Why we can’t trust clinical guidelines

We would like to point out some significant inaccuracies in Lenzer’s1 interpretation of the accumulating trial data on alteplase for acute ischaemic stroke. As Lenzer herself states, the most powerful evidence comes the overall estimate from the Cochrane pooled analysis, yet she cites selectively from the 2009 Cochrane review on the well-recognised hazards: ‘…that showed alteplase increased fatal intracerebral haemorrhage nearly fourfold, and that thrombolytics overall were associated with a significant increase in mortality by the end of follow-up, representing an extra 30 deaths per 1000 treated patients,’ but did not cite the conclusions of that review: ‘Overall, thrombolytic therapy appears to result in a significant net reduction in the proportion of patients dead or dependent in activities of daily living. This overall benefit was apparent despite an increase both in deaths (evident at seven to 10 days and at final follow-up) and in symptomatic intracranial haemorrhages.’2

Furthermore, she did not cite the conclusions of the more recent 2012 update of the review on the effects of alteplase (12 trials, 7012 patients).3 showing no effect on death at final follow-up (679/3548 [19.1%] vs 640/3464 [18.5%], odds ratio 1.06, (95% CI 0.94–1.20, p=0.33). Furthermore, these analyses showed that, despite the early risk of intracranial haemorrhage, there was a net benefit at final follow-up; treatment with rt-PA within 6 hours of stroke significantly increased the odds of being alive and independent (modified Rankin Scale, mRS 0–2) (1611/3483 [46.3%] vs 1434/3404 [42.1%], OR 1.17, 95% CI 1.06–1.29; p=0.001), an absolute increase of 42 (19–66) per 1000 people treated. We emphasise that this evidence of overall benefit up to six hours is not the product of some data-dependent subgroup as Lenzer seeks to imply, but rather the primary pre-specified outcome specified in the protocol for the review.

Despite her assertion that the best evidence comes from the overall estimate from a Cochrane review, Lenzer supports her argument that there is no clear evidence of benefit from alteplase by counting the number of trials that achieve statistical significance at the arbitrary level of p< 0.05. She cites a (non-peer reviewed) website to note that ‘only two of 12 randomised controlled trials of thrombolytics have shown benefit and five had to be terminated early because of lack of benefit, higher mortality, and significant increases in brain haemorrhage.’

This is inaccurate in two major aspects. Firstly, the number of randomised trials of thrombolytics for stroke is incorrect - the Cochrane 2009 review included 26 trials involving 7152 patients. The second is that stating ‘5 trials stopped early’ is misleading, since three of the trials stopped for futility or hazard were trials of agents other that alteplase (streptokinase and desmoteplase), neither of which are approved for use in stroke. The correct number is that two of the 12 trials of alteplase were terminated (ATLANTIS-A and ATLANTIS-B), involving only 485 of the 7201 (7%) patients included in the 2012 updated systematic review.

Lenzer also incorrectly states that ‘the [AHA Stroke] guideline committee did not include the largest study [IST-3] of the treatment to date in its analysis.’ The published guideline does indeed cite both IST-3 and Wardlaw’s updated 2012 systematic review of the trials of rt-PA.4

The evidence on the effects of treatment accumulates, and so the best evidence comes from a properly conducted and up-to-date systematic review of the high quality evidence. While we accept that undeclared conflicts of interest may bias guidelines, we also must point out that selective citation of parts of the evidence is an even more important source of bias. In our view Lenzer does not take a balanced view of the totality of the evidence on alteplase for stroke. Despite the hazards, there is no increase in deaths and overall treatment with alteplase is associated with reduced disability and dependency.

Peter Sandercock, DM, FRCPE, FMedSci, Professor of Medical Neurology, University of Edinburgh. Peter.sandercock@ed.ac.uk

Joanna Wardlaw MD, FRCP, FMedSci, Professor of Applied Neuroimaging, University of Edinburgh. Joanna.wardlaw@ed.ac.uk

Richard Lindley, MD, FRCPE, Professor of Geriatric Medicine, Sydney Medical School – Westmead Hospital and The George Institute for Global Health, University of Sydney, Australia. richard.lindley@sydney.edu.au

Competing interests
JMW, PS, RIL are all co-chief investigators in the IST-3 trial, which was funded by the UK MRC, the Health Foundation and the Stroke Association, and the Australian Health and Medical Research Council. JMW is the author of the Cochrane Review. JMW was a member of the expert neuroradiological adjudication committee for IST-3 and ECASS 3. All three authors have received lecture fees and travel expenses from Boehringer Ingelheim, the manufacturers of alteplase, for occasional lectures at conferences. None of the authors are on industry expert advisor boards or speakers panels or have any relevant financial or other competing interests

Reference List

(1) Lenzer J. Why we canGÇÖt trust clinical guidelines. BMJ 2013 June 14;346.
(2) Wardlaw JM, Murray V, Berge E, del Zoppo GJ. Thrombolysis for acute ischaemic stroke. Cochrane Database Syst Rev 2009;(4):CD000213.
(3) Wardlaw JM, Murray V, Berge E, del ZG, Sandercock P, Lindley RL et al. Recombinant tissue plasminogen activator for acute ischaemic stroke: an updated systematic review and meta-analysis. Lancet 2012 June 23;379(9834):2364-72.
(4) Jauch EC, Saver JL, Adams HP, Bruno A, Connors JJ, Demaerschalk BM et al. Guidelines for the Early Management of Patients With Acute Ischemic Stroke: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke 2013: 44 (3): 870-974.