Su1250 BURN TO ASHES- A PURIFICATION CEREMONY. A SYSTEMATIC REVIEW AND META-ANALYSIS OF ENDOSCOPIC ULTRASOUND- GUIDED RADIOFREQUENCY ABLATION OF PANCREATIC LESIONS

Abstract

Radiofrequency ablation (RFA) has been used to treat various abdominal tumors including pancreatic tumors. Multiple approaches such as laparoscopic, open, and percutaneous have been used for pancreatic tissue ablation. More recently, endoscopic ultrasound (EUS)-guided RFA has emerged as a new technique for pancreatic tissue ablation. The role of EUS-RFA in management of pancreatic lesions is still not well-established. In this meta-analysis, our aim is to assess efficacy and safety of EUS-RFA for management of pancreatic lesions. MEDLINE, Scopus, and Cochrane Library databases were searched to identify studies reporting EUS-RFA of pancreatic lesions with outcomes of interest. Studies with <5 patients were excluded. Clinical success was defined as symptom resolution, decrease in tumor size, and/or evidence of necrosis on radiologic imaging. Efficacy was assessed by the pooled clinical success rate whereas safety was assessed by the pooled adverse events rate. Heterogeneity was assessed using I2. Pooled estimates and the 95% confidence interval (CI) were calculated using random-effect model. Nine studies (4 retrospective and 5 prospective) involving 106 patients with 116 pancreatic lesions were included. The mean age was 58.9 years and 44% of patients were male. 126 RFA procedures were performed with an average of 2.3 RFA applications per patient. The mean lesion size was 20.1 mm. The lesions comprised of 41 (35.3%) non-functional neuroendocrine tumors (NFNETs), 18 (15.5%) were insulinomas, 25 (21.6%) were pancreatic cystic neoplasms (PCNs), and 32 (27.6%) were pancreatic adenocarcinomas. The majority were present in the pancreatic head (40.8%), with 36.7% in the body, 12.2% in the tail, and 10.2% in the uncinate process. Pooled overall clinical success was 92.7% (95% CI: 0.88-0.97, I2= 1.5%) (Figure 1). The symptoms were resolved in all insulinoma cases. Pooled overall adverse events rate was 18% (95% CI: 0.09-0.28, I2= 57.4%) (Figure 2). The most common complication was abdominal pain (12.6%), followed by acute pancreatitis (2.4%). Two cases of pancreatic duct stenosis, peripancreatic fluid collection, and ascites were reported. Only one patient required surgical intervention after EUS-RFA. This study demonstrates that EUS-RFA is an effective treatment modality for pancreatic lesions, especially functional neuroendocrine tumors such as insulinomas. Although the rate of adverse events was slightly higher than other methods (18%), the most common complication was abdominal pain which was usually self-limited and managed symptomatically. Therefore, EUS-RFA appears to be a safe and effective procedure for pancreatic lesions. The relatively small study size and inclusion of non-controlled studies are among the limitations of this study.