The tests done before lung cancer treatment are not repeating each other. They answer three separate questions: what the abnormality actually is, where it is and whether it has spread, and whether you could withstand the operation that would remove it. No single test answers more than one of them, which is why the list is longer than it feels like it should be. This page sets out the sequence, what each step is for, and — if you have been told surgery is not possible — how to find out which of the three questions produced that answer. Dr Lawrence Okiror, Consultant Thoracic and Robotic Surgeon (GMC 6150382), London Bridge Hospital and The Lister Hospital Chelsea.
Last reviewed: August 2026 · Dr Lawrence Okiror FRCS(CTh) FRCSEd(CTh) · GMC 6150382
History, examination and imaging point to the answer. Only a tissue sample gives it. A scan can show that something is there; it cannot say what it is made of.
PET-CT, brain MRI and sampling of the lymph glands in the centre of the chest. Glands are sampled rather than judged on the scan, because scans get this wrong often enough to matter.
Lung function, exercise testing and cardiac assessment. The question is how much function you would have left afterwards — not how much you have today.
Because they are not repeating each other. Each test answers only one of three questions, and all three have to be answered before treatment can be decided.
The first is what the abnormality actually is. The second is where it is, and whether it has gone anywhere else. The third is whether you could withstand the operation that removing it would require.
Nothing on the list answers more than one. A PET scan says nothing about your lung function. A breathing test says nothing about whether the abnormality is cancer. That is why the sequence looks long, and why results arriving one at a time can feel like it is taking forever — the picture only becomes usable when all three parts are in place.
It starts with a conversation, not a scan. How long symptoms have been present, whether anything has changed, what you have been treated for before, what medications you take, and how much you can physically do. Then an examination. A good history changes the interpretation of everything that follows.
A cough that will not settle, chest pain, breathlessness, coughing up blood, and unexplained weight loss. Each has its own page here.
A hoarse or changed voice deserves separate mention, because it is so easily dismissed as a lingering cold. The nerve that supplies the left vocal cord loops down into the chest and passes close to structures near the left lung and main airway. A new change in the voice alongside a lung abnormality can indicate involvement of that nerve, and it changes both the staging and the plan. Mention it even if it seems minor.
Then imaging. Usually a chest X-ray first, then a CT scan with intravenous contrast, and often a PET-CT. What each of these shows, and what it cannot show, is set out on the chest imaging page. If your scan report mentions a shadow or a nodule, the shadow on a lung scan page explains what that language means.
A scan can show that something is there. Only a sample can say what it is made of. How that sample is obtained depends on where the abnormality sits:
This is staging. It establishes whether the cancer is confined to the lung, whether the lymph glands in the centre of the chest are involved, and whether there is anything elsewhere in the body. It determines what treatment is appropriate, and it is completed before any decision about surgery is finalised.
A whole-body scan that shows where tissue is metabolically active. It is the main test for finding disease outside the lung and for identifying which lymph glands warrant sampling. It is a pointer rather than a verdict — and the section below explains why that distinction matters.
Separate from the CT that forms part of a PET-CT, which is a low-dose scan used mainly to locate activity. A diagnostic CT with contrast gives thin-slice anatomical detail showing the relationship between tumour, glands and blood vessels — which is what determines how an operation would be done, and sometimes whether it can be. Current expert consensus recommends both where mediastinal involvement is suspected.
Often already covered by the PET-CT. Where it is not, it is done separately to check the liver, adrenal glands and other abdominal organs.
Lung cancer can spread to the brain without causing any symptoms, and finding that before an operation changes the plan entirely. MRI is more sensitive than CT for this. It is routine in most patients being considered for surgery, and a normal result is what is expected in the great majority.
Because scans get this wrong often enough to matter. In randomised data, the positive predictive value of PET for the lymph glands in the centre of the chest was 64% — meaning that among patients whose scan suggested the glands were involved, around a third turned out not to be when the glands were actually sampled. Inflammation, infection and other benign processes can all light up on a PET scan.
The consequence runs both ways. Treating on a scan finding alone risks giving someone systemic treatment they did not need — or, in the other direction, withholding an operation from someone whose glands were never actually involved. That is why, where there is any concern about the glands, they are sampled.
If there is any concern that the cancer may have spread to the lymph glands in the centre of the chest, the next step is an EBUS — a specialised bronchoscopy that takes samples from those glands through the wall of the airway, using ultrasound to see them. It is performed by the respiratory medicine team. Sometimes it does not produce an answer, and some patients cannot tolerate it under sedation.
An operation to sample those same glands directly, under general anaesthetic, through a small incision at the base of the neck. It is used where an EBUS has not produced a clear answer, where the glands cannot be reached with a needle, where more tissue is needed than a needle provides, or where a patient cannot tolerate EBUS under sedation.
Dr Okiror performs a substantial proportion of the cervical mediastinoscopies carried out in his department. It is usually a day case. Related mediastinal procedures are covered on the mediastinal surgery page.
The order is deliberate: the least invasive test that can answer the question is done first, and the more invasive one follows only if it does not. Nobody goes to a general anaesthetic for a test that a needle could have settled.
This is a separate question from whether the cancer can be removed, and it is assessed separately. The purpose is to estimate how much lung function you would have left after an operation — not simply how much you have now. Someone whose breathing tests look poor may still have good reserve if the part being removed is barely working, and that is exactly what the more detailed tests are for.
In outline, the assessment involves:
Each of these is covered in detail elsewhere: fitness for lung surgery sets out the whole assessment in plain English, borderline lung function deals with the situation where the numbers sit near the threshold, and lung function testing is the clinical reference.
If you have been told that surgery is not possible, the most useful thing you can ask is which of the three questions produced that answer — because they are different kinds of no, with different likelihoods of changing.
This one often resolves. It usually means a sample has not been obtained, or an attempt was not diagnostic. There are several routes to tissue and one may succeed where another has not.
Worth two questions of its own: has the finding been confirmed by sampling, or does it rest on a scan? And has treatment before surgery been considered? Stage can change with treatment, and a scan finding is not the same as a sampled one.
Sometimes final and sometimes not. It depends on which tests were done, whether the calculation was of function now or function afterwards, and whether a smaller operation would achieve the same thing. Fitness for lung surgery →
None of this means a second opinion will change the answer, and often it does not. But knowing which question produced the no tells you whether anything is likely to change — and that is worth establishing rather than assuming. Lung cancer second opinion →
Questions asked by patients working through the investigations before a treatment decision is made.
Book an Assessment →Or call Jo Mitchelson:
020 7952 2882
Appointments at London Bridge Hospital and The Lister Hospital Chelsea are typically available within 2–3 working days. Bring your scans and reports. Dr Okiror reviews imaging personally and will tell you what has been established, what has not, and what would change the answer.
Jo Mitchelson, PA · 020 7952 2882 · pa@lungsurgeon.co.uk
St Thomas’ Hospital #1 UK · Guy’s Hospital #2 UK · London Bridge Hospital #10 UK · Newsweek World’s Best Hospitals 2026
X-ray, CT, PET-CT, MRI and VQ SPECT-CT — what each scan shows and cannot show
Mediastinal SurgerySurgery in the centre of the chest, including sampling of the mediastinal glands
ION Robotic BronchoscopyReaching small or deep nodules through the airways to obtain a diagnosis
Surgical Lung BiopsyKeyhole biopsy where a larger or more certain sample is needed
Biopsy & Surgery in One AnaestheticDiagnosis and treatment in a single visit for selected patients
Fitness for Lung SurgeryThe assessment that establishes whether an operation is safe for you
Shadow on a Lung ScanWhat the words on your scan report actually mean, and what happens next
Lung Cancer Second OpinionIndependent review of imaging, staging and the surgical recommendation