The Microcirculation as a Therapeutic Target in the Treatment of Sepsis and Shock

Vanina S. Kanoore Edul, M.D.; Arnaldo Dubin, M.D., Ph.D.; Can Ince, Ph.D.


Semin Respir Crit Care Med. 2011;32(5):558-568. 

In This Article

Oxygen Delivery in the Microcirculation

The vasoactive control of microcirculatory perfusion ensures regulation of the distribution of blood flow to meet the oxygen demand of the respiring parenchymal cells. The mechanisms by which this blood flow is regulated include myogenic factors (i.e., strain and stress forces), metabolic factors such as pH, pCO2 and potassium, O2, and nitric oxide (NO) and neurohormonal control.[13,14] In addition, there is the control induced by vasoactive compounds in patients.[15]

A critical point is that O2 transport is achieved by two main mechanisms: convection and diffusion.[16] Convection is defined as the number of cells per unit time in a given capillary and measures O2 delivery to the microcirculation by flowing oxygen-carrying red blood cells (RBCs). In addition to the number of hemoglobin (Hb)-carrying RBCs, the affinity of hemoglobin for O2, pH, pCO2, and temperature also plays an important role in controlling O2 delivery to tissues. Diffusion, the second important factor that determines oxygen transport to tissue, is determined by the difference between the capillary and mitochondrial pO2, a factor that is directly proportional to the physical distance between capillaries filled by oxygen-carrying RBCs and the cells to the mitochondria (i.e., the functional capillary density).[17] The means by which the microcirculation can be recruited in this way is to fill previously empty capillaries. We recently showed that reduction of diffusion distances by filling up previous empty capillaries is the way blood transfusions are effective in promoting microcirculatory oxygen transport.[18]

In 1919, the physiologist August Krogh developed a mathematical model of oxygen transport based on the observation that the number of capillaries perfused in a tissue during high metabolic activity was higher than during resting conditions. The model assumed that each capillary supplies a unique tissue volume[18] and that O2 exchange takes place only in the capillary bed. An increase in tissue O2 demand would cause the capillary bed to respond by increasing the number of perfused vessels and the homogeneity of blood flow. This model was rapidly accepted because it provided an accurate description of the physics behind O2 delivery. However, it did not explain the mechanisms by which metabolism-dependent capillary flow changes occur. Moreover, currently we know that O2 diffusion takes place in both arterioles and venules that can participate in convective O2 transport and can even serve as a source of O2 for erythrocyte reoxygenation.[19] O2 diffusion also occurs from capillaries to neighboring capillaries[20] and directly from arterioles to venules. Conventionally, it has been thought that there is a steep oxygen gradient from the microcirculation (typical values of ~30 mm Hg) to the tissue mitochondria, where oxygen pressures were assumed to be below 5 mm Hg. The oxygen conformance theory introduced by Schumacker and colleagues, however, suggested that mitochondrial pO2 levels were much closer to the values of microcirculatory pO2 and adapt their consumption to those levels.[21] Our recent identification of the oxygen-dependent optical properties of endogenous mitochondrial protoporphyrin (Pp)[22] IX and the development of delayed fluorescence quenching of Pp IX as a means of quantitatively measuring mitochondrial pO2 in vivo, confirmed the predictions of the oxygen conformance theory and demonstrated that mitochondrial pO2 in vivo is in fact much higher than previously thought (~20 mm Hg).[23]

Normal blood flow in the microcirculation has been described as both spatially and temporally heterogeneous. During resting conditions, approximately 30% of the available capillaries are perfused. Under stress, however, there is a recruitment of capillaries along with increases in blood flow that finally allow the homogeneous oxygenation of the microcirculation.[24]

Since the description of the Krogh model, extensive research conducted in this area has tried to determine what mechanisms are responsible for the adequacy of tissue oxygenation. Basically, the main question is, in response to what substance does a vessel change its flow or diameter? That is, what is the "O2 sensor"? Hypoxic tissues may release metabolites that further affect arteriolar tone; however, it is difficult to understand how this tissue metabolite can affect upstream flow. Animal studies using intravital microscopy showed that the distribution of blood flow within the microcirculation is determined by the behavior of the parent vessel from which a terminal arteriole originates [i.e., terminal arteriole-fed vessels (TAFs)].[14] Each TAF controls the distribution of RBCs among the respective arterioles, suggesting that these vessels are a key point of regulation because they determine whether or not the capillaries are perfused. Segal proposed that metabolic feedback to the microcirculation is a control system with two main components.[14] The precapillary sphincter regulates the functional surface area for exchange and the distance for oxygen diffusion, which allows for an increase in O2 extraction when metabolic activity increases and is reflected in turn by a fall in venous pO2. During this period, there is little change in regional blood flow.[25] The second component involves the proximal arterioles that control local blood flow. As venous pO2 decreases, blood flow to the tissue increases by vasodilation of the feeding vessel; this response is known as ascending vasodilation.

The endothelium and RBCs play a major role in the vascular tone regulation and have been implicated in upstream vasodilation.[26] The endothelial cells transmit upstream and downstream information by cell-to-cell signaling by sensing flow, hormones, and other substances.[27] This mechanism is accomplished by electrophysiological control by coupled endothelial cells via gap junctions (connexin 40), which have been shown to uncouple during sepsis.[28] The endothelium also controls coagulation and immunological responses that are key functions of the microcirculation; thus it has also been proposed as a potential oxygen sensor.[29]

More recently, there has been an increasing interest in the active role of RBCs in the regulation of local blood flow and O2 distribution. In hypoxic conditions, RBCs release adenosine triphosphate (ATP) and nitric oxide (NO) to cause vasodilation.[30] If these cells were the O2 sensor, then they may elicit immediate changes in local vascular tone according to tissue needs. Aside from releasing ATP, RBCs are also capable of transporting NO produced in the lungs in the form of S-nitrosothiol (SNO) carried by hemoglobin. SNO is a potent vasodilator and is released when hemoglobin saturation falls.[31] The reaction of nitrite with deoxyhemoglobin results in the production of nitric oxide and is thought to be an additional important mechanism by which RBCs affect vascular tone.[32] This finding emphasizes that RBC function is much more complex than as a simple O2 carrier and might be a potential target to monitor and regulate O2 delivery.

Contrary to the Krogh model in which capillaries were the primary site of oxygen diffusion,[18] today we know that most of the oxygen is "lost'' from the precapillary network.[33] There is a longitudinal gradient of O2 saturation in precapillary microvessels, in which approximately two thirds of the total oxygen extraction occurs, at least in resting skeletal muscle. The current view shows extensive diffusive interactions among microvessels. O2 can diffuse from arterioles into adjacent capillaries and venules according to local oxygen gradients.[20] Because venules lie near and parallel to arterioles, a countercurrent mechanism causes a gradient for solutes in venous blood to diffuse to nearby arterioles and O2 to diffuse from arterioles to venules. In addition, O2 leaving the arterioles can in this way also reoxygenate RBCs. In addition, microvascular hematocrit is very heterogeneous according to the Fahraeus-Lindqvist effect.[34] Based on the heterogeneity of microvascular hematocrit, the precapillary drop in O2 saturation, and the diffusional exchange of O2 among microvessels, Ellis et al proposed that blood flow itself is not a good indicator of adequate oxygen delivery to tissues.[5]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: