Endovascular treatment versus medical care alone for ischaemic stroke: systematic review and meta-analysis

MECHANICAL THROMBECTOMY IN ACUTE STROKE – GOOD DEAL FOR A FEW ? SOME MUSINGS ..

Mechanical thrombectomy for acute ischemic stroke due to proximal vessel occlusion is presently on a roll, with a string of sucessful trials this past year. The high level of scientific design, expertise required, transnational recruitment have added to the buzz, and definitely lifted the gloom which surrounds this serious condition.

The MR CLEAN trial was a great example of synergy between stakeholders, and benefits of political will; the results favoured intervention of thrombectomy over the iv t- PA control group in that 32.6 percent had a mRS (modified Rankin score 0 to 6, test for functional ability) score of 0 to 2 versus 19.1 percent in the control group.

However, the trial was not without problems – 9 percent of patients in the intervention group had embolization to a new vascular territory; 13 percent assigned to the intra arterial treatment also had to undergo a simultaneous second revascularization (acute cervical carotid stenting).The stringent requirements of the study and degree of expertise required by the interventionist , make mechanical thrombectomy an option for the select few patients who can access such a facility within the stipulated time.

As soon as the positive results were announced by the MR CLEAN trialists, the ESCAPE trial 2 was closed after an interim analysis. The ESCAPE study was well designed, recruited patients over a year across 22 centres worldwide to assess benefit of stentrievers in stroke due to proximal vessel occlusion; the study was stopped halfway due to achievement of pre stated objectives of 90 day mRS scores in the intervention group. The mechanical thrombectomy trials have been conducted meticulously with fantastic work flow models and a super short ‘picture to puncture time‘. The trials have shown that the complex technically demanding procedure can be done and is a definite option for the carefully selected patient who can access the services of a comprehensive stroke centre. On analysing the study using the GRADE assessment tool, there is a problem of indirectness - the patients were recruited into highly specialised centres where the investigators, trained by the device manufacturing companies, were asking a question which got answered through provision of an industry grant. The trial centres were across continents – each centre recruited an average of 1.5 patient per month, which suggests a rather thin spread, and pooling of data from different social and geographic groups.

Also, the reporting of benefits could be imprecise – by the unblinded patient whose reporting of health and functional condition could be influenced by their perception of the procedure; also, choice of grade between mRS 2 and 3 could be subjective in a given patient. It has been reported that sensitivity of assessment is reported as lower for scores 2 and 3 than other scores 4 . Suppose we redid the statistics clubbing numbers 0 to 3 mRS - would the results look different?? The number of patients with score 0 to 3 in the control group would be 44 out of 147, ie 30 percent, and, in the intervention group, 60 out of 164 ie 36 percent; the difference may not be significant.

The ESCAPE trial was closed early for benefit; the early closure of trials for benefit has caused objections in the past – McCartney 5 says premature stopping of clinical trials might pose a risk to patients and constitute a ‘leaping to conclusions‘. She quotes from Trotta’s 6 findings: ‘trials that were stopped early were used for registration purposes‘. G. Guyatt 7 reported a systematic survey which showed that trials stopped early for benefit yield treatment effects that are often not credible.

There is also the issue of equity - an article published in the NEJM carries significant clout in clinical decision making; the fear is that the industry and corporate hospitals might push this expensive intervention; the perception of the public and policy maker might favour the intervention and cause shift of focus and scarce resources in the poorer countries, from prevention and primary care to sophisticated intervention and cause further imbalance of equity.

These recent successful trials with new generation stentrievers have resulted in a strong recommendation by the American Heart Association 2015 guidelines 8 for mechanical thrombectomy in a carefully selected group of patients. I wonder whether it is at all appropriate to point out that all the trials have been sponsored by the same company Covidien, and whether this could be a basis for bias?

In conclusion, a fair attitude for the individual clinician would be a balance of optimism and informed scepticism.

REFERENCES

1. Berkhemer OA, Fransen PSS , Beumer D ., for the MR CLEAN Investigators ; A Randomized Trial of Intraarterial Treatment for Acute Ischemic Stroke N Engl J Med 2015 ; 372 : 11-20 .

2 . Goyal M . , Demchuk A. Menon B . Randomized Assessment of Rapid Endovascular Treatment of Ischemic Stroke N Engl J Med 2015 ; 372 : 1019-1030 .

3. Balsheim H . , Holland M . , Schunemann GRADE guidelines : 3 . Rating the quality of evidence . J Clin Epidemiol.2011 Apr ; 64 ( 4 ) : 401-6 .

4. Savio K . Pietra G.L.D. and Leone M. Reliability of the modified Rankin Scale applied by telephone .Neurol Int .2013 Feb 11; 5 (1) :e2

5 . McCartney M . Leaping to conclusions BMJ 2008 : 336 :1213

6 .Trotta F , Apolone G , Garattini S , Tafuri G . Stopping a trial early in oncology : for patients or for industry ? Ann Oncol 2008 .Apr 9 [ Epub ahead of print ] . [ PubMed ] .

7. Guyatt GH , Mathias B , Glasziou P , Problems of stopping trials early BMJ 2012 ; 344 :e 3863

8. Powers W . J . A Guideline for Healthcare Professionals From the American Heart Association/ American Stroke Association Stroke . June 2015 ( published online )