Background: Our previous study in a swine polytrauma model suggested that equieffective systemic pressor doses of arginine vasopressin (AVP) versus phenylephrine (PE) have differential effects on the systemic and cerebral microcirculation. The purpose of this study was to directly observe the effects of AVP versus PE on inflammatory changes evoked by tumor necrosis factor alpha (TNF) in the skeletal muscle microcirculation.
Methods: Seventy-five male rats (180-250 g) were anesthetized with isoforane, intubated and mechanically ventilated with 100% oxygen. The cremaster muscle microcirculation was prepared for intravital video microscopy while being suffused with a heated hetastarch-electrolyte solution. Fluorescein isothiocyanate-labeled albumin (100 mg/kg) was administered intravenously (i.v.) before one of five protocols. In series 1 (n = 20), either AVP (0.2 U/mL) or its vehicle was added to the suffusate for 10 minutes, washed out for 30 minutes, then TNF was suffused (5 ng/mL) for 30 minutes. In series 2 (n = 16), the protocol was similar, except AVP (0.2 U/mL) or an equieffective dose of PE (0.04 mg/mL) was administered i.v. (4.5 mL/h) for 15 minutes before, during, and 45 minutes after TNF suffusion. In series 3 (n = 12), the protocol was similar to series 2, except venous hemorrhage preceded i.v. AVP or PE. In series 4 (n = 15), the protocol was similar to series 3, except an AVP antagonist (vaprisol, 1 mg/kg i.v.) or its vehicle was administered after hemorrhage. In the control series (n = 13), inflammation was evaluated either with a different suffusate (lactated Ringers instead of hetastarch solution), different antigen (histamine instead of TNF), or hemorrhage with no antigen.
Results: In series 1, the TNF-evoked increase in leukocyte infiltration (i.e., rolling), leukocyte activation (i.e., sticking), and macromolecular permeability (i.e., albumin extravasation) were attenuated with topical AVP versus vehicle (both p < 0.05), with no effect on venular blood flow (which determines sheer stress). In series 2, the TNF-evoked increase in infiltration, activation, and permeability were all attenuated, and arteriolar blood flow (which determines perfused capillary surface area and hydrostatic pressure) was reduced with i.v. AVP versus i.v. PE (all p < 0.05). In series 3, after hemorrhage to mean arterial pressure <50 mm Hg for 30 minutes, the TNF-evoked increase in infiltration and activation was attenuated, and arteriolar and venular blood flow were both reduced with i.v. AVP versus PE (all p < 0.05). In series 4, after hemorrhage, the TNF-evoked increase in leukocyte activation was potentiated with the vaprisol versus vehicle (p < 0.05) with no effect on arteriolar or venular blood flow. In series 5 (controls), suffusion with lactated Ringers' versus hetastarch solution more than doubled the TNF-evoked increase in activation (p < 0.05).
Conclusion: (1) AVP can attenuate TNF-evoked leukocyte infiltration, activation or permeability changes in the skeletal muscle microcirculation. (2) The mechanism is probably receptor mediated and does not entirely depend on sheer stress in venules or Starling forces in capillaries. (3) The magnitude of this anti-inflammatory effect is influenced by several conditions, including volume status, the colloid or crystalloid suffusion fluid, and is possibly specific to the antigenic stimulus (TNF vs. histamine).