Tissue perfusion assessment at bedside in critically ill patients still depends largely on measurement of global hemodynamic variables such as arterial pressure, cardiac output and mixed venous oxygen saturation (SvO2). However, alterations in distribution of regional perfusion and, even more importantly, alterations in microcirculatory perfusion often occur and impair tissue perfusion even when global perfusion is preserved. How can we measure the microcirculation? Recently developed tools to assess the microcirculation include small hand held microscopes, such as orthogonal polarization spectral (OPS)/sidestream dark field (SDF) imaging (primarily in the sublingual region). These allow direct visualization of red blood cell flowing in the microcirculation, and hence are particularly useful to evaluate perfusion and in particular density of perfused vessels and heterogeneity in perfusion. Of note, white blood cells and platelets are hardly detected. These microscopes can be applied of tissues covered by a thin epithelial layer. In humans it was mostly used in the sublingual area. In the operating field, it can be directly applied to the heart, but great caution should be used to minimize motion artifact and to discard bloody secretions. The use of near-infrared spectroscopy (NIRS), can also be useful but it requires performance of a transient vascular occlusion test to evaluate microvascular reactivity. Hence, it is used in primarily in the thenar eminence. Measurement of tissue oxygenation with actual tools, including NIRS, without performance of vascular occlusion test is of limited interest. Indeed, this is a regional measurement of oxygen saturation but this measurement aggregates multiple vessels O2 saturation and fails to evaluate microcirculatory perfusion especially when perfusion is eterogeneous. Extensive discussion of the different techniques can be found in a recent review paper [1]. In humans, it has been shown that microcirculatory alterations occur in patients with severe sepsis [2] and in patients with severe heart failure [3]. These alterations were no related with systemic hemodynamic variables [4]. These alterations are characterized by heterogeneity of perfusion with capillaries with stop flow in close vicinity to well perfused capillaries. Different mechanisms have been implicated in the development of these alterations including loss of communication between vascular segments, impaired endothelial vasoreactivity, alterations in red and white blood cells rheology, alteration in endothelial glycocalyx, platelet aggregation and microthrombosis. What are the consequences of theses alterations in microvascular perfusion? The presence of stop flow capillaries favours development of zones of tissue hypoxia, even though total perfusion to the organ is preserved. Heterogeneity in perfusion is a crucial problem, leading to inadequate matching of flow to metabolism. In critically ill patients, the severity of microvascular alterations are independently associated with outcome [2;3;5-7]. How can these alterations be influenced? The first point is of course to minimize the factors leading or contributing to their development. Increasing flow without recruiting the microcirculation is ineffective. The role of fluids, transfusions, inotropic agents is quite variable. The use of vasodilatory agents has been proposed but there is still insufficient evidence to support their use. In particular, due to their absence of selectivity, these can also dilate already perfused vessels and lead to a steal phenomenon. Other agents such as anticoagulant agents, vitamin C and statins may improve the microcirculation.