In recent years, we have witnessed an evolving landscape in the management of chronic pancreatitis (CP). Endoscopy has played a pivotal role in the paradigm shift from conventional surgery toward minimally invasive approaches.

CP is a broad term used to describe multiple disease-processes which affect the pancreas and eventually lead to damage, fibrosis and atrophy of the gland. Many patients with CP will eventually develop endocrine insufficiency, in the form of type 3c (pancreatogenic) diabetes, and exocrine pancreatic insufficiency resulting in malabsorption and steatorrhea. A large proportion of patients suffer from debilitating, chronic abdominal pain which is difficult to manage and requires consultation with a pain specialist.¹ Although the cause of pain-associated with CP remains to be fully elucidated, increased pancreatic ductal (PD) pressure due to stones or strictures may be encountered.² The theory of increased PD pressure-induced pain is the basis of various endoscopic and surgical techniques developed to facilitate PD drainage. Although these techniques have been shown to be effective in select patients with anatomic ductal obstruction and dilation, the pathophysiology of pain in CP is complex often involving confounding processes including aberrant processing of pain by the central nervous system³ and the alteration of pancreatic nociceptors.⁴

While management of CP is problematic, there have been advances in endoscopic therapies utilizing endoscopic retrograde cholangiopancreatography (ERCP) and more recently endoscopic ultrasound (EUS) in facilitating PD drainage with acceptable risks and effective pain relief. Aside from chronic pain, patients with CP may develop other adverse events such as pancreatic fluid collections, PD strictures, PD disruptions, and progression to pancreatic carcinoma. Our aim is to review and evaluate the most recent developments in endoscopic therapy in managing complications of CP.

Patients with chronic calcific pancreatitis (CCP) develop calcifications in the parenchyma or in the duct.¹ The clinical rationale behind treatment of PD stones and strictures originates in the principle that decreasing intraductal pressures may result in reduced pain,² which can be achieved with pancreatic drainage surgeries, or endoscopically by removing stones or bypassing and dilating an area of narrowing with a stent. A randomized clinical trial performed by Cahen et al⁵ included 39 patients with painful CCP randomized to endoscopy versus surgery. After 24 months of follow up, patients who had surgical drainage achieved a higher rate of pain control (75%) when compared to pancreatic endotherapy (32%). Long-term follow-up (79 months) found that most patients in the endoscopy group (68%) required re-interventions versus 5% in the surgery group. Improved pain relief persisted in the surgery group, and 47% of endoscopic therapy patients ultimately underwent pancreatic surgery.⁶ The European Society for Gastrointestinal Endoscopy (ESGE) recommends endoscopy as a first-line modality in treatment of painful CP while acknowledging evidence that surgery has superior outcomes, as endoscopic therapy may provide significant pain relief for the patients and carries a low risk of complications when compared to major pancreatic surgery.^7,8

Since the clinical trials by Cahen et al,^5,6 there has been nearly a decade in advances in techniques for pancreatic endotherapy, with more recent studies showing high clinical efficacy in the treatment of pancreatic duct stones.^9,10 Extra corporeal shockwave lithotripsy (ESWL) followed by endoscopic removal of the PD stone fragments has shown excellent outcomes in the treatment of CCP. The largest study showing the success of this technique comes from the Asian Institute of Gastroenterology in Hyderabad, India. Investigators studied a cohort of 636 patients, reporting 68.7% were pain-free at 2–5 years follow-up and 60.3% remained pain-free after 5 years with improved quality of life scores.⁹

A retrospective cross-sectional study from Virginia Mason in Seattle, USA, showcased the center’s experience using ESWL in a western population. The authors noted 85% of patients achieved improved pain control after ESWL followed by endoscopic stone extraction. Moreover, 50% of patients were narcotic-free and the majority reported improved quality of life. Interestingly, a significant proportion of patients had follow-up procedures such as ESWL (29%), ERCP (84%), and surgery (16%); suggesting major surgery was avoided in over 80% of patients.¹⁰

Additional reports confirm the efficacy of ESWL alone for CCP treatment, making it a reasonable first step in treatment of PD obstruction and pain due to PD calcifications. A randomized trial published by Dumonceau et al¹¹ found ESWL alone is safe and effective, with similar outcomes when compared to ESWL plus endotherapy. While endotherapy alone for pancreatic duct stones less than 5 mm is acceptable, stones larger than 5 mm are usually spiculated and adhere to the pancreatic duct wall; larger stones require fragmentation to facilitate removal and reduce the risk of complications.⁷

Proper patient selection is paramount to achieve success as not all CCP patients are appropriate candidates for ESWL and endotherapy. One of the strongest predictors associated with success is pancreatic head predominant disease.¹¹ Patients with calcifications in the pancreatic tail would likely benefit from a distal pancreatic resection as the effectiveness of ESWL and endotherapy in this patient population is unclear. Patients who smoke should be encouraged to stop as abstinence from smoking has been shown to improve narcotic requirements after ESWL plus endotherapy.¹⁰

Pancreatoscopy-guided intraductal lithotripsy (with laser or electrohydraulic lithotriptors) is possible, but most studies showing its effectiveness are small.⁷ An US multicenter study of 28 patients over three years found the technical success of lithotripsy was 79%; and 89% of patients had reduced pain scores and decreased narcotic use.¹² Lithotriptor technology is relatively experimental and the decision to incorporate intraductal lithotripsy in CCP treatment should be made based on local expertise and availability with multidisciplinary input in selection and management.

Patients with large stones that are refractory to ESWL and endotherapy, calcifications in the entire duct and parenchymal calcifications (especially peripheral side-branch stones; Fig. 1), should be referred to a pancreatic surgeon for consideration of surgical decompression or total pancreatectomy with auto-islet cell transplantation.^7,13–15

Eighteen percent of patients with CP will develop main pancreatic duct (MPD) strictures. Significant strictures may cause upstream MPD dilation and increased ductal pressure with resultant pancreatic-type pain. The endoscopic treatment of MPD strictures is indicated for dominant duct strictures in patients with (1) upstream MPD dilation > 6 mm in diameter, (2) obstruction of contrast outflow alongside a 6-french nasopancreatic catheter, or continued pain after infusion of 1 L of saline over a nasopancreatic catheter.¹⁶ Pancreatic duct strictures may be treated with pancreatic sphincterotomy and stent placement across the stricture. In many occasions, MPD stricture dilation with bougies, balloons or a Soehendra stent retriever is necessary to cross the stricture and gain proximal PD access.⁷

Pancreatic stent placement has been associated with high rates of technical success and good clinical outcomes. In a study of 19 patients followed for 5 years after PD stent placement, the investigators found that 17 patients had successful initial endoscopic intervention and all patients reported immediate pain relief. Remarkably, after 5 years 57% of patients reported persistent long term benefit.¹⁷

The largest series of patients comes from Brussels, Belgium where the outcomes of 100 patients who underwent pancreatic duct stricture treatment with PD stent placement was reported. Seventy-percent of patients achieved adequate pain control, although approximately 30% of patients required repeat stent placement within one year. At the end of the follow up period, 40% of patients overall had required a repeat endoscopic intervention.¹⁸

At the initial ERCP session, the goal should be to place a plastic stent of the largest possible diameter to bypass the stricture. Small diameter stents, as well as MPD dilation may be required for tight strictures. The goal should be placing the largest possible plastic stent, for the stricture, by stent exchanges with incremental increases in size over a 1 year period. In patients with high-grade PD strictures which persist after 12 months of single PD stent placement, stricture dilation with multiple plastic stents side-by-side may provide benefit.^7,8 Successful endoscopic drainage may predict clinical success in patients who undergo a future lateral pancreaticojejunostomy.¹⁹ Therefore, monitoring the clinical response to PD stent placement may aid in selecting appropriate candidates for surgery or continued non-surgical treatments, which appropriately balances the risks and benefits of a technically challenging and higher risk operation.

Use of fully covered metallic stents have been investigated for treatment of MPD strictures. In a systematic review, five studies were identified comparing plastic to metal stent placement for MPD strictures. The review included 80 patients in whom metal stents were placed. Overall, the technical success, rates of re-intervention, stent migration and rates of pain relief were similar in patients with metal and plastic stents.²⁰ Uncovered metallic stents should not be used to treat pancreatic duct strictures, as uncovered stents embed in the tissue, making stent removal difficult or impossible.

As for patients with pancreatic duct disruption, the use of plastic stents appears economical, safe and effective. One of the first large studies was a single center retrospective cohort of 97 patients with ductal disruption where stent insertion was successful in 95% of patients with a clinical success rate of 55% where the disruption was bypassed in 92% of cases.²¹ A more recent study, also retrospective and including 107 patients from two different centers, showed a technical success of PD stent placement of 96%. In 44% of cases the PD disruption was bypassed and 75% of patients achieved clinical success (no visible leak on ERCP or in follow-up imaging).²² Key factors associated with clinical success in the treatment of PD disruption with endoscopic therapy include (1) the presence of a partially disrupted duct^21,22 and (2) stent placement bridging the area of ductal disruption.²¹

Twenty to forty percent of patients with CP may develop pancreatic pseudocysts (PCs),^8,23 although the majority of PCs resolve spontaneously or remain asymptomatic. PCs are common adverse events following pancreatectomy. In a retrospective study of over 700 patients, 21% developed PCs with up to 35% requiring intervention.²⁴ Indications for treatment of PCs include symptoms caused by compression on adjacent structures including epigastric pain, early satiety, unintended weight loss, gastroduodenal outlet obstruction (GOO) or biliary obstruction.^7,25 Endoscopic therapy of PCs has become the “new gold standard” due to low risk and improved outcomes when compared to percutaneous and surgical approaches. When compared to conventional surgery, endoscopic transmural drainage has improved outcomes offering the advantage of shorter hospital stays, better quality of life and lower healthcare associated costs.^26,27

The use of EUS to guide transmural drainage and avoid injury to surrounding vasculature and organs is strongly encouraged. A prospective randomized trial of 53 patients who underwent endoscopic transmural drainage and, if unsuccessful, subsequent EUS-guided PC drainage, showed that endoscopic drainage alone is possible and highly successful in patients who had fluid collections in the pancreatic head or body, but not in the pancreatic tail.²⁸ A subsequent study comparing endoscopic drainage alone versus EUS-guided drainage showed that EUS-guidance was significantly more successful than endoscopy alone (100% vs 33%, P < 0.001) with no difference in complications. Interestingly, in a patient who underwent EUS, an alternative diagnosis of a biliary cystadenoma was made and the patient did not undergo endoscopic drainage.²⁹ We have two additional cases of pancreatic fluid collections that had an alternative diagnosis after computed tomography and/or EUS evaluation; these cases are illustrated in Fig. 2.

Traditionally, the endoscopic drainage of PCs was achieved with placement of one or more small-caliber plastic stents. The recent development of lumen apposing metal stents (LAMS), which are covered with a synthetic membrane to prevent tissue embedment, has significantly advanced the field of interventional gastroenterology. LAMS are now frequently employed to treat PCs and walled-off necrotic fluid collections and have been proven to be safe and effective.³⁰ The use of a LAMS for PC drainage is illustrated in Fig. 3.

In a prospective study of 33 patients with PCs who underwent LAMS placement for transmural endoscopic drainage, stent placement was technically successful in 90% of patients, with a 93% resolution rate.³¹ No head-to-head trials comparing plastic stents versus LAMS in drainage of walled-off pancreatic fluid collections have been reported. A retrospective study showed no difference on clinical outcomes; however, there were significant differences in procedural duration (less when using LAMS) and expense (higher when using LAMS); the economical difference in cost was not significantly different when used in drainage of walled-off pancreatic necrosis.³² Recent studies confirm high technical and clinical success rates when using LAMS for drainage of necrotic pancreatic collections.^31,33,34

A recent report comparing outcomes of EUS-guided transmural drainage of PCs versus combined transmural-transpapillary drainage found no added benefit with placement of a transpapillary PD stent.³⁵ This study included 174 patients from 15 centers, of which 79 underwent combined transmural-transpapillary drainage. Patients with CP composed 45% of the cohort. Importantly, transpapillary drainage attempts were associated with reduced resolution of PCs in a multivariate model controlling for age, sex, and etiology of pancreatitis (gallstone vs idiopathic). Limitations of the study include retrospective design and inability to control for patient selection (e.g., presence of ductal disruption).

This procedure was typically performed intraoperatively or guided by conventional transabdominal ultrasound or computed tomography.^36–38 The first reports using EUS to guide in regional nerve injections into the celiac plexus were reported in 1996.^39,40 Since then, EUS-guided techniques targeting the celiac ganglia have become a popular to treat patients with persistent pancreatic-disease related pain. The techniques for EUS-guided celiac plexus injection have evolved throughout the years. By visualizing the celiac plexus with endosonography, a needle may be inserted to deliver a local anesthetic followed by a steroid agent to perform a nerve block (EUS-guided celiac plexus block, EUS-CPB) or absolute alcohol (EUS-guided celiac plexus neurolysis, EUS-CPN) to achieve neurolysis of the celiac plexus. The intention is to cease the transmission of pain nociception from the celiac neuronal ganglia to the central nervous system.

The EUS-CPB technique is most commonly used in patients with CP. In a study of 248 patients with CP who underwent EUS-CPB, 76% of them had a median duration of pain control of 10 weeks. Most patients underwent a median of three procedures and lack of response after the first procedure was predictive of failure in subsequent sessions.⁴¹ In a meta-analysis that included 6 studies and 221 patients with CP, EUS-CPB was effective in 51.46% of patients. EUS-CPN is effective in patients with pancreatic cancer^42–44 and it has been used in patients with CP as well, achieving acceptable pain-relief shortly after the procedure, although the long-term effects are still uncertain. Another meta-analysis including 9 studies with 376 patients showed EUS-CPN procedures provided pain relief in 59% of patients with CP.^45,46 The use of single celiac plexus injection versus two injections has not been found to provide additional benefit.⁴⁷

Recent studies describe a more targeted approach where the injection is performed under EUS guidance directly into the celiac ganglia for neurolysis (EUS-guided celiac ganglia neurolysis, EUS-CGN). The initial data published in 2008 showed that this technique was efficacious and safe in patients with pancreatic cancer and CP.⁴⁸ A randomized control trial comparing the EUS-CPN versus EUS-CGN showed that the endoscopist could visualize the celiac ganglia by EUS in 88% of cases and EUS-CGN had a better response rate (75.5%) when compared to standard EUS-CPN (45.5%).⁴⁹ A recent study evaluating this technique with more detail, showed that celiac plexus ganglia could be visualized by EUS most the times and that the spread of ethanol using EUS-CGN (high volume, 4 mL per ganglion) resulted in a large and bilateral spread of ethanol comparable to the standard EUS-CPN technique. The authors hypothesize that the increased clinical effect of EUS-CGN compared to EUS-CPN could be because the ganglion are affected from the inside out as well as from the outside in.⁵⁰ The usefulness of these techniques in patients with CP continues to be controversial.⁵¹ Further research is needed to describe long-term outcomes of these techniques, and its use should be carefully weighed against procedural risks.

When ERCP is unsuccessful in selective pancreatic duct cannulation, EUS-guided pancreatic duct drainage (EUS-PD drainage) may be used to achieve PD access.

One approach is the EUS-guided rendezvous technique. Using EUS, the PD is visualized and targeted for transmural needle puncture. Subsequently a guidewire is advanced antegrade through the major papilla and the echoendoscope is removed. The pancreatic rendezvous is completed by transpapillary snare capture of the guidewire using a duodenoscope, which then allows retrograde pancreatic duct access for ERCP-mediated therapy.

Direct antegrade treatment by puncturing a dilated pancreatic duct from the stomach or duodenum via EUS followed by transmural stent placement has been described as an alternative. Stent placement across the point of obstruction and/or papilla (transpapillary drainage) depends on stricture location and anatomy of the patient. If bridging the stricture is not feasible, the stent may be placed in the pancreatic duct facilitating pancreatic duct drainage into the gastrointestinal lumen.⁵² It is important to note that in the antegrade technique, tract dilation is necessary to be able to pass the visceral wall and connect the lumen to the pancreatic duct.^52,53 Chen et al⁵⁴ recently published a video case series where different techniques of pancreatic duct drainage are showcased.

The first reports of EUS-PD drainage were published in the early 2000s.^55,56 In an early case series of six patients EUS-PD drainage showed reasonable initial success.⁵⁶ In a single center retrospective evaluation including 20 patients over 10 years, EUS-PD drainage was technically successful in 90% of cases. After median follow-up of 37 months, 72% of patients achieved pain control.⁵⁷ A retrospective study by Fujii et al⁵⁸ reported their single center experience in 43 patients. This report showed an overall technical success of 74% and complete symptom resolution was reported in 83% of cases. The same author recently published a review paper summarizing the outcomes of multiple studies, including 222 patients, where the technique was successful in 77% of cases.⁵³

In general, EUS-PD drainage has adequate technical success rates, although complications are common. These complications include pancreatitis, pancreatic leaks, bleeding, perforation and even fatal events.^59,60 Adverse outcomes reported by the Fujii-Lau and Levy⁵³ review found that 18.9% of patients suffered adverse events associated with EUS-PD drainage. EUS-PD drainage can lead to a significant number of stent-related adverse events ranging from 25% to 55% of cases.⁵² Due to its high technical challenge, significant incidence of adverse events and rare indications for EUS-PD drainage procedures, prospective data are limited. Multicenter, prospective studies would be useful to further refine EUS-PD drainage and define its ultimate value.⁶¹

The term GOO is used to define any mechanical obstruction impairing gastric emptying into the small bowel. The area of obstruction can be in the distal stomach, proximal duodenum or due to extrinsic compression. In the 1970s the most common cause of GOO was pyloric stenosis associated with peptic ulcer disease. With improved treatments for peptic ulcers and helicobacter pylori since the 1980s, the most common etiology of GOO has become pancreatic cancer.⁶² While CP patients can develop gastroparesis in > 40% of cases,⁶³ the development of GOO is rare (when not associated with pseudocyst-induced extrinsic compression).

A video case report was published in 2015 by Law et al⁶⁴ showing the utilization of EUS-guided gastroenterostomy (EUS-GE) in a patient with CP. More recently, two small studies evaluating the efficacy and safety of EUS-GE have been published including patients with benign or malignant GOO. These studies reported the results of 10 patients⁶⁵ and 26 patients,⁶⁶ both describing 90% technical success rates, and clinical success rates of 100% and 85%, respectively.^65,66 The rates of adverse events were 0% in the series by Khashab et al⁶⁵ and 3% in the larger report by Tyberg et al⁶⁶ with one death. It is important to note limited numbers of patients with CP in both studies; 3/10 and 9/26, respectively. At this time, EUS-GE is investigational and reserved for poor surgical candidates; multidisciplinary care is suggested incorporating physicians, surgeons, radiologists and gastroenterologists. EUS-GE remains to be prospectively evaluated, but early results show promise.

EUS elastography is a novel technology that has been explored more extensively in Asia and Europe. It consists in measuring the difference of tissue hardness and strain in the desired area of examination and it can provide objective, quantitative results. This measurement is translated into the EUS screen, where the red, yellow and green spectrum of colors represents softer structures and the blue spectrum represents firm tissues, possibly due to cancer or fibrosis for example. In addition, EUS elastography provides a histogram which is a graphical representation of colors in the area of interest. In a prospective study of 191 patients with epigastric pain syndrome or known CP, EUS elastography had an accuracy of 91.1% in diagnosing CP with a cut-off strain ratio of 2.25, and provided significant objective information to support the EUS findings.⁶⁷ This technology has proven to be useful in differentiating focal masses in patients with CP and pancreatic cancer.⁶⁸ In addition, EUS elastography may help target malignant lymph nodes (classically firm, blue spectrum dense nodes).⁶⁷ EUS elastography alone has a modest capability to discriminate between benign and malignant lesions, thus use as a supplement to fine-needle aspiration cytology is currently advised. Further attention is warranted to better define the role of EUS elastography in diagnostic and therapeutic applications.

This review summarizes published data regarding endoscopic management of CP and associated complications. Proper patient selection is paramount to achieve reasonable balance of risks and benefits of endoscopic therapy, while multidisciplinary care ensures employment of feasible alternatives such as medications and surgery. Due to the limited subset of patient who are ideal candidates for these procedures, multicenter research efforts should be pursued to optimally evaluate the safety and efficacy of minimally invasive endoscopic treatments.

Fig. 2B courtesy of Eric Tamm, MD.