What this video covers
- Why chronic pancreatitis pain is a nerve problem as much as a gland problem
- Where the celiac plexus sits, and why most pain traffic from the pancreas passes through it — and why blocking it still does not help everyone
- How the block is done under X-ray or CT guidance in a pain clinic, and under endoscopic ultrasound in a GI suite
- What the published evidence really shows for the endoscopic ultrasound-guided version: pain relief in about half of patients at first. But only about one in five was still better at twelve weeks, and about one in thirty at six months
- The honest risks: short-lived diarrhea, low blood pressure, and a brief pain flare after a steroid block. Alcohol neurolysis carries higher risks, rarely serious, including bleeding and rare spinal cord injury
- Where celiac plexus neurolysis fits for pancreatic cancer pain, and how it differs from a block
- MEDICAL DISCLAIMER: This content is for educational purposes only and is not medical advice. It does not substitute for professional diagnosis or treatment. Always consult a licensed healthcare provider regarding your condition. Viewing this video does not establish a doctor-patient relationship.
Chronic pancreatitis pain continues after the enzymes go normal for a reason. Repeated inflammation rewires the gland’s nerves into their own pain source. Meanwhile the spinal cord and brain turn up the volume. That signal travels through the celiac plexus, which is what a celiac plexus block targets.
It sits below the breastbone and bores through to your back. You lean forward over your knees because that is the only position that helps. Eating sets it off, so you eat less, and the weight comes off. There have been hospital stays this year — fluids, nothing by mouth, an opioid pump, discharge. Each one ended the same way. Then one night the lipase comes back normal while you are still gripping the bed rail. And the talk in the room changes. This article explains why pancreatic pain can go on after the gland itself has gone quiet. It shows where that pain signal really travels. And it shows what a celiac plexus block can and cannot do about it.
Repeated inflammation changes two different things
The first change is the one everybody already knows, because imaging shows it. Fibrosis — scar tissue — replaces working pancreatic tissue. The duct scars and narrows. Pressure builds behind the narrow spot. That is the structural story, and it is real.
The second change is in the nerves, and nobody mentions it. Sensory nerves inside the gland swell. Immune cells move into the nerve’s outer sheath. Nerve fibers sprout into inflamed tissue. The nerve stops being a passive wire reporting on the gland. It becomes its own source of signal. Meanwhile the spinal cord and brain turn up the volume on everything coming from that area.
The 2017 international consensus guidelines on pain in chronic pancreatitis, from Drewes and colleagues in Pancreatology, say it plainly. Pain here can come from repeated or chronic inflammation, from local complications, or from nerve-driven processes with matching changes in the nervous system. (Established — international consensus guideline.) Three mechanisms. Most patients who reach a pain clinic have been treated for exactly one of them.
That is why the gland can burn itself out — scar, calcify, and go quiet on every test you own — while the pain goes on. It has moved into the nervous system.
Why does pancreatitis pain go through to your back and get worse after eating?
Pancreatic pain does not mainly travel with the vagus nerve. The vagus carries other traffic from this gland. The pain fibers run with sympathetic nerves into the celiac plexus. This is a dense nerve network deep in the belly, wrapped around the aorta at about the twelfth thoracic and first lumbar vertebrae. From there the signal runs back along the greater, lesser, and least splanchnic nerves to the nerve roots in the mid-back.
That anatomy explains what you feel. The plexus sits deep and behind, in front of the spine rather than under the skin. That is why the pain bores straight through to the back instead of feeling like a surface ache. It is a visceral (organ) signal, and organ signals are hard to pin down. You can point at a region, not a spot. And eating is what asks the gland to work, so meals set it off. That is how food avoidance and weight loss get built into the illness itself.
It also explains why patients find on their own that leaning forward over the knees helps. And it explains why an enzyme panel drawn during a flare can be totally normal. Enzyme levels report on sudden, active (acute) inflammation. They do not report on a rewired nerve.
What the standard approach misses
None of this criticizes emergency care, and nothing here says the medicines are worthless. Fluids, bowel rest, and pain relief are the right answers to an acute attack. Drugs that control symptoms truly reduce suffering. What they do not do is treat a pain source that has moved into the nervous system.
Two failures follow from that gap.
The enzyme-normal trap. When the labs are quiet and the pain is not, the chart often stops asking why the patient hurts. It starts asking what they want. A nerve problem gets recorded as a character problem. That is not a diagnosis. It is the absence of one.
The single-mechanism treatment plan. Say a patient’s pain comes from the nerves, and the whole plan aims at inflammation, or at the duct, or at opioid dose alone. The plan can be done well and still miss.
But before anyone moves a needle toward the aorta, the list of other causes has to be finished. This is where honest gatekeeping happens:
- Abdominal wall nerve entrapment. Tensing the belly wall takes ten seconds. If the tenderness gets worse, the source is more likely the wall than the organs. That sign points; it does not prove. But a plexus block does nothing for it.
- Chronic mesenteric ischemia (poor blood flow to the gut), suggested by pain only after meals with food avoidance.
- Peptic disease, biliary obstruction, and malignancy, which must be excluded.
- Opioid-induced hyperalgesia (more pain caused by opioids) — because sometimes the treatment has become the disease.
What should be tried before a celiac plexus block?
The guidelines place plexus procedures after the earlier steps, not instead of them. (Established — international consensus guideline.) In the Drewes consensus, that means quitting alcohol and smoking, which is strongly advised. Pancreatic enzyme therapy and antioxidants may be helpful as a first treatment. It also means a simple step-by-step increase in pain medicine, and endoscopic treatment when the duct is blocked. Then comes this line, which pain clinics rarely read aloud: surgery should be considered early, and after a maximum of five endoscopic interventions. Nerve-destroying procedures and neuromodulation come last, for hard cases.
Read that order carefully. If you have a widened duct and a surgical option, that talk belongs before a pain-procedure talk, not after it.
Quitting alcohol and tobacco is the step with the most to give. No trial has raced it head-to-head against a needle, so we will not dress it up as if one had. It is also the one treatment a doctor cannot do for you.
Enzyme replacement has to be dosed right and taken with food, not an hour later. The guidelines use the word may about pain, and that word is worth keeping. Enzymes reliably treat poor digestion and weight loss. Their effect on pain varies from trial to trial. (Established for maldigestion; inconsistent for pain.) Do not start, stop, or change any medication without consulting your physician.
What is a celiac plexus block?
Only when pain is refractory after those steps does the plexus become the target.
For a block guided by live X-ray, you lie face down. On the screen the twelfth thoracic and first lumbar vertebrae are found. A needle is walked off the outer edge of the bone on each side and moved toward the target. Nothing therapeutic is injected until contrast confirms spread hugging the aorta on both views, with no uptake into a blood vessel. Then comes numbing medicine, usually with a steroid. Some centers do the same block under CT, trading live imaging for better soft-tissue detail. Gastroenterology reaches the same plexus from the front, through the stomach wall, under endoscopic ultrasound. Same target, different door.
Know what the injection is doing. It interrupts a pathway for a while. It does not repair the gland, and it does not remove the disease. Who is a candidate, how it is done, and what the day involves are on the pancreatitis pain treatment page. The technique is covered further on the celiac plexus block page.
Does a celiac plexus block work for chronic pancreatitis pain?
This is the part of the conversation that usually goes missing.
The best pooled data are for the endoscopic version, and they are honest about the ceiling. In 2025, Machicado and colleagues pooled eleven studies and 729 patients in Pancreatology. All got an endoscopic ultrasound-guided celiac plexus block for painful chronic pancreatitis. Overall it worked in 53 percent (95% CI 39–67%). Relief lasted 81 days on average (95% CI 30–217). Over time, 52 percent were better at four weeks, 57 percent at eight weeks, 19 percent at twelve weeks, and 3 percent at twenty-six weeks. Side effects occurred in 8 percent (95% CI 4–15%). The authors concluded that a sham-controlled clinical trial is still needed to prove it relieves pain. (Low-certainty for efficacy: pooled uncontrolled data, no sham comparison.)
Three limits belong right beside those numbers. First, that is a window, not a repair. The drop from 57 percent to 3 percent across eighteen weeks is the real shape of the benefit. Second, the 81-day figure is an average, and its confidence interval (the likely span of the true average) runs from 30 days to 217. That is too wide to plan a life around. Nobody should be promised it. Third, those figures belong to the endoscopic technique. They should not be quoted as if they described the X-ray-guided block done in a pain clinic. How long that block lasts has not been measured with the same care.
In pancreatic cancer the procedure is different, and so is the evidence. There the treatment is alcohol neurolysis, which destroys the plexus instead of interrupting it. The 2011 Cochrane review by Arcidiacono and colleagues pooled six randomized trials of the through-the-skin approach in 358 people. At four weeks the pain difference was −0.42 points on a 10-point scale (95% CI −0.70 to −0.13). At eight weeks it was −0.44 (95% CI −0.89 to −0.01). Four tenths of a point is below what most patients would notice. The review’s own authors called the evidence for better pain relief minimal. What clearly differed was opioid consumption, significantly lower in the neurolysis group. The authors saw that as the more meaningful finding. (Consistent across the six pooled trials for the opioid-sparing effect; the authors’ own word for the pain-score difference is minimal.)
What are the risks of a celiac plexus block?
Risk belongs before consent. And the steroid block and the nerve-destroying procedure must not be blurred together.
After a steroid block, you may get short-lived diarrhea, a drop in blood pressure when you stand, and a brief flare of pain before any relief. In the pooled endoscopic series above, side effects ran 8 percent.
Alcohol neurolysis carries a heavier and older set of numbers. Eisenberg, Carr, and Chalmers pooled 1,145 cancer patients who got a neurolytic celiac plexus block, in their 1995 meta-analysis in Anesthesia and Analgesia. Local pain occurred in 96 percent, diarrhea in 44 percent, and low blood pressure in 38 percent, with complications in 2 percent. The authors concluded severe side effects are uncommon. That paper does not report paraplegia at all, so the paraplegia figure comes from somewhere else. Davies ran a 1993 survey in the Journal of the Royal Society of Medicine. It covered 2,730 neurolytic blocks done in England and Wales between 1986 and 1990. He found four permanent paraplegias — roughly one per 683 blocks. Those are neurolysis figures from the 1990s, and in the Eisenberg data, a cancer population. They do not describe the steroid block. That number is small. It is not zero.
Route hazards apply to any approach to this target, because the plexus sits against the aorta. That is exactly what the contrast run is for. The hazards are injection into a blood vessel and numbing-drug toxicity, bleeding deep in the belly, injury to a kidney or a vessel, a collapsed lung (uncommon but possible), and infection. If you take a blood thinner, that is planned for before a date is set, not on the table. Do not start, stop, or change any medication without consulting your physician.
A word on metabolic and adjunctive approaches — here, more restrictive
Other topics in this series discuss metabolic plans. This one is on purpose more strict, not less.
A high-fat ketogenic diet is a poor fit for a gland that can no longer digest fat. It buys steatorrhea (greasy, loose stools), not relief. Pro-resolving mediators and near-infrared light are untested in this condition: there is no controlled human outcome data in chronic pancreatitis. What matters for nutrition here is dull and specific. Dose enzymes correctly, check fat-soluble vitamins, and eat enough calories to stop the weight loss. A theory will not be dressed up as standard care.
An honest picture of a good outcome
What follows is a composite — a picture built from many patients with this condition, not one person’s chart. It is something seen in practice, not a trial result. Nothing in it should be read as a result you can expect.
An adult with calcified chronic pancreatitis on imaging. Six years of pain. Four hospital stays in eighteen months. An opioid dose tripled without producing a good day.
The first eight weeks held no procedure at all. The enzyme dose turned out to be wrong — about half the needed dose, taken after meals instead of with them. Once fixed, the poor digestion and weight loss improved. How much of the pain change came from the enzymes cannot honestly be sorted out. Tobacco was the hard part. There were two relapses before quitting held.
The first X-ray-guided block caused two days of worse pain. Then came relief that was real but incomplete. It was enough to finish a meal, sleep flat, and begin tapering. It held into the third month. That number needs caution. In the pooled endoscopic data, only about one in five is still better at twelve weeks. And how long the X-ray-guided block lasts has not been carefully measured. Hear it as one possible course, not a figure to plan around. A second block did about the same. No published evidence predicts what a third would do.
What changed was the ratio — fewer emergency visits, a lower dose, weight regained. That is not a cure. The disease is still there. What was bought was function, and enough quiet for the slow work no needle will do. Again, this is something seen in practice, not a trial result. Individual results vary.
Frequently asked questions
Why does my pancreatitis pain continue when my lipase and CT scans look normal?
Because pancreatic pain does not reliably follow pancreatic inflammation. International consensus guidelines describe three separate causes: inflammation, local complications, and nerve-driven changes in the nervous system. Sensory nerves in the gland can swell, fill with immune cells, and sprout. The spinal cord and brain can turn up the volume on that input. Once that happens, the pain can last after the gland has scarred, calcified, and gone quiet on every test. Normal enzymes rule out an acute flare. They do not rule out pain.
How long does a celiac plexus block last?
The fullest data come from the endoscopic ultrasound-guided version. Across eleven pooled studies and 729 patients with chronic pancreatitis, it worked in about 53 percent (95% CI 39–67%). Relief lasted near 81 days on average. But the confidence interval around that average (the likely span of the true average) runs from 30 days to 217, so it is not a number to plan around. Over time, 57 percent were better at eight weeks, 19 percent at twelve weeks, and 3 percent at twenty-six weeks. Those pooled studies had no sham control. The authors said plainly that a sham-controlled trial is still needed. How long the X-ray-guided block in a pain clinic lasts has not been measured as carefully. So be wary of any confident single number. This is a window, not a repair. Individual results vary.
Is this the same procedure people get for pancreatic cancer pain?
No, and the difference matters. For chronic pancreatitis the usual procedure is a block — numbing medicine, usually with a steroid. It interrupts the pathway for a while. For pancreatic cancer pain the procedure is usually alcohol neurolysis, which destroys the plexus. Their risks are not the same. A Cochrane review of through-the-skin neurolysis in cancer found only a 0.42-point difference on a 10-point pain scale at four weeks. The review’s own authors called that evidence minimal. Opioid use was significantly lower. Older neurolysis reports found local pain in 96 percent, diarrhea in 44 percent, low blood pressure in 38 percent, and complications in 2 percent. Permanent paraplegia occurred in roughly one case per 683 blocks. Those neurolysis numbers should never be applied to a steroid block.
Will pancreatic enzymes fix the pain?
They may help, but honestly, their effect on pain varies. Guidelines say enzyme therapy may be helpful as a first treatment, and that word is doing real work. What enzymes reliably treat is poor digestion and weight loss. To have any chance of working, they must be dosed right and taken with food, not an hour after. That is a common mistake, and it can be fixed. Do not start, stop, or change any medication without consulting your physician, and never adjust an opioid regimen on your own.
Am I a candidate for a plexus block right now?
Maybe not yet, and that answer is a service, not a brush-off. The guidelines set the order. First come quitting alcohol and tobacco, enough enzyme therapy, and step-by-step pain medicine. When the duct is blocked, endoscopic or surgical relief of the blockage also comes before a plexus procedure. Surgery should be considered early, and after no more than five endoscopic interventions. Other causes also have to be ruled out first. These include abdominal wall nerve entrapment, chronic mesenteric ischemia, peptic disease, biliary obstruction, cancer, and opioid-induced hyperalgesia. Individual results vary.
What does chronic pancreatitis pain feel like?
It usually sits below the breastbone and bores straight through to the back. It is an organ signal, so you can point to a region but not a spot. Eating asks the gland to work, so meals set it off. Many people eat less and lose weight. Most patients find on their own that leaning forward over the knees helps. An enzyme panel drawn during a flare can still come back normal.
What is the recommended diet for chronic pancreatitis?
The nutrition that matters here is dull and specific. Take pancreatic enzymes in the right dose, with food, not an hour later. Check fat-soluble vitamins. Eat enough calories to stop the weight loss. A high-fat ketogenic diet is a poor fit for a gland that can no longer digest fat. It buys steatorrhea, not relief. Stopping alcohol and tobacco is the step with the most to give.
What is the treatment for chronic pancreatitis pain?
The guidelines set an order. First comes stopping alcohol and smoking. Then come enough enzyme therapy, a step-by-step increase in pain medicine, and endoscopic treatment when the duct is blocked. Surgery should be considered early, and after no more than five endoscopic interventions. A celiac plexus block comes only when pain still does not let up after those steps. It interrupts the pain pathway for a window. It does not repair the gland.
Where is this evaluated, and how do I get seen?
Padda Institute Center for Interventional Pain Management is at 4477 Woodson Rd, Suite 100, St. Louis, MO 63134, right next to St. Louis Lambert International Airport. A second location is at 12174 Natural Bridge Rd, St. Louis, MO 63044. The practice serves the St. Louis region across Missouri and Illinois. Call (314) 481-5000 or text (314) 886-5902, Monday through Friday, 8:00 AM to 5:00 PM. Bring your imaging, your enzyme dosing, and your hospital history. The evaluation depends on all three.
Key takeaways
- Chronic pancreatitis pain has three possible sources: inflammation, local complications, and nerve changes in the nervous system. Most patients have been treated for only one.
- Pancreatic pain fibers travel with sympathetic nerves through the celiac plexus in front of the aorta. That is why the pain bores through to the back, gets worse with eating, and can last after the gland goes quiet.
- A celiac plexus block interrupts that pathway for a while. Pooled endoscopic ultrasound-guided data show it works in roughly 53 percent, with 81 days of relief on average. But the confidence interval on that average runs from 30 to 217 days. Only 19 percent were still better at twelve weeks, and no sham-controlled trial has been done yet.
- Quitting alcohol and tobacco, enough enzyme dosing, step-by-step pain medicine, and opening a blocked duct come first. Guidelines also say surgery should be considered early.
- Normal enzymes with ongoing pain do not prove that nothing is wrong. They are a reason to look at the nerve, not at the patient’s character.
This article is educational and is not a substitute for evaluation, diagnosis, or treatment by a physician. Individual results vary. Do not start, stop, or change any medication without consulting your physician. To have your abdominal pain properly evaluated before any procedure is scheduled, call (314) 481-5000 or text (314) 886-5902.
References
- Machicado JD, Tanner S, Adoor D, Chalhoub JM, Plann-Curley B, Lee UJ, Bang JY, Varadarajulu S, Wilcox CM; US Pancreatic Disease Study Group. Endoscopic ultrasound-guided celiac plexus block for painful chronic pancreatitis: A systematic review and meta-analysis. Pancreatology. 2025 Sep;25(6):860-867. PMID 40813224. PubMed
- Drewes AM, Bouwense SAW, Campbell CM, Ceyhan GO, Delhaye M, Demir IE, Garg PK, van Goor H, Halloran C, Isaji S, Neoptolemos JP, Olesen SS, Palermo T, Pasricha PJ, Sheel A, Shimosegawa T, Szigethy E, Whitcomb DC, Yadav D; Working group for the International (IAP-APA-JPS-EPC) Consensus Guidelines for Chronic Pancreatitis. Guidelines for the understanding and management of pain in chronic pancreatitis. Pancreatology. 2017 Sep-Oct;17(5):720-731. PMID 28734722. PubMed
- Arcidiacono PG, Calori G, Carrara S, McNicol ED, Testoni PA. Celiac plexus block for pancreatic cancer pain in adults. Cochrane Database Syst Rev. 2011 Mar 16;2011(3):CD007519. PMID 21412903. PubMed
- Eisenberg E, Carr DB, Chalmers TC. Neurolytic celiac plexus block for treatment of cancer pain: a meta-analysis. Anesth Analg. 1995 Feb;80(2):290-295. PMID 7818115. PubMed
- Davies DD. Incidence of major complications of neurolytic coeliac plexus block. J R Soc Med. 1993 May;86(5):264-266. PMID 8505748. PubMed
Get the diagnosis before you accept the procedure
Bring your imaging and your history to the Padda Institute Center for Interventional Pain Management in St. Louis. We will tell you which structure is truly causing your pain — and what the evidence does and does not support.
Or call or text us.
Medically reviewed by Dr. Gurpreet Singh Padda, MD, MBA, MHP — Board Certified in Anesthesiology, Pain Medicine, Interventional Pain Management, Addiction Medicine, and Obesity Medicine. Last reviewed July 2026.
Dr. Gurpreet Singh Padda, MD, MBA, MHP
