Levosimendan in early experimental sepsis: effects on the heart and hepatosplanchnic circulation

Sepsis-related cardiovascular dysfunction associated with fluid-unresponsive tissue hypoperfusion might require inotropic treatment. This cardiovascular dysfunction seems to involve calcium desensitization and adrenergic unresponsiveness. We investigated the effects of clinically relevant plasma concentrations of the calcium sensitizer levosimendan during the first 6 h of endotoxemia in a model of experimental sepsis in pigs (21.8-44.0 kg). Levosimendan given to endotoxemic pigs receiving moderate volume resuscitation elicited tachycardia, hypotension and myocardial ischemia, as evidenced by a negative myocardial lactate flux (study I). Likewise, compared to non-treated controls, levosimendan did not improve systemic or hepatosplanchnic perfusion during endotoxin shock: the animals developed signs of tissue hypoperfusion with elevated blood lactate and low oxygen venous saturations (study II). Aggressive volume resuscitation before levosimendan treatment induced a hyperdynamic state that was sustained by levosimendan and norepinephrine treatment, whereas control animals gradually developed shock. Nevertheless, splanchnic blood flow was redistributed and the superior mesenteric artery, the hepatic artery and the portal vein blood flows (PVF) decreased. Similarly to study II, there were signs of systemic and hepatosplanchnic tissue hypoperfusion. In contrast, administration of dobutamine and norepinephrine, increased cardiac output (CO) and oxygen delivery, maintained PVF and improved tissue perfusion (study III). Load independent measurements of cardiac function showed that systolic function was actually enhanced during the first 2 h of sepsis, whereas diastolic function was depressed in both ventricles. The initial decrease seen in CO was a result of volume depletion, and recovered with aggressive volume resuscitation. Although endotoxin-induced lung injury caused early right ventriculovascular uncoupling and increased right ventricular (RV) myocardial oxygen demand, right coronary artery blood flow improved markedly with resuscitation, maintaining adequate myocardial perfusion (study IV). In resuscitated septic pigs, levosimendan supported RV function by increasing RV contractility at a low energy cost. CO and left ventricular ejection fraction increased, and right ventriculovascular coupling and mechanical efficiency tended to improve (study V). In conclusion, early treatment with levosimendan during resuscitated sepsis can increase CO and improves RV contractility at a low energy cost, but it does not improve hepatosplanchnic perfusion significantly, which is better achieved with dobutamine-norepinephrine. In addition, because levosimendan is an inodilator, its use in sepsis should be restricted to thoroughly fluid-resuscitated subjects.