When cancer spreads to bone, it weakens it. The bone can become painful, and it can break under loads it would normally carry easily — a break called a pathological fracture.
Orthopaedic surgery has a specific and limited job in this situation, and it helps to be clear about it from the start. The operation is not treating the cancer. Your oncologist does that. The operation is about the mechanics: keeping the bone able to bear load, keeping you walking or using your arm, keeping you out of pain and out of hospital.
This page explains how that decision is made, what the operations involve, and — just as importantly — when surgery is not the right answer.
The essentials
- The purpose is function, not cure. Success is measured in walking, independence and comfort, not in shrinking the cancer.
- Fixing a bone before it breaks is better than repairing it afterwards. Better survival, fewer complications, and a far better chance of walking normally again.
- Pain when you put weight through a limb is the symptom to report. It is different from general aching and it is the most useful warning that a bone is in mechanical trouble.
- The operation chosen depends heavily on expected survival. Median survival after this surgery varies more than tenfold depending on the type of cancer, and that legitimately changes what implant is right.
- Sometimes not operating is the right decision — and that is a decision made with you, not for you.
Some background
Bone is one of the commonest places for cancer to spread. In an analysis of more than 400,000 patients with cancer affecting bone, the primaries responsible were most often lung (about 25%), prostate (19%), breast (19%), then gastrointestinal and urological cancers.
Not all behave the same way in bone. Ranked by how often they actually cause a long bone to break, the order shifts: kidney cancer heads the list, then myeloma, then gynaecological, lung and breast cancers. The femur — the thigh bone — is the bone most often affected, and because it carries your body weight it is where a fracture does the most damage to daily life.
One piece of context makes sense of everything that follows. People with advanced cancer are living considerably longer than they used to. In a study of 460 patients having surgery for long bone metastases, survival improved significantly in exactly those cancers that respond to modern drug treatment — kidney cancer, melanoma, liver and bowel cancer — with immunotherapy specifically named. That changes the surgical problem: a repair that only needed to last months may now need to last years.
Why operate at all?
A fractured femur in someone with advanced cancer is a catastrophe of a particular kind. It usually means an ambulance, an emergency admission, an operation done urgently rather than planned, weeks of reduced mobility, and often not getting back to walking as well as before. It interrupts cancer treatment. It can end independent living.
Surgery on a bone that is weakened but not yet broken avoids nearly all of that. The evidence for doing it earlier rather than later is consistent:
Survival and complications. In that 400,000-patient analysis, patients treated for an impending fracture had better survival at both 90 days and one year than those treated after the bone had broken, with significantly fewer blood clots in the legs, fewer clots on the lung, fewer urinary infections and fewer blood transfusions.
Getting back to walking. A study comparing nailing of impending against completed femoral fractures found the pre-emptive operations had shorter operating times, less blood loss and shorter hospital stays — and, the outcome that matters most, significantly more patients regained the ability to walk.
One claim that does not hold up, and it is worth being straight about. Prophylactic surgery is often said to be cheaper. The best-adjusted study — 265 patients, carefully matched — found no difference in hospital costs at all. What it did find was that patients who broke the bone first were far more likely to need discharge to a rehabilitation facility: 57% against 30%. The real cost of waiting is not money. It is independence.
How the decision is made
Judging whether a bone will break
Several things are weighed together:
- How much of the bone’s width the lesion occupies
- Whether the outer shell — the cortex — is breached
- Which bone, and where in it — weight-bearing matters, and the region just below the hip is the highest-risk site of all
- Pain when you load the limb — the single most informative symptom, and quite different from pain at rest
A scoring system published by Mirels in 1989 combines four of these into a number and remains the most widely used tool. It has real limitations worth knowing: different observers score the same X-ray differently, and in the arm it has been shown to miss most of the bones that go on to break. It is best understood as a structured prompt for the conversation rather than a verdict, which is why it never replaces looking at the whole picture.
Estimating how long the repair needs to last
This is the part patients rarely hear discussed openly, and it deserves to be, because it drives the choice of operation more than anything else.
In the study of 460 surgical patients, median survival after surgery differed enormously by cancer type: 16.5 months in the better-prognosis group, 9.2 months in the intermediate group, and 3.9 months in the poorest group. A repair expected to be needed for three months and one expected to be needed for three years are genuinely different engineering problems.
That is not a judgement about the value of anyone’s time. It is the honest basis on which a surgeon decides whether to choose the quickest operation with the fastest recovery, or a more substantial reconstruction built to last.
The operations
Several different operations are legitimate here, and experts genuinely differ at the margins. In a survey of specialist tumour surgeons, recommendations for a fracture at the top of the femur split between intramedullary nailing (45%), removing and reconstructing the segment (34%), hemiarthroplasty with a long cemented stem (15%) and plate fixation (7%) — with variation driven by cancer type, the fracture pattern and its exact site. What follows is the logic behind each.
Intramedullary nailing
A metal rod passed down the hollow centre of the bone. Because it supports the bone along its entire length rather than at a single point, it protects the lesion that prompted the operation and any other deposits in the same bone — which matters in a disease that is rarely confined to one spot.
Nails go in through small incisions without opening the tumour, so blood loss is modest and recovery quick. Weight-bearing is generally allowed immediately, which is the whole purpose.
A theoretical concern is that instrumenting the marrow cavity might spread tumour along the tract. This has been studied: among 122 patients, disease progressed around the nail in about 6%, and only 2% ever needed further surgery because of it. Worth monitoring; not a reason to avoid the operation.
Open internal fixation with cement augmentation
Where a lesion has hollowed out a segment of bone, or sits somewhere a nail cannot control, the lesion is opened, the tumour scraped out, and the bone fixed with a plate and screws.
Cement is what makes this work. Bone affected by cancer, and bone that has been irradiated, cannot be relied on to heal — so a repair that depends on healing may simply fail while waiting. Bone cement fills the cavity and provides structural support from the moment it sets, independent of biology. In the surgeons’ survey, 56% recommended cement alongside a nail for these fractures, which gives a fair sense of how routine the principle is.
In one comparison at the shoulder, plate fixation with cement involved considerably less blood loss and shorter surgery than replacing the joint, with better early pain and function scores. That study was small and the groups were not truly comparable — its own authors said so — so it shows the trade-off rather than a winner: fixation is the smaller operation where there is enough bone to hold it.
Joint replacement
When the joint surface itself, or the bone immediately next to it, has been destroyed, there is too little sound bone for a nail or plate to grip. Here the answer is to replace the joint — commonly with a standard hip or knee replacement using a long, cemented stem that also protects the length of bone below.
The advantage is fundamental: a replacement does not rely on bone healing at all. It is stable the day it goes in, allows immediate weight-bearing, and reliably relieves pain.
Endoprosthetic reconstruction
Where a whole segment of bone has been destroyed, that segment and its adjacent joint are removed and replaced with a larger modular implant — an endoprosthesis. These are the same implants used in primary bone cancer surgery, and they are the reason a limb can be preserved when a substantial length of bone has gone.
Why fixation sometimes has to become a replacement
There is a piece of evidence that illuminates the whole decision, and patients are rarely told about it.
A series looked at 26 patients whose fixation for a fracture at the top of the femur later failed and had to be converted to a joint replacement. The conversions happened at a mean of 13 months after the first operation, and the reasons were disease progressing (12 patients), the metal failing (8) and the bone never uniting (6). After conversion, function improved markedly, and the authors concluded that replacement “reliably provides improved quality of life and immediate weight bearing”.
The lesson is not that fixation is wrong. It is that fixation depends on bone that cancer may continue to destroy — so in someone expected to live long enough for that to matter, a replacement chosen at the outset may be the better decision. This is precisely why expected survival features so heavily in the discussion, and why these operations are best planned rather than improvised at 2am.
When not to operate
Knowing when to stop is as much a part of this work as knowing when to operate, and there are several situations where surgery is not the right answer.
When expected survival is very short. In the poorest-prognosis group, median survival after surgery was under four months. Recovery from a major operation can consume a significant part of that. Where the burden of the operation would outweigh the time it could improve, not operating is a legitimate and often kinder decision.
When the lesion is painful but not a mechanical threat. Mirels’ original work showed that lesions below the threshold could be safely treated with radiotherapy alone, without fixation. Radiotherapy is very effective for bone pain and carries none of the recovery cost of surgery.
When the drug treatment is likely to settle it. In cancers that respond well to modern systemic therapy, the treatment itself can improve the bone. This is a genuine reason to wait and reassess rather than to operate immediately, and it is why these decisions are made jointly with your oncologist rather than by a surgeon alone.
When the risks are simply too high. Ninety-day mortality after surgery for a pathological fracture reaches around 10 to 12% in the highest-risk cancer groups. That is not a reason to withhold surgery from people who need it — but it is a reason to be honest that this is significant surgery in people who are unwell.
What this means for you
- Report pain that comes on when you put weight through a limb. Not general aching — the pain that appears when you stand, walk, or push up from a chair. It is the most useful thing you can tell your oncologist, and it is the commonest missed opportunity to act early.
- If a bone is at risk, ask whether it can be dealt with before it breaks. The evidence for doing so is good, and the difference shows up in whether you keep walking.
- Expect the conversation to include how long the repair needs to last. It is not a comment on your prognosis so much as an engineering question, and it is the reason one person gets a nail and another a replacement.
- Expect radiotherapy afterwards, and expect your cancer treatment to continue. The operation fixes the bone; it does not treat the cancer.
- If the recommendation is not to operate, ask why — and expect a real answer about survival, recovery cost and what radiotherapy or drug treatment can achieve instead. A considered decision not to operate is a clinical decision, not a refusal.
Assessment of a bone lesion is set out in tests and diagnosis, and the way a plan is agreed between teams in how treatment is planned. Myeloma affects bone in a related but distinct way and is covered in myeloma and bone. Cancer affecting the spine is assessed and managed differently and is not covered here.
Referral is usually arranged through your oncology team. If a bone is painful on weight-bearing, early assessment — before it breaks — is the single most useful step.