A bunch of clever people at Tufts Medical Centre in Boston, MA, have attempted to influence the outcome of patients calling their community Emergency Medical Service (Ambulance service) for apparent Acute Coronary Syndrome by applying physiological principles and some pre-existing suggestive evidence. They called it the IMMEDIATE Trial and it was published online by JAMA in the end of March. It offers some interesting possibilities. In summary, the IMMEDIATE Trial is a US-based, multi-centre, double blind, placebo-controlled trial (MC-DB-PC) of very early administration of 30% Glucose + 50iU Insulin + 80mEq K+ in 1 litre of 0.9% saline at 1.5ml/kg/hr for 12 hours to patients attended by an ambulance with apparent ACS symptoms. The inclusions were broad and the exclusions few (<30yo, pregnant, active heart or renal failure). The baseline characteristics were even. The median time to drug administration was 90 minutes, with only 6% receiving the therapy within 30 minutes (Hope that does not reflect transport times, but then suggests most patients would not have started their infusion until they were in the ED, which wasn’t really the point). There was no difference in 30 day mortality despite a statistically significant difference in the combined outcome of cardiac arrest and hospital mortality, with the majority of the benefit in the cardiac arrest group, patients > 65yo and patients whose infusion was started within 60 minutes. It was also despite a measureable reduction in infarct size and free fatty acid levels and improvement in LVEF in the study group. Adverse effects were minimal, with no increased incidence of pulmonary oedema nad only a slight increase in incidence of hyperkalaemia >5.5mmol/L. There was a significant increase in the incidence of hyperglycaemia in the GIK group, which could be important, as we know from the DIGAMI (Circulation. 1999;99:2626–2632) and ICONS (J Am Coll Cardiol 2002;40, 1748-1754) studies. Looking through the numbers, there is probably something in it, though the NNT might be quite high for a small benefit. The NNH however, while probably small for serious harm, is hard to be clear about. The lack of 30-day mortality benefit is a major concern but might be too crude on its own and might reflect the study power. As a demonstration of how hard it can be to successfully recruit what might seem to be easy numbers, over 50,000 patients were screened, only 1,400 were elligible, 1,087 were enrolled, 911 got past protocol violations (pre-hospital trials are notoriously difficult to control) and finally 871 were analysed by ITT (411 GIK, 460 placebo). So, should this strategy be employed by pre-hospital emergency medical services? This study probably doesn’t offer enough to change practice. The failure to improve 30-day mortality despite any of the other improvements will be the main barrier, especially as previous studies of GIK for ACS have shown no benefit; although the GIK was usually started later in the management pathway. Why did this trial see a reduction in infarct size, less functional impairment, less post arrest arrythmias and lower in-hospital mortality, but no change in mortality at 30 days? We don’t have that answer. So there will probably be more trials in this area.
You can read the online JAMA article here. You can read the slides used for the presentation of the IMMEDIATE trial at the American College of Cardiology Meeting in Chicago here.
|
- Media
- Upskill