Pathogenesis.

The critical events in both ischemic and nephrotoxic AKI are believed to be (1) tubular injury and (2) persistent and severe disturbances in blood flow⁶⁰ (Fig. 20-23).

• Tubule cell injury: Tubular epithelial cells are particularly sensitive to ischemia and are also vulnerable to toxins. Several factors predispose the tubules to toxic injury, including a vast charged surface for tubular reabsorption, active transport systems for ions and organic acids, a high metabolic rate and oxygen consumption requirement so as to perform these transport and reabsorption functions, and the capability for effective concentration.

FIGURE 20-23 Postulated sequence in acute kidney injury. GFR, glomerular filtration rate; NO, nitric oxide; PGI₂, prostaglandin I2 (prostacyclin).

(Modified from Brady HR et al.: Acute renal failure. In Brenner BM [ed]: Brenner and Rector’s The Kidney, 5th ed, Vol II. Philadelphia, WB Saunders, 1996, p 1210).

Ischemia causes numerous structural and functional alterations in epithelial cells, as discussed in Chapter 1. The structural changes include those of reversible injury (such as cellular swelling, loss of brush border and polarity, blebbing, and cell detachment) and those associated with lethal injury (necrosis and apoptosis). Biochemically there is depletion of ATP; accumulation of intracellular calcium; activation of proteases (e.g., calpain), which cause cytoskeletal disruption; activation of phospholipases, which damage membranes; generation of reactive oxygen species; and activation of caspases, which induce apoptotic cell death. One early reversible result of ischemia is loss of cell polarity due to redistribution of membrane proteins (e.g., the enzyme Na⁺, K⁺-ATPase) from the basolateral to the luminal surface of the tubular cells, resulting in abnormal ion transport across the cells, and increased sodium delivery to distal tubules. The latter incites vasoconstriction via tubuloglomerular feedback, which will be discussed below. In addition, ischemic tubular cells express cytokines (such as monocyte chemoattractant protein 1) and adhesion molecules (such as intercellular adhesion molecule 1), thus recruiting leukocytes that appear to participate in the subsequent injury. In time, injured cells detach from the basement membranes and cause luminal obstruction, increased intratubular pressure, and decreased GFR. In addition, fluid from the damaged tubules leaks into the interstitium, resulting in interstitial edema, increased interstitial pressure, and further damage to the tubule. All these effects, as shown in Figure 20-23, contribute to the decreased GFR.

The patchiness of tubular necrosis and maintenance of the integrity of the basement membrane along many segments allow ready repair of the necrotic foci and recovery of function if the precipitating cause is removed. This repair is dependent on the capacity of reversibly injured epithelial cells to proliferate and differentiate. Re-epithelialization is mediated by a variety of growth factors and cytokines produced locally by the tubular cells themselves (autocrine stimulation) or by inflammatory cells in the vicinity of necrotic foci (paracrine stimulation).⁶¹ Of these, epidermal growth factor, TGF-α, insulin-like growth factor type 1, and hepatocyte growth factor have been shown to be particularly important in renal tubular repair. Growth factors, indeed, are being explored as possible therapeutic agents to enhance re-epithelialization in AKI, although clinical trials to date have been disappointing.⁶¹

Morphology. Ischemic AKI is characterized by focal tubular epithelial necrosis at multiple points along the nephron, with large skip areas in between, often accompanied by rupture of basement membranes (tubulorrhexis) and occlusion of tubular lumens by casts⁶² (Figs. 20-24 and 20-25). The straight portion of the proximal tubule and the ascending thick limb in the renal medulla are especially vulnerable, but focal lesions may also occur in the distal tubule, often in conjunction with casts. Paradoxically, the clinical syndrome of AKI is often associated with lesser degrees of tubular injury. This includes attenuation or loss of proximal tubule brush borders, simplification of cell structure, cell swelling and vacuolization, and sloughing of non-necrotic tubular cells into the tubular lumina (see Fig. 20-25). The severity of the morphologic findings often does not correlate well with the severity of the clinical manifestations.

FIGURE 20-24 Patterns of tubular damage in ischemic and toxic acute kidney injury. In the ischemic type, tubular necrosis is patchy, relatively short lengths of tubules are affected, and straight segments of proximal tubules (PST) and ascending limbs of Henle’s loop (HL) are most vulnerable. In toxic acute kidney injury, extensive necrosis is present along the proximal convoluted tubule segments (PCT) with many toxins (e.g., mercury), but necrosis of the distal tubule, particularly ascending HL, also occurs. In both types, lumens of the distal convoluted tubules (DCT) and collecting ducts (CD) contain casts.

FIGURE 20-25 Acute kidney injury. Some of the tubular epithelial cells in the tubules are necrotic, and many have become detached (from their basement membranes) and been sloughed into the tubular lumens, whereas others are swollen, vacuolated, and regenerating.

(Courtesy of Dr. Agnes Fogo, Vanderbilt University, Nashville, TN.)

Eosinophilic hyaline casts, as well as pigmented granular casts, are common, particularly in distal tubules and collecting ducts. These casts consist principally of Tamm-Horsfall protein (a urinary glycoprotein normally secreted by the cells of ascending thick limb and distal tubules) in conjunction with other plasma proteins. Other findings in ischemic AKI are interstitial edema and accumulations of leukocytes within dilated vasa recta. There is also evidence of epithelial regeneration: flattened epithelial cells with hyperchromatic nuclei and mitotic figures are often present. In the course of time this regeneration repopulates the tubules so that, no residual evidence of damage is seen.

Toxic AKI is manifested by acute tubular injury, most obvious in the proximal convoluted tubules. On histologic examination the tubular necrosis may be entirely nonspecific, but it is somewhat distinctive in poisoning with certain agents. With mercuric chloride, for example, severely injured cells may contain large acidophilic inclusions. Later, these cells become totally necrotic, are desquamated into the lumen, and may undergo calcification. Carbon tetrachloride poisoning, in contrast, is characterized by the accumulation of neutral lipids in injured cells; again, such fatty change is followed by necrosis. Ethylene glycol produces marked ballooning and hydropic or vacuolar degeneration of proximal convoluted tubules. Calcium oxalate crystals are often found in the tubular lumens in such poisoning.

Clinical Course.

The clinical course of AKI is highly variable, but the classic case may be divided into initiation, maintenance, and recovery stages. The initiation phase, lasting for about 36 hours, is dominated by the inciting medical, surgical, or obstetric event in the ischemic form of AKI. The only indication of renal involvement is a slight decline in urine output with a rise in BUN. At this point, oliguria could be explained on the basis of a transient decrease in blood flow and declining GFR.

The maintenance phase is characterized by sustained decreases in urine output to between 40 and 400 mL/day (oliguria), salt and water overload, rising BUN concentrations, hyperkalemia, metabolic acidosis, and other manifestations of uremia. With appropriate attention to the balance of water and blood electrolytes, including dialysis, the patient can be supported through this oliguric crisis.

The recovery phase is ushered in by a steady increase in urine volume that may reach up to 3 L/day. The tubules are still damaged, so large amounts of water, sodium, and potassium are lost in the flood of urine. Hypokalemia, rather than hyperkalemia, becomes a clinical problem. There is a peculiar increased vulnerability to infection at this stage. Eventually, renal tubular function is restored and concentrating abilityimproves. At the same time, BUN and creatinine levels begin to return to normal. Subtle tubular functional impairment may persist for months, but most patients who reach this phase eventually recover completely.

The prognosis of AKI depends on the clinical setting. Recovery is expected with nephrotoxic AKI when the toxin has not caused serious damage to other organs, such as the liver or heart. With current supportive care, 95% of those who do not succumb to the precipitating cause recover. Conversely, in shock related to sepsis, extensive burns, or other causes of multi-organ failure, the mortality rate can rise to more than 50%.

Up to 50% of patients with AKI do not have oliguria and instead often have increased urine volumes. This so-called nonoliguric AKI occurs particularly often with nephrotoxins, and it generally tends to follow a more benign clinical course.

Etiology and Pathogenesis.

The dominant etiologic agents, accounting for more than 85% of cases of urinary tract infection, are the gram-negative bacilli that are normal inhabitants of the intestinal tract.⁶³ By far the most common is Escherichia coli, followed by Proteus, Klebsiella, and Enterobacter. Streptococcus faecalis, also of enteric origin, staphylococci, and virtually every other bacterial and fungal agent can also cause lower urinary tract and renal infection. In immunocompromised persons, particularly those with transplanted organs, viruses such as Polyomavirus, cytomegalovirus, and adenovirus can also be a cause of renal infection.

In most patients with urinary tract infection, the infecting organisms are derived from the patient’s own fecal flora. This is thus a form of endogenous infection. There are two routes by which bacteria can reach the kidneys: (1) through the bloodstream (hematogenous infection) and (2) from the lower urinary tract (ascending infection) (Fig. 20-26). The hematogenous route is the less common of the two and results from seeding of the kidneys by bacteria from distant foci in the course of septicemia or infective endocarditis. Hematogenous infection is more likely to occur in the presence of ureteral obstruction, in debilitated patients, in patients receiving immunosuppressive therapy, and with nonenteric organisms, such as staphylococci and certain fungi and viruses.

FIGURE 20-26 Schematic representation of pathways of renal infection. Hematogenous infection results from bacteremic spread. More common is ascending infection, which results from a combination of urinary bladder infection, vesicoureteral reflux, and intrarenal reflux.

Ascending infection is the most common cause of clinical pyelonephritis. Normal human bladder and bladder urine are sterile; therefore, a number of steps must occur for renal infection to occur:

FIGURE 20-27 Vesicoureteral reflux demonstrated by a voiding cystourethrogram. Dye injected into the bladder refluxes into both dilated ureters, filling the pelvis and calyces.

In the absence of vesicoureteral reflux, infection usually remains localized in the bladder. Thus, the majority of individuals with repeated or persistent bacterial colonization of the urinary tract suffer from cystitis and urethritis (lower urinary tract infection) rather than pyelonephritis.

Clinical Features.

Acute pyelonephritis is often associated with predisposing conditions, some of which were mentioned before. These include the following:

When acute pyelonephritis is clinically apparent, the onset is usually sudden, with pain at the costovertebral angle and systemic evidence of infection, such as fever and malaise. There are usually indications of bladder and urethral irritation, such as dysuria, frequency, and urgency. The urine contains many leukocytes (pyuria) derived from the inflammatory infiltrate, but pyuria does not differentiate upper from lower urinary tract infection. The finding of leukocyte casts, typically rich in neutrophils (pus casts), indicates renal involvement, because casts are formed only in tubules. The diagnosis of infection is established by quantitative urine culture.

Uncomplicated acute pyelonephritis usually follows a benign course, and the symptoms disappear within a few days after the institution of appropriate antibiotic therapy. Bacteria, however, may persist in the urine, or there may be recurrence of infection with new serologic types of E. coli or other organisms. Such bacteriuria then either disappears or may persist, sometimes for years. In the presence of unrelieved urinary obstruction, diabetes mellitus, or immunodeficiency, acute pyelonephritis may be more serious, leading to repeated septicemic episodes. The superimposition of papillary necrosis may lead to acute renal failure.

An emerging viral pathogen causing pyelonephritis in kidney allografts is polyomavirus. Latent infection with polyomavirus is widespread in the general population, but immunosuppression of the allograft recipient can lead to reactivation of latent infection and the development of a nephropathy resulting in allograft failure in as many as 1% to 5% of kidney transplant recipients.⁶⁷ This form of pyelonephritis, now referred to as polyomavirus nephropathy, is characterized by viral infection of tubular epithelial cell nuclei, leading to nuclear enlargement and intranuclear inclusions visible by light microscopy (viral cytopathic effect). The inclusions are composed of virions arrayed in distinctive crystalline-like lattices when visualized by electron microscopy (Fig. 20-31). An interstitial inflammatory response is invariably present. Treatment consists of a reduction in immunosuppression.

FIGURE 20-31 Polyomavirus nephropathy. A, The kidney shows enlarged tubular epithelial cells with nuclear inclusions (arrows) and interstitial inflammation (arrowheads). B, Intranuclear viral inclusions visualized by electron microscopy.

(Courtesy of Dr. Jean Olson, Department of Pathology, University of California San Francisco, San Francisco, CA.)

Reflux Nephropathy.

This is by far the more common form of chronic pyelonephritic scarring. Renal involvement in reflux nephropathy occurs early in childhood as a result of superimposition of a urinary infection on congenital vesicoureteral reflux and intrarenal reflux. Reflux may be unilateral or bilateral; thus, the resultant renal damage may cause scarring and atrophy of one kidney or involve both, leading to chronic renal insufficiency. Vesicoureteral reflux occasionally causes renal damage in the absence of infection (sterile reflux), but only when obstruction is severe.

Chronic Obstructive Pyelonephritis.

We have seen that obstruction predisposes the kidney to infection. Recurrent infections superimposed on diffuse or localized obstructive lesions lead to recurrent bouts of renal inflammation and scarring, resulting in a picture of chronic pyelonephritis. In this condition, the effects of obstruction contribute to the parenchymal atrophy; indeed, it is sometimes difficult to differentiate the effects of bacterial infection from those of obstruction alone. The disease can be bilateral, as with posterior urethral valves, resulting in renal insufficiency unless the anomaly is corrected, or unilateral, such as occurs with calculi and unilateral obstructive anomalies of the ureter.

Morphology. The characteristic changes of chronic pyelonephritis are seen on gross examination (Figs. 20-32 and 20-33). The kidneys usually are irregularly scarred; if bilateral, the involvement is asymmetric. This contrasts with chronic glomerulonephritis, in which both kidneys are diffusely and symmetrically scarred. The hallmarks of chronic pyelonephritis are coarse, discrete, corticomedullary scars overlying dilated, blunted, or deformed calyces, and flattening of the papillae (see Fig. 20-33). The scars can vary from one to several in number and may affect one or both kidneys. Most are in the upper and lower poles, consistent with the frequency of reflux in these sites.

FIGURE 20-33 A, Chronic pyelonephritis. The surface (left) is irregularly scarred. The cut section (right) reveals characteristic dilation and blunting of calyces. The ureter is dilated and thickened, a finding that is consistent with chronic vesicoureteral reflux. B, Low-power view showing a corticomedullary renal scar with an underlying dilated deformed calyx. Note the thyroidization of tubules in the cortex.

The microscopic changes involve predominantly tubules and interstitium. The tubules show atrophy in some areas and hypertrophy or dilation in others. Dilated tubules with flattened epithelium may be filled with colloid casts (thyroidization). There are varying degrees of chronic interstitial inflammation and fibrosis in the cortex and medulla. In the presence of active infection there may be neutrophils in the interstitium and pus casts in the tubules. Arcuate and interlobular vessels demonstrate obliterative intimal sclerosis in the scarred areas; and in the presence of hypertension, hyaline arteriosclerosis is seen in the entire kidney. There is often fibrosis around the calyceal epithelium as well as a marked chronic inflammatory infiltrate. Glomeruli may appear normal except for periglomerular fibrosis, or exhibit a variety of changes, including ischemic fibrous obliteration and secondary changes related to hypertension. Individuals with chronic pyelonephritis and reflux nephropathy who develop proteinuria in advanced stages show secondary focal segmental glomerulosclerosis, as described later.

Xanthogranulomatous pyelonephritis is an unusual and relatively rare form of chronic pyelonephritis characterized by accumulation of foamy macrophages intermingled with plasma cells, lymphocytes, polymorphonuclear leukocytes, and occasional giant cells. Often associated with Proteus infections and obstruction, the lesions sometimes produce large, yellowish orange nodules that may be grossly confused with renal cell carcinoma.

Clinical Features.

Chronic obstructive pyelonephritis may be insidious in onset or present with clinical manifestations of acute recurrent pyelonephritis, such as back pain, fever, frequent pyuria, and bacteriuria. Chronic pyelonephritis associated with reflux may have a silent onset. These patients come to medical attention relatively late in the course of their disease because of the gradual onset of renal insufficiency and hypertension or because of the discovery of pyuria or bacteriuria on routine examination. Reflux nephropathy is often discovered when hypertension in children is investigated. Loss of tubular function—in particular of concentrating ability—gives rise to polyuria and nocturia. Radiographic studies show asymmetrically contracted kidneys with characteristic coarse scars and blunting and deformity of the calyceal system. Significant bacteriuria may be present, but it is often absent in the late stages.

Although proteinuria is usually mild, some individuals with pyelonephritic scars develop secondary focal segmental glomerulosclerosis with significant proteinuria, even in the nephrotic range, usually several years after the scarring has occurred and often in the absence of continued infection or persistent vesicoureteral reflux. The appearance of proteinuria and focal segmental glomerulosclerosis is a poor prognostic sign, and patients with these findings may proceed to chronic or end-stage renal failure. The glomerulosclerosis, as we have discussed, may be attributable to the adaptive glomerular alterations secondary to loss of renal mass caused by pyelonephritic scarring (renal ablation nephropathy).

Pathogenesis.

Many features of the disease suggest an immune mechanism. The immune response is idiosyncratic and not dose-related. Clinical evidence of hypersensitivity includes the latent period, the eosinophilia and rash, the fact that the onset of nephropathy is not dose-related, and the recurrence of hypersensitivity after re-exposure to the same or a cross-reactive drug. In some patients, serum IgE levels are increased, and IgE-containing plasma cells and basophils are present in the lesions, suggesting that the late-phase reaction of an IgE-mediated (type I) hypersensitivity may be involved in the pathogenesis (Chapter 6). In other cases, mononuclear or granulomatous infiltrate, together with positive results of skin tests to drug haptens, suggest a T cell–mediated delayedhypersensitivity reaction (type IV).

The most likely sequence of events is that the drugs act as haptens, which covalently bind to some cytoplasmic or extracellular component of tubular cells and become immunogenic. The resultant injury is then due to IgE and/or cell-mediated immune reactions to tubular cells or their basement membranes.

Morphology. On histologic examination the abnormalities are in the interstitium, which shows variable but frequently pronounced edema and infiltration by mononuclear cells, principally lymphocytes and macrophages. Eosinophils and neutrophils may be present (Fig. 20-34), often in clusters and large numbers, and plasma cells and basophils are sometimes found in small numbers. With some drugs (e.g., methicillin, thiazides), interstitial non-necrotizing granulomas containing giant cells may be seen. “Tubulitis,” the infiltration of tubules by lymphocytes, is common. Variable degrees of tubular necrosis and regeneration are present. The glomeruli are normal except in some cases caused by NSAIDs, when minimal-change disease and the nephrotic syndrome develop concurrently (see the discussion of NSAIDs later in the chapter).

FIGURE 20-34 Drug-induced interstitial nephritis, with prominent eosinophilic and mononuclear cell infiltrate.

(Courtesy of Dr. H. Rennke, Brigham and Women’s Hospital, Boston, MA.)

Clinical Features.

It is important to recognize drug-induced renal failure because withdrawal of the offending drug is followed by recovery, although it may take several months, and irreversible damage occurs occasionally in older subjects. It is also important to remember that while drugs are the leading identifiable cause of acute interstitial nephritis, in many affected patients (approximately 30% to 40%) an offending drug or mechanism cannot be identified.

Pathogenesis.

Papillary necrosis is readily induced experimentally by a mixture of aspirin and phenacetin, usually combined with water depletion. It is now clear that in the sequence of events leading to renal damage, papillary necrosis occurs first, and cortical tubulointerstitial nephritis follows as a consequence of impeded urine outflow. The phenacetin metabolite acetaminophen, which can deplete cells of glutathione, then injures these cells by subsequent generation of oxidative metabolites. Aspirin induces its potentiating effect by inhibiting the vasodilatory effects of prostaglandins, predisposing the papillae to ischemia. Thus, the papillary damage may be due to a combination of direct toxic effects of phenacetin metabolites and ischemic injury to both tubular cells and vessels.

Morphology. In gross appearance the kidneys are either normal or slightly reduced in size, and the cortex shows depressed areas representing cortical atrophy overlying necrotic papillae. The papillae show various stages of necrosis, calcification, fragmentation, and sloughing. This gross appearance contrasts with the papillary necrosis seen in diabetic patients, in which all papillae are at the same stage of injury. On microscopic examination the papillary changes may take one of several forms. In early cases there is patchy necrosis, but in the advanced form the entire papilla is necrotic, often remaining in place as a structureless mass containing “ghosts” of tubules and foci of dystrophic calcification (Fig. 20-35). Segments of entire portions of the papilla may then be sloughed and excreted in the urine.

FIGURE 20-35 Analgesic nephropathy. A, The brownish necrotic papilla, transformed to a necrotic, structureless mass, fills the pelvis. B, Microscopic view. Note the fibrosis in the medulla.

(Courtesy of Dr. F.J. Gloor, Institut für Pathologie, Kantonsspital, St. Gallen, Switzerland.)

The cortical changes consist of loss and atrophy of tubules and interstitial fibrosis and inflammation. These changes are mainly due to obstructive atrophy caused by the tubular damage in the papillae. The cortical columns of Bertin are characteristically spared from this atrophy.

Clinical Features.

Analgesic nephropathy is more common in women than in men and is particularly prevalent in individuals with recurrent headaches and muscle pain, in psychoneurotic patients, and in factory workers. Early renal findings include inability to concentrate the urine (hyposthenuria), as would be expected for papillary lesions. Acquired distal renal tubular acidosis contributes to the development of renal stones. Headache, anemia, gastrointestinal symptoms, and hypertension are common accompaniments of analgesic nephropathy. Urinary tract infection complicates about 50% of cases. On occasion, entire tips of necrotic papillae are excreted, and these may cause gross hematuria or renal colic due to obstruction of the ureter by necrotic fragments. Magnetic resonance and computed tomographic imaging are helpful in detecting papillary necrosis and calcifications. Progressive impairment of renal function may lead to chronic renal failure, but with drug withdrawal, renal function may either stabilize or actually improve.

Unfortunately, a small percentage of patients with analgesic nephropathy develop transitional papillary carcinoma of the renal pelvis. Whether the carcinogenic effect is due to a metabolite of phenacetin or to some other component of the analgesic compounds is unsettled.

Papillary necrosis is not specific for analgesic nephropathy. It is also seen in diabetes mellitus, as was mentioned earlier, as well as in urinary tract obstruction, sickle cell disease or trait (described later), and focally in renal tuberculosis. Table 20-9 lists certain features of papillary necrosis in these conditions.

TABLE 20-9 Causes of Papillary Necrosis

Clinical Features.

Clinically, the renal manifestations are of several types. In the most common form, chronic renal failure develops insidiously and usually progresses slowly during a period of several months to years. Another form occurs suddenly and is manifested by acute renal failure with oliguria. Precipitating factors in these patients include dehydration, hypercalcemia, acute infection, and treatment with nephrotoxic antibiotics. Bence Jones proteinuria occurs in 70% of individuals with multiple myeloma; the presence of significant non–light-chain proteinuria (e.g., albuminuria) suggests AL amyloidosis or light-chain deposition disease.

Pathogenesis.

Two processes participate in the arterial lesions:

Morphology. The kidneys are either normal or moderately reduced in size, with average weights between 110 and 130 gm. The cortical surfaces have a fine, even granularity that resembles grain leather (Fig. 20-38). The loss of mass is due mainly to cortical scarring and shrinking.

FIGURE 20-38 Close-up of the gross appearance of the cortical surface in benign nephrosclerosis illustrating the fine, leathery granularity of the surface.

On histologic examination there is narrowing of the lumens of arterioles and small arteries, caused by thickening and hyalinization of the walls (hyaline arteriolosclerosis) (Fig. 20-39). Corresponding to the fine surface granulations are microscopic subcapsular scars with sclerotic glomeruli and tubular dropout, alternating with better preserved parenchyma. In addition, the interlobular and arcuate arteries show a characteristic lesion that consists of medial hypertrophy, reduplication of the elastic lamina, and increased myofibroblastic tissue in the intima, which combine to narrow the lumen. This change, called fibroelastic hyperplasia, often accompanies hyaline arteriolosclerosis and increases in severity with age and in the presence of hypertension.

FIGURE 20-39 Hyaline arteriolosclerosis. High-power view of two arterioles with hyaline deposition, marked thickening of the walls, and a narrowed lumen.

(Courtesy of Dr. M.A. Venkatachalam, Department of Pathology, University of Texas Health Sciences Center, San Antonio, TX.)

Consequent to the vascular narrowing, there is patchy ischemic atrophy, which consists of (1) foci of tubular atrophy and interstitial fibrosis and (2) a variety of glomerular alterations. The latter include collapse of the GBM, deposition of collagen within the Bowman space, periglomerular fibrosis, and total sclerosis of glomeruli. When the ischemic changes are pronounced and affect large areas of parenchyma, they can produce regional scars and histologic alterations that may resemble those seen in renal ablation injury, mentioned earlier.

Clinical Features.

It is unusual for uncomplicated benign nephrosclerosis to cause renal insufficiency or uremia. There are usually moderate reductions in renal blood flow, but the GFR is normal or only slightly reduced. On occasion, there is mild proteinuria. However, three groups of hypertensive patients with benign nephrosclerosis are at increased risk of developing renal failure: people of African descent, people with more severe blood pressure elevations, and persons with a second underlying disease, especially diabetes. In these groups renal insufficiency may supervene after prolonged benign hypertension, but more rapid renal failure results from the development of the malignant or accelerated phase of hypertension, discussed next.

Pathogenesis.

The basis for this turn for the worse in hypertensive subjects is unclear, but the following sequence of events is suggested. The initial insult seems to be some form of vascular damage to the kidneys. This might result from long-standing benign hypertension, with eventual injury to the arteriolar walls, or the initiating injury may spring de novo from arteritis, a coagulopathy, or some injury causing acute exacerbation of the hypertension. In any case, the result is increased permeability of the small vessels to fibrinogen and other plasma proteins, endothelial injury, focal death of cells of the vascular wall, and platelet deposition. This leads to the appearance of fibrinoid necrosis of arterioles and small arteries, swelling of the vascular intima, and intravascular thrombosis. Mitogenic factors from platelets (e.g., PDGF), plasma, and other cells cause hyperplasia of intimal smooth muscle of vessels, resulting in the hyperplastic arteriolosclerosis that is typical of malignant hypertension and further narrowing of the lumens. The kidneys become markedly ischemic. With severe involvement of the renal afferent arterioles, the renin-angiotensin system receives a powerful stimulus; indeed, patients with malignant hypertension have markedly elevated levels of plasma renin. This sets up a self-perpetuating cycle in which angiotensin II causes intrarenal vasoconstriction, and the attendant renal ischemia perpetuates renin secretion. Other vasoconstrictors (e.g., endothelin) and loss of vasodilators (nitric oxide) may also contribute to vasoconstriction. Aldosterone levels are also elevated, and salt retention undoubtedly contributes to the elevation of blood pressure. The consequences of the markedly elevated blood pressure on the blood vessels throughout the body are known as malignant arteriosclerosis, and the renal disorder is malignant nephrosclerosis.

Morphology. On gross inspection the kidney size depends on the duration and severity of the hypertensive disease. Small, pinpoint petechial hemorrhages may appear on the cortical surface from rupture of arterioles or glomerular capillaries, giving the kidney a peculiar “flea-bitten” appearance.

Two histologic alterations characterize blood vessels in malignant hypertension (Fig. 20-40):

• Fibrinoid necrosis of arterioles. This appears as an eosinophilic granular change in the blood vessel wall, which stains positively for fibrin by histochemical or immunofluorescence techniques. This change represents an acute event; it may be accompanied by limited inflammatory infiltrate within the wall, but prominent inflammation is not seen. Sometimes the glomeruli become necrotic and infiltrated with neutrophils, and the glomerular capillaries may thrombose.

• In the interlobular arteries and arterioles, there is intimal thickening caused by a proliferation of elongated, concentrically arranged smooth muscle cells, together with fine concentric layering of collagen and accumulation of pale-staining material that probably represents accumulations of proteoglycans and plasma proteins. This alteration has been referred to as onion-skinning because of its concentric appearance. The lesion, also called hyperplastic arteriolitis, correlates well with renal failure in malignant hypertension. There may be superimposed intraluminal thrombosis. The arteriolar and arterial lesions result in considerable narrowing of all vascular lumens, ischemic atrophy and, at times, infarction distal to the abnormal vessels.

FIGURE 20-40 Malignant hypertension. A, Fibrinoid necrosis of afferent arteriole (PAS stain). B, Hyperplastic arteriolitis (onion-skin lesion).

(Courtesy of Dr. H. Rennke, Brigham and Women’s Hospital, Boston, MA.)

Clinical Features.

The full-blown syndrome of malignant hypertension is characterized by systolic pressures greater than 200 mm Hg and diastolic pressures greater than 120 mm Hg, papilledema, retinal hemorrhages, encephalopathy, cardiovascular abnormalities, and renal failure. Most often, the early symptoms are related to increased intracranial pressure and include headaches, nausea, vomiting, and visual impairments, particularly scotomas or spots before the eyes. “Hypertensive crises” are sometimes encountered, characterized by episodes of loss of consciousness or even convulsions. At the onset of rapidly mounting blood pressure, there is marked proteinuria and microscopic or sometimes macroscopic hematuria but no significant alteration in renal function. Soon, however, renal failure makes its appearance. The syndrome is a true medical emergency requiring the institution of aggressive and prompt antihypertensive therapy to prevent the development of irreversible renal lesions. Before the introduction of current antihypertensive drugs, malignant hypertension was associated with a 50% mortality rate within 3 months of onset, progressing to 90% within a year. At present, however, about 75% of patients survive 5 years, and 50% survive with restoration of pre-crisis renal function.

Pathogenesis.

The classic experiments of Goldblatt and colleagues⁷⁸ showed that constriction of one renal artery in dogs results in hypertension and that the magnitude of the effect is roughly proportional to the amount of constriction. Later experiments in rats confirmed these results, and in time it was shown that the hypertensive effect, at least initially, is due to stimulation of renin secretion by cells of the juxtaglomerular apparatus and the subsequent production of the vasoconstrictor angiotensin II. A large proportion of individuals with renovascular hypertension have elevated plasma or renal vein renin levels, and almost all show a reduction of blood pressure when given drugs that block the activity of angiotensin II. Furthermore, unilateral renal renin hypersecretion can be normalized after renal revascularization, usually resulting in a decrease in blood pressure. Other factors, however, may contribute to the maintenance of renovascular hypertension after the renin-angiotensin system has initiated it, including sodium retention and possibly endothelin and loss of nitric oxide.

Morphology. The most common cause of renal artery stenosis (70% of cases) is occlusion by an atheromatous plaque at the origin of the renal artery. This lesion occurs more frequently in men, and the incidence increases with advancing age and diabetes mellitus. The plaque is usually concentrically placed, and superimposed thrombosis often occurs.

The second type of lesion leading to stenosis is so-called fibromuscular dysplasia of the renal artery. This is a heterogeneous group of lesions characterized by fibrous or fibromuscular thickening and may involve the intima, the media, or the adventitia of the artery. These lesions are thus subclassified into intimal, medial, and adventitial hyperplasia, the medial type being by far the most common (Fig. 20-41). The stenoses, as a whole, are more common in women and tend to occur in younger age groups (i.e., in the third and fourth decades). The lesions may consist of a single well-defined constriction or a series of narrowings, usually in the middle or distal portion of the renal artery. They may also involve the segmental branches and may be bilateral.

FIGURE 20-41 Fibromuscular dysplasia of the renal artery, medial type (elastic tissue stain). The media shows marked fibrous thickening, and the lumen is stenotic.

(Courtesy of Dr. Seymour Rosen, Beth Israel Hospital, Boston, MA.)

The ischemic kidney is usually reduced in size and shows signs of diffuse ischemic atrophy, with crowded glomeruli, atrophic tubules, interstitial fibrosis, and focal inflammatory infiltrates. The arterioles in the ischemic kidney are usually protected from the effects of high pressure, thus showing only mild arteriolosclerosis. In contrast, the contralateral nonischemic kidney may show more severe arteriolosclerosis, depending on the severity of the hypertension.

Clinical Course.

Few distinctive features suggest the presence of renal artery stenosis, and in general, these patients resemble those with essential hypertension. On occasion, a bruit can be heard on auscultation of the affected kidneys. Elevated plasma or renal vein renin, response to angiotensin-converting enzyme inhibitor, renal scans, and intravenous pyelography may aid with diagnosis, but arteriography is required to localize the stenotic lesion. The cure rate after surgery is 70% to 80% in well-selected cases.

Pathogenesis.

Within the thrombotic microangiopathies, two pathogenetic triggers dominate: (1) endothelial injury, and (2) platelet activation and aggregation. As will be discussed, endothelial injury appears to be the primary cause of HUS, whereas platelet activation may be the inciting event in TTP.

Typical (epidemic, classic, diarrhea-positive) Hemolytic-Uremic Syndrome.

This is the best-characterized form of HUS. Most cases occur following intestinal infection with strains of E. coli (the most common being O157:H7) that produce Shiga-like toxins,⁸³ so-called because they resemble those made by Shigella dysenteriae (Chapter 17). Epidemics have been traced to various sources, most commonly the ingestion of contaminated ground meat (as in hamburgers), but also drinking water, raw milk, and person-to-person transmission. However, most cases of typical HUS caused by E. coli are sporadic. Less commonly, infections by other agents, including Shigella dysenteriae, can give rise to a similar clinical picture.

Typical HUS can occur in adults, particularly the elderly, but it affects children preferentially, in whom it is one of the main causes of acute renal failure. Following a prodrome of influenza-like or diarrheal symptoms, there is a sudden onset of bleeding manifestations (especially hematemesis and melena), severe oliguria, and hematuria, associated with microangiopathic hemolytic anemia, thrombocytopenia, and (in some patients) prominent neurologic changes. Hypertension is present in about half the patients.

Shiga-like toxin injures endothelial cells, inducing increased expression of leukocyte adhesion molecules; increased endothelin and decreased nitric oxide production; and in the presence of cytokines such as TNF, endothelial apoptosis. These alterations lead to platelet activation and induce vasoconstriction, resulting in the characteristic microangiopathy. There is also some evidence that Shiga-like toxins bind and directly activate platelets.

In typical HUS, if the renal failure is managed properly with dialysis, most patients recover normal renal function in a matter of weeks. However, due to underlying renal damage the long-term (15 to 25 year) outlook is more guarded. In one study, only 10 of 25 patients with prior epidemic HUS had normal renal function, and 7 had chronic kidney disease.

Atypical (non-epidemic, diarrhea-negative) Hemolytic-Uremic Syndrome.

Atypical HUS occurs mainly in adults in a number of different settings. More than half of those affected have an inherited deficiency of complement-regulatory proteins, most commonly factor H, which normally breaks down the alternative pathway C3 convertase and protects cells from damage by uncontrolled complement activation (Chapter 2).⁸² A small number of patients have mutations in two other proteins that regulate complement, complement factor I and CD46 (membrane cofactor protein). Patients with genetic mutations in complement-regulatory proteins may develop HUS at any age. Roughly half of affected individuals have a course marked by multiple relapses and progression to end-stage renal disease. As the deficiencies in complement-regulatory factors are life-long, it is a mystery why the onset of HUS is delayed; additional unknown co-factors that trigger the development of HUS are suspected.

A variety of miscellaneous conditions or exposures are occasionally associated with atypical forms of HUS. These include:

Patients with atypical HUS do not fare as well as those with typical HUS, in large part because the underlying conditions may be chronic and difficult to treat.⁸⁰ As in typical HUS, some patients have neurologic symptoms; the disease in these patients can be distinguished from TTP by the presence of normal ADAMTS13 levels in the plasma (see below).

Thrombotic Thrombocytopenic Purpura.

TTP is classically manifested by the pentad of fever, neurologic symptoms, microangiopathic hemolytic anemia, thrombocytopenia, and renal failure.⁸⁰ As discussed above, it is usually caused by antibodies (either autoimmune or drug-induced) or genetic defects that lead to functional deficits in ADAMTS13.⁸² The most common cause of deficient ADAMTS13 activity is inhibitory autoantibodies, and the majority of those with such antibodies are women. Regardless of cause, most patients present as adults at ages younger than 40.

In TTP, central nervous system involvement is the dominant feature, whereas renal involvement occurs in only about 50% of patients. The clinical findings are dictated by the distribution of the microthrombi, which are found in arterioles throughout the body. Untreated, the disease was once highly fatal, but in those with autoantibodies exchange transfusions and immunosuppressive therapy have reduced the mortality to less than 50%. As in HUS associated with inherited deficiencies of complement regulatory proteins, it is not understood why those with life-long genetic deficiency of ADAMTS13 present in adulthood. Such patients tend to follow a relapsing and remitting course.

Agenesis of the Kidney.

Bilateral agenesis, which is incompatible with life, is usually encountered in stillborn infants. It is often associated with many other congenital disorders (e.g., limb defects, hypoplastic lungs) and leads to early death. Unilateral agenesis is an uncommon anomaly that is compatible with normal life if no other abnormalities exist. The opposite kidney is usually enlarged as a result of compensatory hypertrophy. Some patients eventually develop progressive glomerular sclerosis in the remaining kidney as a result of the adaptive changes in hypertrophied nephrons, discussed earlier in the chapter, and in time, chronic kidney disease ensues.

Hypoplasia.

Renal hypoplasia refers to failure of the kidneys to develop to a normal size. This anomaly may occur bilaterally, resulting in renal failure in early childhood, but it is more commonly encountered as a unilateral defect. True renal hypoplasia is extremely rare; most cases reported probably represent acquired scarring due to vascular, infectious, or other parenchymal diseases rather than an underlying developmental failure. Differentiation between congenital and acquired atrophic kidneys may be impossible, but a truly hypoplastic kidney shows no scars and has a reduced number of renal lobes and pyramids, usually six or fewer. In one form of hypoplastic kidney, oligomeganephronia, the kidney is small with fewer nephrons that are markedly hypertrophied.

Ectopic Kidneys.

The development of the definitive metanephros may occur in ectopic foci, usually at abnormally low levels. These kidneys lie either just above the pelvic brim or sometimes within the pelvis. They are usually normal or slightly small in size but otherwise are not remarkable. Because of their abnormal position, kinking or tortuosity of the ureters may cause some obstruction to urinary flow, which predisposes to bacterial infections.

Horseshoe Kidneys.

Fusion of the upper or lower poles of the kidneys produces a horseshoe-shaped structure that is continuous across the midline anterior to the great vessels. This anatomic anomaly is common and is found in about 1 in 500 to 1000 autopsies. Ninety percent of such kidneys are fused at the lower pole, and 10% are fused at the upper pole.

Genetics and Pathogenesis.

A wide range of different mutations in PKD1 and PKD2 has been described, and this allelic heterogeneity has complicated genetic diagnosis of this disorder.

The pathogenesis of polycystic disease is not established, but the hypothesis that is currently favored places the cilia-centrosome complex of tubular epithelial cells at the center of the disorder (Fig. 20-46).^91-93 The epithelial cells of the kidney each contain a single nonmotile primary cilium, a 2–3 μm long hairlike organelle that projects into the tubular lumen from the apical surface of tubular cells. The cilium is made up of microtubules, and arises from and is attached to a basal body derived from the centriole. The cilia are part of a system of organelles and cellular structures that sense mechanical signals. It is believed that the apical cilia function in the kidney tubule as a mechanosensor to monitor changes in fluid flow and shear stress, while intercellular junctional complexes monitor forces between cells, and focal adhesions sense attachment to extracellular matrices. In response to external signals, these sensors regulate ion flux (cilia can induce Ca2+ flux in cultured kidney epithelial cells) and cellular behavior, including cell polarity and proliferation. The hypothesis that defects in mechanosensing, Ca2+ flux, and signal transduction underlie cyst formation is supported several observations.

FIGURE 20-46 Possible mechanisms of cyst formation in polycystic kidney disease (see text).

Polycystin-1 and polycystin-2 may form a protein complex that acts to regulate intracellular Ca2+ in response to fluid flow, perhaps because fluid moving through the kidney tubules causes ciliary bending that opens Ca2+ channels.^91,93 Mutation of either of the PKD genes would lead to loss of the polycystin complex or the formation of an aberrant complex. The consequent disruption of normal polycystin activity then leads to changes in intracellular Ca2+ level and, given the second-messenger effects of Ca2+, to changes in cellular proliferation, basal levels of apoptosis, interactions with the ECM, and secretory function of the epithelia that together result in the characteristic feature of ADPKD. The interaction of PKD1 and PKD2 gene products probably accounts for the similar phenotype in the disease induced by mutations in either of the two genes.⁹¹ The increase in the number of cells caused by abnormal proliferation, and the expanding volume of intraluminal fluid caused by abnormal secretion from epithelial cells lining the cysts, result in progressive cyst enlargement. In addition, cyst fluids have been shown to harbor mediators, derived from epithelial cells, that enhance fluid secretion and induce inflammation. These abnormalities contribute to further enlargement of cysts and the interstitial fibrosis characteristic of progressive polycystic kidney disease.

Morphology. In gross appearance, the kidneys are usually bilaterally enlarged and may achieve enormous sizes; weights as high as 4 kg for each kidney have been reported. The external surface appears to be composed solely of a mass of cysts, up to 3 to 4 cm in diameter, with no intervening parenchyma (Fig. 20-47A and B). However, microscopic examination reveals functioning nephrons dispersed between the cysts. The cysts may be filled with a clear, serous fluid or, more usually, with turbid, red to brown, sometimes hemorrhagic fluid. As these cysts enlarge, they may encroach on the calyces and pelvis to produce pressure defects. The cysts arise from the tubules throughout the nephron and therefore have variable lining epithelia. On occasion, papillary epithelial formations and polyps project into the lumen. Bowman capsules are occasionally involved in cyst formation, and glomerular tufts may be seen within the cystic space.

FIGURE 20-47 A and B, Autosomal-dominant adult polycystic kidney disease (ADPKD) viewed from the external surface and bisected. The kidney is markedly enlarged and contains numerous dilated cysts. C, Autosomal-recessive childhood PKD, showing smaller cysts and dilated channels at right angles to the cortical surface. D, Liver cysts in adult PKD.

Clinical Features.

Many of these patients remain asymptomatic until renal insufficiency announces the presence of the disease. In others, hemorrhage or progressive dilation of cysts may produce pain. Excretion of blood clots causes renal colic. The enlarged kidneys, usually apparent on abdominal palpation, may induce a dragging sensation. The disease occasionally begins with the insidious onset of hematuria, followed by other features of progressive chronic kidney disease, such as proteinuria (rarely more than 2 gm/day), polyuria, and hypertension. Patients with PKD2 mutations tend to have an older age at onset and later development of renal failure. Both genetic and environmental factors influence disease severity. Progression is accelerated in blacks (largely correlated with sickle-cell trait), in males, and in the presence of hypertension.

Individuals with polycystic kidney disease also tend to have extrarenal congenital anomalies.⁸⁷ About 40% have one to several cysts in the liver (polycystic liver disease) that are usually asymptomatic. The cysts are derived from biliary epithelium. Cysts occur much less frequently in the spleen, pancreas, and lungs. Intracranial berry aneurysms, presumably from altered expression of polycystin in vascular smooth muscle, arise in the circle of Willis, and subarachnoid hemorrhages from these account for death in about 4% to 10% of individuals. Mitral valve prolapse and other cardiac valvular anomalies occur in 20% to 25% of patients, but most are asymptomatic. The clinical diagnosis is made by radiologic imaging techniques.

This form of chronic renal failure is remarkable in that patients may survive for many years with azotemia slowly progressing to uremia. Ultimately, about 40% of adult patients die of coronary or hypertensive heart disease, 25% of infection, 15% of a ruptured berry aneurysm or hypertensive intracerebral hemorrhage, and the rest of other causes.

Pathogenesis.

At least seven responsible gene loci have been identified. Three genes, NPH1, NPH2, and NPH3, are mutated in the juvenile forms of nephronophthisis.⁹⁵ The protein products of NPH1 and NPH3–NPH6 have been identified (collectively called nephrocystins), but their functions are not yet known. As discussed earlier, these proteins are present in the primary cilia, basal bodies attached to these cilia, or the centrosome organelle from which the basal bodies originate. The NPHP2 gene product has been identified as inversin, which mediates left-right patterning during embryogenesis.⁹¹ Two genes (MCKD1 and MCKD2), with autosomal dominant transmission, have been identified as causing medullary cystic disease that is characterized by progression to end-stage kidney disease in adult life.⁸⁷

Morphology. The kidneys are small, have contracted granular surfaces, and show cysts in the medulla, most prominently at the corticomedullary junction (Fig. 20-48). Small cysts are also seen in the cortex. The cysts are lined by flattened or cuboidal epithelium and are usually surrounded by either inflammatory cells or fibrous tissue. In the cortex there is widespread atrophy and thickening of the basement membranes of the proximal and distal tubules, together with interstitial fibrosis. Some glomeruli may be hyalinized, but in general, glomerular structure is preserved.

FIGURE 20-48 Medullary cystic disease. Cut section of kidney showing cysts at the corticomedullary junction and in the medulla.

There are few specific clues to diagnosis, because the medullary cysts might be too small to be visualized radiographically. The disease should be strongly considered in children or adolescents with otherwise unexplained chronic renal failure, a positive family history, and chronic tubulointerstitial nephritis on biopsy.

Clinical Features.

Acute obstruction may provoke pain attributed to distention of the collecting system or renal capsule. Most of the early symptoms are produced by the underlying cause of the hydronephrosis. Thus, calculi lodged in the ureters may give rise to renal colic, and prostatic enlargements may give rise to bladder symptoms.

Unilateral complete or partial hydronephrosis may remain silent for long periods, since the unaffected kidney can maintain adequate renal function. Sometimes its existence first becomes apparent in the course of intravenous pyelography. It is regrettable that this disease tends to remain asymptomatic, because in its early stages, perhaps the first few weeks, relief of obstruction leads to reversion to normal function. Ultrasonography is a useful noninvasive technique in the diagnosis of obstructive uropathy.

In bilateral partial obstruction the earliest manifestation is inability to concentrate the urine, reflected by polyuria and nocturia. Some patients have acquired distal tubular acidosis, renal salt wasting, secondary renal calculi, and a typical picture of chronic tubulointerstitial nephritis with scarring and atrophy of the papilla and medulla. Hypertension is common in such patients.

Complete bilateral obstruction results in oliguria or anuria and is incompatible with survival unless the obstruction is relieved. Curiously, after relief of complete urinary tract obstruction, postobstructive diuresis occurs. This can often be massive, with the kidney excreting large amounts of urine that is rich in sodium chloride.

Cause and Pathogenesis.

There are four main types of calculi⁹⁸ (Table 20-12): (1) calcium stones (about 70%), composed largely of calcium oxalate or calcium oxalate mixed with calcium phosphate; (2) another 15% are so-called triple stones or struvite stones, composed of magnesium ammonium phosphate; (3) 5% to 10% are uric acid stones; and (4) 1% to 2% are made up of cystine. An organic mucoprotein matrix, making up 1% to 5% of the stone by weight, is present in all calculi. Although there are many causes for the initiation and propagation of stones, the most important determinant is an increased urinary concentration of the stones’ constituents, such that it exceeds their solubility (supersaturation). A low urine volume in some metabolically normal patients may also favor supersaturation.

TABLE 20-12 Prevalence of Various Types of Renal Stones

Calcium oxalate stones (Table 20-12) are associated in about 5% of patients with hypercalcemia and hypercalciuria, such as occurs with hyperparathyroidism, diffuse bone disease, sarcoidosis, and other hypercalcemic states. About 55% have hypercalciuria without hypercalcemia. This is caused by several factors, including hyperabsorption of calcium from the intestine (absorptive hypercalciuria), an intrinsic impairment in renal tubular reabsorption of calcium (renal hypercalciuria), or idiopathic fasting hypercalciuria with normal parathyroid function. As many as 20% of calcium oxalate stones are associated with increased uric acid secretion (hyperuricosuric calcium nephrolithiasis), with or without hypercalciuria. The mechanism of stone formation in this setting involves “nucleation” of calcium oxalate by uric acid crystals in the collecting ducts. Five percent are associated with hyperoxaluria, either hereditary (primary oxaluria) or, more commonly, acquired by intestinal overabsorption in patients with enteric diseases. The latter, so-called enteric hyperoxaluria, also occurs in vegetarians, because much of their diet is rich in oxalates. Hypocitraturia, associated with acidosis and chronic diarrhea of unknown cause, may produce calcium stones. In a variable proportion of individuals with calcium stones, no cause can be found (idiopathic calcium stone disease).

Magnesium ammonium phosphate stones are formed largely after infections by bacteria (e.g., Proteus and some staphylococci) that convert urea to ammonia. The resultant alkaline urine causes the precipitation of magnesium ammonium phosphate salts. These form some of the largest stones, as the amounts of urea excreted normally are huge. Indeed, so-called staghorn calculi occupying large portions of the renal pelvis are almost always a consequence of infection.

Uric acid stones are common in individuals with hyperuricemia, such as gout, and diseases involving rapid cell turnover, such as the leukemias. However, more than half of all patients with uric acid calculi have neither hyperuricemia nor increased urinary excretion of uric acid. In this group, it is thought that an unexplained tendency to excrete urine of pH below 5.5 may predispose to uric acid stones, because uric acid is insoluble in acidic urine. In contrast to the radiopaque calcium stones, uric acid stones are radiolucent.

Cystine stones are caused by genetic defects in the renal reabsorption of amino acids, including cystine, leading to cystinuria. Stones form at low urinary pH.

It can therefore be appreciated that increased concentration of stone constituents, changes in urinary pH, decreased urine volume, and the presence of bacteria influence the formation of calculi. However, many calculi occur in the absence of these factors; conversely, individuals with hypercalciuria, hyperoxaluria, and hyperuricosuria often do not form stones. It has therefore been postulated that stone formation is enhanced by a deficiency in inhibitors of crystal formation in urine. The list of such inhibitors is long, including pyrophosphate, diphosphonate, citrate, glycosaminoglycans, osteopontin, and a glycoprotein called nephrocalcin.

Morphology. Stones are unilateral in about 80% of patients. The favored sites for their formation are within the renal calyces and pelves (Fig. 20-51) and in the bladder. If formed in the renal pelvis they tend to remain small, having an average diameter of 2 to 3 mm. These may have smooth contours or may take the form of an irregular, jagged mass of spicules. Often many stones are found within one kidney. On occasion, progressive accretion of salts leads to the development of branching structures known as staghorn calculi, which create a cast of the pelvic and calyceal system.

FIGURE 20-51 Nephrolithiasis. A large stone impacted in the renal pelvis.

(Courtesy of Dr. E. Mosher, Brigham and Women’s Hospital, Boston, MA.)

Clinical Features.

Stones are of importance when they obstruct urinary flow or produce ulceration and bleeding. They may be present without producing any symptoms or they may cause significant renal damage. In general, smaller stones are most hazardous, because they may pass into the ureters, producing colic, one of the most intense forms of pain, and ureteral obstruction. Larger stones cannot enter the ureters and are more likely to remain silent within the renal pelvis. Commonly, these larger stones first manifest themselves by hematuria. Stones also predispose to superimposed infection, both by their obstructive nature and by the trauma they produce.

Epidemiology.

Tobacco is the most significant risk factor. Cigarette smokers have double the incidence of renal cell carcinoma, and pipe and cigar smokers are also more susceptible. An international study has identified additional risk factors, including obesity (particularly in women); hypertension; unopposed estrogen therapy; and exposure to asbestos, petroleum products, and heavy metals.^101,102 There is also an increased incidence in patients with chronic renal failure and acquired cystic disease (see earlier) and in tuberous sclerosis.

Most renal cancer is sporadic, but unusual forms of autosomal dominant familial cancers occur, usually in younger individuals. Although they account for only 4% of renal cancers, familial variants have been enormously instructive in studying renal carcinogenesis.

Classification of renal cell carcinoma: histology, cytogenetics, and genetics.

The classification of renal cell carcinoma is based on correlative cytogenetic, genetic, and histologic studies of both familial and sporadic tumors.^103,104 The major types of tumor are as follows (Fig. 20-52):

FIGURE 20-52 Cytogenetics (blue) and genetics (red) of clear cell versus papillary renal cell carcinoma.

(Courtesy of Dr. Keith Ligon, Brigham and Women’s Hospital, Boston, MA.)

New variants of renal cell carcinoma that are distinctive (histologically, genetically, and clinically) are being recognized as a result of molecular profiling, illustrating how application of these techniques may improve our clinical understanding and management of these neoplasms.¹⁰⁹

Morphology. Renal cell carcinomas may arise in any portion of the kidney, but more commonly affects the poles. Clear cell carcinomas arise most likely from proximal tubular epithelium, and usually occur as solitary unilateral lesions. They are spherical masses, which can vary in size, composed of bright yellow-gray-white tissue that distorts the renal outline. The yellow color is a consequence of the prominent lipid accumulations in tumor cells. There are commonly large areas of ischemic, opaque, gray-white necrosis, and foci of hemorrhagic discoloration. The margins are usually sharply defined and confined within the renal capsule (Fig. 20-53). Papillary tumors, thought to arise from distal convoluted tubules, can be multifocal and bilateral. They are typically hemorrhagic and cystic, especially when large. Papillary carcinomas are the most common type of renal cancer in patients who develop dialysis-associated cystic disease.

FIGURE 20-53 Renal cell carcinoma. Typical cross-section of yellowish, spherical neoplasm in one pole of the kidney. Note the tumor in the dilated thrombosed renal vein.

As tumors enlarge they may bulge into the calyces and pelvis and eventually may fungate through the walls of the collecting system to extend into the ureter. One of the striking characteristics of renal cell carcinoma is its tendency to invade the renal vein (see Fig. 20-53) and grow as a solid column of cells within this vessel. Further growth may produce a continuous cord of tumor in the inferior vena cava that may extend into the right side of the heart.

In clear cell carcinoma the growth pattern varies from solid to trabecular (cordlike) or tubular (resembling tubules). The tumor cells have a rounded or polygonal shape and abundant clear or granular cytoplasm, which contains glycogen and lipids (Fig. 20-54A). The tumors have delicate branching vasculature and may show cystic as well as solid areas. Most tumors are well differentiated, but some show marked nuclear atypia with formation of bizarre nuclei and giant cells. Papillary carcinoma is composed of cuboidal or low columnar cells arranged in papillary formations. Interstitial foam cells are common in the papillary cores (Fig. 20-54B). Psammoma bodies may be present. The stroma is usually scanty but highly vascularized. Chromophobe renal carcinoma is made up of pale eosinophilic cells, often with a perinuclear halo, arranged in solid sheets with a concentration of the largest cells around blood vessels (Fig. 20-54C). Collecting duct carcinoma is a rare variant showing irregular channels lined by highly atypical epithelium with a hobnail pattern. Sarcomatoid changes arise infrequently in all types of renal cell carcinoma and are a decidedly ominous feature.

FIGURE 20-54 Renal cell carcinoma. A, Clear cell type. B, Papillary type. Note the papillae and foamy macrophages in the stalk. C, Chromophobe type.

(Courtesy of Dr. A. Renshaw, Baptist Hospital, Miami, FL.)

Clinical Features.

The three classic diagnostic features of renal cell carcinoma are costovertebral pain, palpable mass, and hematuria, but these are seen in only 10% of cases. The most reliable of the three is hematuria, but it is usually intermittent and may be microscopic; thus, the tumor may remain silent until it attains a large size. At this time it is often associated with generalized constitutional symptoms, such as fever, malaise, weakness, and weight loss. This pattern of asymptomatic growth occurs in many patients, so the tumor may have reached a diameter of more than 10 cm when it is discovered. Currently, an increasing number of tumors are being discovered in the asymptomatic state by incidental radiologic studies (e.g., computed tomographic scan or magnetic resonance imaging) usually performed for nonrenal indications.

Renal cell carcinoma is classified as one of the great mimics in medicine, because it tends to produce a diversity of systemic symptoms not related to the kidney. In addition to fever and constitutional symptoms mentioned earlier, renal cell carcinomas produce a number of paraneoplastic syndromes (Chapter 7), ascribed to abnormal hormone production, including polycythemia, hypercalcemia, hypertension, hepatic dysfunction, feminization or masculinization, Cushing syndrome, eosinophilia, leukemoid reactions, and amyloidosis.

One of the common characteristics of this tumor is its tendency to metastasize widely before giving rise to any local symptoms or signs. In 25% of new patients with renal cell carcinoma, there is radiologic evidence of metastases at the time of presentation. The most common locations of metastasis are the lungs (more than 50%) and bones (33%), followed in frequency by the regional lymph nodes, liver, adrenal, and brain.

The average 5-year survival rate of persons with renal cell carcinoma is about 45% and as high as 70% in the absence of distant metastases. With renal vein invasion or extension into the perinephric fat, the figure is reduced to approximately 15% to 20%. Nephrectomy has been the treatment of choice, but partial nephrectomy to preserve renal function is being done with increasing frequency and similar outcome.