The precise decrement in GFR that constitutes acute renal failure is ill-defined, and varies in different studies. Using absolute increments in the plasma creatinine concentration (P Cr ) is especially problematic (see next slide).
This slide depicts the inverse relationship between P Cr and GFR (measured by inulin clearance) in a large number of subjects with varying degrees of renal function. The hyperbolic relationship between P Cr and GFR complicates the use of absolute increments in P Cr (e.g., > 0.5 or 1.0 mg/dl) as yardsticks for defining acute renal failure.
To function properly, the kidney requires: (1) normal blood flow; functioning glomeruli and tubules to separate and process an ultafiltrate containing waste products from the blood; and (3) drainage and elimination of formed urine from the body. The sudden interruption of any of these processes will lead to Acute Renal Failure (ARF) . Disorders causing ARF are classified on the basis their primary site of interference with these processes. Conditions which interfere with blood delivery to the kidney are called Prerenal, and are most commonly functional (and potentially reversible) in nature (e.g., ECF volume contraction, congestive heart failure) but on occasion may be structural (e.g., renal artery stenosis). Diseases which cause intrinsic injury to the kidney proper (glomeruli, tubules, interstitium, small blood vessels) are grouped under Renal causes of ARF (e.g., acute glomerulonephritis, acute tubular necrosis, acute interstitial nephritis or small vessel vasculitis). Acute Tubular Necrosis is a distinctive clinicopathological syndrome in which the tubules are the primary site of injury. The terms ARF and ATN should not be used interchangeably. Finally, conditions which interfere with normal drainage and elimination of formed urine are classified as Postrenal (e.g., prostatic outlet obstruction, bilateral ureteral obstruction). Pre-renal ARF (also commonly referred to as “pre-renal azotemia”) and acute tubular necrosis (ATN) are the most common causes of acute renal failure in hospitalized patients.
Some combination of hypovolemia, hypotension and diminished renal perfusion is the most common cause of ARF in hospitalized patients. Identification of pre-renal (functional) ARF is important because it is generally reversible. Pre-renal ARF may evolve from blood loss, sodium depletion (due to diarrhea, excessive diuresis or congenital or acquired salt-wasting disorders), redistribution of plasma volume to a so-called “third space” (e.g., ascites in patients with hemorrhagic pancreatitis or hepatic cirrhosis), or reductions in effective arterial blood volume with consequent renal hypoperfusion (as in congestive heart failure, hepatic cirrhosis or nephrotic syndrome). In other situations, especially when for one reason or another renal perfusion is tenuous or already compromised, drugs which affect afferent and/or efferent arteriolar resistance (e.g., NSAIDs, ACE inhibitors) can precipitate pre-renal azotemia.
The determinants of renal blood flow (RBF) are summarized in this slide. In the absence of renal artery stenosis, renal arterial pressure (RAP) is the same as systemic mean arterial pressure (commonly referred to by nephrologists as “renal perfusion pressure”). Renal venous pressure (RVP) is usually, but not always, low and relatively constant. The glomerular afferent and efferent arterioles are the major sites of renal vascular resistance (R aff and R eff , respectively); changes in either will affect renal blood flow (RBF). In the “universal pressure;flow relationship”, F = flow; P = change in pressure; R = resistance. RBF = renal blood flow; RAP = renal arterial pressure; RVP = renal venous pressure; R aff = afferent arteriolar resistance; R eff = efferent arteriolar resistance
It is useful to think about glomerular capillary hydrostatic pressure (P GC ), and its primary role in determining GFR (blue arrow), in terms of the ratio of efferent to afferent arteriolar resistance (i.e., R eff /R aff ). Increases in R eff /R aff increase P GC and therefore tend to increase GFR. Similarly, decreases in R eff /R aff decrease P GC and therefore tend to decrease GFR.
The kidney is able to maintain a relatively constant glomerular filtration rate (GFR) and to a lesser extent renal blood flow (RBF) despite modest reductions in mean arterial pressure. This process, termed renal autoregulation , derives largely from hormonally-mediated adjustments in afferent and efferent arteriolar resistance. By increasing R eff /R aff angiotensin-II and prostaglandin-E maintain P GC and GFR at the expense of RBF. As a result, filtration fraction (GFR/RBF) and post-glomerular capillary oncotic pressure increase, which facilitates Na and water reabsorption from the adjacent proximal tubule. Angiotensin-II also directly increases proximal tubular Na reabsorption and by stimulating aldosterone synthesis and release increases Na reabsorption in more distal tubular segments as well. At the same time, volume-related antidiuretic hormone release leads to enhanced water and urea reabsortion in the collecting duct. Urine formed under these conditions is of reduced volume , highly concentrated , and contains scant amounts of sodium . These characteristics are the basis of tests (see later slides) for distinguishing pre-renal azotemia from acute tubular necrosis. Renal autoregulation breaks down as mean arterial pressure falls below about 80 mm Hg, at which point further adjustments in intra-renal hemodynamics are unable to maintain GFR and RBF in the face of a progressive reduction in renal perfusion pressure.
Ischemia has been recognized as a cause of acute renal failure since the 1940s. This form of ATN includes any pathophysiologic state in which volume depletion, impaired cardiac output, or vascular pooling results in renal hypoperfusion. The degree and duration of renal hypoperfusion (ischemia) necessary to cross from a readily reversible form of ARF (pre-renal) to a less reversible or irreversible form of ARF despite correction of the hypoperfusion (post- ischemic acute tubular necrosis) is highly variable. In addition to ischemia, there are a variety of endogenous and exogenous toxins that can lead to acute tubular necrosis.
Acute tubular necrosis showing focal loss of tubular epithelial cells (arrows) and partial occlusion of tubular lumens by cellular debris (D) (H&E stain). The histologic findings in ischemic acute tubular necrosis may be disproportionately small in comparison to the magnitude of renal dysfunction. Focal necrosis of single proximal tubular cells and clusters of cells is seen in the majority of cases of ATN. Portions of tubules not involved by necrosis commonly show effacement of the proximal tubule brush border. Other pathologic findings include flattening of the tubular epithelium and dilatation of the tubular lumina. Many of the flattened tubular cells exhibit signs of regeneration – i.e., mitoses and large hyperchromatic nuclei. Regenerative changes and foci of necrosis are often seen in the same biopsy specimen. Distal nephron segments are characteristically occluded by urinary casts of hyaline, granular, and pigmented varieties.
Another example of ischemic ATN (PAS stain).
Some of the more common nephrotoxins associated with acute tubular necrosis are listed on this slide.
Tubular epithelial degeneration and hyaline amphophilic casts (positive with immunologic stains for myoglobin) in a patient with rhabdomyolysis and myoglobinuric acute tubular necrosis.
The tubular epithelial cells show extensive cytoplasmic vacuolar change in a case of ethylene glycol poisoning.
Acute interstitial nephritis (AIN) is most often caused by drugs (in such cases it is often called “allergic intersitial nephritis”), although bacterial (including Legionella and leptospirosis) and viral infections may also be responsible. Sarcoidosis causes granulomatous interstitial nephritis. Although the number of drugs associated with AIN is large, relatively few have been reported to cause AIN with any frequency. AIN was initially reported with methicillin but this antibiotic is rarely used today. However, other beta-lactam antibiotics (penicillins and cephalosporins) remain a frequently reported cause of AIN. Of the quinolones, AIN has been most often seen with ciprofloxacin. NSAIDs can cause either AIN, nephrotic syndrome, or both. Although reported with cimetidine, AIN is extremely rare with other H 2 receptor blockers such as ranitidine. Sulfonamide-containing antibiotics (sulfamethoxazole in the trimethoprim-sulfamethaxazole combination) and diuretics derived from sulfonamides (thiazides, bumetamide, furosemide) can also cause AIN.
Most patients with allergic intersitial nephritis have will have fever and in many cases this will be accompanied by a rash. However, the other “classic” findings listed on this slide are much more variable, and all one may see is fever and otherwise unexplained acute renal failure. In NSAID-induced allergic interstitial nephritis fever, rash and eosinophilia are typically absent and acute renal failure and nephrotic level proteinuria may be the only findings. Significant proteinuria is unusual in other forms of allergic interstitial nephritis. The sediment will usually show WBC (with a sterile culture), WBC casts, and RBCs. Previously eosinophiluria was regarded as highly specific for AIN, but experience has not borne this out.
Drug-induced allergic interstitial nephritis (H&E stain). Note the diffuse interstitial infiltrate, many red-staining eosinophils, and sparing of the glomerulus (on the left).
Cholesterol embolization may occur after a “sentinel” procedure (e.g., cardiac catherization) or be associated with slowly progressive renal failure over a period of months or years. Whatever the time course, it is generally irreversible. Cholesterol emboli usually lodge in vessels 100 to 200 microns in diameter, and are visualized as clear spaces where the cholesterol crystals have been dissolved by routine processing. The early response of platelets and occasional mononuclear cells is seen in the medium size artery occluded by acute cholesterol emboli in the left panel (Jones Silver stain). In later stages of organization the lumen may have more fibrous reorganization surrounding the cholesterol clefts, as shown in the right panel (PAS stain). Chlosterol embolization is an often unrecognized cause of acute renal failure which may be indistinguishable from the “bland” variety of allergic interstitial nephritis except by biopsy. In addition to ARF, other manifestations of the occlusion of small arteries by atheroembolic material include skin mottling (livedo reticularis), blue toes and distal digital infarcts with intact peripheral pulses. Transient eosinophilia, hypocomplementemia, and an elevated sedimentation rate are sometimes also seen.
In addition to ischemia, contrast media remains one of the most common forms of acute tubular necrosis. It’s reported prevalence varies depending on study design (retrospective or prospective), the criteria used to define ARF, and if general or high-risk populations are evaluated. The single most important risk factor is underlying renal insufficiency, which may not always be obvious even with routine laboratory studies.
The greater the degree of renal insufficiency, the greater the risk for contrast-induced ARF. Although diabetes is commonly listed as a risk factor, there is little increased risk for contrast ARF in diabetics with normal renal function. However, it does appear that diabetics with impaired renal function are at greater risk than non-diabetics with comparable degrees of renal insufficiency. Multiple myeloma is more of a historical risk factor for contrast ARF and was initially reported with early preparations of contrast media which had diminished solubility or enhanced precipitation with Tamm-Horsfall protein compared to modern day contrast media. Contrast media used in the 1970s and 1980s were negatively charged benzene rings and thus were coupled with cations such as sodium or meglumine. Thus, two molecules were required to deliver the iodine attached to each benzene ring, therefore increasing the osmolality of the contrast media. These early versions are referred to as ionic or high-osmolar media and are less commonly used today. Current contrast media are iodine containing benzene rings which are not charged and thus their osmolality is roughly half of the high-osmolar compounds. These compounds are referred to as non-ionic or low osmolar contrast media. In patients with underlying renal insufficiency especially those with diabetes (but not in patients with normal renal function), prospective studies have shown a reduced incidence of ARF with non-ionic compared to ionic contrast media. Although there is no ‘threshold’ dose of contrast media which if exceeded will significantly increase the risk of ARF, in general the greater the volume of contrast given the greater the risk for ARF.
For the majority of patients with contrast-induced ARF, onset as manifested by a rise in serum creatinine is seen within the first 24 hrs after contrast exposure. A small number of additional cases may require 48 hrs after exposure to become apparent. If ARF is not manifested by 48 hrs after exposure to contrast, it is very unlikely to occur. Fortunately for the majority of patients (~ 75%) with contrast ARF, the renal failure is mild, its duration is brief, most patients are non-oliguric, and dialysis is rarely needed. The urinary sediment may contain the “muddy-brown” pigmented casts and renal tubular cells typical of ATN (to be discussed later) or may be quite bland. Somewhat atypical for acute tubular necrosis, the fractional excretion of sodium is often low (also to be discussed later). Finally, in patients who are administered contrast media through an arterial vessel, contrast media ARF needs to be distinguished from atheroembolic (cholesterol) ARF.
Since the administration of contrast media is predictable, numerous prophylactic strategies have been proposed to prevent ARF. In general, radiologic procedures requiring parenteral contrast media administration should be avoided, especially in “high-risk” patients, when other imaging procedures (ultrasound, MRI, CT without contrast) are able to provide equivalent information. Current evidence suggests that the imaging agent gadolinium used in magnetic resonance imaging is not nephrotoxic. Prophylactic strategies are usually reserved for patients at “high-risk” for contrast ARF (i.e., have renal insufficiency). Hydration with 1/2 normal or normal saline remains the standard for prophylaxis. The optimal volume and duration of hydration have not been definitively determined but most protocols require 8-12 hrs of hydration before and after contrast exposure. The use of low-osmolar contrast media at the smallest possible volume is obviously important. N-acetylcysteine or the selective D-1 dopamine receptor agonist fenoldopam are currently popular prophylactic strategies, but large prospective randomized studies supporting their efficacy are lacking. Mannitol, furosemide and atrial natriuretic peptide appear ineffective.
The nature of the obstructing lesion, the site of the obstruction, the rapidity of onset, and the magnitude of the obstruction are all important determinants of the presentation of postrenal ARF. Since postrenal ARF is often reversible, it is essential that the clinician quickly recognize and correct the cause of obstruction. In addition to a careful history and physical examination and examination of the urinary sediment, renal ultrasound and spiral computed tomography are the diagnostic tools most helpful in detecting obstruction. Because of ‘compensatory’ increases in GFR in the contralateral non-obstructed kidney, unilateral ureteral obstruction does not usually result in a rise in the serum creatinine concentration.
A careful microscopic examination of the urine sediment, quantification of the urine volume, determination of urinary electrolytes, and a variety of radiologic studies are the tools the clinician uses in conjunction with a thorough history and physical examination to determine the cause of ARF in any given patient.
A carefully performed urinalysis is a critical tool in determining the cause of acute renal failure.
RBC casts when present are virtually diagnostic of glomerulonephritis or vasculitis. However, the absence of RBC casts does not exclude glomerulonephritis. In recent years RBC morphology has been used to distinguish the source of hematuria. Dysmorphic RBC strongly suggest glomerular bleeding. Examples of dysmorphism include a’punched-out’ central defect (‘doughnut -like’), diverticular projections from the cell rim (‘blebs’), and partially destroyed and fragmented RBC.
Two examples of red blood cell casts, typical of glomerular bleeding.
Monomorphic (non-dysmorphic) RBC suggest non-glomerular source of bleeding – i.e., bleeding from the calyces, pelvis, ureter(s), bladder, prostate or urethra. Dysmorphic red blood cells suggest glomerular injury.
Dysmorphic red blood cells are best viewed under phase-contrast microscopy (as in this example). Note the ‘blebs’ and punched-out centers (arrows).
Another example of dysmorphic RBC (viewed by scanning electron microscopy).
The presence of WBC alone or in association with WBC casts suggests either allergic interstitial nephritis or acute pyelonephritis. A urine culture can usually differentiate these two possibilities. Eosinophiluria is suggestive of allergic interstitial nephritis.
White bood cells are sometimes called “glitter” cells because of the Brownian motion of the neutrophilic particles.
Note the clear cell outlines and granular cytoplasm.
If the urine demonstrates pigmented granular (“muddy-brown”) casts and/or renal tubular epithelial (RTE) cells in large numbers or in casts, acute tubular necrosis should be strongly considered.
Pigmented granular (“muddy brown”) casts are characteristic of acute tubular necrosis. Although the exact pathogenesis of cast formation is not known, the major component is Tamm-Horsfall glycoprotein, a strongly anionic macromolecule secreted by the ascending thick limb of Henle. Hyaline and waxy casts are also prominent in this particular urinary sediment.
The volume of urine excreted over a 24 hour period is often helpful in narrowing the differential diagnosis of ARF. Anuria is defined by some authors as the virtual absence of urine while others apply the term for daily urine volumes less than 100 ml. Prolonged anuria is uncommon in ATN but sometimes can be seen in the early stages of ATN if hypotension or volume depletion leads to intense salt and water reabsorption.
Oliguria is defined by a urine volume between 100 and 500 ml/24h*. The two most common causes of oliguria in hospitalized patients are prerenal azotemia (ECF volume depletion, for example) and ATN. Non-oliguric ARF (sometimes called “polyuric” ARF) is most commonly due to ATN, but partial urinary outflow obstruction must also be considered. *Normal individuals can concentrate their urine to a maximum of about 1200 mOsm/L. To remain in solute balance on an average diet requires the daily elimination of about 600 mOsm of electrolytes. Thus, under physiologic conditions, the daily solute load could be eliminated in about 500 ml of urine.
Prior to the article by Anderson (reference below), it was believed that most cases of ATN were oliguric. Today, we know that ATN can present with oliguria or non-oliguria and that both presentations are common. Any cause of ATN can present with nonoliguria ; non-oliguria is more likely with nephrotoxic causes of ATN such as aminoglycosides, contrast media, cis-platinum, and amphotericin. The increased incidence of non-oliguric ATN during the past 25 years is most likely due to the increased usage of nephrotoxins, more frequent chemical testing, and more aggressive use of fluids, potent diuretics, and vasodilators in the management of ATN. The reduced mortality of non-oliguric compared to oliguric ATN is probably not because of the increased urine volume but rather due to a lower associated mortality of the conditions causing non-oliguric compared to oliguric ATN. (Data from Anderson et al: Non-oliguric Acute Renal Failure. New Engl J Med 296:134, 1977.)
As discussed, the main differential in the oliguric patient with acute renal failure is between pre-renal azotemia (PR) and acute tubular necrosis (ATN). Although a careful history and physical examination coupled with a careful urinalysis will often distinguish between these two conditions, the use of urinary electrolytes provides further information. The basis of the urinary electrolytes is the different tubular responses to salt and water conservation. Tubular function with pre-renal azotemia is normal allowing maximum tubular sodium and water reabsorption resulting in a concentrated urine that is low in sodium. In acute tubular necrosis, tubular dysfunction leads to sodium wasting and an inability to concentrate the urine. The ratio of urine to plasma creatinine concentrations (U/P) Cr has also been proposed as a discriminating marker. In pre-renal ARF, U Cr is high due to water reabsorption without Cr reabsorption and P Cr is usually only mildly increased resulting in a high (U/P) Cr ratio. In ATN, the U Cr is lower due to an inbility to concentrate the urine and the P Cr is increased in prpoprtion to the degree of renal failure so the U/P cr is generally lower than that seen in pre-renal azotemia. Unfortunately, as shown in the slide, there is considerable overlap (“grey zone”) in all these indices. The so-called “renal failure index” [RFI = U Na /(U/P) Cr ] and the more commonly employed fractional excretion of Na [FE Na = 100(U Na X P Cr ) / (P Na X U Cr )], which combine different “single” indices, provide better discrimination. Recent data have demonstrated that the fractional excretion of sodium (FE Na ) , which expresses the fraction of filtered sodium that escapes reabsorption and eventually appears in the urine, is a more discriminating test to distinguish between pre-renal azotemia and oliguric ATN. FE Na > 1 % strongly suggests ATN while FE Na < 1 % suggests pre-renal azotemia. However, the FE Na is not infallible, and there are a number of exceptions where pre-renal azotemia can be associated with FE Na values greater than 1 % - e.g., recent diuretic use or pre-renal azotemia superimposed on chronic renal insufficiency. In a similar manner, many instances of ATN will have FE Na < 1 % - e.g., early phase of contrast-induced ATN, rhabdomyolysis or septic ATN. Thus, the clinician should utilize the FE Na in conjunction with the overall clinical picture and other lab tests and should not be ‘wedded’ to a particular FE Na result when other data suggest a different cause for ARF.
Intravenous pyelogram showing bilateral hydronephrosis and hydroureter (arrow) due to ureterovesical obstruction. Radiologic studies such as the IVP ( intravenous pyelogram; also called an excretory urogram) or routine CAT scans which require intraveous contrast media are now rarely performed as the initial screening tests to see if post-renal (i.e., obstructive) ARF is present. As discussed earlier, the risk of ATN from contrast media is greatly enhanced in the presence of a reduced GFR, which is the case in patients with ARF. In addition, contrast media is eliminated by glomerular filtration – thus, in the presence of reduced GFR from obstruction, delayed films (several hours) after injection are often required before enough contrast has been filtered and fills the pelvocalyceal system. Current radiologic methods to screen for obstruction use renal ultrasounds or in some cases helical or spiral CAT scans without IV contrast.
The normal kidney is approximately 10 cm in its longitudinal length and 5-6 cm in its transverse diameter. The relatively echo-free renal parenchyma surrounds a central section of tightly compacted echodensities caused by the urothelium of the pelvis and calyces. Hydronephrosis is detected by ultrasonography as an enlarged echo-free area that spreads apart the normal central echodensities. As the obstruction persists and worsens, dilated calyces appear as echo-free fluid pouches extending from the periphery of the expanded echo-free pelvis. In the setting of obstruction, the renal ultrasound has been demonstrated to have a sensitivity of approximately 98% and a specificity of 75%. This high sensitivity makes the renal ultrasound an excellent screening test to look for obstruction. Reasons for false negative exams in addition to performing the study immediately after obstruction include dehydration, extreme obesity, bowel gas or retroperitoneal fibrosis. The lower specificity and higher false positive rate is due to the difficulty in distinguishing a normal extra-renal pelvis from mild hydronephrosis.
This longitudinal ultrasound shows a kidney with moderate hydronephrosis. The dilation of the collecting system extends from the renal pelvis to the calyces. The parenchyma is relatively normal in thickness. Dilatation of the collecting system is detectable by ultrasonography within 24 hrs of urinary outflow obstruction. Thus, it is possible (but very unusual) that patients evaluated within only a couple hours of the onset of obstruction may still have normal renal ultrasounds.
Abdominal CAT scan with contrast demonstrates right hydronephrosis and hydroureter as a consequence of ureteral obstruction. CAT scans can often demonstrate hyodronephrosis even without intravenous contrast. However, CAT scans are much more expensive than renal ultrasounds and should be used for screening purposes only if the renal ultrasound cannot be performed for technical reasons (e.g., extreme obesity).
15 Cox Acute Renal Failure
Acute Renal Failure Malcolm Cox, M.D.
Acute Renal Failure Definition <ul><li>Acute decrement in GFR </li></ul><ul><li>May heal partially or completely or progress to more severe renal insufficiency, including end-stage renal disease </li></ul>
Pre-Renal Azotemia Pathophysiology <ul><li>Renal hypoperfusion </li></ul><ul><ul><li>Decreased RBF and GFR </li></ul></ul><ul><ul><li>Increased filtration fraction (GFR/RBF) </li></ul></ul><ul><li>Increased Na and H 2 O reabsorption </li></ul><ul><ul><li>Oliguria, high U osm , low U Na </li></ul></ul><ul><ul><li>Elevated BUN/Cr ratio </li></ul></ul>