Dear Editor,

The editorials by Hoffman and Cooper,[1] and Fatovich[2] attempt to critically appraise the recently published International Stroke Trial 3 (IST3) stroke thrombolysis trial,[3] as well as provide personal opinions on the current status of tissue plasminogen activator (tPA) stroke thrombolysis. Unfortunately, these contain factual errors, statistical inaccuracies, and demonstrate a reliance on arguments that are perhaps of historical interest, but have long been discredited. The authors again remind us of the extensive discussions that took place in the late 1990s on the validity of the National Institute of Neurological Disorders and Stroke (NINDS) tPA stroke trial results that led to a statistical re-analysis of the original data.[4] This fully independent re-analysis reconfirmed (and indeed improved on) the original findings with a significant benefit for tPA stroke thrombolysis. In 2009, Hoffman and Schriger attempted an alternative re-analysis of the NINDS data.[5] However, this re-analysis demonstrated a failure to understand the nature of the data and used statistical methods that were flawed.[6] Many publications and systematic reviews have since followed, and have been accepted into international stroke treatment guidelines, including the Australasian College for Emergency Medicine.

The authors are critical of the methodology and inclusion criteria and analysis of IST3, all of which were appropriate, pre-planned, and indeed are recognised as strengths of IST3. The IST3 outcomes are further supported by an updated stroke thrombolysis systematic review and meta-analysis in an accompanying publication in the Lancet. It confirms the benefit from early treatment with tPA stroke thrombolysis, with a treatment effect very similar to the previous systematic review data. The authors, nevertheless, feel that systematic reviews in stroke thrombolysis are inherently unreliable. The totality of the evidence is generally regarded by expert clinical trialists and clinical trial epidemiologists as being best represented in rigorously conducted systematic reviews. Systematic reviews reduce random error (from underpowered trials) and avoid undue emphasis on small positive or negative studies. This form of analysis is therefore clearly acceptable, if not essential, in the statistical evaluation of the evidence of benefit of a given treatment. The authors comment that IST3 relied on a secondary exploratory analysis ‘to promote a supposedly positive outcome’. Again, this is inaccurate. It was a planned, specified secondary outcome, and clearly indicated in the pre-published statistical analysis plan, which the authors fail to reference.[7] Therefore, to suggest that tPA stroke thrombolysis is not supported by the totality of the evidence is clearly nonsensical. The authors also fail to understand the complex clinico-pathophysiology of cerebral reperfusion after tPA. It is overly simplistic to state that with ‘tPA caused more brain swelling which goes against any physiological mechanism of benefit’. The Australian EPITHET study, and associated publications, have demonstrated that the correlation between clinical outcome, vessel reperfusion and infarct size/swelling is variable and related to many factors (e.g. time to treatment, size of pretreatment infarction, presence of reperfusion). Indeed, the key issues in stroke thrombolysis now requiring research and critical scrutiny relate to using brain imaging to assist in patient selection for treatment, mechanical thrombectomy, reducing early intracerebral haemorrhage, and the search for newer and more effective fibrinolytic drugs.

It is perhaps an ‘inconvenient truth’ for these authors that stroke care has now well and truly moved on from these historical arguments. A simple question to be answered could be this: who is more likely to be correct? Is it the many international stroke societies, stroke organisations, medical colleges, government drug regulatory authorities, peer-reviewed journals, hospital ethics committees that all function independently and look at the totality of the evidence, and who have given approval for tPA stroke thrombolysis … or is it the authors of these opinion pieces (Hoffman and Cooper, and Fatovich)?

Competing interests

CB has received advisory board fees from Bayer Australia. CL has received consultancy fees from Boehringer-Ingelheim and is a member of the Clinical Council of the National Stroke Foundation. MP has received advisory board fees from Bayer Australia and honoraria from Boehringer-Ingelheim.


  1. Top of page
  2. References