Annual chest x-rays to screen for tuberculosis (TB) should be stopped even though the disease is once again on the rise.

The call for a major change in hospital practice has been made by Professor Lee Reichman at the New Jersey Medical School. He says that there is no evidence that regular and expensive screening makes very much difference in detecting the disease.

The practice of annual chest x-rays in hospitals is widespread, even though several studies have been published that argue against its benefits.

X-rays would have to be performed "absurdly often, such as daily or weekly" if they are to detect TB, he says in an editorial published in the Lancet.

Prof Reichman's views are supported by the World Health Organization (WHO), by the US Food and Drug Administration (FDA) and by the Centers for Disease Control and Prevention.

The WHO estimated that radiographs, which are extremely expensive, represent up to 10 per cent of a nation's total expenditure on healthcare.

In a separate study, the FDA concluded: "In the absence of clinical symptoms, repeated chest radiographic examinations have not been shown to be of clinical value or productivity to justify their continued use."

Prof Reichman believes the practice is a throwback to the days before chemotherapy was used to treat TB, when it was thought that the risk of TB activation was life long, and so regular x-rays were needed to constantly monitor it.

But even by the early 1950s the practice was questioned because of the high incidence of incorrect readings, which were estimated to be as high as 43 per cent.

A large trial in 1963 found that bacteriological examinations were a better way of detecting TB, and even smear tests were picking up far more cases than were x-rays (Lancet, 1999; 353: 319-20).