Image-guided radiofrequency ablation for adrenal metastatic tumors: a systematic review and meta-analysis

Abstract

Purpose

To evaluate the safety and clinical efficacy of image-guided radiofrequency ablation (RFA) in patients exhibiting adrenal metastatic tumors (AMTs).

Methods

The PubMed, Web of Science, and Wanfang databases were searched for relevant studies published as of November 2022, and study results were pooled for subsequent analysis. The endpoints of this meta-analysis included primary and secondary technical success, local hemorrhage, pneumothorax, hypertensive crisis, local recurrence, and 1- and 3-year overall survival (OS) rates.

Results

This analysis incorporated 11 studies enrolling 351 patients that underwent RFA treatment for 373 AMTs. The pooled primary and secondary technical success, local hemorrhage, pneumothorax, hypertensive crisis, local recurrence, and 1- and 3-year OS rates in these patients were 84%, 91%, 4%, 6%, 7%, 19%, 82%, and 46%, respectively. The 1-year OS (I² = 75.2%, p = 0.003) and 3-year OS (I² = 81.4%, p = 0.001) endpoints were subject to significant heterogeneity. Subgroup analyses suggested that primary technical success rates were < 80% in patients with tumors exhibiting a mean diameter ≥ 4 cm. Guidance type and tumor size had no impact on a hypertensive crisis or local recurrence rates.

Conclusions

These data indicate that image-guided RFA represents a safe and effective means of treating AMTs.

Keywords:

• Radiofrequency
• adrenal
• metastatic
• meta-analysis
• microwave

Introduction

Tumors often metastasize to the adrenal glands [1–3], with up to 27% of patients with known malignancies exhibiting adrenal metastatic tumors (AMTs) identified on autopsy [4]. AMTs are often linked to poorer patient prognostic outcomes [1]. While adrenalectomy in patients with isolated AMTs can improve survival outcomes [5,6], those patients exhibiting severe comorbidities may not be eligible for surgical tumor resection [1].

A range of ablation strategies has been designed to facilitate local AMT control, including cryoablation, laser ablation, radiofrequency ablation (RFA), and microwave ablation (MWA) [1,2,4,7]. Prior meta-analyses have suggested that image-guided ablation can safely and effectively treat patients harboring adrenal tumors [1,2], but these analyses are subject to certain limitations. The meta-analysis performed by Pan et al. [1] incorporated multiple ablation techniques, and the results were subject to potential selection bias. While Nadeem et al. [2] only analyzed outcomes associated with RFA, their analyses focused on both primary adrenal tumors and AMTs, contributing to a high risk of bias. There is thus a need for a specific meta-analysis focused solely on the ablation of AMTs using a single technique.

Here, a meta-analysis was thus performed to evaluate the clinical safety and efficacy of image-guided RFA as a means of treating patients with AMTs.

Materials and methods

Study selection

This Meta-analysis Of Observational Studies in Epidemiology (MOOSE) checklist [8] was used to guide the design of this study, which was registered at inplasy.com (No. INPLASY2022120050).

The PubMed, Web of Science, and Wanfang databases were searched for relevant studies published through November 2022 using the following terms: (radiofrequency) AND (((ablation) AND (adrenal)) AND ((metastasis) OR (metastatic))).

Studies eligible for inclusion were those that (a) focused on the image-guided RFA treatment for AMTs; (b) sample size > 10; (c) studies reporting at least one or more of the endpoints of interest (rates of technical success, hypertensive crisis, pneumothorax, local hemorrhage, or 1- or 3-year overall survival [OS]). Studies were excluded if they: (a) were animal studies; (b) employed RFA in combination with other ablation techniques; or (c) were reviews or meta-analyses.

Data extraction

Relevant data were independently extracted from selected studies by two authors, with any discrepancies being resolved via discussion and consensus with a third researcher. Baseline study data are compiled in Table 1.

Table 1. Baseline data of the included studies.

	Year	Country	Patients number	AMT number	Mean age	M/F	Mean diameter	Primary tumors	Mean follow-up	Follow-up methods
Hasegawa [9]	2015	Japan	35	41	64.7 y	25/10	3.3 cm	HCC, LC, RCC, CC, others	30.1 m	CT
Huang [10]	2019	China	22	22	53.0 y	21/1	4.0 cm	HCC	10 m	Not given
Li [11]	2013	China	30	30	64.2 y	22/8	4.2 cm	HCC, LC, RCC, GC, BC	Not given	CT
Liu [12]	2020	China	29	29	52.9 y	25/4	Not given	HCC, LC, RCC, others	24.5 m	CT and US
Liu B [13]	2020	China	16	16	53.3 y	15/1	3.6 cm	HCC	Range: 3-55 m	CT and MRI
Sui [14]	2019	China	70	70	54.4 y	38/32	Not given	HCC	Not given	Not given
Wang [15]	2009	China	19	24	58 y	13/6	3.5 cm	HCC, LC, RCC, CC	10 m	CT
Yuan [16]	2018	China	38	38	54.2 y	26/12	3.3 cm	HCC	26.3 m	CT and MRI
Zhang [17]	2011	China	13	13	59 y	10/3	4.5 cm	HCC	14.2 m	CT
Zhou [18]	2018	China	33	38	60 y	17/16	Not given	HCC, LC, RCC, CC, OC	22.8 m	CT and MRI
Zhou K [19]	2018	China	46	52	64.4 y	28/18	3.31 cm	HCC, LC, RCC, CC, OC	11.5 m	CT and MRI

AMT: adrenal metastatic tumor; BC: breast cancer; CC: colonic cancer; CT: computed tomography; F: female; GC: gastric cancer; HCC: hepatocellular carcinoma; LC: lung cancer; M: male; MRI: magnetic resonance imaging; OC: ovarian cancer; RCC: renal cell carcinoma; RFA: radiofrequency ablation; US: ultrasound.

The primary technical success of RFA was defined by the proportion of target AMTs that were successfully eliminated based on the disappearance of tumor enhancement upon contrast-enhanced CT or magnetic resonance imaging (MRI) analyses following the completion of the ablation procedure [2]. Secondary technical success was defined as technical success following a second round of RFA to treat residual tumors [2]. Local recurrence was defined by the development of a new area of AMT-associated enhancement observed during follow-up after an ablation procedure that initially achieved technical success [4]. Hypertensive crisis incidence was defined by systolic or diastolic blood pressure values greater than or equal to 180 mmHg or 110 mmHg, respectively [20]. Overall survival (OS) was defined by the period of time between initial RFA treatment and all-cause death.

Quality assessment

Non-randomized controlled trial study quality was analyzed using the Newcastle-Ottawa scale (NOS) [21], with points being assigned based on selection (4 scores), comparability (2 scores), and outcome (3 scores) criteria. High-quality studies were those exhibiting a total score ≥ 7.

Statistical analyses

Random-effects models were employed to assess pooled endpoint data. The Q test and I² values were used to assess potential heterogeneity, with I² > 50% as the threshold for heterogeneity being considered significant. A leave-one-out approach was used to conduct sensitivity analyses, while subgroup analyses were conducted based on differences in guidance methods, primary tumor, and mean diameter. Publication bias was assessed with Egger’s test, and p < 0.05 was the significance threshold. Pooled analyses were performed with Stata 12.0.

Results

Study inclusion

An initial search identified 369 potentially relevant studies, of which 11 were ultimately incorporated into this meta-analysis (Figure 1). Of these studies, one was performed in Japan [9] and 10 were performed in China [10–19]. These studies included 351 patients (240 males, 111 females; mean age: 52.9–64.7 years) undergoing RFA to treat 373 AMTs (Table 1). The mean AMT diameter in these patients ranged from 3.3 − 4.5 cm. The details of parameters of RFA were shown in Table 2. CT guidance was performed in 6 studies [9,14–16,18,19], while 5 employed ultrasound guidance [10–13,17]. The specific number of RFA probes was not provided in any of the included studies. Hydro dissection was mentioned in 5 studies [9,10,12,13,16], while 6 studies did not mention the situation of hydro dissection [11,14,15,17–19].

Figure 1. Study flowchart.

Table 2. Baseline data of the RFA.

	Guidance	RFA devices	Electrode size	Maximum power	Hydro dissection
Hasegawa [9]	CT	Cooltip	Not given	Not given	Yes
Huang [10]	US	Cooltip	17 G	200 W	Yes
Li [11]	US	Valleylab	Not given	150 W	Not mentioned
Liu [12]	US	LeVeen, Starburst XL, Cooltip	Not given	Not given	Yes
Liu B [13]	US	Cooltip	Not given	Not given	Yes
Sui [14]	CT	Not given	Not given	60 W	Not mentioned
Wang [15]	CT	Starburst XL	16 G	150 W	Not mentioned
Yuan [16]	CT	RITA	Not given	Not given	Yes
Zhang [17]	US	Radionic	17.5 G	200 W	Not mentioned
Zhou [18]	CT	CelonLab	Not given	250 W	Not mentioned
Zhou K [19]	CT	CelonLab	Not given	250 W	Not mentioned

CT: computed tomography; RFA: radiofrequency ablation; US: ultrasound.

Five of these studies focused on the primary tumor of hepatocellular carcinoma (HCC) [10,13,14,16,17]. In the remaining studies, each patient also only had one type of primary tumor and each study contained primary HCC patients, primary lung cancer (LC) patients, primary renal cell carcinoma (RCC) patients, and patients with other primary tumors. The details of the primary tumor types were shown in Table 1. All 11 studies exhibited NOS scores from 7–8.

Primary technical success

All studies reported primary technical success rates, with a pooled 84% rate (95% CI: 0.8–0.88, Figure 2(a)). Low heterogeneity was detected (I² = 0.0%, p = 0.583), although Egger’s test revealed significant publication bias (p = 0.001).

Figure 2. The results of pooled analyses of (a) primary technical success rate, (b) secondary technical success rate, (c) local hemorrhage rate, (d) pneumothorax rate, (e) hypertensive crisis rate, (f) local recurrence rate, (g) 1-year OS rate, and (h) 3-year OS rate.

Secondary technical success

Four studies reported secondary technical success rates [9–12], with a pooled 91% rate (95% CI: 0.85–0.96, Figure 2(b)). Low heterogeneity was detected (I² = 0.0%, p = 0.675), although Egger’s test revealed significant publication bias (p = 0.001).

Local hemorrhage

Local hemorrhage rates were reported in three studies [9,18,19], with a pooled 4% rate (95% CI: 0.00–0.08, Figure 2(c)). Low heterogeneity was detected (I² = 13.6%, p = 0.314). Publication bias could not be assessed with Egger’s test by Stata 12.0.

Pneumothorax

Five studies reported pneumothorax rates [14,16–19], with a pooled 6% rate (95% CI: 0.03–0.09, Figure 2(d)). Low heterogeneity was detected (I² = 0.0%, p = 0.998). Publication bias could not be assessed with Egger’s test by Stata 12.0.

Hypertensive crisis

Hypertensive crisis incidence was reported in all 11 studies, with a pooled 7%rat (95% CI: 0.04–0.10, Figure 2(e)). Low heterogeneity was detected (I² = 24.3%, p = 0.212), and there was no evidence of significant publication bias (Egger’s test, p = 0.238).

Local recurrence

Local recurrence rates were reported in eight studies [10–13,15,16,18,19], with an overall pooled rate of 19% (95% CI: 0.14–0.24, Figure 2(f)). Low heterogeneity was detected (I² = 0.0%, p = 0.886), and there was no evidence of significant publication bias (Egger’s test, p = 0.231).

1-Year OS

Rates of 1-year OS was reported in five studies [9,10,12,14,16], with a pooled rate of 82% (95% CI: 0.72–0.92, Figure 2(g)). High heterogeneity was detected (I² = 75.2%, p = 0.003), as was evidence of publication bias (p = 0.039). The study by Huang et al. [10] was identified as the source of heterogeneity, and when it was excluded the pooled 1-year OS rate was 88% (95% CI: 0.82–0.94), with low heterogeneity (I² = 38.3%, p = 0.182).

3-Year OS

Analyses of the results of four studies [9,12,14,16] revealed a pooled 3-year OS rate of 46% (95% CI: 0.29–0.63, Figure 2(h)). High heterogeneity was detected (I² = 81.4%, p = 0.001), although there was no evidence of publication bias (p = 0.214). Sensitivity analyses did not clarify the source of this significant heterogeneity.

Subgroup analyses

Subgroup analysis results for the primary technical success rate are summarized in Table 3. Pooled rates for patients that underwent ultrasound and CT guidance were 80% and 85%, respectively, and pooled rates for patients with a mean tumor diameter <4 cm and ≥4 cm were 86% and 79%.

Table 3. Subgroup analysis of primary technical success rate.

	Studies (n)	Primary technical success rate	95% CI	I^2
Guidance methods
US	5	80%	0.72–0.87	0.0%
CT	6	85%	0.81–0.90	0.0%
Mean diameter
<4 cm	5	86%	0.81–0.91	0.0%
≥4 cm	3	79%	0.70–0.89	0.0%

CI: confidence interval; CT: computed tomography; US: ultrasound.

Subgroup analysis results for hypertensive crisis rates are summarized in Table 4. Pooled rates for patients that underwent ultrasound and CT guidance were 7% and 6%, respectively, and pooled rates for patients with a mean tumor diameter <4 cm and ≥4 cm were 7% and 5%.

Table 4. Subgroup analysis of hypertension crisis rate.

	Studies (n)	Hypertension crisis rate	95% CI	I^2
Guidance methods
US	5	7%	0.02–0.12	0.0%
CT	6	6%	0.03–0.09	44.9%
Mean diameter
<4 cm	5	7%	0.03–0.11	55.5%
≥4 cm	3	5%	0.00–0.10	0.0%

CI: confidence interval; CT: computed tomography; US: ultrasound.

Subgroup analysis results for local recurrence rates are summarized in Table 5. Pooled rates for patients that underwent ultrasound and CT guidance were 21% and 18%, respectively, and pooled rates for patients with a mean tumor diameter <4 cm and ≥ 4 cm were 17% and 20%.

Table 5. Subgroup analysis of local recurrence rate.

	Studies (n)	Local recurrence rate	95% CI	I^2
Guidance methods
US	4	21%	0.13–0.29	0.0%
CT	4	18%	0.12–0.25	0.0%
Mean diameter
<4 cm	4	17%	0.10–0.24	0.0%
≥4 cm	2	20%	0.09–0.31	0.0%

CI: confidence interval; CT: computed tomography; US: ultrasound.

Discussion

This meta-analysis showed the safety and efficacy of RFA as a treatment for patients harboring AMTs. While some prior meta-analyses have also explored the value of the image-guided percutaneous ablation of adrenal tumors [2,3], this study only focused on the use of RFA for patients with AMTs. Firstly, RFA is the most commonly used method for AMTs [1,2]. In addition, only including RFA and AMTs might decrease the risk of bias.

In this patient population, the pooled primary technical success rate was 84%, consistent with the good short-term efficacy of RFA as a treatment for AMTs. The 16% of primary technical failure may be attributed to the fact that RFA heat may be reduced by the abundant blood flow of the AMTs [13]. The high pooled rate of secondary technical success (91%) was also consistent with the repeatability of this image guidance-based ablation strategy. These pooled primary and secondary technical success rates were lower than values reported in prior studies of the cryoablation- and MWA-based treatment of AMTs (90.3–94.9% and 100%) [4,22–24]. However, those prior studies were relatively limited in number such that they may not fully capture the potential range of AMT responses to these ablation strategies.

Subgroup analyses for the pooled rates of primary technical success indicated that these rates were consistent > 80% irrespective of guidance type (CT vs. ultrasound), suggesting that these factors do not influence the efficacy of RFA. In contrast, primary success rates were significantly higher for patients with tumors <4 cm in diameter relative to those with larger tumors (86% vs. 79%). This is presumably attributable to the limited target area for RFA.

Local hemorrhage and pneumothorax are rare complications when adrenal tumors are treated via percutaneous ablation [4,22–27]. Consistently, the pooled local hemorrhage and pneumothorax rates calculated in the present study were just 4% and 6%, respectively. Hypertensive crisis is the most severe outcome associated with percutaneous adrenal tumor ablation, but it is most common in individuals harboring primary adrenal malignancies, occurring in up to 67.5% of patients [27,28]. The pooled hypertensive crisis rate in this study was just 7%, indicating that RFA can be safely used for AMT treatment.

In this analysis, the pooled local recurrence rate was 19%, in line with rates of 19–23% reported in prior analyses of WMA- and cryoablation-based AMT treatment [4,22,24]. This suggests that RFA may offer similar value to these two other ablation techniques with respect to the local control of AMTs. Liu et al. [12] reported that RFA was related to a significantly higher local recurrence incidence as compared to adrenalectomy (24% vs. 6.5%, p = 0.048). Given that RFA is a minimally invasive procedure, however, it can be used for patients who were not suitable for surgical resection.

Subgroup analysis results revealed that guidance methods and AMT diameter did not influence local recurrence rates, in line with prior results from a study focused on the percutaneous ablation of AMTs [29].

Pooled respective 1- and 3-year OS rates in the present meta-analysis were 82% and 46%, revealing that RFA can effectively facilitate disease control in patients harboring AMTs. However, these results were subject to significant heterogeneity, underscoring a need for additional prospective studies as a means of reducing the risk of bias and achieving more reliable results.

This meta-analysis is subject to some limitations. Firstly, the analyzed studies were all retrospective in design, and the associated risk of bias was high. Additional prospective validation will be essential. Secondly, high heterogeneity was observed for the 1- and 3-year OS endpoints and the source of heterogeneity for 3-year OS was not established, potentially because this endpoint was reported in just 4 studies. Thirdly, all included studies were conducted in Asia, and therefore, these results could not reflect the clinical effectiveness of RFA for AMTs all over the world.

Conclusion

In summary, these meta-analysis results demonstrate that image-guided RFA can safely and effectively treat AMT, but the long-term outcomes associated with this interventional strategy warrant further investigation.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Additional information

Funding

This work is supported by the Innovation Program of Xuzhou Commission of Health (No. XWKYHT20220111).