Dr Udesen opened her talk by stating the clinical relevance of testing vasoactive drugs concomitant with mechanical unloading in cardiogenic shock (CS). She highlighted that, according to the 2017 European Society of Cardiology guideline for the management of acute MI (AMI) in patients with ST-segment elevation MI, both inotropes and short-term mechanical support devices might be considered for haemodynamic stabilisation (Class IIb); however, the level of evidence is weak.1,2 A recent publication assessing the temporal trends of CS in AMI in Denmark reported the increasing use of mechanical circulatory support with Impella devices, and the use of ≥1 vasoactive drug in about 90% of cases.3 She described the clinical conundrum involving patients with CS and low perfusion pressure and the decision to use an inotrope or vasoconstrictor to increase end-organ perfusion. This increase in perfusion pressure comes at the expense of increased cardiac afterload, and hence an increase in left ventricular (LV) workload.
To aid in the choice of the vasoactive drug, Dr Udesen compared the effect of norepinephrine (NA), epinephrine (AD), dopamine (DA) and phenylephrine (PE) on pressure–volume area (PVA), LV workload (product of PVA and heart rate) and metabolism in a pig model of ischaemic cardiogenic shock supported by Impella CP (n=10).
CS was induced using stepwise injections of polyvinyl microspheres and was defined as mixed venous oxygen saturation (SvO2) to <30% or ≤50% of baseline value and/or sustained cardiac index <1.5l/min/m2 for ≥10 minutes. The multiple steps of the experiment included instrumentation, development of CS, initiation of support with Impella CP for 30 minutes, administration of minimal NA (if the mean arterial pressure declined to <50 mmHg), blinded crossover to drug infusion for 30 minutes each (AD 0.1 µg/kg/min, NA 0.1 µg/kg/min and DA 10 µg/kg/min), PE infusion for 20 minutes and euthanasia. A linear mixed model was constructed using individual animals as subjects for random factors and sequential experimental stages as fixed repeated measurements. The reference time was set to 30 minutes after initiation of Impella CP support.
Concomitant administration of NA with Impella CP resulted in a leftward shift of the pressure–volume loop (PVL) with an increase in stroke work. Similar results were observed with DA (slightly more pronounced) and AD. However, concomitant administration of PE with Impella CP resulted in a rightward shift of the PVL, with an increase in end-diastolic pressure (LVEDP). Compared to treatment with Impella (reference), AD increased heart rate by 1.2-fold, DA and PE by about 1.4-fold, while no difference was observed with NA. LVEDP increased significantly only with PE and not with reference and other vasoactive drugs. In contrast, LV stroke work increased to different degrees with AD, DA and NA, with no difference with PE. Also, the PVA increased with all four drugs, compared to treatment with Impella. These data indicate that the total LV workload increased with all four drugs compared to treatment with Impella alone, and reached statistical significance in all drug treatment groups, except NA.
The arterial lactate concentration and renal and cerebral oxygen saturation were measured to assess the status of end-organ perfusion. Compared to reference, SvO2 increased with AD, DA and NA, but decreased significantly with PE. Treatment with PE significantly increased arterial lactate levels and decreased renal venous oxygen compared to treatment with Impella. Treatment with DA significantly increased cerebral SvO2 compared to treatment with Impella.
Taken together, the results suggest that if the perfusion pressure remains low after initiating support with Impella CP in CS, NA should be the first vasoactive drug of choice because it exhibits mild inotropic and potent vasopressor effects. The use of PE, which exhibits only vasopressor effects, should be avoided.