Evaluation of gravimetric techniques to estimate the microvascular filtration coefficient

Microvascular permeability to water is characterized by the microvascular filtration coefficient (K(f)). Conventional gravimetric techniques to estimate K(f) rely on data obtained from either transient or steady-state increases in organ weight in response to increases in microvascular pressure. Both techniques result in considerably different estimates and neither account for interstitial fluid storage and lymphatic return. We therefore developed a theoretical framework to evaluate K(f) estimation techniques by 1) comparing conventional techniques to a novel technique that includes effects of interstitial fluid storage and lymphatic return, 2) evaluating the ability of conventional techniques to reproduce K(f) from simulated gravimetric data generated by a realistic interstitial fluid balance model, 3) analyzing new data collected from rat intestine, and 4) analyzing previously reported data. These approaches revealed that the steady-state gravimetric technique yields estimates that are not directly related to K(f) and are in some cases directly proportional to interstitial compliance. However, the transient gravimetric technique yields accurate estimates in some organs, because the typical experimental duration minimizes the effects of interstitial fluid storage and lymphatic return. Furthermore, our analytical framework reveals that the supposed requirement of tying off all draining lymphatic vessels for the transient technique is unnecessary. Finally, our numerical simulations indicate that our comprehensive technique accurately reproduces the value of K(f) in all organs, is not confounded by interstitial storage and lymphatic return, and provides corroboration of the estimate from the transient technique.     

Evaluation of Starling forces in the equine digit

A pump-perfused extracorporeal digital preparation was used to evaluate blood flow, arterial pressure, venous pressure, isogravimetric capillary filtration coefficient, capillary pressure, and vascular compliance in six normal horses. From these data, pre- and postcapillary resistances and pre- and postcapillary resistance ratios were determined. Vascular and tissue oncotic pressures were estimated from plasma and lymph protein concentrations, respectively. By use of the collected and calculated data, tissue pressure in the digit was calculated using the Starling equation. In the isolated equine digit, isogravimetric capillary pressure averaged 36.7 mmHg, plasma and lymph oncotic pressures averaged aged 19.12 and 6.6 mmHg, respectively, interstitial fluid pressure averaged 25.6 mmHg, and the capillary filtration coefficient averaged 0.0013 ml.min-1.mm-1.100 g-1. Our results indicate that digital capillary pressure in the laterally recumbent horse is much higher than in analogous tissues in other species such as dog and human. However, the potential edemagenic effects of this high digital capillary pressure are opposed by at least two mechanisms: 1) a high tissue pressure and 2) a low microvascular surface area for fluid exchange and/or a low microvascular permeability to filtered fluid.     

The microvascular pathophysiology of chronic venous insufficiency.

Severe chronic venous insufficiency (CVI) demonstrates as chronic, hard-to-heal wounds of the lower extremity. The wound is the result of poor skin perfusion due to a complex series of pathologic events, often initiated by a deep vein thrombosis (DVT). As years pass, the DVT causes venous valvular damage and incompetence. The calf muscle pump fails to augment venous return, and venous blood pressure is chronically elevated upon standing. Mechanisms that normally prevent the transmission of venous hypertension back upstream to the dermal microcirculation are lost. Early dermal microvascular responses include increased fluid filtration and edema. An inflammatory response induces white cell activation and adhesion. It is thought that activated white cells are trapped in dermal capillaries and increase microvascular permeability. Plasma proteins leak into the tissue space, increasing the edema. Ischemic damage to the epidermis leads to epithelial cell necrosis and ulceration. The ulcer is often slow to heal, due to inadequate perfusion and delivery of substrates required for proper wound healing. Current treatments aim to improve calf pump function, reduce edema, improve perfusion, and enhance wound healing.     

Lung edema increases transvascular filtration rate but not filtration coefficient

To determine whether the accelerated rate of lobe weight gain during severe pulmonary edema is attributed to increased permeability of the microvascular barrier or a loss of tissue forces opposing filtration, the effect of edema on capillary filtration coefficient (Kf,C), interstitial compliance (Ci), and the volume of fluid filtered after a step increase in microvascular pressure (delta Vi) were determined in eight isolated left lower lobes of dog lungs perfused at 37 degrees C with autologous blood. After attaining a base-line isogravimetric state, the capillary pressure (Pc) was increased in successive steps of 2, 5, and 10 cmH2O. This sequence of vascular pressure increases was repeated three times. Edema accumulation was expressed as weight gained as a percent of initial lobe weight (% delta Wt), and Kf,C was measured by time 0 extrapolation of the weight gain curve. An exponential rate constant for the decrease in the rate of weight gain with time (K) was calculated for each curve. Ci was then calculated by assuming that the capillary wall and interstitium constitute a resistance-capacitance network. Kf,C was not increased by edema formation in any group. Between mild (% delta Wt less than 30%) and severe edema states (% delta Wt greater than 50%) respective mean Ci increased significantly from 3.54 to 9.12 ml.cmH2O-1.100 g-1, K decreased from 0.089 to 0.036 min-1, and delta Vi increased from 1.28 to 2.4 ml.cmH2O-1.100 g-1. The delta Vi during each Pc increase was highly correlated with Kf,C and Ci when used together as independent variables (r = 0.99) but less well correlated when used separately.(ABSTRACT TRUNCATED AT 250 WORDS)     

Mechanics of interstitial-lymphatic fluid transport: theoretical foundation and experimental validation

Interstitial fluid movement is intrinsically linked to lymphatic drainage. However, their relationship is poorly understood, and associated pathologies are mostly untreatable. In this work we test the hypothesis that bulk tissue fluid movement can be evaluated in situ and described by a linear biphasic theory which integrates the regulatory function of the lymphatics with the mechanical stresses of the tissue. To accomplish this, we develop a novel experimental and theoretical model using the skin of the mouse tail. We then use the model to demonstrate how interstitial–lymphatic fluid movement depends on a balance between the elasticity, hydraulic conductivity, and lymphatic conductance as well as to demonstrate how chronic swelling (edema) alters the equipoise between tissue fluid balance parameters. Specifically, tissue fluid equilibrium is perturbed with a continuous interstitial infusion of saline into the tip of the tail. The resulting gradients in tissue stress are measured in terms of interstitial fluid pressure using a servo-null system. These measurements are then fit to the theory to provide in vivo estimates of the tissue hydraulic conductivity, elastic modulus, and overall resistance to lymphatic drainage. Additional experiments are performed on edematous tails to show that although chronic swelling causes an increase in the hydraulic conductivity, its greatly increased distensibility (due to matrix remodeling) dampens the driving forces for fluid movement and leads to fluid stagnation. This model is useful for examining potential treatments for edema and lymphatic disorders as well as substances which may alter tissue fluid balance and/or lymphatic drainage.     

Transcapillary fluid shifts in tissues of the head and neck during and after simulated microgravity.

To understand the mechanism, magnitude, and time course of facial puffiness that occurs in microgravity, seven male subjects were tilted 6 degrees head-down for 8 h, and all four Starling transcapillary pressures were directly measured before, during, and after tilt. Head-down tilt (HDT) caused facial edema and a significant elevation of microvascular pressures measured in the lower lip: capillary pressures increased from 27.7 +/- 1.5 mmHg (mean +/- SE) pre-HDT to 33.9 +/- 1.7 mmHg by the end of tilt. Subcutaneous and intramuscular interstitial fluid pressures in the neck also increased as a result of HDT, whereas interstitial fluid colloid osmotic pressures remained unchanged. Plasma colloid osmotic pressure dropped significantly by 4 h of HDT (21.5 +/- 1.5 mmHg pre-HDT to 18.2 +/- 1.9 mmHg), suggesting a transition from fluid filtration to absorption in capillary beds between the heart and feet during HDT. After 4 h of seated recovery from HDT, microvascular pressures in the lip (capillary and venule pressures) remained significantly elevated by 5-8 mmHg above baseline values. During HDT, urine output was 126.5 ml/h compared with 46.7 ml/h during the control baseline period. These results suggest that facial edema resulting from HDT is caused primarily by elevated capillary pressures and decreased plasma colloid osmotic pressures. The negativity of interstitial fluid pressures above heart level also has implications for maintenance of tissue fluid balance in upright posture.     

Relationship of fluid filtration to lung vascular pressure during edema

Effect of edema on the relationship between rate of fluid filtration and vascular pressure was studied in ventilated isolated dog lung lobes blood-perfused at constant flow. Constant rate of lobe weight gain (S), representing transvascular fluid flux, was obtained at different venous pressures (Pv) as Pv was increased stepwise from 2 to 40 and then similarly decreased from 40 to 2 Torr (n = 6). In another group (n = 6), edema was maximized by reversing the sequence of Pv change; S was obtained during similar Pv steps as Pv was decreased from 40 to 2 and then returned to 40 Torr. In both groups, delta S was disproportionately greater for delta Pv at higher Pv's, with S vs. Pv fit by an exponential curve (P less than 0.001). The exponential relationship was independent of lung hydration inasmuch as greater edema on the second limb of Pv change did not alter the curve (P greater than 0.05). At 144% weight gain, interstitial compliance was 55.5 +/- 26.8 ml.100 g-1.Torr-1 (n = 10). Interstitial pressure reportedly remains constant, i.e., fails to increase to further buffer fluid filtration, after transition of the lung interstitium from low to high compliance at approximately 40% lung weight gain. If so, then the exponential S vs. Pv relationship observed in the present study at elevated interstitial compliance does not appear related to tissue pressure-buffering effects.     

Multiscale modeling of lymphatic drainage from tissues using homogenization theory

Lymphatic capillary drainage of interstitial fluid under both steady-state and inflammatory conditions is important for tissue fluid balance, cancer metastasis, and immunity. Lymphatic drainage function is critically coupled to the fluid mechanical properties of the interstitium, yet this coupling is poorly understood. Here we sought to effectively model the lymphatic-interstitial fluid coupling and ask why the lymphatic capillary network often appears with roughly a hexagonal architecture. We use homogenization method, which allows tissue-scale lymph flow to be integrated with the microstructural details of the lymphatic capillaries, thus gaining insight into the functionality of lymphatic anatomy. We first describe flow in lymphatic capillaries using the Navier-Stokes equations and flow through the interstitium using Darcy's law. We then use multiscale homogenization to derive macroscale equations describing lymphatic drainage, with the mouse tail skin as a basis. We find that the limiting resistance for fluid drainage is that from the interstitium into the capillaries rather than within the capillaries. We also find that between hexagonal, square, and parallel tube configurations of lymphatic capillary networks, the hexagonal structure is the most efficient architecture for coupled interstitial and capillary fluid transport; that is, it clears the most interstitial fluid for a given network density and baseline interstitial fluid pressure. Thus, using homogenization theory, one can assess how vessel microstructure influences the macroscale fluid drainage by the lymphatics and demonstrate why the hexagonal network of dermal lymphatic capillaries is optimal for interstitial tissue fluid clearance.     

Protein washdown as a defense mechanism against myocardial edema

Myocardial edema occurs in many pathological conditions. We hypothesized that protein washdown at the myocardial microvascular exchange barrier would change the distribution of interstitial proteins from large to small molecules and diminish the effect of washdown on the colloid osmotic pressure (COP) of interstitial fluid and lymph. Dogs were instrumented with coronary sinus balloon-tipped catheters and myocardial lymphatic cannulas to manipulate myocardial lymph flow and to collect lymph. Myocardial venous pressure was elevated by balloon inflation to increase transmicrovascular fluid flux and myocardial lymph flow. COP of lymph was measured directly and was also calculated from protein concentration. Decreases occurred in both protein concentration and COP of lymph. The proportion of lymph protein accounted for by albumin increased significantly, whereas that accounted for by beta-lipoprotein decreased significantly. The change in the calculated plasma-to-lymph COP gradient was significantly greater than the change in the measured COP gradient. We conclude that the change in the distribution of interstitial fluid protein species decreases the effect of protein washdown on interstitial fluid COP and limits its effectiveness as a defense mechanism against myocardial edema formation.     

Physiological and pharmacological evidence for the regulation of permeability

Local intraarterial infusions of histamine-type mediators produce increases in microvascular pressure (Pmv), protein efflux, and net fluid filtration that promote edema formation. The rise in Pmv is not the primary determinant of edema formation inasmuch as mediator-stimulated edema formation develops without an increase in Pmv. The inflammatory mediators increase the hydraulic conductivity of the microvascular membrane as evidenced by a large increase in the capillary filtration coefficient (CFC) subsequent to an increase in permeability. The development of inflammatory edema is primarily attributable to the increase in protein efflux, which decreases the lymph-to-plasma total-protein ratio (L/P ratio), virtually eliminating the transmural colloid osmotic pressure gradient. Hence, fluid filtration is increased at almost any level of Pmv. Noninflammatory vasodilators and venous occlusion produce increases in Pmv and protein clearance, but fail to increase the L/P ratio. The increase in protein efflux and L/P ratio is attributable to a nonhemodynamic action of the inflammatory mediators, an increase in microvascular permeability to macromolecules. The increase in protein efflux, CFC, and net fluid filtration produced by various inflammatory mediators is largely inhibited by cooling, treatment with endothelial cell stabilizers, or perfusion with blood from hemorrhaged animals. This inhibition is independent of changes in hemodynamics and must be ascribed to a direct effect on the microvascular membrane, providing evidence for a variable macromolecular transport pathway. In contrast, increases in protein clearance produced by increasing Pmv are not inhibited by these maneuvers, which provides evidence for a static macromolecular transport pathway. These findings correlate well with those from microscopic studies supporting the concept that macromolecular permeability may be directly regulated at the level of the venular endothelial cell subsequent to the modulation of interendothelial cell junction gap size.     

Effects of vasomotion and venous pressure elevation on capillary-tissue fluid exchange across hetero-porous membrane

Flow autoregulation in the arteriolar network serves to maintain the capillary-tissue fluid balance by regulation of capillary pressure. In the present study, we have examined theoretically the effects of arteriolar vasomotion and venous pressure elevation on the capillary fluid exchange, the interstitial fluid pressure, and the interstitial osmotic pressure during capillary pressure regulation. We used Starling's hypothesis and extended it to include a consideration of a parallel hetero-porous pathway and to determine the effects of plasma protein filtration on interstitial fluid pressure and osmotic pressure. We have found that arteriolar vasomotion plays a primary role in protecting the capillary-tissue fluid balance during the elevation of capillary flow and that it is a secondary mechanism for the regulation of capillary arterial pressure.     

A model of unsteady-state transvascular fluid and protein transport in the lung

Models of steady-state fluid and solute transport in the microcirculation are used primarily to characterize filtration and permeability properties of the transport barrier. Important transient relationships, such as the rate of fluid accumulation in the tissue, cannot be predicted with steady-state models. In this paper we present three simple models of unsteady-state fluid and protein exchange between blood plasma and interstitial fluid. The first treats the interstitium as a homogeneous well-mixed compliant compartment, the second includes an interstitial gel, and the third allows for both gel and free fluid in the interstitium. Because we are primarily interested in lung transvascular exchange we used the multiple-pore model and pore sizes described by Harris and Roselli (J. Appl. Physiol.: Respirat . Environ. Exercise Physiol. 50: 1-14, 1981) to characterize the microvascular barrier. However, the unsteady-state transport theory presented here should apply to other organ systems and can be used with different conceptual models of the blood-lymph barrier. For a step increase in microvascular pressure we found good agreement between theoretical and experimental lymph flow and lymph concentrations in the sheep lung when the following parameter ranges were used: base-line interstitial volume, 150-190 ml; interstitial compliance, 7-10 ml/Torr; initial interstitial fluid pressure, -1 Torr; pressure in initial lymphatics, -5 to -6 Torr; and conductivity of the interstitium and lymphatic barrier, 4.25 X 10(-4) ml X s-1 X Torr-1. Based on these values the model predicts 50% of the total change in interstitial water volume occurs in the first 45 min after a step change in microvascular pressure.(ABSTRACT TRUNCATED AT 250 WORDS)     

Functional recovery of fluid drainage precedes lymphangiogenesis in acute murine foreleg lymphedema

Secondary lymphedema in humans is a common consequence of axillary lymph node dissection (ALND) to treat breast cancer. It is commonly hypothesized that lymphatic growth is required to increase fluid drainage and ameliorate lymphedema. Although there is a pronounced alteration in the balance of interstitial forces regulating fluid transport that sustains the chronic form of lymphedema, it is presently unknown whether changes occur to the balance of interstitial forces during acute lymphedema that may play a role in the recovery of fluid drainage. Here, we compared the relative importance of lymphangiogenesis of lymphatic vessels and interstitial flows for restoring fluid drainage and resolving acute lymphedema in the mouse foreleg after ALND. We found that removal of the axillary lymph nodes reduced lymph drainage in the foreleg at days 0 and 5 postsurgery, with fluid tracer spreading interstitially through subcutaneous tissues. Interstitial fluid drainage returned to normal by day 10, whereas functional regrowth of lymphatic vessels was first detected by indocyanine green fluorescence lymphography at day 15, demonstrating that the recovery of interstitial fluid drainage preceded the regrowth of lymphatic vessels. This was confirmed by the administration of VEGF receptor-3-neutralizing antibodies, which completely blocks lymphatic regrowth. It was found that the recovery of interstitial fluid drainage and the natural resolution of acute lymphedema produced by ALND were not hindered by VEGF receptor-3 neutralization, demonstrating that interstitial fluid drainage recovery and the resolution of acute lymphedema are lymphangiogenesis independent. The data highlight the central role of the interstitial environment in adapting to lymphatic injury to increase fluid drainage.     

Microvascular membrane permeability in high surface tension pulmonary edema

Pulmonary edema was induced in dogs by an aerosol of detergent dioctyl sodium sulfosuccinate. The permeability of the pulmonary microvascular membrane was assessed by cannulating an afferent tracheobronchial lymphatic and comparing the lymph-to-plasma total protein concentration (CL/CP) during high lymph flows induced by increasing left atrial (LA) pressure after detergent aerosol. Base-line CL/CP of 0.69 +/- 0.02 fell to 0.55 +/- 0.03 with increased LA pressure alone. CL/CP fell to 0.47 +/- 0.02 when LA pressure was increased following detergent, 0.51 +/- 0.04 following an aerosol of the vehicle in which the detergent was dissolved, and 0.73 +/- 0.10 following intravenous alloxan. In additional animals protein concentration of the airway edema fluid was compared with that of plasma. The ration of protein concentration of airway fluid to plasma was 0.63 +/- 0.08 following detergent aerosol, 0.64 +/- 0.10 following increased LA pressure, and 0.94 +/- 0.09 following administration of alloxan. These data indicate no major increase in pulmonary microvascular permeability following detergent aerosol and support the concept that pulmonary edema is the consequence of reduced interstitial perimicrovascular hydrostatic pressure caused by increased alveolar surface tension.     

Influence of interstitial fluid dynamics on growth and therapy of angiogenic tumor. Analysis by mathematical model

We have developed a spatially distributed mathematical model of angiogenic tumor growth in tissue with account of interstitial fluid dynamics and bevacizumab monotherapy. In this model the process of neovascularization is initiated by tumor cells in a state of metabolic stress, vascular endothelial growth factor (VEGF) being its main mediator. The model takes into consideration the convection flows arising in dense tissue due to active proliferation and migration of tumor cells as well as interstitial fluid inflow from blood vascular system, its outflow through lymphatic system and redistribution in the area of tumor growth. The work considers the diffusive approximation of interstitial fluid dynamics in tumor and normal tissue. Numerical study of the model showed that in absence of therapy a peritumoral edema is formed due to the increase of interstitial fluid inflow from angiogenic capillaries. In the case of rapid interstitial fluid outflow through lymphatic system and its fast transport from necrotic zone to normal tissue the regimes of full growth stop are observed in case of low-invasive tumor. Under bevacizumab monotherapy the peritumoral edema vanishes and low-invasive tumor may not only decelerate its growth, but also start shrinking for a large range of parameters.     

Isolation of pulmonary interstitial fluid in rabbits by a modified wick technique

Interstitial fluid protein concentration (C(protein)) values in perivascular and peribronchial lung tissues were never simultaneously measured in mammals; in this study, perivascular and peribronchial interstitial fluids were collected from rabbits under control conditions and rabbits with hydraulic edema or lesional edema. Postmortem dry wicks were implanted in the perivascular and peribronchial tissues; after 20 min, the wicks were withdrawn and the interstitial fluid was collected to measure C(protein) and colloid osmotic pressure. Plasma, perivascular, and peribronchial C(protein) values averaged 6.4 +/- 0.7 (SD), 3.7 +/- 0.5, and 2.4 +/- 0.7 g/dl, respectively, in control rabbits; 4.8 +/- 0.7, 2.5 +/- 0.6, and 2.4 +/- 0.4 g/dl, respectively, in rabbits with hydraulic edema; and 5.1 +/- 0.3, 4.3 +/- 0.4 and 3.3 +/- 0.6 g/dl, respectively, in rabbits with lesional edema. Contamination of plasma proteins from microvascular lesions during wick insertion was 14% of plasma C(protein). In control animals, pulmonary interstitial C(protein) was lower than previous estimates from pre- and postnodal pulmonary lymph; furthermore, although the interstitium constitutes a continuum within the lung parenchyma, regional differences in tissue content seem to exist in the rabbit lung.     

Changes in microvascular fluid filtration capacity during 120 days of 6 degrees head-down tilt.

We used venous congestion strain gauge plethysmography (VCP) to measure the changes in fluid filtration capacity (K(f)), isovolumetric venous pressure (Pv(i)), and blood flow in six volunteers before, on the 118th day (D118) of head-down tilt (HDT), and 2 days after remobilization (Post). We hypothesized that 120 days of HDT cause significant micro- and macrovascular changes. We observed a significant increase in K(f) from 3.6 +/- 0.4 x 10(-3) to 5.7 +/- 0.9 x 10(-3) ml. min(-1). 100 ml(-1). mmHg(-1) (+51.4%; P < 0.003), which returned to pretilt values (4.0 + 0.4 x 10(-3) ml. min(-1). 100 ml(-1). mmHg(-1)) after remobilization. Similarly, Pv(i) increased from 13.4 +/- 2.1 mmHg to 28.9 +/- 2.8 mmHg (+105.8%; P < 0.001) at D118 and was not significantly different at Post (12.4 +/- 2.6 mmHg). Blood flow decreased significantly from 2.3 +/- 0.3 to 1.3 +/- 0.2 ml. min(-1). 100 ml tissue(-1) at D118 and was found elevated to 3.4 +/- 0.7 ml. min(-1). 100 ml tissue(-1) at Post. We believe that the increased K(f) is caused by a higher microvascular water permeability. Because this may result in edema formation, it could contribute to the alterations in fluid homeostasis after exposure to microgravity.     

Tumor necrosis factor-alpha does not cause lung edema in rabbits.

Although tumor necrosis factor-alpha (TNF) is a key mediator in the pathophysiology of sepsis and septic shock, its role in lung microvascular injury is controversial. In isolated blood-perfused rabbit lungs, we studied the microvascular effects of human recombinant TNF by measuring the capillary filtration coefficient (Kf,c) as an index of microvascular leakiness and the arterial and venous resistances and occlusion pressures to define the microvascular pressure profile. At the end of the experiments, the lung wet-to-dry weight ratio (W/D) was determined as an index of edema. TNF increased the pulmonary venous resistance slightly but did not affect Kf,c or W/D. Furthermore, TNF at different doses failed to increase W/D less than or equal to 8 h after in vivo administration. Our data suggest that 1) the pulmonary microvascular response to TNF differs from the systemic response, which is characterized by arteriolar vasodilation, and 2) TNF is insufficient to cause lung edema, both in vivo and in vitro. Thus the development of lung microvascular injury may require the combined action of TNF and other mediators.