Health Check: 'Tonsillectomies show the fine line between doing good and harm'

A tonsillectomy is one of the simplest and safest of operations - or should be. Not for Crawford Roney, though, a tousle-haired two-year-old who died after a tonsil operation at a private hospital in 2001.

Medical care is supposed to have become progressively safer, as knowledge and experience have grown and science has advanced. But sometimes treatment becomes more risky.

About 45,000 tonsillectomies are performed each year, mostly on children. This is far fewer than in the operation's heyday in the 1950s, but it is still a common procedure. The cause of Crawford's death was a massive haemorrhage which occurred a few days after the operation while he was at home in his cot. The surgeons blamed the use of disposable instruments, introduced on the orders of the Department of Health to reduce the potential risk of transmission of variant CJD, the infective agent of which had been found in tonsils.

Six months later Eileen Basham, 33, from Middlesbrough died in similar circumstances following a tonsillectomy. The Department of Health recommended the withdrawal of disposable instruments on the grounds that their use had increased the known risk of haemorrhage.

Crawford's parents have campaigned for an independent inquiry into the introduction of disposable instruments and his father, Peter Roney, has said he believed they were introduced out of political expediency because ministers felt they had to do something about potential CJD risks.

Now they have some important ammunition for their campaign. Ministers ordered an audit of tonsillectomies which has already revealed two disturbing findings. The National Institute for Clinical Excellence (Nice) which commissioned the audit, decided the findings were sufficiently worrying to issue an urgent alert to all surgeons last week. First, the audit found one in 10 surgeons is still using disposable instruments, despite being told to stop doing so. But second, it also revealed that operations involving "hot" techniques - diathermy and coblation, which use heat to burn through tissue and seal blood vessels - carried at least double the risk of haemorrhage compared with the the traditional "cold steel" method in which a scalpel is used for cutting and the blood vessels are then tied off or packed.

Hot techniques are widely used in many types of surgery and because of their success there was no reason to think they would not work equally well in tonsillectomy. But this has turned out to be a fatal assumption.

The audit of 12,000 tonsillectomies carried out between July 2003 and February 2004 found 389 (3 per cent) suffered a haemorrhage. Two-thirds of those (241) had to return to the operating theatre for further surgery.

Totalled for the whole country that is nearly 1,000 children and adults a year who suffered bleeding so badly following the operation that they had to be rushed back to theatre, while anxious parents and relatives waited. Between 600 and 700 of them might have avoided the trauma - and danger - of a haemorrhage if they had been operated on by the cold steel method.

It is thought the danger of haemorrhage is higher in the tonsils because of the presence of bacteria in the throat and food's passage down the throat.

Nice stressed its warning only applied to the use of diathermy in tonsillectomies. Cauterising remained a safe way of sealing blood vessels in other circumstances, it said.

The episode illustrates the fragile balance between doing good and harm in medicine - and the need to ensure the balance is maintained in favour of the patient. Although tonsillectomies are less safe, complex children's heart surgery has improved.

The Bristol children's heart surgery disaster of the 1990s exposed the way that surgeons were operating without proper monitoring of their performance. This month the British Medical Journal published the first fully validated results from the 13 children's heart units in the UK, showing all units performing well with a better than 91 per cent survival rate one year after surgery.

It is the first time such figures have been reported anywhere and they show there were no bad units. While Nice's alert last week demonstrated how a simple operation could become more dangerous, the heart surgery audit has shown how, out of one of medicine's worst tragedies, has come a world-class service.