Management of Chronic Pancreatitis - Loyola …mistaken for acute pancreatitis. In this phase, clear-cut evidence of chronic pancreatitis can be lacking. With time, pain can become

swt

DISEA

SEA

ND

THER

APY

OF

PA

NCREA

TICD

ISORD

ER

GASTROENTEROLOGY 2013;144:1282–1291

Management of Chronic Pancreatitis

Christopher E. Forsmark

Division of Gastroenterology, Hepatology, and Nutrition, University of Florida, Gainesville, Florida

adtp

Advances in our understanding of chronic pancreati-tis have improved our care of patients with this dis-ease. Although our therapies are imperfect and manypatients remain symptomatic, appropriate medicalcare improves the quality of life in these patients.Proper management requires an accurate diagnosis,recognition of the modifiable causes of disease, assess-ment of symptoms and complications, treatment ofthese symptoms and complications utilizing a multi-disciplinary team, and ongoing monitoring for theeffect of therapy and the occurrence of complications.

Keywords: Chronic; Pancreatitis; Management.

Chronic pancreatitis is a syndrome characterized byinflammation, fibrosis, and loss of acinar and islet

cells. The syndrome can produce symptoms (pain) and,with sufficient tissue destruction, exocrine or endocrineinsufficiency. Not all patients with chronic pancreatitisdevelop symptoms or exocrine or endocrine insufficiency.Disease progression often begins with an early phase thatis characterized by episodes of abdominal pain and can bemistaken for acute pancreatitis. In this phase, clear-cutevidence of chronic pancreatitis can be lacking. With time,pain can become more persistent and severe, imaging testscan show evidence of chronic pancreatitis, and exocrineand endocrine insufficiency might develop. This processcan take many years.1 Although chronic pancreatitiseems to first require acute pancreatitis, most patientsith acute pancreatitis of various etiologies do not go on

o develop chronic pancreatitis2; the risk of progression tochronic disease is greatest among patients who smoke orhave had alcohol-related acute pancreatitis.

The management of chronic pancreatitis is challenging,and most patients remain symptomatic despite therapy.There are no effective methods to stop progression orreverse this syndrome. However, a number of new insightshave improved therapy and provided some evidence basedon which therapy to choose.

Proper management of chronic pancreatitis starts with

tients with early-stage disease, who do not always havedefinitive evidence of chronic pancreatitis. Next, it is im-portant to determine etiologic agents that contribute todisease so that these can be addressed if possible. Individ-ualized treatment of symptoms and exocrine and endo-crine insufficiency is necessary, because patients have vari-able expression of disease. Finally, ongoing monitoringprovides opportunities to detect complications at earlierstages so that more effective therapy is possible. Optimalmanagement of these patients requires a multidisciplinaryteam that includes gastroenterologists, surgeons, endocri-nologists, dieticians, pain management, psychiatrists, so-cial workers, and patient support groups.

DiagnosisChronic pancreatitis is usually diagnosed based on

results of imaging analyses or tests of pancreatic function.Although histologic evidence of fibrosis and tissue lossmight be considered the most definitive diagnostic crite-ria, it is rarely available. In addition, similar changes inhistologic features are observed in patients without symp-toms of chronic pancreatitis, such as very elderly patients,those with diabetes or renal failure, and those who smoke.These changes can result from normal “wear and tear” onthe pancreas. Diagnosis requires recognition of the varietyof symptoms (if present) and complications that consti-tute the syndrome,3 in conjunction with appropriate im-

ging or functional tests (Table 1). In early stages ofisease progression, results from these tests can be nega-ive or inconclusive. In later stages, marked changes inancreatic structure and function make diagnosis much

Abbreviations used in this paper: CT, computed tomography; DPPHR,duodenum-preserving pancreatic head resection; ERCP, endoscopicretrograde cholangiopancreatography; ESWL, extracorporeal shockwave lithotripsy; EUS, endoscopic ultrasonography; MRCP, magneticresonance cholangiopancreatography; MRI, magnetic resonance imag-ing; TIGAR-O, Toxic-Metabolic, Idiopathic, Genetic, Autoimmune, Recur-rent and Severe Acute Pancreatitis, Obstructive.

© 2013 by the AGA Institute0016-5085/$36.00

http://dx.doi.org/10.1053/j.gastro.2013.02.008

sfiw

peaa

mcsaitMooaae

paash

bsciAact

lphaj

S ts

c

DIS

EASE

AN

DTH

ERA

PY

OF

PA

NCREA

TIC

DIS

ORD

ER

May 2013 MANAGEMENT OF CHRONIC PANCREATITIS 1283

easier and more accurate. Of course, at this later stage,there is less opportunity to interrupt disease progression.

Accurate diagnosis is therefore especially importantduring the early stages of disease. Patients arrive at manypancreas specialty clinics with a chronic abdominal painsyndrome and normal results from pancreatic imagingtests, yet they have been incorrectly diagnosed withchronic pancreatitis and may have been subjected to dan-gerous and ultimately futile interventions, such as endo-scopic retrograde cholangiopancreatography (ERCP) withstent placement or prescriptions for narcotics.

Characteristic findings from imaging techniques suchas computed tomography (CT) or ultrasonography in-clude atrophy of the pancreas, a dilated pancreatic duct,and pancreatic calcifications (Figure 1). These features arepathognomonic of chronic pancreatitis and can take 5 to10 years or more to develop.1 Pancreatic calcificationseem to be much more common in patients with certainorms of chronic pancreatitis, especially those with hered-tary pancreatitis (caused by variants of PRSS1), patientsho smoke or drink alcohol, or patients with tropical

Table 1. Diagnostic Tests for Chronic Pancreatitis

Imaging tests

EUS Two systems of reporting are used: standaexamination of both the pancreatic paren

MRI with MRCP Administration of secretin during MRCP impancreatic secretory capacity to be estim

CT CT images the pancreatic parenchyma wellERCP ERCP provides the most detailed images oUltrasonography Ultrasonography has limited ability to imag

Functional tests

Secretin test Administration of a supraphysiologic dosecollected with a Dreiling tube or an endo

Fecal elastase Low levels in the stool (�200 �g/g of stooerum trypsin Low levels (�20 mg/dL) are seen in patien

Figure 1. CT scan shows a dilated pancreatic duct and a large pan-

ancreatitis. Patients with chronic pancreatitis of othertiologies can also develop calcifications, detected by im-ging tests, although these are observed less frequentlynd develop more slowly.

Additional methods to image the pancreas includeagnetic resonance imaging (MRI), magnetic resonance

holangiopancreatography (MRCP, with or without theecretagogue secretin), endoscopic ultrasonography (EUS),nd ERCP. MRI with MRCP has replaced CT at manynstitutions. One advantage of MRI is that it can be usedo characterize the pancreatic ductal anatomy (with

RCP), but it often misses calcifications. Administrationf secretin during MRCP improves the quality of imagesf the pancreatic duct, although the gain in diagnosticccuracy with this technique has not been defined. Inddition, the quantity of pancreatic secretions can bestimated using MRI with secretin.4

EUS allows for detailed imaging of the pancreatic ductand parenchyma (Figure 2). EUS images can be inter-preted using the traditional 9-feature scheme or the newlyproposed Rosemont criteria.5 These systems do not ap-

ear to differ in diagnostic accuracy for chronic pancre-titis. However, it is important to recognize that subtlebnormalities detected by EUS are nonspecific and notufficient for diagnosis of chronic pancreatitis. EUS has aigh level of sensitivity and a low level of specificity.Imaging techniques such as CT, MRI, and EUS can also

e used to exclude other conditions that produce similarymptoms to, or complicate, chronic pancreatitis, espe-ially pancreatic ductal adenocarcinoma, intraductal pap-llary mucinous neoplasm, and other cystic neoplasms.lthough ERCP provides detailed images of the pancre-tic duct, it is not appropriate for use in diagnosis ofhronic pancreatitis, although it might be used therapeu-ically.

Biochemical tests of pancreatic function are used muchess commonly than imaging analyses. In a direct test ofancreatic function, patients are given an infusion of theormone secretin, and pancreatic secretions are collectednd analyzed for bicarbonate concentrations; pancreaticuice is collected using a Dreiling tube or an endoscope.

erminology and Rosemont criteria.5 EUS allows detailedyma and the pancreatic duct.es the quality of imaging of the pancreatic duct and may allow thed. MRI cannot visualize calcification.

ith less ductal detail than MRI.e pancreatic duct but is rarely used for diagnosis.e pancreas, but it is of low cost and has no ionizing radiation.

ecretin produces maximal pancreatic stimulation; pancreatic juice ispe and analyzed for bicarbonate concentration.re seen in patients with advanced chronic pancreatitis.with advanced chronic pancreatitis.

rd tch

provate, wf the th

of sscol) a

lthough there is evidence that this test can detect

tpm

fcappwiynec

k

cettpip

aatstameppipa

tosa

owoaw

DISEA

SEA

ND

THER

APY

OF

PA

NCREA

TICD

ISORD

ER

1284 CHRISTOPHER E. FORSMARK GASTROENTEROLOGY Vol. 144, No. 6

chronic pancreatitis at earlier stages than imaging studies,it is only performed at a few centers and is not available tomost patients. Indirect assessment of pancreatic functionusing tests such as fecal elastase is also possible but notaccurate for diagnosis. A diagnostic algorithm has beenrecently proposed6 that first uses commonly availableests (CT, MRI, and EUS) and reserves direct tests ofancreatic function to clarify equivocal results (Supple-entary Figure 1).Analysis of the fluid collected during a direct pancreatic

unction test, in addition to bicarbonate concentration,an allow additional insights into mechanisms of diseasend identify new diagnostic tools. Differentially expressedroteins can be identified in pancreatic secretions fromatients with chronic pancreatitis compared with patientsith nonpancreatic pain.7 These proteins include many

nvolved in inflammation, fibrosis, and pain. Similar anal-ses are possible in patients with malignant or premalig-ant pancreatic disease. Whether these approaches willvolve to accurate diagnostic tests in early chronic pan-reatitis is not yet known.

Determination of EtiologyIn the past, alcohol consumption was believed to

cause most cases of chronic pancreatitis. Recent evidenceindicates that although alcohol contributes significantlyto pathogenesis, it is not the main cause of disease formost patients. The Toxic-Metabolic, Idiopathic, Genetic,Autoimmune, Recurrent and Severe Acute Pancreatitis,Obstructive (TIGAR-O) classification system8 categorizes

Figure 2. EUS image shows subtle sonographic features of chronicpancreatitis, with a hyperechoic duct margin and hyperechoic foci in thegland. These features are nonspecific and cannot be used alone todiagnose chronic pancreatitis.

nown causes and factors that contribute to chronic pan- h

reatitis (Supplementary Table 1). Determining diseasetiology(s) provides opportunities for focused and specificreatments, such as corticosteroids for patients with au-oimmune pancreatitis, relief of duct obstructions foratients with chronic ductal strictures, and lifestyle mod-

fications (abstinence) to slow disease progression andreserve pancreatic tissue.Long-term ingestion of large quantities of alcohol (on

verage, �5 drinks per day) is required for development oflcohol-associated chronic pancreatitis. Smoking appearso be an equally injurious toxin. Continued alcohol use ormoking accelerates the progression of chronic pancreati-is and the development of pancreatic calcifications andlso increases the risk of pancreatic and nonpancreaticalignancy. Physicians should make vigorous efforts to

ncourage sustained abstinence for patients with chronicancreatitis as early in the clinical course of disease asossible. A randomized trial showed that even simple

nterventions (a 30-minute discussion), delivered in a re-eated and systematic way, helped patients abstain fromlcohol and reduced recurrence of pancreatitis.9 Because

many gastroenterologists may not have the expertise orexperience to provide support for smoking cessation andalcohol abstinence, referral of patients to appropriate sup-port groups should be consistently provided.

Treatment of PainMost patients with chronic pancreatitis have ab-

dominal pain. Pain takes many forms; a common presen-tation is a chronic and continuous pain, with or withoutexacerbations. Pain severity, character, timing, and re-sponse to therapy vary. Pain can change over time inindividual patients and in some patients can eventuallydisappear. Pain can develop during early stages of chronicpancreatitis, before development of easily visible struc-tural abnormalities of the pancreas. In this situation, itcan be difficult to determine if pain is actually caused bychronic pancreatitis. Pain is the symptom that most fre-quently causes patients to seek medical care; it accountsfor most chronic pancreatitis–associated medical costsand causes the greatest reductions in social function andquality of life. Recent studies have found that certain painpatterns are most detrimental to quality of life. Continu-ous pain, even if less intense than intermittent pain, isassociated with a lower quality of life, increased disability,and greater health care utilization.10 It is helpful to quan-ify pain severity and character, determine baseline levelsf pain, and measure the effects of therapies. Some as-essment of quality of life and disability is also appropri-te on a periodic basis.

In the past, pain was believed to be mainly caused bybstruction of the pancreatic duct (by stricture or stone),hich results in high pressure or ischemia above thebstruction. Therapies that relieve the obstruction, suchs ERCP with stent placement or surgical ductal drainage,ere therefore used to treat pain. Although this approach

DIS

EASE

AN

DTH

ERA

PY

OF

PA

NCREA

TIC

DIS

ORD

ER

May 2013 MANAGEMENT OF CHRONIC PANCREATITIS 1285

predictable. Of course, patients with early-stage diseasecan have severe pain without any evidence of ductal ob-struction, and patients with a stone or stricture causingduct obstruction may have no pain.

More recent studies have moved beyond this simple“plumbing” viewpoint of pathophysiology. The pancreascontains many nociceptive neurons. In patients withchronic pancreatitis, these neurons are increased in sizeand often surrounded by inflammatory infiltrates. In ad-dition to the usual stimuli that cause pain (ischemia, heat,inflammation), the pancreatic enzyme trypsin stronglyactivates these nociceptors. The primary nociceptive neu-rons communicate with second-order neurons in the spi-nal cord, which can develop abnormal function and be-come sensitized. Sensitization produces hyperalgesia (amagnified pain response to nociceptive stimuli) and allo-dynia (a pain response to a normal physiologic stimulus).These second-order neurons communicate with third-or-der neurons in the brain, which connect to the limbicsystem and somatosensory cortex; this causes the physicalsensation of pain and the emotional response (suffering)caused by pain.

There are significant changes in brain function, micro-structure, and macrostructure in patients with painfulchronic pancreatitis, producing changes in electroenceph-alography patterns, brain-evoked potentials, and corticalorganization and thickness. These changes are similar tothose observed in patients with other conditions thatcause chronic visceral or somatic pain. These changes innociception indicate that pancreatitis-related pain couldbe primarily a “wiring” problem, rather than a plumbingproblem, that would not respond to treatments that sim-ply remove ductal obstructions.11 These changes couldaccount for the benefits of agents such as gabapentoids inreducing pain in patients with chronic pancreatitis.12

Their effects are likely to result from their abilities tomodulate nociception. Finally, it is worth noting that theplacebo response in these patients is at least 20%.

The first step in treating pain is to search for treatablecomplications of chronic pancreatitis. These could in-clude a pancreatic pseudocyst, obstruction of a surround-ing hollow viscus (bile duct or duodenum), or cancer.Although these complications do not always cause pain,they do have specific treatments. Cross-sectional imagingwith CT or MRI can be used to identify complications andscreen for conditions that produce symptoms that aresimilar to those of chronic pancreatitis (pancreatic ade-nocarcinoma or intraductal papillary mucinous neo-plasm).

Pain treatment begins with medical therapy. One im-portant goal of effective medical therapy is to slow diseaseprogression, which involves focused efforts to achieveabstinence from tobacco and alcohol if they have roles indisease development. Alcohol abstinence usually reducespain, although the magnitude of the effect is unpredict-able. More importantly, abstinence slows disease progres-sion, reduces the likelihood of complications such as

Most patients with pain will require analgesics. Cross-sectional and cohort studies estimate that approximatelyhalf of all patients with chronic pancreatitis will betreated with opioids. The precise risk of addiction is notknown, but studies of patients with other chronic painsyndromes indicate that it is less than 20%. Patients withprevious addictive behaviors, including alcohol abuse orsmoking, are most prone to dependence, abuse, or addic-tion; the rate of addiction and other aberrant drug-relatedbehaviors thus varies among patients with chronic pan-creatitis.

Notwithstanding the risk of addiction, the main goal ispain relief. It is appropriate to begin treatment with less-potent opioids. Tramadol is commonly used for this pur-pose in dosages of 200 to 400 mg daily, although higherdosages are given to some patients. More potent narcoticsare often required, and it is appropriate to slowly increasepotency and frequency, with a goal of reducing but noteliminating pain. A number of other agents are given withopioids to manage chronic pain syndromes. These includetricyclic antidepressants, selective serotonin reuptake in-hibitors, serotonin-norepinephrine reuptake inhibitors,and gabapentoids. Of these, only pregabalin has beenstudied in a randomized controlled trial in patients withchronic pancreatitis.12 Patients treated with pregabalin(up to 300 mg twice daily) had reduced pain comparedwith those given placebo and were able to reduce opioiduse. Side effects were more common in the pregabalingroup (lightheadedness or a feeling of being drunk). Pre-liminary studies suggest that pregabalin inhibits centralsensitization. It is not clear whether other adjunct agentsare equally effective or if combinations of these agents aremore effective. Nonetheless, use of adjuvant agents isreasonable for patients who require opioids for pain con-trol.

Additional medical options for pain include adminis-tration of pancreatic enzymes, octreotide, antioxidants,and various nontraditional therapies. Pancreatic enzymeshave been studied in 7 small randomized trials with cross-over designs, with mixed results. These studies had signif-icant heterogeneity in patient selection, enzyme dosageand formulation, and outcome measures.14 Octreotidehas been studied in 4 randomized trials with mixed resultsand is rarely used. Although enzyme therapy may notprovide substantial benefit, it is often tried because of itssafety and the lack of other highly effective treatments.

The effects of antioxidants were tested in 2 relativelylarge randomized trials.15,16 Although serum levels of an-tioxidants increased in both trials, they produced differ-ent results for the outcome of pain relief. The trialsincluded different types of patients; the trial with thepositive results included much younger patients withmainly idiopathic pancreatitis, whereas the trial that pro-duced negative results included older patients with alco-hol and smoking as the primary etiologies.17 Antioxidantsmight reduce pain, but further studies are needed to

p

aprapat

adidennn

nanspptcpshpnag

nidbvaasmptl

DISEA

SEA

ND

THER

APY

OF

PA

NCREA

TICD

ISORD

ER

1286 CHRISTOPHER E. FORSMARK GASTROENTEROLOGY Vol. 144, No. 6

Patients with chronic pancreatitis should be educatedabout healthy lifestyle choices. They have increased risksof osteoporosis, osteopenia, and fractures,18 so periodicassessments of bone density and vitamin D levels as wellas routine dietary supplementation with vitamin D andcalcium are recommended.

There is one form of chronic pancreatitis that hasspecific medical therapy. Autoimmune pancreatitis occursin 2 forms.19 Type 1 is characterized by the presence ofimmunoglobulin G4 –positive plasma cells in affected or-gans, and some patients have increased serum levels ofimmunoglobulin G4. The most common presentation isobstructive jaundice, but a variety of organs can be in-volved, including salivary glands, bile ducts, kidneys, andlungs. Type 2 is not associated with altered levels ofimmunoglobulin G4 and involves only the pancreas. Bothtypes of autoimmune pancreatitis respond to corticoste-roid therapy, although relapse can occur and require otherimmunosuppressive therapies. Corticosteroid therapy, ifpromptly initiated, may prevent the development of exo-crine or endocrine insufficiency.

For patients who do not respond to medical therapy,options include endoscopic therapy, nerve block or neu-rolysis, and surgery. Endoscopic therapy aims to removeobstructions (strictures or stones) in the main pancreaticduct. This approach requires careful patient selection anddetailed evaluation of pancreatic duct anatomy. Ductanatomies that are most amenable to endoscopic therapyinclude a dilated main pancreatic duct (usually more than5– 6 mm) with an obstructing stone in the head or aductal stricture in the head. It is not clear what propor-tion of patients with chronic pancreatitis has ductal anat-omy that is amenable to endoscopic therapy but is cer-tainly far less than half. It is essential to rule out cancerbefore treating an isolated pancreatic duct stricture. Inpatients with multiple stones or impacted stones, endo-scopic therapy is more challenging and less effective. Itshould be noted that ductal abnormalities such as dilata-tion and stones do not correlate with symptoms. Further-more, changes in duct diameter after endoscopic therapy(effective duct decompression should theoretically reduceduct diameter) do not correlate with relief of symptoms.

Endoscopic therapy comprises pancreatic and biliarysphincterotomy, stricture dilation and stenting, stone ex-traction, and lithotripsy. Stones and strictures that aretoo far from the ampulla are usually not amenable toendoscopic therapy. Large stones or impacted stones usu-ally require extracorporeal shock wave lithotripsy (ESWL)or intraductal lithotripsy; these techniques apply shockwaves to break up stones.

A number of large retrospective studies have evaluatedthe efficacy of endoscopic therapy for painful chronicpancreatitis. A retrospective analysis of 1000 patientstreated with a variety of techniques, including ESWL,found that endoscopic therapy was technically successfulfor 69% of patients, with lower rates of success associatedwith more complex ductal anatomies.20 After a mean

eriod of 4.9 years, 66% of patients were pain-free, with a

nother 19% reporting mild pain. Approximately 25% ofatients underwent surgery, generally for failure of painelief after endoscopic therapy. A separate retrospectivenalysis of 146 patients found that approximately 50% ofatients who received endoscopic therapy had pain reliefnd that half of the nonresponders required surgery forheir pain.21 Pain was also reduced in approximately 33%

of patients who received only medical therapy.Because these were retrospective studies, which did not

use standardized or validated measures to assess response,they can only estimate the effectiveness of endoscopictherapy. Nonetheless, it appears that endoscopic therapyis technically successful (approximately 80%) for carefullyselected patients with appropriate ductal anatomy andthat 50% to 70% of these patients have some pain relief.

Lithotripsy is necessary as an adjunct to endoscopictherapy for many patients with pancreatic duct stones.Interestingly, a randomized trial compared ESWL alonewith ESWL followed by ERCP in removal of pancreaticduct stones.22 Pain relief was equivalent between groups,

nd there were significant cost savings for the group thatid not receive ERCP. These findings indicate that break-

ng up obstructing stones could be sufficient to improveuctal drainage and relieve pain. However, an alternativexplanation is that lithotripsy reduces pain by a mecha-ism other than fracturing stones, perhaps by changingociception through some effect on intrapancreaticerves.Another endoscopic treatment option is nerve block or

eurolysis with EUS. Neurolysis delivered by EUS guid-nce is safer and more effective than CT-guided tech-iques. EUS-guided delivery of bupivacaine and cortico-teroids to block nerves has limited effectiveness inatients with chronic pancreatitis. Approximately 50% ofatients have pain relief with a duration of a few weeks, sohis approach is not recommended for patients withhronic pancreatitis. It is more effective for patients withain from pancreatic cancer who have not developedensitization. Neurolysis with injection of absolute alco-ol is not recommended for patients with painful chronicancreatitis. Neurolysis with thoracosocopic splanch-icectomy is rarely used for patients with chronic pancre-titis and its effectiveness is limited, similar to the EUS-uided nerve block.

Surgical therapy is considered for patients who haveot responded to medical or endoscopic therapy. Surgery

s appropriate for pain, local complications such as duo-enal or biliary obstruction, and when cancer is suspectedut cannot be excluded in preoperative evaluations. Aariety of surgical options are available. Selecting theppropriate surgical approach requires analysis of pancre-tic ductal anatomy, consideration of local complicationsuch as duodenal or bile duct compression, and assess-

ents of available surgical expertise. The most commonlyerformed procedure is the lateral pancreaticojejunos-omy or modified Puestow operation. This involves aongitudinal incision of the anterior pancreas and pancre-

fTpfmp

psp

a

tTdpa

DIS

EASE

AN

DTH

ERA

PY

OF

PA

NCREA

TIC

DIS

ORD

ER

May 2013 MANAGEMENT OF CHRONIC PANCREATITIS 1287

pancreatic tail to as close to the duodenum as is feasible.Ductal stones are removed and strictures are incised, andthe incision is overlaid with a defunctionalized roux limb.This operation requires a dilated pancreatic duct (usuallymore than 6 mm in diameter). Approximately 80% ofpatients have pain relief immediately after this operation,but only approximately 50% still have pain relief 5 or moreyears after follow-up.

Several surgical approaches involve drainage of the pan-creatic duct along with localized resection of the pancre-atic head, forms of a duodenum-preserving pancreatichead resection (DPPHR) or a pancreaticoduodenectomy(Whipple operation). These procedures are most fre-quently considered for patients with an inflammatorymass of the pancreatic head, often with associated ob-struction of the bile duct or duodenum. Several random-ized controlled trials reported a similar effect on painfollowing these operations but found that the Whippleprocedure was more frequently associated with postoper-ative diabetes and lower quality of life.23,24 There are 3orms of DPPHR: the Frey, Beger, and Berne operations.hese appear to be equally efficacious, and selection de-ends on local surgical expertise. Short-term pain reliefrom these operations appears equivalent to that from the

odified Puestow operation. Long-term pain relief ap-ears to be better.24 Patients who undergo DPPHR fre-

quently develop exocrine and endocrine insufficiency asconsequences of the surgical resection and continued pan-creatitis in the remnant pancreas.

Distal pancreatectomy, with resection of the pancreatictail, is rarely performed but can be appropriate for pa-tients with disease limited to the pancreatic tail (such asductal stricture from trauma, with obstructive chronicpancreatitis in the tail). Patients with a nondilated pan-creatic duct can be treated with a variation on the mod-ified Puestow operation, in which a V-shaped incision ismade in the anterior surface of the pancreas to include thenondilated pancreatic duct; this is overlaid with a Rouxlimb. This procedure, called a V-plasty or Hamburg pro-cedure, is performed mainly in Europe.

Total pancreatectomy is also rarely performed and canbe coupled with autotransplantation of islet cells. In thisprocedure, the resected pancreas is digested, and the isletsare collected and infused into the portal vein; they im-plant in the liver and reduce the severity of diabetes afterpancreatectomy. This operation has been considered a lastresort for patients who have failed to respond to previoussurgical therapies, although it is increasingly consideredfor patients who have not undergone surgery. Pain reliefoccurs in approximately 66% of patients after total pan-createctomy, indicating the complex extrapancreaticmechanisms of pain.25,26 A study reported that 45% of

atients who underwent this procedure were initially in-ulin independent after surgery, although insulin inde-endence decreased over time.25 The risk of diabetes was

inversely related to islet cell yield from the resected pan-creas, and previous pancreatic drainage procedures (such

yield. It is not clear whether this reduction in yield re-sulted from difficulties in digesting the pancreas to obtainislets following the modified Puestow operation or fromthe patients’ stage of disease (more advanced disease withfewer residual surviving islets at the time of total pancre-atectomy). Patients must be carefully selected for totalpancreatectomy with autotransplantation of islets so, forthe moment, it is a last resort.

Two trials have compared endoscopic and surgical ther-apies for pain from chronic pancreatitis. The first in-cluded 72 patients treated with endoscopy (comprisingsphincterotomy, stricture dilation and stenting, and stoneextraction with lithotripsy if necessary) or surgery (simpledrainage or DPPHR and Whipple operations).27 One yearlater, pain relief was equivalent between groups. After 5years, 34% of the surgery group and 15% of the endoscopygroup still had complete relief of pain. Partial relief ofpain was equivalent between groups (approximately 50%).

A trial of 39 patients was stopped early because patientsin the surgery group had significant improvements inoutcome.28 After a median follow-up period of 24months, 75% of patients in the surgery group and 32% inthe endoscopy group had complete or partial relief ofpain. Five years later, 68% of patients in the endoscopygroup underwent surgery, which remained superior inproviding pain relief.29 Costs, quality of life, and pancre-

tic function were equivalent in the follow-up analysis.These trials included patients with ductal anatomies

hat were amenable to endoscopic or surgical therapy.hey primarily included patients with a dilated pancreaticuct and an obstructing stone or stricture. For theseatients, surgery appears to be somewhat more effectivend more durable than endoscopic therapy.30 These re-

sults are appropriate to discuss with patients, many ofwhom will still opt for endoscopic therapy out of a desireto avoid surgery.

The treatment of pain remains the most difficult chal-lenge in managing patients with chronic pancreatitis.17

Medical therapy is appropriate for all patients, and paincan be reduced to a manageable level in many. Endoscopyand surgery are options for select patients. An algorithmoutlining a reasonable approach is presented (Figure 3).

Management of Exocrine InsufficiencyExocrine insufficiency most commonly develops

after 5 to 10 years of chronic pancreatitis but can alsodevelop in patients with other pancreatic disorders (Sup-plementary Table 2). Exocrine insufficiency should besuspected in patients with these conditions who haveclinical features of or laboratory test results that suggestmalabsorption. These include diarrhea, steatorrhea,weight loss, metabolic bone disease, or vitamin or mineraldeficiency. Exocrine insufficiency is not identified in manypatients who have it, and many who are diagnosed withexocrine insufficiency are undertreated. A 72-hour analy-sis of fecal fat concentrations on a high-fat diet is neces-

DISEA

SEA

ND

THER

APY

OF

PA

NCREA

TICD

ISORD

ER

1288 CHRISTOPHER E. FORSMARK GASTROENTEROLOGY Vol. 144, No. 6

Figure 3. Management algorithm for chronic pancreatitis. IPMN, intraductal papillary mucinous neoplasm; QOL, quality of life; SSRI, selective

stt

atacmeB

emenaatuAntpfNteatdca

uot

sampatispw

ZCPP

UV

DIS

EASE

AN

DTH

ERA

PY

OF

PA

NCREA

TIC

DIS

ORD

ER

May 2013 MANAGEMENT OF CHRONIC PANCREATITIS 1289

formed. Low levels of fecal elastase (�200 �g/g stool) orerum trypsin (�20 ng/mL) are usually observed in pa-ients with exocrine insufficiency and can confirm theypical clinical signs of this disorder.

A baseline evaluation of nutritional status is appropri-te when patients begin pancreatic enzyme therapy. Pa-ients’ weight and body mass index should be determined,nd basic laboratory tests should be performed, includingomplete blood counts (with differential), comprehensiveetabolic panel, international normalized ratio, and lev-

ls of albumin, prealbumin, carotene, and vitamin D.one mineral density testing is also appropriate.Treatment of exocrine insufficiency requires pancreatic

nzyme replacement. It is estimated that the healthy hu-an pancreas produces 900,000 USP units of lipase with

ach meal and that approximately 10% of this amount isecessary to achieve relatively normal absorption of fatnd fat-soluble vitamins. In the past, there was confusionbout this estimate, because 2 different systems were usedo measure the lipase content of products (IU and USPnits). Enzymes are sold in the United States by USP unit.pproximately 90,000 USP units of lipase per meal areeeded for appropriate fat absorption. In many patients,he pancreas still produces some lipase, or there are com-ensatory increases in secretion of gastric lipase, so theull 90,000 USP units per meal is always not required.onetheless, studies from several countries have found

hat many patients do not receive sufficient amounts ofnzymes. A number of agents are available (Table 2). Mostre enteric-coated microsphere capsules, but one is inablet form and is not enteric coated. Lipase is sensitive toegradation by acid, so the nonenteric product requiresotreatment with an agent to suppress gastric acid, suchs an H2-blocker or proton pump inhibitor.

Although pancreatic exocrine insufficiency can interferewith digestion of fats, proteins, and carbohydrates, theeffect is most pronounced on fat and fat-soluble vitaminabsorption. It is appropriate to begin therapy with at least40,000 to 50,000 USP units of lipase with each meal andhalf that amount with snacks. The enzymes should betaken during and after the meal; splitting the doses be-tween these times is commonly recommended. The doseof enzymes can be adjusted based on their effects (reduc-tions in diarrhea or steatorrhea, weight gain) and byincreases in levels of fat-soluble vitamins and nutritionalmeasures. If there is evidence of an insufficient response,the dosage can be increased up to 90,000 USP units of

Table 2. Pancreatic Enzyme Products

Product Formulation

enpep Enteric-coated porcinereon Enteric-coated porcineancreaze Enteric-coated porcineertzye Enteric-coated porcine mixed with bicarbonate

granulesltresa Enteric-coated porcine

lipase with each meal and occasionally more. Supplemen-tation with fat-soluble vitamins is also appropriate.

There are quantitative methods to determine the effi-cacy of fat absorption, including a 72-hour fecal fat mea-surement and breath tests, but these are rarely performedoutside of clinical research studies. Pancreatic enzymetherapy can be ineffective because of inadequate doseadministration, asynchrony in delivery of enzymes andfood simultaneously to the intestine (especially in pa-tients who have undergone pancreatic surgery), or acidinactivation of lipase. Non– enteric-coated preparationsrequire coadministration with acid-reducing agents, butthese can also increase the efficacy of enteric-coated prep-arations. Acid-reducing agents allow enteric-coated en-zymes to be released from their pH-sensitive delivery sys-tem more proximally in the duodenum, the site of mostnormal fat and fat-soluble vitamin absorption.31 Contin-ued failure of therapy despite an adequate dose and co-treatment with an H2-blocker or proton pump inhibitor

sually indicates the presence of another cause of diarrhear malabsorption; small intestinal bacterial overgrowth ishe most common.

Management of Endocrine InsufficiencyLike exocrine insufficiency, endocrine insufficiency

is typically a consequence of longstanding chronic pan-creatitis. The relationship between diabetes and chronicpancreatitis is complex. Some patients with earlier stagesof chronic pancreatitis have type 2 diabetes mellitus be-cause of obesity and the metabolic syndrome. Some pa-tients with long-term type 1 diabetes mellitus also developchronic pancreatitis, but little is known about the mech-anisms of this process. Patients with long-term chronicpancreatitis or extensive pancreatic resection can developtype 3 diabetes mellitus,32 characterized by a lack of in-ulin and other counter-regulatory islet hormones, suchs glucagon. These features cause more frequent treat-ent-induced hypoglycemia and low levels of glucagon,

ancreatic polypeptide, and gastric inhibitory peptidend, on rare occasions, ketoacidosis.32 Chronic pancreati-is is a risk factor for pancreatic cancer, and diabetes canncrease this risk.33 The use of an insulin-sensitizing agentuch as metformin may reduce the risk of cancer in theseatients. The exact prevalence of diabetes among patientsith chronic pancreatitis is not known, but diabetes and

Manufacturer Lipase content (USP)/pill or capsule

Aptalis 3000, 5000, 10,000, 15,000, 20,000Abbott 3000, 6000, 12,000, 24,000Ortho-McNeil-Janssen 4200, 10,500, 16,800, 21,000Digestive Care 8000, 16,000

Aptalis 13,800, 20,700, 23,000

t

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

DISEA

SEA

ND

THER

APY

OF

PA

NCREA

TICD

ISORD

ER

1290 CHRISTOPHER E. FORSMARK GASTROENTEROLOGY Vol. 144, No. 6

impaired glucose tolerance are observed in many, depend-ing on the stage of disease.

Diagnosis of diabetes in patients with pancreatitis relieson the same criteria as for all forms of diabetes: fastingplasma glucose level �126 mg/dL, 2-hour oral glucoseolerance test result �200 mg/dL, or hemoglobin A1c

�6.5%. Although there are no screening and surveillanceguidelines for chronic pancreatitis, it is reasonable toperform tests for fasting plasma levels of glucose andhemoglobin A1c, along with the standard glucose toler-ance test, if findings are inconclusive. It is also reasonableto repeat these tests on a yearly basis.

Treatment of diabetes in patients with pancreatitis usu-ally involves referral to an endocrine specialist. There arebenefits to insulin therapy, but treatment-induced hypo-glycemia is a potential disadvantage. Metformin may re-duce the long-term risk of cancer but is often inadequateas a single agent to treat diabetes. Patients with low levelsof insulin and C-peptide and low body mass would prob-ably benefit most from insulin therapy but also have thehighest risk of hypoglycemia.

Many additional complications can develop in patientswith chronic pancreatitis, including osteoporosis, pancre-atic pseudocyst formation, and pancreatic cancer. Al-though these are not discussed in detail, some aspects ofmanagement are outlined in Supplementary Table 3.

ConclusionThe management of patients with chronic pancre-

atitis has undergone significant changes over the pastdecade. Strategies to increase the accuracy of diagnosis,better assess symptoms and complications, and managepatients using multidisciplinary teams have improved thecare of these patients. Consistent delivery of the bestevidence-based care has the potential to improve thehealth of many patients with chronic pancreatitis. Unfor-tunately, we still have many patients who remain symp-tomatic despite our best efforts. Connecting these pa-tients with support groups, such as the National PancreasFoundation (www.pancreasfoundation.org), can increasethe effects of treatment.

Supplementary Material

Note: To access the supplementary materialaccompanying this article, visit the online version ofGastroenterology at www.gastrojournal.org, and at http://dx.doi.org/10.1053/j.gastro.2013.02.008.

References

1. Layer P, Yamamoto H, Kalthoff L, et al. The different courses ofearly- and late-onset idiopathic and alcoholic chronic pancreatitis.Gastroenterology 1994;107:1481–1487.

2. Yadav D, O’Connell M, Papachristou GI. Natural history followingthe first attack of acute pancreatitis. Am J Gastroenterol 2012;107:1096–1103.

3. Whitcomb DC. What is personalized medicine and what should it

4. Sanyal R, Stevens T, Novak E, et al. Secretin-enhanced MRCP:review of technique and application with proposal for quantifica-tion of exocrine function. AJR Am J Roentgenol 2012;198:124–132.

5. Stevens T. Update on the role of endoscopic ultrasound in chronicpancreatitis. Curr Gastroenterol Rep 2011;13:117–122.

6. Conwell DL, Wu BU. Chronic pancreatitis: making the diagnosis.Clin Gastroenterol Hepatol 2012;10:1088–1095.

7. Paolo JA, Kadiyala V, Lee LS, et al. Proteomic analysis (GeLC-MS/MS) of ePFT-collected pancreatic fluid in chronic pancreatitis. JProteome Res 2012;11:1897–1912.

8. Etemad B, Whitcomb DC. Chronic pancreatitis: diagnosis, classi-fication, and new genetic developments. Gastroenterology 2011;120:682–707.

9. Nordback I, Pelli H, Lappalainen-Lehto R, et al. The recurrence ofalcohol-associated pancreatitis can be reduced: a randomizedtrial. Gastroenterology 2009;136:848–855.

0. Mullady DK, Yadav D, Amann ST, et al; NAPS2 Consortium. Typeof pain, pain-associated complications, quality of life, disabilityand resource utilisation in chronic pancreatitis: a prospectivecohort study. Gut 2011;60:77–84.

1. Pasricha PJ. Unraveling the mystery of pain in chronic pancreatitis.Nat Rev Gastroenterol Hepatol 2012;9:140–151.

2. Olesen SS, Bouwense SA, Wilder-Smith OH, et al. Pregabalinreduces pain in patients with chronic pancreatitis in a randomized,controlled trial. Gastroenterology 2011;141:536–543.

3. Lowenfels AB, Maisonneuve P. Defining the role of smoking inchronic pancreatitis. Clin Gastroenterol Hepatol 2011;9:196–197.

4. Winstead NS, Wilcox CM. Clinical trials of pancreatic enzymereplacement for painful chronic pancreatitis—a review. Pancrea-tology 2009;9:344–350.

5. Siriwardena AK, Mason JM, Sheen AJ, et al. Antioxidant therapydoes not reduce pain in patients with chronic pancreatitis: theANTICIPATE study. Gastroenterology 2012;143:655–663.

6. Bhardwaj P, Garg PK, Maulik SK, et al. A randomized controlledtrial of antioxidant supplementation for pain relief in patients withchronic pancreatitis. Gastroenterology 2009;136:149–159.

7. Forsmark CE, Liddle RA. The challenging task of treating painfulchronic pancreatitis. Gastroenterology 2012;143:533–535.

8. Tignor AS, Wu BU, Whitlock TL, et al. High prevalence of low-trauma fracture in chronic pancreatitis. Am J Gastroenterol 2010;105:2680–2686.

9. Sah RP, Chari ST. Autoimmune pancreatitis: an update on classi-fication, diagnosis natural history, and management. Curr Gastro-enterol Rep 2012;14:95–105.

0. Rosch T, Daniel S, Sholz M, et al. Endoscopic treatment of chronicpancreatitis: a multicenter study of 1000 patients with long-termfollow-up. Endoscopy 2002;34:765–771.

1. Clarke B, Slivka A, Tomizawa Y, et al. Endoscopic therapy iseffective for chronic pancreatitis. Clin Gastroenterol Hepatol2012;10:795–802.

2. Dumonceau JM, Costamanga G, Tringali A, et al. Treatment forpainful chronic pancreatitis: extracorporeal shock wave lithotripsyversus endoscopic treatment: a randomized controlled trial. Gut2007;5:545–552.

3. Diener MK, Rahbari NN, Fischer L, et al. Duodenum-preservingpancreatic head resection versus pancreatoduodenectomy forsurgical treatment of chronic pancreatitis: a systematic reviewand meta-analysis. Ann Surg 2008;247:950–961.

4. Buchler MW, Warshaw AL. Resection versus drainage in treatmentof chronic pancreatitis. Gastroenterology 2008;134:1605–1607.

5. Bramis K, Gordon-Weeks AN, Friend PJ, et al. Systematic review oftotal pancreatectomy and islet cell autotransplantation for chronicpancreatitis. Br J Surg 2012;99:761–766.

6. Harris H. Systematic review of total pancreatectomy and islet cellautotransplantation for chronic pancreatitis. (Br J Surg 2012;99:

2

2

3

3

ISEA

SEA

ND

THER

APY

OF

PA

NCREA

TIC

DIS

ORD

ER

May 2013 MANAGEMENT OF CHRONIC PANCREATITIS 1291

27. Dite P, Ruzicka M, Zboril V, et al. A prospective, randomized trialcomparing endoscopic and surgical therapy for chronic pancreati-tis. Endoscopy 2003;35:553–558.

8. Cahen DL, Gouma DJ, Nio Y, et al. Endoscopic versus surgicaldrainage of the pancreatic duct I painful chronic pancreatitis.N Engl J Med 2007;356:676–684.

9. Cahen DL, Gouma DJ, Laramee P, et al. Long-term outcomes ofendoscopic vs surgical drainage of the pancreatic duct in patientswith chronic pancreatitis. Gastroenterology 2011;141:1690–1695.

0. Ahmed Ali U, Pahlplatz JM, Nealon WH, et al. Endoscopic orsurgical intervention for painful chronic pancreatitis. CochraneDatabase Syst Rev 2012;1:CD007884.

1. Dominguez-Munoz JE, Iglesias-Gasrcia J, Iglesias-Rey M, et al.Optimising the therapy of exocrine pancreatic insufficiency by theassociation of a proton pump inhibitor to enteric coated pancre-

32. Cui Y, Andersen DK. Pancreaticogenic diabetes: special consider-ations for management. Pancreatology 2011;11:279–294.

33. Cui Y, Andersen DK. Diabetes and pancreatic cancer. EndocrRelat Cancer 2012;19:F9–F26.

Received October 17, 2012. Accepted February 5, 2013.

Reprint requestsAddress requests for reprints to: Chris E. Forsmark, MD, Division of

Gastroenterology, Hepatology, and Nutrition, University of Florida, Box100214, Room HD 602, 1600 SW Archer Road, Gainesville, Florida32610-0214. e-mail: [email protected]; fax: (352)392-3618.

Conflicts of Interest

D

ostic algorithm for chronic pancreatitis.

1291.e1 CHRISTOPHER E. FORSMARK GASTROENTEROLOGY Vol. 144, No. 6

Supplemental Figure 1. A diagn

May 2013 MANAGEMENT OF CHRONIC PANCREATITIS 1291.e2

Supplementary Table 1. TIGAR-O Classification System

Toxic metabolicAlcoholicTobacco smokingHypercalcemiaHyperlipidemiaChronic renal failure

IdiopathicTropicalCause unknown

GeneticAutosomal dominant

Cationic trypsinogenAutosomal-recessive/modifier genes

CFTR mutationsSPINK1 mutationsChymotrypsin-COthers

AutoimmuneType 1Type 2

Recurrent and severe acute pancreatitisPostnecrotic (severe acute pancreatitis)Vascular diseases/ischemiaPostradiation exposure

ObstructivePancreas divisum (controversial)Sphincter of Oddi dysfunction (controversial)

Supplementary Table 2. Conditions Associated With Pancreatic Exocrine Insufficiency

Conditions Presumed mechanism Comments

Chronic pancreatitis Acinar cell loss or injury May be reversible (ie, corticosteroid treatment ofautoimmune pancreatitis)

Severe acute (necrotizing) pancreatitis Acinar cell loss Transient exocrine insufficiency may developafter severe pancreatitis without necrosis

Shwachman–Diamond syndrome Acinar cell dysfunction Usual presentation in childhood with exocrineinsufficiency, skeletal abnormalities, bonemarrow dysfunction, and short stature

Pancreatic surgery Acinar cell removal, often withasynchronous delivery ofenzymes

Intestinal surgery (eg, gastric bypass) Asynchronous delivery ofenzymes with meal

Duct dysfunction Cystic fibrosis The majority of patients, but not all, arepancreatic insufficient

Duct blockage Ductal adenocarcinoma,intraductal papillarymucinous neoplasm,stricture

Zollinger–Ellison syndrome Enzyme destruction

NOTE. Courtesy of David C. Whitcomb, MD, PhD.

1291.e3 CHRISTOPHER E. FORSMARK GASTROENTEROLOGY Vol. 144, No. 6

Supplementary Table 3. Complications of Chronic Pancreatitis

Complication Usual clinical presentation Management

Pseudocyst May occur at any timeUsual presentation is abdominal pain that is

worse or different from chronic painNeed to differentiate from cystic neoplasms

Make sure the cystic collection does not represent acystic neoplasm (history, EUS)

Do not treat pseudocysts not causing symptoms,regardless of size

Endoscopic and surgical drainage techniques aremost commonly used, depending on anatomy andavailable expertise

Pancreatic cancer The overall lifetime risk is up to 4%, butcertain groups are at increased risk,including those who smoke, havecoexistent diabetes, and have hereditarypancreatitis

Worsening pain, weight loss, or obstructivejaundice

May be difficult to visualize a mass in thebackground of chronic pancreatitis; a combinationof EUS and CT or MRI is most accurate

Duodenal obstruction Most commonly occurs in those with aninflammatory mass in the head of thepancreas

Usually requires surgery with a DPPHR or Whippleoperation

Biliary obstruction Most commonly occurs in those with aninflammatory mass in the head of thepancreas

May respond to temporary endoscopic stenting,using multiple plastic stents of fully covered metalbiliary stents

May require surgery with a DPPHR, Whippleoperation, or choledochojejunostomy

Osteopenia and osteoporosis Seen in those with exocrine insufficiency,with similar risk as in patients withinflammatory bowel disease

Risk increases with disease duration; mayhave associated deficiency of other fat-soluble vitamins

Baseline and periodic assessment of fat-solublevitamin levels and bone density

Routine supplementation with calcium and vitamin Dis appropriate

Gastroparesis A consequence of both chronic pancreatitisand narcotic analgesia

Can mimic some symptoms of chronic pancreatitis,including nausea and abdominal discomfort

Small intestinal bacterial Diarrhea, steatorrhea, bloating; mechanism Can mimic symptoms of exocrine insufficiency with