General Surgery

ERCP

ERCP is now a treatment rather than a test, and its main risk is pancreatitis in roughly one person in ten. A single suppository cuts that risk by close to half. What the procedure treats, what it should not be used for, and what to ask before you agree.

About This Department

A randomized trial gave 602 high-risk patients a single suppository before their ERCP and cut pancreatitis afterwards from 16.9 percent to 9.2 percent, a result that changed the procedure more than any advance in the instruments has. It illustrates the two things worth understanding about ERCP, which are that it is now a treatment rather than a test, and that its main risk is largely preventable if the people doing it are paying attention.

Free consultation

Ask whether you need an ERCP or a scan first

Send the ultrasound, MRCP or CT reports with the images themselves, your liver blood tests including bilirubin and the alkaline phosphatase, any previous endoscopy report, a note of whether you still have your gallbladder and whether you have had any abdominal surgery including weight loss surgery, your medication list with blood thinners and aspirin named, and whether you have ever had pancreatitis. A specialist reviews the file and tells you whether an ERCP is the right next step, whether a scan should come first, and what would be done during it. No fee, no obligation, and a coordinator replies in your own language, usually within the same working day.

WhatsApp: +90 549 648 55 11 Send your scans and reports +90 549 648 55 11

No longer a test

For thirty years ERCP was how doctors looked inside the bile duct and the pancreatic duct. Two other things now do that job without putting an instrument into either, which are magnetic resonance imaging of the ducts and ultrasound performed from inside the stomach with a special endoscope. Pooling 18 studies covering 2,366 people being investigated for stones in the bile duct, ultrasound from inside gave a sensitivity of 0.95 and a specificity of 0.97, while magnetic resonance gave 0.93 and 0.96, and the two were statistically indistinguishable. Neither carries any risk of pancreatitis.

The consequence is that being sent for a purely diagnostic ERCP in 2026 should prompt a question, because the information is almost always obtainable another way. What has not changed is that ERCP remains the only way to take a stone out of the bile duct, place a stent through a blockage, or open a narrowed drainage channel, all through the mouth and without an incision. So the modern question is never whether ERCP can see the problem, and always whether there is something that needs doing about it. Prediction is imperfect either way. Applying the standard high-probability criteria to 498 patients suspected of having a bile duct stone, only 56.3 percent of those meeting the criteria actually had one, while 34.8 percent of those who did not meet them did, giving an overall accuracy of 62.1 percent.

The complication rate

ERCP is the riskiest endoscopic procedure in routine use, and being told the number in advance is more useful than being reassured. Two large analyses give the shape of it.

Across 21 prospective studies covering 16,855 patients, complications of any kind occurred in 6.85 percent and death in 0.33 percent. Pancreatitis accounted for 3.47 percent, infection 1.44 percent, bleeding 1.34 percent and perforation 0.60 percent. Severe complications made up 1.67 percent of the total.

Researchers approached it differently in a second analysis, pooling the placebo and no-stent arms of 108 randomized trials to get a clean figure for pancreatitis specifically, and across 13,296 such patients it occurred in 9.7 percent, with death in 0.7 percent, and among the 2,345 classified as high risk it reached 14.7 percent. Severe pancreatitis was rare in both analyses, at 0.5 percent overall and 0.8 percent in the high-risk group. The gap between 3.47 and 9.7 percent is explained by what was being counted and in whom, since trial populations are enriched for difficult cases and use stricter definitions. What both agree on is that pancreatitis is the dominant risk, that most of it is mild, and that a small number of people become seriously unwell. Neither figure is a reason to refuse a procedure you need, and both are a reason to ask what the unit does to bring its own number down. The rest of this page is largely about the things that move those numbers.

The trial that changed it

Anti-inflammatory drugs given rectally at the time of the procedure turned out to prevent a large share of the pancreatitis, and the evidence for that is unusually clean.

Six hundred and two high-risk patients were randomized to rectal indomethacin or placebo. Pancreatitis occurred in 9.2 percent against 16.9 percent, and moderate or severe pancreatitis in 4.4 percent against 8.8 percent. A single suppository, given once.

Chinese investigators then asked a sharper question, which was whether the drug should be given to everybody beforehand or only to selected patients afterwards. Randomizing 2,600 people across multiple hospitals, giving it routinely before the procedure produced pancreatitis in 4 percent against 8 percent, a relative risk of 0.47 with a confidence interval of 0.34 to 0.66. The benefit held in the average-risk group as well as the high-risk one, at 3 percent against 6 percent. That average-risk finding is the part most often overlooked, because it means the drug is not reserved for the small group of people who look obviously vulnerable beforehand. That is the finding worth carrying into a consultation, because it means the drug is not only for people who look high risk beforehand, and because a unit that gives it selectively after the event is following an approach that has been tested against the alternative and lost.

Preventing pancreatitis

The small plastic tube

Where the pancreatic duct has been entered or injected, a thin plastic stent left in it keeps the duct draining while it recovers, and it falls out on its own within days. Pooling ten randomized trials covering 1,176 high-risk patients, that stent gave odds of 0.25 for pancreatitis, with a confidence interval of 0.17 to 0.38. The harder and more current question is whether it adds anything on top of the suppository, and a trial across 20 centers randomized 1,950 patients to find out. Pancreatitis occurred in 14.9 percent with the drug alone and 11.3 percent with drug plus stent, a difference of 3.6 percentage points, and the drug alone failed to meet the trial's threshold for being no worse. Serious adverse events, intensive care admissions and length of stay did not differ, and a later analysis suggested the drug alone was actually inferior, though that analysis was done after the trial rather than planned into it. The practical reading is that in genuinely high-risk cases the stent is still worth placing, and that this is a fair thing to ask about beforehand.

The intravenous fluid

Giving generous intravenous fluid around the procedure was an attractive idea and it has not survived proper testing. A small pilot randomized 62 patients and reported no pancreatitis at all in the aggressively hydrated group against 17 percent in the standard one, which was widely quoted and came from 39 patients receiving the intervention. A Dutch trial then randomized 826 patients across 22 hospitals to aggressive fluid plus an anti-inflammatory drug or the drug alone, and found pancreatitis in 8 percent against 9 percent, a relative risk of 0.84 with a confidence interval running from 0.53 to 1.33. Complications related to the fluid itself, intensive care admission and death at 30 days were all no different. Read those two together as an object lesson in why a striking result from 62 patients needs a larger trial before it changes anything, and as a reason not to be impressed by a unit that lists aggressive hydration among its safety measures.

How the risk stacks up

Your personal risk of pancreatitis depends on who you are and on what is done during the procedure, and both halves have been quantified in a prospective study of 1,963 procedures where pancreatitis occurred in 6.7 percent.

Wide table. Swipe or drag it sideways on a narrow screen, because it scrolls instead of shrinking.

What raises the risk of pancreatitis after ERCP, from a study of 1,963 procedures
Factor	Odds	What it means for you
Pancreatitis after a previous ERCP	5.4	The single strongest patient factor. Say so before the procedure rather than assume it is in the notes.
Stretching the sphincter with a balloon	4.5	A technique choice rather than something about you. Worth asking whether it is planned and why.
Difficulty getting into the duct	3.4	Unknowable in advance and heavily influenced by experience. This is where the endoscopist's caseload shows.
Cutting the pancreatic sphincter	3.1	Sometimes necessary. Combined with a protective stent it becomes considerably safer.
Injecting contrast into the pancreatic duct	2.7	Modern practice uses a guidewire instead wherever possible, for exactly this reason.
Suspected sphincter of Oddi dysfunction	2.6	The highest risk indication and, as a later section explains, the one where the procedure was shown not to work.
Being female, and having a normal bilirubin	2.5 and 1.9	Nothing you can change, and both are markers of a duct that was never really obstructed in the first place.
Not having chronic pancreatitis	1.9	Counterintuitive and real. A scarred pancreas is less able to become inflamed than a healthy one.

Getting into the duct

That opening into the bile duct is a few millimeters across and sits at an angle. Everything about the safety of this procedure turns on how the first few minutes at that opening are handled.

It usually works first time, and not always. Across 13,018 procedures by 85 endoscopists, conventional cannulation succeeded in 89.8 percent and overall success reached 95.6 percent once other techniques were used, while a separate national registry reported 93.8 percent, with difficult access encountered in 32.1 percent of untouched papillae.
A guidewire beats injecting dye. Pooling 12 randomized trials covering 3,450 participants, leading with a wire rather than contrast reduced pancreatitis with a relative risk of 0.51, improved first-attempt success and reduced the need for a cutting technique. The benefit was clearest in trials that did not let endoscopists switch methods midway.
Cutting early is safer than persisting. Where access is proving difficult, making a small cut to reach the duct sooner rather than continuing to probe reduced pancreatitis across seven randomized trials, with a relative risk of 0.57. Overall success was unchanged. Bleeding and perforation were no different, though the analysis could not confirm those secondary results.
Volume shows up here more than anywhere. In that 13,018-procedure network, endoscopists performing more than 239 a year had odds of 2.79 for overall success. In an older prospective study, those doing more than two a week achieved 96.5 percent bile duct cannulation against 91.5 percent.
Difficulty is not the same as danger, and it is related. Difficult cannulation carried odds of 3.4 for pancreatitis in that study, and nobody can predict it beforehand, which is the whole argument for having the procedure somewhere that handles difficulty routinely rather than somewhere that rarely meets it.
What to ask. How many ERCPs the endoscopist performs in a year, what their cannulation success rate is, and what they do when access proves difficult. Three short questions with three factual answers, and a unit that answers all three without hesitating has told you a good deal more about itself than any brochure will.

What happens on the day

Thirty minutes to over an hour, and you will remember very little of it.

You arrive fasted, and you are sedated. Deep sedation or a general anesthetic depending on the unit and on you. Cardiopulmonary events related to sedation accounted for 1.33 percent of complications across 16,855 patients, with fatal events in 0.07 percent, which makes the anesthetic assessment worth taking seriously.

The suppository goes in at the start. Rectal indomethacin or diclofenac, given before or at the beginning rather than afterwards, on the evidence in the section above. If nobody mentions it, ask.

A side-viewing endoscope goes down to the duodenum. Through the mouth, past the stomach, to the small opening where the bile and pancreatic ducts arrive. Nothing is cut on the outside and there is no wound, which is why people underestimate this procedure and why the observation period afterwards catches them by surprise.

A guidewire is threaded into the bile duct under X-ray. This is the critical step and the one described in the previous section. Contrast is injected only once the wire is where it should be.

The sphincter is usually cut, and then the work is done. A small electrical cut widens the opening so instruments can pass. Stones are pulled out with a balloon or a basket, a stent is pushed through a narrowing, or a leaking duct is decompressed.

You wake up and are watched. Most people go home the same day or the following morning. The observation is for pancreatitis, which declares itself within hours rather than days, and the next section explains what that feels like.

The other complications

Bleeding from the cut

Cutting the sphincter divides small blood vessels that cannot be seen until they open, and bleeding that matters clinically occurred in 2.0 percent of 2,347 patients in the landmark study of this procedure, within an overall complication rate of 9.8 percent. Five things predicted trouble in that study, which were a suspected sphincter problem as the indication, cirrhosis, difficult cannulation, needing a cutting technique to gain access, and combining the procedure with a route through the skin. Aspirin is the question patients ask about most often. A study of 804 patients found bleeding in 9.7 percent of those who continued aspirin, 9.5 percent of those who stopped it a week beforehand and 3.9 percent of those who had never taken it, with no difference between continuing and stopping. That was a retrospective study, so treat it as a reason to discuss your own blood thinners specifically rather than as permission to stop or continue anything.

Perforation and infection

Making a hole is the complication people fear most and it is the rarest, at 0.60 percent across 16,855 patients, and what matters more than the rate is where the hole is and how quickly it is found. In a series of 6,040 procedures, 40 perforations were classified into three kinds, and those made by a guidewire, 14 cases, were all recognized immediately and treated without surgery, with an average stay of three and a half days. Those beside the ampulla, 22 cases, were mostly drained endoscopically and none needed an operation when found early. The four that involved the duodenal wall itself all needed surgery. Infection is the other one, with cholangitis affecting 8.8 percent of 227 patients in one series where older age, low albumin and removing a stent were the associated factors, and infections overall accounting for 1.44 percent in the larger pooled analysis. Fever with shaking chills after an ERCP is a same-day problem rather than a next-week one.

Bile duct stones

This is what most ERCPs are done for, and there is a genuine choice about how to combine it with removing the gallbladder.

Narrow screens scroll this table sideways. Drag or swipe it to reach the last column.

Ways of dealing with stones in the bile duct alongside gallbladder removal
Approach	What the randomized evidence shows	How to read it
ERCP first, gallbladder out afterwards	Compared against a single operation clearing both, across 285 and 295 patients, death, complications and retained stones were all statistically indistinguishable.	The commonest approach worldwide, and a legitimate one. It involves two procedures and two anesthetics.
One operation clearing both	Pooling 8 randomized trials and 1,130 patients, stone clearance favored the single operation with odds of 1.56, hospital stay was about a day shorter and operating time 17 minutes shorter.	Complications and deaths were no different. It requires a surgeon trained in exploring the bile duct, which not every unit has.
Gallbladder out first, ERCP afterwards	Retained stones occurred in 25 percent of that group against 9 percent where one operation cleared both, with odds of 0.28.	The weakest of the three sequences on the evidence, and it is still sometimes chosen for practical reasons.
Open surgery instead of ERCP	Retained stones were less common with open surgery, at 6 percent against 16 percent, with odds of 0.36, while mortality and morbidity did not differ.	Historical rather than current practice. It is quoted here because it shows ERCP leaves more stones behind than an operation does.
How soon the gallbladder comes out	Randomizing patients to removal within three days of the ERCP or a month later, recurrent biliary symptoms affected 1 patient in the early group and 12.71 percent of the delayed group.	If you are being sent home to wait a month for your gallbladder operation, this is the number to raise.
One caveat on all of it	The Cochrane review covering 16 trials and 1,758 participants states plainly that every included trial carried a high risk of bias.	Which means local expertise reasonably decides this, and a unit that does one approach well is a better answer than a theoretical preference.

Blocked by a tumor

Where a tumor is squeezing the bile duct shut and the disease cannot be removed, a stent placed through the blockage relieves the jaundice and the itching. Two questions about that are worth knowing.

Metal stents stay open much longer than plastic ones. A trial randomized 105 patients with unresectable disease and found the first stent stayed open for a median of 273 days with metal against 126 with plastic, and among those needing a second stent, none of the metal ones blocked against 48 percent of the plastic.
And they mean fewer trips back. Choosing metal first reduced the number of endoscopic procedures by 28 percent in that trial. Survival was no different, at a median of 149 days overall, which is the honest framing since the point of the stent is comfort rather than time.
Ultrasound-guided drainage is now a real alternative. Pooling five randomized trials covering 519 patients, draining the duct under ultrasound guidance from inside the stomach rather than through the ampulla gave technical and clinical success no different from ERCP.
Its advantage is in what happens later. In that analysis, needing a repeat procedure carried a relative risk of 0.58 in favor of the ultrasound-guided route, while adverse events overall were no different. One-year stent patency did not differ significantly.
Which is not yet a reason to insist on it. Those trials are few and the technique needs specific expertise and specific stents. It belongs in the conversation as an option rather than as a standard, and asking whether a unit offers both is more useful than asking which is better.

What else it treats

Four other situations bring people to an ERCP, and the evidence for each is worth a line.

Sideways scrolling here on a small screen. Swipe or drag the table to see every column.

The other reasons an ERCP is performed, and how well each is supported
Situation	What the evidence shows
Bile leaking after gallbladder removal	Across 178 patients treated with a sphincter cut and a plastic stent, the leak closed in 91.0 percent. A high-grade leak was the only independent predictor of failure. Of the 16 failures, multiple stents closed 10 and a covered metal stent closed the remaining 6.
A benign narrowing of the bile duct	Randomizing 112 patients, mostly after liver transplant, a covered metal stent resolved the narrowing in 92.6 percent against 85.4 percent for repeated plastic stents, meeting the trial's threshold for being no worse. It took fewer procedures, at 2.14 against 3.24.
Acute infection of the bile duct	Drainage timing matters and is unevenly delivered. Across 12,613 American admissions, delay beyond 48 hours was associated with previous weight loss surgery at odds of 2.10, and with being Black at odds of 1.53 or Asian at 1.22.
Anatomy rebuilt by previous surgery	Across 132 procedures in 78 patients who had a gastric bypass, reaching the duct with a long enteroscope succeeded in 64.1 percent, while going through a temporary opening into the old stomach succeeded in 89.5 to 100 percent depending on the technique.
During pregnancy	Pooling 27 observational studies covering 1,307 pregnant patients, adverse outcomes of any kind occurred in 15.9 percent, with a confidence interval of 13.2 to 19.1. Fetal adverse outcomes did not differ significantly between techniques using X-rays and those avoiding them.

Where it should not be used

Pain after gallbladder removal

A substantial number of people continue to have upper abdominal pain after their gallbladder is removed, and for years the explanation offered was a malfunctioning sphincter at the end of the bile duct, treated by cutting it during an ERCP. That belief was tested properly. A trial randomized 214 such patients two to one between a real sphincter cut and a sham procedure, and success was achieved in 23 percent of those who had the cut and 37 percent of those who had nothing, an adjusted difference of 15.6 percentage points against the treatment. Measuring the pressure inside the sphincter, which was the test used to select patients, did not predict who responded. Pancreatitis occurred in 11 percent of those treated. Its authors wrote that the findings do not support the procedure for these patients, and that conclusion has not been overturned, so if you are being offered an ERCP for pain after gallbladder surgery with normal blood tests and a normal duct, this trial is the one to name.

Looking rather than doing

The other place ERCP should not be used is where nothing needs treating. Magnetic resonance imaging of the ducts and ultrasound from inside the stomach both answer the diagnostic question with sensitivities and specificities in the mid-nineties and no risk of pancreatitis, so an ERCP performed only to see what is there exposes you to roughly a one in ten chance of pancreatitis for information available another way. That does not mean an ERCP is wrong when a scan has already shown a stone, since going in to look and remove it in the same sitting is exactly what the procedure is for. It means the sequence matters, and the question to ask is what will be done if the ERCP finds what everybody expects it to find. A clear answer naming a stone extraction, a stent or a sphincter cut is a good sign, and an answer that amounts to seeing how things look is a reason to ask for the scan first.

Chronic pancreatitis

Two trials that both favored surgery

Where chronic inflammation has left the pancreatic duct blocked and painful, it can be opened endoscopically with stents and stone fragmentation, or drained surgically by joining the duct to a loop of bowel. Two randomized trials compared them and both pointed the same way. The first, of 39 patients, found pain scores over two years of 51 with endoscopy against 25 with surgery, complete or partial pain relief in 32 percent against 75 percent, and a median of 8 procedures against 3. A larger Dutch trial then randomized 88 patients across 30 hospitals to early surgery or an endoscopy-first approach, and found pain scores over 18 months of 37 against 49, a difference of 12 points in favor of surgery, with a median of 1 intervention against 3. Complications occurred in 27 percent against 25 percent and nobody died in either arm.

What to do with that

Neither trial was blinded, the first was small at 39 patients, and the second measured pain over 18 months and left open whether the difference persists. Its authors said as much. What the two together establish is that repeated endoscopic procedures are not a soft alternative to an operation, since they add up to more procedures for less pain relief, and that being offered a third or fourth ERCP for the same problem is a moment to ask about surgery rather than to book the next one. That conversation belongs with a pancreatic surgeon and a gastroenterologist together rather than with whoever performed the last procedure. Neither trial says endoscopy has no place, and both say it should not become an indefinite series of appointments that nobody ever steps back from and reviews.

Who does it matters

Three separate studies measured the same thing in different ways and reached the same answer. In a network of 13,018 procedures, endoscopists performing more than 239 a year had odds of 2.79 for overall cannulation success, and in a prospective study of 1,963 procedures, those doing more than two a week reached the bile duct in 96.5 percent of cases against 91.5 percent. And in a study of 2,347 sphincter cuts, endoscopists doing more than one a week had complications in 8.4 percent against 11.1 percent, with severe complications in 0.9 percent against 2.3 percent.

Those are modest-looking differences that compound, because the procedures that go wrong are disproportionately the difficult ones, and difficulty is exactly what experience handles rather than avoids. One useful nuance sits inside the same data. In the prospective study, lower caseload predicted failure to reach the duct but did not independently predict pancreatitis, which suggests the volume effect works through success and difficulty rather than through carelessness. None of that means a lower-volume unit cannot do a straightforward procedure competently, and it does mean the case that turns out to be difficult is the one where the difference shows. The practical version is short. Ask how many the endoscopist does in a year, ask what their cannulation rate is, and ask whether the unit keeps a register of its own complications, because a unit that measures itself is telling you something regardless of the number it reports. A register also means somebody there knows what their own pancreatitis rate is, which is a different and better answer than quoting the published average back at you.

Before your ERCP

Four things to sort out beforehand, and the first two take one sentence each.

Say if you have ever had pancreatitis

Pancreatitis after a previous ERCP was the strongest patient-related risk factor in a study of 1,963 procedures, carrying odds of 5.4. Pancreatitis from any other cause matters too. Do not assume it is in your notes, particularly if it happened in another country or several years ago, because this is the one piece of history that most changes what the endoscopist does during the procedure itself.

Confirm the suppository is planned

Given routinely to everybody before the procedure rather than selectively afterwards, on a trial of 2,600 patients that found 4 percent against 8 percent. This is a one-sentence question with a one-word answer and it is worth asking.

Bring the imaging, not just the reports

Whether an ERCP is the right procedure at all depends on what the scans show, and a written report cannot substitute. If you have had a magnetic resonance scan of the ducts, bring the images, since it may make the ERCP unnecessary or may change what is planned during it.

Name every blood thinner

Aspirin, clopidogrel, warfarin and the newer anticoagulants each need a separate decision and the answers differ, and in one retrospective study of 804 patients, stopping aspirin a week beforehand made no measurable difference to bleeding, so this is a conversation to have rather than a rule you should apply to yourself.

Afterwards, hour by hour

Pancreatitis after an ERCP declares itself quickly, which is why the observation period is measured in hours rather than days.

First two hours. Waking up, a sore throat, and some bloating from the air used during the procedure. Bloating that eases as you walk around is normal, and it is the commonest thing people mistake for the start of a complication, which is why the distinction in the next line is worth reading before you leave the recovery area.
Two to six hours. This is the window in which pancreatitis usually announces itself, as steady boring pain in the upper abdomen going through to the back, often with nausea, and it is different from bloating because it does not ease with walking.
Six hours onward. Most units check a blood test at this point if there is any pain, and many discharge people who are comfortable and eating. Somebody who is uncomfortable stays, which is the correct decision rather than an ominous one.
That first night. Most people go home the same day or the following morning. Have somebody with you, since the sedation lingers considerably longer than it feels like it does, you should not drive, and you should not make any decision that afternoon that you would want to be sharp for.
Days two to seven. Fever with shaking chills, jaundice, black stools or vomiting blood all mean contacting somebody the same day. Pancreatitis that develops after 24 hours is uncommon but not impossible.
If a stent was left in. Small pancreatic stents are designed to fall out on their own and are usually checked with an X-ray a week or two later, while biliary stents are planned for removal or exchange, and that date should be written down before you leave rather than left to be arranged.

Reading a quote

No figure appears on this page. ERCP is quoted as a single procedure and it is often not one, so the clauses matter.

Drag this table sideways on a narrow screen, since it scrolls rather than shrinking to fit.

What has to be in writing before you compare two quotes for an ERCP
Ask for	Why it changes the number
Whether a second ERCP is included	Large stones, difficult access and stent exchanges routinely require a second session. A quote for one procedure covering a problem that needs two is not a cheaper quote.
Whether the stent is included, and which type	Metal and plastic stents differ considerably in cost and in how long they last, at a median of 273 days against 126 in the randomized comparison.
Whether treating pancreatitis is covered	It happens to roughly one person in ten, most cases are mild and need two or three days in hospital, and a small number need much longer.
Whether an overnight stay is included	Many units keep people overnight after a therapeutic ERCP as a matter of routine, and a day-case quote will not cover it.
Whether anesthesia is separate	Deep sedation or general anesthesia is usual for this procedure and is sometimes billed apart from the endoscopy itself.
What is excluded	Packages published by Turkish hospitals and medical travel agencies generally exclude flights, insurance, complications, extra nights, subsequent procedures and any follow-up beyond the first days.

Coming to Istanbul

Five to eight days covers assessment, the procedure, the observation period and a safe departure, and longer where a stent will need exchanging or where more than one session is expected.

Do not fly the same day

Pancreatitis declares itself within hours, and a flight taken that evening removes the safety net at exactly the wrong moment. Allow at least two clear days after a therapeutic ERCP before a long flight, and longer if a sphincter cut was made or if any pain has developed.

Send these things first

The ultrasound, magnetic resonance or CT images themselves along with their reports, your liver blood tests, any previous endoscopy report, a note of whether you still have your gallbladder, details of any abdominal or weight loss surgery, your medication list naming every blood thinner, and whether you have ever had pancreatitis.

What is arranged for you

One coordinator holds your file from the first message to discharge, and the international patients team works in English, Arabic, French, Russian, Serbian, Romanian and Spanish, with other languages arranged on request. Ask in your first message if you would prefer a female physician. Someone can stay overnight with you where an admission is needed, the rooms carry a companion bed, meals cover halal, vegetarian and diabetic diets, and a prayer room is available. The office books accommodation, arranges the airport transfer and daily transport, and prepares the appointment confirmation and the invitation letter naming the hospital and your treating doctor for a medical visa application, around ten days before travel.

Follow-up after returning home

Leave with the procedure report naming exactly what was done, any images, the discharge summary and a written plan. If a stent was left in, the plan must name the removal or exchange date and who arranges it, because a biliary stent left indefinitely blocks and causes infection. If your gallbladder is still in place after stones were cleared, book that operation before you fly, since recurrent symptoms affected 12.71 percent of a delayed group against one patient in an early one. Your coordinator stays reachable on the same WhatsApp number afterwards.

Questions to ask

Four questions, all short, and the first one is the one most often skipped.

Is there something to treat, or is this to look?

If the answer is to look, ask whether a magnetic resonance scan or an ultrasound from inside would answer the same question, given that both reached sensitivities and specificities in the mid-nineties across 2,366 patients and neither causes pancreatitis.

How many do you do a year, and what is your success rate?

Both are facts and both are known. Endoscopists above 239 a year had odds of 2.79 for success in one large network, and success is what keeps a procedure short and safe.

Will I get the anti-inflammatory suppository, and might I need a pancreatic stent?

The suppository should be routine. The stent is for higher risk cases, and a trial of 1,950 patients found the drug alone did not meet the threshold for being no worse than drug plus stent.

What is the plan if you cannot get into the duct?

Conventional cannulation failed in about one in ten procedures in a network of 13,018. A clear answer, naming a technique or a second session or a different route, tells you the unit has thought about it in advance rather than deciding on the day with you already asleep on the table.

When to call

Six things that mean contacting somebody the same day rather than waiting for an appointment. Keep this list with you for the first week.

Steady upper abdominal pain going through to the back. The characteristic pain of pancreatitis. It builds rather than comes in waves, it does not ease with walking or with passing wind, and it usually starts within six hours.

Fever with shaking chills. Infection in the bile duct, which affected 8.8 percent of patients in one series. It responds well to antibiotics given quickly and badly to antibiotics given late, which is why this belongs on a list you keep with you rather than in a leaflet you read once and file away somewhere.

Black tarry stools or vomiting blood. Bleeding from the sphincter cut, which occurred in 2.0 percent of 2,347 patients and can appear up to a week later rather than immediately.

Yellow eyes or dark urine returning. A stent that has blocked or a duct that has narrowed again. Not an emergency in the same way as the first three, and it needs an appointment within days rather than weeks.

Severe pain with a rigid abdomen and inability to keep anything down. Uncommon, at a perforation rate of 0.60 percent across 16,855 patients, and the reason to go to an emergency department rather than to telephone.

A stent removal date that has passed. Not a symptom, and the commonest avoidable problem after an ERCP abroad. A biliary stent left in indefinitely will eventually block and cause infection, so chase the appointment rather than waiting to feel unwell.

ERCP FAQ

How likely is pancreatitis after an ERCP?

Around 3.5 percent across 16,855 patients in prospective studies, and 9.7 percent in the placebo arms of 108 randomized trials, which enrol more difficult cases. In high-risk patients it reached 14.7 percent. Most cases are mild, and severe pancreatitis affected 0.5 percent.

Can that be prevented?

Largely. A rectal anti-inflammatory suppository cut pancreatitis from 16.9 to 9.2 percent in one trial of 602 high-risk patients, and giving it routinely to everybody beforehand gave 4 percent against 8 percent in a trial of 2,600. A small pancreatic stent adds further protection in high-risk cases.

Do I actually need an ERCP, or would a scan do?

If nothing needs treating, a scan will do. Pooling 18 studies of 2,366 patients, ultrasound from inside the stomach gave a sensitivity of 0.95 and magnetic resonance 0.93 for bile duct stones, with specificities of 0.97 and 0.96, and neither causes pancreatitis.

What are the other risks?

Across 16,855 patients, complications of any kind occurred in 6.85 percent and death in 0.33 percent. Infection accounted for 1.44 percent, bleeding 1.34 percent, perforation 0.60 percent and events related to sedation 1.33 percent.

Should I stop my aspirin?

Ask rather than decide. In a retrospective study of 804 patients, bleeding occurred in 9.7 percent of those who continued aspirin and 9.5 percent of those who stopped it a week beforehand, against 3.9 percent of those who had never taken it. Other blood thinners need separate decisions.

When should my gallbladder come out?

Soon. In a randomized comparison, removing it within three days of the ERCP left one patient with recurrent biliary symptoms against 12.71 percent of those who waited a month, with no difference in conversion or operative difficulty.

I still have pain after my gallbladder operation. Will an ERCP help?

On the best evidence, no. A trial randomized 214 such patients between a sphincter cut and a sham procedure, and success was reached in 23 percent of those treated and 37 percent of those who had nothing. Its authors concluded the findings do not support the procedure for these patients.

Does the experience of the endoscopist matter?

Measurably. Endoscopists performing more than 239 a year had odds of 2.79 for overall success across 13,018 procedures. In a study of 2,347 sphincter cuts, those doing more than one a week had complications in 8.4 percent against 11.1 percent and severe complications in 0.9 against 2.3 percent.

Can I have an ERCP if I am pregnant?

It is done where the indication is strong. Pooling 27 observational studies covering 1,307 pregnant patients, adverse outcomes of any kind occurred in 15.9 percent, and fetal outcomes did not differ significantly between techniques using X-rays and those avoiding them.

How long should I stay in Turkey?

Five to eight days, and longer if more than one session is expected or a stent needs exchanging. Pancreatitis declares itself within hours, so allow at least two clear days after a therapeutic ERCP before a long flight.

References

Elmunzer BJ, Scheiman JM, Lehman GA, Chak A, Mosler P, Higgins PDR, Hayward RA, Romagnuolo J, Elta GH, Sherman S, Waljee AK, Repaka A. A randomized trial of rectal indomethacin to prevent post-ERCP pancreatitis. The New England Journal of Medicine. 2012;366(15):1414-1422.
Luo H, Zhao L, Leung J, Zhang R, Liu Z, Wang X, Wang B, Nie Z, Lei T, Li X, Zhou W, Zhang L. Routine pre-procedural rectal indometacin versus selective post-procedural rectal indometacin to prevent pancreatitis in patients undergoing endoscopic retrograde cholangiopancreatography, a multicentre, single-blinded, randomised controlled trial. The Lancet. 2016;387(10035):2293-2301.
Elmunzer BJ, Foster LD, Serrano J, Cote GA, Edmundowicz SA, Wani S, Shah R, Bang JY, Varadarajulu S, Singh VK, Khashab M, Kwon RS. Indomethacin with or without prophylactic pancreatic stent placement to prevent pancreatitis after ERCP, a randomised non-inferiority trial. The Lancet. 2024;403(10425):450-458.
Shi QQ, Ning XY, Zhan LL, Tang GD, Lv XP. Placement of prophylactic pancreatic stents to prevent post-endoscopic retrograde cholangiopancreatography pancreatitis in high-risk patients, a meta-analysis. World Journal of Gastroenterology. 2014;20(22):7040-7048.
Buxbaum J, Yan A, Yeh K, Lane C, Nguyen N, Laine L. Aggressive hydration with lactated Ringer's solution reduces pancreatitis after endoscopic retrograde cholangiopancreatography. Clinical Gastroenterology and Hepatology. 2014;12(2):303-307.
Sperna Weiland CJ, Smeets XJNM, Kievit W, Verdonk RC, Poen AC, Bhalla A, Venneman NG, Witteman BJM, da Costa DW, van Eijck BC, Schwartz MP, Romkens TEH, van Geenen EJM. Aggressive fluid hydration plus non-steroidal anti-inflammatory drugs versus non-steroidal anti-inflammatory drugs alone for post-endoscopic retrograde cholangiopancreatography pancreatitis (FLUYT), a multicentre, open-label, randomised, controlled trial. The Lancet Gastroenterology and Hepatology. 2021;6(5):350-358.
Kochar B, Akshintala VS, Afghani E, Elmunzer BJ, Kim KJ, Lennon AM, Khashab MA, Kalloo AN, Singh VK. Incidence, severity, and mortality of post-ERCP pancreatitis, a systematic review by using randomized, controlled trials. Gastrointestinal Endoscopy. 2015;81(1):143-149.
Andriulli A, Loperfido S, Napolitano G, Niro G, Valvano MR, Spirito F, Pilotto A, Forlano R. Incidence rates of post-ERCP complications, a systematic survey of prospective studies. The American Journal of Gastroenterology. 2007;102(8):1781-1788.
Freeman ML, DiSario JA, Nelson DB, Fennerty MB, Lee JG, Bjorkman DJ, Overby CS, Aas J, Ryan ME, Bochna GS, Shaw MJ, Snady HW. Risk factors for post-ERCP pancreatitis, a prospective, multicenter study. Gastrointestinal Endoscopy. 2001;54(4):425-434.
Freeman ML, Nelson DB, Sherman S, Haber GB, Herman ME, Dorsher PJ, Moore JP, Fennerty MB, Ryan ME, Shaw MJ, Lande JD, Pheley AM. Complications of endoscopic biliary sphincterotomy. The New England Journal of Medicine. 1996;335(13):909-918.
Tse F, Yuan Y, Moayyedi P, Leontiadis GI. Guidewire-assisted cannulation of the common bile duct for the prevention of post-endoscopic retrograde cholangiopancreatography pancreatitis. Cochrane Database of Systematic Reviews. 2012;12(12):CD009662.
Tang Z, Yang Y, Yang Z, Meng W, Li X. Early precut sphincterotomy does not increase the risk of adverse events for patients with difficult biliary access, a systematic review of randomized clinical trials with meta-analysis and trial sequential analysis. Medicine. 2018;97(36):e12213.
Giljaca V, Gurusamy KS, Takwoingi Y, Higgie D, Poropat G, Stimac D, Davidson BR. Endoscopic ultrasound versus magnetic resonance cholangiopancreatography for common bile duct stones. Cochrane Database of Systematic Reviews. 2015;2015(2):CD011549.
Makmun D, Fauzi A, Shatri H. Sensitivity and specificity of magnetic resonance cholangiopancreatography versus endoscopic ultrasonography against endoscopic retrograde cholangiopancreatography in diagnosing choledocholithiasis, the Indonesian experience. Clinical Endoscopy. 2017;50(5):486-490.
Adams MA, Hosmer AE, Wamsteker EJ, Anderson MA, Elta GH, Kubiliun NM, Kwon RS, Piraka CR, Scheiman JM, Waljee AK, Hussain HK, Elmunzer BJ. Predicting the likelihood of a persistent bile duct stone in patients with suspected choledocholithiasis, accuracy of existing guidelines and the impact of laboratory trends. Gastrointestinal Endoscopy. 2015;82(1):88-93.
Peng C, Nietert PJ, Cotton PB, Lackland DT, Romagnuolo J. Predicting native papilla biliary cannulation success using a multinational ERCP quality network. BMC Gastroenterology. 2013;13:147.
Pecsi D, Hegyi P, Szentesi A, Godi S, Pakodi F, Vincze A. The role of endoscopy registries in quality health care, the first data from the Hungarian ERCP registry. Orvosi Hetilap. 2018;159(37):1506-1515.
Hui CK, Lai KC, Yuen MF, Wong WM, Lam SK, Lai CL. Does withholding aspirin for one week reduce the risk of post-sphincterotomy bleeding. Alimentary Pharmacology and Therapeutics. 2002;16(5):929-936.
Howard TJ, Tan T, Lehman GA, Sherman S, Madura JA, Fogel E, Swack ML, Kopecky KK. Classification and management of perforations complicating endoscopic sphincterotomy. Surgery. 1999;126(4):658-663.
Boonsinsukh T, Viriyaroj V, Yodying H. Risk factors for post endoscopic retrograde cholangiopancreatography cholangitis. Journal of the Medical Association of Thailand. 2016;99(Suppl 8):S166-S170.
Dasari BVM, Tan CJ, Gurusamy KS, Martin DJ, Kirk G, McKie L, Diamond T, Taylor MA. Surgical versus endoscopic treatment of bile duct stones. Cochrane Database of Systematic Reviews. 2013;2013(12):CD003327.
Zhu HY, Xu M, Shen HJ, Yang C, Li F, Li KW, Shi WJ, Ji F. A meta-analysis of single-stage versus two-stage management for concomitant gallstones and common bile duct stones. Clinics and Research in Hepatology and Gastroenterology. 2015;39(5):584-593.
El Nakeeb A, Ezzet H, Askar W, El Hanafy E, Hamdy E, Atef E, Youssef M, Talaat H, Hamed H, Abdallah T. Early versus late cholecystectomy after clearance of common bile duct stones by endoscopic retrograde cholangiopancreatography, a prospective randomized study. Surgical Laparoscopy, Endoscopy and Percutaneous Techniques. 2016;26(3):202-207.
Atarere JO, Mensah B, Kunkle B, Nwaneki C, Annor E, Vasireddy R, Orhurhu V, Weisman D, Thompson C. Factors associated with delayed endoscopic retrograde cholangiopancreatography among patients with acute cholangitis. Journal of Clinical Gastroenterology. 2025.
Wang M, Wadhwani SI, Cullaro G, Lai JC, Rubin JB. Racial and ethnic disparities among patients hospitalized for acute cholangitis in the United States. Journal of Clinical Gastroenterology. 2023;57(7):731-736.
Davids PH, Groen AK, Rauws EA, Tytgat GN, Huibregtse K. Randomised trial of self-expanding metal stents versus polyethylene stents for distal malignant biliary obstruction. The Lancet. 1992;340(8834-8835):1488-1492.
Khoury T, Sbeit W, Fumex F, Marasco G, Eusebi LH, Fusaroli P, Chan SM, Shahin A, Basheer M, Gincul R, Leblanc S, Teoh AYB. Endoscopic ultrasound- versus ERCP-guided primary drainage of inoperable malignant distal biliary obstruction, systematic review and meta-analysis of randomized controlled trials. Endoscopy. 2024;56(12):955-963.
Canena J, Horta D, Coimbra J, Meireles L, Russo P, Marques I, Ricardo L, Rodrigues C, Capela T, Carvalho D, Loureiro R, Dias AM. Outcomes of endoscopic management of primary and refractory postcholecystectomy biliary leaks in a multicentre review of 178 patients. BMC Gastroenterology. 2015;15:105.
Cote GA, Slivka A, Tarnasky P, Mullady DK, Elmunzer BJ, Elta G, Fogel E, Lehman G, McHenry L, Romagnuolo J, Menon S, Siddiqui UD. Effect of covered metallic stents compared with plastic stents on benign biliary stricture resolution, a randomized clinical trial. JAMA. 2016;315(12):1250-1257.
Cotton PB, Durkalski V, Romagnuolo J, Pauls Q, Fogel E, Tarnasky P, Aliperti G, Freeman M, Kozarek R, Jamidar P, Wilcox M, Serrano J. Effect of endoscopic sphincterotomy for suspected sphincter of Oddi dysfunction on pain-related disability following cholecystectomy, the EPISOD randomized clinical trial. JAMA. 2014;311(20):2101-2109.
Issa Y, Kempeneers MA, Bruno MJ, Fockens P, Poley JW, Ahmed Ali U, Bollen TL, Busch OR, Dejong CH, van Duijvendijk P, van Dullemen HM, van Eijck CH. Effect of early surgery versus endoscopy-first approach on pain in patients with chronic pancreatitis, the ESCAPE randomized clinical trial. JAMA. 2020;323(3):237-247.
Cahen DL, Gouma DJ, Nio Y, Rauws EAJ, Boermeester MA, Busch OR, Stoker J, Lameris JS, Dijkgraaf MGW, Huibregtse K, Bruno MJ. Endoscopic versus surgical drainage of the pancreatic duct in chronic pancreatitis. The New England Journal of Medicine. 2007;356(7):676-684.
Ghazi R, Razzak FA, Kerbage A, Brunaldi V, Storm AC, Vargas EJ, Bofill-Garcia A, Chandrasekhara V, Law RJ, Martin JA, Ghanem OM, Petersen BT, Abu Dayyeh BK. Endoscopic retrograde cholangiopancreatography approach for patients with Roux-en-Y gastric bypass, a comparative study between four ERCP techniques with proposed management algorithm. Surgery for Obesity and Related Diseases. 2024;20(1):53-61.
Azab M, Bharadwaj S, Jayaraj M, Hong AS, Solaimani P, Mubder M, Yeom H, Yoo JW, Volk ML. Safety of endoscopic retrograde cholangiopancreatography in pregnancy, a systematic review and meta-analysis. Saudi Journal of Gastroenterology. 2019;25(6):341-354.

Editor's note

Written by the Biruni Hospital medical editorial team. Reviewed by Assistant Professor Emir NEKAY, General Surgery.

Medically reviewed by

Assistant Professor Emir NEKAY

General Surgery

Related Treatments

View All