- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Nephrectomy (radical or partial) remains the most proven effective treatment for patients with RCC, with excellent long-term freedom from recurrence and cancer-related death . Percutaneous renal cryoablation has been shown to be an effective treatment option for patients with small renal tumours who have major medical comorbidities, or for patients who have previously undergone surgery on the ipsilateral kidney [2–5]. In fact, the AUA consensus guidelines now include percutaneous cryoablation as a potential treatment option for high surgical risk patients with T1a renal tumours . Long-term efficacy of renal cryoablation has yet to be established, but as patient interest and physician acceptance continue to grow, indications for renal cryoablation are beginning to expand. It is therefore important for urologists and interventional radiologists to develop an understanding of the efficacy and complications of percutaneous cryotherapy vs. surgery in the treatment of RCC.
There are well established efficacy and complication rates for radical and partial nephrectomy [1,7–9]. These surgical results are primarily based on evaluation of patients with solitary, sporadic RCCs (Nx/N0/M0) and are often subdivided by T-stage. This includes patients with T1a tumours (≤4 cm), T1b tumours (>4 cm and ≤7 cm), and T2 tumours (>7 cm). To date, percutaneous cryoablation studies have included a diverse group of patients with benign and malignant renal masses, solitary and multifocal tumours, history of RCC in the ipsilateral and/or contralateral kidney, and even patients with known metastatic disease at the time of renal cryoablation. In an initial effort to compare and contrast efficacy and complications of renal cryoablation vs. surgery, we reviewed our experience with percutaneous cryoablation in the treatment of patients without history of RCC and a unifocal, biopsy-proven RCC (i.e. sporadic RCC). The treatment effectiveness and procedural complication rates with renal cryoablation were also subdivided according to the size (T-stage) of the RCCs.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Approval for the present study was obtained from our institutional review board, and the study was compliant with the Health Insurance Portability and Accountability Act. A retrospective review of our percutaneous renal cryoablation database found that a total of 343 renal tumours in 302 patients had been treated with percutaneous cryoablation at our institution between November 2003 and November 2010. Of these, 94/302 (31%) patients with 130/343 (38%) tumours had a history of RCC and were excluded from analysis. An additional 13 patients without a history of RCC presented with multiple tumours and were excluded. Of the remaining 195 patients with solitary, sporadic tumours, 181 had biopsy of their tumour and 116 of the 181 (64%) biopsies showed RCC. These 116 patients with solitary, sporadic RCC represent the primary study population.
All patients underwent formal consultation in the Department of Urology before treatment. Primary clinical indications for percutaneous renal cryoablation included contraindication to surgery owing to significant medical comorbidities in 82 patients (71%) and patient preference in 34 patients (29%). Tumour characteristics are shown in Table 1. Central renal masses were defined as those extending to, or within, the renal sinus fat. Exophytic renal masses were defined as those with >50% of the tumour circumference outside the renal capsule, and intraparenchymal renal masses were those with <50% of the tumour circumference outside the renal capsule. Nephrometry scores were calculated using a system that quantifies renal tumour characteristics based on tumour radius, exo/endophyticity, nearness to the collecting system, and location with reference to the polar lines and renal midline . The RENAL nephrometry score for a tumour is the combination of points (ranging from 1 to 3) in each of these four categories.
Table 1. Tumour characteristics (N= 116)
|Median (range) size, cm||3.4 (1.3–8.9)|
|Stage, n (%)|| |
| T1a||83 (72)|
| T1b||27 (23)|
| T2||6 (5)|
|Location, n|| |
| Right kidney||61|
| Left kidney||55|
| Upper pole||26|
| Mid kidney||54|
| Lower pole||36|
|Median (range) RENAL nephrometry score||7 (4–10)|
|Subtype and grade of RCC, n|| |
| Papillary RCC||19|
| Clear-cell RCC||81|
| Chromophobe RCC||1|
| RCC (not otherwise specified)||15|
| Grade 1||25|
| Grade 2||54|
| Grade 3||13|
| Grade 4||3|
| No pathological grade reported||21|
A single cryoablation procedure was performed for treatment of each patient by one of six interventional radiologists with between 3 and 10 years of experience performing percutaneous renal ablations. The cryoablations were performed in a hospital CT suite (GE Hispeed CT/i system [GE Healthcare, Waukesha, WI, USA] or Somatom Sensation Open 40-MDCT system [Siemens Medical Solutions, Malvern, PA, USA]) with the patients under general anaesthesia. Ultrasonography (Acuson Sequoia, Siemens Medical Solutions) and/or intermittent CT guidance was used to place the cryoprobes into the renal mass. Probes were placed with 1–2 cm spacing within the tumour in a configuration to achieve a confluent ice ball with at least a 0.5-cm margin around the entire tumour. One or two core biopsies were then obtained from the tumour using an 18-gauge x 2-cm biopsy device (Bard Monopty, CR Bard, Riverside, NJ, USA).
In 26 of the 116 cases (22%), the bowel was within 2 cm of the target tumour, and hydrodisplacement was used to move the colon or small bowel a safe distance away from the expected ablation zone. This was performed by infusing sterile saline through a 19-gauge needle placed between the bowel and the renal mass using ultrasonography and/or CT guidance . In eight patients (7%) pre-ablation selective arterial embolization of the RCC was performed with 150–250 micron polyvinyl alcohol particles (Ivalon, Uniport Industries, High Port, NC, USA) . The goal of this procedure was to eliminate as much arterial blood flow to the tumour as possible to reduce the amount of bleeding during the subsequent cryoablation procedure. At the same time, the embolization was performed as selectively as possible to preserve as much normal kidney as possible. Although pre-ablation tumour embolization is not uniformly implemented across our practice, it is now typically performed the day before the ablation procedure in patients with tumours >5 cm in diameter.
The Endocare cryoablation system and Perc-24 cryoablation probes (Endocare, Inc, Irvine, CA, USA) were used in all cases. A median (range) of 3 (1–8) cryoprobes was used to ablate each RCC. The Perc-24 cryoablation probes are 2.4 mm in diameter (13-gauge) and will generate an ice ball ∼3 cm in diameter × 5 cm along the shaft. A freeze-passive thaw–freeze cycle was performed for cryoablation of each tumour. The length of each freeze time was based on the size of the ice ball required to completely encompass the mass. The median (range) initial freeze time was 10 (5–14) min, the median (range) passive thaw time was 5 (5–15) min, and the median (range) final freeze time was 8 (4–17) min.
Non-contrast CT monitoring of the size and location of the ablation ice ball was performed at 2–4-min intervals during each freeze cycle. Images were obtained at 2.5–5.0 mm slice thickness using standard CT technique (120 kV peak and ∼240 mA) and viewed at body window and level settings (W400, L40). The CT images were often reconstructed in different planes during the ablation to ensure that the ice ball completely encompassed the tumour and did not extend to involve adjacent critical structures, including the bowel and ureter. After the second freeze, the ice ball was actively thawed and the cryoprobes were removed.
Non-contrast CT was obtained immediately after removal of the cryoprobes to assess for potential complications. Very large perinephric haematomas were seen in three patients, who were immediately transferred for angiography and embolization. The other 113 patients underwent contrast-enhanced CT or MRI of the abdomen within 24 h of the procedure to ensure technical success of the ablation. In 94 of 113 patients (83%), triphasic contrast-enhanced CT was performed after the ablation procedure while the patient was still under anaesthesia. In the other 19 patients (17%), MRI was performed later that day or the next morning without and with i.v. gadolinium enhancement on a 1.5T system (Signa Excite, GE Healthcare).
As adapted from the International Working Group on Image-Guided Tumor Ablation , technical success was defined as extension of the ice ball at least 0.5 cm beyond the margins of the tumour on monitoring non-contrast CT performed during the procedure and extension of the ablation zone beyond the margins of the tumour on contrast-enhanced CT or MRI performed within 3 months of the ablation. Local tumour recurrence was defined as a hyperenhancing or enlarging soft tissue nodule in, or around, the ablation zone on contrast-enhanced CT or MRI performed ≥3 months after the ablation. Follow-up imaging was recommended at 3, 6 and 12 months after ablation, and yearly thereafter; however, imaging follow-up was often individualized based on patient-specific factors, including size, subtype, and grade of the treated RCC. A total of 88 patients (76%) had contrast-enhanced CT or MRI imaging follow-up available for review ≥3 months after their ablation. Evidence for tumour recurrence was based on imaging only, and routine post-ablation biopsies were not performed.
Complications were assessed using the revised Clavien–Dindo classification system , which grades surgical complications based on the degree of necessity for unanticipated intervention during the normal postoperative course. Any grade ≥3 complication was considered a major complication. After recovery from anaesthesia, all patients were admitted to the urology hospital service.
Standard descriptive statistics were used to summarize the sample data. Overall survival was modelled from the time of ablation to death or last clinical follow-up. Recurrence-free survival was modelled from the time of ablation to imaging-confirmed local or metastatic tumour recurrence, or to the date of last imaging without evidence for tumour recurrence. The patient with a technical failure was considered to have an immediate tumour recurrence. Patients without imaging follow-up ≥3 months after their cryoablation were censored at 0. The product limit (Kaplan–Meier) estimator was used to estimate each survival function after ablation. The Greenwood formula was used to estimate the point-wise CI at 1, 2 and 3 years after ablation. Analyses were conducted using The SAS System (version 9.22). The Cochran-Armitage trend test was used to test if tumour size was associated with the occurrence of a major complication.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Of the 116 patients treated for a biopsy-proven RCC during the 7-year study period, 80 of the patients (69%) were male and 36 (31%) were female. The median (range) patient age at the time of ablation was 72 (45–92) years. In all, 83 patients (72%) were treated for a stage T1a RCC, 27 patients (23%) for a stage T1b RCC, and six patients (5%) for a stage T2 RCC.
Technical success was achieved in the treatment of 115 of 116 (99%) renal tumours. The single technical failure occurred with treatment of a 4.3-cm central RCC. The residual tumour was first identified on 1-month post-ablation CT along the deepest margin of the ablation zone, but in retrospect, the residual tumour was evident on the immediate post-ablation CT as well. Although this residual tumour would have been amenable to repeat cryoablation, the patient opted for laparoscopic radical nephrectomy and remains alive without evidence for recurrent tumour >3 years after surgery.
Of the 88 patients with follow-up imaging beyond 3 months (median 21 months; range 3–73), a single patient (1%) treated for a 2.7-cm central RCC had evidence of recurrent tumour on CT imaging performed 11 months after the cryoablation. This patient was retreated with cryoablation for a 1.0-cm biopsy-proven RCC recurrence along the deepest margin of the ablation zone. CT imaging at 12 months after re-treatment showed no evidence of recurrent tumour.
Figure 1 shows the Kaplan–Meier estimates of overall survival and recurrence-free survival. A total of 16 patients (14%) died during the follow-up period, but none of the deaths were related to RCC. Six deaths were observed in the first year with the last death event at ∼ 8 months after ablation. The estimated 1-year overall survival for the cohort was 94.9% (95% CI: 88.1–97.8%). At 2 years after the ablation, four additional deaths had occurred, and the 2-year overall survival was estimated at 89.4% (95% CI: 80.4–94.4%). At year three, the risk set size was 32, and the estimated overall survival function was 81.7% (95% CI: 69.9–89.3%). The final three deaths occurred after year three, but owing to the limited risk set size, the interpretation of these data is limited. Cancer-specific survival was 100%. Two patients developed local residual/recurrent tumour (one immediately and one at 11 months), but no patients developed metastatic disease. The estimated 1-year recurrence-free survival was 97.3% (95% CI: 88.9–99.4%). Recurrence-free survival was 98% at a median of 21 months in the 88 patients with follow-up imaging beyond 3 months. Metastasis-free survival was 100%.
Although cryoablation was successful in 67 of 69 (97%) patients treated for a central RCC, both cases of residual/recurrent tumour occurred along the deepest margin of the ablation zone in treatment of a central RCC. Tumour size was not a statistically significant factor in predicting residual/recurrent tumour in the study group, since the technical failure occurred with treatment of a 4.3-cm tumour and the local tumour recurrence occurred after treatment of a 2.7-cm tumour.
Major complications occurred in nine patients (8%). Seven of these nine major complications were attributable to severe immediate post-procedural bleeding. All seven of the patients required immediate angiography and embolization. The other two major complications included a post-procedural pulmonary embolism and a post-procedural myocardial infarction. There were no bowel or ureteric injuries, and no procedural-related deaths. The frequency of major complications was associated with tumour size. Major complications were observed in 4% (3/83), 15% (4/27), and 2/6 of the patients with T1a, T1b, and T2 tumours, respectively (P= 0.003 for trend).
Ice ball fractures were identified on non-contrast CT during treatment of six (5%) patients. Three of these six patients went on to have major bleeding complications requiring angiography and embolization for treatment. These ice ball cracks generally occurred during treatment of larger RCCs. The median (range) tumour size in the patients with fractured ice balls was 4.6 (1.7–8.8) cm. The 1.7-cm tumour was the only tumour treated <4 cm in size where an ice ball fracture was identified.
All patients were observed in the hospital overnight after the ablation treatment. The median (range) length of hospital stay was 1 (1–10) day. Of the 116 patients, 96 (83%) were discharged from the hospital the next day. All nine patients with major complications required an extended hospital stay. This includes the patient with a 10-day hospital admission related to post-cryoablation bleeding and periprocedural myocardial infarction. The other 11 patients who required an extended hospital stay of 2 or 3 days included four patients with prolonged post-procedural pain, four patients with urinary retention, one patient with hyperkalemia, one patient with hypertension not controlled by his usual dose of antihypertensive medications, and one patient for close observation while off anticoagulation owing to his history of pulmonary embolism.
Outside of the 116 patients in our study group with solitary, sporadic, biopsy-proven RCCs, we have treated an additional 79 patients for a solitary sporadic renal tumour without a biopsy diagnosis of RCC. Of these 79 patients, 14 (18%) did not undergo biopsy (for technical reasons), 27 (34%) had non-diagnostic biopsies, 22 (28%) had oncocytomas, 14 (18%) had oncocytic neoplasms (where the pathologist could not make a clear determination of oncocytoma vs. RCC), and two (3%) had benign angiomyolipomas. All 79 of these patients had technically successful cryoablations, and none of them has had evidence for recurrent tumour on imaging follow-up.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
A recent consensus document detailing guidelines for management of small (stage T1) renal masses was sponsored by the AUA and published in The Journal of Urology. This document included a meta-analysis of 114 articles published on renal surgery, renal ablation, and active surveillance. Local disease-free survival was 90.5% in 10 renal cryoablation studies, which included 644 patients, at a mean of 19.5 months' follow-up. In comparison, local disease-free survival for renal surgery (open, laparoscopic, partial and radical) ranged from 98.0 to 99.2% in 66 studies at a mean of 22.9–59.3 months' follow-up.
As one considers such meta-analyses, it is important to recognize the heterogenous population of patients treated with thermal ablation, largely related to current limited indications, as reflected in the AUA guidelines. Specifically, and as found in our overall patient population, many patients treated with thermal ablation have a history of RCC; nearly a third in our experience. Given the increased surgical risk associated with repeat operation in these patients [15,16], such patients are commonly referred for ablation. This selection bias toward patients with a history of recurrent renal tumours may affect the oncological outcomes and survival in the evaluation of ablation efficacy. Accordingly, this published effort was designed to focus on a homogeneous patient and tumour population to optimally reflect the role of cryoablation in the primary treatment of sporadic RCC. The local disease-free survival rate in study was 98% at a median of 21 months' follow-up, which is better than that reported in other renal cryoablation studies, and compares favourably with the surgical results.
Although the present renal cryoablation study is much smaller in size and with considerably shorter follow-up, preliminary results also compare favourably with a recent study from our institution reporting outcomes after surgical resection of T1 renal tumours . That study showed 94–100% cancer-specific survival at 5 years for 637 patients treated with partial nephrectomy for a solitary, sporadic RCC ≤7 cm in size, compared with100% cancer-specific survival in the present study during the median 21-month follow-up. In addition, no one in the present study developed metastatic disease during the relatively short period of follow-up, which also compares favourably with the 5.6% RCC metastasis rate after partial nephrectomy, as reported in a recent meta-analysis .
Both cases of residual/recurrent tumour in the study group occurred along the deepest margin of the ablation zone in treatment of central RCCs. Although cryoablation was a successful treatment in 97% of patients with a central RCC, these two cases again illustrate the previously reported importance of paying special attention to extending the cryoablation zone beyond the deepest tumour margin when treating a central renal mass .
The expanded version of the consensus guidelines for treatment of stage 1 renal masses by the AUA cited a 9.9% major complication rate for cryoablation of T1 renal tumours, determined from the studies included in the meta-analysis . The 7.8% major complication rate in the present study was slightly lower than that reported figure and was also slightly lower than the reported 8.5% major complication rate for open partial nephrectomy and the 13.5% major complication rate for laparoscopic partial nephrectomy, also determined in the meta-analysis.
The present study shows increased major complications with cryoablation of larger (higher T-stage) RCCs. The great majority of these major complications were attributable to post-procedural bleeding. One of the causes of bleeding complications was ice ball fractures. While ice ball fractures have been relatively infrequently seen during percutaneous renal cryoablation procedures to date , they are likely to become more common as we continue to ablate larger renal masses. Cryoablation of larger renal masses requires use of multiple probes, which has been shown to be one of the main risk factors for development of ice ball fractures .
A small retrospective study from our institution suggests that selective arterial tumour embolization before cryoablation of tumours ≥5 cm decreases post-cryoablation bleeding . However, additional studies are certainly necessary to determine the efficacy of pre-ablation arterial tumour embolization for renal cryoablation. One of the patients in the present study required post-procedural embolization for a major haemorrhage after cryoablation of an 8.3-cm RCC despite prophylactic tumour embolization the day before the cryoablation procedure.
A major limitation of the present study is the 24% of patients without imaging follow-up for our review >3 months after ablation. Our tertiary care centre serves patients from a wide geographical area, and many of these patients choose to have their imaging and clinical follow-up performed closer to home. We attempt to obtain outside follow-up imaging on all ablation patients, but are sometimes unsuccessful. In addition, comparison of these retrospective single-institution cryoablation results with those of previous surgical studies is clearly limited because of differences in the patient populations and differences in the length and method of patient follow-up.
In conclusion, percutaneous renal cryoablation of RCC can be performed with high technical success in patients with tumours up to, and beyond 7 cm in maximum diameter. The tumour recurrence rate after percutaneous renal cryoablation was low, and recurrence was not related to tumour size in this group of patients. Statistically significant higher complication rates, however, were seen with treatment of larger (higher T-stage) RCCs. The 33% major complication rate in the present study for cryoablation of RCCs >7 cm would suggest that cryoablation of these larger T2 renal masses should be reserved for patients without other treatment options.