Glomerular morphology

The glomerular capillary wall or filtration barrier consists of three components: the capillary endothelium, basement membrane, and visceral epithelium (Fig. 2-2). The glomerulus is a unique vascular structure consisting of a capillary bed interposed between two arterioles: the afferent and efferent arterioles. The glomerular capillary divides into several branches, each of which forms a lobule of the glomerulus. The capillary endothelium of the glomerulus is fenestrated by openings 50 to 100 nm in diameter. These openings exclude cells from the ultrafiltrate, but macromolecules are not restricted based on size. The luminal surface of the endothelium is covered by negatively charged sialoglycoproteins that contribute to the charge selectivity of the filtration barrier.

Figure 2-2 Schematic representation of the glomerulus demonstrating the afferent and efferent arterioles, juxtaglomerular apparatus, and glomerular capillary loops. At the vascular pole, an afferent arteriole (AA) enters and an efferent arteriole (EA) leaves the glomerulus. At the urinary pole, the Bowman space (BS) becomes the tubular lumen of the proximal tubule (PT). The epithelial cells composing the Bowman capsule (B) enclose the Bowman space. Smooth muscle cells proper of the arterioles and all cells derived from smooth muscle are shown in black, including the granular cells (G).The afferent arteriole is innervated by sympathetic nerve terminals (N). The extraglomerular mesangial cells are located at the angle between AA and EA and continue into the mesangial cells (M) of the glomerular tuft. The glomerular capillaries are outlined by fenestrated endothelial cells (EN) and covered from the outside by the epithelial cells (EP) with foot processes (F). The glomerular basement membrane (BM) is continuous throughout the glomerulus. At the vascular pole, the thick ascending limb touches the macula densa (MD), the extraglomerular mesangium.³¹

The glomerular basement membrane is composed of the lamina rara interna on the endothelial side, the central lamina densa, and the lamina rara externa on the epithelial side. The lamina rara interna and lamina rara externa contain polar noncollagenous proteins that contribute to the negative charge of the filtration barrier. The lamina densa contains nonpolar collagenous proteins that contribute primarily to the size selectivity of the filtration barrier. The filtration barrier is permeable to molecules with effective molecular radii less than 2 nm and impermeable to those with radii greater than 4 nm.

The visceral epithelial cells or podocytes constitute the outermost portion of the filtration barrier. They cover the glomerular basement membrane and glomerular capillaries on the urinary side of the barrier with their primary and interdigitating secondary foot processes. Filtration slits, 10 to 30 nm in width, are located between the secondary foot processes. The podocytes are phagocytic and may engulf macromolecules trapped by the filtration slits. They are invested with a negatively charged sialoglycoprotein coat that contributes to the charge selectivity of the filtration barrier. It is believed that the visceral epithelial cells synthesize the glomerular basement membrane.

The mesangium is not a part of the filtration barrier but a stabilizing core of tissue, forming an anchor for the glomerulus at the vascular pole and along the axes of the capillary lobules. The mesangial cells are in contact with the basement membrane in areas where there is no capillary endothelium. The extraglomerular mesangium fills the space between the macula densa and the glomerular arterioles and constitutes part of the juxtaglomerular apparatus (JGA). The mesangial cells contain microfilaments and can contract in response to specific hormones (e.g., angiotensin II), thus altering the surface area available for filtration. They also synthesize prostaglandins that contribute to renal vasodilatation. The mesangium also contains macrophages that can clear filtration residues from the mesangial space by phagocytosis.

The glomerular capillary wall is a size-selective and a charge-selective barrier to filtration. Its size selectivity resides primarily in the lamina densa of the glomerular basement membrane. The glomerulus generally excludes molecules with radii greater than 4 nm. Inulin, with a molecular mass of 5200 daltons and radius of 1.4 nm, permeates freely, whereas serum albumin, with a molecular mass of 69,000 daltons and radius of 3.6 nm, permeates minimally.

The charge selectivity of the glomerulus resides in the negatively charged sialoglycoproteins (e.g., laminin and fibronectin) and peptidoglycans (e.g., heparan sulfate) of the capillary endothelium, lamina rara interna, lamina rara externa, and visceral epithelium. At any given effective molecular radius, negatively charged macromolecules experience greater restriction to filtration than neutral ones. Positively charged macromolecules experience less restriction to filtration than neutral ones of the same size (Fig. 2-3).

Figure 2-3 Effect of electrostatic charge on filtration of macromolecules across the glomerular capillary wall.

(Drawing by Tim Vojt.)

Determinants of glomerular filtration

The term glomerular filtration rate refers to the total filtration rate of both kidneys and represents the sum of the single-nephron glomerular filtration rates (SNGFRs) of all nephrons. The number of nephrons per kidney reflects the size of the animal. The feline kidney has approximately 200,000 nephrons, the canine kidney approximately 400,000, and the human kidney approximately 1,200,000 nephrons. SNGFR may differ among some groups of nephrons under normal conditions, and additional changes may occur in response to such factors as water deprivation, increased water intake, increased salt intake, or increased protein intake. Superficial cortical nephrons have short loops of Henle with little or no penetration into the renal medulla. These nephrons tend to excrete relatively more solute and water. Juxtamedullary nephrons have long loops of Henle that penetrate the inner medulla, and these nephrons tend to conserve solute and water. All of the nephrons in the canine and feline kidneys are thought to have long loops of Henle.

The glomerular ultrafiltrate is a protein-free ultrafiltrate of plasma containing water and all of the crystalloids of plasma in concentrations similar to those in plasma. The concentrations are not exactly the same because of the Gibbs-Donnan effect. The same Starling forces that govern the movement of fluid across other capillaries in the body determine SNGFR, but there are some important differences in the glomerulus that account for the relatively high rate of filtration:

where P_GC is the hydrostatic pressure in the glomerular capillary, which falls slightly along the length of the glomerular capillary, averaging 55 mm Hg; P_T is the hydrostatic pressure in the Bowman space, which is higher than systemic interstitial pressure, averaging 20 mm Hg; π_GC is the oncotic pressure in the glomerular capillary, which increases along the length of the capillary because of loss of protein-free ultrafiltrate into the Bowman space, averaging 20 mm Hg; and π_T is the oncotic pressure in the Bowman space and is negligible because the ultrafiltrate is nearly protein free. If π_T is neglected, the formula for SNGFR simplifies to:

These relationships are depicted in Figure 2-4, in which average pressure values are those reported for dogs³⁶ and cats.⁷ If the average pressures just described are considered alone, it can be seen that the net filtration pressure in the glomerulus is approximately 15 mm Hg, which is similar to values obtained for systemic capillaries. The fact that GFR is so much higher than the movement of fluid across systemic capillaries is explained by different values for K_f.

Figure 2-4 Graphic representation of the generation of net filtration pressure in the glomerulus as governed by Starling forces.

(Drawing by Tim Vojt.)

The ultrafiltration constant, K_f, is dependent on the surface area available for filtration and the permeability per unit area of capillary to crystalloids and water. The morphology of the glomerulus is such that the surface area available for filtration is much greater than that found in the capillary beds of skeletal muscle, and the unit permeability of the glomerular endothelium is more than 100 times that of skeletal muscle capillaries. This much higher value for K_f in glomerular capillaries than in systemic capillaries accounts for the much higher rate of filtration. The ultrafiltration coefficient, K_f, is not constant and can change as a result of disease and in response to hormones that cause mesangial cells to contract (e.g., angiotensin II).

Changes in the resistance of the afferent (preglomerular) and efferent (postglomerular) arterioles may have a marked effect on GFR. Alterations in resistance in the afferent arterioles lead to parallel changes in GFR and renal blood flow (RBF), but changes in resistance in the efferent arterioles lead to divergent changes in GFR and RBF (Fig. 2-5). The interplay of the effects of neural and hormonal factors on vascular tone in the kidneys is complex, but the main purpose of these effects is to minimize even slight changes in GFR, which could have drastic adverse effects on the volume and composition of the extracellular fluid.

Figure 2-5 Effects of alterations in afferent and efferent arteriolar tone on renal blood flow and glomerular filtration rate.

(Drawing by Tim Vojt.)

The resistance of these arterioles is regulated by the autonomic nervous system and by numerous vasoactive mediators (Table 2-1). Stimulation of the sympathetic nervous system results in release of norepinephrine from nerves terminating on the afferent and efferent arterioles. Norepinephrine can cause afferent and efferent vasoconstriction, but efferent arteriolar constriction usually predominates. As a result, RBF decreases with minimal changes in GFR (i.e., filtration fraction [FF] increases). Angiotensin II also causes efferent more than afferent vasoconstriction and has similar effects on RBF and GFR. Stimulation of dopaminergic receptors causes afferent and efferent vasodilatation and increased RBF with little change in GFR at low concentrations of dopamine. Norepinephrine, angiotensin II, and antidiuretic hormone (ADH, vasopressin) cause vasoconstriction, at the same time promoting the production of prostaglandins that cause vasodilatation. These prostaglandins (PGE₂ and PGI₂) play an important role in maintaining RBF in hypovolemic states when angiotensin II and norepinephrine concentrations are increased. The effects of these prostaglandins are limited to the kidneys because they are rapidly metabolized in the pulmonary circulation. Nonsteroidal anti-inflammatory drugs that inhibit generation of prostaglandins by the cyclooxygenase pathway may cause renal ischemia and acute renal insufficiency in hypovolemic patients.^10,12 Locally produced kinins also cause vasodilatation and favor redistribution of RBF to inner cortical nephrons. Mediators produced locally by the vascular endothelium also contribute to afferent and efferent vasoconstriction (e.g., endothelin and thromboxane) and vasodilatation (e.g., nitric oxide and prostacyclin).

Table 2-1 Effects of Selected Vasoactive Mediators on Glomerular Hemodynamics

Measurement of glomerular filtration rate

Consider a substance that is filtered by the glomeruli but neither reabsorbed nor secreted by the tubules. Under steady-state conditions, the following mass balance equation may be written:

where P_x is the plasma concentration of x (milligrams per milliliter), U_x is the urine concentration of x (milligrams per milliliter), V is the urine flow rate (milliliters per minute), and GFR is the glomerular filtration rate (milliliters per minute). Dividing both sides of the equation by P_x:

Note that this equation is the same as the formula for clearance presented before. Thus, the renal clearance of a substance that is neither reabsorbed nor secreted is equal to GFR. Inulin is a polymer of fructose with a molecular mass of 5200 da. It is not bound to plasma proteins and is freely filtered by the glomeruli. It is neither reabsorbed nor secreted by the tubules. It is not metabolized by the kidneys or any other organ. It is uncharged and not subject to the Gibbs-Donnan effect. In summary, inulin is an ideal substance for the measurement of GFR, and inulin clearance is the laboratory standard for GFR determination. Normal values for GFR as measured by inulin clearance are 3 to 5 mL/min/kg in the dog^16,21 and 2.5 to 3.5 mL/min/kg in the cat.^16,45

Inulin clearance is not used clinically because it requires intravenous infusion of inulin and an assay that is not routinely available in most clinical pathology laboratories. Creatinine is produced endogenously in the body and excreted primarily by glomerular filtration, so its clearance can be used to estimate GFR in the steady state. The only requirements for determination of endogenous creatinine clearance are an accurately timed urine sample (usually 24 hours), determination of the patient’s body weight, and measurement of serum and urine creatinine concentrations.

In the dog and cat, creatinine is filtered by the glomeruli and is neither reabsorbed nor secreted by the tubules.^18–20,22 In most clinical pathology laboratories, creatinine is measured by the alkaline picrate reaction. This reaction is not entirely specific for creatinine and measures another group of substances collectively known as noncreatinine chromogens. These substances are found in plasma, where they may constitute up to 50% of the measured creatinine at normal serum creatinine concentrations, but only small amounts appear in urine.^21,22 When the creatinine concentration is determined using the alkaline picrate reaction, the presence of noncreatinine chromogens causes endogenous creatinine clearance to underestimate GFR. This problem may be avoided by using more accurate methods (e.g., peroxidase-antiperoxidase) to measure the creatinine concentration.²² Values for endogenous creatinine clearance in the dog and cat are approximately 2 to 5 mL/min/kg.^5,17,22

To circumvent the problem of noncreatinine chromogens and to improve accuracy, some investigators have advocated determination of exogenous creatinine clearance. In this test, which is somewhat more cumbersome, creatinine is administered subcutaneously to the animal to increase the serum creatinine concentration and reduce the relative effect of the noncreatinine chromogens. For example, a normal dog may have a serum creatinine concentration of 1.0 mg/dL, of which 0.5 mg/dL represents noncreatinine chromogens. This measurement represents a 50% error. If, however, the dog’s serum creatinine concentration is increased to 10 mg/dL by subcutaneous administration of creatinine, the noncreatinine chromogens still represent only 0.5 mg/dL, and the error is reduced to 5%. Exogenous creatinine clearance exceeds endogenous creatinine clearance and more closely approximates inulin clearance in the dog.²⁰

The amount of any substance excreted by the kidneys is the algebraic sum of the amount filtered and the amount handled by the tubules:

where T_x is the amount handled by tubules (milligrams per minute).

The term T_x is a positive number if the substance experiences net secretion and a negative number if it experiences net reabsorption. Dividing both sides of the equation by P_x yields the familiar clearance formula:

Thus, the clearance of a substance experiencing net reabsorption is less than GFR (T_x is negative), and the clearance of a substance experiencing net secretion is greater than GFR (T_x is positive). The ratio of the clearance of a substance to inulin clearance gives an indication of the net handling of that substance by the kidneys. If the ratio is less than 1.0, the substance experiences net reabsorption; if it is greater than 1.0, it experiences net secretion.

Autoregulation

Autoregulation refers to the intrinsic ability of an organ to maintain blood flow at a nearly constant rate despite changes in arterial perfusion pressure. In the kidneys, between perfusion pressures of 80 and 180 mm Hg, GFR and RBF vary less than 10% (Fig. 2-7). Flow (Q) is equal to pressure (P) divided by resistance (R). As pressure increases, flow can remain constant only if resistance increases proportionately. The site of this resistance change in the kidneys is the afferent arteriole. Autoregulation is intrinsic to the kidneys and occurs in the isolated, denervated kidney and in the adrenalectomized animal. However, it is impaired by anesthesia in proportion to the depth of anesthesia. The afferent arterioles are maximally dilated at mean arterial pressures of 70 to 80 mm Hg, and at lower pressures, GFR declines linearly with RBF (i.e., autoregulation is lost). It is likely that autoregulation of RBF is a consequence of the need to regulate GFR closely and thus maintain tight control over water and salt balance.Two physiologic mechanisms contribute to autoregulation. The myogenic mechanism is based on the principle that smooth muscle tends to contract when stretched and relax when shortened. As the afferent arteriole is stretched by increased perfusion pressure, it constricts, thus limiting transmission of this increased pressure to the glomerulus and minimizing any change in glomerular capillary hydrostatic pressure and SNGFR. The myogenic mechanism represents a coarse control that operates with a delay of 1 to 2 seconds.

Figure 2-7 Autoregulation of renal blood flow and glomerular filtration rate.

(Drawing by Tim Vojt.)

Tubuloglomerular feedback represents a local intrarenal negative feedback mechanism for individual nephrons. The morphologic basis for this physiologic mechanism is the JGA. Increased sodium chloride concentration or transport in the distal tubule is sensed by the extraglomerular mesangial cells of the JGA as they monitor sodium chloride transport across the tubular cells of the macula densa. Transport of NaCl by the tubular cells of the macula densa requires functional NKCC2 (the Na⁺, K⁺, 2Cl^– cotransporter) and ROMK (a potassium channel) in the luminal membranes and functional Na⁺, K⁺-ATPase in the basolateral membranes.⁴⁶ Transcellular transport of NaCl causes generation of adenosine, which together with angiotensin II causes afferent arteriolar constriction in the parent glomerulus. The afferent arteriolar constriction causes SNGFR to decrease, thus decreasing filtration and minimizing NaCl loss in that nephron. This effect occurs locally in the region of the juxtaglomerular interstitium. Tubuloglomerular feedback represents a fine control that operates with a 10- to 12-second delay.

Measurement of renal blood flow and renal plasma flow

Consider the following mass balance equation⁵²:

Amount entering the kidneys = amount leaving the kidney

where P_AX is the renal arterial plasma concentration of x, RPF_A is the arterial renal plasma flow (RPF), P_VX is the renal venous plasma concentration of x, RPF_V is the venous RPF, U_X is the urine concentration of x, and V is the urine flow.

If we ignore the slight difference between renal arterial and venous plasma flow (with probably less than 1% error), the equation is simplified to:

If we choose a substance that is completely removed from the blood in one pass through the kidneys, P_VX is zero and RPF = U_X V/P_AX. If the substance x is not metabolized and is not excreted by any organ other than the kidneys, its concentration in any peripheral vessel equals P_AX. Thus, RPF = U_XV/P_X.

PAH is filtered by the glomeruli and secreted by the peritubular capillaries into the tubules so that approximately 90% of it is removed in one pass through the kidneys. It is not metabolized or excreted by any other organ. Thus, it approximately meets the preceding assumptions and RPF = U_PAHV/P_PAH. Now it can be seen that the clearance of PAH is an estimate of RPF. When PAH is infused during a clearance study, it is essential that P_PAH be maintained at a concentration much below the tubular transport maximum (T_max) for PAH. If not, P_VX cannot be neglected.

Some blood flows through regions of the kidneys that do not remove PAH (e.g., renal capsule, perirenal fat, and renal pelvis), and as a result, P_VX is not really zero. Thus, the term effective RPF is more appropriately used when speaking of PAH clearance. Furthermore, only 90% of PAH is removed from the blood during a single pass through the kidneys. This also contributes to the fact that P_VX for PAH is not really zero. A closer approximation of RPF can be determined by sampling renal arterial and venous blood and measuring their respective PAH concentrations. The extraction ratio for PAH is then determined:

A more accurate calculation of RPF is then:

The extraction ratio for PAH is 0.9 because approximately 90% of it is removed from the blood in a single pass through the kidneys. Notice that if we substitute the equation for E_X into the preceding equation, we get RPF = U_XV/(P_AX − P_VX), which is the same equation as derived before for RPF.

Another way to determine RPF is by use of the Fick principle, which states that the amount of a substance (V) removed by an organ is equal to the blood flow to the organ (Q) times the arteriovenous concentration difference of the substance in question (C_A − C_V):

Using the kidneys as an example and equating the amount of the substance removed to the amount excreted (U_XV):

Note that this equation is identical to that derived before using the mass balance principle.

If the hematocrit is known, RBF can be calculated from the RPF by using the following equation:

In the dog and cat, normal values for RPF are 7 to 20 mL/min/kg and 8 to 22 mL/min/kg, respectively.^38,39,45

If all of the plasma were filtered in one pass of blood through the glomeruli, an immovable mass of red blood cells would be all that remained behind at the efferent arteriole of the glomerular capillary. This does not occur because π_GC increases along the length of the capillary and, in conjunction with P_T, effectively opposes further filtration. The filtration fraction is the fraction of plasma flowing through the kidneys that is filtered into the Bowman space. It is determined by the following equation:

In the dog and cat, values for FF are 0.32 to 0.36 and 0.33 to 0.41, respectively. These values are higher than those observed in humans in whom FF is approximately 0.20.