Cryoablation

Percutaneous cryoablation (PCA) was first performed by Uchida and colleagues in 1994 on canine kidneys. It was not until 1995 that Uchida’s group described the initial application of PCA in two patients who had metastatic RCC, resulting in a decrease in their tumor size by 20% and 81%, respectively. The practical limitations of cryotechnology at that time, namely, comparatively larger (6.8 mm) probes that used an exclusively liquid nitrogen medium, detracted from the interest and accessibility in applying this approach more broadly. The advent of smaller (2.4 mm) probes that use an argon gas medium, which allows for increased treatment precision by halting ice ball growth at the termination of the freeze, has allowed this technology to become more amenable to percutaneous applications. It is with these improvements and concurrent advances in real-time imaging modalities that we have seen a renewed interest in cryoablation as a formidable and more available treatment option for renal mass lesions.

Cryoablation is the best known and most well studied of all thermal ablative modalities. After renal mass biopsy, vacuum-insulated liquid nitrogen or argon-cooled probes are inserted directly into the renal lesions and an ice ball is rapidly generated around this core as these media are forced through a small aperture at the tip of the probe. The ensuing freeze-thaw cycles lead to complete and reproducible necrosis of renal parenchyma occurring at temperatures of −19.4°C or lower. Rapid freezing causes cytotoxic intracellular and extracellular ice crystal formation leading to an increased extracellular osmotic concentration, which ultimately results in intracellular pH changes and protein denaturation and extracellular mechanical disruption of the cell membrane. Acute injury to the vasculature with resultant thrombosis and ischemia, particularly at the microcirculatory level, leads to delayed tissue necrosis over the ensuing hours to days after cryotherapy. In addition to these mechanisms, apoptosis and freeze-induced immunologic sensitization are other means by which tumor cell destruction occurs.

Chosy and colleagues found that normal renal parenchyma needed exposure to a temperature at or lower than −19.4°C for complete cell destruction. Given that cancer cells may need even lower temperatures for reliable cell death, most clinical protocols err on the side of caution and freeze to −40°C. In experimental settings, these temperatures were reached only 3.1 mm inside the edge of the ice ball. As a result, in clinical practice, the ice ball is usually extended 5 to 10 mm beyond the tumor margin. The end points of temperature and ice-ball monitoring can be determined by thermocouples at the tumor margin and real-time intraoperative ultrasound (US), CT, or MRI, respectively. In addition, most large clinical series use a double freeze-thaw cycle, which has demonstrated greater clinical efficacy when compared with a single cycle.

Fortunately, not all renal structures are equally cryosensitive. This was demonstrated by Sung and colleagues, who observed that in the absence of mechanical laceration with the cryoprobe, the collecting system heals in a watertight manner even when encompassed within the ice ball. Several researchers have duplicated these findings but have also discovered that hyperthermal ablation techniques (ie, RFA) were associated with an increased risk for urinary leak and fistula formation when compared with hypothermal cryoablation. Proponents of RFA believe that the increased rate of urinary leak and fistula formation is secondary to the multiple tines used to measure impedance and monitor the ablation. Theoretically, the increased number of tines (up to 10) surrounding each electrode, as opposed to two to three cryoprobes, may make it easier to enter the collecting system, and ultimately damage it during ablation.

Currently, renal cryoablation is performed by one of two main minimally invasive approaches: laparoscopic cryoablation (LCA) and PCA. In LCA, the lesions can be approached in a transperitoneal or retroperitoneal fashion. In terms of facilitating exposure, tumors in an anterior or medial location are generally best approached in a transperitoneal manner, whereas posteriorly located tumors lend themselves to a retroperitoneal approach. LCA allows for direct intraoperative visualization, thereby permitting precise cryoprobe positioning and monitoring of ice-ball formation with the aid of a deflectable laparoscopic US transducer. The US scan depicts the ice ball as hypoechoic with an advancing hyperechoic crescent with a posterior acoustic shadow. Once cryoablation has been accomplished and the probe has been withdrawn, hemostasis can be achieved by local compression with bioadhesive bolsters, argon beam coagulation, or a combination of both.

Laparoscopic Cryoablation

Davol and colleagues reported their clinical experience with cryoablation, using an open technique in the first 24 patients and subsequently evolving their surgical method into LCA for the next 24 patients in their data set. The median tumor size in these 48 patients was 2.6 cm (range: 1.1–4.6 cm) with a median follow-up of 64 months (range: 36–110 months). The overall and cancer-specific survival rates were 89.5% and 100%, respectively. The disease-free survival rate after a single cryoablation procedure was 87.5%, which improved dramatically to 97.5% after a repeat procedure. There were no major adverse events and seven minor adverse events, which included four capsular fractures, 2 patients requiring postoperative packed red blood cell transfusions, and 1 patient who had postoperative ileus.

Schwartz and colleagues presented their data using cryoablation, again, first with an open approach and subsequently with a laparoscopic approach. In the 85 consecutive patients who were treated at multiple centers, 70 procedures were performed in a laparoscopic fashion. The mean tumor size in this cohort was 2.6 cm (range: 1.2–4.7 cm), with a mean follow-up of 10 months (range: 3–36 months). The mean estimated blood loss (EBL) was 33 mL (range: <50–550 mL), with 2 patients requiring packed red blood cell transfusion. A total of seven laparoscopic cases were converted to an open procedure, two of which were considered technical failures. Abnormal postoperative enhancement occurred in 2 patients at 3 and 12 months. Radical nephrectomy in the first patient revealed no viable tumor, whereas needle biopsy in the second patient demonstrated RCC, which prompted nephrectomy.

Lawatsch and colleagues reported on their experience with LCA of 81 tumors in 59 patients. The median tumor size in this group was 2.5 cm (range: 1.0–5.0 cm), with a median follow-up of 27 months. Mean pre- and postoperative creatinine levels were stable at 1.11 and 1.17 mg/dL, respectively. Two patients required open conversion, one of which required nephrectomy for bleeding. There were two recurrences identified by residual enhancement of the cryolesion on a contrast-enhanced CT scan. Both of these patients underwent salvage nephrectomy and had no evidence of disease (NED) at the time of publication.

Cestari and colleagues presented data on 104 patients who underwent LCA with a mean tumor size of 2.2 cm (range: 0.7–6.0 cm). The mean operating time was 202 minutes (range: 90–320 min), with a mean EBL of 212 mL (range: 10–3200 mL). Thirty-six and 11 patients have had follow-up for 5 and 7 years, respectively. Delayed complications included one case of ureteropelvic junction (UPJ) obstruction requiring open pyeloplasty 8 months after surgery. One patient underwent open radical nephrectomy at 12 months for suspected tumor recurrence. Progressive reduction of the cryoablated lesions on MRI scans was demonstrated in all patients, with only a renal scar visible after 24 months of follow-up.

Aron and colleagues recently reported an updated series from Gill and colleagues on long-term (5-year minimum) outcomes after LCA in 82 patients who had sporadic single renal mass lesions. Mean tumor size was 2.3 cm (range: 0.9–5.0 cm), with a median follow-up of 83 months (range: 60–120 months). Follow-up consisted of MRI scans performed on postoperative day 1; at months 3, 6, and 12; and then annually thereafter. CT-guided biopsy of the cryolesion was performed 6 months after surgery and then repeated if MRI findings were abnormal. Of the 88 patients who underwent LCA and met the criterion of at least 5 years of follow-up, the mean operating time was just greater than 3 hours (range: 75–420 minutes), with a mean EBL of 106 mL (range: 10–800 mL). The 5-year actual overall, cancer-specific, and disease-free survival was 83%, 95%, and 78%, respectively, whereas the estimated Kaplan-Meier cancer-specific survival at 10 years was a promising 88%.

Modeled after the established principles of its open surgical counterpart, laparoscopic partial nephrectomy (LPN) has emerged as the main MINSS technique to address the SRM. Although the cytocidal effect and durability of LCA seem promising, the fundamental difference between these two MINSS procedures is that whereas LCA leaves the ablated tumor in situ, LPN is extirpative. It is within this clinical context that Desai and colleagues compared the perioperative and short-term outcomes of LPN (group 1, n = 153) versus LCA (group 2, n = 78) in patients who had small renal tumors (≤3 cm). LPN was associated with greater blood loss (211 vs 101 mL) and a higher incidence of delayed complications after discharge (16.3% vs 2.2%). Both groups were comparable with regard to operating time, intraoperative and postoperative complications, hospital stay, convalescence, and postoperative serum creatinine level. Local recurrence was detected over a mean follow-up of 5.8 months in group 1 (0.6%) and 24.6 months in group 2 (3%).

Expounding on Desai’s work, Hruby and colleagues compared 12 LPNs with 11 LCAs in the management of small tumors (≤3.2 cm) located in close proximity (≤5 mm) to the renal hilum. All 23 cases were successfully completed laparoscopically. The mean operating time for LPN and LCA was 2.8 hours and 2.3 hours, respectively. The mean EBL was 197 mL for LPN and 70 mL for LRC, whereas the mean hospital stay for the LPN and LCA groups was 3.9 days and 3.2 days, respectively. There were no apparent intraoperative complications in either group. Nine postoperative complications were seen in the LPN group (four of which were urine leaks), whereas the LCA group had no complications. No disease recurrence was seen in either group during the follow-up period of approximately 1 year.

Percutaneous Cryoablation

The percutaneous approach to cryoablation is even less invasive than LCA. In fact, several institutions have demonstrated that PCA can be successfully performed with just intravenous sedation and local anesthesia. This last point is under debate, however, because others believe that general anesthesia optimizes patient tolerance and allows greater control of respiratory motion during probe placement. In any case, PCA can be performed using real-time imaging modalities under US, CT, or MRI guidance. Although US is the most cost-effective and widely available of the imaging techniques, the attenuation of the ice ball limits cryoablation monitoring secondary to the posterior acoustic shadow. This factor precludes imaging deep to the superficial ice ball margin, which is an important consideration in terms of treatment efficacy and complications. As a result, PCA is currently performed with the use of open-gantry MRI or CT scan guidance. The requisite for an open-access MRI unit or fluoroscopy-capable CT scanner for intraoperative monitoring could potentially limit this technique in terms of widespread availability.

Gupta and colleagues reported on their clinical experience with CT-guided PCA in 27 renal tumors 5 cm or less in maximal diameter. The average tumor size was 2.5 cm (range: 1.0–5.0 cm), with 11 small (≤3 cm) and 5 large (>3 cm) tumors. The procedure was accomplished successfully in all patients using conscious sedation. Of the 20 treatments, there was only one complication (perinephric hematoma) requiring blood transfusion. Although mean follow-up imaging time was relatively short (5.9 months), 15 of 16 cryoablated lesions showed no residual enhancement. It should be noted that the single tumor that did show residual enhancement occurred in a patient who had a centrally located 4.6-cm tumor.

Silverman and colleagues presented their initial experience using MRI-guided PCA in 23 patients. The 26 treated renal tumors had a mean diameter of 2.6 cm (range: 1.0–4.6 cm). Using general anesthesia, 24 of 26 tumors were successfully ablated, with a mean follow-up of 12 months (range: 3–43 months). In two cases, a small enhancing nodule located at the ablation zone margin proved to be recurrent tumor. The researchers noted that intraprocedure CT fluoroscopy seems promising in ablating renal tumors in one session.

Sewell and Shingleton reported on their updated clinical experience in 103 patients treated with MRI-guided PCA over a 5-year period. A total of 120 tumors were cryoablated in a broad range of 103 patients, including those with solitary kidneys, multiple or bilateral tumors, and renal insufficiency. Tumor size ranged from 1.1 to 7.5 cm, with a mean follow-up of 35 months (range: 4–99 months). Overall and disease-specific survival was 90.2% and 97%, respectively. Based on their results, 72 tumors were completely ablated with a single procedure, with 18 tumors requiring a second cryoablation procedure, 4 tumors requiring a third cryoablation procedure, and 11 tumors requiring a fourth cryoablation procedure. Ten patients had residual disease (or recurrence) at the time of death; however, only three of these cases were disease-related. Although these results are promising in terms of overall and disease-specific survival, they also underscore the importance of patient counseling on the possibility for additional procedures to treat renal tumors adequately and effectively using this MINSS therapeutic modality.

Atwell and colleagues recently reported on their experience using CT-guided PCA in 110 patients. A total of 115 tumors with a mean tumor size of 3.3 cm (range: 1.5–7.3 cm) were treated. Of 90 renal mass biopsies performed, 52 (58%) showed RCC. All patients were admitted to the hospital after cryoablation, with most (87%) being discharged on postoperative day 1 (range: 1–12 days). There were seven major complications (6%) of 113 total procedures performed. Technical success was achieved in 97% of the treated tumors, with three failures occurring in centrally located tumors. Of these failures, 1 patient died while in hospice of nonrelated complex medical issues, 1 is being observed, and 1 has been lost to follow-up. There has been no local progression of 80 tumors (100% treatment success) followed for a mean of 13.3 months (range: 3–39 months). It should be noted that 25% of the tumors treated were 4.0 cm or larger, thereby making a case for technical feasibility in successfully ablating relatively larger tumors. The investigators also demonstrated that with the correct changes in patient positioning (eg, supine versus decubitus), and even fluid injection into the anterior pararenal space (ie, the creation of a “salinoma”), the overlying bowel can be displaced to treat anterior tumors successfully.

Complications of Cryoablation

The morbidity associated with cryoablation of renal tumors seems to be relatively low. With LCA, the most recent publication by Cestari and colleagues reported only 3 major complications of 104 cases treated: one perinephric hematoma that required a blood transfusion, one abscess treated successfully with percutaneous drainage, and one UPJ obstruction that required open pyeloplasty. Although the results from this Italian group are promising, a recent study by Lehman and colleagues reported that LCA of larger (≥3.0 cm) renal masses was associated with increased morbidity. This retrospective study of 44 procedures on 51 tumors stratified the subjects into two groups based on tumor size. Group 1 included 30 tumors in 23 patients with a maximum tumor diameter less than 3.0 cm (mean = 1.8 cm, range: 0.7–2.8 cm), and group 2 included 21 tumors in 21 patients with a maximum tumor diameter greater than or equal to 3.0 cm (mean = 4.0 cm, range: 3.0–7.5 cm). Although group 1 had no complications, group 2 had 13 (62%) complications, including two deaths. The most common complication was postoperative hemorrhage requiring blood transfusion (38%), with the two deaths secondary to aspiration pneumonia and myocardial infarction on postoperative days 16 and 37, respectively. With a mean follow-up of 9 months, there were no recurrences in group 1, whereas group 2 had one recurrence. The investigators concluded that although LCA was equally effective for larger renal masses (>3.0 cm) in terms of successful ablation, the laparoscopic approach to cryoablation should be performed only for lesions less than 3.0 cm because of the significantly higher complication rate.

One of the larger and more recent PCA studies by Atwell and colleagues reported a complication rate of 6%, similar to the other PCA studies mentioned previously. The seven complications included worsening preexisting hypertension and pulmonary edema, delayed urosepsis, pulmonary embolus, hematuria requiring ureteral stent placement, and three patients who had retroperitoneal hematomas requiring angiography and red blood cell transfusion. The researchers mentioned the value of preablation selective angioembolization of renal tumors in decreasing the rate of major postoperative hematomas. Although the small number of tumors that were embolized before surgery precluded any meaningful statistical analysis, this does give some food for thought in terms of whether or not this optimizes patient safety by decreasing the number of major postoperative retroperitoneal hematomas and what effect this would have on renal function and the overall cost of the procedure.

In a multi-institutional retrospective series of 271 ablative cases, Johnson and colleagues analyzed the complications of 139 cryoablation and 133 RFA procedures, of which 181 were performed percutaneously, with the remaining 90 procedures performed laparoscopically. A total of 20 complications (14.4%) occurred in the cryoablation series (4 after LCA and 16 after PCA). The cryoablative group reported 2 major complications: one significant postoperative hemorrhage requiring reoperation and blood transfusion after PCA and one open conversion attributable to the inability to access the tumor laparoscopically. Minor complications included 10 cases of flank pain and paresthesias at the operative site, two postoperative urinary tract infections, one small self-limited hematoma, one case of postoperative pneumonia, and one case of elevated serum creatinine. Most of these patients had a complete recovery. The researchers noted that more than half of the complications in this study occurred in the first third of the procedures performed, suggesting a learning curve associated with this new MINSS technique.

The theory that cryoinjury predisposes the kidney to further loss of renal function secondary to necrosis and scarring seems to be largely unfounded based on all clinical trials to date. Johnson and colleagues showed promising results after cryoablation with only 1 of 139 patients sustaining a significant elevation in serum creatinine. In a retrospective review with a mean follow-up of 20.6 months, Carvalhal and colleagues reported no significant differences in pre- and postoperative creatinine levels, blood pressure, or estimated creatinine clearance.

In a retrospective review, Finley and colleagues compared the peri- and postoperative outcomes of PCA and LCA. A total of 37 patients underwent treatment for 43 renal masses; 24 tumors were approached laparoscopically (mean tumor size = 3.0 cm, range: 1.1–5.4 cm), whereas the remaining 19 tumors were treated with CT-guided PCA (mean tumor size = 2.7 cm, range: 1.7–4.7 cm). The transfusion rates in the PCA and LCA groups were 11.1% and 27.8%, respectively. Operative time was significantly longer in the LCA group (mean = 147 minutes, range: 89–209 minutes) than in the PCA group (mean = 250 minutes, range: 151–360 minutes). The overall complication rate was lower in the PCA group (22%) when compared with the LCA group (40%). Hospital stay was also significantly shorter in the PCA group than in the LCA group (1.3 versus 3.1 days; P <.0001). Among the patients who had biopsy-proved RCC, during a median follow-up of 11.4 and 13.4 months, the PCA and LCA groups had 100% cancer-specific survival and treatment failure rates of 5.3% and 4.2%, respectively. Given these findings, the investigators concluded that PCA seems to be superior to the laparoscopic approach to cryoablation for small renal tumors.

Radiofrequency Ablation

RFA acts through the conversion of radiofrequency waves into heat, resulting in thermal damage to tissue. A high-frequency monopolar alternating current flows from the needle electrode to the target tissue, resulting in ionic agitation and heat-producing molecular friction as an inverse function of the tissues’ impedance (resistance to energy flow). Heat is not directly supplied by the probe itself. RFA leads to cell death through instantaneous and irreversible protein denaturation, tissue cross-linking, and desiccation. The thermal effects of tissue desiccation and coagulative necrosis occur almost immediately when temperatures reach higher than 60°C. Exposure to these high temperatures also has direct effects on cellular components, including cell membrane disintegration and cell nucleus unrest. In addition, RFA has a direct effect on tissue perfusion secondary to microvascular and arteriolar occlusion within the ablated zone, leading to ischemia of the treated tissue.

Originally, RFA was applied in a “dry” fashion by an electrode causing rapid increases in temperature, with tissue desiccation and charring around the tip of the probe. Once tumors reach 4 cm and larger, however, this technique increases tissue impedance and results in an inefficient radiofrequency lesion. To circumvent this problem and make ablation more time-efficient, “wet” RFA was developed, in which saline is irrigated through the tissue to increase conductivity and help to dissipate energy, thereby resulting in a larger radiofrequency lesion.

Because of the formation of microbubbles and the low intrinsic contrast between normal and ablated tissues, real-time imaging of the developing thermal lesion with intraoperative US, CT, or MRI has proved difficult. The lesion created by the radiofrequency probe, however, is a function of its deployed diameter and minimum time activated; as a result, proponents of RFA argue that real-time monitoring of the ablation zone is not necessary, because activating the probe for longer periods does not change the lesion size. RFA is generally monitored by measuring impedance and temperature changes, with feedback provided from thermosensors integrated at the end of radiofrequency tines around the radiofrequency probes. Imaging immediately after the procedure can be difficult to interpret in terms of successful ablation because of peripheral inflammation; however, one study did demonstrate that MRI allowed differentiation of treated versus untreated renal parenchyma to within 2 mm when compared with pathologic examination. Typically, the success of RFA is assessed by a contrast-enhanced CT scan or MRI 4 to 6 weeks after treatment by demonstrating a paucity of enhancement within the ablated tissue.

As with cryoablation, RFA can be used by means of an open, laparoscopic, or percutaneous approach, with the latter two techniques being the predominant ways in which radiofrequency waves are currently delivered. As with cryoablation, laparoscopic radiofrequency ablation (LRFA) has the advantage of tumor mobilization and placement of the probe under direct vision. This increased exposure and visualization of the kidney would theoretically allow for the avoidance of adjacent organ damage.

Early experience with LRFA was presented by Jacomides and colleagues on 13 patients with a total of 17 renal tumors. In 5 patients, the tumor was subsequently excised completely, whereas in 7 patients, it was left in situ. Mean tumor size was 1.96 cm (range: 0.9–3.6 cm). Pathologic analysis revealed RCC in 10 patients, angiomyolipoma in 2 patients, and oncocytoma in 1 patient. None of the 8 patients who had RCC and at least 6 weeks of follow-up had any evidence of persistent tumor enhancement on surveillance CT scan. There was one focal positive margin in a patient who underwent RFA and excision of RCC, but the patient has remained disease-free for 12 months after treatment.

The evolution of thinner diameter radiofrequency probes has rendered percutaneous radiofrequency ablation (PRFA) the most prevalent method currently for renal RFA, using CT or MRI guidance. Matsumoto and colleagues reported on a series of 91 patients who had 109 small renal tumors; 63 were treated with PRFA, and the other 46 were treated with LRFA. The combined mean tumor size for the two groups was 2.4 cm (range: 0.8–4.7 cm), with a combined initial success rate of 98% (107 of 109 tumors). There was one local recurrence (1.7%) detected during a mean follow-up of 19.4 months (range: 12–33 months), with incomplete ablations and the local recurrence successfully treated with reablation. Paralleling the results from the CA literature, the percutaneous approach had a shorter mean operative time (105.8 versus 147.5 minutes) and lower mean EBL (4.2 versus 25.1 mL) than LRFA. Although the PRFA group reported no major complications, the LRFA group had three major complications, consisting of a UPJ injury that ultimately required nephrectomy, a urine leak, and a lower pole infarct.

Gervais and colleagues reported on their clinical experience with CT- and MRI-guided PRFA for 100 renal masses in 85 patients over a period of more than 6 years. All the 68 peripheral exophytic lesions (52 of which were ≤3 cm) were ablated successfully after one treatment. This was in contrast to 78% (7 of 9 tumors) of central tumors and only 61% (11 of 18 tumors) of mixed-type tumors that had both central and exophytic components, which were completely ablated after one session. In addition, the researchers noted that small (<3 cm) peripheral lesions had a successful ablation rate of 100%, whereas larger tumors (>3 cm) and those that were centrally located had successful ablation rates of 92% and 25%, respectively. Results of the ad hoc analysis suggested that small size ( P <.0001) and noncentral location ( P <.005) proved to be independent predictors of complete necrosis after a single PRFA session.

Stern and colleagues recently compared the intermediate-term outcomes of patients who had clinical T1a renal tumors treated with partial nephrectomy (open and laparoscopic) or RFA (laparoscopic and percutaneous). A total of 37 partial nephrectomies (30 open, 7 laparoscopic) were compared with 40 RFAs (14 laparoscopic, 26 percutaneous), with a mean tumor size of 2.41 cm and 2.43 cm, respectively. The mean (range) follow-ups for the RFA and partial nephrectomy groups were 30 and 47 months, respectively. There was one incomplete ablation and two recurrences in the RFA group, whereas the partial nephrectomy group had two recurrences (one local and one in the contralateral kidney). There were no disease-specific deaths in either group. The overall actuarial disease-free probability for the partial nephrectomy and RFA groups was 95.8% and 93.4%, respectively.

Levinson and colleagues recently reported their long-term oncologic and overall outcomes of PRFA in 31 high-risk surgical patients who had solitary renal masses. Median tumor size was 2.0 cm (range: 1.0–4.0 cm), and patients had a mean follow-up of 60.5 months (range: 41–80 months). There was one primary treatment failure, which was successfully reablated, and three recurrences at 7, 13, and 39 months after PRFA. The recurrences were managed with repeat PRFA, PCA, and laparoscopic radical nephrectomy, respectively. The overall recurrence-free survival was 90.3%, with 100% metastasis-free and disease-specific survival. The difference between pretreatment and last known serum creatinine levels was 0.15 mg/dL ( P = .06). The researchers concluded that in patients who have a limited life expectancy or are high-risk surgical candidates, RFA provides reasonable long-term oncologic control.

Complications of Radiofrequency Ablation

In the multi-institutional review by Johnson and colleagues, of the 133 cases that underwent RFA, there were a total of 10 complications (7.6% complication rate) directly attributable to the procedure (3 major, 7 minor). The LRFA group had all 3 major complications, consisting of urine leak, UPJ obstruction, and ileus. The minor complications included four patients who had probe site pain or paresthesias, postoperative pneumonia, or self-limited liver-burn injury and one patient with an elevated serum creatinine level.

As with cryoablation, renal function after RFA is preserved in the most cases. In a study by Lucas and colleagues, a total of 242 consecutive patients were treated for solitary renal masses 4 cm or less in diameter using RFA (86 patients), partial nephrectomy (85 patients), and nephrectomy (71 patients). Before surgery, stage 3 chronic kidney disease was identified in a total of 65 patients, including 26.7%, 27.1%, and 26.8% who underwent RFA, partial nephrectomy, and radical nephrectomy, respectively. After intervention, the 3-year freedom from a glomerular filtration rate decrease of less than 60 mL per minute per 1.73 m ² for RFA, partial nephrectomy, and radical nephrectomy was 95.2%, 70.7%, and 39.9%, respectively ( P <.001). This article highlights the prevalence of impaired renal function in patients who have SRMs and the important consideration of long-term kidney function, particularly in those patients in whom renal function is already compromised.

Overall, the complications that were defined in the review by Johnson and colleagues are similar to those reported by other smaller single-institution series. By far, the most common minor complication is pain and paresthesias at the percutaneous probe insertion site. In terms of major complications, most are secondary to thermal injury to the renal collecting system. Traver and colleagues reported two cases of proximal ureteral strictures from a series of 73 tumors that were treated with PRFA. Single cases of ureteral stricture were also seen in the PRFA series of Matsumoto and colleagues, Gervais and colleagues, and Weiser and colleagues. This last point further emphasizes the need to apply any thermal ablation technology in the vicinity of the proximal ureter or renal pelvis judiciously for the reasons discussed previously. In most of the reported clinical series, the preponderance of major complications secondary to cryoablation has to do with postoperative hemorrhage. The ability of RFA to avoid tract bleeding and tumor seeding by coagulating the puncture channel during probe withdrawal results in the decreased number of major bleeding complications associated with this procedure.

There are a few clinical studies in which renal tumors have been removed or biopsied after RFA that have called the success of this technology into question. In a recent study by Weight and colleagues, they sought to correlate radiographic imaging with histopathologic examination after LCA and PRFA. There were a total of 109 lesions treated with PRFA, whereas 192 lesions were treated with LCA. Radiographic success at 6 months was 85% and 90% for PRFA and LCA, respectively. At 6 months, almost half of the lesions were biopsied and success in the PRFA cohort dropped to 64.8%, whereas success in the LCA group remained fairly high at 93.8%. Perhaps most alarming was that 6 (46.2%) of 13 patients in the PRFA group with a positive biopsy demonstrated no enhancement on posttreatment MRI or CT scans. In the patients treated with LCA, 100% of the positive biopsies revealed posttreatment enhancement on imaging just before biopsy. The investigators concluded that because of the poor correlation between postoperative imaging and biopsy to define success after RFA, all patients undergoing RFA should have a postoperative biopsy to rule out residual cancer cells left behind after ablation.

Standard histologic staining using hematoxylin and eosin (H&E) may be somewhat misleading, however, immediately after RFA. Rather, staining with NADH diaphorase has been postulated to be a more accurate assessment of cell viability, because the activity of this ubiquitous cellular enzyme has been shown to cease several hours after cell death. Essentially, the temperatures used during RFA are high enough to cause cell death and not necessarily acute architectural damage throughout the ablated lesion. Marcovich and colleagues demonstrated this point by staining resected porcine renal tumors that had first been treated with RFA with H&E and NADH diaphorase. What they found was that although RFA produced discernable histologic changes on H&E, these were variable and patchy and alternated with areas in which the architecture had been well preserved. When these corresponding areas were then stained with NADH diaphorase, there was a complete absence of staining, indicating that cell death had taken place.

Taking these data into account, Michaels and colleagues treated 20 tumors in 15 patients using a temperature-based electrode immediately before open partial nephrectomy. Although gross visualization indicated that all tumors were completely treated, 4 of 5 tumors stained with NADH diaphorase suggested residual activity. Matlaga and colleagues also evaluated tumor cell destruction after RFA using histologic analysis. Ten patients underwent impedance-based US-guided RFA immediately before partial or radical nephrectomy. Overall, 8 of 10 tumors demonstrated no NADH activity within the tumor or surrounding rim of renal parenchyma. Of the 2 tumors that stained positively for NADH diaphorase, the first did not achieve an adequate increase in temperature (41°C), whereas the other was 8.0 cm in maximal diameter. The first failure was attributed to heat sink phenomena, whereas the second was not expected to be successful based on the size of the tumor.

The failures of renal RFA in these studies have largely been attributed to flaws in surgical and staining technique and immature technology used at that time. Among the reasons cited to explain the ablative failures are inappropriate use of NADH diaphorase staining, real-time US to monitor the ablation zone, impedance-based electrodes prematurely ceasing tissue ablation, short ablation times, and inadequate wattage. Recently, Stern and colleagues reassuringly performed post-RFA biopsies on 20 lesions at a mean follow-up of 26.9 months (range: 13.1–58 months). All biopsies were performed on patients who clinically had NED (absence of lesion growth or contrast enhancement by CT) 1 year or more after RFA. Preablation biopsies confirmed that 17 of 20 tumors were RCC, whereas the remaining 3 tumors were oncocytomas. At the time of repeat biopsy, all 20 specimens showed unequivocal cell death with no evidence of cellular viability.

Microwave Ablation

Microwave ablation (MWA) involves the insertion of flexible antennae placed directly into target tissue and channels microwave energy to create a rapidly alternating electromagnetic field. In turn, this leads to the oscillation of ions, with a subsequent increase in kinetic energy that is converted to heat and ultimately results in coagulative necrosis. Most of the published literature on MT describes its use in achieving hemostasis at partial nephrectomy without the need to clamp hilar vessels. Until recently, no clinical experience had been reported regarding its use in renal tumor ablation as a curative approach. Liang and colleagues recently evaluated the feasibility, safety, and efficacy of US-guided percutaneous MWA for small (<4.0 cm) RCCs. A total of 12 patients who had pathologically proved RCC on preoperative biopsy underwent US-guided percutaneous MWA. Mean tumor size was 2.5 cm (range: 1.3–3.8 cm), with 10 tumors being exophytic and the other 2 being intraparenchymal. Immediate treatment efficacy was assessed using thermocouples to monitor temperature in real time during ablation and 24 hours after ablation with contrast-enhanced US. Short-term efficacy was then assessed by a contrast-enhanced CT scan or US at 1, 3, and 6 months and every 6 months thereafter. All tumors were successfully ablated in one session, and there were no reported complications. No residual tumor recurrence was observed at a median follow-up of 11 months (range: 4–20 months). The researchers discuss some potential advantages of MWA, which include a broad zone of active heating and the fact that its transmission through tissue is not limited by desiccation and charring. MT is also less affected by the perfusion-mediated heat sink effect. Although the study is promising in terms of ablation success rate, complication rate, and short-term efficacy, it does have some limitations that must be addressed in subsequent series. The inclusion criteria were quite strict in terms of tumor size (<4.0 cm) and location (tumors near the hilum or adjacent to important structures, such as bowel or ureter, were excluded), and the relatively small sample size makes it somewhat difficult to interpret the results.

High-Intensity Focused Ultrasound

In HIFU, an US wave is propagated through biologic tissues, whereby it is progressively absorbed. This energy is converted into heat, leading to protein denaturation, coagulative necrosis, and subsequent cell death. The transducer is used for delivery and monitoring of the ablative therapy. The most vital aspect of this technology (which allows it to be applied in an extracorporeal fashion) is that the energy generated by means of the ultrasonic waves drops sharply outside of the focal zone, such that nearby tissues remain largely unchanged. This seemingly narrow range of focus also proves to be HIFU’s main limitation in terms of lesion localization and targeting, especially when applied to renal tumors.

This last point was demonstrated by Marberger and colleagues, who looked at 18 kidney tumors treated with HIFU. All tumors were subsequently removed, and histologic analysis found incomplete ablation in every case. The disappointing results were largely attributable to the difficulty in tissue targeting secondary to respiratory movement, intervening structures of the abdominal wall and ribs, and acoustic characteristics of the tumors themselves. In another study, Hacker and colleagues applied HIFU to the healthy tissue of 24 kidneys monitored by US. For an in vivo study, 14 kidneys were removed immediately after ablation, with the other 10 kidneys removed after 1, 7, and 10 days of therapy. Side effects included grade 3 skin burns in two of the patients. In one case, there was a thermal lesion of the small intestine, which the investigators attributed to poor focusing. HIFU effects in the focal zone immediately after application were interstitial hemorrhage, fiber rupture, shrinking of collagen fibers, and coagulative necrosis. These effects occurred sporadically, and their number and size did not correlate with the number of HIFU pulses that were applied to the target tissue. After 7 and 10 days, there was a well-demarcated area of coagulative necrosis in vivo. Based on the cumulative clinical experiences of these and other researchers, it would seem that continued refinements in HIFU are essential to establish this ablative technology as a noninvasive treatment option for renal masses.

Radiosurgery

In contrast to the thermal ablation techniques discussed heretofore, radiation destroys dividing tumor cells by mitosis-linked apoptosis. At higher doses, radiation results in complete ablation of tissue, and thus provides the basis for solid mass ablation. The caveat to radiation’s use for ablation was the unacceptable collateral damage that occurred to normal tissue at the treatment margins. In contrast to traditional external beam radiation, however, radiosurgical therapy is delivered using a frameless image-guided device that utilizes a linear accelerator mounted on a robotic arm and divides the high-dose radiation into more than 1000 beams. Therefore, the individual dose of each beam is relatively benign to skin and adjacent tissue, whereas at the focal point of the beams, the dose is additive and the desired ablative dose is attained.

As described previously, one of the major limitations to extracorporeal solid organ ablation of renal tumors is the inherent back-and-forth movement of the kidney associated with the respiratory cycle. This limitation is circumvented by a highly advanced radiosurgical image guidance system that allows the radiation beam to follow and “correct” for a moving target in real time. A unique tracking and compensation system uses external “fiducial” markers in conjunction with diagnostic radiographic images to guide the robotic arm so that the beam remains aligned with its target. With the aid of this technology, renal tumor motion can be tracked in three-dimensional (3D) space, allowing for exceptionally precise treatment (within 0.3 mm).

Still in its relative infancy, radiosurgical ablation is devoid of any long-term studies; however, short-term analysis has been promising. In 2003, Ponsky and colleagues were the first to report their initial evaluation with stereotactic radiosurgery technology in eight swine. In each pig, bilateral kidneys were treated at predetermined sites (each approximately 2 cm in diameter) using a single dose of 24 to 40 Gy followed by organ harvest and histologic examination 4 to 8 weeks later. All targeted sites revealed complete tissue ablation surrounded by a small zone of partial fibrosis. Perhaps just as important, the remainder of the surrounding renal parenchyma was completely unharmed, and gross inspection at the time of kidney procurement failed to reveal any evidence of radiation injury to the body wall or surrounding organs. After this initial study, a clinical protocol was designed to address the safety of renal radiosurgery in humans. Ponsky and colleagues treated three patients with a mean tumor size of 2.03 cm using 16 Gy divided into four fractions over 2 consecutive days. All patients had a CT scan 8 weeks after radiosurgery and subsequently underwent a partial nephrectomy for their renal mass. Histopathologic examination of the excised specimens demonstrated a cavity without evidence of viable tumor in one patient, whereas the remaining two showed evidence of residual RCC. This trial confirmed the safety of renal radiosurgery, because there were no acute or chronic toxicities using this low dose with a follow-up time of longer than 1 year. With a clinical protocol in place for gradual dose escalation, results are forthcoming with regard to clinical efficacy.

Promising work with radiosurgical technology has also been reported by Hong and colleagues. Their initial evaluation included 14 patients with a mean tumor diameter of 4.1 cm treated with 21 Gy divided into three fractions. Patients were then imaged every 3 months with serial CT scans. Tumor volume decreased by a mean of 44%, with no signs of disease progression at 12 months of follow-up. Given that multiple radiosurgical units are currently in operation at radiation oncology centers worldwide, further experimental and clinical data should continue to become available for this promising ablative technology.