Neuroendocrine Tumor - NET Surgery

Free consultation

Two numbers change the whole plan, the grade and the Ki-67

Send the pathology report with the grade and the Ki-67 index written in it, since those two figures decide more about the operation than the size of the tumour does. Alongside them we want the cross-sectional imaging on a disc, any somatostatin receptor scan you have had, and the chromogranin A and urinary 5-HIAA results if they were measured. Tell us whether you have flushing, diarrhoea, wheezing, low blood sugar or ulcers that will not settle, because a tumour that secretes something is a different operation from one that does not. Say how long the tumour has been known about and whether earlier scans exist, since a lesion that has not moved in three years is telling you something an isolated scan cannot. The review costs nothing and commits you to nothing.

Two rules that run backwards

Neuroendocrine tumours arise from hormone-producing cells scattered through the gut, the pancreas and the lungs, and their defining characteristic is that most of them grow slowly enough to change what a surgeon should do about them. A pancreatic adenocarcinoma doubles in weeks and gives you no time to think. A well differentiated neuroendocrine tumour of the same size may not have changed measurably in three years, and that difference in tempo is why the surgical logic that governs every other abdominal cancer does not simply transfer across. It is also why so much of the advice patients find online is subtly wrong for them, since it was written for the fast diseases. This one is slow. The word neuroendocrine covers an enormous range, from a five millimetre lesion in the stomach that will never do anything to a poorly differentiated carcinoma that behaves like a small cell lung cancer and is not treated surgically at all. Between those extremes sits the disease this page is about, the well differentiated tumour of grade one or grade two arising in the pancreas, the small bowel, the appendix, the stomach, the rectum or the lung. For those, surgery remains the only treatment that removes the disease, and the questions worth arguing about are which patients need it, how much of it they need, and when. Three questions, one argument.

When the right operation is none

Cross-sectional imaging has become so good and so widely used that a great many pancreatic neuroendocrine tumours are now found by accident, in someone scanned for kidney stones or back pain who has no symptoms whatsoever.

The instinct is to remove them.

Pancreatic surgery, though, is not a small undertaking, and the operation that removes a one centimetre lesion from the tail of the pancreas carries a real risk of a leak, of diabetes, of a splenectomy and of the vaccinations and infection risk that follow one. Weighing that against a tumour that may be biologically inert requires evidence rather than instinct.

The evidence exists.

That finding is not a licence to ignore every small tumour, and the companion evidence matters as much. Researchers examined 612 patients from a national registry whose non-functioning pancreatic tumours measured two centimetres or less and who all had surgery. Nodal spread was present in 72 of them, which is 11.7 percent, and distant metastases in 35, which is 5.7 percent, so roughly one in nine of these supposedly harmless lesions had already reached the lymph nodes.

Three factors predicted it.

A tumour in the body or tail instead of the head, a grade of three or four, and younger age at diagnosis. The authors read their own data as supporting watchful waiting for most patients while arguing for individualised risk assessment in the group carrying those features, which is exactly the right conclusion and exactly the conversation you should be having. Raise all three.

Surveillance means something specific and it is not neglect. It means a scan at defined intervals, usually every six to twelve months at first and stretching out if nothing moves, with an agreed threshold that triggers surgery. Growth beyond two centimetres, growth of more than half a centimetre between scans, a rise in grade on repeat biopsy, a duct that becomes obstructed or the appearance of a suspicious node all move a patient from the watching column to the operating one.

What makes surveillance safe is that the threshold is written down in advance. What makes it dangerous is drifting into it because nobody made a decision. Write the threshold down.

Grade, and whether it secretes

Only two properties of the tumour do most of the work in deciding what happens, and neither is size. The first is grade, which comes from the Ki-67 index, a measure of what proportion of the tumour cells are actively dividing at the moment the tissue was fixed. The second is whether the tumour produces a hormone in quantities that reach the bloodstream and cause symptoms.

A tumour that does is called functioning, and functioning tumours are operated on for a reason that has nothing to do with cancer, which is that the hormone itself is making the patient ill. Cancer is not the reason.

This table scrolls sideways on a narrow screen. Swipe or drag to see every column.

One warning about grade is worth carrying with you. Ki-67 measured on a needle biopsy comes from a few cores of a tumour that is not uniform, and it tends to understate the true grade of the whole lesion, sometimes considerably. A biopsy reading of 2 percent does not guarantee the resected specimen will read 2 percent. This is one of the reasons surveillance decisions are made on the full picture of size, site, growth over time and symptoms, never on a single laboratory figure, and it is a reason to ask how much tissue the grade was calculated from. One core is not the tumour.

Where it started changes everything

Site changes this operation more than anything else does. Tumours with identical grade and identical size behave differently depending on which organ they arose in, and the operations bear almost no resemblance to one another. A patient reading about neuroendocrine tumour surgery in general will find advice that applies to somebody else entirely, so the first question to settle is site. Everything else follows.

This table scrolls sideways on a narrow screen. Swipe or drag to see every column.

Why the scan is not the survey

Small bowel neuroendocrine tumours have a habit that makes them unlike almost anything else in abdominal surgery. They are commonly multiple, scattered along several metres of bowel as separate primaries rather than as spread from one, and the additional lesions are frequently only a few millimetres across. Somatostatin receptor imaging, usually a gallium DOTATATE positron emission tomography scan, is the most sensitive tool oncology has for finding this disease anywhere in the body, and it is genuinely excellent. It is still not good enough to be trusted with this particular question. Hands still beat it.

Those numbers deserve stating plainly, because it cuts against the direction most surgical marketing points. When the scan says there is more than one tumour it is usually right. When the scan says there is only one, it is wrong more than a third of the time.

And a negative predictive value of 29 percent for lymph nodes means a clean-looking scan tells you almost nothing about whether the nodes are involved. For small bowel disease that is an argument for an open operation in which the surgeon runs the whole small intestine through their fingers from the ligament of Treitz to the caecum, and against a keyhole approach chosen for the sake of a smaller scar. A missed second primary becomes a recurrence a few years later, and the second operation is performed through adhesions.

Nobody wants that one.

What the operation involves

Specifics vary enormously by site. The sequence of a laparotomy for a small bowel or pancreatic tumour follows a recognisable shape.

Laparoscopic and robotic approaches have a real place in this disease, particularly for distal pancreatectomy, where they reduce complications and shorten stay against the open equivalent. The argument for keyhole surgery is weakest exactly where the palpation evidence is strongest, which is the small bowel. A surgeon who offers you a minimally invasive small bowel resection should be able to explain how they intend to find the second and third tumours, and if the answer is that the scan was clear, you now know what that scan is worth. Ask anyway.

The anaesthetic problem

Handling a serotonin-secreting tumour can release a surge of hormone into the circulation, and the result on the operating table is a sudden and severe swing in blood pressure known as carcinoid crisis. It has a real mortality, it is provoked by exactly the manipulation the operation requires, and it is the single reason this surgery belongs in a centre that does it regularly.

Most patients never hear about it before their operation. It is worth hearing about, because the way a unit answers a question about it tells you how often they do this. Volume shows in the answer.

A French centre studied 81 patients operated on for small bowel neuroendocrine tumours under a standardised protocol of continuous octreotide infusion started before surgery. Fifty-nine had liver metastases. Forty-nine had a prior carcinoid syndrome and seven had carcinoid heart disease. This was not a low risk group. They recorded 139 episodes of intraoperative carcinoid syndrome in 45 of the patients, of which 45 met their strictest definition. Ninety-one percent of the episodes were hypertensive and 29 percent hypotensive. No factor they tested, including the use of vasopressors, predicted who would have one. Crucially, no patient suffered a full carcinoid crisis, every operation was completed, and the authors read this as evidence that the standardised prophylaxis protocol works.

One patient died of cardiac failure four days afterwards.

That has two consequences. The first is that blood pressure swings during these operations are common rather than exceptional, so an anaesthetist who is expecting them and has octreotide drawn up handles something routine, while one who is not is handling an emergency. The second concerns the heart. Long standing serotonin exposure damages the right sided heart valves in a proportion of patients with carcinoid syndrome, and carcinoid heart disease that has gone unrecognised is a far more dangerous thing to take to theatre than the crisis itself. An echocardiogram before surgery in anyone with a functioning midgut tumour is not a formality, and the one death in that series was cardiac. Book the echocardiogram.

Operating on spread disease

Here is the second inversion. In most cancers, liver metastases mean surgery on the primary tumour has little left to offer. In well differentiated neuroendocrine disease that reasoning does not hold, and a substantial body of work supports operating in the metastatic setting for two separate reasons that are worth keeping apart in your mind. One is symptom control, since a large volume of hormone-producing tissue causes flushing, diarrhoea and eventually valve damage, and removing most of it reduces all three. The other is survival, and the evidence there is observational rather than randomised, which is a limitation nobody should hide from you. A recent review of surgical management of neuroendocrine liver metastases sets out where the field has moved. Preoperative gallium DOTATATE scanning combined with gadoxetic acid enhanced liver magnetic resonance imaging has improved the mapping of how much liver disease there actually is, which in turn improves how completely it can be cleared. Parenchyma-sparing techniques, in which deposits are enucleated or wedged out instead of removing whole anatomical segments, have widened the number of patients who can be treated. The threshold for what counts as worthwhile debulking has been lowered from the older requirement to remove ninety percent of the tumour burden. And removal of the primary tumour is associated with improved survival regardless of whether the liver metastases are themselves treated, which is the finding that most directly contradicts the ordinary rule.

Honestly framed, cytoreduction of liver metastases is rarely curative and is nonetheless associated with better symptom control and longer survival, delivered as part of a package that includes liver-directed treatments such as embolisation and ablation alongside systemic therapy with somatostatin analogues, targeted drugs and peptide receptor radionuclide therapy. Surgery is one component of that package rather than the whole of it. A unit that presents an operation as the answer, without naming what comes before and after it and who will deliver those parts, is describing half a plan. Ask who does the other half.

Recovery and what changes

Length of stay tracks the operation rather than the diagnosis.

An endoscopic removal of a small gastric or rectal lesion is a day case, while enucleation of a pancreatic tumour averages under six days, against just over seven for a distal pancreatectomy, and the enucleation group also has less blood loss and shorter operating time, though a higher rate of pancreatic leak. A small bowel resection with mesenteric clearance runs to six or ten days. A pancreatoduodenectomy is ten to fourteen. Liver debulking sits between seven and twelve depending on how much parenchyma was removed. Recovery to ordinary activity follows the same ranking, from a few days after an endoscopic procedure to six or eight weeks after a major pancreatic or hepatic resection. Some consequences are permanent and belong in the conversation before surgery, and mentioning them afterwards is too late. Removing the terminal ileum, which is common in small bowel disease, means bile salts are no longer reabsorbed there, and the result is loose stools that respond well to a bile acid binder but do not resolve on their own. It also stops absorption of vitamin B12, so lifelong injections are needed. Losing part of the pancreas can produce diabetes or poor digestion of fat requiring enzyme replacement with meals, and how likely that is depends on how much gland was taken and how healthy the remainder was. If the spleen came out, vaccinations and an understanding of infection risk follow. If the gallbladder came out alongside the resection, expect looser stools for some months. Ask beforehand.

There is one change nobody warns patients about and it is worth naming. In a functioning tumour that has been causing flushing and diarrhoea for years, those symptoms often disappear within days of surgery, and the difference can be startling. Patients frequently describe realising only afterwards how much of what they had accepted as their normal state was the hormone.

That is the argument for operating on functioning tumours stated better than any survival curve states it. Patients notice within days.

Coming to Istanbul

How long you need to stay in Istanbul depends on which operation is planned, and the honest ranges are two to three weeks for an endoscopic or laparoscopic procedure and three to five weeks for open small bowel, pancreatic or liver surgery, hotel nights included. The first week goes on assessment. Imaging is reviewed or repeated, a somatostatin receptor scan is arranged if you have not had one or if yours is more than a few months old, biochemistry is measured, the pathology is re-read by our own pathologists where slides exist, and an echocardiogram is done for anyone with a functioning midgut tumour. The case then goes to a multidisciplinary meeting that includes the surgeon, the endocrinologist, the medical oncologist, the nuclear medicine physician, the radiologist and the pathologist before anything is fixed. Somatostatin analogue treatment is worth planning around specifically, since the long acting monthly injection is usually started or continued in the perioperative period and it needs to be arranged with whoever will give it once you are home. Peptide receptor radionuclide therapy, where it forms part of the plan, is delivered in cycles over months and is not something to attempt to compress into a single trip. We will tell you which parts of the plan belong here and which belong at home, and that answer comes from the multidisciplinary discussion. Not from a coordinator.

Flying home is reasonable once you are eating, the wound is dry, any drains are out and the pathology has been discussed with you face to face, which for open abdominal surgery generally lands at twelve to eighteen days after the operation. Clot risk is raised after cancer surgery in the abdomen, so compression stockings, a defined period of anticoagulation and moving around the cabin all matter on the way back. Interpreting is arranged in advance in English, Arabic, Russian, French and German.

You go home with the operative note, the full pathology including grade and Ki-67, the imaging on a disc, the drug plan, and a written surveillance schedule. Check nothing is missing before you fly.

What moves the cost

Neuroendocrine tumour surgery spans an unusually wide range of operations, from an endoscopic resection lasting forty minutes to a combined bowel, mesenteric and liver procedure lasting most of a day, so a single figure attached to the phrase would be meaningless. Anybody quoting one before the site, the grade, the functional status and the extent of liver involvement are established is quoting for an operation they have not yet identified. What follows are the variables that actually move it, all of them clinical. None of them administrative.

• Which organ, and therefore whether this is an endoscopic procedure, a segmental bowel resection, an enucleation, a distal pancreatectomy or a pancreatoduodenectomy.
• Whether the liver is involved and whether it is being treated in the same sitting, which turns one operation into two.
• Whether the tumour is functioning, since that adds perioperative octreotide, cardiology assessment and closer intraoperative monitoring.
• Whether a somatostatin receptor scan is needed here or has already been done to an acceptable standard elsewhere.
• Whether the approach is open, laparoscopic or robotic, and whether a second consultant team is operating alongside.
• Whether an intensive care bed is required after surgery, which depends on the operation and on cardiac status.
• How much pathology is needed, since grading, receptor staining and sometimes molecular work all sit behind the final report.
• Whether drug therapy or radionuclide treatment forms part of the plan and where it will be delivered.

Any written quotation should say which of those it covers. The exclusions catch people out, so ask whether the figure includes the somatostatin receptor scan, the full pathology with Ki-67 and receptor staining, intensive care if it proves necessary rather than only if it was predicted, treatment of a pancreatic leak should one occur, and how many follow-up appointments and scans sit inside it. Ask what happens to the figure if the multidisciplinary meeting changes the plan, which in this disease it frequently does, because the operation proposed on the strength of an outside scan is not always the operation that turns out to be right. A number that survives those questions is worth something. One that does not was never a number.

Follow-up once you are home

Surveillance after neuroendocrine surgery runs longer than for most cancers because recurrence can appear a decade later, and a five year discharge is inappropriate here. Expect cross-sectional imaging and biochemistry every six to twelve months for several years, stretching out but rarely stopping, with a somatostatin receptor scan repeated when something on conventional imaging is unclear. Chromogranin A and, for midgut tumours, urinary 5-HIAA are followed as trends rather than as single values, since both are affected by diet, by proton pump inhibitors and by kidney function, and a solitary raised result means much less than a rising line. Trends matter. Single values rarely do.

Your own oncologist can deliver all of this, and it is easier for everyone if they do. Six documents make that practical. Leave with all of them.

• The operative note describing precisely what was removed and what was deliberately left behind.
• The pathology report with grade, Ki-67, margin status and the number of nodes examined and involved.
• The somatostatin receptor status, which determines whether radionuclide therapy is available to you later.
• A note of whether the terminal ileum, spleen or gallbladder was taken, and what supplementation follows from each.
• The drug schedule with doses and dates, including who gives the monthly injection and when the next one falls due.
• A written surveillance plan naming the specific scans, the blood tests and the intervals.

Our team stays reachable for your doctor's questions afterwards, and a scan that looks equivocal three years from now is something we would rather look at with you than hear about later. Send it over.

Frequently asked questions about neuroendocrine tumour surgery