Medical Physiology, 3rd Edition

Renal Blood Flow

Renal blood flow (RBF) is ~1 L/min out of the total cardiac output of 5 L/min. Normalized for weight, this blood flow amounts to ~350 mL/min for each 100 g of tissue, which is 7-fold higher than the normalized blood flow to the brain (see p. 558). Renal plasma flow (RPF) is



Given a hematocrit (Hct) of 0.40 (see p. 102), the “normal” RPF is ~600 mL/min.

Increased glomerular plasma flow leads to an increase in GFR

At low glomerular plasma flow (Fig. 34-6A), filtration equilibrium occurs halfway down the capillary. At higher plasma flow (i.e., normal for humans), the profile of net ultrafiltration forces (PUF) along the glomerular capillary stretches out considerably to the right (see Fig. 34-6B) so that the point of equilibrium would be reached at a site actually beyond the end of the capillary. Failure to reach equilibrium (filtration disequilibrium) occurs because the increased delivery of plasma to the capillary outstrips the ability of the filtration apparatus to remove fluid and simultaneously increase capillary oncotic pressure. As a result, πGC rises more slowly along the length of the capillary.


FIGURE 34-6 Dependence of the GFR on plasma flow. imageN34-10


Dependence of Glomerular Filtration Rate and Filtration Fraction on Renal Plasma Flow

Contributed by Walter Boron

imageN34-9 described our computer model for generating plots of πGC versus distance along the glomerular capillary. The curves in Figure 34-6A are the same as in Figure 34-5C, and represent a low RPF of 70.6 mL/min. For this condition of low RPF, we computed the πGC curve using a image of 0.0015 (mm Hg)−1. (Remember from imageN34-9 that image is the embodiment of both the microscopic Kf and RPF.) As described in imageN34-9, summing the individual filtration events along the capillary yields the FF, which is 37.5% in this case. This is one of the points along the left end of the curve in Figure 34-6E. Multiplying RPF and FF yields the macroscopic GFR, which is the rather low value of 26.47 mL/min. This is one of the points along the left end of the curve in Figure 34-6D.

The curves in Figure 34-6B represent a normal RPF of 600 mL/min. For this condition of normal RPF, we computed the πGC curve using a image of 0.0001765 (mm Hg)−1. Summing the individual filtration events along the capillary yields the FF, which is 20.8% in this case. This is the identified point along the curve in Figure 34-6E. Multiplying RPF and FF yields the macroscopic GFR, which is the normal value of 125 mL/min. This is identified point along the curve in Figure 34-6D.

The curves in Figure 34-6C represent a high RPF of 1200 mL/min. For this high RPF, we computed the πGC curve using a image of 0.00008825 (mm Hg)−1, that is, half the value in the “normal” example. Summing the individual filtration events along the capillary yields the FF, which is 11.9% in this case. This is one of the points along the right end of the curve in Figure 34-6E. Multiplying RPF and FF yields the macroscopic GFR, which is the elevated value of 142.6 mL/min. This is one of the points along the right end of the curve in Figure 34-6D.

Using the approach outlined in imageN34-9 we generated πGC curves—and thus FF and GFR values—for a wide range of RPF values. The plots in Figure 34-6D and E are the results of these simulations.

As glomerular blood flow increases from low (see Fig. 34-6A) to normal (see Fig. 34-6B), the site of filtration equilibrium shifts distally (i.e., toward the efferent arteriole). This shift has two important consequences: First, as one progresses along the capillary, PUF (and hence filtration) remains greater when blood flow is higher. Second, more of the glomerular capillary is exposed to a net driving force for filtration, which increases the useful surface area for filtration. Thus, the end of the capillary that is “wasted” at low plasma flow rates really is “in reserve” to contribute when blood flow is higher.

A further increase in plasma flow stretches out the πGC profile even more, so that PUF is even higher at each point along the capillary (see Fig. 34-6C). Single-nephron GFR (SNGFR) is the sum of individual filtration events along the capillary. Thus, SNGFR is proportional to the yellow area that represents the product of PUF and effective (i.e., nonwasted) length along the capillary. Because the yellow areas progressively increase from Figure 34-6A to Figure 34-6C, SNGFR increases with glomerular plasma flow. However, this increase is not linear. Compared with the normal situation, the GFR summed for both kidneys increases only moderately with increasing RPF, but decreases greatly with decreasing RPF (see Fig. 34-6D). Indeed, clinical conditions causing an acute fall in renal perfusion result in an abrupt decline in GFR.

The relationship between GFR and RPF also defines a parameter known as the filtration fraction (FF), which is the volume of filtrate that forms from a given volume of plasma entering the glomeruli:



Because the normal GFR is ~125 mL/min and the normal RPF is ~600 mL/min, the normal FF is ~0.2. Because GFR saturates at high values of RPF, FF is greater at low plasma flows than it is at high plasma flows.

The dependence of GFR on RPF is analogous to the dependence of alveolar O2 and CO2 transport on pulmonary blood flow (see pp. 671–673).

Afferent and efferent arteriolar resistances control both glomerular plasma flow and GFR

The renal microvasculature has two unique features. First, this vascular bed has two major sites of resistance control, the afferent and the efferent arterioles. Second, it has two capillary beds in series, the glomerular and the peritubular capillaries. As a consequence of this unique architecture, significant pressure drops occur along both arterioles (Fig. 34-7), glomerular capillary pressure is relatively high throughout, and peritubular capillary pressure is relatively low. Selective constriction or relaxation of the afferent and efferent arterioles allows for highly sensitive control of the hydrostatic pressure in the intervening glomerular capillary, and thus of glomerular filtration.


FIGURE 34-7 Pressure profile along the renal vasculature.

Figure 34-8A provides an idealized example in which we reciprocally change afferent and efferent arteriolar resistance while keeping total arteriolar resistance—and thus glomerular plasma flow—constant. Compared with an initial condition in which the afferent and efferent arteriolar resistances are the same (see Fig. 34-8A, top panel), constricting the afferent arteriole while relaxing the efferent arteriole lowers PGC(see Fig. 34-8A, middle panel). Conversely, constricting the efferent arteriole while relaxing the afferent arteriole raises PGC (see Fig. 34-8A, lower panel). From these idealized PGC responses, one might predict that an increase in afferent arteriolar resistance would decrease the GFR and that an increase in efferent arteriolar resistance should have the opposite effect. However, physiological changes in the afferent and efferent arteriolar resistance usually do not keep overall arteriolar resistance constant. Thus, changes in arteriolar resistance generally lead to changes in glomerular plasma flow, which, as discussed above, can influence GFR independent of glomerular capillary pressure.


FIGURE 34-8 Role of afferent and efferent arteriolar resistance on pressure and flows. In A, the sum of afferent and efferent arteriolar resistances is always 2, whereas in B and C, the total resistance changes.

Figure 34-8B and C show somewhat more realistic effects on RPF and GFR as we change the resistance of a single arteriole. With a selective increase of afferent arteriolar resistance (see Fig. 34-8B), both capillary pressure and RPF decrease, which leads to a monotonic decline in the GFR. In contrast, a selective increase of efferent arteriolar resistance (see Fig. 34-8C) causes a steep increase in glomerular capillary pressure but a decrease in RPF. As a result, over the lower range of resistances, GFR increases with efferent resistance as an increasing PGC dominates. On the other hand, at higher resistances, GFR begins to fall as the effect of a declining RPF dominates. These opposing effects on glomerular capillary pressure and RPF account for the biphasic dependence of GFR on efferent resistance.

The examples in Figure 34-8B and C, in which only afferent or efferent resistance is increased, are still somewhat artificial. During sympathetic stimulation, or in response to ANG II, both afferent and efferent resistances increase. Thus, RPF decreases. The generally opposing effects on GFR of increasing both afferent resistance (see Fig. 34-8B) and efferent resistance (see Fig. 34-8C) explain why the combination of both keeps GFR fairly constant despite a decline in RPF.

In certain clinical situations, changes in either the afferent or the efferent arteriolar resistance dominate. A striking example is the decrease in afferent arteriolar resistance—and large increase in RPF—that occurs with the loss of renal tissue, as after a nephrectomy in a kidney donor. As a result, GFR in the remnant kidney nearly doubles (see Fig. 34-8B). Another example is a situation that occurs in patients with congestive heart failure, who thus have a tendency toward decreased renal perfusion and—for reasons not well understood—greatly increased levels of vasodilatory prostaglandins (PGE2 and PGI2). In these patients, GFR depends greatly on prostaglandin-mediated afferent arteriolar dilatation. Indeed, blocking prostaglandin synthesis with nonsteroidal anti-inflammatory drugs (NSAIDs; see p. 64) often leads to an acute fall in GFR in such patients.

An example of an efferent arteriolar effect occurs in patients with congestive heart failure associated with decreased renal perfusion. GFR in such patients depends greatly on efferent arteriolar constriction due to increased endogenous angiotensin levels. Administering angiotensin-converting enzyme inhibitors (ACE-Is) to such patients often leads to an abrupt fall in GFR. If we imagine that the peak of the GFR curve in Figure 34-8C represents the patient before treatment, then reducing the resistance of the efferent arteriole would indeed cause GFR to decrease.

Peritubular capillaries provide tubules with nutrients and retrieve reabsorbed fluid

Peritubular capillaries originate from the efferent arterioles of the superficial and juxtamedullary glomeruli (see Fig. 33-1C). The capillaries from the superficial glomeruli form a dense network in the cortex, and those from the juxtamedullary glomeruli follow the tubules down into the medulla, where the capillaries are known as the vasa recta (see pp. 722–723). The peritubular capillaries have two main functions. First, these vessels deliver oxygen and nutrients to the epithelial cells. Second, they are responsible for taking up from the interstitial space the fluid and solutes that the renal tubules reabsorb.

The Starling forces that govern filtration in other capillary beds apply here as well (Fig. 34-9A). However, in peritubular capillaries, the pattern is unique. In “standard” systemic capillaries, Starling forces favor filtration at the arteriolar end and absorption at the venular end (see pp. 471–472). Glomerular capillaries resemble the early part of these standard capillaries: the Starling forces always favor filtration (see Fig. 34-9B, yellow area). The peritubular capillaries are like the late part of standard capillaries: the Starling forces always favor absorption (see Fig. 34-9C, brown area).


FIGURE 34-9 Starling forces along the peritubular capillaries. imageN34-11


Comparison of Starling Forces along the Glomerular versus Peritubular Capillaries

Contributed by Walter Boron

Figure 34-9B shows the same pair of curves as Figure 34-6B. These represent a normal RPF of 600 mL/min. imageN34-10

What makes peritubular capillaries unique is that glomerular capillaries and the efferent arteriole precede them. Glomerular filtration concentrates the plasma proteins and thereby elevates the oncotic pressure of blood entering the peritubular capillary network (πPC) to ~35 mm Hg. In addition, the resistance of the efferent arteriole decreases the intravascular hydrostatic pressure (PPC) to ~20 mm Hg (see Fig. 34-9A). Interstitial oncotic pressure (πo) is 4 to 8 mm Hg, and interstitial hydrostatic pressure (Po) is probably 6 to 10 mm Hg. The net effect is a large net absorptive pressure at the beginning of the peritubular capillary. Along the peritubular capillary, πPC falls modestly because of the reabsorption of protein-poor fluid from the interstitium into the capillary, and hydrostatic pressure probably falls modestly as well. Even so, the Starling forces remain solidly in favor of absorption along the entire length of the peritubular capillary, falling from ~17 mm Hg at the arteriolar end to ~12 mm Hg at the venular end (see Fig. 34-9C).

The net absorptive force at the beginning of the peritubular capillaries is subject to alterations in upstream arteriolar resistances and FF. For instance, expansion of the ECF volume (Fig. 34-10) inhibits the renin-angiotensin system (see pp. 841–842), leading to a relatively larger decrease in efferent than in afferent arteriolar resistance, and therefore to a rise in PPC. The fall in total arteriolar resistance causes a rise in RPF that is larger than the increase in GFR, thereby causing a fall in FF. Thus, more fluid remains inside the glomerular capillary, and the blood entering the peritubular capillaries has an oncotic pressure that is not as high as it otherwise would be (e.g., πPC < 35 mm Hg). The fall in efferent arteriolar resistance also raises the PPC (e.g., PPC > 20 mm Hg). As a consequence of the low πPC and high PPC, the peritubular capillaries take up less interstitial fluid (see pp. 763–765). imageN34-6 The reverse sequence of events takes place during volume contraction and in chronic heart failure.


FIGURE 34-10 Effect of volume expansion on fluid uptake by the peritubular capillaries.


Effects of Hemodynamic Changes on Fluid Reabsorption by the Peritubular Capillaries

Contributed by Emile Boulpaep, Walter Boron

The paragraph in the text analyzes an example in which expansion of the ECF volume inhibits the renin-angiotensin system, producing hemodynamic changes—that is, decreases in both afferent and efferent arteriolar resistance, but more so the latter. Thus, the fractional rise in GFR is less than the fractional rise in RPF, so that the FF falls. The analysis in the text focuses on forces inside the peritubular capillaries: PPC and πPC. In addition, the hemodynamic changes in our example also lead to altered forces outside the peritubular capillaries. These changes in interstitial forces reduce fluid absorption—from the tubule lumen to the interstitium—by the proximal tubule. Thus, in this example, an increase in ECF volume leads to changes in renal hemodynamics, resulting in a parallel rise in GFR and a fall in fluid reabsorption from the proximal tubule to its peritubular capillaries. The result is an appropriate increase in fluid excretion into the urine.

We discuss the effect of hemodynamic changes on Na+ and fluid transport in more depth beginning on page 763.

Lymphatic capillaries are mainly found in the cortex, in association with blood vessels. They provide an important route for removing protein from the interstitial fluid. Proteins leak continuously from the peritubular capillaries into the interstitial fluid. Total renal lymph flow is small and amounts to <1% of RPF.

Blood flow in the renal cortex exceeds that in the renal medulla

Measurements of regional blood flow in the kidney show that ~90% of the blood leaving the glomeruli in efferent arterioles perfuses cortical tissue. The remaining 10% perfuses the renal medulla, with only 1% to 2% reaching the papilla. The relatively low blood flow through the medulla, a consequence of the high resistance of the long vasa recta, is important for minimizing washout of the hypertonic medullary interstitium and thus for producing a concentrated urine (see pp. 813–815).

The clearance of para-aminohippurate is a measure of RPF

As discussed in Chapter 33, for any solute (X) that the kidney neither metabolizes nor produces, the only route of entry to the kidney is the renal artery, and the only two routes of exit are the renal vein and the ureter (see Fig. 33-7):



The foregoing equation (a restatement of Equation 33-1) is an application of the Fick principle used for measurements of regional blood flow (see p. 423). To estimate arterial RPF (RPFa)—or, more simply, RPF—we could in principle use the clearance of any substance that the kidney measurably excretes into the urine, as long as it is practical to obtain samples of systemic arterial plasma, renal venous plasma, and urine. The problem, of course, is sampling blood from the renal vein.

However, we can avoid the need for sampling the renal vein if we choose a substance that the kidneys clear so efficiently that they leave almost none in the renal vein. Para-aminohippurate (PAH) is such a substance (see pp. 731–732). Because PAH is an organic acid that is not normally present in the human body, PAH must be administered by continuous intravenous infusion. Some PAH binds to plasma proteins, but a significant amount remains freely dissolved in the plasma and therefore filters into Bowman's space. However, the kidney filters only ~20% of the RPF (i.e., FF = ~0.2), and a major portion of the PAH remains in the plasma that flows out of the efferent arterioles. PAH diffuses out of the peritubular capillary network and reaches the basolateral surface of the proximal-tubule cells. These cells have a high capacity to secrete PAH from blood into the tubule lumen against large concentration gradients. This PAH-secretory system (see pp. 779–781) is so efficient that—as long as we do not overwhelm it by infusing too much PAH—almost no PAH (~10%) remains in the renal venous blood. We can therefore assume that all the PAH presented to the kidney appears in the urine:



In the example of Figure 34-11, the concentration of filterable PAH in the arterial blood plasma (PPAH) is 10 mg/dL or 0.1 mg/mL. If RPF is 600 mL/min, then the arterial load of PAH to the kidney is 60 mg/min. Of this amount, 12 mg/min appears in the glomerular filtrate. If the tubules secrete the remaining 48 mg/min, then the entire 60 mg/min of PAH presented to the kidney appears in the urine. That is, in the idealized example of Figure 34-11, during a single passage of blood the kidneys clear 100% of PAH presented to them. In practice, the kidneys excrete only ~90% of the arterial load of PAH, provided PPAH does not exceed 12 mg/dL. The excretion of 90% rather than 100% of the arterial load of PAH reflects the 10% of the RBF that perfuses the medulla, where the tubules do not secrete PAH.


FIGURE 34-11 Renal handling of PAH.

As long as we do not infuse too much PAH—that is, as long as virtually no PAH remains in the renal venous blood—Equation 34-7 reduces to the equation for the clearance of PAH, as introduced in Table 33-2B:



If we apply this equation to the example in Figure 34-11,



To compute RPF, we need to collect a urine sample to obtain (image) and a blood sample to obtain PPAH. However, the blood sample need not be arterial. For example, one can obtain venous blood from the arm, inasmuch as skeletal muscle extracts negligible amounts of PAH.