2: Renal Blood Flow and Glomerular Filtration

• ▸ Define renal blood flow, renal plasma flow, glomerular filtration rate, and filtration fraction, and give normal values.

• ▸ State the formula relating flow, pressure, and resistance in an organ.

• ▸ Identify the successive vessels through which blood flows after leaving the renal artery.

• ▸ State the relative resistances of the afferent arterioles and efferent arterioles.

• ▸ Describe how changes in afferent and efferent arteriolar resistances affect renal blood flow.

• ▸ Describe the 3 layers of the glomerular filtration barrier and define podocyte, foot process, and slit diaphragm.

• ▸ Describe how molecular size and electrical charge determine filterability of plasma solutes; state how protein binding of a low-molecular-weight substance influences its filterability.

• ▸ State the formula for the determinants of glomerular filtration rate, and state, in qualitative terms, why the net filtration pressure is positive.

• ▸ State the reason glomerular filtration rate is so large relative to filtration across other capillaries in the body.

• ▸ Describe how arterial pressure, afferent arteriolar resistance, and efferent arteriolar resistance influence glomerular capillary pressure.

• ▸ Describe how changes in renal plasma flow influence average glomerular capillary oncotic pressure.

• ▸ Define autoregulation of renal blood flow and glomerular filtration rate.

The amount of blood flowing through the kidneys is huge relative to their size. Renal blood flow (RBF) is about 1 L/min. This constitutes 20% of the resting cardiac output through tissue that constitutes less than 0.5% of the body mass! Considering that the volume of each kidney is less than 150 mL, this means that each kidney is perfused with over 3 times its total volume every minute. All of this blood is delivered to the cortex. About 10% of the cortical blood flow is then directed to the medulla.

Blood enters each kidney at the hilum via a renal artery. After several divisions into smaller arteries blood reaches arcuate arteries that course across the tops of the pyramids between the medulla and cortex. From these, cortical radial arteries project upward toward the kidney surface and give off a series of afferent arterioles (AAs), each of which leads to a glomerulus within Bowman's capsule (Figure 2–1). These arteries and glomeruli are found only in the cortex, never in the medulla. In most organs, capillaries recombine to form the beginnings of the venous system, but the glomerular capillaries instead recombine to form another set of arterioles, the efferent arterioles (EAs). The vast majority of the EAs soon subdivide into a second set of capillaries called peritubular capillaries. These capillaries are profusely distributed throughout the cortex intermingled with the tubular segments. The peritubular capillaries then rejoin to form the veins by which blood ultimately leaves the kidney.

EAs of glomeruli situated just above the corticomedullary border (juxtamedullary glomeruli) do not branch into peritubular capillaries the way most EAs do. Instead these arterioles descend downward into the outer medulla. Once in the medulla they divide many times to form bundles of parallel vessels called vasa recta (Latin recta for “straight” and vasa for “vessels”). These bundles of vasa recta penetrate deep into the medulla (see Figure 2–1). Vasa recta on the outside of the vascular bundles “peel off” and give rise to interbundle networks of capillaries that surround Henle's loops and the collecting ducts in the outer medulla. Only the center-most vasa recta supply capillaries in the inner medulla; thus, little blood flows into the papilla. The capillaries from the inner medulla re-form into ascending vasa recta that run in close association with the descending vasa recta within the vascular bundles. The structural and functional properties of the vasa recta are rather complex, and will be elucidated further in Chapter 6.

Blood flow through the vasa recta into the medulla is far less than cortical blood flow, perhaps 0.1 L/min. Although low relative to cortical blood flow, medullary blood flow is not low in an absolute sense and is quite comparable to blood flow in many other tissues. The significance of the differences between cortical and medullary blood flow and the vascular anatomy is the following: The high blood flow and peritubular network in the cortex maintain the interstitial environment of the cortical renal tubules very close in composition to that of blood plasma throughout the body. In contrast, the lower blood flow and grouping of vascular bundles in the medulla permit an interstitial environment that is quite different from blood plasma. As described in Chapter 6, the interstitial environment in the medulla plays a crucial role in regulating water excretion.

Blood flow in the kidneys obeys the same hemodynamic principles as found in other organs throughout the body. The basic equation for blood flow through any organ is as follows:

where Q is organ blood flow, ΔP is the mean pressure in the artery supplying the organ minus mean pressure in the vein draining that organ, and R is the total vascular resistance in that organ. The resistance of an organ is determined by the resistance of individual vessels and their series/parallel connections. The resistance of any single vessel is a function of blood viscosity, vessel length, and most of all, vessel radius. As described by Poiseiulle's law, the resistance of a cylindrical vessel varies inversely with the fourth power of vessel radius. It takes only a 19% decrease or increase in vessel radius to double or halve vessel resistance. Because the kidneys contain so many parallel pathways, that is, glomeruli and associated vessels, total renal vascular resistance is low. In turn this accounts for the high RBF.

The presence of 2 sets of arterioles (afferent and efferent) and 2 sets of capillaries (glomerular and peritubular) makes the vasculature of the cortex unusual. The resistances of the afferent and EAs are about equal in most circumstances and account for most of the total renal vascular resistance. Resistances in arteries preceding AAs (ie, cortical radial arteries) and in the capillaries play some role, but we concentrate on the arterioles because these resistances are variable and are the sites of regulation. A change in the resistance of an AA or EA has the same effect on blood flow because these vessels are in series. When the 2 resistances both change in the same direction (the most common state of affairs), their effects on RBF are additive. When they change in different directions—one resistance increasing and the other decreasing—the changes offset each other.

Hydrostatic pressures are much higher in the glomerular capillaries than in the peritubular capillaries. As blood flows through any vascular resistance the pressure progressively decreases. Pressure at the beginning of a given AA is close to mean systemic arterial pressure (∼100 mm Hg) and decreases to about 60 mm Hg at the point where it feeds a glomerulus. Because each glomerulus contains so many capillaries in parallel, pressure decreases very little during flow through those capillaries and glomerular capillary pressure stays close to 60 mm Hg. Then pressure decreases again during flow through an EA, to about 20 mm Hg at the point where it feeds peritubular capillaries (Figure 2–2). The high glomerular pressure of about 60 mm Hg is necessary to drive glomerular filtration, whereas the low peritubular capillary pressure of 20 mm Hg is equally necessary to permit the reabsorption of fluid from the renal interstitium.

The glomerular filtrate contains most inorganic ions and low-molecular-weight organic solutes in virtually the same concentrations as in the plasma. It also contains small plasma peptides and a very limited amount of albumin. Filtered fluid must pass through a 3-layered glomerular filtration barrier. The first layer, the endothelial cells of the capillaries, is perforated by many large fenestrae (“windows”), like a slice of Swiss cheese, which occupy about 10% of the endothelial surface area. They are freely permeable to everything in the blood except cells and platelets. The middle layer, the capillary basement membrane, is a gel-like acellular meshwork of glycoproteins and proteoglycans, with a structure like a kitchen sponge. The third layer consists of epithelial cells (podocytes) that surround the capillaries and rest on the capillary basement membrane. The podocytes have an unusual octopus-like structure. Small “fingers,” called pedicels (or foot processes), extend from each arm of the podocyte and are embedded in the basement membrane (see Figure 1–5D). Pedicels from a given podocyte interdigitate with the pedicels from adjacent podocytes. The pedicels are coated by a thick layer of extracellular material, which partially occludes the slits. Extremely thin processes called slit diaphragms bridge the slits between the pedicels. Slit diaphragms are widened versions of the tight junctions and adhering junctions that link all contiguous epithelial cells together and are like miniature ladders. The pedicels form the sides of the ladder, and the slit diaphragms are the rungs. Spaces between slit diaphragms constitute the path through which the filtrate, once it has passed through the endothelial cells and basement membrane, travels to enter Bowman's space.

Both the slit diaphragms and basement membrane are composed of an array of proteins, and while the basement membrane may contribute to the selectivity of the filtration barrier, integrity of the slit diaphragms is essential to prevent excessive leak of plasma protein (albumin). Some protein-wasting diseases are associated with abnormal slit diaphragm structure.

The selectivity of the filtration barrier is crucial for renal function. The barrier has to be leaky enough to permit free passage of everything that should be filtered, such as organic waste, yet restrictive to plasma proteins that should not be filtered. Selectivity of the barrier is based on both molecular size and electrical charge. Let us look first at size.

The filtration barrier of the renal corpuscle provides no hindrance to the movement of molecules with molecular weights less than 7000 Da (ie, solutes this small are all freely filtered). This includes all small ions, glucose, urea, amino acids, and many hormones. The filtration barrier almost totally excludes plasma albumin (molecular weight of approximately 66,000 Da). (For simplicity we use molecular weight as the reference for size; in reality, it is molecular radius and shape that is critical.) The hindrance to plasma albumin is not 100%, however, and the glomerular filtrate does contain extremely small quantities of albumin, on the order of 10 mg/L or less. This is only about 0.02% of the concentration of albumin in plasma and is the reason for the use of the phrase “nearly protein-free” earlier. Some small substances are partly or mostly bound to large plasma proteins and are thus not free to be filtered, even though the unbound fractions can easily move through the filtration barrier. This includes hydrophobic hormones of the steroid and thyroid categories and about 40% of the calcium in the blood.

For molecules with a molecular weight ranging from 7000 to 70,000 Da, the amount filtered becomes progressively smaller as the molecule becomes larger (Figure 2–3). Thus, many normally occurring small- and medium-sized plasma peptides and proteins are actually filtered to a significant degree. Moreover, when certain small proteins appear in the plasma because of disease (eg, hemoglobin released from damaged erythrocytes or myoglobin released from damaged muscles), considerable filtration of these may occur as well.

Electrical charge is the second variable determining filterability of macromolecules. For any given size, negatively charged macromolecules are filtered to a lesser extent, and positively charged macromolecules to a greater extent, than neutral molecules. This is because the surfaces of all the components of the filtration barrier (the cell coats of the endothelium, the basement membrane, and the cell coats of the slit diaphragms) contain fixed polyanions, which repel negatively charged macromolecules during filtration. Because almost all plasma proteins bear net negative charge, this electrical repulsion plays a very important restrictive role, enhancing that of purely size hindrance. In other words, if either albumin or the filtration barrier were not charged, even albumin would be filtered to a considerable degree (see Figure 2–3B). Certain diseases that cause glomerular capillaries to become “leaky” to protein do so because negative charges in the membranes are lost.

It must be emphasized that the negative charges in the filtration membranes act as a hindrance only to macromolecules, not to mineral anions or low-molecular-weight organic anions. Thus, chloride and bicarbonate ions, despite their negative charge, are freely filtered.

Direct Determinants of Glomerular Filtration Rate

The value of glomerular filtration rate (GFR) is a crucial determinant of renal function. It affects the excretion of waste products, and because its normal value is so large, it affects the excretion of all the substances that are handled by downstream tubular elements, particularly salt and water. Regulation of the GFR is straightforward in terms of physical principles but very complex functionally because there are many signals impinging on the controllable elements.

The rate of filtration in any capillary bed, including the glomeruli, is determined by the hydraulic permeability of the capillaries (including in this case, all elements of the filtration barrier), their surface area, and the net filtration pressure (NFP) acting across them.

Because it is difficult to estimate the surface area of a capillary bed, a parameter called the filtration coefficient (K_f) is used to denote the product of the hydraulic permeability and surface area.

The NFP is the algebraic sum of the hydrostatic pressures and the osmotic pressures resulting from protein—the oncotic, or colloid osmotic pressures—on the 2 sides of the capillary wall. There are 4 pressures to consider: 2 hydrostatic pressures and 2 oncotic pressures. These are known as Starling forces. In the glomerular capillaries

where P_GC is glomerular capillary hydrostatic pressure, π_BC is the oncotic pressure of fluid in Bowman's capsule, P_BC is the hydrostatic pressure in Bowman's capsule, and π_GC is the oncotic pressure in glomerular capillary plasma, shown schematically in Figure 2–4, along with typical average values.

Because there is normally little total protein in Bowman's capsule, π_BC may be taken as zero and not considered in our analysis. Accordingly, the overall equation for GFR becomes

Figure 2–5 shows that the hydrostatic pressure is nearly constant within the glomeruli. This is because there are so many capillaries in parallel, and collectively they provide only a small resistance to flow, but the oncotic pressure in the glomerular capillaries does change substantially along the length of the glomeruli. As water is filtered out of the vascular space it leaves most of the protein behind, thereby increasing protein concentration and, hence, the oncotic pressure of the unfiltered plasma remaining in the glomerular capillaries. Mainly because of this large increase in oncotic pressure, the NFP decreases from the beginning of the glomerular capillaries to the end (Table 2–1).

The NFP when averaged over the whole length of the glomerulus is about 16 mm Hg. This average NFP is higher than that found in most nonrenal capillary beds. Taken together with a very high value for K_f, it accounts for the enormous filtration of 180 L of fluid/day (compared with 3 L/day or so in all other capillary beds combined).

The GFR is not constant but shows fluctuations in differing physiological states and in disease. Its value must be tightly controlled. To understand control of the GFR, it is essential to see how a change in any one factor affects GFR under the assumption that all other factors are held constant.

Table 2–2 presents a summary of these factors. It provides a checklist to review when trying to understand how diseases or vasoactive chemical messengers and drugs change GFR. It should be noted that the major cause of decreased GFR in renal disease is not a change in these parameters within individual nephrons, but rather a decrease in the number of functioning nephrons. This reduces K_f.

Filtration Coefficient (K_f)

Changes in K_f are caused most often by glomerular disease, but also by normal physiological control. The details are still not completely clear, but chemical messengers released within the kidneys cause contraction of glomerular mesangial cells. Such contraction may restrict flow through some of the capillary loops, effectively reducing the area available for filtration, K_f, and, hence GFR.

Glomerular Capillary Hydrostatic Pressure (P_GC)

Hydrostatic pressure in the glomerular capillaries is influenced by many factors. We can depict the situation using the analogy of a leaking garden hose. If pressure in the pipes feeding the hose changes, the pressure in the hose and, hence, the rate of leak will be altered. Resistances in the hose also affect the leak. If we kink the hose upstream from the leak, pressure at the region of leak decreases and less water leaks out. However, if we kink the hose beyond the leak, this increases pressure at the region of leak and increases the leak rate. These same principles apply to P_GC and GFR. First, a change in renal arterial pressure causes a change in P_GC in the same direction. If resistances remain constant, P_GC rises and falls as renal artery pressure rises and falls. This is a crucial point because arterial blood pressure shows considerable variability. Second, changes in the resistance of the afferent and EAs have opposite effects on P_GC. An increase in afferent arteriolar resistance, which is upstream from the glomerulus, is like kinking the hose above the leak—it decreases P_GC. An increase in efferent arteriolar resistance is downstream from the glomerulus and is like kinking the hose beyond the leak—it increases P_GC. Of course dilation of the AA raises P_GC, and hence GFR, whereas dilation of the EA lowers P_GC and GFR. It should also be clear that when the afferent and efferent arteriolar resistances both change in the same direction (ie, they both increase or decrease), they exert opposing effects on P_GC.

What is the significance of this? It means the kidney can regulate P_GC and, hence, GFR independently of RBF. The effects of changes in arteriolar resistances are summarized in Figure 2–6.

Hydrostatic Pressure in Bowman's Capsule (P_BC)

Changes in pressure within Bowman's space are usually of very minor importance. However, obstruction anywhere along the tubule or in the external portions of the urinary system (eg, the ureter) increases the tubular pressure everywhere proximal to the occlusion, all the way back to Bowman's capsule. The result is to decrease GFR.

Oncotic Pressure in Glomerular Capillary Plasma (π_GC)

Oncotic pressure in the plasma at the very beginning of the glomerular capillaries is the oncotic pressure of systemic arterial plasma. Accordingly, a decrease in arterial plasma protein concentration, as occurs, for example, in liver disease, decreases arterial oncotic pressure and tends to increase GFR, whereas increased arterial oncotic pressure tends to reduce GFR.

However, recall that π_GC is the same as arterial oncotic pressure only at the very beginning of the glomerular capillaries; π_GC then increases somewhat along the glomerular capillaries as protein-free fluid filters out of the capillary, concentrating the protein left behind. This means that NFP and, hence, filtration progressively decrease along the capillary length. Accordingly, anything that causes a steeper increase in π_GC tends to lower average NFP and hence GFR.

Such a steep increase in oncotic pressure occurs in conditions of low RBF. When RBF is low, the filtration process removes a relatively larger fraction of the plasma, leaving a smaller volume of plasma behind in the glomeruli still containing all the plasma protein. The π_GC reaches a final value at the end of the glomerular capillaries that is higher than normal. This lowers average NFP and, hence, GFR. Conversely, a high RBF, all other factors remaining constant, causes π_GC to increase less steeply and reach a final value at the end of the capillaries that is less than normal, which increases the GFR. Table 2–1 shows typical values of the forces affecting glomerular filtration under normal conditions.

As blood is composed of cells and plasma, we can describe the flow of plasma per se, the renal plasma flow (RPF). Variations in the relative amounts of plasma that are filtered can be expressed as a filtration fraction: the ratio RPF/GFR, which is normally about 20%, that is, about 20% of the plasma entering the kidney is removed from the blood and put into the Bowman's space. The increase in π_GC along the glomerular capillaries is directly proportional to the filtration fraction (ie, if relatively more of the plasma is filtered, the increase in π_GC is greater). If the filtration fraction has changed, it is certain that there has also been a proportional change in π_GC and that this has played a role in altering GFR.

Filtered Load

A term we use in other chapters is filtered load. It is the amount of substance that is filtered per unit time. For freely filtered substances, the filtered load is the product of GFR and plasma concentration. Consider sodium. Its normal plasma concentration is 140 mEq/L, or 0.14 mEq/mL. A normal GFR in healthy young adult males is 125 mL/min, so the filtered load of sodium is 0.14 mEq/mL × 125 mL/min = 17.5 mEq/min. We can do the same calculation for any other substance, being careful in each case to be aware of the unit of measure in which concentration is expressed. The filtered load is what is presented to the rest of the nephron to handle. The filtered load varies with plasma concentration and GFR. An increase in GFR, at constant plasma concentration, increases the filtered load, as does an increase in plasma concentration at constant GFR. Variations in filtered load play a major role in the renal handling of many substances.

It is important for the kidneys to keep the GFR within a limited range. If GFR is too low, there will be insufficient excretion of waste substances, whereas if GFR is too high, this will overwhelm the capacity of the tubules to reabsorb salt and water. Most of the time this means keeping GFR relatively constant in the face of factors that change it. The most important of the factors tending to change GFR is renal artery pressure. In the healthy kidney, renal artery pressure is virtually the same as systemic arterial pressure. This pressure is by no means constant. Its potential effect on GFR is so strong that urinary excretion would tend to vary widely with ordinary daily excursions of arterial pressure. Also, vascular pressure in the thin-walled glomerular capillaries is higher than in capillaries elsewhere in the body and hypertensive damage ensues if this pressure is too high. To protect the glomerular capillaries from hypertensive damage and to preserve a healthy GFR at different arterial pressure values, changes in GFR and RBF are minimized by several mechanisms that we collectively call autoregulation.

Consider a situation in which mean arterial pressure rises 20%. Such a modest rise occurs many times throughout the day in association with changes in excitement level and activity. If all renal vascular resistances remained constant, RBF would rise 20% also (actually slightly more if pressure in the renal vein is unaffected). What would this do to GFR? It would rise much more than 20%, in fact almost 50%. This is because NFP would rise almost 50%. In effect, fractional changes in upstream pressure (in the renal artery) are magnified in terms of NFP. Why is this? In the glomerulus, capillary hydrostatic pressure averages about 60 mm Hg. The pressures opposing filtration sum to 36 mm Hg, yielding an NFP of about 24 mm Hg (Table 2–1). With an increase in arterial pressure to 120 mm Hg, glomerular capillary pressure would rise to about 71 mm Hg, but there would be no increase in the pressures that oppose filtration–plasma oncotic pressure and Bowman's capsule pressure. Therefore, NFP would rise to 71 − 36 = 35 mm Hg (an increase of almost 50%). The higher NFP would cause a parallel increase in GFR. This emphasizes why it is so important to maintain glomerular capillary pressure at the right value.

Now, what actually happens in the face of changes in mean arterial pressure? As is the case in many organs, blood flow does not change in proportion to changes in arterial pressure. The changes are blunted. A rise in driving pressure is counteracted by a rise in vascular resistance that almost offsets the rise in pressure. The word “almost” is crucial here. Higher driving pressures do indeed lead to higher flow but not proportionally. Consider Figure 2–7. Within the range of mean arterial pressures commonly found in the human body (in the region labeled “Autoregulatory range”), GFR and RBF vary only modestly when mean arterial pressure changes. This is partly a result of the myogenic response, which is the contraction or relaxation of arteriolar smooth muscle in response to changes in vascular pressures. Autoregulation is also partly the result of rather complicated intrarenal signals that affect vascular resistance and mesangial cell contraction called tubuloglomerular feedback (tubuloglomerular [TG] feedback, see Chapter 7). The myogenic response is very fast-acting (within 1 second) and protects the glomeruli from short-term fluctuations in blood pressure, while TG feedback helps maintain an appropriate filtered load of sodium and waste products.

• The kidneys have a very large blood flow relative to their mass that is regulated for functional reasons rather than metabolic demand.

• Glomerular capillary pressure is determined by the relative resistances of afferent arterioles, which precede the glomerulus, and efferent arterioles, which follow it.

• Glomerular filtration proceeds through a 3-layered barrier that restricts filtration of large macromolecules such as albumin.

• Negative surface charge on the filtration barrier restricts filtration of negatively charged solutes more than positively charged solutes.

• The GFR is determined by the permeability of the filtration barrier and NFP.

• NFP varies mainly with hydrostatic and oncotic pressures in the glomerular capillaries.

• Control of the resistances of the afferent and efferent arterioles permits independent control of glomerular filtration rate and renal blood flow.

• Autoregulation of vascular resistances keeps GFR within limits in the face of large variations in arterial pressure.