Feasibility, safety, and long-term efficacy of stereotactic radiofrequency ablation for tumors adjacent to the diaphragm in the hepatic dome: a case-control study

RF ablation has been increasingly accepted as a curative alternative to surgical resection in the management of primary or metastatic liver tumors [1, 2]. The achievement of a sufficient safety margin, i.e., complete coverage of the coagulation zone with a margin of at least 5 mm, is crucial for good local tumor control and clinical outcome [3]. Treatment of subdiaphragmatically located tumors located in the hepatic dome is challenging when the target tumor is difficult to visualize, awkward to access, or situated adjacent to vulnerable structures. Treatment of tumors located close to the diaphragm implies a higher risk of complications such as pneumothorax or diaphragmatic injuries. Several studies reported severe right shoulder pain after treatment of hepatic dome tumors [4, 5].

Common strategies include angulated approach sparing the pleura [5] or transthoracic or transpleural access [6‐8]. Hydrodissection is another possible protective method in the hepatic dome [9]. In addition, these tumors located posteriorly and superiorly also pose a significant technical challenge for laparoscopic liver resection, demanding special approaches and techniques [10]. There are several studies reporting limited local tumor control in subcapsular locations with conventional RF ablation [11], higher complication rates (mainly bleeding), as well as tumor recurrence [12, 13]. Stereotaxy has proven useful for planning and execution of complex or difficult ablations. Difficult access routes may be specifically facilitated and more precise coverage of the target tumor and safety margin accomplished with frameless stereotactic navigation systems [14].

The purpose of the current study was to assess the feasibility, safety, and clinical outcome of multi-probe SRFA of liver tumors close to the diaphragm and to compare the results with a matched control group. Our hypothesis was that there is no difference in terms of safety and local control for tumors in the hepatic dome vs. non-hepatic dome tumors.

This retrospective analysis with a propensity score–matched control cohort suggests that SRFA of subphrenic tumors can be accomplished with comparable safety and efficacy rates achieved for tumors treated elsewhere in the liver. The technical success rate, i.e., completion according to plan, in our study was 100%, which is in line with reports from the literature [5, 18]. The primary technical efficacy was 97.5%, also comparable with reported primary technical efficacy rates after conventional RF ablation in the hepatic dome ranging from 86.7% [5] to 93.2–96% [19, 20]. The conventionally treated tumors were smaller (median diameter 1.7 cm and maximal diameter 4 cm) compared with our study with a median tumor size of 2.2 cm and a maximal diameter of 10 cm. We observed local tumor progression in 21 (8.8%) of 238 tumors, with no significant difference in local tumor treatment outcomes compared with the control group (p = 0.414). Kim et al [5] reported local tumor progression in 13.3% after RF ablation of subdiaphragmatic HCCs with a transthoracic approach despite the inclusion of smaller lesions. Cha et al [21] compared the therapeutic outcomes of conventional US-guided RF ablation for subcardiac and non-subcardiac HCCs in 73 patients and reported no significant difference observed between both groups, with a cumulative local tumor progression rate > 15% in the subcardiac group. Recently, Vo Chieu et al [18] showed local tumor progression rates of up to 13.6% in the “risk group” after microwave ablation of tumors abutting the diaphragm. Microwave ablation (MWA) generally requires fewer overlapping ablations compared with RFA. Asvadi et al [22] treated 48 hepatic dome lesions in 46 patients with conventional CT-guided MWA and reported a complete response rate of 94%. Transarterial chemoembolization (TACE) is the recommended locoregional treatment for patients with intermediate HCC [23, 24]. Several studies reported the feasibility and benefit of RF ablation after transarterial embolization with complete response rates of 76–82% in large HCC [25, 26]. We attribute the good local tumor control after SRFA to consistent achievement of a sufficient safety margin of at least 5 mm [3]. SRFA offers three-dimensional ablation planning to achieve optimal alignment of RF probes to create multiple overlapping coagulation volumes. Usage of an aiming device facilitates very precise path alignment and targeting [14]. In case of poor tumor visibility, SRFA planning may include fusion with previously acquired MR images. Temporary disconnection of the endotracheal tube facilitates good control of respiratory motion [27]. Immediate postablation contrast-enhanced CT fusion with the planning CT allows for rapid reliable judgment of the ablation results with the option for re-ablation. Bale et al [28] achieved complete pathological response after SRFA (i.e. no evidence of tumor) with these techniques in 183 of 188 (97.3%) hepatocellular carcinomas with a median size of 2.5 cm (range, 1–8) in a histopathological study in explanted livers.

The major complication rate in our study was 12.3%, slightly but not significantly higher than the 8.8% rate in the control group with random non-subdiaphragmatic tumors. The majority of complications such as pleural effusion or pneumothoraces were relatively easy to treat. Significant bleeding in all but one instance could be immediately managed with angiographic coiling in the same general anesthesia session. Rhim et al [29] reported a low complication rate of 4% (1/25) with artificial ascites during RF ablation of subdiaphragmatic tumors, and Kim et al [5] described no major complications in a small series of 15 patients treated by RF ablation with an angulated transhepatic approach. More recently, Vo Chieu et al [18] reported major complication rates of up to 57.9% with microwave ablation in subcapsular hepatic dome tumors with a transpleural approach while Ding et al [20] reported major complication rate of 10% (6/60) with a transhepatic approach. Asvadi et al [22] showed that creation of artificial ascites is a valuable option for avoiding diaphragm-associated complications. Application of this technique would have most likely prevented most of the complications related to thermal injury of the diaphragm in our patient series. In unresectable HCCs, Zhang et al (16) reported a major complication rate of 4.4% after combined TACE-RF and Zhao et al (19) of 2.3% after TACE and 9.4% after RF ablation, respectively. The complication rate of our study is moderately higher, which may be due in part to the larger size of our treated tumors and the larger number of inserted needles (median 9 needles, ranging from 3 to 22). Supporting this assumption, a subanalysis of our data showed that the complication rate in larger tumors treated with more RF probes was significantly higher. However, the major complication rate of the present study compares favorably with the reported rates following laparoscopic or open liver resection of primary or metastatic liver tumors in the posterosuperior segments of the liver. According to the in-surgical studies that primarily used Clavien-Dindo classification [30], the major complicate rate in the present study was 4.9% (10/204). Major complication rates after laparoscopic RFA were 10% and 12.4%, respectively [20, 31]. Several surgical studies report significantly higher major complicate rates of 10–27% after LLR, and 18–37% after open resection, respectively [10, 32‐35].

Overall survival rates after SRFA of colorectal liver metastases [16], breast cancer liver metastases [36], intrahepatic cholangiocellular carcinomas [37], and melanoma liver metastases [38] are comparable with those of liver resection. In the current study, patients with exclusively hepatic dome HCCs showed OS rates at 1, 3, and 5 years from the date of the first SRFA of 87.1%, 76.2%, and 58.2% and 91%, 75.8%, and 55.2% for matched controls. The median OS rates were 66.9 months and 72 months, respectively. Ding et al [20] treated 60 patients with HCC with percutaneous RFA and 56 patients with laparoscopic RFA. The OS rates for the percutaneous RF ablation group were 91.7%, 56.7%, and 36.7 at 1, 3, and 5 years with a median OS of 44 months. The corresponding rates for the laparoscopic RF ablation group were 89.2%, 57.1%, and 44.6 at 1, 3, and 5 years with a median of 57 months, respectively. Kim et al [39] reported high 3- and 5-year OS rates of 84% and 72.7% in HCC patients with 12% of tumors abutting the diaphragm with a mean tumor size of 2.1 cm. Llovet et al [40] showed in a RCT the survival benefit of TACE vs. a control group with significantly higher 1- and 2-year OS of 82% and 63% vs. 63% and 27%, respectively. Especially in larger tumors, several studies reported evidence for the benefit of a combination therapy of TACE and RF ablation; Zhang et al [26] reported 1-, 2-, and 3-year OS rates of 89%, 61%, and 43% after combined RF ablation and TACE in patients with HCC beyond the Milan criteria.

The DFS for patients with exclusively hepatic dome HCC was 78%, 47.5%, and 35.6% at 1, 3, and 5 years and for matched controls 67.2%, 50.8%, and 35.5%, respectively. These data compare with Kim et al [39], who reported similar recurrence-free survival rates after RF ablation of 66.5%, 20.4%, and 17% at 1, 3, and 5 years.