Immediate Endoscopic Necrosectomy for Walled‐Off Pancreatic Necrosis: A First‐Choice Option?

Walled-off necrosis (WON) of the pancreas develops as a local complication of acute necrotizing pancreatitis. When symptomatic WON does not respond to conservative management, patients are generally referred to drainage-based interventions. With the increasing popularity and availability of endoscopic ultrasound (EUS)-guided transluminal procedures, endoscopic treatment has become the preferred first-line treatment for symptomatic WON at many centers [1]. Prior to the clinical implementation of lumen-apposing metal stents (LAMSs), direct endoscopic necrosectomy was reserved for patients who were not amenable to drainage alone (termed “step-up approach”). The step-up approach is expected to avoid unnecessary necrosectomy sessions and associated adverse events. LAMSs provide a stable transmural port that allows for safe and effective necrosectomy following EUS-guided drainage [2]. Their lumen-apposing design is particularly advantageous in cases of WON located somewhat apart from the gastrointestinal tract. Consequently, in the era of LAMSs, the optimal timing of necrosectomy has become a major topic of debate [3]. Intuitively, performing necrosectomy during or shortly after the initial drainage (termed “immediate necrosectomy”) may expedite WON resolution and improve clinical outcomes. However, there have been long-standing concerns about the elevated risk of procedure-related adverse events associated with this aggressive approach (e.g., bleeding and peritonitis due to perforation of the WON wall). In this issue of Digestive Endoscopy, Yuen and colleagues reported a meta-analysis of 15 cohort studies involving a total of 1290 patients, which aimed to compare immediate necrosectomy and the step-up approach in terms of clinical outcomes [4]. As only one study compared the clinical outcomes of both treatment strategies within the same cohort, the rates of a given outcome were pooled separately for each group. Overall, no statistically significant differences were noted between the groups: the pooled rates or means (95% confidence intervals) for the immediate necrosectomy and step-up approach groups were 90% (83%–94%) vs. 94% (91%–96%), respectively, for clinical success; 16% (10%–24%) vs. 16% (10%–24%), respectively, for adverse events; 0.99 (0.48–1.51) vs. 1.48 (0.80–2.17), respectively, for the number of reinterventions; and 1.24 (0.57–1.92) vs. 0.93 (0.45–1.39), respectively, for the number of additional necrosectomy sessions. These findings support the comparable clinical outcomes between the treatment strategies. In contrast with the abovementioned concerns, immediate necrosectomy appears to be associated with neither an increased risk of adverse events nor a higher number of necrosectomy sessions. The authors concluded that immediate necrosectomy could be considered a viable treatment option, demonstrating safety and effectiveness profiles similar to those of the step-up approach. These somewhat unexpected results might be attributed to the clinical implementation of LAMSs for more feasible necrosectomy and the maturation of our procedures of EUS-guided drainage and adjunctive treatment. If immediate necrosectomy and the step-up approach demonstrate comparable safety and effectiveness, which strategy should be prioritized as the first-choice option? Accumulating evidence supports the additional clinical benefits of the immediate necrosectomy-driven treatment algorithm. In the current meta-analysis, the lack of data inhibited an analysis of the time to clinical treatment success (WON resolution)—an important yet largely uninvestigated clinical endpoint in which immediate necrosectomy may be advantageous by facilitating earlier resolution of WON. In a multi-institutional cohort of 227 patients with WON within the WONDERFUL consortium [5], the early implementation of endoscopic necrosectomy was associated with a shorter time to clinical treatment success compared to the drainage-oriented approach, with medians of 68 and 92 days, respectively. A prolonged treatment period predisposes patients to the risk of complications due to persistent pancreatic and peripancreatic inflammation. The risk of a fistula of the gastrointestinal tract may be elevated [6]. Prolonged placement of an LAMS may provoke the formation of a pseudoaneurysm, which is a potentially lethal adverse event in this setting [2]. In addition, the risk of sarcopenia—characterized by reduced skeletal muscle mass and function—may increase, potentially leading to adverse clinical outcomes [7]. Therefore, immediate necrosectomy may be beneficial by shortening the time to WON resolution while simultaneously reducing complications. While the current meta-analysis reports comparable (but nonsuperior) outcomes, emerging randomized controlled trials (RCTs) have demonstrated the superiority of immediate necrosectomy. In the multicenter DESTIN trial involving 70 patients [8], immediate necrosectomy significantly reduced the number of reinterventions compared to the step-up approach, with medians (25th to 75th percentiles) of 1 (0–1) and 2 (1–4), respectively. Of note, there was no significant between-group difference in mortality or procedure-related adverse events. In a single-center ACCELERATE trial that was published more recently and thus not included in the current analysis [9], a composite endpoint of death, major complications, or a prolonged hospitalization (> 58 days) was investigated in patients with WON > 15 cm. The trial reported an absolute risk reduction of 53.5%, and this dramatic difference prompted early termination after the enrollment of 25 patients. In our multicenter WONDER-01 trial involving 70 patients [10], immediate necrosectomy significantly shortened the treatment period compared to the drainage-oriented step-up approach: the median times to clinical success were 29 days (95% CI 19–34 days) and 44 days (95% CI 38–52 days), respectively. Again, the risk of mortality or procedure-related adverse events did not differ significantly between the groups. Given these RCT data with the noninferiority demonstrated by the large sample size in the current analysis, immediate necrosectomy may be considered a preferable first-line option. Several considerations should be addressed before immediate endoscopic necrosectomy can be incorporated as a standard of care in treatment protocols. First, although the current meta-analysis provides evidence with the highest possible quality based on a systematic review of the literature, its findings warrant validation due to the limited availability of data and resultant methodological limitations. A head-to-head comparison between the treatment strategies could not be performed, precluding the calculation of risk ratios. Given the wide variations in treatment indications and protocols for WON across centers, this limitation is likely to be substantial. Namely, the outcomes of immediate necrosectomy in certain cohorts may not be compared directly with those of the step-up approach in independent cohorts. Indeed, the inter-institutional heterogeneity was evident in the current analysis, as the exclusion of a single study in a sensitivity analysis shifted the outcomes toward statistical significance for clinical success and the number of reinterventions. Furthermore, publication bias could not be assessed for the comparative risk estimates. Taken together, these limitations may account for the discrepancies between the findings of the current meta-analysis and those of the recent RCTs. Therefore, a meta-analysis based on direct head-to-head comparisons is warranted when evidence accumulates. Second, immediate necrosectomy requires substantial procedural expertise, which currently limits its generalizability. Performing necrosectomy simultaneously during the EUS-guided drainage can be technically difficult and cause LAMS migration when the stent does not fully expand. Procedural techniques and dedicated devices should be investigated to facilitate immediate necrosectomy. Third, endoscopic necrosectomy is a resource-intensive procedure that involves various specialists, resulting in high medical costs. On the other hand, if immediate necrosectomy shortened the time to clinical success and patients' discharge, it would have the potential to improve cost-effectiveness in the management of patients with WON. Therefore, the treatment strategies should be compared from a health economics perspective, considering broad healthcare costs (including personnel costs). Fourth, future research should investigate the long-term outcomes with a focus on pancreatic endocrine and exocrine functions and patients' quality of life. Immediate necrosectomy may resolve pancreatic inflammation more rapidly and thereby help maintain those functions. Last but not least, given the variations in treatment indications across studies, the criteria for selecting immediate necrosectomy over the step-up approach need to be clearly defined. Further research is needed to identify the subgroup of patients who are more likely to benefit from immediate necrosectomy until this approach can be recommended as a universally applicable first-line strategy. The RCTs with positive findings adopted various cut-off points of WON size for inclusion [8-10], and the optimal size threshold for recommending immediate necrosectomy has yet to be established. In conclusion, the current meta-analysis, reinforced by emerging evidence from recent RCTs, suggests that immediate necrosectomy may be noninferior—and possibly superior—to the step-up approach as a first-line treatment for symptomatic WON. If validated, these findings could make a paradigm shift in the endoscopic management of patients presenting with this challenging condition. All authors: conceptualization, design, and approval of the final version of the manuscript. Y.N.: obtained funding. T.H.: literature search and drafting of the manuscript. T.F. and Y.N.: editing and critical revision of the manuscript for important intellectual contents. T.H. and Y.N.: supervision. Y.N. receives research funding and honoraria from Boston Scientific Japan. The other authors declare that they have no conflicts of interest. This article is linked with “Comparison of direct endoscopic necrosectomy with endoscopic step-up approach after endoscopic drainage of pancreatic walled-off necrosis: systematic review and Meta-analysis,” http://doi.org/10.1111/den.70021.