Issue 25 / 14 July 2014

ABOUT 70 000 Australians have paid nearly $200 each for asymptomatic carotid artery stenosis screening, but a recent systematic review has found no evidence of benefit from such screening in the general population.

The review, published in the Annals of Internal Medicine and conducted for the US Preventive Services Task Force (USPSTF), examined evidence about screening and treating asymptomatic adults for carotid artery stenosis (CAS). It found several major flaws among 56 studies of CAS screening in the general population using ultrasound, and of carotid endarterectomy (CEA) or stenting for people with positive results. (1)

The reviewers wrote that duplex ultrasonography was a widely available, non-invasive screening test but its reliability was questionable because accuracy varied considerably among laboratories. Its use on a low-prevalence population also resulted in many false-positive results.

“Although our meta-analyses of [randomised control trials] that compared CEA with medical therapy found a reduction in perioperative stroke or death or any subsequent stroke (and other outcomes), the applicability of the evidence to current practice is substantially limited”, they wrote.

None of the trials reviewed compared endarterectomy or stenting with current best medical therapy using statins and antihypertensives, which had substantially reduced the ipsilateral stroke rate worldwide.

“The best recent evidence suggests that the incidence rate of ipsilateral stroke is nearing 1% per year, approaching the rate achieved in the surgical groups of trials that compared CEA with medical therapy”, the reviewers wrote. “This would significantly reduce the potential benefits of surgery.”

They warned of a very high false-positive rate and significant harms with screening. For instance, they estimated that among 100 000 adults in the general population, CAS screening would result in 940 true-positive results and 7920 false-positive results. If all positive tests were followed by angiography, as many as 1.2% of those people could have a resulting stroke.

Based on the review, the USPSTF has updated its recommendation statement, standing by its previous recommendation against screening for asymptomatic CAS in the general population. (2)

The statement is consistent with numerous international guidelines and with the RACGP Red Book. (3)

However, a company marketing CAS and other vascular screening in Australia told MJA InSight it did not accept the authority of the systematic review or the expert recommendations.

Screen for Life’s Australian medical director, Professor Scott Kitchener from Griffith University’s School of Medicine, said: “There are similarly ‘experts around the world’ who recommend for this type of screening, and evidence supportive of the benefits of the approach.”

He said 70 000 Australians had been screened by the company’s sonographers, mostly through a $199 vascular screening “package”, which included testing for CAS, abdominal aortic aneurysm, atrial fibrillation and peripheral artery disease.

In a statement with the company’s US-based chief medical officer Dr Andrew Manganaro, Professor Kitchener suggested the review’s conclusions were not relevant because they were based on patients undergoing vascular surgery after a positive result, whereas in reality, most positive cases would be treated with lifestyle interventions or medication.

“People with abnormal screening results are referred to their GP for diagnosis, monitoring and potentially treatment”, the statement said.

They cited an audit of patients who had attended Screen for Life as evidence that undergoing screening could “improve behaviour change in lifestyle interventions by making risk factors more tangible and real”.

However, this claim was dismissed by Dr John Quinn, director of vascular surgery at Brisbane’s Princess Alexandra Hospital.

“You don't need an expensive test to be told you need to change your lifestyle if you are over 60 and overweight”, Dr Quinn said.

“There is no evidence that general population screening is worthwhile.”

Professor Paul Glasziou, professor of evidence-based medicine at Bond University, Queensland, said he knew of several people who had received an “invitation” from Screen for Life, and said he would not personally undergo screening unless he developed symptoms, such as a TIA.

“When patients have asked me about screening, I have asked about their standard cardiovascular risk factors — blood pressure, cholesterol, smoking, diabetes, etc”, he said. “These are simpler [to assess], proven, and we can do something direct about them. I suggest they avoid additional screening.”

An Annals of Internal Medicine editorial accompanying the systematic review said the population attributable risk for stroke related to asymptomatic CAS was just 0.7%, and was dwarfed by other population attributable risk (PAR) factors such as hypertension (PAR > 95%), atrial fibrillation (PAR, 1.5%‒24%), cigarette smoking (PAR, 1.2%‒14%) and hyperlipidaemia (PAR about 9%). (4)

“All risk factors should be managed aggressively, regardless of the presence or absence of an asymptomatic CAS”, it said.

National Stroke Foundation Clinical Council member Dr Bruce Campbell said addressing hypertension, cholesterol and the usual risk factors was the appropriate course for patients worried about cardiovascular disease.

 

1. Ann Intern Med 2014; Online 7 July
2. Ann Intern Med 2014; Online 7 July
3. RACGP: Guidelines for preventive activities in general practice 8th edition
4. Ann Intern Med 2014; Online 7 July


Poll

Would you recommend screening for carotid artery stenosis in an asymptomatic patient?
  • No (68%, 54 Votes)
  • Yes - in certain circumstances (29%, 23 Votes)
  • Yes - if the patient asked (4%, 3 Votes)

Total Voters: 80

Loading ... Loading ...

5 thoughts on “Stroke screening “no benefit”

  1. Sue Ieraci says:

    DIdn’t we already know this? From Bayesian reasoning – the application of a test to a low prevalence population results in a high number of false positives, which can lead to more harm than might be prevented. I was taught to limit screening to a population with symptoms or a bruit. Plain CT for detection of calcium in asymptomatic people has shown us that almost all normal adults have coronary calcium. Invasive tests for coronary disease carry significant morbidity and some mortality. We must resist the temptation to use a test  ”just because it’s there”.

  2. colross@tpg.com.au says:

    Exactly, Sue. I raised this issue in the medical media last year when my wife, a healthy 45 yo, was mailed a personally addressed Screen For Life promo which contained no mention of any possible harms of screening. At the time, Screen For Life apologised, saying they promoted screening only to over-50s. But this year, at age 45, sure enough, I received my own promotional brochure.

    As for arguing “The review’s conclusions were not relevant because they were based on patients undergoing vascular surgery after a positive result”, they can’t have it both ways. Any population screening test is only justified if it leads to increased intervention which would not otherwise have occurred: indeed, Screen For Life uses this as the marketing basis for their product. So they can’t then discount a study showing poor outcomes for that treatment and instead claim we should instead only consider outcomes for lifestyle advice. As Prof Glasziou points out, that advice would have occurred anyway, without the screening.

    One can’t help wondering how many of the ‘experts around the world’ have financial ties to private vascular screening services.

  3. Scott Kitchener says:

    This is the fourth attempt to actually contribute to this discussion directly. The data was submitted twice to this journal to allow discussion of the findings to develop a way forward – and rejected as not being of priority to publish, yet criticism without the data seems to be a priority. It is extraordinary that this journalist piece is published referring incorrectly to the data and still the data isn’t requested for at least some informed comments. 

    The journalist piece also has misquoted the response to the evidence and even a response to clarify was not published. It will be interesting to see whether this is published to get at least some balanced discussion.

  4. Richard Gerraty says:

    I received the Screen for Life invitation in the mail. Having been once before invited by mail to enter a screening program to detect bowel cancer, I might have been forgiven for thinking that this too was evidence-based and government sponsored, but being in the trade I knew that this proposal was a commercial venture. Caveat emptor. While historically there have been trials demonstrating a modest benefit of surgery for asymptomatic carotid artery stenosis, it is far from certain that the benefit exceeds that of modern medical therapy. Selecting cases by transcranial Doppler ultrasound embolus detection might identify a higher risk group, but that is not a proven strategy either. Even with the cases that are turned up inadvertently a lot of harm can be done, including iatrogenic death and stroke. Certainly there is no evidence to support a deliberate population screening program for carotid stenosis.

  5. kryan says:

    Reply to Scott Kitchener

    Thank you for your comment.

    Re the priority of this topic for publication, MJA InSight is the MJA’s weekly news letter. For this news story we were reporting on a large review of the international literature and the recommendations of a peak body. We cannot comment on the quality or importance of papers submitted to the MJA and not published. The MJA editors are obliged to reject more than 80% of research papers they receive and the usual advice to authors of rejected articles is to seek publication elsewhere.

    We also thank you for providing our journalist with some comments. It is difficult to respond to your concerns about the accuracy of the story without specific examples.

    Ruth Armstrong

Leave a Reply

Your email address will not be published.