Will surgery actually fix a blocked nose?

What the trial found, with its comparator attached

It recruited 378 adults from 17 NHS hospitals in Great Britain. Of those, 188 were randomised to septoplasty, with or without turbinate reduction on the other side, and 190 to six months of a steroid nasal spray with saline. It was open label, so nobody was blinded to which they had.

That is a large effect by the standards of surgical trials. The sample size had been built to detect a difference of 9 points on that questionnaire, so what was measured is more than twice what the trial was designed to find. The arm averages among those analysed were 19.9 and 39.5.

Who was actually in the trial

To be entered, a patient needed a septal deflection visible to a specialist at nasendoscopy and a nasal obstruction score of at least 30. An external bony deformity of the nose was an exclusion, and so was any previous septal surgery.

In their own limitations the authors record that more than 80 per cent of those recruited scored in the severe or extreme band, which they put down to milder cases settling in primary care or declining to take part. Elsewhere, answering a separate worry that the most deviated patients may have been under-referred, they say the trial recruited across a broad range of severities. Greater severity at the start was associated with greater improvement.

A blocked nose is not one condition

The septum is one cause. The nasal valve, the narrowest part of the airway, is another, and it can collapse inwards on breathing in while the septum sits straight. Turbinate enlargement and inflammation of the lining are others again, and more than one can be present at once. The trial could not exclude participants who also had rhinitis.

So the honest statement is about who was studied rather than who will fail. Everyone in that trial had a septal deviation a specialist could see. No trial we could open has asked the same question of a nose that is blocked for a different reason.

The American Academy of Otolaryngology, Head and Neck Surgery states that operations such as septoplasty may be complementary to nasal valve repair but are not effective substitutes for it, because they do not address valve dysfunction. That is a specialty body stating a position, not a trial reporting a result, and the asymmetry matters: septoplasty for a deviated septum has a large randomised trial behind it, while nasal valve repair, in the sources this page cites, has a position statement. More at nasal valve collapse.

What the same trial reports about harm

At six months, 7 of 174 participants in the surgery arm, or 4 per cent, had been readmitted to hospital with nasal bleeding, none of whom needed a further operation. Infection requiring antibiotics was reported by 20 of 172, or 12 per cent. Decreased sense of smell was reported by 19 of 171 and numbness of the upper teeth by 18 of 171, both 11 per cent. Six septal perforations and seven adhesions inside the nose were recorded at six and twelve months. Each of those figures carries its own denominator, because each item was answered by a slightly different number of people.

That last figure rarely appears in accounts of this trial and it should. A functional operation is still an operation on the shape of the nose, which is also why the two are so often planned together. See septorhinoplasty for what that involves.

The authors add a caution that is easy to miss: the operations in the trial were performed by experienced surgeons, whereas in NHS practice septoplasty is frequently performed by junior trainees, albeit often supervised. The trial measured what the operation can do, not what any given theatre list does.