Journal
Notes on the evidence, policy and practice shaping thoracic surgery

The money in longevity is chasing a way to regrow the thymus. The two things that already work are quieter — keeping the organ when we operate, and reading it on the scans we already take. What the new evidence does, and does not, change in the operating theatre.

MARS2 found against surgery for pleural mesothelioma and changed British practice within months. Its own lead recruiter then restaged half the surgical arm and found that only one in three would have been operated on under current criteria — and that they lived far longer than the rest. What that changes, from a surgeon who stopped.

Severe emphysema is two problems in one — the exacerbations and the mechanics — assessed by different people, never in the same room. New biologics that settle eosinophilic exacerbations can move a patient back across the line into candidacy for endobronchial valves or surgery. The narrow subgroup it applies to, and why it matters.

A woman whose breathlessness and chest pain arrive with her period — inhaler doing nothing, lung-function tests normal — is carrying a diagnosis written in her menstrual history. Why thoracic endometriosis is still read as asthma, and reaches a surgeon years too late.

A new British Thoracic Society quality standard puts a clock on the referral to thoracic surgery — 48 hours when pleural infection fails to drain, 24 in a prolonged air leak. What that changes, seen from the end that receives the call.

A study at this year's ASCO meeting linked weight-loss jabs to slower cancer progression. For a surgeon the useful question is a different one — what these drugs mean on the morning of an operation, and why a treatable cancer should not be left in place while the evidence catches up.

Two studies this month seem to reach opposite verdicts on medical AI. They do not: one tests the models on their own, the other against the products built for doctors. What neither measures is the judgement that decides who should have an operation.

A screening scan that finds a nodule sets off a sequence the patient never sees. A new Society of Thoracic Surgeons consensus, and fast-moving evidence on biopsy, are redrawing the surgeon's first job — from operating to find out, to making sure no one has to.

The government has pledged £20 million to put AI chest X-ray into every NHS trust by 2029. The LungIMPACT trial shows why a faster reading does not become a faster diagnosis — and why better detection lands, ultimately, on the surgical pathway.

Lung volume reduction surgery still carries the reputation it earned in 2003: dangerous, narrow, a last resort. Modern selection, imaging and recovery have moved the operation on — and the reputation has not kept up.