“Investigators, led by Dr Harry Selker (Tufts Medical Center, Boston, MA), are pleased with the results, believing that after years of futile study, they have finally found some clinical evidence to support the experimental data suggesting that GIK [glucose-insulin-potassium] myocardial metabolic support could protect the heart in the ACS setting.”
…which lead to the press release tweet of “Intravenous GIK Slashes Death Risk in Acute Coronary Syndrome: CHICAGO – Glucose, insulin, and potassium given i…” by @ACEPNews. That press release can be seen here.
There have been trials enrolling over 20,000 patients to date that have been negative.
Despite all these previous negative trials, the authors believed the problem was timeliness – the critical time in which to provide metabolic support to the infarcting myocardium was in the prehospital setting, upon the earliest recognition of ACS. The original goal was to enroll 15,450 patients. They ended up with 880. Then, after data collection, they changed the primary endpoint from all-cause mortality to progression to myocardial infarction at 30 days and at 1 year. And they only have the 30 day data right now, they’ll get back to us with the 1 year outcomes. How this made the cut for publication in JAMA is outside the scope of my speculative powers.
So, they enrolled folks prehospital with signs and symptoms of potential acute coronary syndrome whose prehospital EKG was read as STEMI or met the ACI-TIPI prediction instrument probability threshold of 75%. They received the GIK solution with 90 minutes, on average. And, the primary outcome measure was negative for progression to MI, trend favoring GIK with OR 0.88 (CI 0.66-1.13). Negative for 30 day mortality, OR 0.72 (0.40-1.29). For STEMI patients, negative for progression to MI, OR 0.74 (0.40-1.38), and negative for 30 day mortality, OR 0.63 (0.27-1.49).
So, yes, there is a trend. And some subgroups even had significant trends in favor of GIK. But for JAMA and the rest of the internet to be promoting this as practice-changing at this juncture is absolutely inappropriate.
“Out-of-Hospital Administration of Intravenous Glucose-Insulin-Potassium in Patients With Suspected Acute Coronary Syndromes: The IMMEDIATE Randomized Controlled Trial”
http://jama.ama-assn.org/content/early/2012/03/21/jama.2012.426.full
One thought on “Glucose-Insulin-Potassium For MI?”
Comments are closed.
Thanks for pointing out the change from the original protocol – it had escaped my attention. (http://millhillavecommand.blogspot.com/2012/03/immediate-trial-shoukd-ems-give-glucose.html)
Still, the purported "positive" result is so weak that no amount of protocol-shifting could really substantively save the study.
It's bizarre how accepting the medical press is of the "spin" that the authors are presenting, especially when, in their conclusion, the authors demonstrate that they are acutely aware of the tenuous nature of their results.