It is estimated that >200 000 cases of RCC are diagnosed annually worldwide, making it the third most diagnosed urothelial cancer behind bladder and prostate . Most RCCs occur in the fifth to seventh decade of life, with a male to female ratio of 3 : 2 and a greater incidence in patients with von Hippel-Lindau disease, horseshoe kidneys, acquired renal cystic disease, end-stage renal disease and obesity. The disease carries the highest mortality rate (40%) of the genitourinary malignancies, resulting in 100 000 deaths each year. Estimates have suggested that the incidence of RCC is increasing at a rate of 2.5%, which is attributable to early detection and increased median lifetime survival [2,3]. Because RCC is resistant to non-surgical treatments (chemotherapy, hormonal therapy and radiation) radical nephrectomy (RN) remains the reference standard of treatment. Advances in imaging methods, surgical technique and molecular biology have enabled the detection of disease at earlier stages, the treatment of more aggressive disease and the offer of promising future therapeutics for metastatic or recurrent disease.
The first series report of 130 patients by Sir Eric Riches in 1963 concluded that prognosis of RCC depends on histological grade, venous invasion, lymphatic spread, perinephric fat invasion and distant metastasis . These conclusions were tested in the landmark paper by Robson et al., where 88 patients with RCC had RN, primarily via thoracoabdominal incisions with lymph node resection and a 3-, 5- and 10-year follow-up. The surgical tenets of RN (early ligation of renal vessels, en bloc kidney removal including Gerota’s fascia, and ipsilateral adrenal and regional lymphadenectomy) originate from this paper. Additional contributions by urologists in treating RCC are noted in Table 1.
Table 1. The historical contribution of surgeons to RN
Wolcott and Stoddard
First nephrectomy in patient with suspected liver cyst
First planned nephrectomy for benign disease (ureterovaginal fistula)
First planned nephrectomy for malignant disease
First partial nephrectomy: perirenal fibrolipoma
Reports nephrectomy for hydronephrosis in 21 patients; 38% mortality
Describes the anatomy of the renal fascia
‘Brodel’s Line’; avascular plane posterior to midline
First IVC thrombectomy: transverse abdominal incision
Retrospective review of 88 patients undergoing RN
Fray and Krane
CBP-DHCA for removal of atrial thrombus
First laparoscopic nephrectomy
These basic principles in urological surgery have come under scrutiny, as tumour extension and location are better defined via superb imaging techniques, thus promoting minimally invasive surgery with respectable survival outcomes. Current reports question the need for all components of the original surgery (adrenalectomy, complete kidney removal, extensive lymph node dissection.) Preservation of renal parenchyma via limited extirpative techniques can be seen as ‘reno-protective’ in the setting of an ageing population with increased rates of chronic renal failure.
With the advent of modern and excellent cost-effective imaging techniques, many renal lesions are being detected incidentally during evaluations for abdominal pain. Many recent series estimate that 15–48% of RCC cases are discovered incidentally . This has led to an improved prognosis due to early-stage detection. In essence the classic clinical triad (flank pain, haematuria and palpable mass) might be better served as the ‘elusive triad.’ RCC has often been described as the ‘internist’s tumour’ due to the many paraneoplastic manifestations :
• Hypertension; increased renin production.
• Elevated erythrocyte sedimentation rate.
• Weight loss/cachexia.
• Fever of unknown origin.
• Elevated alkaline phosphatase;
• Hypercalcaemia; release of parathyroid hormone-related hormone or osteolytic bone metastasis.
• Staufer syndrome; elevated liver function tests, fever and hepatic necrosis (resolves after nephrectomy).
The diagnosis can be delayed or attributed to many other medical conditions, leading to a metastases at diagnosis. Incidentolomas are more common but 20–30% of patients still present with metastatic disease. The 5-year survival in the latter population is <10%.
The overall goal of preoperative imaging is to determine the malignant potential of the lesion, assess tumour size and location, determine capsular invasion, and to identify regional lymphadenopathy or metastases. In addition, the renal vein and inferior vena cava (IVC) need to be assessed for tumour thrombus, as 4–10% of RCC will present with venous tumour involvement. Of the latter group, 1% might also have extension to the level of the right atrium .
CT remains the primary imaging to detect and differentiate renal lesions. RCC can display hypo-, iso-, and hyperdense characteristics on native CT. Previous reports describe RCC as having a significant contrast enhancement of 115 ± 48 Hounsfield units on corticomedullary phases (CMP) and 62 ± 25 Hounsfield units on the excretory phase. Using these parameters the sensitivity and specificity in differentiating RCC from other renal tumour subtypes was 74% and 100% for CMP phase, and 84% and 91% for the excretory phase . Contrast enhancement often shows intratumoral areas of necrosis. These estimates are apt to change with the incorporation of helical CT techniques and three-dimensional (3D) capabilities. The reconstructive capability of CT has provided surgeons with precise anatomical locations and we advocate their use in almost all of our cases.
MRI has great promise in the evaluation of renal masses, multifocality, tumour staging, vessel and collecting involvement. Huang et al. correctly staged 29 of 30 patients with renal cell cancer using 3D-MRI. MRI has been shown to be equal to CT for detecting renal masses, with MRI showing advantages when discerning complicated cysts from hypo- or avascular tumours. MRI is well suited for individuals with renal impairment or allergy to contrast media.
The routine use of arterial and venous direct imaging has fallen from favour compared with CT and MRI. At our institution we choose to evaluate venous tumour thrombus extension with 3D CT reconstructions and MR venography. Arteriography can determine the extent of arterialization of tumours and provide a means for preoperative embolization. Classic findings in RCC include neovascularization, arteriovenous fistula, contrast media pooling, and capsular vessel accentuation.
Imaging studies have led to an increase in the number of renal lesions; in many instances these findings are cystic and need no further follow-up. For solid masses, it is estimated that >90% are RCC. Cysts are classified according to the Bosniak system (Table 2). For cysts with internal septations and calcifications a close follow-up is mandated if they are not removed for ‘peace of mind’. Masses in the setting of primary tumours can be biopsied to exclude metastatic disease. In other situations, the diagnosis of lymphoma must be considered. The last two disease states are the only ones whereby needle biopsy of large renal masses might be warranted, due to the different treatment approaches. The use of biopsy in diagnosing RCC for smaller renal lesions remains a debated topic.
Clinical staging of RCC involves a radiological evaluation of the tumour at a local level; extracapsular invasion, venous thrombus or nodal involvement. The pulmonary and hepatic systems also need to be evaluated, preferably with CT. RCC is staged using the 2002 TNM Classification (Table 3). The 5-year cancer-specific survival from many contemporary studies is estimated as 73–96%, 63–95%, 38–70% and 11–32% for stages I–IV (Table 4) [12–16]. Although too detailed for this section, numerous authors (ourselves included) are collecting data and reviewing long-term outcomes in an effort to enhance the current staging system.
Table 3. The 2002 TNM Classification of RCC
T: Primary tumour
Primary tumour cannot be assessed
No evidence of primary tumour
Tumour <7 cm in diameter, limited to kidney
Tumour 0–4 cm in greatest diameter confined to kidney
Tumour 4–7 cm in greatest diameter confined to kidney
Tumour > 7 cm in greatest diameter confined to kidney
Tumour extends into major veins or invades adrenal gland or perinephric tissues but not beyond Gerota’s fascia
Tumour directly invades adrenal gland or perirenal and/or renal sinus fat but not beyond Gerota’s fascia
Tumour grossly extends into the renal vein or its segmental (i.e. muscle containing) branches, or the IVC below the
Tumour grossly extends into the IVC above the diaphragm or invades the wall of the IVC
Tumour invades beyond Gerota’s Fascia
N: Regional lymph nodes
Regional lymph nodes cannot be assessed
No regional lymph node metastasis
Metastasis in a single regional lymph node
Metastasis in more than 1 regional lymph node
M: Distant metastases
Metastases cannot be assessed
No distant metastases
Table 4. Multi-centre comparisons of 5-year survival for RCC according to stage (Stage I, T1N0M0; Stage II, T2N0M0; Stage III, T1 or T2, N1M0, T3, N0 or N1 M0; Stage IV, T4, Any N, M0, Any T, N2M0, Any T, Any N, M1)
RN involves the en bloc removal of the respective kidney, adrenal gland, perirenal fat, proximal ureter and Gerota’s fascia in >95% of case. Surgeons need to be aware that vascular anomalies are actually the rule rather that the exception. The kidneys lie in the retroperitoneal space in an oblique position, with the right kidney lower than the left, due to the downward displacement by the liver during development. Immediately posterior to the kidneys are the psoas and quadratus lumborum muscles, which are encased by ribs 11 and 12. Both kidneys are surrounded by a renal capsule, which is embedded in perirenal fat contained within a second enveloping lining known as Gerota’s fascia. The latter also encompasses the ipsilateral adrenal gland and has an intimate relationship with the posterior peritoneum. The right kidney is closely apposed to the liver superiorly and adjacent to the hepatic flexure of the colon and duodenum. The left kidney contacts the tail of pancreas, splenic flexure of the colon and the spleen (Fig. 1) .
Traditional anatomy consists of two renal arteries from the aorta lying dorsal to the renal vein, with the right traditionally longer than the left, the latter with a retrocaval course. Each kidney has four constant end-artery segments, i.e. apical, segmental, posterior and basilar. Brodel’s avascular plane exists along the posterior surface of the kidney at the posterior-anterior junction. Multiple renal arteries occur unilaterally and bilaterally in 23% and 10% of the general population. The left renal vein is longer that the right and receives drainage from the adrenal, phrenic, gonadal and lumbar veins occasionally, while the shorter right receives no tributaries before joining the IVC. Intrarenal venous collaterals permit ligation of venous segments without subsequent infarction. The renal lymphatics follow the arterial system and coalesce in the renal sinus before draining into regional lymph nodes .
SURGICAL APPROACHES TO THE KIDNEY
Tumour location and size, in addition to body habitus and surgical history, can influence the choice of incision. The three incisions used for RN are flank, abdominal and thoracoabdominal. Flank incisions are ideal for obese patients whose peritoneum one does not wish to violate, but access to the renal pedicles is marginal. Extraperitoneal approaches (flank and thoracoabdominal) are of great utility in patients with previous abdominal surgery or a chronically infected kidney. Transperitoneal incisions permit direct access to the renal vascular pedicles through the posterior peritoneum and should be considered for advanced renal tumours, which are often hypervascular. Thoracoabdominal incisions are ideal for larger upper-pole tumours and venous tumour thrombus. Pulmonary metastasectomy can also be performed for ipsilateral lesions via this incision. Cardiopulmonary bypass with or without deep hypothermic circulatory arrest (CPB-DHCA) might be required for patients with tumour thrombus at or above the level of the hepatic veins. Wood et al. and Nethercliff et al.[19,20] provide superb in-depth reviews of surgical approaches and relevant anatomy for surgical house officers in early training.
Anterior approaches (subcostal, paramedian and midline) can be used for transperitoneal and extraperitoneal approaches, with the latter providing optimal exposure in the case of RN (Fig. 2) . A midline incision from the xiphoid process to the pubic symphysis is made and the underlying peritoneum entered. Access to both renal pedicles can be achieved through the posterior peritoneum. The right kidney is accessed via a Kocher manoeuvre after taking down the colon via the white line of Toldt at the hepatic flexure (Figs 3 and 4). The left kidney can be accessed through a window in the posterior peritoneum below the ligament of Treitz, or through the lesser sac after dividing the gastrocolic ligament (Fig. 5). Once exposed the renal artery should be taken first, unless embolized before surgery. If the renal vein does not decompress at this point, one needs to seek additional arterial branches perfusing the kidney. The ureter, gonadal vessels and periurethral fat are then dissected free and taken separately. At this point the kidney and associated Gerota’s fascia is dissected free at the superior and inferior margins, using large haemostatic clips. Once free from the retroperitoneum, the renal vein is taken and the specimen passed off the table. After copious irrigation of the surgical bed to insure haemostasis, a regional lymphadenectomy is performed. The wound is closed in layers with no drains left in place.
Flank and subcostal approaches and their variations are commonplace in renal surgery. For this review the thoracoabdominal approach will be described (Fig. 6a–e). Thoracoabdominal incisions provide maximum exposure for managing large renal tumours requiring RN. Proper positioning in a modified flank position is imperative to limit pressure ulcers and provide working access to the peritoneum and thoracic cavities. A curvilinear incision is carried out over rib 10 or 11 from the mid-axillary line to the rectus fascia. The abdomen is entered after dividing the rectus muscle, external oblique, internal oblique, lattisimus dorsi muscle, transversalis abdominis muscle and peritoneum. An extraperitoneal approach can be achieved if the peritoneum is not violated but swept medially and mobilized from the underlying diaphragm, transversalis muscle and anterior rectus fascia. In the absence of metastatic deposits the pleura is opened below the intercostal margins and the diaphragm divided, using caution to avoid the phrenic nerve coursing on the thoracic side. At this point both thoracic and abdominal cavities are exposed. The colon is next mobilized and the vessels identified and taken in a similar fashion to the midline incision. Closing the thoracoabdominal incision must use a sound technique to prevent diaphragmatic hernia.
Percutaneous renal artery embolization has been in clinical practice since the 1970s and is currently advocated as a preoperative or palliative adjunct in the management of RCC . We advocate embolization in our management of larger renal tumours with hypervascular characteristics. By incorporating 3D imaging, multiple vessels can be identified and embolized before surgical intervention. The procedure itself carries little risk, with pain, haematuria, transient hypertension and fever being reported afterward, the last of which is attributable to cytokine cascades . Embolization can be achieved with Gelfoam (Pharmacia, Kalamazoo, MI, USA) sponges, wire coils, polyvinyl foam or absolute ethanol, the latter being the preferred choice at the authors’ institution. Embolization is often done 7–10 days before surgery, with an interim imaging study. Arterial embolization affords the surgeon the opportunity to take the renal vein before the artery and minimizes blood loss during mobilization at the superior and inferior margins.
TREATMENT OPTIONS FOR LOCALLY ADVANCED DISEASE: RCC WITH CAVAL TUMOUR THROMBUS
In the absence of nodal or metastatic disease, patients with caval tumour thrombus have acceptable disease-free survival rates . Although seldom reported, varicocele, lower extremity oedema, deep vein thrombosis, proteinuria, recurrent pulmonary emboli and caput medusae, should raise the index of suspicion. The surgical technique for thrombus involving the renal vein is similar to RN. The complexity of the procedure increases with cephalad extension, at times requiring CPB-DHCA and transplantation techniques for liver mobilization and caval exposure. Preoperative embolization should also be considered to limit blood loss and allow the artery to be taken after to the renal vein.
Our institution has performed >200 RN in cases requiring IVC thrombectomy. Cancer-specific survival differences, when stratified by level of tumour thrombus (renal vein vs caval) lend support to the need for revision of the TNM 2002 classification system (Fig. 7). A subgroup survival analysis of patients undergoing CPB-DHCA at our institution for pT3b and pT3c disease lend further support for the role of RN and IVC thrombectomy for extensive tumour thrombus in the absence of overt metastases (Fig. 8) .
The role of RN for metastatic disease is limited to palliation, enrolment in experimental protocols and in the presence of solitary metastatic lesions, often pulmonary nodules. Studies are beginning to suggest that the cytoreductive surgery in combination with interleukin-2 immunotherapy and autologous tumour vaccines might provide a survival advantage .
Lymphadenectomy in the treatment of RCC is controversial, but the prognostic role is without question. The boundaries for the procedure are the crus of the diaphragm to the aortic bifurcation. The presence of disease in the lymph nodes portends a poor prognosis for patients with RCC, with an estimated 5-year survival of 11–35%. Histologically confirmed false-positive findings with CT have been estimated at 40% by Studer et al., while false-negatives were 4%. There are many reports of retrospective studies that do and do not show a survival advantage for regional lymphadenectomy. The only prospective randomized controlled study (EORTC 30881) failed to show a survival difference between patients treated with and without lymph node dissection at the 5-year follow-up .
The 6th edition of the American Joint Committee on Cancer Staging Manual, updated in 2002 , specified that histological examination should include eight or more nodes for regional lymphadenectomy. In addition, there is no definitive imaging method that can differentiate between hyperplastic and metastatic nodes . There are numerous reports advocating a specific number of nodes to be removed, but it is often the discretion and experience of the surgeon that dictates the extent of lymphadenectomy [30,31]. At our institution, we perform the procedure to provide a more accurate prognosis for our patients.
Several retrospective studies have assessed the incidence of adrenal metastasis after RN and have come to similar conclusions. In general, the incidence of adrenal metastasis is 2–10% in the absence of widespread disease . Autorino et al. showed that tumour stage correlated with the probability of adrenal involvement. In this same review, the authors assessed the results of 13 studies involving 5351 patients, and in all but two of the studies was adrenalectomy recommended at all times. Many groups assessed the preoperative efficacy of CT in predicting adrenal involvement and determined the study to have sensitivity and specificity of 87.5–100% and 76–98%, respectively. Our personal preference is to remove suspicious adrenals found on preoperative imaging or during the procedure. We also remove the gland for larger upper-pole tumours. We have begun to adopt adrenal-sparing RN, recognizing the importance of adrenocortical reserve to minimize hormonal disorders .
COMPLICATIONS OF RN
Due to the well-protected retroperitoneal location of the kidney and closeness to vascular, endocrine and digestive organs, the rates of complications can be high with cavalier approaches. In many instances failure to appreciate the variations in anatomy, either congenital or acquired, result in complications. The choice of incision can be paramount during RN, to minimize the degree of dissection. The incidence of pneumothorax in cases other than thoracoabdominal approaches, where the pleura is intentionally entered, is 1.5–25%. Incisions at or above the 11th rib carry a greater risk of pneumothorax, and the latter must be excluded before closure. With the fascia closed, the wound is filled with irrigant and the lungs hyper-inflated. Bubbles will indicate a leak and the pleura should be closed with 4/0 chromic suture, unless a large defect is noted, and then a formal chest tube is placed.
Liver lacerations during right-sided RN are often attributable to retraction, and should be recognized immediately and repaired. For more superficial lesions one can use electrocautery or the argon-laser beam for fulguration. For urologists comfortable with the right upper quadrant, a Pringle manoeuvre can be used to minimize bleeding during extensive repairs; otherwise, a general or hepato-biliary operative consultation should be sought. The spleen is apt to injury in larger left-sided tumours and has been estimated to occur in 5–10% of cases, with intraperitoneal approaches incurring higher rates . Superficial packing with electrocautery, Surgicel (Ethicon, Cornelia, GA, USA) or argon laser can be used to treat minor tears. Ongoing haemorrhage is best managed with splenectomy . Overzealous mobilization of the pancreas during left-sided RN can led to pancreatic ductal injury. These injuries are best managed with distal pancreatectomy. Manipulation can also lead to pancreatitis and these patients should be managed conservatively with hyper-alimentation when appropriate.
Injury to the colon and associated mesocolon are rare unless a significant degree of inflammation or tumour extension is present. Mechanical bowel preparations are a prudent choice for selected cases. In most instances, colonic injury can be repaired in a standard two-layered closure with generous irrigation and drains left in place. Duodenal injuries likewise should be closed in a standard two-layer fashion with a Lembert suture on the outside.
Vascular injuries for the most part are recognized during RN and managed then. Delayed arterial bleeding is a very worrisome complication and might require secondary surgery. Embolization is often a futile effort, as the main concern is bleeding from the renal artery stumps, which are take close to the aorta. One particular area of interest for the urologist is the closeness of the left renal artery to the superior mesenteric artery, which is often a fatal mistake if taken in place of the renal artery.
MINIMALLY INVASIVE AND NEPHRON-SPARING SURGERY
Both practising urologists and those in training need to be flexible with respect to treatment options for RCC, given the promising results of minimally invasive techniques, e.g. laparoscopic partial nephrectomy, radiofrequency ablation, cryoablation, high-intensity focused ultrasound, microwave ablation, and laser-induced thermotherapy.
The increased incidence of early-stage RCC has enabled urologists to use minimally invasive techniques on smaller tumours, preserving renal parenchyma and function. Steinbach et al. reported that up to 3.4% of patients undergoing RN develop a contralateral tumour. By minimizing operative times and lessening recovery periods, minimally invasive approaches become feasible options for sicker patients unable to tolerate open procedures.
Laparoscopic partial nephrectomy has gained increased attention in the treatment of tumours of <4 cm that are preferentially located along the periphery. Fergany et al. reported a disease-specific survival of 97% at 42 months in their series from the Cleveland Clinic. Centres of excellence should serve as a referral base for community urologists with limited laparoscopic experience, given the difficulty of learning and the lack of an established technique. Oncological and functional outcomes of laparoscopic partial nephrectomy provide solid evidence that laparoscopic partial nephrectomy serves as an excellent alternative to open procedures in properly screened patients .
Cryoablation has the longest track record of the ablative techniques currently in use, and can be done as an open, laparoscopic or percutaneous procedure. The lack of histopathological evidence of tumour eradication remains an area of controversy within ablative therapy. Proponents of the procedure cite the increased sensitivity of MRI in detecting recurrence. The long-term follow-up of numerous ongoing trials will have a dramatic impact and should be of interest to all urologists.
Energy-ablative approaches have emerged after good clinical outcomes were achieved in prostate and liver diseases. The approaches use heat to induce irreversible cellular damage. The mode of delivery is different for radiofrequency ablation, high-intensity focused ultrasound and laser-induced thermotherapy, but the physiological impact is essentially the same. The lack of pathological evaluation has created scepticism and that might abate with a longer follow-up. Drs Kaouk, McDougal and Marberger provide superb reviews of these methods elsewhere in this issue.
IMMUNOTHERAPY FOR RCC
The poor response rates of RCC to traditional chemotherapy and radiotherapy have enabled immunotherapy to take centre stage as an alternative or adjunct to surgical intervention. Collective data from ongoing clinical trials evaluating cytokine-based therapy for metastatic RCC have shown a partial response in 15% and a complete response in 5%. Interferon-α and/or interleukin-2, both cytokine-based immunotherapies, constitute the most effective currently available agents. Both agents result in non-specific activation of T lymphocytes (CD4 and CD8) and natural killer cells. Another promising avenue, adoptive cellular therapy, relies on ex vivo activation of lymphocytes and killer cells with subsequent reinfusion. Stem cell transplantation is another option due to the highly immunogenic nature of RCC.
The good news was the approval of sunitinib maleate and sorafenib earlier this year for metastatic RCC. Both sunitinib and sorafenib offer several mechanisms of activity, including inhibition of vascular endothelial growth factor and platelet-derived growth factor . Inhibition of the former might have a more profound impact, given the increased rates of angiogenesis associated with vascularized malignancies like RCC. These advances need to be embraced by urologists, who might be called upon more frequently for cytoreductive surgery in the future. Please see the accompanying chapters on medical therapy for RCC in this issue for more in-depth reporting.
The role of RN in treating RCC has and will be revised in the future as the demand for less-invasive and renal-sparing procedures continues to rise. A sound understanding of the surgical anatomy and technique required to perform a RN need to be part of any urologist’s skills; these skills will facilitate difficult cases and enable surgeons to adapt to the increasing variations of procedures. We have had great success with RN in treating a wide range of RCC and hope that this review will serve as guide for our readers.
We thank Francis E. Steckel, Art Editor and Principal Illustrator of the Lahey Clinic Medical Center. Mr Steckel has provided our clinical departments with an extraordinary array of illustrations that have and will remain a centrepiece of academics. We are saddened by his loss this past year but look forward to his remembrance by incorporating his collections into future literature.