Previous studies have reported that a history of abdominal surgery was the main reason to technical failure of artificial ascites because postoperative lesions were present in up to 93% of the patients [10, 11, 17, 18]. The difficulty of separating adhesions can increase the volume of the artificial ascites and the range of liver movement. Any movement of the omentum and gastrointestinal tract in relation to the index tumor can increase the difficulty of percutaneous catheterization. Percutaneous balloon catheters have been used to separate the index tumor from gastrointestinal tract, but they can be difficult to position, especially in in patients with tumors distant from body surface [19, 20]. This method also requires special instruments that complicate its adoption. Laparoscopy can assist ablation, but it is more costly and invasive than artificial ascites and postoperative adhesions may be more likely[21]. A small dose of ethanol injected into the marginal tissue of the tumor can reduce thermal damage of the gastrointestinal tract but it is not a curative treatment 24, 25]. Artificial ascites is a relatively safe and effective option for tumors abutting the gastrointestinal tract. Further study is needed to confirm the usefulness and clinical benefit of this procedure to assist in local ablation of hepatic tumors in patients with previous abdominal surgery.
CT can clearly show the anatomical relationship between the lesion and the surrounding structure. It is believed by some clinicians that treatment under the CT control is safer. However, we used ‘ablation, Gastrointestinal, liver’ as keywords and searched on Pubmed, and we did not find any literatures clearly stated that hepatic lesions must be ablated under CT control, or ultrasound-guided ablation will increase the incidence of complications. Besides, There are two literatures compared the application of CEUS/US and CT guided RFA for hepatocellular carcinoma, the results showed that there was no comparative difference in the incidence of complications between the two groups[22, 23]. Most articles using artificial ascites to assist ablation of liver cancers adjacent to gastrointestinal tract were conducted under the guidance of US, because it has the advantage of real-time monitoring, which is conducive to the formation of artificial ascites, and can real-time and accurately assess the depth and the position of needle[10, 12–17, 24]. Thus, we consider that ultrasound-guided artificial ascites assisted thermal ablation is an option for liver cancer tumors adjacent to gastrointestinal tract for in patients with previous abdominal surgery.
Artificial ascites was successfully implemented in 95%, of the cases in this study, which is consistent with previous reports of from 78%-92.7% success [10, 12–17]. The success rate of artificial ascites implementation was improved by some modifications implemented before and during this study. The puncture site for instilling the artificial ascites was near the tumor, and the catheter was located between the index tumor and adjacent gastrointestinal tract if possible to ensure the accumulation of liquid in the space between them. An intravenous catheter was used because it is soft and can maintain the saline infusion and its position during ablation without real-time monitoring. If artificial ascites did not accumulate between the tumor and adjacent gastrointestinal tract an 18 G PTC needle was placed between them and liquid as perfused to form a local thermal barrier. Direct injection of US contrast agent through the needle or catheter facilitates confirmation of correct needle or catheter position and the cavity morphology [25]. That allowed use of intracavitary CEUS to evaluate the distribution of artificial ascites. If liquid was present between the tumor and adjacent gastrointestinal tract, then perfusion could continue to remove thermal energy from the site. The study results confirm that these modifications in the use of artificial ascites were successful in separating the index tumor and gastrointestinal tract in more than 90% of the patients. Artificial ascites was a feasible technique to assist thermal ablation for liver cancer adjacent to the gastrointestinal tract in patients with previous abdominal surgery.
Mesothelial damage caused by inflammation or surgical trauma can trigger the formation of postoperative adhesions during peritoneal wound healing[21, 26]. Adhesions likely to form between an incision and the adjacent gastrointestinal tract. In this study there were 29 patients with a history of hepatectomy, but there were no significant differences in the successful implementation of artificial ascites in those with the index tumor on the same side or different side as the hepatectomy. The result shows that, in either situation, the adhesion could be broken and did not influence the distribution of the artificial ascites.
Major complications were associated with 5.1% of the study procedures, which is higher than reported in previous studies [10, 12–15, 17, 27]. One of the two complication cases experienced an intestinal fistula accompanied by infection that developed in a failed implementation of artificial ascites. The tumor had a maximum diameter > 3 cm. The large ablative zone might have increased the risk of gastrointestinal tract injury because the artificial ascites failed to separate the liver from the gastrointestinal tract. However, it was difficult to distinguish the intestinal fistula caused by infection or ablation. We considered that avoiding infection may be help reduce complications. Besides, when the artificial ascites cannot be performed,, accurate calculation of the ablative zone may help complete the ablation, but the process may be restricted. In such cases, laparoscopy, laparotomy, or ethanol ablation might be helpful.
Some studies considered that saline can conduct electrical current due to its ionic composition, which may lead to non-target tissue heating while using RFA. However, based on our previous experience [28, 29], we did not have any complications associated with the choice of normal saline as artificial ascites. In addition, we did not find any articles about the complications of using normal saline as artificial ascites in human beings. We think it may be related to the following factors. Firstly, the RFA does not heat the normal saline directly, thus may less likely to conduct electrical current. Secondly, most cases of artificial ascites in our study were floating, which may reduce the local energy accumulation. Besides, some patients in our study have diabetes and were not suitable for D5W. Artificial ascites has been reported to increase the risk of intraperitoneal bleeding and tumor seeding because it washes coagulation substances away from the puncture site and decreases the pressure of the abdominal wall against the liver, facilitating the dissemination of tumor cells[10]. Bleeding and tumor seeding were not observed in this study. Cauterizing the needle track when withdrawing the electrode or antenna may help to prevent these complications[30]. As in previous reports, residual ascites disappeared spontaneously without additional diuretics or paracentesis.
All tumors in this study were ablated completely. The technical efficacy rate was 100%, indicating that artificial ascites was effective in to assisting thermal ablation. The 1-, 2-, and 3-year LTP rates were 2.9%, 5.7% and 5.7%, and the 1-, 2-, and 3-year OS rates were 97.1%, 86.8% and 69.5%. Both were in line with previous studies and the experience at the study center, and support the use of artificial ascites can help achieve an effective therapeutic effect.
The study limitations include its single-arm retrospective design. A controlled trial should be designed and conducted; Secondly, the sample size was small. Thirdly, as the median follow-up was 16 months, additional monitoring is needed to better support the therapeutic benefit. Consequently, further research is needed to validate the clinical value of artificial ascites implementation in patients with a history of abdominal surgery.