Myocardial blood flow in patients with sepsis.

The high mortality of systemic infection results from its associated cardiovascular depression. Cardiac depression typically normalizes if the patient recovers from the infection, but recent studies suggest that these patients have an increased long-term risk of developing cardiovascular disease after the septic episode. These findings have reignited interest in sepsis-associated cardiac function and myocardial blood flow, which remain poorly understood in humans. We studied cardiac function and myocardial microvascular perfusion using gadolinium-contrast magnetic resonance imaging in a cohort of patients during the initial recovery period of sepsis (n = 16) or septic shock (n = 5) and after full recovery 1-2 months later (n = 13). In addition, hepatic, splenic, and renal cortical perfusion were also assessed. With infection, cardiac output (+27%), the rate-pressure-product (+21%), and the left ventricle (LV) peak-ejection (+36%) and peak-filling (+35%) rates increased compared to full recovery (all p < 0.05). Onset of LV myocardial perfusion and the time to peak LV myocardial perfusion of gadolinium-contrast occurred earlier during initial than after recovery, with a numerically higher LV myocardium wash-in rate (23 ± 22 vs. 14 ± 14 s^-1; p = 0.07). LV myocardial fibrosis was not seen in any patients. Renal cortical, splenic, and hepatic perfusion parameters including onset, time-to-peak, and wash-in rates of gadolinium contrast were comparable between initial and full recovery, except for a lower hepatic wash-in rate during initial recovery (13 ± 10 vs. 22 ± 16 s^-1; p = 0.03). Our study supports that myocardial microvascular dysfunction is unlikely to contribute to cardiovascular disease after severe infection. Conversely, hepatic hypoperfusion during sepsis may explain the commonly observed hepatic dysfunction in sepsis.