Ken Milne, MD, is chief of emergency medicine and chief of staff at South Huron Hospital, Ontario, Canada. He is on the Best Evidence in Emergency Medicine faculty and is creator of the knowledge translation project the Skeptics Guide to Emergency Medicine.
Meta-analyses (x3) are cited in the 2018 AHA guidelines as supporting TPA in the 3-4.5 hour window. Incredibly, by my reading, none of the meta-analyses supports benefit:The Cochrane review plainly states no benefit out to 4.5 hours as the confidence interval crosses 1: Odd Ratio of 0.93 [CI 0.66 to 1.32])showed a symptomatic ICH rate of 5%does not include IST-3The first Lancet article only studys 0-3 and 3-6 hours, no mention of 3-4.5 hour window analysisThe second Lancet meta-analysis did show OR of 1.4 (CI 1.1–1.9) but this study does not include IST-3 in its analysis!!
Thanks for posting this information. SRMA are only as good as the studies included. These types of analysis have their own thread to validity. One is that biases in the original studies can be compounded by the SRMA process. It can result in a misleading point estimate with a significant p-value. This can give an illusion of certainty when certainly does not exist.
I have elicited criticism for sharing this article in EM department meetings. There is considerable inertia toward continuing to continue what “we” have been doing.
As Jerry Hoffman put it: “I wish I knew.”
3 Responses to “After Re-Analysis, No Trials Show Efficacy of tPA in Acute Ischemic Stroke”
October 3, 2020
Brian TenneyMeta-analyses (x3) are cited in the 2018 AHA guidelines as supporting TPA in the 3-4.5 hour window. Incredibly, by my reading, none of the meta-analyses supports benefit:The Cochrane review plainly states no benefit out to 4.5 hours as the confidence interval crosses 1: Odd Ratio of 0.93 [CI 0.66 to 1.32])showed a symptomatic ICH rate of 5%does not include IST-3The first Lancet article only studys 0-3 and 3-6 hours, no mention of 3-4.5 hour window analysisThe second Lancet meta-analysis did show OR of 1.4 (CI 1.1–1.9) but this study does not include IST-3 in its analysis!!
October 12, 2020
Ken MilneThanks for posting this information. SRMA are only as good as the studies included. These types of analysis have their own thread to validity. One is that biases in the original studies can be compounded by the SRMA process. It can result in a misleading point estimate with a significant p-value. This can give an illusion of certainty when certainly does not exist.
January 13, 2021
M Bruce ParkerI have elicited criticism for sharing this article in EM department meetings. There is considerable inertia toward continuing to continue what “we” have been doing.
As Jerry Hoffman put it: “I wish I knew.”