Therapeutic pancreatic endoscopy

1. Introduction 

Endoscopic therapy for pancreatic disorders has, for the most part, focused on the management of chronic pancreatitis. Indeed, stricture dilation and extraction of pancreatic duct stones have been easy targets for endoscopic therapy as it represents a significantly less invasive approach to the alternative surgical options. Owing in large part to the refinement of techniques for endolumenal therapeutics and therapeutic endoscopic ultrasound, the past decade has seen the emergence of additional indications for therapeutic pancreatic endoscopy, including the management of severe narcotising pancreatitis. This review will highlight many of the areas, both old and new, in which therapeutic endoscopists can play a critical role in the management of pancreatic disorders.

2. Endoscopic pancreatic sphincterotomy 

Endoscopic pancreatic duct (PD) sphincterotomy (ES) is utilized in a variety of settings including the treatment of relapsing pancreatitis associated with sphincter of Oddi dysfunction (SOD) and pancreas divisum [1]. In both conditions, ES is performed, either to the major or minor papilla, as definitive therapy once other etiologies of acute pancreatitis have been excluded [2]. The most common indication for pancreatic sphincterotomy is not for therapy, however, but rather as a means to providing access to the pancreatic duct in order to facilitate stone fragmentation and removal, treatment of an obstructing stenosis, or to approach a ductal disruption and its consequences.

Endoscopic sphincterotomy of the pancreatic duct can be performed through a variety of techniques; however, most endoscopists perform a pull-type sphincterotomy. Although an initial biliary sphincterotomy will often delineate the pancreaticobiliary septum and help define the length of the incision [3], it is not imperative. PD ES without a biliary ES is performed by angling the direction of the cut to the one o’clock position and taking care not to cut into the pancreatic parenchyma. PD ES undertaken with a pull sphincterotome is almost invariably undertaken over a slick guidewire within the PD and using pure cut current to minimize damage to the pancreatic parenchyma or subsequent sphincterotomy stenosis. An alternative to the use of a pull-type sphincterotome for the performance of PD ES involves initial placement of a small (3 or 5Fr) PD endoprosthesis followed by needle-knife sphincterotomy, taking care not to cut inferior to the level of the stent [2], [3], [4]. Stent placement is often used for PD ES in chronic pancreatitis following procedure completion, however, to both treat an underlying leak or stricture and to prevent edematous closure of the sphincterotomy following multiple endoscopic manipulations [5], [6].

Although PD ES can be associated with an acute pancreatitis flare in up to 10% of patients in whom the procedure is performed for SOD, multiple studies suggest that ES is associated with a much lower risk in chronic pancreatitis [1], [2], [4]. Additional immediate complications of PD ES include acute procedural bleeding, local perforation and infection. Late complications include stenosis of the sphincterotomy and ductal stricturing related to parenchymal cautery burn or incision.

3. Pancreatic stricture treatment 

Stricture formation within the pancreatic duct may be the end-result of several different inflammatory processes. Previous severe acute pancreatitis complicated by ductal disruption, relapsing acute pancreatitis or chronic calcific pancreatitis may all be complicated by stricture formation. In the case of chronic pancreatitis, strictures in the PD may be associated with an upstream calculus. Although more than one PD strictures may be present at any point along the PD, their clinical significance can be highly variable, even asymptomatic.

The clinical work-up of an identified stricture of the PD initially revolves around an attempt to determine whether the lesion is benign or malignant. Techniques including directed brush cytology, fluoroscopic-assisted endoscopic biopsy, and even direct pancreatoscopy may all be employed for this purpose. The use of serum tumour markers, high-resolution pancreas-protocol CT scanning as well as magnetic resonance imaging (MRI/MRCP) may also be helpful in this regard. In cases where an abrupt transition point is noted within the pancreatic duct either on pancreatography or cross-sectional imaging, endoscopic ultrasound (EUS) may be helpful in ruling out a small mass lesion. Once malignancy has been excluded, a determination as to whether or not a stricture plays a role in obstructive pancreatitis and chronic pancreatic pain is required. Typically, symptomatic strictures are characterized by upstream ductal dilation, an impacted stone at the stenosis site, or a ductal disruption, pseudocyst or pancreatic fluid collection proximal to the stricture [2], [7], [8]. It is often the case, however, that determining the clinical significance of a PD stricture can only be made through a “therapeutic trial” of dilation and stent placement.

Most strictures are initially approached using a wire guide placed into the pancreatic duct following a pancreatic sphincterotomy. Stenoses can be dilated using a variety of methods including graduated dilating catheters and controlled radial expansion polyethylene balloons (e.g., Quantum™ balloon, Wilson-Cook, Inc., Winston-Salem, NC; MaxiForce™, Boston Scientific, Natick, MA) ranging from 4 to 10mm in diameter and 2–4cm in length (Fig. 1). The latter are utilized more frequently and the size of balloon selected should approximate that of the downstream PD. The use of a balloon which is too large for a given pancreatic duct can lead to rupture of the more normal portion of the PD. Although dilation is performed to waist effacement, there are no data to suggest that balloon inflation (and subsequent waist effacement) for longer periods of time (e.g. 1–3min) leads to superior clinical outcomes as compared to inflation for 10–15s.

View Large Image Download to PowerPoint

Fig. 1. CT (A) demonstrating a large pancreatic duct stone in the head of the pancreas (arrow). Pancreatogram obtained through the minor papilla (B) demonstrates “pseudodivisum” anatomy with the ventral duct occluded by the large stone. A severe stenosis is present in the dorsal pancreatic duct which is subsequently dilated (C) using a hydrostatic balloon dilator. Two stents are placed into the dorsal pancreatic duct following ESWL and stone extraction (D).

In strictures which are acutely angulated or impassable by graduated dilating catheters or balloon dilators, the use of a Soehendra Stent Extractor™ (Wilson-Cook, Inc., Winston-Salem, NC) may be initially required to “drill through” the stricture prior to dilation. This is often the case with large PD calculi. Following dilation, most endoscopists attempt to place a 5–10Fr polyethylene stent approximating the downstream diameter of the duct. In patients who are symptomatically improved, prostheses are retrieved at 2–3 months followed by re-dilation and upsizing of the stent, if possible. This process is repeated several times over a year's period, and if there is no stricture resolution and the patient becomes stent dependent, good risk surgical patients should be considered for surgical referral.

It has been our practice to place multiple, parallel 5–7Fr prostheses into the PD following stricture dilation (Fig. 1). Although there are no clinical data to suggest that this is superior to a single larger stent, anecdotal experience suggests that this approach allows continued drainage between stents at time of inevitable stent occlusion. In turn, this likely reduces the risk of both obstructive pancreatic infections and recurrent obstructive pancreatitis in the setting of stent occlusion. The recent introduction of fully covered self-expanding metal stents (SEMS) into clinical endoscopic practice has raised the possibility of using these larger caliber devices as a removable endoprosthesis for benign strictures of the pancreaticobiliary tree. A small case series [9] has found recurrent stricture formation in 6 of 7 patients who had fully covered SEMS placed and subsequently removed for treatment of benign pancreatic duct strictures. Additional study is clearly required prior to advocating the routine use of these significantly more costly devices for benign pancreatic duct strictures.

Several published series now exist which demonstrate a 60–80% reduction in attacks of relapsing pancreatitis following endotherapy for PD strictures [7], [10], [11], [12], [13], [14], [15]. A decrease in or resolution of chronic pain has also been suggested, however, this is qualified by the fact that few studies have used true objective measurements such as quality of life indices, reproducible pain scales, or pre- and post-treatment analgesic requirements. PD stenting is not without its own risk of iatrogenic complications. Side branch occlusion, parenchymal atrophy and fibrosis are all possibilities when stents are placed into relatively normal ducts [16], [17]. Moreover, inflammation or “ductitis” induced by a stent side flap or pressure from the internal stent tip can lead to ductal stenoses. Pancreatic stents are associated with bacterial colonization of the duct [18]. Not only can stent occlusion therefore result in recurrence of pancreatic symptoms and duct blow-out with formation of pseudocysts, but also with infectious complications such as abscess formation and pancreatic sepsis.

The existing literature shows at best a 50% rate of resolution when PD strictures are endoscopically treated for 1 year with most series suggesting rates closer to 20–30% [7], [10], [11], [12], [13], [14], [15]. It is important to recognize, however, that the presence of a stricture does not imply symptoms and, indeed, 60–80% of patients become asymptomatic following endoscopic therapy, even in the presence of a residual stenosis. Why this occurs remains unclear, however, an increased flow of pancreatic juice, accelerated chronic pancreatitis in side branch parenchyma with resultant diminution of pain from capsular distention, and resolution of small upstream stenoses have all been hypothesized.

4. Endoscopic ultrasound-guided rendezvous 

In some patients, the pancreatic duct cannot be accessed through either the major or minor papilla due to a severe stenosis or calculus in the head of the pancreas. In cases where the main pancreatic duct is dilated upstream from the stenosis, transpapillary access can often be obtained through the use of an endoscopic ultrasound-guided “rendezvous” procedure [19], [20]. A 19-guage needle can be passed under EUS control into the dilated pancreatic duct through which a slick 0.025in. or 0.035in. guidewire can be passed in an antegrade fashion through the stricture into the duodenum. After exchanging for a duodenoscope, cannulation can be achieved alongside the guidewire or the wire can be grasped using a snare or forceps, brought out through the accessory channel of the duodenoscope over which therapeutic accessories can be subsequently be passed. This technique can also be used in patients with altered anatomy who require pancreatic intervention. One such group of patients are those who have undergone pancreaticoduodenectomy (Whipple procedure) in whom an anastamotic stricture has developed at the pancreaticojejunostomy.

In cases where transpapillary access cannot be achieved in an antegrade fashion using the rendezvous technique, trans-gastric drainage of the pancreatic duct has been reported [19] (Fig. 2). A guidewire is left coiled within the dilated pancreatic duct over which the tract is dilated either using a needle knife, cystotome, graduated dilating catheter or Sohendra stent extractor. A pigtail or straight stent can then be passed over the guidewire into the dilated pancreatic duct to allow for drainage into the stomach [19].

View Large Image Download to PowerPoint

Fig. 2. CT scan (A) demonstrating a dilated pancreatic duct secondary to obstruction from an ampullary carcinoma and metal bile duct stent. A trans-gastric pancreatogram is obtained using endoscopic ultrasound (B) subsequent to which a guidewire is placed into the pancreatic duct (C). Transpapillary access could not be obtained using the guidewire and a trans-gastric stent was subsequently placed to decompress the pancreatic duct (D).

5. Endoscopic treatment of pancreatic duct stones 

As opposed to their biliary counterparts, PD calculi represent a more vexing clinical problem. There are several factors which contribute to this, including a significantly harder composition than bile duct stones in addition to commonly being located upstream from a PD stricture. Indeed, less than one half of patients with PD calculi have stones which are amenable to endoscopic extraction alone; the other 50% require some form of lithotripsy to facilitate removal. Finally, the PD upstream from impacted calculi can be markedly dilated; the resultant intraductal hypertension can lead to ductal disruption and the development of pancreatic pseudocysts, ascites and pleural effusions.

Like other complications arising from chronic pancreatitis, the management of PD calculi begins with obtaining high-quality cross-sectional imaging in the form of CT or MRI/MRCP. This allows for pre-procedural planning, estimation of stone size as well as the identification of sequelae of ductal disruption. Smaller stones can be removed by standard extraction techniques utilizing an extraction balloon or basket passed alongside or over a hydrophilic wire guide (Fig. 3). Dilation of ductal strictures – using a graduated dilating catheter or hydrostatic balloon – is required when stones are impacted upstream from a stenosis. In the case of severe stenosis which will not allow passage of a catheter or balloon dilator, the stricture and stone can be “drilled” through using a Sohendra stent extractor (Cook Endoscopy, Winston-Salem, NC). In the majority of cases, endoscopic pancreatic sphincterotomy is performed to facilitate stone extraction. In some cases of obstructive chronic pancreatitis, a “pseudodivisum” may be present in which the ventral pancreatic duct is obstructed by a large stone or severe stricture (Fig. 1). In such cases, access and drainage must typically be achieved through the minor papilla.

View Large Image Download to PowerPoint

Fig. 3. A large pseudocyst is seen which invades into the left lobe of the liver (A). At ERCP, a leak from the main pancreatic duct is identified (B) and stented (C). CT scan obtained several weeks later demonstrates complete resolution of the pseudocyst after transpapillary stent placement and endoscopic cystgastrostomy (D).

As mentioned earlier, 50% of patients will require some form of lithotripsy to facilitate stone extraction. Mechanical lithotripsy can be difficult with pancreatic duct stones, especially in the presence of a PD stricture as the basket must be passed upstream from the calculus to allow for full deployment and stone capture. Probe-based lithotripsy – either electrohydraulic or laser – can be performed at the time of pancreatoscopy. The fragility of the instruments required as well as the need to pass a pancreatoscope into an often strictured PD to access the stone make this a more difficult, if not less attractive, option.

Owing to the difficulties associated with both mechanical and probe-based lithotripsy, the majority of patients with large PD stones initially undergo ESWL for fragmentation prior to extraction. Stone localization is required prior to ESWL, although some calculi will be evident on plain abdominal film. For those that are not, placement of a pancreatic duct stent or nasopancreatic drain can be performed, the latter having the advantage of allowing for a pancreatogram to be performed to ensure fragmentation following treatment. It is our practice to perform ERCP immediately following ESWL at which time a PD sphincterotomy and stone extraction are performed (Fig. 1). Saline lavage at the time of post-ESWL ERCP and nasopancreatic drain placement at the time of ERP both seem to facilitate stone fragment passage and may preclude the need for multiple ERCPs. Regardless of whether ESWL was performed, placement of one or more endoprostheses into the pancreatic duct is performed at the time of ERCP to allow for passage of additional stone fragments and allow for ductal decompression and prevention of pancreatitis secondary to edema from the performance of pancreatic ES.

Although there are data to suggest that endoscopic stone extraction leads to clinical improvement [21], [22], not all patients with chronic calcific pancreatitis should undergo attempted stone extraction. Poor candidates include those with a major burden of small stones within the pancreatic head (pseudotumour), patients without a dilated pancreatic duct upstream from the calculus and patients with ductal calculi within the setting of a disconnected duct. Such individuals should be considered for surgery [8] or managed medically.

Most recent series suggest that approximately two-thirds of patients who have successful stone removal from the MPD will have a significant improvement in chronic pain and a decrease or elimination of attacks of relapsing pancreatitis [21], [22]. Our group reported the results of ESWL pre- or post-ERCP in 40 patients with obstructing PD stones over a 5-year period [23]. Patients required an average of just over two ERCP to completely clear the PD of stones. Complications occurred in 20% of the patients; always minor, to include GI bleeding, exacerbation of pancreatitis, and bacteraemia. At a mean follow-up approximating 2½ years, 80% of patients avoided surgery and there was a statistically significant decrease in pancreatitis-related hospitalisation, narcotic use and improvement in an analog pain scale. Despite this success and that of others [21], [22], the requirement for careful patient selection cannot be emphasized enough. Those that are poor candidates for endoscopic therapy (see above) and those in who repeated ERCP is required over a number of years to manage complications should be strongly considered for surgery. Even in an era of effective endoscopic therapy for chronic pancreatitis and its associated complications, there are reasonable data to conclude that, in the appropriate candidate, surgical resection o decompression is superior to endotherapy in the long-term management of pain in patients with chronic obstructive pancreatitis [24].

6. Pancreatic duct disruption 

Disruption of the main pancreatic duct or its side branches can be seen in both acute and chronic pancreatitis. Virtually every case of acute pancreatitis involves some form of duct leak which may or not persist (Fig. 4). Persistent leaks in the setting of acute pancreatitis can lead to pancreatic ascites and high-amylase pleural effusions, pancreaticobiliary fistula, as well as the disconnected duct syndrome. In patients with chronic pancreatitis, leaks are invariably associated with a downstream calculus or stricture. Endoscopy plays a significant role in the management of duct leaks in each of these clinical scenarios.

View Large Image Download to PowerPoint

Fig. 4. A small pancreatic duct stone (A) is identified on CT scan (arrow). Pancreatography demonstrates the small calculus (B) which is subsequently extracted using a stone basket (C and D).

There are now several published series reporting transpapillary stents for ductal disruptions in the setting of pancreatic ascites and high-amylase pleural effusions [25], [26] which suggest that over 90% of patients resolve their fluid collection without complication or recurrence. Patients may require simultaneous large volume paracentesis or concomitant pseudocyst drainage. It is likely that transpapillary stenting works less by leak occlusion than by bypassing potential areas of downstream obstruction and converting the duodenum to the path of least resistance to flow of pancreatic juice. Pancreaticobiliary fistulae almost invariably respond to concomitant pancreaticobiliary stenting [27] for 4–6 weeks, assuming that the fistula does not arise from the upstream portion of a disconnected gland.

7. Pancreatic pseudocysts 

In patients with both acute and chronic pancreatitis, disruption of the main pancreatic duct or its side branches leads to the leakage of amylase-rich fluid into the peri-pancreatic tissues and retroperitoneum (Fig. 4). Collections which persist for greater than 6 weeks with a characteristic round, encapsulated appearance on cross-sectional imaging are dubbed pseudocysts (so called due to the lack of a true epithelium of the cyst wall). Although they are often asymptomatic, pseudocysts can become infected, dissect into adjacent vascular structures, obstruct the gastric outlet as well as cause significant abdominal pain. The presence of these symptoms is indication for drainage. While size alone is not a criterion for drainage, larger pseudocysts tend to be the most symptomatic due to their space-occupying nature and resultant compression onto, or erosion into contiguous structures (Fig. 4).

Pseudocyst drainage can be accomplished through a variety of methods including surgery, interventional radiology and endoscopy. There are no data to suggest the superiority of any of these modalities and the choice of technique is typically based on local expertise. Endoscopic drainage can be performed with or without the use of EUS and can be approached through a transenteric or transpapillary approach. There are a number of series that have looked at either EUS or standard luminal endoscopy, or surgery for pseudocyst drainage [28], [29], [30], [31]. Varadarajula et al. have randomised various endoscopic techniques of pseudocyst drainage concluding that EUS procedures were more successful [32]. This same group demonstrated that EUS-facilitated procedures had comparable efficacy and complications when compared to surgery although endoscopically treated patients had statistically significant decreased costs and resource utilization as compared to patients treated surgically [33].

Endosonographic guidance is typically required in cases where a bulge is not visible within the gastric or duodenal lumen, although three-dimensional imaging has changed this from an absolute to relative requirement. In the presence of a luminal bulge, the wall of the pseudocyst can be punctured using a sclerotherapy needle following which contrast is injected under fluoroscopic control to ensure that the lesion has been entered. A needle knife or cystotome can then be utilized to access the pseudocyst and allow for wire guide placement. Subsequent to this, the tract is enlarged using graduated dilating catheters and/or through the scope balloon dilators. The cystgastrostomy or cystduodenostomy is then maintained by the placement of at least 2 transenteric double pigtail stents. When endoscopic ultrasound is required, the cyst is initially accessed using a 19-gauge fine needle aspiration needle though which a wire guide is placed. Dilation and stent placement can then proceed as above, either using the echoendoscope or exchanging for a therapeutic upper endoscope or duodenoscope. Transpapillary drainage can be accomplished in cases where the pseudocyst communicates directly with the main pancreatic duct. A guidewire and subsequent stent are placed directly into the pseudocyst cavity across the major or minor papilla. Small pseudocysts associated with a pancreatic duct leak may resolve following transpapillary stent placement to bridge the ongoing leak and decrease transpapillary pressures.

In cases where transenteric drainage is performed, pancreatography – either endoscopic or magnetic resonance – should be performed in order to define ductal anatomy and to determine the presence of leaks or strictures, the treatment of which may allow faster and more durable resolution of pseudocysts. Pancreatography in the setting of pancreatic pseudocysts also helps to define the duration of transenteric stent placement. In patients where there is no communication between the pseudocyst and the main pancreatic duct, transenteric stents can likely be removed in 6–8 weeks after the pseudocyst has been completely drained. In cases where there is communication with the main pancreatic duct, drainage will be required for as long as the duct leak persists [34].

The major complications associated with pseudocyst drainage – infection, bleeding, stent migration and perforation – occur in up to 37% of patients [30], [35], [36], [37]. Infection is by far the most common complication seen, and as such, peri-procedural prophylactic antibiotics are typically administered. Infection typically occurs due to incomplete evacuation of the cyst cavity – this occurs in the case of blocked stents and inadvertent drainage of organized pancreatic necrosis [38], [39]. The former can typically be avoided through the placement of at least 2 large-caliber double pigtail stents into the cyst cavity. Mistaken identification of a pancreatic fluid collection is common – especially in low volume centres where radiologists may not be accustomed to reviewing imaging of patients with pancreatitis and its associated complications. It is imperative to review high-resolution cross-sectional imaging – typically a CT scan – with an experienced radiologist prior to drainage. Collections which are debris-filled, irregular in appearance or have thick septations are more likely to be organized pancreatic necrosis or cystic neoplasms rather than pseudocysts and are therefore are at high risk for infection with transenteric drainage alone.

Bleeding associated with pseudocyst drainage is usually a result of puncture of blood vessels during the drainage procedure. Patients at high risk are those with concomitant gastric varices in whom the use of endoscopic ultrasound with colour Doppler can theoretically reduce the risk of inadvertent puncture of adjacent blood vessels. The use of electrocautery to create the internal fistula may also elevate the bleeding risk for small vessels within the duodenal or gastric wall. Another potentially severe (but rare) cause of bleeding with pseudocyst drainage is inadvertent puncture of an unsuspected pseudoaneurysm. In most cases, however, bleeding is mild and can be treated using endoscopic techniques such as infiltration with epinephrine, heater probe application or hemoclip placement. In cases of severe haemorrhage, angiography or surgery or use of cautery applied through the 10Fr cystotome (Cook Endoscopy, Winston-Salem, NC) may be required [40].

Other complications such as free abdominal perforation and stent migration are rare. Free perforation is usually seen when the pseudocyst is located greater than 1cm away from the gastric or duodenal wall. On occasion, the cyst wall may not be fully mature and cyst contents may leak into the abdominal cavity following puncture. Surgical consultation is advisable when perforation is encountered, although some cases may be managed conservatively with bowel rest antibiotics, and, if necessary, percutaneous drainage.

8. Necrotising pancreatitis 

Necrotising pancreatitis is a severe condition which complicates up to 25% of cases of acute pancreatitis; it has an associated mortality of 17% [41], [42]. Historically, infected necrosis has been a surgical disease with open surgical necrosectomy being the operation of choice. Owing to a significant associated operative morbidity, need for repeat surgery, and development of pancreaticocutaneous fistula, the last several years have witnessed the emergence of alternative strategies for debridement of infected tissue. In addition to minimally invasive surgical techniques and percutaneous drainage [43], [44], endoscopic necrosectomy has emerged as an alternative to surgical necrosectomy for patients with infected pancreatic necrosis [45], [46], [47], [48].

Building on the experience with transenteric drainage of pancreatic pseudocysts, the technique for endoscopic necrosectomy involves initial guidewire access to the necroma either via direct puncture (in the case of a visible luminal bulge) or through the use of endoscopic ultrasound with fine needle aspiration. Once wire guide access is obtained, the tract is dilated either using a graduated dilating catheter, needle-knife sphincterotome or cystotome. Subsequent dilation to 15–20mm is performed using a balloon dilator to allow passage of an upper endoscope through the newly created tract into the necroma. Necrosectomy is then performed using a combination of endoscopic accessories including large-caliber grasping forceps, snare cautery and needle knife. The tract is then maintained by the placement of 2 or more pigtail stents into the cavity across the gastric or duodenal wall following which repeat access may be required for additional debridement. In some cases, a nasocystic tube may be placed for continuous lavage.

Although endoscopic necrosectomy has the potential advantage of avoiding surgery, there are several drawbacks to the technique. First, the procedure is resource intensive and may require multiple procedures performed on consecutive days in the same patient [47], [48]. This technique may not be possible in a busy endoscopy unit with limited fluoroscopic availability. Second, the procedure-related morbidity with this technique is high [48]. The largest series to date [48] reported a procedure-related morbidity rate of 26% with a mortality rate of 7%, including 2 deaths directly related to the procedure. Because of these drawbacks, less invasive endoscopic techniques such as combined modality therapy using both endoscopic and percutaneous drainage have been investigated and described [49], [50]. It is unlikely that a randomised prospective trial will be done comparing any of the currently available modalities for treating pancreatic necrosis, or that the results can be generalized to institutions with different levels of subspecialty strengths and skills. As such, it is imperative for those who care for such patients to work with a multidisciplinary team that includes endoscopy, surgery, and interventional radiology.

9. Conclusions 

Although technological advances have, in some ways, allowed for new endoscopic approaches to acute and chronic pancreatitis, the primary complications associated with these disease processes – stenoses, leaks, and stones – have not changed. In other words, the last decade has seen some new approaches to what is fundamentally a very old problem. In contemplating endotherapy for patients with pancreatitis, it is imperative to maintain a dogmatic approach – dedication, procedural planning and accurate, high-quality cross-sectional imaging are key. Patients with pancreatitis-related complications are often extremely ill, utilize a significant number of hospital resources and present a significant clinical challenge. As such, the use of a multidisciplinary team of experienced endoscopists, surgeons, interventional radiologists as well as hospitalists is, above all, essential to achieving consistently good clinical outcomes.

Conflict of interest statement 

The authors declare that they have no conflict of interest.