Pain management ladder — from mechanism assessment through medical, endoscopic, surgical, and neurolytic therapy
Chronic pancreatitis pain is notoriously difficult to treat because it arises from at least two distinct, often coexisting mechanisms: mechanical ductal hypertension from obstructing stones/strictures, and neuropathic/centrally sensitized pain from progressive perineural inflammation and remodeling of pain-processing pathways in the spinal cord and brain. Correctly weighting these two contributors at the outset determines whether the patient is more likely to benefit from duct-directed intervention or from a primarily pharmacologic/neuromodulatory strategy.
Chronic pancreatitis pain does not have a single unifying pathophysiology, which is precisely why a one-size-fits-all treatment fails so often. Two broad, overlapping mechanisms are now recognized:
1. Ductal/parenchymal hypertension (the "plumbing" problem): recurrent inflammation leads to protein plugs, calcifications, and stone formation within the main pancreatic duct, as well as strictures at sites of prior necrosis or inflammation. These obstruct outflow of pancreatic secretions, raising intraductal and interstitial pressure — analogous to compartment syndrome within the gland. Pain from this mechanism is often described as postprandial (worsened by eating, which stimulates secretion into an obstructed system), and classically improves — sometimes dramatically — with mechanical decompression, whether endoscopic or surgical.
2. Neuropathic / centrally sensitized pain (the "wiring" problem): chronic inflammation damages intrapancreatic sensory nerves, which become hypertrophied and infiltrated by inflammatory cells (a histologic hallmark of chronic pancreatitis). Repeated nociceptive barrage from the pancreas drives neuroplastic changes in the dorsal horn and higher CNS pain centers — central sensitization — such that pain can become self-sustaining and persist even after the original ductal pathology has "burned out" (extensive fibrosis/atrophy with exocrine and endocrine failure). This explains the clinically frustrating observation that roughly half of patients with end-stage pancreatic atrophy continue to have significant pain despite an anatomically "quiescent" gland.
In practice, most patients have a mixture of both mechanisms, and the assessment task is to judge which one currently dominates — using imaging (duct diameter and morphology on CT/MRCP/EUS), the character and timing of pain, and the trajectory of prior treatment response — to select the most appropriately targeted next step.
Cross-sectional imaging is central to mechanism assessment and to planning any duct-directed therapy:
• CT pancreas protocol: identifies calcifications, gland atrophy, main duct dilation, and pseudocysts; also screens for pancreatic malignancy, which must always be excluded in a patient with new or changing pain patterns • MRCP (magnetic resonance cholangiopancreatography), often with secretin stimulation: the best non-invasive test to characterize ductal anatomy — duct diameter, number and location of strictures, side-branch ectasia — and to assess exocrine reserve via the secretin-stimulated duodenal fill grade • Endoscopic ultrasound (EUS): most sensitive test for early parenchymal and ductal changes (using the Rosemont criteria), and allows tissue sampling if a mass is identified • Main pancreatic duct diameter is the key anatomic threshold driving the decompression-vs-neuromodulation decision tree: a duct ≥6–7mm strongly favors a ductal hypertension mechanism amenable to endoscopic or surgical drainage, while a normal-caliber or minimally dilated duct in the setting of extensive parenchymal fibrosis points toward a neuropathic-dominant picture better addressed pharmacologically or with neurolysis.
Regardless of which mechanism dominates, medical management is the foundation upon which every other therapy is built. Complete cessation of alcohol and tobacco is the single most effective disease-modifying intervention available, slowing the rate of further calcification and functional loss, while a structured analgesic approach — deliberately incorporating neuropathic agents earlier than in typical acute pain algorithms — addresses the centralized component of pain from the outset.
Alcohol is the leading cause of chronic pancreatitis in most series, and continued drinking accelerates the calcification, fibrosis, and pain trajectory of the disease. Complete abstinence — not merely reduction — is associated with slower disease progression, fewer acute exacerbations superimposed on chronic disease, and in some studies, measurably improved pain scores over time. Structured counseling, and where needed, pharmacologic support for alcohol use disorder (naltrexone, acamprosate) and referral to addiction services, should be offered at every visit, since abstinence rates achieved through brief advice alone are low.
Smoking is now recognized as an independent risk factor for chronic pancreatitis progression and calcification, separate from and additive to alcohol — smokers with chronic pancreatitis progress to calcific disease and exocrine/endocrine insufficiency significantly faster than non-smokers with equivalent alcohol exposure. Smoking cessation should be pursued with the same urgency as alcohol cessation, using nicotine replacement, varenicline, or bupropion as needed.
A modified WHO-style analgesic ladder is used, but with an important departure from typical acute pain management: neuropathic agents are introduced early rather than reserved for refractory cases, because central sensitization is common even in earlier disease stages.
Step 1 — Non-opioid analgesia: acetaminophen and, cautiously, NSAIDs (mindful of GI and renal risk in a population with nutritional compromise) form the base layer.
Step 2 — Neuropathic/adjuvant agents, introduced early: pregabalin and gabapentin have the best evidence in this population. The landmark RCT by Olesen et al. (Gastroenterology, 2011) showed pregabalin significantly reduced pain scores in chronic pancreatitis compared with placebo, supporting a mechanistic rationale (targeting central sensitization) rather than simply symptomatic control. Tricyclic antidepressants (nortriptyline) and SNRIs (duloxetine) are also used, particularly when mood/sleep disturbance coexists, which is common in chronic pain populations.
Step 3 — Opioids: reserved for breakthrough or severe pain not controlled by the above, used at the lowest effective dose with close monitoring, given the well-documented risks of tolerance, dependence, opioid-induced hyperalgesia, and constipation/motility effects that can worsen the malabsorptive symptoms already present in many patients. Chronic high-dose opioid use is itself associated with worse long-term outcomes and should prompt reassessment for a duct-directed or neurolytic intervention rather than simple dose escalation.
Pancreatic enzyme replacement therapy, while primarily indicated for malabsorption, is also given a trial in some pain-predominant patients under the "pancreatic rest" hypothesis (feedback inhibition of pancreatic secretion), though evidence for pain benefit specifically is weaker and inconsistent across trials.
Because chronic pancreatitis pain has a strong neuropathic/centrally sensitized component in many patients, treating it purely as nociceptive pain — escalating opioids without addressing central sensitization — is a common and consequential error. Early pregabalin/gabapentin use is now considered standard of care alongside lifestyle modification.
When ductal hypertension is judged to be the dominant pain mechanism — a dilated main pancreatic duct with obstructing stones or a dominant stricture — endoscopic therapy offers a lower-risk first-line decompressive option before surgery is considered. ERCP with pancreatic sphincterotomy, stone extraction (often combined with extracorporeal shock wave lithotripsy for large or impacted stones), and stricture stenting can meaningfully relieve pain in appropriately selected patients.
Endoscopic pancreatic duct decompression follows a stepwise technique:
1. Pancreatic sphincterotomy: cutting the pancreatic sphincter (analogous to biliary sphincterotomy) to enlarge the outflow orifice, facilitating both stone extraction and stent placement. Carries a meaningful risk of post-ERCP pancreatitis (higher than biliary sphincterotomy), so is performed by experienced pancreaticobiliary endoscopists with periprocedural rectal indomethacin and often pancreatic duct stent placement for prophylaxis.
2. Stone extraction: small, mobile stones can be extracted directly with balloon or basket catheters after sphincterotomy. Large (>5mm), impacted, or densely calcified stones typically cannot be extracted directly and require fragmentation first.
3. Extracorporeal shock wave lithotripsy (ESWL): focused shock waves delivered from outside the body fragment large pancreatic duct stones into pieces small enough for endoscopic extraction or spontaneous passage. This is considered first-line for large obstructing stones (typically performed before or in conjunction with ERCP) and has good published success rates for achieving stone clearance and pain improvement when duct anatomy is favorable.
4. Stricture management: dominant strictures are treated with balloon dilation followed by placement of one or multiple plastic stents (single large-bore or multiple side-by-side stents), typically exchanged every 3 months over 6–12 months to achieve durable stricture resolution. Fully covered self-expanding metal stents are used selectively in refractory dominant strictures.
Endoscopic therapy works best in a specific phenotype: dominant main pancreatic duct obstruction from stones or a single dominant stricture, in patients without extensive parenchymal calcification/atrophy or multiple strictures throughout the duct ("chain of lakes" morphology), and without a markedly dilated duct that would better be served by a larger surgical anastomosis.
Limitations: • Multiple ERCP sessions are typically required to complete stone clearance and stricture dilation/stenting, and repeat procedures carry cumulative risk of post-ERCP pancreatitis, bleeding, and perforation • Durability is a genuine concern: while initial response rates are good (60–80% meaningful pain improvement), a substantial proportion of patients experience symptom recurrence over subsequent years and require further endoscopic sessions or eventual surgery • Diffuse, multifocal ductal disease with a "chain of lakes" pattern (multiple strictures alternating with dilated segments) is generally poorly suited to endoscopic single-point decompression and is better served by a longitudinal surgical drainage procedure that decompresses the entire length of the duct in one operation
This sets up the central comparative question addressed by trials like ESCAPE (Stage 4): for patients who are reasonable candidates for either approach, does starting with endoscopy versus proceeding directly to surgery produce better long-term outcomes?
When endoscopic therapy fails, is technically not feasible, or when the ductal disease is extensive enough that surgery is expected to be more durable up front, surgical options are considered. The choice of operation depends on whether the disease is duct-dominant with a widely dilated duct amenable to longitudinal decompression, or head-dominant with an inflammatory mass requiring targeted resection — and increasingly, evidence supports offering surgery earlier rather than only after endoscopy has been exhausted.
The modified Puestow procedure (Partington-Rochelle modification) is the classic drainage operation for duct-dominant chronic pancreatitis with a widely dilated main pancreatic duct (generally >7mm), often with a "chain of lakes" pattern of alternating dilations and strictures along its length:
• The main pancreatic duct is opened longitudinally along its entire length, from the tail to as close to the head as safely possible • A Roux-en-Y limb of jejunum is anastomosed side-to-side along this opened duct, creating a wide, low-pressure drainage channel for pancreatic secretions directly into the bowel • This decompresses the entire ductal system in a single longitudinal anastomosis, addressing multifocal strictures that endoscopic single-point stenting cannot adequately treat • Pancreatic parenchyma is preserved (no resection), making this the least destructive surgical option when duct anatomy is favorable • Pain relief rates are good (approximately 70–80% short-term), though — similar to endotherapy — durability can wane over years as underlying parenchymal disease progresses
A substantial proportion of chronic pancreatitis patients have disease concentrated in the pancreatic head, which is thought to act as a "pacemaker" driving pain and downstream ductal obstruction — analogous to how an inflamed appendix drives symptoms disproportionate to its size. For these patients, targeted head resection (with duct decompression) outperforms simple duct drainage alone:
• Frey procedure: a local resection (coring out) of the pancreatic head combined with a longitudinal pancreaticojejunostomy along the opened duct — effectively combining head decompression with Puestow-style ductal drainage in one operation, without a formal duodenum-preserving pancreatic head resection. It preserves the duodenum, bile duct, and stomach, avoiding the complexity of a Whipple-type resection.
• Beger procedure: a more extensive duodenum-preserving pancreatic head resection, in which the pancreatic head is subtotally excised (leaving a thin rim along the duodenum and bile duct) and reconstructed with a Roux-en-Y jejunal limb. Technically more demanding than Frey but may be preferred for larger inflammatory head masses or when biliary/duodenal involvement is present.
Both procedures achieve pain relief rates in the range of 80–90% at 1–2 years in appropriately selected patients, with Frey generally preferred for its technical simplicity and lower morbidity when a formal resection is not strictly necessary.
The ESCAPE trial (Issa et al., JAMA 2020) directly addressed a long-standing controversy: for patients with dilated-duct chronic pancreatitis who are candidates for either approach, is it better to start with endoscopy (reserving surgery for endoscopic failures) or to proceed directly to surgery?
This multicenter Dutch RCT randomized 88 patients with chronic pancreatitis, a dilated main pancreatic duct, and recent-onset pain to either early surgical intervention or an optimized step-up endoscopic approach (with surgery reserved for endoscopic failure). At 18 months, complete or partial pain relief was achieved in approximately 58% of the surgery-first group compared with approximately 39% of the endoscopy-first group — a clinically and statistically meaningful difference favoring early surgery, achieved without an increase in major complications or mortality.
This trial, building on earlier similar findings (the Dutch ESCAPE and earlier trials by Cahen et al.), has shifted practice meaningfully: rather than viewing surgery strictly as a last resort after endoscopy has failed, current guidelines increasingly support considering surgery as a reasonable — sometimes preferred — first-line option in patients with a dilated duct and a relatively short symptom duration, since earlier surgical decompression may also help preserve remaining pancreatic function by halting the cycle of obstruction-driven parenchymal injury.
The ESCAPE trial reframed the endoscopy-vs-surgery debate: rather than "surgery only after endoscopy fails," current evidence supports discussing early surgical drainage as a first-line option in duct-dominant chronic pancreatitis, since delaying effective decompression may itself contribute to a lower chance of durable pain relief.
For patients whose pain persists despite duct-directed therapy, or whose pain is predominantly neuropathic/centrally driven rather than mechanical, celiac plexus intervention offers a way to directly interrupt the visceral afferent pathway carrying pain signals from the pancreas to the spinal cord — either as a temporizing diagnostic/therapeutic block or as a longer-lasting neurolytic procedure.
The celiac plexus is a dense network of sympathetic and visceral afferent nerve fibers surrounding the celiac artery origin at the T12-L1 level, through which the majority of nociceptive signals from the pancreas, stomach, liver, and proximal bowel travel en route to the spinal cord. Interrupting this plexus can substantially reduce the perceived intensity of visceral pancreatic pain, independent of the underlying ductal anatomy.
Two techniques are used:
• Celiac plexus block (CPB): injection of local anesthetic (often combined with corticosteroid) around the celiac plexus, providing temporary pain relief lasting weeks to a few months. Because the effect is reversible, this is often used first as a diagnostic/prognostic test — if a patient responds well to a temporary block, this predicts a higher likelihood of benefit from a more permanent neurolytic procedure or surgical splanchnicectomy.
• Celiac plexus neurolysis (CPN): injection of a neurolytic agent (absolute alcohol or phenol) to chemically destroy the nerve fibers, producing longer-lasting pain relief (typically months, sometimes longer) at the cost of being less easily reversible. This is generally reserved for patients who have already demonstrated response to a diagnostic block.
Endoscopic ultrasound (EUS)-guided delivery has become the preferred approach over the older posterior percutaneous (fluoroscopy or CT-guided) technique, because EUS allows direct real-time visualization of the celiac axis and surrounding structures from within the stomach, improving accuracy and reducing the risk of inadvertent vascular or neural injury.
Celiac plexus intervention is most useful in two overlapping clinical scenarios:
1. As an adjunct in patients who have already undergone appropriate duct-directed therapy (endoscopic or surgical) but continue to have significant residual pain — reflecting the centrally sensitized/neuropathic component of their disease that anatomic decompression alone cannot fully address.
2. As a primary option in patients whose pain is judged, from the outset (Stage 1 assessment), to be predominantly neuropathic rather than ductal — for example, patients with a "burned out" atrophic, minimally dilated pancreas who nonetheless have severe pain, where duct-directed intervention offers little mechanical target.
Efficacy in chronic pancreatitis is more modest and less durable than in the well-established use of celiac plexus neurolysis for pancreatic cancer pain — reflecting the more complex, multifactorial, and centrally reinforced nature of chronic (benign) pancreatitis pain compared with the more purely nociceptive pain of malignant invasion. Response rates for meaningful pain reduction range widely (40–70%) across studies, and relief is typically temporary, often prompting repeat procedures. For carefully selected patients with a robust response to blockade, thoracoscopic bilateral splanchnicectomy (surgically dividing the splanchnic nerves proximal to the celiac plexus) offers a more durable — though still not permanent — surgical alternative, generally reserved for those who have exhausted or are poor candidates for further duct-directed intervention.
Celiac plexus intervention is best understood not as a competitor to duct-directed therapy but as a complementary tool addressing the neuropathic axis of chronic pancreatitis pain — reinforcing that the two mechanisms identified in Stage 1 often require two different, concurrently applied treatment strategies rather than a single sequential algorithm.