Each year, according to the World Health Organisation, more than 300m operations are performed worldwide. In England about 60% of people can expect to go under the knife at least once in their lives. Some of this work plainly saves lives: transplanted organs, cut-out tumours. But as the numbers grow, a few people inside the profession have begun asking how much of the rest is necessary at all. The answer, when it has been sought honestly, is less comfortable than anyone in a white coat would like.
For a long time the answer could not be had, because surgery was almost never put to the test that drugs must pass. Few randomised controlled trials were done on operations, so nobody could say whether patients fared better with surgery than without it. That is changing. The number of surgical trials funded by Britain’s National Institute for Health and Care Research rose from 34 in 2011 to 188 in 2023, and national surgical-trials programmes now run across Europe and in Australia and Canada.
What the trials have found should give any prospective patient pause. Removing an inflamed appendix — an operation 5-10% of people have had — turns out to be no better for most patients than a course of antibiotics. Cheaper, simpler techniques have beaten dearer, more complex ones. And some of the most widely used procedures, spinal fusions and rotator-cuff repairs among them, produce effects indistinguishable from placebo or from non-surgical care such as physiotherapy.
Some operations do worse than nothing. A British trial found that prostate-cancer surgery had no effect on mortality 15 years on, while worsening sexual and urological problems. A Finnish trial of a common knee procedure concluded that, ten years later, patients who had the real operation had more trouble in the affected knee than patients who had been anaesthetised, given a superficial cut and sewn up again. David Ring, an orthopaedic surgeon at the University of Texas at Austin, reckons most operations in his field may be unnecessary.
The patient heals; the surgeon bills
How, then, do so many patients leave the operating table feeling better? In 2022, in the journal JAMA, researchers pooled data from 100 surgical trials covering 32 interventions. Two-thirds of the improvement patients reported, they calculated, came either from the body healing itself or from the placebo effect of being prepared, swabbed and cared for. One-third came from the procedure itself. “Surgeons think they’re effective because they are administering a potent placebo,” says Seth Leopold, an orthopaedic surgeon at the University of Washington.
The placebo can be remarkably elaborate. In a landmark 2002 trial of 165 patients, led by Bruce Moseley of the Baylor College of Medicine in Houston, the surgical team splashed water in a dish to mimic the sound of debris being flushed from the arthritic knee — on the chance that even an unconscious brain might register it. The sham operation relieved pain exactly as well as the real one. And in the JAMA study, where real and sham surgeries could be compared with ordinary non-surgical care, the placebo explained only a small part of the sham group’s improvement. Most of it was the body, given time, mending on its own.
You might expect such findings to change what surgeons do. They have not, much. Surgeons go on performing operations shown to be no better than placebo at the same rate as before. Ian Harris, a surgeon at the University of New South Wales in Sydney, says they find it hard to accept that work they have done for years, and which seems to help, is useless. In 2014 a group of surgeons wrote in the journal Arthroscopy to protest that in sham-surgery trials “Patients who may not be of entirely sound mind are selected as research subjects, and research performed on such individuals would not be generalisable to mentally healthy patients.” It is an old story: when the facts threaten the trade, the facts are said to be mad.
“You have to remember surgeons have no uncertainty,” says Jane Blazeby, a surgeon at the University of Bristol who has run many such trials. “They always know what’s best.” The certainty is not earned. Research on one common knee operation found that surgeons’ predictions about which patients would improve were as accurate as a coin toss. Stefan Lohmander of Lund University in Sweden points out why the illusion persists: the patients who come back for check-ups are the ones who got better, and many of them do not want to disappoint their doctor.
Dr Blazeby got her colleagues to enrol patients in trials only by teaching them to “confidently be unconfident” — to tell patients that an operation performed for years might not be right for them. There are signs of movement. Studies in America and Australia find that younger surgeons are reader to abandon a procedure a trial has discredited; the longer a surgeon has done it, the less likely he is to stop.
Money explains much of the rest. Discredited operations wither fastest where surgeons are salaried, as in Britain and Scandinavia, rather than paid per operation as in America, where a surgeon is in effect a small business. After trials found that shaving a bone spur off the shoulder relieved pain no better than sham surgery, British operations for it fell from 28,000 in 2016-17 to 5,720 in 2019-20. In America the procedure stayed popular.
Patients, too, press for the knife. Scans show “abnormalities” that Dr Ring compares to wrinkles and grey hair — signs of age, not disease, and usually not the cause of the pain. Studies find such abnormalities in the knees, backs and shoulders of 20-40% of people with no symptoms at all. Yet the promise of the quick fix wins. “It is becoming more and more difficult to say ‘no’ to the patient,” says Dr Lohmander. In America some surgeons fear a bad online review, and insurers still pay for ineffective procedures lest customers defect.
More trials would stiffen spines. For now, most non-essential operations remain untested against either sham surgery or simple non-surgical care; a 2020 review found that just 1% of the trials on surgery for chronic musculo-skeletal pain compared operating with not operating at all. The old oath binds doctors to do no harm. Keeping it begins with a plainer question than the profession likes to ask: how much good is the knife actually doing?