Multidisciplinary management of patients with post-inflammatory pancreatic necrosis

Snthlingm Jegtheeswrn , Joe Gerghty , , , AjithK Siriwrden , ,

a Regional Hepato-Pancreato-Biliary Surgery Unit, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK

b Department of Gastroenterology, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK

c Faculty of Biology, Medicine and Health, University of Manchester, Manchester M13 9PL, UK

Current knowledge of the pathophysiology of acute pancreatitis indicates that pancreatic injury originates at the acinar cell level and then extends through a spectrum of damage ranging from mild peri-acinar inflammatory infiltration and edema to extensive pancreatic parenchymal and peri-pancreatic necrosis [ 1 , 2 ].Clinical acute pancreatitis correlates closely with this range of injury with the majority of patients experiencing mild disease,some having transient organ dysfunction which typically recovers after adequate resuscitation (moderate acute pancreatitis) and a variable minority exhibiting sustained organ failure together with radiological evidence of pancreatic necrosis (severe acute pancreatitis) [ 2 , 3 ]. Worldwide, the management of this latter category of patients with severe acute pancreatitis remains a challenge.There is no effective direct medical treatment and there is no role for early pancreatic debridement [ 3 , 4 ]. This article provides a concise summary of current multidisciplinary management of patients with post-inflammatory pancreatic necrosis.

The mainstays of initial management are the establishment of a correct diagnosis, organ support, symptom control and nutrition [5].

Management of severe acute pancreatitis can be divided into an early phase of care and a later phase which involves treatment of ongoing disease. In the early phase, an important step is confirmation of diagnosis. Computed tomography (CT) although not mandatory in all cases of acute pancreatitis is of value when there is diagnostic uncertainty. In patients with severe disease, CT assessment of pancreatic and peri-pancreatic inflammation is also of value as a prognostic aid [ 6 , 7 ].

In parallel to confirmation of diagnosis, one of the key requirements is adequate fluid resuscitation. There is some evidence that the type of fluid administered may be relevant but more important is adequate (rather than aggressive) fluid resuscitation [ 8 , 9 ]. Once fluid resuscitation is underway, it is critically important to monitor organ systems with a close overview of heart rate, blood pressure,oxygen saturation and urine output. Symptom control (analgesia,antiemetics) is a central component of care. Finally, consideration should be given to the correct setting for treatment which must be in a unit that can provide adequate monitoring and intervention [5].

As these early phase issues are addressed, a proportion of patients will recover. However, the management of those with ongoing illness and pancreatic necrosis remains complex.

In the modern era, the management of patients with severe acute pancreatitis requires a multidisciplinary team (MDT) approach [10]. The core membership should include specialist pancreatic surgeons, interventional radiologists, pancreatologists, nutritionists, critical care physicians and specialist nurses. The MDT should outline a treatment plan which is reviewed and updated at regular intervals. Sharing this information with the patient’s relatives is also important.

In terms of specialist interventions at this stage of the disease, a meta-analysis supported early endoscopic retrograde cholangiography (ERC) in biliary acute pancreatitis, and the authors concluded that treating 26 such patients with ERC + endoscopic sphincterotomy is predicted to save one life [11]. Prophylactic antibiotics are not effective in improving the outcome of severe acute pancreatitis and guidelines do not recommend use [5]. There is evidence that antibiotic over-utilization in acute pancreatitis is widespread globally as differentiation of infection from the systemic inflammatory response can be difficult [12]. A procalcitonin-based algorithm to guide antibiotic use is being evaluated and results are awaited [13]. Antibiotics are appropriate for the treatment of secondary infections and should be guided by antimicrobial sensitivities where possible.

Nutritional support is important. Patients with mild disease do not need to cease oral intake but patients with moderate and severe disease will need supplementation. This can be delivered by nasogastric or nasojejunal feeding tube but some critically ill patients with marked gastrointestinal stasis will require parenteral nutritional support [ 14 , 15 ].

Early fluid collections in the absence of necrosis were termed acute peripancreatic fluid collections in the 2012 update of the Atlanta consensus criteria [2]. Early fluid collections within four weeks of onset of disease in patients with pancreatic necrosis were termed acute necrotic collections [2]. Both of these are part of the pathophysiologic spectrum of severe acute pancreatitis and typi-cally neither requires intervention. Magnetic resonance (MR) scanning is useful in differentiating solid from liquid components of pancreatic necrosis. Thus MR has an important role in diagnosis of post-inflammatory pancreatic necrosis and in directing appropriate therapy [16].

After the initial stage of the disease, the mainstay of care remains supportive but also increasingly seeks to detect evidence of infection in peri-pancreatic necrosis. Fine-needle aspiration of pancreatic necrosis to assess for infection is no longer routinely used in many centers. The reasons for this are multiple: needle aspiration is associated with both false positive and false negative results and may introduce infection, and the major determinant of the requirement for intervention is the clinical status rather than the microbiological profile of necrotic collections.

Perhaps the greatest change in severe acute pancreatitis is in the management of patients with infected necrosis. The diagnosis is typically based on a combination of clinical signs and CT evidence (for example collections containing gas). It is wellestablished that it is better to wait for collections to mature and become walled-offwith a cleardemarcation of liquidcontent before undertaking intervention. Before this time (ideally beyond the fourth week of the episode) management should remain conservative. Historically, pancreatic necrosis with a clinical index of suspicion of infection was managed by open surgical necrosectomy [17].This was a major surgical undertaking and this was reflected in a study which showed deterioration in organ dysfunction scores after open necrosectomy with a procedure-related mortality of 40%being reported after open necrosectomy as recently as 2002 [17].An understanding that the nature of the intervention itself contributed to the deterioration in organ dysfunction led to the evaluation of less traumatic treatments. The PANTER study was a landmark randomized controlled trial comparing a step-up approach to the then-standard option of open necrosectomy [18]. The step-up approach utilized percutaneous or endoscopic drainage followed by retroperitoneal necrosectomy. This study established that a third of patients allocated to the step-up approach could be effectively managed by percutaneous catheter drainage alone. Second, this study showed that although episode-related mortality was not different between arms, the endoscopic group had a significantly lower incidence of new-onset organ failure, diabetes mellitus and late incisional hernia. Given the good outcomes seen with the step-up approach it was logical to compare endoscopic step-up to the surgical step-up approach [19]. This trial, undertaken by the Dutch Pancreatitis Study Group screened 418 patients with post-inflammatory pancreatic necrosis between 2011 and 2015 with patients being allocated to either endoscopic step-up [endoscopic ultrasound (EUS)-guided transluminal drainage followed by endoscopic necrosectomy]or surgical step-up (percutaneous catheter drainage followed by videoscopically-assisted retroperitoneal necrosectomy). The results showed equivalence in terms of major complications and mortality but a lower rate of pancreatic fistula and shorter in-patient stay in the endoscopic group. These important studies led to the widespread acceptance of the minimally invasive approach to the treatment of pancreatic necrosis[ 18 , 19 ].

Endoscopic drainage initially involves leaving pigtail stents in the cavity to permit continued drainage. This is very effective for the drainage of non-necrotic collections (pseudocysts) but less so for managing infected necrosis [20]. This is because of stent occlusion with solid necrosis and secondary infection. A large retrospective study of 211 patients reported success rate of 94% for pseudocyst drainage compared to 63% for necrosis [21]. In order to improve debridement, staged endoscopic drainage can be undertaken after dilatation of the track between the stomach and the necrotic cavity. This technique is termed direct endoscopic necrosectomy [22].

Continued refinement of endoscopic drainage equipment leads to the development of lumen apposing stents that allows improved drainage and crucially also permits repeated endoscopic access for continued debridement [23]. These double flanged stents, shaped as a dumbbell, oppose the visceral and cyst walls and thus reduce the chance of migration and leakage. Long-term outcome data are awaited. Direct comparison of efficacy of lumen apposing and plastic stents has shown mixed results. A meta-analysis reported better resolution of walled-offnecrosis with lumen apposing stents [24]. However a recent randomized trial in 60 patients comparing lumen apposing stents to plastic stents (up to three were placed) showed a higher incidence of procedure-related complications and greater cost with the newer technique [25]. The protocol was modified during the study to recommend repeat CT at 3 weeks after placement of metal endoprosthesis followed by stent removal if the collection had been resolved [25]. In overview, lumen opposing stents may provide better access to necrosis and specifically, facilitate repeat procedures. It is unclear whether this is achieved at the cost of a greater incidence of procedure-related complications such as hemorrhage. Retention of metal stents after successful drainage seems unnecessary and may be associated with erosion-related complications.

In cases where endoscopic access is not possible (if the distance from the visceral wall to cavity is＞15 mm or there are interposed vessels) then percutaneous drainage should be performed. A combined approach using both endoscopic and percutaneous drainage can be very effective, especially when there are deep pelvic collections to access.

Endoscopic techniques continue to develop and in the multiple transluminal gateway technique two or three transmural tracts are created by using EUS guidance between the necrotic cavity and the gastrointestinal lumen. While one tract is used to flush saline via a nasocystic catheter, multiple stents are deployed in other tracts to facilitate drainage of necrotic contents with promising early results being reported in a small series [26]. With newer techniques such as this, more experience is required to assess their safety, utility and role.

In the modern era, the role for direct, primary surgical drainage of infected necrosis has changed. Surgery is rarely the first choice intervention but retains an important role in patients who are critically ill, have failed endoscopic therapy or have co-existent complications [27]. In terms of surgical approach, video-assisted retroperitoneal debridement using a modified nephroscope was an early “minimally” invasive approach and was evaluated in the classic PANTER study [18]. Other minimally invasive approaches such as laparoscopic necrosectomy have also been reported and minimally invasive operative approaches to the debridement of acute necrotizing pancreatitis are preferred to open surgical necrosectomy where possible [10].

In addition to intervention for infected necrosis, drainage can also be required for sterile necrosis in patients with persistent failure to thrive or ongoing mass effect from necrosis causing biliary or gastric obstruction [28].

After treatment of the index episode and management of necrosis, there may yet be a requirement for treatment of later complications such as disconnected duct syndrome, sinistral portal hypertension from splenic vein occlusion and pseudoaneurysm of the splenic artery. Patients treated for severe acute pancreatitis thus require careful clinical and nutritional follow-up after recovery.

In conclusion, this article provides a succinct overview of the current evidence-based management of patients with pancreatic necrosis complicating severe acute pancreatitis. The mainstays of care are early confirmation of diagnosis, organ system support, nutrition and avoidance of inappropriate antibiotics in the early phase of the disease. For those patients with persistent illness, debridement of necrosis becomes an important component of care. This should be carefully managed to avoid premature intervention before the fourth week of the episode. The demonstration that major surgical interventions resulted in an increase in post-procedure organ dysfunction led to the development of less traumatic procedures. Endoscopic intervention utilizing stents and intra-cavitary drainage and debridement has become the mainstay of care for drainage of walled-offpancreatic necrosis. Multidisciplinary care is essential for the successful management of this challenging disease.

Looking to the future, there remains scope for development of specific therapies for treatment of severe acute pancreatitis, optimal management of necrosis in terms of timing and technique can be further refined and as in other areas of medicine, more consideration must be given to patient-reported outcome measures.

Acknowledgments

None.

CRediT authorship contribution statement

Santhalingam Jegatheeswaran:Data curation, Methodology,Writing - original draft, Writing - review & editing.Joe Geraghty:Data curation, Methodology, Validation, Visualization, Writing -original draft, Writing - review & editing.Ajith K Siriwardena:Conceptualization, Methodology, Project administration, Supervision, Visualization, Writing - original draft, Writing - review & editing.

Funding

None.

Ethical approval

Not needed.

Competing interest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.