Percutaneous Microwave Ablation versus Cryoablation for Small Renal Masses (≤4 cm): 12-Year Experience at a Single Center

Abstract

Purpose:

To determine whether microwave ablation (MWA) has equivalent outcomes to those of cryoablation (CA) in terms of technical success, adverse events, local tumor recurrence, and survival in adult patients with solid enhancing renal masses ≤ 4 cm.

Materials and Methods:

A retrospective review was performed of 279 small renal masses (≤4 cm) in 257 patients (median age, 71 years; range, 40–92 years) treated with either CA (n = 191) or MWA (n = 88) between January 2008 and December 2020 at a single high-volume institution. Evaluations of adverse events, treatment effectiveness, and therapeutic outcomes were conducted for both MWA and CA. Disease-free, metastatic-free, and cancer-specific survival rates were tabulated. The estimated glomerular filtration rate was employed to examine treatment-related alterations in renal function.

Results:

No difference in patient age (P = .99) or sex (P = .06) was observed between the MWA and CA groups. Cryoablated lesions were larger (P < .01) and of greater complexity (P = .03). The technical success rate for MWA was 100%, whereas 1 of 191 cryoablated lesions required retreatment for residual tumor. There was no impact on renal function after CA (P = .76) or MWA (P = .49). Secondary analysis using propensity score matching demonstrated no significant differences in local recurrence rates (P = .39), adverse event rates (P = .20), cancer-free survival (P = .76), or overall survival (P = .19) when comparing matched cohorts of patients who underwent MWA and CA.

Conclusions:

High technical success and local disease control were achieved for both MWA and CA. Cancer-specific survival was equivalent. Higher adverse event rates after CA may reflect the tendency to treat larger, more complex lesions with CA.

Percutaneous ablation of renal tumors has been shown to be safe, provide excellent local disease control, and preserve renal function. Multiple retrospective reviews have demonstrated favorable outcomes for percutaneous thermal ablation of cT1a renal tumors when compared with those for partial nephrectomy and radical nephrectomy, without any significant difference in local recurrence and metastasis-free survival (MFS) (1–8). These observations have led the American Urological Association to update guidelines, recommending percutaneous thermal ablation as a viable alternative to partial nephrectomy and active surveillance in the treatment of cT1a solid renal masses <3 cm (9).

The largest bodies of evidence exist for the use of radiofrequency (RF) ablation and cryoablation (CA), the 2 ablation modalities with the longest clinical use. In recent years, microwave ablation (MWA) has emerged as a powerful heat-based ablation modality and has largely replaced RF ablation in the liver in the United States. MWA has several advantages over RF ablation: (a) it causes rapid and sustained heating of a volume of tissue, (b) the electromagnetic energy can penetrate all tissue types without impedance, and (c) it is less susceptible to a heat sink effect (10–12). MWA has been shown to be safe in the kidney, and oncologic outcomes have been promising (12–15). The purpose of this study was to determine whether MWA has equivalent outcomes to CA in terms of technical success, adverse events, local tumor recurrence, cancer-specific survival, MFS, and overall survival (OS) in adult patients with renal tumors ≤4 cm.

MATERIALS AND METHODS

Study Population

After institutional review board (IRB) approval of Thomas Jefferson University, a retrospective comparative analysis of 279 consecutive percutaneous ablative treatments in 257 patients treated for a contrast-enhancing solid renal mass ≤4 cm between February 2008 and December 2020 at a single center was performed. All patients with lesions deemed clinically suspicious by imaging for cT1a renal cell carcinoma (RCC) were included in the study. All patients were evaluated by a urologist prior to referral to interventional radiology (IR). Poor surgical candidates (high anesthesia risk, significant comorbidities, advanced chronic kidney disease, single kidney, or personal or family history of RCC) or those who declined surgery were referred to IR for possible thermal ablation. Cross-sectional computed tomography (CT) or magnetic resonance (MR) imaging of the abdomen was reviewed and used for procedure planning.

Procedural Details and Follow-up

Patients underwent either CA (n = 191) or MWA (n = 88) at the discretion of the interventional radiologist. Biopsy of the mass was performed concurrent with the ablation procedure using an 18-gauge core biopsy needle. In the first few years of the program, procedures were performed under moderate sedation (intravenous midazolam and fentanyl) administered and monitored by a dedicated IR nurse. From 2013 onward, all procedures were performed under general anesthesia. The first small renal mass MWA was performed in 2013. On the morning of ablation, complete blood count, basic metabolic panel, prothrombin time/international normalized ratio, and a blood type and screen were acquired. Anticoagulation and antiplatelet medications were managed in accordance with Society of Interventional Radiology (SIR) standards of practice guidelines on coagulation status and hemostasis risk (16). INR (1.5) and platelet (>50,000) thresholds were adopted from SIR standards of practice guidelines on coagulation status and hemostasis risk (17). All ablations were performed using CT guidance. Hydrodissection using contrast-doped saline (2%) was performed in select cases to prevent nontarget injury to adjacent structures. Pyeloperfusion (retrograde or antegrade) was performed if the lesion was located close to the ureteropelvic junction. In the case of a retrograde approach, an open-ended ureteral catheter was placed by urology; if an antegrade approach was used, the interventional radiologist pyeloperfused through a 22-gauge needle placed in the renal collecting system under CT or ultrasound (US) guidance immediately prior to ablation. Warm or cold fluid was infused during the ablation depending on whether CA or MWA was being performed. In the early years of the program, all patients were admitted for overnight observation and discharged the following day if no adverse events occurred. In more recent years, patients were sent home the same day barring any adverse events.

Ablation Devices

The goal of ablation was to encompass the mass and a 5-mm margin. Smaller and exophytic lesions were more likely to undergo MWA, and larger, perihilar, and/or endophytic lesions were more likely to undergo CA. MWA was performed using 17-gauge probes (Neuwave PR Probe; NeuWave Medical, Madison, Wisconsin). Ablation was performed at 65 W with duration based on lesion size and number of probes used. The tract was cauterized on a case-by-case basis at the discretion of the interventional radiologist. CA was performed using 13-gauge cryoprobes (PCS-24; Cryocare System; Varian, Palo Alto, California). The number of cryoprobes placed depended on the size and location of the lesion. Two freeze-thaw treatment cycles (10-minute freeze, 8-minute passive thaw, and 10-minute freeze followed by a final 5-minute active thaw) were applied. After completion of either ablation modality, a final CT with contrast was performed to evaluate the effectiveness of the procedure.

Patients were seen in the IR clinic (in person or via teleconsult) 2 weeks after ablation. CT or MR imaging of the abdomen with and without contrast was performed at 4, 8, 12, 18, and 24 months after ablation. Imaging was performed annually thereafter for up to 10 years. Annual staging with thorax CT or chest radiography was performed.

Qualitative and Quantitative Characteristics

Data on patient demographics (date of birth, age, and sex), body mass index, and tumor histology were collected from the electronic medical record. The Charlson Comorbidity Index was calculated from the patient’s medical history (18). Tumors were categorized according to the radius, exophytic/endophytic, nearness to collecting system or sinus, anterior/posterior, and location relative to polar lines (RENAL) nephrometry score (determined from preoperative MR imaging or CTof the abdomen) (19). Technical success of the procedure was defined as an ablation zone completely overlapping or encompassing the target tumor plus an ablative margin as determined by cross-sectional imaging at the end of the procedure (20). Primary effectiveness was defined as the percentage of tumors successfully eradicated after the initial procedure, and secondary effectiveness was defined as the percentage of all tumors that were successfully treated, including tumors that underwent repeat ablation. Local recurrence and survival analyses were performed in those with biopsy-proven RCC. Local tumor recurrence was defined as nodular enhancement in or adjacent to the ablation cavity after a previously documented successful treatment. Disease-free survival was defined as time from treatment to disease recurrence or death. Cancer-specific survival was defined as the percentage of patients who did not die from progression of the ablated RCC. OS was the percentage of patients who died of any cause, including progression of the ablated RCC. MFS was the percentage of patients without metastatic lesions from the ablated RCC. The procedure duration was extracted from the electronic medical record and was defined as the time elapsed from the “procedure start time” to “procedure end.” Adverse events were classified according to the SIR adverse event classification system (21).

Statistical Analysis

Statistical analyses were performed using Prism 9 (Graph-Pad Software, San Diego, California). Quantitative variables were collected and analyzed via descriptive statistics, giving mean, SD, median, and interquartile range (IQR). Categorical variables were tallied and reported as frequencies and percentages. Differences between the MWA and CA groups were compared using unpaired t-tests. As a secondary analysis, propensity score matching across groups was performed in Stata version 15.1 (StataCorp, College Station, Texas). Patients in each cohort (CA vs MWA) were matched based on Charlson Comorbidity Index, tumor size, number of probes, nephrometry score, and pyeloperfusion because these factors were found to be statistically significantly different across treatment groups. Kaplan-Meier survival curves were used to assess disease-free survival, MFS, and OS in those with biopsy-proven RCC. The Mantel–Cox test was used to compare survival distributions between groups. P values <.05 were considered significant.

RESULTS

A total of 279 ablations in 257 patients were analyzed. Of these, 191 lesions were treated with CA, and 88 were treated with MWA. Sixty-seven percent of patients were men. The median age of the study group was 71 years (range, 40–92 years; IQR, 11 years). The mean Charlson Comorbidity Index score was significantly higher for CA than for MWA (P < .01). Approximately 50% of patients had Stage 3–5 chronic kidney disease (Table 1). Three patients underwent ablation of 3 lesions, 15 underwent ablation of 2 lesions, and the remaining 238 underwent ablation of a single lesion; all ablations were performed during separate sessions. A local recurrence after partial nephrectomy or laparoscopic ablation was the indication for percutaneous ablation in 15 patients.

Table 1.

Patient Demographics and Tumor Characteristics (n = 279)

	Cryoablation	Microwave ablation	P value
No. of ablations	191	88
Sex, n (%)
Male	123 (64.4)	66 (75.0)	.06
Female	68 (35.6)	22 (25.0)
Body mass index, n (%)
Normal weight (18.5–24.9)	43 (22.5)	21 (23.9)
Overweight (25.0–29.9)	63 (33.0)	34 (38.6)	.51
Obese (>30.0)	80 (41.9)	33 (37.5)	.37
No data	5 (2.6)	0 (0.0)	.49
Chronic kidney disease stage, n (%)
Stage 3	87 (45.5)	33 (37.5)	.20
Stage 4	3 (1.6)	4 (4.5)	.22
Stage 5	4 (2.1)	3 (3.4)	.55
No chronic kidney disease, n (%)	97 (50.8)	48 (54.6)	.56
Ablations in patients on dialysis, n (%)	2 (1.0)	2 (2.2)	.49
Native kidney ablations in transplant recipients, n (%)	1 (0.5)	2 (2.2)	.30
Ablations in patients with single kidney, n (%)	17 (8.9)	7 (7.9)	.79
Age (y), median (IQR)	72 (65–75)	71 (64–77)	.99
Charlson Comorbidity Index score, median (IQR)	5 (3–7)	4 (3–6)	<.01
Tumor characteristics
Nephrometry score, n (%)
Low complexity (score, 4–6)	124 (64.9)	72 (81.8)	<.01
Medium complexity (score, 7–9)	53 (27.8)	15 (17.1)	.04
High complexity (score, 10+)	14 (7.3)	1 (1.1)	<.01
Tumor size (cm), mean ± SD	2.61 ± 0.75	2.22 ± 0.83	<.01
Anterior location, n (%)	32 (16.8)	24 (27.3)	.06
100% endophytic location, n (%)	80 (41.9)	31 (35.2)	.29

IQR = interquartile range.

Of the 191 patients who underwent CA, 109 (57.1%) received general anesthesia compared with 87 (98.9%) of 88 patients who underwent MWA (P < .01). Cryoablated lesions were significantly larger (mean size [CA], 2.61 cm; mean size [MWA], 2.22 cm; P < .01) and had a higher nephrometry score (mean RENAL score [CA], 6.07; mean RENAL score [MWA], 5.63; P = .03) (Table 1 and Figs 1, 2). The mean number of probes used per ablation was 2.61 for CA and 1.6 for MWA (P < .01). Pyeloperfusion was performed in 2 (2.3%) of 88 MWAs and 15 (7.9%) of 191 CAs (P = .03). Hydrodissection was performed in 30 (34.1%) of 88 MWAs and 75 (39.3%) of 191 CAs (P = .37). The mean MWA time was 5.8 minutes. The mean procedure time was 128 minutes for CA compared with 79 minutes for MWA (P < .001). The median length of stay was 1 day for CA (range, 0–13 days; IQR, 0 days) and 0 days for MWA (range, 0–10 days; IQR, 0 days).

Figure 1.

(a) An unenhanced abdominal computed tomography (CT) axial image demonstrated a 2.5-cm endophytic mass (arrow) in the interpolar region of the left kidney. The mass was biopsy-proven clear cell renal cell carcinoma Grade I. (b) Contrast-enhanced CT of the abdomen immediately prior to ablation demonstrated an enhancing mass. A retrograde catheter was placed for pyeloperfusion (arrow). (c) Coronal reconstruction of intraprocedural CT during cryoablation showed the 3 probes placed via lateral approach. (d) Abdominal magnetic resonance (MR) contrast-enhanced T1-weighted images in arterial phase and (e) portal venous phase performed 1 year after ablation demonstrated hypointense ablation cavity (circle) with no evidence of tumor recurrence in the interpolar region of the left kidney.

Figure 2.

(a) Abdominal magnetic resonance (MR) contrast-enhanced T1-weighted image in portal venous phase demonstrated a 2-cm exophytic solid mass (arrow) arising from the anterior cortex upper pole of the right kidney. (b) An abdominal computed tomography (CT) contrast-enhanced axial image with the patient placed left side down facilitated probe placement via intercostal approach posterior to the liver. (c) An unenhanced abdominal CT axial image demonstrated a single microwave probe placed via intercostal approach through the base of the lesion. A 5-minute ablation was performed. (d) Abdominal MR contrast-enhanced T1-weighted image 8 months after ablation showed a hypointense ablation cavity (circle) encompassing the lesion and a wedge of the renal cortex—indicative of successful treatment with no evidence of local recurrence.

The technical success rates for MWA and CA were 100% (88/88 lesions) and 100% (191/191 lesions), respectively. Primary technical effectiveness was 100% for MWA and 99.48% for CA, with 1 of 191 cryoablated lesions requiring retreatment for residual tumor. Secondary technical effectiveness was 100% for CA. Overall, 29 adverse events were documented—4 after MWA and 25 after CA, with CA having a significantly higher rate (P = .010) (Table 2). More than two thirds (20/29) were minor, and of these, 14 did not require treatment and were managed according to standard of care. There were 9 major adverse events, including 6 bleeds, an injury to the ureteropelvic junction, pulmonary embolus, and needle tract tumor seeding. The latter was identified 2 years after ablation. The patient declined all forms of treatment and subsequently developed liver metastases. Of the 6 patients with bleeds, 2 were managed with blood transfusions, 2 required transarterial embolization, 1 developed hematuria after resuming anticoagulation and antiplatelet therapy and required placement of an internal ureteral stent, and 1 (the sixth patient) developed a subcapsular hematoma resulting in page kidney, acute-on-chronic renal failure culminating in long-term hemodialysis.

Table 2.

Adverse Events

Adverse event category	Cryoablation n = 191	Microwave ablation n = 88	P value
Mild	17 (8.9%)	3 (3.4%)	.054
Moderate	3 (1.6%)	0 (0.0%)	.083
Severe	5 (2.6%)	1 (1.1%)	.360

Of those with biopsy-proven RCC (Table 3), recurrence was observed with 2.7% (3/113) of CAs and 3.5% (2/57) of MWAs (P = .77) (Table 4). Median follow-up interval was 3.9 years for CA (range, 0–11.9 years; IQR, 4.8 years) and 1.1 years for MWA (range, 0–5.7 years; IQR, 1.8 years). Four of the 5 local recurrences occurred after ablation of tumor recurrences at a prior partial nephrectomy site. All 5 were either intermediate or high complexity based on their nephrometry score. The time to recurrence ranged from 0.8 to 1.3 years. Disease-free survival rates for CA and MWA were comparable using the Mantel–Cox test (P = .32) (Fig 3). Cancer-specific survival was 100% for MWA and 98.2% for CA (P = .31). Two patients developed metastatic disease. In both cases, the index lesion had been treated with CA. One patient developed tumor seeding along the probe track, declined all treatments, and subsequently developed remote metastases. The second patient developed a 3.5-cm local recurrence <1 year after the treatment of the index lesion and, on restaging, was diagnosed with lung metastases. OS was 70.8% (80/113) for those who underwent CA and 91.2% (52/57) for those who underwent MWA (P < .001). Secondary analysis using propensity score matching demonstrated no significant differences in local recurrence rates (coefficient, −0.025 [SE ± 0.029]; z −0.86; P = .39), adverse event rates (coefficient, −0.086 [SE ± 0.067]; z = −1.28; P = .20), cancer-free survival (coefficient, −0.0035 [SE ± 0.0119]; z = −0.30; P = .76), or OS (coefficient, −0.064 [SE ± 0.048]; z −1.35; P = .19) when comparing matched cohorts of MWA and CA patients.

Table 3.

Pathology Results

Pathology result		Total no. of biopsies (n = 239)
Malignant	170	Renal cell carcinoma
		Clear cell	107
		Chromophobe	13
		Papillary	51
		Other	1
Benign	41	Angiomyolipoma	4
		Oncocytoma	30
		Other	6
Nondiagnostic	28

Table 4.

Local Recurrence

Patient no.	Ablation modality	Personal history of renal cell carcinoma	Size of the primary lesion (cm)	Nephrometry score	Pathology of the primary lesion	Size of the local recurrence (cm)	Time to recurrence (y)	Management of local recurrence	Pathology of the local recurrence
1	MWA	Recurrence after prior partial nephrectomy Contralateral Radical nephrectomy	2.1	10p	Clear cell Grade 2	0.8	1.3	MWA	Benign renal cortex
2	MWA	Recurrence after prior partial nephrectomy	3.5	7p	Clear cell Grade 2	1	0.8	Total nephrectomy	Clear cell Grade 2 1.4-cm local recurrence at the ablation site and 2.5-cm recurrence at the superior aspect of the partial nephrectomy site
3.	CA	Treatment-naive	3.5	10x	Clear cell Grade 2	3.5	0.8	Patient declined further treatment*	N/A
4	CA	Recurrence after prior partial nephrectomy	2.7	10x	Clear cell Grade 1	1.1	0.8	CA	Benign interstitial inflammation and fibrosis
5	CA	Recurrence after prior partial nephrectomy	3	7p	Clear cell Grade 2	2.5	1.1	CA	Clear cell Grade 2

CA = cryoablation; MWA = microwave ablation; NA = not applicable.

^*The patient developed concurrent lung metastases.

Figure 3.

Disease-free survival of cryoablation (CA) versus microwave ablation (MWA).

A paired t-test comparing preablation and postablation estimated glomerular filtration rates showed no significant difference in renal function after CA (P = .76) or MWA (P = .49).

DISCUSSION

In the early years of small renal mass ablation, CA was the only ablation modality used, and most patients were treated using moderate sedation. With the advent of MWA in 2013, a need to treat more complex lesions, and an increasing number of obese patients (>40%) in clinical practice, general anesthesia became the standard of care for these procedures. This retrospective review showed a tendency to treat smaller, less complex lesions with MWA. Although anterior and exophytic lesions were more likely to be treated with MWA and endophytic lesions were more likely to be treated with CA, the differences between the 2 groups were not significant. The operator bias may be attributed to greater experience with CA, the ability to clearly identify the treatment zone (ice ball) with CA compared with gas cloud with MWA, and a concern for injury to the collecting system when treating central lesions with MWA—a concern also raised by other IR groups (22–24).

The proximity to adjacent structures did not serve as a contraindication for the utilization of either modality; hydrodissection was used in over a third of cases, whereas pyeloperfusion was performed in just 6% of cases overall. Owing to the reduced number of probes needed, infrequent reliance on pyeloperfusion, and notably briefer ablation duration in MWA than in CA, this translated into a significantly shorter procedural timeframe, a finding that has also been reported by others (12,24,25). The longer length of stay for patients undergoing CA relates primarily to the standard overnight observation followed in the early years of the program.

Both modalities demonstrated excellent technical success outcomes. The higher adverse event rates observed among cryoablated lesions may be attributed to a few factors. Much of the expertise and development of skills were acquired using CA in the first 5 years of the program. Patients treated with CA were more likely to have a greater number of comorbidities, and their tumors were more technically challenging. As the customary approach involves conducting a percutaneous biopsy simultaneously with the ablation, determining the actual risk of bleeding solely from the ablation proves challenging. Bleeding adverse events were also more common after CA and may be related to the fact that tumors treated with CA were generally larger and more likely to be centrally located, thus requiring more probes. This hypothesis was reinforced by propensity score matching, with a well-balanced cohort showing no significant difference (P = .20) in adverse event rates when matched with patients based on Charlson Comorbidity Index score, tumor size, number of probes, nephrometry score, and pyeloperfusion. Additionally, the cryoprobe employed possessed a larger gauge (13 gauges) than that of the MWA probe (17 gauges), and there existed no capability to cauterize the tract upon the removal of the cryoprobes. In the course of MWA, dielectric heating at the electrode tip initiates thermal coagulation, leading to coagulation necrosis in the tissues. Consequently, the likelihood of postcoagulation bleeding in tissues subjected to MWA is reduced.

The adverse event rate of 4.5% after MWA is comparable with that reported by other groups (Table 5) (26–31). The only major adverse event incurred after MWA was not attributable to tumor location but rather an overzealous treatment performed early in the MWA program. In a retrospective review by Maciolek et al (13), 151 T1a RCCs were treated with MWA. An overall adverse event rate of 13.5% was reported, with 2.7% classified as major. There were no nontarget organ injuries attributed to tumor location. Such injuries were mitigated through appropriate patient positioning and judicious use of hydrodissection. Similarly, An and Arellano (32) reported no significant difference in primary technical effectiveness or adverse event rate after CT-guided MWA of central RCC compared with that of peripheral RCC when adjunctive procedures (hydrodissection and pyeloperfusion) were used.

Table 5.

Outcomes after Microwave Ablation of Small Renal Masses

Author	Year	Patient/tumor number	Tumor size (cm)	Primary effectiveness (%)	Secondary effectiveness (%)	Median follow-up (mo)	Adverse events (Clavien-Dindo classification)	Outcomes
Guo and Arellano (26)	2021	106/119	Mean (range) 2.4 (1.7–3.1)	95.3	100	24	5 Grade I 2 Grade III	OS, 1 y, 99%; 2 y, 97.7%; 3 y, 94.6% PFS, 1 y, 100%; 2 y, 97.7%; 3 y, 94.6% CSS, 100% LR, 5.7%
De Cobelli et al (25)	2020	28/32	Median (IQR) 2.2 (1.1–3.3)	94		20	6.3% (2/32) 2 CIRSE Grade 1	LR, 3.3%
Zhou and Arellano (12)	2018	38/44	Mean (range) 2.5 (1.2–6.1)	100	N/A	1	13% (5/44) 5 Grade I	Decreased procedure time, MWA vs CA and RF ablation Decreased conscious sedation dosage requirements for MWA No significant difference in renal function before and after MWA
John et al (27)	2021	113/113	Median (IQR) 2.5 (2.0–3.2)	100	N/A	12	Grade I 18% Grade II 1.8% Grade IIIb 0.9% Grade IV 0.9%	LR, 0.9% Metastases, 1.8% (2/113)
Wilcox et al (28)	2021	101/110	Median (IQR) 2.0 (1.5–2.6)	96	99	12	2 Grade I 1 Grade II 1 Grade III	LR, 2.4%
Wells et al (29)	2016	23/24	Median (IQR) 2.8 (2.1–3.3)	96	100	12	1 Grade I 2 Grade II	LR, 0% CSS, 97%
Aarts et al (30)	2020	77/85	Median (IQR) 2.8 (2.2–3.5)	89	99	19	6 Grade I 1 Grade II	LR, 2% Metastases, n = 4 CSS, 100%
Ruiz et al (31)	2020	93/101	Median (range) 2.5 (1.0–4.2)*	96.2	97.1	Mean, 25	5.6% 4 Grade I 1 Grade II 1 Grade IIIa	LR, n = 1 Metastases, n = 2
Yuetal (15)	2011	185/192	Mean, 2.3 (SD ± 0.5)	93.2	N/A	42	2.2% 3 Grade III 1 Grade IV	LR, 3.2% CSS, 97.8% Metastases, 4.3%
Maciolek et al (13)	2019	148/151	Median (IQR) 2.4 (1.9–3.0)	100	N/A	27	13.5% 4 Grades III and IV	LR, 4% 3-y RFS, 95% 3-y CSS, 100% 3-y OS, 96%
Hao et al (14)	2018	162/171	Mean, 2.6 (SD ± 0.8)	98.2	100	45.5	N/A	LR, 3%
Sun et al (present study)		83/88	Mean, 2.2 (SD ± 0.8)	100	N/A	13	4.5% 3 Grade I 1 Grade IIIa	LR, 3.5% CSS, 100 % Metastasis-free survival, 100%

CA = cryoablation; CIRSE = Cardiovascular and Interventional Radiological Society of Europe; CSS = cancer-specific survival; IQR = interquartile range; LR = local recurrence rate; MWA = microwave ablation; OS = overall survival; PFS = progression-free survival; RF = radiofrequency; RFS = recurrence free survival.

^*The study group included 4 T1b lesions.

Local recurrence rates were similar for the 2 groups and comparable with recurrence rates reported after partial nephrectomy (12,25,32–34). The MWA local recurrence rate was comparable with that reported by other institutions (Table 5). MWA and CA cohorts demonstrated no significant differences in local recurrence rates before and after propensity score matching (P = .77 and P = 39, respectively) based on ablated malignant lesions. All recurrences were reported in the first 2 years and would suggest inadequate treatment of the index lesion. Four of the 5 recurrences occurred in patients treated for a local recurrence after partial nephrectomy. The size of the postsurgical recurrences ranged from 2.1 to 3.5 cm. Nephrometry scores were medium to high complexity. The higher recurrence rate among these tumors may reflect a more aggressive tumor biology and/or difficulty in identifying the true extent of the disease with cross-sectional imaging. In the case of Patient 2, the recurrence after surgical resection was identified 5 years after partial nephrectomy. Pathology from completion of nephrectomy showed a second tumor (not visualized with MR imaging of the abdomen) at the superior margin of the partial nephrectomy bed in addition to the recurrence at the thermal ablation site.

Two of the lesions deemed recurrent (Table 4, Patients 1 and 4) were reablated; however, the biopsy histology was benign. The sizes of the local recurrences were 1.1 and 0.8 cm. Hao et al (14) reported a 3% recurrence rate per patient over a median follow-up of 45.5 months after MWA of 171 lesions in 162 patients. In their analysis, they found higher recurrence rates (8%) among lesions in “dangerous locations” (the distance between the tumor margin and bowel or renal pelvis was <5 mm as measured by US) than in those in safe locations (0.8%). Maciolek et al (13) reported 4% (6/151) of local recurrences at a median follow-up of 27 months. In their analysis, they reported that tumor location (anterior vs posterior) had no impact on local tumor control. In terms of tumor size, there was no significant impact on recurrence for MWA, similar to that reported by Efthymiou et al (35). The survival data showed no deaths from RCC after treatment with MWA compared with 2 deaths from metastatic disease in patients treated with CA. The cancer-specific survival rate of 100% is similar to that reported by other institutions (Table 5). Additional analysis showed no significant difference between the MWA and CA cohorts in cancer-specific survival before and after propensity score matching analysis (P = .31 and P = .76, respectively) (Table 6).

Table 6.

Summary Data

Variable	Overall dataset			After PSM
	MWA	CA	P value	MWA	CA	P value
Tumor size (cm), mean ± SD	2.22 ± 0.83	2.61 ± 0.75	<.01	2.22 ± 0.83	2.26 ± 0.84	.75
Charlson Comorbidity Index Score, median (IQR)	4 (3–6)	5 (3–7)	<.01	4 (3–6)	4 (3–6)	.24
Overall adverse event rate (%)	4.5	13.1	.01	4.5	9.4	.20
Local recurrence rate (%)	3.5	2.7	.77	3.5	1.5	.39
Cancer-specific survival (%)	100	98.2	.31	100	98.9	.76

CA = cryoablation; IQR = interquartile range; MWA = microwave ablation; PSM = propensity score matching.

This study is limited by the retrospective collection of data, treatment selection bias, difference in the size of the 2 treatment groups, and shorter duration of follow-up available for MWA than for CA. In addition, the overall power of this study may be limited by the sample size because factors such as adverse event rate, primary technical effectiveness, and local tumor recurrence are rare at baseline. Risk factors for local tumor progression with MWA are difficult to analyze given the low number of recurrences and shorter follow-up interval. Furthermore, CA was the only modality in use in the early years of the program, when technology may not have been as advanced. This is further compounded by variable ablation selection at the discretion of the treating interventional radiologist through different periods of time. Future prospective studies with predetermined treatment arms may be necessary to further evaluate effectiveness and factors impacting local recurrence.

Over a 12-year period of using CT-guided thermal ablation, both percutaneous MWA and CA have proven to be safe and effective in treating kidney tumors, resulting in outstanding local tumor control. MWA and CA were viewed as complementary rather than competing technologies in renal applications, with a preference toward treating smaller, less complex tumors with MWA. Local recurrences occurred almost exclusively in cases where the index lesion manifested as a postsurgical recurrence.

RESEARCH HIGHLIGHTS.

• Cryoablation and microwave ablation both provided high technical success and local disease control of cT1a (≤4 cm) small renal masses with no significant impact on renal function.

• Cryoablation lesions at the authors’ institution were significantly larger and of greater complexity, but cancer-specific survival was comparable between the 2 modalities.

• Microwave ablation and cryoablation can be viewed as complementary technologies in the kidney, with a lean toward treating smaller, less complex lesions with microwave ablation and larger, more complex lesions with cryoablation.

ABBREVIATIONS

CA

cryoablation

CT

computed tomography

IQR

interquartile range

IR

interventional radiology

IRB

institutional review board

MFS

metastasis-free survival

MR

magnetic resonance

MWA

microwave ablation

OS

overall survival

RCC

renal cell carcinoma

RENAL

radius, exophytic/endophytic, nearness to collecting system or sinus, anterior/posterior, and location relative to polar lines

RF

radiofrequency

US

ultrasound