Profess finally published

Sacco RL et al. NEJM 359:1238-1251 2008
Yusuf et al IBID 1225-1237 2008
editorial DM Kent and DE Thaler IBID 1287-88

Aggrenox v. clopidogrel (Sacco et al.) As a neurologist not a statistician I will cut to the chase. They studied over 20,000 patients in a double blind 2x2 factorial design, randomly assigned to aggrenox or plavix. There was not evidence of the trial meeting the predefined criteria for noninferiority and there was no evidence that either treatment was superior to the other in the prevention of recurrent stroke.

Telmisartan to prevent recurrent stroke and cardiovascular events (Yusuf paper). Same 20,332 patients were randominzed to telmisartan 80 mg/day v. placebo. Interval to randomize was 15 days. Therapy continued 2.5 years . There was no evidence that therapy with telmasartan lowered the rate of recurrent stroke, major cardiovascular events or diabetes.

Editorial note.(Kent and Thaler). titled "Insights from Incoherence" authors properly note the tendency of some to make indirect comparisons based on different clinical trials (showing a better effect of aggrenox in this case) were mistaken. They also properly note that the "compelling logic of the transitive property, so reliable in mathematics, has little authority in the often illogical world of clinical trials" (I love it, but credit goes in a cite to Baker and Kramer). However Kent and Thaler liken the result to the Zen koan, a result incomprehensible to rational understanding, the contemplation of which may lead to a deeper realization. They summarize realization with the original haiku "For stroke prevention/use an antiplatelet drug/Treat hypertension."

Robert Frost, look out!

Late alteplase 3- 4.5 hours

Two articles and one editorial
Hacke W. et al. NEJM 359:1317-1329 2008. Also called ECASS III
Trial replicated in many ways the seminal NINDS trials except studied the 3-4.5 hour period. 821 patients were enrolled and randomized 1:1 with primary endpoint being favorable outcome at 90 days (modified Rankin score of 0 or 1). Safety endpoints were death, symptomatic ICH and other SAE's. Median time of enrollment was 3 hours 59 minutes. Treated group had more favorable outcomes (52.4 v. 45.2 %) or stated otherwise number to treat to have one more good outcome is about 14. sICH occurred in 2.4 v. 0.2 % (NTT for one more symptomatic hemorrhage is 45). Overall hemorrhage was 27 v. 17.6 % (NTT for one more hemorrhage is about 11).

Criticism-- did not study severe strokes with high NIHSS scores. Editorial by Lyden P NEJM op cit. 1393-5 also noted study tended not to enroll as many diabetics. Editorial also pointed out data showing earlier treatment is more beneficial earlier and less beneficial later; treatment in 3-4.5 hour bloc is successively less successful than treatment in each earlier time bloc. Authors allowed use of small doses of SQ heparin after alteplase to prevent DVT's.

The article itself explains that prior trials of 3-4.5 hour window ECASS II and Atlantis, were not powered to detect a positive finding in 7-10 % range.

Alteplase in first 90 minutes after stroke is nearly twice as effective as in second ninety minute bloc. Odds ratio for global outcome in first 90 minites is 2.81, in second ninety minutes is 1.55, in the third 90 minutes is 1.40 (based on pooled analysis of prior studies) and in current study is 1.34 in third 90 minutes. Time to needle is still paramount.

The previous article from Lancet cited (Wahlgren et al. Lancet 2008, dubbed the SITS-ISTR trial Safe Implementation of Thrombolysis in Stroke) compared 664 patients receiving alteplase in 3-4.5 hour time period to 11,685 who received it in 3 hour window. The outcome was symptomatic intracranial hemorrhage in first 24 hours with NIHSS deterioration of 4 points or more, and 90 day outcome based on modified Rankin score of 0-2. The result was no difference in any outcome measure and the conclusion is that alteplase is safe when given 3-4.5 hours.