Preview Mode Links will not work in preview mode

Coda Change

Oct 24, 2016

Mervyn Singer discusses the use of biomarkers in critical care. Multiple biomarkers - physiological, biochemical, biological - can prognosticate early in critical illness, even in the ED. These biomarkers are numerous - lipids, progesterone, troponin, thyroid stimulating hormone, inflammatory cytokines, mitochondrial dysfunction… so on and so forth! Prognostication can happen as early as the Emergency Department. Studies from the States have found high levels of inflammatory cytokines can predict death, separately from clinical presentation. Therefore, we can predict when critically ill patients are destined to die. So, does this mean that we are just prolonging the life of those destined to die in critical care? Perhaps. Mervyn discusses this being the possible reason for many failed ICU studies. Concurrently, the only progress in critical care in the past 20 years may be due only to less iatrogenic harm. Furthermore, he explains his experiments with rats demonstrating the use of cardiovascular parameters, cytokines, troponins and even cholesterol being accurate prognostic biomarkers. Then, Mervyn goes on to identify the use of steroids in sepsis. He talks about research that demonstrates a benefit to steroid use, but only in those patients predicted to die using the aforementioned biomarkers. This could be a key to selecting an appropriate patient group to allocate a specific treatment too. Furthermore, we examine treating sepsis with beta blockers. Giving beta blockers to everyone has no effect at best and a harmful effect at worst. However, giving beta-blockers to those who were predicted to die conferred benefit! In conclusion, we can predict outcome early in disease. This may allow better selection of patients for certain treatments! We thus need to adopt a completely different strategy for such patients predetermined to die. This also applies to trial design, especially where survival is the endpoint.

For more like this, head to