepsis. Finger dissection between the trunk of the vein and the skin is helpful to ensure accurate placement of the incision and, thus, to avoid development of these flaps (see Fig. 10-2). On the day before operation, the course of the saphenous vein is outlined with an indelible pen while the patient is standing.
A 20-cm-long vein graft with an outside diameter of 4 to 6 mm is usually adequate for reconstruction of the renal artery. Excess vein should always be available for revision of any intraoperative technical problems that may occur during anastomosis. The vein is handled gently without stretching or tearing its branches. The tributaries are tied in continuity with fine silk before they are divided. The areolar tissue is dissected from the specimen, and the adventitia is left undisturbed.
To decrease transmural ischemia, the vein graft remains4 in situ until the renal vessels are mobilized and it is ready to be used. If the graft is inadvertently removed prematurely, it is placed in cold Ringer’s lactate solution or autologous blood, even if only a short period of time will ensue. The distal end of the vein is transected, cannulated with a Marks needle, and secured with a silk tie (Fig. 10-3). A dilute heparinized solution of autologous blood distends the vein graft before the proximal is transected. This step helps to identify any untied tributaries or unrecognized leakage and washes out any residual blood clots. The vein is distended to a minimal diameter of 5 to 6 mm by exerting gentle pressure on the syringe. The proximal end of the vein is transected, and the vein graft is now ready for use. The thigh incision is not closed until the bypass procedure has been completed to ensure that any delayed bleeding caused by the heparinized state is identified and controlled.
Technique of Insertion of Saphenous Vein Graft
Heparin is initially given systemically after the surgical dissection has been completed and approximately 30 minutes before the arteries are clamped. The saphenous vein graft should be oriented properly to avoid misalignment during implantation. Either an end-to-end or an end-to-side anastomosis can be accomplished, depending on the anatomic situation encountered. An end-to-end anastomosis is preferred under usual circumstances because it permits the best laminar flow.
The aorta, which has already been mobilized and exposed from the renal arteries to the level of the inferior mesenteric artery, is carefully palpated to determine a suitable soft location for the anastomosis that is relatively free of atherosclerotic plaque. A medium-sized DeBakey clamp is placed on the anterolateral portion of the infrarenal aorta in a tangential manner. A vertical 13- to 16-mm aortotomy is made without excising any of the aortic wall or attempting to perform a localized endarterectomy (Fig. 10-4), which may dislodge intimal plaque fragments that can form emboli to the lower extremities when the clamp is released.
Excision of the aortic wall is not necessary because intraluminal aortic pressure spreads the edge of the linear aortotomy to the appropriate dimensions when the clamp is released. The vein graft is anastomosed to the aorta with continuous 5-0 Proline suture after it has been satisfactorily spatulated (Fig. 10-5). A microvascular Schwartz clamp is placed on the end of the saphenous vein graft, and the aortic clamp is released. The graft is allowed to lie anterior to the vena cava on the right side or anterior to the renal vein on the left side. Although it is preferable to leave the vein too long than too short, it should not be so long as to bend into an acute angle at any point. The renal artery is secured distally with a smooth-jawed Schwartz microvascular clamp placed on either the distal main renal artery or its branches. The proper site for the arterial anastomosis is selected. An end-to-end anastomosis is performed utilizing a continuous 6-0 Proline suture or interrupted sutures of the same material, depending on the diameter of the anastomosis (Fig. 10-6). When the saphenous vein graft is being anastomosed with two branches 3 mm or less in size, interrupted sutures are chosen. An interrupted suture line is also selected in children to prevent a pursestring effect with growth of the vessels when the patients become older. This effect may also occur with running synthetic monofilament sutures when too much tension is applied during the creation of the anastomosis. The pursestring effect can be avoided by placing sutures at four quadrants in the arterial wall before beginning the anastomosis. Operating loupe magnification and fiberoptic headlamps are very helpful at this point in the operation to allow precise placement of the sutures, particularly when exposure in the renal artery is difficult.
Alternative Arterial Bypass Grafts
Extensive atherosclerosis, previous aortic surgery, and complete thrombosis of the aorta may preclude the use of the aortorenal bypass procedure for renal artery reconstruction. When the surgeon is treating a patient with stenosis of the right renal artery in association with these pathologic limitations on the aorta, a splenorenal or hepatic-to-renal artery saphenous vein bypass or gastroduodenal-to-renal artery bypass procedure can be selected.
Splenorenal Arterial Bypass
Splenorenal arterial bypass has many desirable features as a substitute for aortorenal bypass in patients with stenosis of the left renal artery. It is particularly suitable for patients who have diffuse atherosclerotic disease or thrombosis of the aortic lumen and for those who have previously undergone difficult aortic reconstructions.
The splenic artery has the advantages of being an autogenous artery that has not been separated from its nutrient vaso vasorum, of being exposed without difficulty by a relatively uncomplicated anatomic dissection, and of requiring only one vascular anastomosis. Carefully monitored oblique and lateral angiography of the celiac axis is required to determine the patency of this artery because atherosclerosis can affect the arterial lumen early in the patient’s life. Surgical exploration and intraoperative evaluation by palpation and measurement of splenic blood flow are also helpful in establishing its suitability for renal revascularization. If the blood flow is less than 125 ml/min, the splenic artery should probably not be utilized for renal artery bypass.
We now prefer to expose the splenic artery through a supracostal 11th-rib flank incision (Fig. 10-8). The dissection is continued along the upper border of the rib. The overlying latissimus dorsi, the serratus posterior inferior, and the intercostal muscles are divided. Division of the intercostal ligament permits the rib to move freely. The external, internal oblique, and transversus abdominis muscles are divided, and the intercostal muscle attachments on the distal 1 inch of the rib are divided carefully until the corresponding intercostal nerve is identified. The investing fascia around the nerve is entered. Dissection in this plane allows an extrapleural approach and generally avoids entry into the pleural cavity. This approach also allows excellent exposure, for the ribs are free to pivot downward in a “bucket-handle” fashion (Fig. 10-9).
The plane between Gerota’s fascia and the adrenal gland posteriorly and the pancreas anteriorly is entered. The splenic artery is identified at the upper border of the pancreas. Its enveloping fascia is entered, and the splenic artery is mobilized by a purely retroperitoneal approach. Several small pancreatic branches are identified, isolated, ligated, and divided. The splenic artery can usually be mobilized from the splenic hilum to the celiac axis without difficulty, and it provides sufficient length to reach the left renal artery.
After the splenic artery is mobilized, a sponge soaked with papaverine is placed on it to permit it to dilate. The artery is divided just proximal to its primary bifurcation in the hilum of the spleen, after a suitable vascular clamp has been applied to the origin of the artery. If necessary, the artery may be dilated with a Gruntzig balloon or Fogarty catheter intraoperatively to obtain maximum caliber. Removal of the spleen is not necessary because it continues to receive adequate blood flow from the short gastric arteries. The left kidney is approached posteriorly, and the left renal artery is identified and mobilized (Fig. 10-10). The renal artery is ligated at the aorta, and an end-to-end anastomosis between the splenic artery and the distal renal artery is carried out using continuous or interrupted 6-0 Proline sutures (Fig. 10-10). We have employed this approach in nearly 100 patients and now prefer it to the traditional transabdominal technique.

Splenic artery disease, the risk of pancreatitis, and the formation of a pancreatic pseudocyst are some of the limitations that have restricted splenorenal bypass as a routine procedure in the management of disease of the left renal artery.
Hepatorenal Bypass Graft
Arising from the celiac axis and continuing along the upper border of the pancreas, the hepatic artery reaches the portal vein and divides into an ascending and a descending limb. The ascending limb is a continuation of the main hepatic artery upward within the lesser omentum; it lies in front of the portal vein and to the left of the biliary tree. The descending limb forms the gastroduodenal artery. In the porta hepatis, the hepatic artery ends by dividing into the right and left hepatic branches, which supply the corresponding lobes of the liver (Fig. 10-12). The anatomic variations in the hepatic circulation must be appreciated before this procedure can be utilized. The right hepatic artery is more variable than the left. It may be anterior (24% of patients) or posterior (64% of patients) to the common bile duct, and in 12%, this artery arises from the superior mesenteric artery (Fig. 10-13). The hepatic artery lies anterior (91% of patients) or posterior (9% of patients) to the portal vein. In addition, the left hepatic artery arises from the left gastric artery in 11.5% of patients.
Careful dissection of the porta hepatis is essential, and the common hepatic, gastroduodenal, and right and left hepatic arteries should be identified before an anastomotic procedure is attempted. Vascular elastic loops are placed about these vessels, and the common bile duct and portal vein are identified.
After careful dissection and mobilization of the renal artery, clamps are placed on the proximal portion of the common hepatic artery and its distal branches. The gastroduodenal artery is divided (Fig. 10-14). The inferior surface of the hepatic artery is mobilized from the underlying portal vein and the common bile duct. An arteriotomy, 10 to 12 mm in length, is made in the anterior inferior wall of the common hepatic artery, beginning at the ostium of the gastroduodenal artery. A reversed autogenous saphenous vein is inserted with an end-to-side anastomosis between the vein graft and the hepatic artery. This maneuver is usually accomplished with a continuous 6-0 Proline suture. A microvascular clamp is placed on the vein graft after it has been filled with heparin and after the proper alignment and length for the renal artery anastomosis has been determined. The clamps are removed from the hepatic circulation, and a small Schwartz microvascular clamp is placed on the distal renal artery. The vein graft is anastomosed to the right renal artery in an end-to-end fashion. When the gastroduodenal artery is used, it is divided, and an end-to-end anastomosis between the gastroduodenal artery and the renal artery is accomplished.
We have employed this procedure in approximately 50 patients with good results. Postoperative angiography has demonstrated the absence of a renal–hepatic steal syndrome. Liver function has not been compromised in any of our patients to date. We no longer advocate the use of the gastroduodenal artery in adult patients, but it is a perfectly acceptable bypass procedure in the pediatric patients.
We have also utilized the superior mesenteric-to-renal artery saphenous vein bypass as a “bailout procedure” as well with good results (Fig. 10-15). An iliac-to-renal bypass graft has been done as an alternative to the aortorenal bypass procedure in ten of our patients, with favorable results (Fig. 10-16).
Renal Autotransplantation and Ex Vivo Bench Surgery
On rare occasions, kidneys with lesions of the renal artery or its branches are not amenable to in situ reconstruction. In these circumstances, temporary removal of the kidney, ex vivo preservation, microvascular repair (bench surgery), and autotransplantation may permit salvage.
Autotransplantation developed as an outgrowth of the technique in renal transplantation. The simultaneous development of an apparatus that could preserve kidneys extracorporeally for long periods of time and of preservation solutions also led to the technique of extracorporeal renal repair.
Autotransplantation and ex vivo repair should be considered in patients with traumatic arterial injuries, when disease of the major vessels extends beyond the bifurcation of the main renal artery into the segmental branches, and when multiple vessels supplying the affected kidney are involved. Bench surgery may also be required in patients who have very large aneurysms, arteriovenous fistulas, or malformations (Fig. 10-17).
Other indications for autotransplantation that usually do not require ex vivo repair include abdominal aortic aneurysms that involve the origin of the renal arteries and extensive atheromatous aortic disease, when an operation on the aorta itself may prove hazardous. In the last case, the patients usually have extensive internal iliac artery disease that precludes utilization of this artery for autotransplantation. However, we have noted that in these instances, the external iliac artery is spared extensive atherosclerosis and is suitable for autotransplantation, with an end-to-side renal artery anastomosis or an iliac-to-renal bypass graft.
Techniques for Autotransplantation
Autotransplantation can be accomplished through a large single midline incision or two separate flank and iliac fossa incisions. When the kidney is removed, care is taken to preserve the maximum length of renal vessels and ureter. If the transabdominal approach is selected, ureteral continuity can be retained, necessitating only vascular anastomosis afer the kidney is flipped over. If ex vivo surgery requires transection of the ureter, ureteroneocystostomy is necessary in addition to vascular anastomosis.
When ureteral continuity is preserved, autotransplantation is performed as illustrated in Fig. 10-18. When the kidney has been excised completely, the standard techniques for renal homotransplantation are used.
During dissection of the iliac vessels, meticulous care is taken to ligate the lymphatics in this area to prevent the development of a lymphocele. The external iliac vein is freed to the point where it is crossed by the internal iliac artery (Fig. 10-19A). The renal vein is anastomosed end-to-side to the external iliac vein using 5-0 Proline sutures (Fig. 10-19B). If the renal artery is free of atherosclerotic disease, it is then anastomosed end-to-end to the internal iliac artery, employing 6-0 Proline sutures (Fig. 10-19C,D). If the internal iliac artery is diseased, the renal artery is anastomosed end-to-side to the external iliac artery.
When the ureter requires reimplantation, we prefer a modification of the Politano–Leadbetter ureteroneocystostomy. Saline solution, 2 to 3 ml, is injected submucosally, raising a mucosal bleb (Fig. 10-20A). A small segment of mucosa is removed from the inferior portion of the bleb (Fig. 10-20B). A right-angle clamp is inserted into this opening, and a 3-cm-long submucosal tunnel is created (Fig. 10-21A). At the apex of the tunnel, the right-angle clamp is rotated 180 degrees to pierce the detrusor muscle. The ureter is brought to lie in the submucosal tunnel (Fig. 10-21B). The distal ureter is cut at a 45-degree angle, and the ureter is anastomosed to the bladder with interrupted 4-0 or 5-0 Dexon or Vicryl sutures (Fig. 10-21C,D).
In the future, the use of endovascular prostheses in maintaining the effect of luminal balloon dilatation is a very promising technique if long-term evaluation confirms the preliminary results.
OUTCOMES
Complications
Complications of renal vascular surgery can be classified as early or delayed. Early complications include bleeding, thrombosis of the artery, embolization of the branch vessels, subintimal dissection, false aneurysms, and loss of the kidney. Postoperative bleeding requiring operative intervention is to some extent a technical failure but may also be a function of the structural integrity of the arterial wall in diseased segments. It is important to recognize the enlarged perihilar vessels that are seen in high-grade stenosis and to be cognizant of the adrenal venous channels. Delayed bleeding may be the result of false aneurysm formation or erosion of the graft anastomosis into the duodenum or other bowel.
Renal artery thrombosis is the most common complication of renal vascular reconstruction and is most common after either placement of a dacron graft or endarterectomy. Predisposing factors include small dacron grafts, renal atrophy associated with thin-walled diseased arteries and high intrarenal vascular resistance, hypotension, or hypovolemia. It has been shown that the thrombosis of both venous and synthetic grafts is partially affected by the adequacy of the peripheral runoff as well as the adequacy of resection of the endothelial plaques in atheroschlerotic vessels. Embolization of plaque to the distal extremities or aortic thrombosis is rare.
Results
Balloon angioplasty is primarily used to treat patients with mural dysplasia.2,10 This modality of treatment is 80% to 85% effective in the management of these patients at our institution. Balloon angioplasty has a very limited role in the management of atherosclerotic renovascular disease at our institution.
The combination of balloon angioplasty for the younger, healthier patients suffering from mural dysplasia, the advent of alternative bypass procedures, and the concept of revascularization for preservation and restoration of renal function have dramatically changed the nature of the patient population being referred to our institution for renal revascularization. We are now frequently being called on to revascularize more elderly, higher-risk patients with diffuse atherosclerosis who have failed aggressive antihypertensive therapy in order to improve their renal function. We have recently reported a series of more than 100 patients who have undergone renal revascularization for preservation and restoration of renal function with an 85% success rate in this very high-risk patient population.5
CONCLUSION
Renovascular disease is a rapidly changing clinical entity. Considerable progress is being made in screening and diagnosis, primarily as a result of the development of less- or noninvasive studies. Renovascular disease is a potentially curable cause of hypertension and one of the few curable or preventable causes of renal failure. The indications for angioplasty and surgical revascularization are better known, and interventional therapy is justified when anesthesia and surgery represent an acceptable risk. We look forward to the short- and long-term results of renal artery atherectomy and wall stenting with anticipation.

Success rates for kidney transplantation continue to improve. Graft survival at 1 year is now 83% for all cadaver transplants done in the United States and 93% for living related transplants.9 Some programs report greater than 90% survival even for cadaver grafts, and greater than 85% survival at 2 years.1 Despite these good results, they are not yet 100%, so a percentage of patients transplanted will fail. The fate of the kidney after transplant failure is the subject of this chapter.
DIAGNOSIS
The diagnosis of a failed renal transplant is not difficult because the patients are azotemic and require dialysis. The common causes of graft failure are technical, such as arterial or venous thrombosis, and rejection, acute or chronic. Irreversible rejection can occur despite optimum immune suppression. Sometimes rejection results from the necessity to reduce or stop immune suppression to treat a life-threatening infection. Acute rejections occurring more than 1 year after the transplant are often a result of patient noncompliance with drugs. In any of these situations, patients will usually have undergone some radiologic investigation such as Doppler ultrasound or renal scan to ascertain whether there was a treatable cause for graft dysfunction. Renal biopsy is often performed once it is clear that there is no technical cause of graft dysfunction to determine the degree of acute or chronic rejection present. Once it is certain that renal failure is irreversible, immune suppression is withdrawn, chronic dialysis is reinstituted, and a decision is made as to whether or not it is necessary to remove the allograft.
INDICATIONS FOR SURGERY
Grafts failing within the first 12 months are almost always removed prophylactically regardless of the cause of graft loss and whether or not there are specific symptoms present. This is because if the graft is left in place, significant symptoms necessitating its removal will almost always occur.3 Most of these are either technical failures, which occur in the first few weeks after transplant, or acute rejection, which is manifest in the first 2 to 3 months. Those grafts failing after 12 months are left in place initially, immunosuppression is stopped, and patients are followed. Approximately 50% of these patients will develop symptoms consistent with acute rejection such as fever, malaise, graft tenderness, or gross hematuria. These symptoms can be confused with an infection and can be difficult to distinguish from rejection. onetheless, these symptoms occurring in a patient who has recently had a failed kidney transplant are usually a result of rejection, and so a prolonged workup, searching for the source of fever, is not usually necessary or productive. These symptoms can occur many months after the patient is back on dialysis and many after the transplantation, which can make the diagnosis more difficult. Sometimes proceeding with nephrectomy is the only way to distinguish the symptoms. There has been some suggestion that repeat transplants may do worse in patients who have had the primary graft removed.1 The data on this point are not absolutely convincing, although they further support the notion of doing the nephrectomy only for specific indications.
A rare indication for allograft nephrectomy is the presence of a mass lesion of the transplant kidney. We have seen this twice in approximately 1,500 transplants. The diagnosis is usually made by ultrasound done for routine evaluation of the transplant, which detects the mass, followed by CT scan and biopsy to confirm the diagnosis. This may involve lesions unintentionally transmitted from the donor or lesions arising de novo in the transplant kidney. In the early era of transplantation, nephrectomy was often required because of technical complications such as urinary fistula or graft hemorrhage, which were not lethal to the kidney but could not be repaired. This almost never occurs now.
ALTERNATIVE THERAPY
One alternative to allograft nephrectomy is watchful waiting, as described earlier. In those whose graft is lost beyond 1 year from the transplant, about half the time no symptoms occur, and the kidney shrinks and sometimes will calcify. In those chronic cases where symptoms do occur, a short course of oral steroids is sometimes given over a week or two with prednisone, 15 mg/day initially and quickly tapered. This will ameliorate symptoms and, in a rare case, may eliminate the need for nephrectomy.
Another alternative is radiologic embolization, which we have performed in a few cases. The drawbacks are having a large kidney occluded and uncertainty as to whether this would make a subsequent transplant on that side more difficult in the future. Another drawback is that it may be hard to angiographically identify the renal artery(ies) and catheterize it (them) in the small chronic kidney, which commonly has significant intimal thickening from the rejection process. One group has reported the combination of ethanol and stainless steel coil transvenous catheter ablation in 14 patients.4 Postembolization syndrome occurred in 11 of 14 patients, and one patient had an abscess develop in the graft.
SURGICAL TECHNIQUE
The patient is placed on the operating room table in the supine position. In most cases the previous transplant incision is a lower quadrant incision (Fig. 9-1). The nephrectomy incision is over that incision and extended laterally if necessary. If the transplant had been done through a midline transperitoneal incision, it may be necessary to reopen that incision and go transperitoneally to get to the kidney. When the transplant has been transperitoneal, it is often swollen and palpable in the lower abdomen, and thus, it is possible to make the incision directly over the kidney, incise the external and internal obliques, and reach the kidney capsule. Because the kidney is usually stuck to its surrounding structures, it is possible in this situation to stay extraperitoneal and remove the kidney. Routinely, in reopening a lower quadrant incision, the scars in the external and internal obliques are reincised. Once the internal oblique is incised, the kidney should be approached as laterally as possible because the peritoneum is often draped over the kidney in the line of the incision. If the peritoneum is opened and the patient is on peritoneal dialysis. this creates a problem to use the P-D catheter until the peritoneum reseals, necessitating temporary vascular access for dialysis. This is particularly problematic in children, and it is helpful to avoid this circumstance.
In cases where the kidney has rejected, it is sometimes swollen to two to three times its normal size. The kidney is palpable superficially, but the bulk of the size is hidden in the flank. This may be the situation even where there has been chronic rejection because acute rejection is often superimposed. A very long incision may not create the amount of access needed to dissect the kidney under direct vision because of the lie of the kidney in the abdomen and flank. The subcapsular approach, in addition to helping with the kidney being stuck to surrounding structures, helps with these very large kidneys because it allows the mobilization of the kidney to be done blindly and delivers the large kidney into the incision so the pedicle can be dissected under direct vision. This is not unlike the dissection of the prostate in a retropubic or suprapubic prostatectomy for BPH.
Kidneys that have been lost in the first few days after transplant for technical reasons can be removed in toto, including the whole capsule, because these kidneys are not usually stuck to the surrounding tissues. Kidneys that have undergone rejection are usually stuck to the pelvic side wall, iliac vessels, and peritoneum. Attempts to remove these extracapsularly can result in injury to vessels or bowel in addition to being very difficult to do.6 A subcapsular approach simplifies the operation considerably, reducing risk of damage to adjacent structures.7 The renal vessels are ligated well into the hilum.
After the internal and external obliques are opened, the surface of the kidney can be balloted and often has a bluish tint. A tiny incision in the capsule is made with the knife, confirming the presence of renal parenchyma (Fig. 9-2). The Metzenbaum scissors are used to dissect bluntly under the capsule and enlarge the capsulotomy. A finger sliding under the capsule should meet a plane that dissects easily (Fig. 9-3). An exception is the rare circumstance where the kidney is small and partly calcified, where the subcapsular plane may be quite difficult to appreciate. In the more usual circumstance, the plane between capsule and parenchyma is extended around the entire kidney except the hilum, and the kidney is delivered into the wound. It may be necessary to extend the fascial incisions to allow this depending on the size of the kidney. The kidney will now be tethered by the vessels and the ureter. Care must be taken not to avulse these in delivering the kidney into the wound.
The peritoneum can be entered here, so it is best to make this incision close to the renal parenchyma. The other advantage is that ligating the vessels well into the hilum minimizes the risk of damage to the iliac artery. A theoretical disadvantage to this approach is that donor material is left in situ. Blowout of this remnant is a possibility but appears to occur only if the suture line itself becomes infected.8 Therefore, leaving this material in place usually causes no problem. The only other drawback is that if a third transplant is ever required on that side, it may be necessary to dissect this scar out at the time of the transplant. The potential problems that this presents are outweighed by the complexity of the nephrectomy and risk of trying to control the proximal and distal iliac artery and vein in the presence of a large, friable failed allograft and can sometimes be avoided by going higher on the vessels, to the common iliacs or aorta and vena cava.
The hilar structures may be densely adherent to one another, but it is usually possible to dissect arteries, veins, and ureters free for individual ligation (Fig. 9-5). It may be possible to obtain only enough length to handle the proximal vessels without ligating the kidney side. A finger over the kidney side may be sufficient until all the vessels have been ligated. Ligation plus suture ligation with 2-0 or 3-0 cardiovasculars wedged-on sutures provides security that there will be no major bleeding postoperatively. Silk, Prolene, or some other nonabsorbable suture can be used. The vessels themselves are often friable, and care must be taken not to saw through them. Occasionally a vessel is torn flush with the dense scar at the base of the kidney overlying the iliac artery or vein, which can not be seen. A figure-of-eight stitch can control it, but care must be taken not to pass the needle too deeply into the scar because the iliac vessels may be very superficial.
Once all the hilar structures have been ligated, hemostasis of the capsule must be obtained. This can be done by using the flat part of the electrocautery or an argon-beam coagulator. The entire capsular surface should be inspected. The unacceptable alternative, unless it is absolutely impossible to obtain a dry field, is to drain the space, which runs the risk of potentiating infection of the space, even if a closed drain is used. Spray thrombin can be liberally applied after coagulation. Care should be taken not to rub off the carefully obtained surface hemostasis. Antibiotic irrigation is carried out, although the evidence that this prevents infection in this situation is lacking. The wound is closed in one or two layers, depending on what is possible. I prefer interrupted 2-0 Prolene, but absorbable and running sutures are possible. Usually there is no value in specifically closing the remaining renal capsule.
OUTCOMES
Complications
Complications reported in earlier series included major perioperative bleeding, wound infections, and intraoperative injury to adjacent structures.5 The incidence of hemorrhage following transplant nephrectomy is 5% to 6%.2 Routine use of the subcapsular approach can reduce these complications to less than 1%.6
The other type of complication is less related to the operation than it is to immune suppression. Even though patients undergoing allograft nephrectomy have had their immune suppression stopped before the nephrectomy, it may have been only a few days before. The sequelae of immune oversuppression may take days or weeks to manifest themselves. In the UCLA series, two perioperative deaths were from B-cell lymphoma related to immune oversuppression. Persistent fever after nephrectomy should be carefully evaluated.
Results
Transplant nephrectomy can be performed with minimal morbidity, though it can be extremely challenging in the postrejection setting. The issue of the value of transplant nephrectomy to reduce the possibility of rejection of subsequent renal allografts remains controversial.

Renal cell carcinoma (RCC) is the most common malignancy of the kidney and accounts for about 3% of all adult neoplasms. The estimated number of new cases of renal cell carcinoma in the United States in 1997 is 28,000 with a projected 11,300 deaths, and this incidence is expected to increase as a result of the expanded use of radiographic imaging.7 Because renal cell carcinoma is relatively refractory to chemotherapy and radiation therapy, surgery in general and radical nephrectomy in particular has evolved as the primary treatment in patients with clinically localized and locally advanced disease.
Radical nephrectomy is defined as resection of Gerota’s fascia and its entire contents including the kidney, perinephric fat and lymphatics, and ipsilateral adrenal gland. In theory, complete surgical excision of all tumor with negative surgical margins would offer the best opportunity for cure in patients with renal cell carcinoma. This argument would favor radical nephrectomy over simple nephrectomy, given the frequent propensity of the tumor to extend microscopically outside of the renal capsule and into perinephric fat. Thus, although no randomized trial has demonstrated the superiority of radical nephrectomy over simple nephrectomy, multiple series have documented improved survival in patients with renal cell carcinoma treated with radical nephrectomy over the past 30 years.2,3,4 and 8

DIAGNOSIS
Between 85% and 90% of all solid renal masses are renal cell carcinoma, and, therefore, the diagnosis of renal cell carcinoma should be considered in all patients with a suspected solid renal mass. A renal mass detected on either intravenous pyelography or ultrasound is usually confirmed by computed tomography (CT scan). Typically, renal cell carcinomas are characterized on CT scan by a solid parenchymal mass with a heterogeneous density and enhancement with intravenous contrast injection (between 15 and 40 Hounsfield units). However, despite modern imaging, some benign tumors of the kidney may be indistinguishable and confirmed only after surgical excision. Additionally, metastatic deposits from a variety of malignancies including lung and breast cancers may involve the kidney and should be considered in patients with a known primary. The role of percutaneous biopsy or needle aspiration in differentiating an indeterminate renal mass remains controversial, and the absence of malignant cells on biopsy does not rule out the possibility of a neoplasm. For this reason, percutaneous renal biopsy for the purpose of diagnosis should be used only in selected cases.
Clinical staging in patients suspected of renal cell carcinoma usually includes a contrast-enhanced CT scan of the abdomen; however, MRI is used occasionally and is particularly useful in patients with a history of a contrast allergy, renal insufficiency, or a suspected vena caval thrombus. From these imaging modalities, a number of factors can be determined, including the size and resectability of the primary, the presence or absence of lymphadenopathy or metastasis, involvement of adjacent structures, and the status of the contralateral kidney. A chest x-ray is obtained to rule out lung metastasis. Bone scans are performed in any patients with symptoms referable to the bone, as well as an elevation in serum alkaline phosphatase or hypercalcemia. In cases of suspected vena caval involvement, Doppler ultrasound is a useful screening tool. If results are equivocal, or if a vena caval thrombus is confirmed, a vascular phase MRI is usually able to determine the level of extension of the tumor thrombus, which allows the surgeon to properly plan an operative strategy.
INDICATIONS FOR SURGERY
The indication for radical nephrectomy is a clinically localized solid renal mass in a patient with a normal contralateral kidney. Patients with solitary kidneys, renal insufficiency, and bilateral renal masses should be considered candidates for nephron-sparing surgery. A thorough preoperative history and physical examination should be performed before the procedure. If significant comorbidities are suspected, consultation with the appropriate physician is recommended. The patient should be expected to physically withstand the operation and have a reasonable overall performance status and a 5-year life expectancy.
In general, radical nephrectomy in patients with metastatic disease is performed for palliation, such as those patients with intractable pain or life-threatening hemorrhage who fail conservative treatment. Also, radical nephrectomy may be performed in the setting of an approved investigational protocol. Radical nephrectomy as a potential adjunct to enhance the effectiveness of biological response modifiers in patients with metastatic renal cell carcinoma remains experimental. The role of radical nephrectomy in patients with a solitary metastatic site is controversial; however, 5-year survival rates of 30% have been reported in selected patients, with best results reported in patients with solitary pulmonary metastases.2
Although local extension of primary renal cell carcinoma into the perinephric fat, vena cava, or ipsilateral adrenal gland may portend a worse prognosis, in the absence of metastatic disease these factors alone should not dissuade the surgeon from attempting a radical nephrectomy. Additionally, radical nephrectomy has been successfully performed in the setting of direct extension of the tumor into adjacent organs such as the liver, colon, or psoas muscle. However, surgical removal in this setting is technically difficult and is associated with a higher morbidity and a potentially poor prognosis. Therefore, it should be attempted only in selected patients without obvious nodal or metastatic disease and in cooperation with appropriate surgical consultants. The role of regional lymphadenectomy at the time of radical nephrectomy remains controversial, and presently most patients should undergo only a limited unilateral lymphadenectomy for the purpose of staging.2
ALTERNATIVE THERAPY
Surgery remains the only effective and potentially curative form of therapy for primary RCC. Along this line, the main challenge to radical nephrectomy in the near future appears to be from more conservative surgical approaches. Partial nephrectomy, enucleation, and wedge resection have recently been proposed in small, clinically localized RCC, with excellent early results (Chapter 6).6 Arguments against partial nephrectomy include the potential for local recurrence, tumor multifocality, and the potential for increased complications.
The routine removal of the ipsilateral adrenal gland at the time of radical nephrectomy has also been questioned, particularly in small tumors and tumors not involving the upper pole.9 The use of laparoscopic techniques is currently expanding, and recently laparoscopic nephrectomy has been reported in patients with small tumors, although concerns over potential for tumor spillage and alteration of pathologic staging remain to be addressed.5 Currently, chemotherapy and radiation therapy have proven to be inadequate treatment in primary renal cell carcinoma.2
SURGICAL TECHNIQUE
There are a variety of factors that influence the choice of incision during radical nephrectomy. These include location of the affected kidney, tumor size and characteristics, body habitus, and physician preference. There are advantages and disadvantages to each incision, and it is important to be familiar with several approaches to the kidney, as no one incision is appropriate in all settings. The most commonly used incisions for radical nephrectomy are the flank, thoracoabdominal, and transabdominal (subcostal or chevron) (Fig. 7-1).
Flank Incision
The flank approach is an excellent choice for a variety of reasons. First, it allows direct access to the retroperitoneum and kidney, and the entire procedure can often be performed in an extrapleural and extraperitoneal fashion. Additionally, the incision is anatomic in that it follows the track of the intercostal nerves with minimal risk of denervation. However, in large tumors, tumors involving the upper pole, or in situations where vena cava access is critical, a flank approach may be suboptimal. Although a flank approach may be performed through a subcostal incision, an 11th or 12th rib incision is superior for exposure of the upper pole and ipsilateral adrenal gland during radical nephrectomy.
The patient is positioned on an inflatable mattress in the lateral decubitus position with the upper chest at about a 45 degree angle. An axillary roll is placed under the patient to cushion against pressure on the brachial plexus, and the elbows are padded to prevent ulnar nerve injury. The upper arm is draped across the body and placed on a Mayo stand or a padded support. The lower leg is flexed at 90 degrees, and the upper leg is extended over one or two pillows. The kidney rest is raised and the table is flexed to elevate the flank, and the table is adjusted to make the flank horizontal to the floor. The inflatable mattress is then activated, and the patient is secured with wide tape.
An 11th or 12th rib incision is made based on several factors including the kidney position, the cephalad extent of the tumor, and the patient’s body habitus. A general rule is to incise over the rib that, when extended medially, will position the incision over the renal hilum. The incision is then made over the rib from the posterior axillary line to the tip and extended medially as far as necessary, which usually stops short of the lateral border of the rectus abdominis Figure 7-2). The latissimus dorsi is divided, and the upper portion of the incision is carried down to the rib. At this point a partial rib resection may be accomplished as shown in (Fig. 7-3). An Alexander periosteal elevator is used to deflect the periosteum from the bone to avoid injury to the intercostal bundle located under the inferior portion of the rib. A Doyen elevator is then used to strip the periosteum from the entire undersurface of the rib to be resected. Next, a rib cutter is used to divide the proximal segment of the rib. Alternatively, the incision may be created between the ribs in the intercostal space.
The posterior layer of the periosteum is then incised carefully, and the pleura is protected superiorly. Anteriorly, the external and internal oblique muscles are divided, and the transversus abdominis muscle is split in the direction of its fibers, taking care not to enter the peritoneum. The peritoneum is swept medially, and the intermediate stratum of the retroperitoneal connective tissue is incised sharply to expose the paranephric space. Approaching this in a posterior fashion with early identification of the psoas muscle helps to keep proper orientation. A self-retaining retractor such as a Finochetto or Balfour helps to maintain exposure. A radical nephrectomy is then performed.
The wound is closed after checking to ensure that no injury to the pleura has occurred (see complications). The table flex is released, and the kidney rest is lowered. The posterior layer consisting of the fascia of the transversus abdominis and the internal oblique is closed in a running fashion with #1 PDS or Prolene. The anterior layer of external oblique fascia is closed with a running #1 PDS or Prolene. Alternatively, interrupted figure-of-eight sutures of #1 Vicryl can be used for both layers. The skin is closed in accordance with surgeon preference.
Thoracoabdominal Incision
The thoracoabdominal approach allows for excellent exposure of large tumors as well as upper-pole tumors, particularly on the left. Additionally, it affords easy access to the adrenal gland and thoracic cavity. The patient is positioned with the hips flat and with the break of the table located just above the iliac crest. The pelvis can be torqued up to about 30 degrees if necessary. The patient’s ipsilateral shoulder is rotated 45 degrees, and the ipsilateral arm is extended over the table and properly supported on a Mayo stand or padded arm rest Figure 7-4). It is important to properly pad all pressure points including between the legs and the contralateral shoulder. The kidney rest may be elevated to accentuate the proper extension, and the break in the table is made to optimize the incision. After positioning, the patient is secured with wide adhesive tape.
The thoracoabdominal incision is made over the bed of the eighth, ninth, or tenth rib, depending on the surgeon’s preference based on patient and tumor characteristics. The incision may be made between the ribs, or a portion of the rib may be removed. The incision is made over the rib beginning at the posterior axillary line. The incision is carried medially across the costal cartilage margin to the midline and then carried down the midline to the umbilicus. Alternatively, the medial portion of the incision may be carried across the midline or combined with a low midline to form a “T.” The latissimus dorsi is divided, and the upper portion of the incision is carried down to the rib. At this point, a rib resection can be performed as previously described (Fig. 7-3).
The peritoneum may be entered by incising the external and internal obliques, the transversus abdominis, and the ipsilateral rectus belly. Next, the costochondral cartilage at the inferior portion of the upper thoracic incision is divided, and the chest is entered along the entire length of the periosteal bed. The pleural space is entered, and care should be taken not to injure the lung. The lung is protected by pads, and the diaphragm is divided in the direction of the muscle fibers, which helps to avoid injury to the phrenic nerve. A self-retaining retractor such as a Finochetto or a Balfour is properly padded and placed to maintain exposure. A radical nephrectomy is then performed.
After completion of the radical nephrectomy, the table flex is removed, and the diaphragm is closed with interrupted 2-0 silk sutures with knots placed on the inferior side. After a #32 chest tube has been inserted through a separate incision and properly positioned, the ribs are reapproximated with 2-0 chromic pericostal sutures. The thoracic portion of the incision is closed with interrupted figure-of-eight 1-0 Vicryl sutures through all layers of the chest wall. The medial portion of the intercostal muscle closure should include at least a small portion of the diaphragm. An intercostal nerve block is administered before closure and may be accomplished by injecting approximately 10 ml of 0.5% lidocaine or bupivicaine hydrochloride into the intercostal space of the incision and two interspaces above and below. The costal cartilage can be reapproximated with 0 chromic suture. The peritoneum is closed with a running 2-0 chromic, although this is optional. The posterior rectus fascia, the fascia of the transversus abdominis, and the internal oblique muscles are closed with a running or interrupted #1 PDS suture. The anterior rectus and the external oblique fascia are closed with either a running or interrupted #1 PDS or Maxon suture. Skin closure is determined by surgeon preference. The chest tube is secured in place with a 0 silk and taped securely in place.
Transabdominal (Chevron or Anterior Subcostal)
Anterior incisions offer several advantages including excellent exposure of the renal pedicle and access to the entire abdomen and contralateral retroperitoneum. With the patient in the supine position, the operative side is elevated slightly with a flank roll, and the patient hyperextended to accentuate the line of incision. An incision is made from near the tip of the 11th or 12th rib on the ipsilateral side two fingerbreadths below the costal margin and extended medially to the xyphoid process. The incision is then gently curved across the midline and as far laterally as necessary for exposure up to near the tip of the contralateral 11th rib.
Occasionally, only a portion of the contralateral side will be incised just across the rectus abdominis. The incision is carried down to the anterior rectus fascia, which is then divided (Fig. 7-5). Next, the external and internal oblique fascia and muscles are divided, and the fibers of the transversus abdominis split. The rectus muscle and posterior rectus sheath are divided with electrocautery by placing a straight clamp or army–navy retractor underneath and gently elevating it. The superior epigastric artery is ligated with 2-0 silk and divided when encountered. The peritoneal cavity is then entered, and the falciform ligament is ligated between two Kelly clamps, divided, and tied with 0 silk suture. To facilitate exposure, the lower aspect of the incision is rotated caudally with a rolled towel placed underneath the skin, and the fascia sutured to the lower portion of the abdominal wall with two #2 nylon sutures. Use of a self-retaining retractor such as a Wishbone Omni-Tract is helpful. A radical nephrectomy is then performed.
Closure of the wound is performed after the table is returned to the horizontal position. The wound is then closed in two layers. The posterior layer consisting of the fascia of the transversus abdominis and the internal oblique laterally along with the posterior rectus fascia medially is closed with two running #1 PDS sutures, each starting at the lateral aspect and running medially to the midline. The anterior layer of external oblique and anterior rectus fascia is closed in a similar fashion with #1 PDS. Alternatively, the layers can be closed with interrupted #1 Vicryl. Occasionally, it is helpful to place a U stitch of #1 Prolene at the apex of the chevron incision before closure, which includes the rectus fascia on either side of the midline, securing this suture after the anterior fascia has been approximated. The skin is then closed according to the surgeon’s preference.
Radical Nephrectomy
Irrespective of the choice of incision, certain caveats are universal for the safe and successful completion of a radical nephrectomy. This includes a systematic approach with careful mobilization of Gerota’s fascia and early vascular control. For a flank approach, the posterior peritoneum lateral to the colon is incised along the length of the descending colon (left side) or ascending colon (right side) and reflected medially. For left-sided exposure, the lienorenal ligament is incised to mobilize the spleen cephalad. On the right side, the hepatic flexure of the colon is mobilized. The ureter is identified and encircled with a vessel loop. The gonadal vein is ligated and divided. The plane between the mesentery of the colon and Gerota’s fascia is then developed using a combination of sharp and blunt dissection. On the right side, the vena cava is exposed by Kocherizing the duodenum. Using blunt dissection, the retroperitoneal fat overlying the renal vessels is separated, exposing the renal hilum. It is often helpful to ligate and divide the ureter before this to allow for mobilization and upward displacement of the lower pole of the kidney.
The dissection is then carried cephalad along the vena cava (right side) or aorta (left side). On the right side, the right renal vein is identified exiting from the vena cava, isolated, and encircled with a right-angle clamp and a 0 silk suture and tagged. After identification of the renal artery (exposure may be enhanced by the use of a vein retractor on the renal vein), the artery is dissected free and cleaned for a distance of approximately 2 to 3 cm. With a right-angle clamp, the renal artery is encircled, and 2-0 silk ties are passed (Fig. 7-6). The sutures are then separated and tied, allowing a safe distance for division of the artery. A small hemoclip or a 3-0 silk suture ligature may be placed on the proximal aspect of the artery before division. A right-angle clamp is placed under the artery to be divided and gently elevated, and the artery is cut with either a knife (#15 blade) or Metzenbaum scissors. The right renal vein is then ligated in a similar fashion with 0 silk sutures.

On the left, the renal vein is isolated as it courses over the aorta. The left adrenal and gonadal veins are identified emanating from the left renal vein, and, if present, a posteriorly directed lumbar venous tributary is noted. A right-angle clamp is passed around the renal vein, followed by a 0 silk suture proximal to the tributaries, and tagged. The venous tributaries are then individually ligated and divided with 2-0 or 3-0 silk and small hemoclips where necessary, leaving the 2-0 silk suture on the main renal vein tagged (Fig. 7-7). The left renal artery and vein are then ligated similarly to the technique described above for the right side.
Gerota’s fascia is then mobilized posteriorly and superiorly using a combination of sharp and blunt dissection. Hemoclips along the superior and medial border are useful to control any potential bleeding during this portion of the procedure. The adrenal hilum is then dissected from caudal to cranial with the aid of either hemoclips or straight clamps and ties. On the right side, the short posteriorly located right adrenal vein should be anticipated as it exits directly from the vena cava. When encountered, the right adrenal vein is isolated, ligated, and divided. The specimen is then delivered, and meticulous hemostasis is achieved.
OUTCOMES
Complications
The potential for bleeding during radical nephrectomy necessitates careful patient preparation and preoperative planning to significantly reduce the chances. Any medications that interfere with platelet function or clotting should be discontinued, and patients should be type and cross matched for 2 units of packed red blood cells. The patient should have either two large-bore peripheral intravenous lines or a central venous line to allow for rapid infusion of fluids or blood products.
Bleeding during radical nephrectomy may be from a variety of locations including the renal hilum, collateral tumor vessels, or adjacent structures. Venous bleeding is usually the most problematic. The first maneuver is to apply direct pressure to the area of bleeding. The point of bleeding is then carefully exposed and controlled by a suture ligature. In the case of venal caval injury, a Satinsky clamp is placed, and a vascular 5-0 or 6-0 Prolene suture is used to oversew the defect. Lumbar veins should be exposed by gentle retraction of the vena cava, appropriately clamped, ligated with vascular silk suture, and divided. Renal artery bleeding may be controlled by direct pressure on the aorta proximally until adequate exposure can be obtained, and the artery is then ligated. Only in rare circumstances will a pedicle clamp or mass ligature be necessary.
Adrenal tears may result in significant hemorrhage during radical nephrectomy, particularly on the right side, where the short adrenal vein enters into the vena cava directly posterior. Control of the right adrenal vein should be attempted only after control of the vena cava, adequate exposure, and proper suction. The vein is then ligated with a 2-0 or 3-0 silk tie or a vascular Prolene. Venous bleeding from a torn adrenal gland can be oversewn with a running suture or stopped by placement of surgical clips. However, removal of the ipsilateral adrenal may be the most expeditious method of controlling bleeding.
Failure to recognize a rent in the pleura during flank incision will result in a pneumothorax. Small openings in the pleura may be recognized by filling the flank wound with sterile water and administering a deep inspiratory breath. Small tears recognized intraoperatively can be managed by closing the pleura with a 3-0 chromic pursestring suture over a 12 Fr or 14 Fr Robinson catheter. Before it is removed, all air is aspirated from the pleural cavity either by suction or by placing the Robinson catheter under water and administering a deep inspiratory breath. The air is evacuated from the pleural space, and the tube is removed while the pursestring suture is simultaneously tied in place. Alternatively, the Robinson catheter can be temporarily left in place with the chromic suture secured and the fascial layers closed around the catheter, which exits from the corner of the wound. Just before skin closure, after all air has been evacuated under water seal as described above, the catheter is removed. The latter technique is helpful when the pleura is attenuated or contains multiple small holes that are not easily closed. Alternatively, a 22 Fr or 24 Fr chest tube may be placed and left to suction.
An upright end-expiratory chest x-ray is obtained after all flank incisions to ensure that no significant pneumothorax exists. A small (usually less than 15%), asymptomatic pneumothorax can be followed conservatively with serial chest x-rays and oxygen therapy. In a symptomatic or large pneumothorax, aspiration of the pleural space using a needle or a central venous catheter (Seldinger technique) introduced just over the rib in the anterior fourth or fifth interspace can be therapeutic. However, if these attempts are not successful, a chest tube should be inserted and placed on suction.
Injuries to the colon during radical nephrectomy are uncommon. In locally advanced tumors suspected of extension into either the colon or mesentery, patients should undergo a mechanical and antibiotic bowel preparation. Segmental colon resection and primary anastomosis should be possible in most cases. Inadvertent injury to the colon during radical nephrectomy can usually be repaired primarily; however, in situations where there is gross spillage of fecal contents or a devascularized segment, a diverting colostomy should be considered, and a general surgery consultation is advisable. Defects in the mesentery of colon should be closed to prevent internal herniation of peritoneal contents.
Right radical nephrectomy is also associated with the potential for injury to the duodenum and liver. The duodenum must be carefully mobilized, and care must be taken to properly pad retractors to prevent injury to the bowel and adjacent structures including the head of the pancreas. The second portion of the duodenum may be injured during a right radical nephrectomy. Duodenal hematomas should only be observed, but rapidly enlarging hematomas will require control of the bleeding, and an intraoperative general surgery consultation should be obtained. Duodenal lacerations should be repaired in multiple layers with interrupted nonabsorbable sutures for the mucosal and serosal layers.10 When possible, an omental wrap may provide additional support, and all patients should be managed with a nasogastric tube during the postoperative period.
Superficial liver lacerations are repaired with absorbable horizontal mattress sutures utilizing a Surgicel or Hemopad bolster. Deep liver lacerations, which may involve the hepatic ducts, could result in bile leakage and should be drained following repair. Direct invasion of the liver by renal cell carcinoma is rare; however, resection including en-bloc removal is possible in selected cases. If a major lobectomy or a partial hepatectomy is to be performed because of either direct extension or major hemorrhage, a general surgeon should be present.
Splenic injury is one of the most common intraoperative complications during a left nephrectomy, with an incidence as high as 10% in some series. Most superficial lacerations or tears can be managed conservatively without the need for splenectomy. Although minor tears may require only some gentle pressure and the application of a Hemopad or Surgicel with Avitene application, closure of a moderate splenic capsular tear is facilitated through the use of nonabsorbable sutures over bolsters of Surgicel. Major hemorrhage secondary to severe splenic lacerations may require splenectomy. The splenic artery and vein are controlled by compressing these structures, located in the splenic hilum near the tail of the pancreas. Initially, this can be accomplished manually by compressing the tail of the pancreas between the thumb and the forefinger. Once bleeding has been temporarily controlled, the spleen is mobilized by dividing the splenocolic and splenorenal ligaments as well as taking down the peritoneal attachments to the diaphragm. The short gastric vessels are then ligated, and the hilum of the spleen is dissected free from the tail of the pancreas. The splenic artery and vein are ligated and divided. The pancreas should be inspected closely to rule out inadvertent injury. Following splenectomy, patients will have a reduced resistance to pneumococcal organisms and should receive Pneumovax and Hibtiter on a yearly basis.
Results
Surgical excision remains the only effective and potentially curative therapy for clinically localized RCC. Pathologic staging remains the best prognostic variable in terms of patient survival, and the two most commonly used staging systems are the Robson classification and the American Joint Committee on Cancer recommendations (TNM) classification.1,8 Both staging systems have demonstrated an inverse relationship between survival and increasing stage, but the TNM is generally thought to be more accurate because it more precisely defines the extent of disease.
In patients treated with radical nephrectomy and found to have tumors confined to the kidney (Robson stage I), the 5-year survival is between 60% and 90% compared with 47% to 67% in patients whose RCC is confined to Gerota’s fascia (Robson stage II). Survival for patients with distant metastases is poor with 5-year survival of between 5% and 10%.2 Under the TNM staging system, the 5-year survival for patients with organ-confined tumors treated with radical nephrectomy for T1N0M0 tumors is between 80% and 91%, whereas that for T2N0M0 tumors is 68% to 92%.3,4 For those patients with T3aN0M0 (tumor invading into the adrenal gland) and T3bN0M0 (tumor invading into the renal vein) carcinomas, the 5-year survival is 77% and 59%, respectively. Finally, patients with node-positive disease (N1–3M0) have a 5-year survival between 15% and 52%.
Although radical nephrectomy remains the standard of care for unilateral renal cell carcinoma, more conservative surgical options have been proposed. Recently, excellent results have been seen in patients treated with nephron-sparing surgery, with 5-year cancer-specific survivals of greater than 90% for small, unilateral, stage I tumors.9 The ultimate choice of surgical treatment in patients with these favorable clinical features, and in particular the risk of local recurrence and cancer-specific survival, remains to be determined through long-term follow-up. Currently, radical nephrectomy remains the treatment of choice in patients with clinically localized renal cell carcinoma and the standard against which future alternative surgical strategies will be measured.