Pressure differentials govern fluid movement across physiologic semi-permeable membranes, and two of these forces are hydrostatic/hydraulic pressure and osmotic pressure. The third factor is the permeability of the capillary membranes. There will be an escape of water and solute into the interstitial space resulting in interstitial edema whenever the hydrostatic pressure is much higher than the osmotic pressure inside the intravascular space. Edema also occurs when there is capillary leakage due to impaired membrane integrity such as in burns or anaphylaxis.
Hydrostatic pressure stems from the action of gravity of a column of fluid while hydraulic pressure refers to the action delivered by a pump. Together, these two forces contribute to blood pressure and fluid movement into and out of the vascular space. Regulation becomes particularly important at the level of the capillary, the point in the circulatory system where permeability exists to both solute and water.
Osmotic pressure relies on selective permeability in membranes. Take two of the major ions of the extracellular fluid: Na+ and Cl-, which can move rapidly between plasma and interstitial fluid spaces, thereby making them ineffective osmotic agents. Proteins, by contrast, are mostly restricted to the plasma compartment, making them effective osmotic agents in the ability to draw water from the interstitial space (where protein concentration is low) to the plasma compartment (where protein concentration is high). The effective osmotic pressure in this example exerted by the plasma proteins on the fluid movement between the two compartments represents colloid osmotic pressure or the plasma oncotic pressure.
While the filtration-reabsorption balance model is the classically taught version, recent studies have shown that it has limitations in accurately depicting microcirculation in most tissues. Michael and Phillips challenged the traditional model when they used capillaries of frog mesentery to demonstrate that fluid absorption occurred transiently when hydrostatic pressure in the capillary (Pc) fell below plasma oncotic pressure (πc). In the steady-state, though, the fluid dynamics changed. When hydrostatic pressure in the capillary (Pc) was lower than plasma oncotic pressure (πc) in this setting, no absorption occurred. This evidence that capillaries in low-pressure organ systems can absorb fluid only transiently and not continuously made it apparent that an additional factor that influences dynamics, namely the interplay between oncotic pressure of the interstitium and capillary filtration rate. Exceptions exist, in which net absorption does take place in the steady-state, as has been shown in lymph nodes, peritubular capillaries of the cortex, and the ascending vasa recta of the medulla.
A discussion of fluid movement across membranes would be incomplete without a discussion of Starling forces. Blood pressure within a capillary (approximately 36 mmHg), referred to as the capillary hydrostatic pressure(P), constitutes an outward filtration force from the plasma space to the interstitium. The opposing force, meaning the hydrostatic pressure exerted by the interstitium (P) towards the capillary is normally close to zero, making it non-contributory to net fluid movement across capillary membranes. The major reabsorptive force in this system comes from the colloid osmotic pressure within the capillary (π), normally around 24 mmHg, whereas the colloid osmotic pressure of the interstitium (πi) drawing fluid out of the vasculature is normally close to zero. A balance normally exists between the blood pressure in the capillary and the plasma colloid oncotic pressure, resulting in a constant vascular volume within the system over time. In reality, filtration exceeds reabsorption by roughly 10%, with the excess non-reabsorbed filtrate being returned to the vascular system via lymphatics. The last contributors to this system are coefficients for a filtration (K), which coverts the hydraulic pressure differentials to flow, and a reflection coefficient (σ) that relates to the membrane's impermeability. The Starling equation can then be written as below:
Net flow of fluid across a capillary wall = (K) * [filtration forces - reabsorptive forces]
Net flow = (K) * [(P + π) - (P + π)]
P = blood pressure, π = colloid osmotic pressure; and the subscripts: c = capillary, i = interstitial fluid
These forces change along the length of the capillary, with the greatest changes occurring with blood pressure. At the arteriolar end of the capillary, the blood pressure is roughly 36 mmHg and falls to about 15 mm Hg at the venous end of the capillary. The colloid osmotic pressure at the arteriolar end remains relatively constant at about 25 mmHg
Importantly, the Starling forces only describe the movement of water across membranes in the vascular system and the mechanism behind constancy in vascular volume.
Colloid osmotic pressure can be calculated using the Van’t Hoff factor equation. The usefulness of this calculation, though, becomes complicated in abnormal physiologic conditions due to several factors including the lack of proportional changes in protein and salts, heterogeneity in the proteins involved, and interaction between the protein. This difficulty warrants measuring the colloid osmotic pressure directly in certain situations.
One method of direct measurement of interstitial colloid osmotic pressure is the wick method, which involves the sampling of interstitial COP with multifilamentous nylon wicks, which are first washed and soaked in priming solution before being sewn into the subcutaneous tissue of an animal being studies. After a certain period, the wicks are pulled out, and the wick fluid isolated by centrifugation.
Another method of measurement called the crossover method involves priming the wicks in several different solutions of various concentrations. COP in the fluid within the wick increases during implantation only in wicks primed with fluid with lower protein concentration than the ISF. By plotting the COP of the priming fluid against the COP of wick fluid after implantation, a linear plot can be constructed with the crossing point of the two representing the true COP of the interstitium.
Normal variation in colloid osmotic pressure has been a topic of research. For example, mean colloid osmotic pressure is 21.1 mmHg in those younger than 50 years old, and significantly higher 19.7 mm Hg in those between ages 70 and 89. Males also had significantly higher COP than females across age groups.
Critical Care and Congestive Heart Failure
Critical care is a setting in which the clinical manifestations of abnormal fluid balance are seen and have a crucial influence on patient outcomes. Pulmonary edema, for example, can result when the gradient between COP and pulmonary artery wedge pressure (PAWP) is reduced – PAWP in this example representing the outward hydrostatic pressure in the pulmonary vascular space. Rackow showed that the greater the decrease in COP-PAWP gradient, the greater the increase in the severity of pulmonary edema. They extrapolated from this that COP-PAWP gradient predicted mortality in shock patients but did not influence outcomes in patients with pulmonary edema without shock.
In left ventricular failure, due to the significantly elevated left ventricular end-diastolic volume and pressure, the PAWP is proportionately increased resulting in the reduced COP-PAWP gradient. Fluid enters the pulmonary interstitial space, i.e., pulmonary edema. During such circumstances, the edema fluid will be more in the dependent areas because the patient experiences increased shortness of breath when lying down (orthopnea). Clinically it will be different from other edema states secondary to reduced plasma protein concentration which results in edema in all interstitial spaces and, therefore, generalized clinical edema (anasarca).
We can measure colloid osmotic pressure to better understand the mechanism of pulmonary edema in left ventricular failure. The primary insult, an increase in left ventricular filling pressure, causes a sequence of counterreactions aimed at restoring fluid balance. A filtrate depleted of protein passes by ultrafiltration through the lung capillaries, thereby creating a higher COP in plasma that may partly counterbalance the elevated hydrostatic pressure accumulated. The lymphatic system of the lungs provides a safety mechanism to remove fluid from the air spaces until this mechanism is saturated.
Investigators have tried to manipulate the COP-PAWP gradient by increasing the plasma COP via albumin infusion as a way of restoring intravascular blood volume and reversing the fluid loss to the interstitium. After all, albumin accounts for roughly 80% of the total oncotic pressure exerted by blood plasma on interstitial fluid. Infusion of albumin alone may produce improvement in 40% critically ill patients, according to one study, while adding a potent diuretic like ethacrynic acid improved that percentage to 70%. From this, it is worth noting that albumin infusion alone as a means of correcting fluid balance is an oversimplification of the backbone physiological concepts outlined by Starling. The quality of membranes involved, transcapillary escape of albumin after infusion, changes in plasma volume, and other factors come into play.
Hypoalbuminemia may occur clinically as a result of impaired albumin absorption (Kwashiorkor) or albumin loss from the gut (protein-losing enteropathy), impaired protein synthesis by the liver (chronic liver disease), or protein loss through the kidneys(nephrotic syndrome). Under such circumstances, the colloid osmotic pressure will be significantly reduced resulting in water and solutes escaping into the interstitial space from the capillaries. These are all causes of generalized anasarca resulting from reduced colloid osmotic pressure.
Pregnancy is another physiologic circumstance in which fluid shifts take place between intravascular and interstitial spaces, with COP playing a role. An increase in plasma volume takes place in normal pregnancy, which accounts for a fall in COP assuming there is no corresponding increase in colloids. Red cell volume increases during pregnancy as well, although less than plasma volume, which causes a decrease in hematocrit during the first and second trimester of normal pregnancy. Wu and colleagues directly measured serum total solids (STS) as a marker of COP, since the principal component of STS is albumin, the main colloid determinant of COP. They found that STS (and therefore COP) fell gradually during pregnancy to a low point between 30 to 34 weeks gestation and proceeded to rise toward term, following a quadratic equation parabolic trend. Correlating this with mean blood pressure indicates the direction of fluid shifts throughout pregnancy.
Type 1 Diabetes
Damage to the microvasculature is one of many physiologic changes that occur in long-standing diabetes. The disturbance in capillary permeability to proteins, in particular, leads to changes in the transcapillary colloid osmotic gradient. Patients with long-standing Type 1 diabetes without nephropathy had reduced interstitial colloid osmotic pressure with increased transcapillary osmotic gradient compared to normal subjects.  Increased microvascular permeability to proteins should in itself increase the amount of protein in the interstitium, thereby increased interstitial oncotic pressure. The reason for the opposite finding in this study was thought to be due to increased net capillary filtration, either because of increased capillary filtration coefficient or increased hydrostatic capillary pressure, and the resultant lymphatic wash-out of proteins from the interstitium. An increased colloid osmotic gradient between vascular space and interstitium facilitates the preservation of plasma volume and limits the development of edema.