In healthy tissue, the rate and extent of capillary perfusion are governed by the tone of upstream arterioles and precapillary sphincters. The number of capillaries open to perfusion (and precapillary sphincter tone) is usually determined by the metabolic demands of the tissue and its need for oxygen, which is consistent with the classic concept that alterations in functional capillary density allow local modulation of O₂ exchange area and capillary-to-cell diffusion distances. This finely tuned process appears to be compromised during acute and chronic inflammatory states. An extreme form of this effect of inflammation is seen in sepsis, which is associated with a loss of functional capillary density (number of perfused capillaries), impaired regulation of oxygen delivery, and a rapid onset of tissue hypoxia despite adequate delivery of oxygen to the organ. The systemic inflammatory response that accompanies sepsis also leads to disseminated intravascular coagulation and the formation of microclots that obstruct capillaries. Inflammation-mediated events occurring in arterioles, capillaries, and venules can all result in impaired capillary perfusion (Figure 5.1) [25,31,80].

FIGURE 5.1

Roles of arterioles, capillaries, and venules in the reduced capillary perfusion that accompanies an inflammatory response. The increased arteriolar tone, leukocyte–capillary plugging, and leukocyte-mediated fluid filtration from venules all tend (more...)

Intravital microscopic observations of leukocyte movement through normal capillaries have revealed a slow passage that can result in a “pile-up” of red cells upstream in the same vessel as well as an intermittent (stop–go) flow in the vessels. This steric hindrance related malperfusion of capillaries by slow-moving leukocytes is exacerbated by inflammation. Reductions in local perfusion pressure, systemic, or local leukocyte activation, as well as the contraction of pericytes that surround capillary endothelium, could explain the impaired capillary perfusion that accompanies inflammation. While the initial trapping of leukocytes in capillaries may be mechanical (steric hindrance), the subsequent of activation of these slow-moving leukocytes by mediators released from the inflamed tissue induce the rapid expression of adhesion molecules (e.g., CD11b/CD18) on their surface and also make the cell less deformable. The rigid phenotype assumed by the activated leukocyte is a result of the polymerization of actin within the cell. The rigid leukocytes that escape the inflamed tissue also have the potential to lodge in capillaries of a downstream vascular bed, such as liver and lung. The low pressure and long narrow capillaries of the lung make this vascular bed an ideal filter for the entrapment of activated leukocytes that gain access to the systemic circulation [81–84].

Activated neutrophils have been shown to leave the inflamed gut and accumulate in liver sinusoids, causing a reduction in the number of perfused sinusoids (capillaries) and tissue hypoxia in the liver of WT mice. The leukostasis and tissue hypoxia are blunted in mice that are deficient in either ICAM-1, P-selectin, or CD11/CD18, which suggests that simple steric hindrance is less important than leukocyte–endothelial cell adhesion in mediating this response. Other studies have also implicated the intravascular Kuppfer cell (macrophages) in this response, as well as a role for T-lymphocytes. Neutrophil accumulation, loss of perfused sinusoids, or tissue hypoxia is not observed in immunodeficient SCID mice; however, the adoptive transfer of WT T-cells into the SCID restores the inflammatory responses. The T-lymphocyte derived cytokine IFN-γ has also implicated in the loss of perfused liver capillaries caused by gut-derived neutrophils [85–88].

The reduced capillary perfusion caused by obstruction from activated leukocytes and/or from aggregates of leukocytes and platelets has also been implicated in the “capillary no-reflow” phenomenon that has been described in postischemic tissues. The propensity for capillary plugging with leukocytes in ischemic tissues has been attributed to a number of events, including increased stiffness of activated leukocytes, endothelial cell swelling, leukocyte–endothelial cell adhesion, and low driving pressures for leukocyte movement along the capillaries. It is likely that the relative contribution of each of these factors to capillary no-reflow varies between organs. Thus, organs (e.g., heart) that are perfused by capillaries with small internal diameters will be more sensitive to leukocyte plugging during and following periods of hypotension. Other tissues (e.g., skin) that have large arterial–venous anastomoses, which shunt leukocytes past the capillary bed, will be less prone to leukocyte capillary plugging. The physical restriction or trapping of leukocytes within capillaries has been implicated as a major contributor to the leukocyte accumulation observed in postischemic myocardium, liver, brain, kidney, and skeletal muscle. The strong correlation between the percentage of capillaries exhibiting no-reflow and the percentage of capillaries that contain granulocytes in postischemic tissues suggests that leukocyte trapping likely accounts for capillary no-reflow. Additional supportive evidence is provided by reports that demonstrate virtual elimination of capillary no-reflow in animals that are either rendered neutropenic or receive antibodies that interfere with leukocyte–endothelial cell adhesion in postcapillary venules. Since the level of adhesion molecule expression on capillary endothelium is quite low (compared to venular endothelium), the ability of adhesion molecule-directed antibodies to attenuate capillary no-reflow has been attributed to an action on the downstream postcapillary venules. Adherent leukocytes in postcapillary venules appear to promote leukostasis in upstream capillaries by enhancing fluid and protein filtration across venular endothelium. The resulting interstitial edema raises interstitial fluid pressure to a level sufficient to occlude the capillary lumen and thereby facilitate the entrapment of leukocytes [4,28,82,85,89–91].