Renal cancer surgery: historical considerations

Successful attempts to surgically cure renal tumors were reported widely after World War II, with surgical strategies designed to address the renal capsular and perinephric fat infiltration observed in up to 70% of the tumors. Using a thoracoabdominal incision, Mortensen reported the first radical nephrectomy in 1948, an operation that removed all of the contents of Gerota fascia. Radical nephrectomy was popularized in the 1960s by Robson, who described the perifascial resection of the tumor-bearing kidney with its perirenal fat, regional lymph nodes, and ipsilateral adrenal gland. In 1969, Robson and colleagues reported RN results in a series of 88 patients and described a 65% survival rate for tumors confined within Gerota fascia (Robson stage 1 and 2), but the finding of regional nodal metastases led to a less than 30% 5-year survival rate.

RN, ipsilateral adrenalectomy, and extensive regional lymphadenectomy, usually through large abdominal or transthoracic incisions, became the standard approach to renal tumors for the next 20 years as major centers began reporting favorable results. In this era, the imaging studies used to diagnose a patient, who was usually symptomatic with a large renal tumor, was intravenous urograms, retrograde pyelograms, and arteriograms. These techniques were unable to detect small tumors and the incidental tumor detection rate was less than 5%. Despite this acceptance of RN by urologic surgeons, convincing data did not exist establishing the therapeutic impact of the component parts of the operation (ie, the need for adrenalectomy or the need for and extent of lymph node dissection). Historical series were subjected to selection biases, and the virtues of randomized trials in clinical investigation to address the many questions in kidney tumor surgery were not yet realized. Although a subsequent report with longer follow-up by Robson in 1982 projected declining long-term survival rates in the range of 40%, there was no doubt that the surgical techniques associated with the safe removal of large renal tumors were established, well described, and reproducible, making RN the only effective treatment for RCTs. Today, at major centers with a commitment to renal tumor surgery, despite the previously described imaging induced stage and tumor size migration, RN is still required in approximately 20% to 30% of patients with renal tumors not amenable to kidney-sparing approaches.

In the past 20 years, surgical oncologists have increasingly used organ-sparing and limb-sparing approaches to treat malignancies, such as breast cancer and sarcoma, often with adjuvant chemotherapy or radiation therapy, with equivalent survival rates to their more radical predecessors. This approach was also applied to RCTs using PN, first for tumors of 4 cm or smaller and recently for tumors of 7 cm or smaller, with equivalent cancer-specific survival rates. Approximately 20% of resected renal tumors will be benign (ie, angiomyolipoma, oncocytoma, metanephric adenoma, or hemorrhagic cyst), 25% will be indolent (papillary, chromophobe carcinoma) with limited metastatic potential, and 55% will have the more potentially malignant conventional clear cell carcinoma that accounts for 90% of the metastatic renal tumors. Even for patients with T1 conventional clear cell carcinomas, a 90% long-term survival is anticipated. Using preoperative predictive nomograms coupled with the emerging view that preservation of renal function, particularly in elderly or comorbidly ill patients, is equally if not more important to the patient’s long-term health than resection of a small renal mass, the use of active surveillance strategies is increasingly accepted.

The introduction of laparoscopic RN by Clayman and colleagues in 1991 began a new era of minimally invasive kidney surgery. McDougal and colleagues investigated minimally invasive PN in the laboratory setting using a pig model, and in 1993 Winfield and colleagues successfully performed the first laparoscopic partial nephrectomy (LPN). LPN is a technically demanding operation being increasingly used with success. Advances in minimally invasive instrumentation, including robotic assistance and dual-teaching counsels, coupled with sophisticated training simulators, may enhance the use of these techniques in the future.

Renal tumor surgery: patient selection and preoperative evaluation

To a large extent, the modern imaging studies of CT, ultrasound, and MRI that have been so effective in creating the era of the “incidentaloma” and the associated tumor size and stage migration, also provide the surgeon with an accurate description of the extent of disease before operation. At Memorial Sloan Kettering Cancer Center (MSKCC), tumors routinely selected for RN include those large and centrally localized tumors that have effectively replaced most of the normal renal parenchyma, often associated with regional adenopathy and renal vein extension. In approximately 10% of patients with renal cell carcinoma, tumor involvement of major branched renal veins, the main renal vein, or extension into the inferior vena cava occurs. In the vast majority of these cases, preoperative imaging reveals the tumor thrombus and allows the surgical team to carefully plan its approach. In most cases, resection without venovenous bypass and cardiac bypass is possible with the prognosis directly related to the extent of disease. For patients with right atrial or ventricular renal tumor thrombus, open heart bypass with or without hypothermic arrest is required to complete the resection. Perioperative mortality can be up to 10% in those cases and completeness of the thrombectomy depends largely on whether the tumor thrombus is free floating in the caval lumen or has infiltrated the caval wall. For patients without regional nodal or distant metastatic disease, long-term survival is possible in 50% to 60% of patients, whereas the presence of either reduces survival to that of those patients undergoing cytoreductive nephrectomy. In addition, RN is performed on patients with metastatic disease referred by medical oncology for cytoreductive nephrectomy before the initiation of systemic therapy. Patients with extensive metastatic disease and a poor Karnofsky Performance Status are advised to undergo percutaneous needle biopsy of the primary tumor or a metastatic site and are subsequently referred for systemic therapy.

Although PN was initially restricted to patients with essential indications (tumor in an anatomic or functional solitary kidney, bilateral renal tumors, hereditary renal tumor syndromes), reports from the United States and abroad indicated that PN was not compromising the local tumor control or survival when compared with RN for patients with T1a renal tumors (4 cm or smaller) across all histologic subtypes with progression-free survival and metastasis-free survival the same whether PN or RN was performed. The rationale for expanding the indications for PN to larger RCTs of 4 to 7 cm was articulated and initial reports were met with similarly favorable results both in the United States and Europe. A recently published report that combined the Mayo Clinic and MSKCC databases, evaluated 1159 patients with renal tumor between 4 and 7 cm treated with RN (n = 873, 75%) and PN (n = 286, 25%) and demonstrated no significant difference in survival between the groups. We recently reported a series of even larger PN for T2 disease performed in 34 patients. Interestingly, 6 patients (16.2%) had benign and 12 patients (35%) had indolent (papillary or chromophobe) pathology. After 17 months of follow-up, 71% of patients with a malignant diagnosis are alive without evidence of disease. Although the resection of massive renal tumors (in this series up to 18 cm in diameter) is largely a function of favorable tumor location and careful case selection, this study indicates that local tumor control can be effectively achieved with results similar to those found in series of tumors of 7 cm or smaller. Mayo Clinic investigators recently reported similar results in 276 patients with clinical stage higher than T2 treated with either PN (n = 69) or RN (n = 207). These data indicate clearly that the ultimate oncological threat of a given tumor depends on biologic factors of the tumor, including histologic subtype, the presence or absence of symptoms, tumor grade, and tumor size with prognostic nomograms and algorithms available to predict outcomes with reasonable accuracy. PN is clearly on firm oncological footing for T1 renal tumors. Over the past 3 years at our center, 1030 surgical nephrectomies have been performed with 21% RN and 79% PN, reflecting our center’s commitment to kidney-sparing approaches whenever possible.

Before the operation, routine serum chemistries, coagulation profile, type and cross match (or autologous blood donation), and chest x-ray are obtained. Routine brain imaging and bone scanning are not performed unless site-specific abnormalities in the history, physical, or routine preoperative laboratory examination are discovered. For patients with significant comorbid conditions, particularly cardiac and pulmonary related, appropriate consultations are obtained with an effort made to optimize patients for operation whenever possible. Patients with significant coronary or carotid artery disease may require revascularization before operation. For patients with compromised pulmonary status, consultation with anesthesiology is requested for consideration of epidural postoperative analgesia.

Renal tumor surgery: patient selection and preoperative evaluation

To a large extent, the modern imaging studies of CT, ultrasound, and MRI that have been so effective in creating the era of the “incidentaloma” and the associated tumor size and stage migration, also provide the surgeon with an accurate description of the extent of disease before operation. At Memorial Sloan Kettering Cancer Center (MSKCC), tumors routinely selected for RN include those large and centrally localized tumors that have effectively replaced most of the normal renal parenchyma, often associated with regional adenopathy and renal vein extension. In approximately 10% of patients with renal cell carcinoma, tumor involvement of major branched renal veins, the main renal vein, or extension into the inferior vena cava occurs. In the vast majority of these cases, preoperative imaging reveals the tumor thrombus and allows the surgical team to carefully plan its approach. In most cases, resection without venovenous bypass and cardiac bypass is possible with the prognosis directly related to the extent of disease. For patients with right atrial or ventricular renal tumor thrombus, open heart bypass with or without hypothermic arrest is required to complete the resection. Perioperative mortality can be up to 10% in those cases and completeness of the thrombectomy depends largely on whether the tumor thrombus is free floating in the caval lumen or has infiltrated the caval wall. For patients without regional nodal or distant metastatic disease, long-term survival is possible in 50% to 60% of patients, whereas the presence of either reduces survival to that of those patients undergoing cytoreductive nephrectomy. In addition, RN is performed on patients with metastatic disease referred by medical oncology for cytoreductive nephrectomy before the initiation of systemic therapy. Patients with extensive metastatic disease and a poor Karnofsky Performance Status are advised to undergo percutaneous needle biopsy of the primary tumor or a metastatic site and are subsequently referred for systemic therapy.

Although PN was initially restricted to patients with essential indications (tumor in an anatomic or functional solitary kidney, bilateral renal tumors, hereditary renal tumor syndromes), reports from the United States and abroad indicated that PN was not compromising the local tumor control or survival when compared with RN for patients with T1a renal tumors (4 cm or smaller) across all histologic subtypes with progression-free survival and metastasis-free survival the same whether PN or RN was performed. The rationale for expanding the indications for PN to larger RCTs of 4 to 7 cm was articulated and initial reports were met with similarly favorable results both in the United States and Europe. A recently published report that combined the Mayo Clinic and MSKCC databases, evaluated 1159 patients with renal tumor between 4 and 7 cm treated with RN (n = 873, 75%) and PN (n = 286, 25%) and demonstrated no significant difference in survival between the groups. We recently reported a series of even larger PN for T2 disease performed in 34 patients. Interestingly, 6 patients (16.2%) had benign and 12 patients (35%) had indolent (papillary or chromophobe) pathology. After 17 months of follow-up, 71% of patients with a malignant diagnosis are alive without evidence of disease. Although the resection of massive renal tumors (in this series up to 18 cm in diameter) is largely a function of favorable tumor location and careful case selection, this study indicates that local tumor control can be effectively achieved with results similar to those found in series of tumors of 7 cm or smaller. Mayo Clinic investigators recently reported similar results in 276 patients with clinical stage higher than T2 treated with either PN (n = 69) or RN (n = 207). These data indicate clearly that the ultimate oncological threat of a given tumor depends on biologic factors of the tumor, including histologic subtype, the presence or absence of symptoms, tumor grade, and tumor size with prognostic nomograms and algorithms available to predict outcomes with reasonable accuracy. PN is clearly on firm oncological footing for T1 renal tumors. Over the past 3 years at our center, 1030 surgical nephrectomies have been performed with 21% RN and 79% PN, reflecting our center’s commitment to kidney-sparing approaches whenever possible.

Before the operation, routine serum chemistries, coagulation profile, type and cross match (or autologous blood donation), and chest x-ray are obtained. Routine brain imaging and bone scanning are not performed unless site-specific abnormalities in the history, physical, or routine preoperative laboratory examination are discovered. For patients with significant comorbid conditions, particularly cardiac and pulmonary related, appropriate consultations are obtained with an effort made to optimize patients for operation whenever possible. Patients with significant coronary or carotid artery disease may require revascularization before operation. For patients with compromised pulmonary status, consultation with anesthesiology is requested for consideration of epidural postoperative analgesia.

Open renal tumor surgery: surgical anatomy and operative considerations

The kidneys are retroperitoneal organs located in the lumbar fossa. They are covered with a variable amount of perinephric fat and Gerota fascia and lay in proximity to the psoas major and quadratus lumborum muscles, and diaphragm. Depending on the patient’s body habitus, the size of the tumor, and its relationship to adjacent organs such as the colon, duodenum, pancreas, and spleen, different surgical incisions can be used including subcostal, thoracoabdominal, 11th rib flank, and midline abdominal. Transabdominal approaches are preferred for massive tumors with regional adenopathy and renal vein and inferior vena caval extension because access to major renal vessels and aorta and inferior vena cava can be optimized. Early ligation and division of the renal artery effectively decrease blood flow to the tumor-involved kidney, decompress fragile and engorged tumor parasitic vessels, and enhance the mobility of the kidney during its resection, all of which can facilitate the tumor resection and decrease intraoperative bleeding. This approach negates the need for preoperative renal artery embolization, which can be a painful and costly procedure. Before its division, the renal vein should be inspected and gently palpated to exclude a renal vein tumor thrombus. For smaller tumors requiring an RN and for all PNs, a “miniflank incision” has the advantage of speedy entry into the retroperitoneum and avoidance of the rib resection with a decreased likelihood (<5%) of subsequent atony of the flank muscles and bulge.

Although the traditional RN described by Robson included ipsilateral adrenalectomy and regional lymph node dissection, no convincing evidence exists that these component parts offer a therapeutic advantage. Contemporary survival data indicate that the presence of lymph node metastasis and/or adrenal metastasis has a similar negative prognostic impact to distant metastatic disease with median survival rates of less than 12 months. For patients undergoing RN for massive renal tumors, the rationale for ipsilateral adrenalectomy and regional node dissection is to maximize local tumor control and provide accurate pathologic staging information that may lead to entry into adjuvant systemic therapy clinical trials using the newly developed mTOR and tyrosine kinase inhibitors. It is our practice to perform ipsilateral adrenalectomy and regional node dissection to maximize local tumor control, decrease the chance of local recurrence if these tissues are harboring micrometastatic disease, and provide maximum pathologic staging to allow entry into ongoing adjuvant clinical trials.

Once the tumor resection is accomplished, postoperative nomograms are available that incorporate clinical presentation, tumor histologic subtype, size, and stage to provide a clinical prognosis. Results for our center indicate 5-year survival rates following resection of nonmetastatic tumors range from 30% to 98% depending on the previously mentioned clinical and pathologic features. These postoperative nomograms have been extremely useful in patient counseling, tailoring cost-effective follow-up strategies, and designing clinical trials.

Minimally invasive approaches to early-stage renal tumor

Following its introduction in 1991 by Clayman and colleagues, laparoscopic radical nephrectomy (LRN) offered a minimally invasive alternative to the classical open RN with dividends of less wound pain and morbidity, decreased analgesic requirement, decreased hospitalization, more rapid convalescence, and faster return to normal activities. Survival rates were directly comparable with those achieved with open RN. At the time of its introduction, RN in general was the preferred treatment for all renal masses and PN was largely reserved for only essential cases where RN would put a patient at risk for dialysis.

However, at centers with expertise in both open and minimally invasive renal surgical approaches, published experiences revealed inconsistencies in the management of small renal tumors. Open surgeons were more likely to perform PN, and laparoscopic surgeons were more likely to perform RN. These reports suggested that minimally invasive surgery learning curves were being conquered by RN applied to small renal tumors (<4 cm) despite the previously described clinical data that this was surgical overkill and deleterious to the patients’ overall renal function. Unique issues relative to minimally invasive renal surgery, such as the problem of tumor-bearing kidney retrieval (ie, morcellation vs an open extraction incision for removal) were debated in the literature. The case load required to extend the surgical limits for minimally invasive surgery was not known and the decision to perform open versus minimally invasive surgery kidney procedures more often depended on the relative surgical expertise of the individual surgeon rather than more clearly defined guidelines relating to tumor size and concerns about future renal function.

By 2000, because of the emerging literature supporting elective PN, several minimally invasive surgery groups began concerted efforts to develop laparoscopic PN techniques intending to closely simulate the open procedure, initially with smaller, exophytic renal tumors and, with time and increasing experience, to more complex centrally located or cystic renal tumors. Valiant attempts to duplicate the renal protective effects of cold ischemia, readily obtained in open PN, were reported and included cold renal arterial and ureteral perfusions, and recently, laparoscopic ice slush placement. Nonetheless, the vast majority of LPNs (many of which are complex) continue to be performed under warm ischemic conditions with the hope that rapid completion of the operation will limit any long-term ischemic effects on the kidney. Even for these expert surgeons, laparoscopic PN is currently described as a “complex” or “advanced” operation with published complication rates that are 3 to 4 times higher than for their open counterparts. Interestingly, LPN teams report similar rates of resected benign lesions (20%–30%) and similar beneficial effects on overall renal function, as described previously in the open PN experience.

Investigators from the Mayo Clinic, Cleveland Clinic, and Johns Hopkins pooled their data on 1800 PNs of which 771 were LPNs and 1028 were open PNs for T1 tumors from 1998 to 2005. Even though the surgeons at these centers were experts in their respective laparoscopic and open operations, careful case selection was apparent. Patients who had open PN had larger tumors that were more likely centrally located and malignant, were at higher risk of perioperative complications as defined by their older age, increased comorbidities, decreased performance status, and decreased baseline renal function, all of which may have contributed to longer hospital stays for open PN (5.8 days) versus LPN (3.3 days). Patients in the LPN group were more likely to have elective indications rather than essential indications for PN, yet LPN was associated with longer ischemic time, more postoperative complications, particularly urologic, and increased number of subsequent procedures to treat complications. This comprehensive study leaves little doubt that the LPN is a technically challenging operation, even in the hands of such experts, where careful case selection may decrease the chance of surgical and urological complications.

As minimally invasive surgeons gain more experience and instrumentation improves, coupled with further refinements in case selection, it is expected that complications related to LPN will decrease and larger tumors and those in the near the renal hilum will be increasingly resectable. Surgeons are increasingly using robotic-assisted techniques to perform RN and PN with outcomes related to estimated blood loss and complications similar to classical laparoscopic techniques but, not surprisingly, with operating time and costs significantly greater in the robotic-assisted cases. At this time, despite the published enthusiasm of many surgeons, little evidence exists that robotic techniques have a substantial advantage over standard laparoscopic techniques.

Renal tumor ablative modalities, including percutaneous and laparoscopic approaches to radiofrequency ablation (RFA) and cryoablation, are offered selectively to some patients with renal tumors that are exophytic and not encroaching upon renal hilar vessels or collecting system elements. Patients considered by many as ideal candidates for ablation are often old or comorbidly ill individuals harboring small renal tumors, the very patients ideally suited for active surveillance. Although the concept of nonsurgical ablation is appealing, the literature has serious deficiencies, including up to 40% of patients not having preablation confirmation of tumor owing to nondiagnostic or nonexistent biopsy, short overall follow-up, and high rates of tumor recurrence compared with PN ranging from 7.45-fold to 18.23-fold greater for RFA and cryotherapy respectively. Additionally, because most studies lacked pathologic confirmation to confirm the completeness of the ablation, it is not known whether changes in radiological images after ablation represent complete or partial tumor destruction or simply a renal tumor, partially treated and not in active growth. The Cleveland Clinic experience reports difficult salvage operations after failed ablations, which usually lead to RN as the final outcome in a patient population initially candidates for either active surveillance or PN, an outcome that now must be viewed as unfavorable from both oncological and renal functional points of view. Carefully designed “ablate and resect” clinical protocols need to be done, much like those done in the 1990s for cryotherapy in localized prostate cancer, to determine the true effectiveness of these approaches.