You are on page 1of 9

The n e w e ng l a n d j o u r na l of m e dic i n e

review article

current concepts

Normotensive Ischemic Acute Renal Failure


J. Gary Abuelo, M.D.

A 
cute renal failure is defined as a rapid decrease in the glomer- From the Division of Kidney Disease and
ular filtration rate, occurring over a period of minutes to days. Because the Hypertension, Department of Medicine,
Rhode Island Hospital and the Warren
rate of production of metabolic waste exceeds the rate of renal excretion in Alpert Medical School of Brown Univer-
this circumstance, serum concentrations of markers of renal function, such as urea sity, Providence, RI. Address reprint re-
and creatinine, rise. The causes of acute renal failure are classically divided into quests to Dr. Abuelo at the Division of
Kidney Disease and Hypertension, Rhode
three categories: prerenal, postrenal (or obstructive), and intrinsic. Prerenal azotemia Island Hospital, 593 Eddy St., Providence,
is considered a functional response to renal hypoperfusion, in which renal structure RI 02903, or at jgabuelo@lifespan.org.
and microstructure are preserved. Postrenal azotemia — obstruction of the urinary
N Engl J Med 2007;357:797-805.
tract — is initially accompanied by few microscopical changes (early hydronephrosis, Copyright © 2007 Massachusetts Medical Society.
with enlargement of the pelvic cavity and minimal distention or blunting of the
renal papilla) or none. In contrast, intrinsic renal azotemia is due to parenchymal
injury of the blood vessels, glomeruli, tubules, or interstitium. In prerenal and
postrenal azotemia, complete recovery may be seen 1 to 2 days after relief of the
offending lesion, provided that normal perfusion or urinary outflow is reestablished
before structural changes occur.
A research focus group organized by the American Society of Nephrology recent­
ly recommended that the term “acute kidney injury” replace the term “acute renal
failure.”1 However, the group left the actual definition of acute kidney injury to be
determined in the future. Thus, whether acute kidney injury refers only to acute
tubular necrosis or includes prerenal and postrenal azotemia and parenchymal dis­
eases such as acute glomerulonephritis remains unclear. Most clinicians still use
the term acute renal failure as defined above.
The two forms of ischemic acute renal failure, prerenal azotemia and acute
tubular necrosis, account for more than half the cases of renal failure seen in
hospitalized patients and are familiar to most clinicians.2-4 Yet in many patients
with acute renal failure, the contribution of ischemia is initially unrecognized.
Patients with ischemic acute renal failure typically have low systemic perfusion,
sometimes caused by volume depletion, although their blood pressure may not fall
dramatically but instead may remain within the normal range (in an adult, systolic
blood pressure >90 to 100 mm Hg). In such cases, in the absence of frank hypo­
tension, the clinician may speculate that an unobserved drop in blood pressure must
have caused the renal failure. Although this scenario cannot be ruled out, other
causative mechanisms can usually be identified. This type of ischemic acute renal
failure (termed normotensive, because the patient’s blood pressure is — at least
temporarily — within the normal range) can occur as a result of several processes,
most of which involve increased renal susceptibility to modest reductions in perfu­
sion pressure. Fortunately, the factors that lead to ischemic renal failure in patients
with apparently normal blood pressure are discernible in most instances. Recogni­
tion of these factors allows the physician to make an early diagnosis and facilitates
the interventions that can help to reestablish normal renal hemodynamics.
The importance of addressing even mild renal failure is illustrated by a recent
study showing that hospitalized patients with a modest increase in the serum

n engl j med 357;8  www.nejm.org  august 23, 2007 797


The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

creatinine level (0.3 to 0.4 mg per deciliter [26.5


to 35.4 μmol per liter]) have a 70% greater risk of 100
Normal autoregulation of GFR
death than persons without any increase.5 This

GFR (% of normal rate)


article reviews the renal response to ischemia, 90
notes the risk factors for the development of nor­
motensive ischemic acute renal failure, and dis­ 80 Normotensive renal failure
cusses the diagnosis of this condition.
70

R ena l R e sp onse t o Is che mi a


0
The salient feature of the renal response to a drop 0 80 100 120
in perfusion pressure is autoregulation — main­ Mean Arterial Pressure (mm Hg)
tenance of normal blood flow and glomerular
filtration rate, even with mean arterial pressure Figure 1. Normal and Impaired Autoregulation
of the AUTHOR: Abuelo
ICMGlomerular Filtration Rate during Reduction
RETAKE 1st
as low as 80 mm Hg (Fig. 1).6 Each of the million 2nd
of Mean
REG F Arterial
FIGURE:Pressure.
1 of 3
or so glomeruli per kidney has an afferent (incom­ 3rd
In normal
CASE autoregulation, the glomerular filtration
Revised rate
ing) and an efferent (outgoing) arteriole, and pres­ (GFR) is maintained until theLine 4-C pressure
mean arterial
EMail SIZEfalls
sure within the glomerular capillaries is affected below
Enon
ARTIST: ts
80 mm Hg. However,H/T H/Twith impaired
in patients 22p3
Combo
by the resistances in these two arterioles as they autoregulation, the GFR falls below normal values
AUTHOR, PLEASE NOTE: within the
respond to a variety of vasoconstrictor and vaso­ while the mean arterial pressure remains
Figure has been redrawn and type has been reset.
dilatory factors, including a myogenic stretch normal range, resulting in normotensive
Please check carefully. ischemic
acute renal failure.
reflex in the afferent arteriole. Autoregulation dur­
JOB: 35708 ISSUE: 08-23-07
ing a decrease in renal-artery pressure derives
mainly from a drop in afferent glomerular arterio­ Tamm–Horsfall protein, which is normally secret­
lar resistance, mediated in large part by prosta­ ed by the loop of Henle, forming a gel and con­
glandins (Fig. 2A and 2B). This drop in afferent tributing to cast formation.10,12
resistance sustains glomerular capillary pressure, The mechanism whereby ischemia and oxygen
the driving force of filtration. Glomerular capil­ depletion injure tubular cells starts with ATP de­
lary pressure is also partly supported by an in­ pletion, which activates a number of critical alter­
crease in efferent glomerular arteriolar resistance, ations in metabolism (Fig. 3).13,14 Cytoskeletal
mediated largely by angiotensin II.7-9 disruption leads to loss of brush-border micro­
As renal perfusion pressure drops below the villi and cell junctions and to mislocation of inte­
autoregulatory range, endogenous vasoconstric­ grins and sodium–potassium ATPase from the
tors increase afferent arteriolar resistance. This basal surface to the apical surface. As a result,
reduces glomerular capillary pressure and the brush-border membranes and cells slough and
glomerular filtration rate, resulting in functional may obstruct tubules downstream. ATP depletion
prerenal azotemia.7,9,10 The postglomerular cap­ also activates harmful proteases and phospho­
illary bed, which perfuses the tubules, has dimin­ lipases, which, with reperfusion, cause oxidant
ished blood flow and pressure, but the tubules injury to tubular cells. Similar damage occurs in
remain intact. However, increasing severity and endothelial cells of the peritubular capillaries,
duration of ischemia may cause structural tubular especially in the outer medulla, which is margin­
injury, further impairing renal function. Sloughed ally oxygenated under normal circumstances.
tubular epithelial cells and brush-border–mem­ This oxidant injury, together with a shift in the
brane debris form casts that obstruct tubules, balance of vasoactive substances toward vasocon­
and experimental data indicate that glomerular strictors such as endothelin, results in vasocon­
filtrate leaks from the tubular lumen across de­ striction, congestion, hypoperfusion, and expres­
nuded tubular walls into capillaries and the circu­ sion of adhesion molecules.14 The expression of
lation (a phenomenon called back-leak).3,9-11 In adhesion molecules, in turn, initiates leukocyte
addition, impaired sodium reabsorption by in­ infiltration, augmented by proinflammatory and
jured tubular epithelial cells increases the sodium chemotactic cytokines generated by ischemic tu­
concentration in the tubular lumen. The increased bular cells. These leukocytes obstruct the micro­
intratubular sodium concentration polymerizes circulation and release cytotoxic cytokines, reac­

798 n engl j med 357;8  www.nejm.org  august 23, 2007

The New England Journal of Medicine


Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
current concepts

A Normal perfusion pressure B Decreased perfusion pressure


Arteriolar resistances

Afferent Efferent
arteriole arteriole

Increased Increased
vasodilatory angiotensin II
prostaglandins

Glomerulus

Tubule

Normal GFR Normal GFR maintained

C Decreased perfusion pressure in the presence of NSAIDs D Decreased perfusion pressure in the presence of ACEI or ARB

Decreased Increased Slightly increased Decreased


vasodilatory angiotensin II vasodilatory angiotensin II
prostaglandins prostaglandins

Low GFR Low GFR

Figure 2. Intrarenal Mechanisms for Autoregulation of the Glomerular Filtration Rate under Decreased Perfusion Pressure and Reduction
of the Glomerular Filtration Rate by Drugs.
Panel A shows normal conditions and a normal glomerular filtration rate (GFR). Panel B shows reduced perfusion pressure within the
COLOR FIGURE

autoregulatory range. Normal glomerular capillary pressure is maintained by afferent vasodilatation and efferent vasoconstriction.
Draft 3 Panel C
08/03/07
shows reduced perfusion pressure with a nonsteroidal antiinflammatory drug (NSAID). Loss of vasodilatory prostaglandins
Author increases affer-
Abuelo
ent resistance; this causes the glomerular capillary pressure to drop below normal values and the GFR to decrease.
Fig #Panel 2D shows reduced
Title GFR regulation
perfusion pressure with an angiotensin-converting–enzyme inhibitor (ACEI) or an angiotensin-receptor blocker (ARB). Loss of angiotensin II
ME
action reduces efferent resistance; this causes the glomerular capillary pressure to drop below normal values DEand the Ingelfinger
GFR to decrease.
Artist KMK
AUTHOR PLEASE NOTE:
Figure has been redrawn and type has been reset
Please check carefully

tive oxygen species, and proteolytic enzymes, 08/23/07


Fac t or s Incr e a sing R enaIssue
l date
which damage the tubular cells.14 Sus cep t ibil i t y t o Is che mi a
This tubular damage is commonly known as
acute tubular necrosis. The term is misleading, The kidneys are most vulnerable to moderate hypo­
however, because only some tubular cells are ne­ perfusion when autoregulation is impaired (Fig. 1),
crotic; most are viable (either healthy or revers­ which occurs most often when afferent arteriolar
ibly injured), and some are apoptotic (i.e., under­ resistance does not decrease appropriately or even
going an orderly programmed death). “Ischemic increases (Table 1). This phenomenon may be seen
acute kidney” and “acute tubular injury” have in elderly patients or in patients with atherosclero­
been proposed as better terms14; neither of these sis, hypertension, or chronic renal failure, in whom
newer terms is in common use. hyalinosis and myointimal hyperplasia cause

n engl j med 357;8  www.nejm.org  august 23, 2007 799


The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

Proximal convoluted Distal convoluted


Oxygen depletion
tubule tubule
ATP depletion
Metabolic changes

Tubular injury Cast obstructing lumen


Endothelial Proinflammatory
injury and chemotactic Increased
cytokines intralumenal
sodium

Leukocyte Polymerizing of
infiltration Tamm–Horsfall
protein
Peritubular
capillaries
Collecting
Release of
duct
cytokines, enzymes,
Increase in
and reactive oxygen
adhesion
species
molecules
Denuded tubular walls
Thick
Cytoskeletal disruption ascending
limb
Backleak
of fluid Loss of microvilli

Necrotic cell

Increased
vasoconstrictors Decreased
Mislocation renal function
Decreased of integrins
vasodilators
Apoptotic Mislocation of Na+/K+–ATPase
cell
Congestion
Hypoperfusion

Figure 3. Pathophysiological Mechanisms of Ischemic Acute Tubular Necrosis. COLOR FIGURE

Tubular injury is a direct consequence of metabolic pathways activated by ischemia but is potentiated by inflammation
Draftand
6 microvascular 08/07/07
compromise. The inset shows shedding of epithelial cells and denudation of the basement membrane in the proximal
Author tubule,
Abuelo with back-
Fig # 3
leak of filtrate (inset, left) and obstruction by sloughed cells in the distal tubule (inset, right).
Title Acute tubular necrosis
ME Perkins
DE Ingelfinger
Artist KMK
structural narrowing of the arterioles.7,9,15
In­ dal antiinflammatory drugs (NSAIDs)
AUTHOR PLEASE NOTE: or cyclo­
Figure has been redrawn and type has been reset
creased susceptibility to renal ischemia may also oxygenase-2 (COX-2) inhibitors, which reduce the
Please check carefully

date 08/23/07
occur in malignant hypertension because of inti­ synthesis of vasodilatory Issue
prostaglandins in the
mal thickening and fibrinoid necrosis of the small kidneys (Fig. 2C). 16,18-20 When this occurs, angio­
arteries and arterioles.16 In addition, in chronic tensin II, norepinephrine, and other vasoconstric­
kidney disease, afferent arterioles in the function­ tors released in low-perfusion states may act on
ing glomeruli become dilated, which increases the afferent arterioles unopposed, further decreas­
their filtration rate (hyperfiltration). Although an ing glomerular capillary pressure. In other situ­
increased glomerular filtration rate per nephron ations, sepsis,21,22 hypercalcemia,23,24 severe liver
compensates for the loss of nephrons, the inabil­ failure,25,26 calcineurin inhibitors,27 and radiocon­
ity to vasodilate further markedly impairs the kid­ trast agents28 can act through various vasocon­
ney’s ability to autoregulate the glomerular filtra­ strictor mediators to increase afferent arteriolar
tion rate in low-perfusion states.17 resistance. In addition, sepsis and contrast agents
Failure of afferent resistance to decrease can may have direct toxic effects on the tubules.21,28
also occur when a patient is receiving nonsteroi­ Decreased renal perfusion may also cause an ex­

800 n engl j med 357;8  www.nejm.org  august 23, 2007

The New England Journal of Medicine


Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
current concepts

Table 1. Factors Increasing Susceptibility to Renal Table 2. Causes of Low-Perfusion States.


Hypoperfusion.
Hypovolemic causes
Failure to decrease arteriolar resistance
Fluid loss to the third space
Structural changes in renal arterioles and small arteries
Tissue damage (e.g., pancreatitis)
Old age
Hypoalbuminemia (e.g., the nephrotic syndrome)
Atherosclerosis
Bowel obstruction
Chronic hypertension
Chronic kidney disease Blood loss
Malignant or accelerated hypertension Fluid loss to the outside
Reduction in vasodilatory prostaglandins Gastrointestinal causes
Nonsteroidal antiinflammatory drugs Renal causes (e.g., diuretics, adrenal insufficiency,
hypercalcemia)
Cyclooxygenase-2 inhibitors
Dermal causes (e.g., burns, sweating)
Afferent glomerular arteriolar vasoconstriction
Cardiovascular causes (congestive heart failure)
Sepsis
Myocardial causes (e.g., infarction, cardiomyopathy)
Hypercalcemia
Pericardial causes (e.g., tamponade)
Hepatorenal syndrome
Pulmonary vascular causes (e.g., embolism)
Cyclosporine or tacrolimus
Arrhythmia
Radiocontrast agents
Valvular disease
Failure to increase efferent arteriolar resistance
Distributive causes (reduced vascular resistance)
Angiotensin-converting–enzyme inhibitors
Angiotensin-receptor blockers Sepsis
Hepatorenal syndrome
Renal-artery stenosis
Overdose of drugs (e.g., barbiturates)
Vasodilators (e.g., nitrates, antihypertensive agents)
aggerated drop in the glomerular filtration rate Local renal hypoperfusion
— for example, when angiotensin II does not raise Renal-artery stenosis (atherosclerosis or fibromuscular
efferent resistance in patients who are receiving hyperplasia)
angiotensin-receptor blockers or angiotensin-con­ Malignant hypertension
verting–enzyme (ACE) inhibitors (Fig. 2D).15,16,29‑31
Narrowing of the main renal arteries due to
atherosclerosis can increase susceptibility to renal because of factors that increase renal susceptibil­
ischemia. In the case of unilateral renal-artery ity to ischemia, as described above (Table 1).
stenosis, the contralateral, normally perfused kid­ A less common mechanism for normotensive
ney maintains the glomerular filtration rate; how­ renal failure is severe hypoperfusion in the pres­
ever, renal failure may occur if renal-artery steno­ ence of increased levels of vasoconstrictive sub­
sis in a solitary kidney or in both kidneys is stances that limit the drop in blood pressure but
greater than 70% of the lumen.16 Renovascular that may occasionally increase the measured
insufficiency affecting the total renal mass may blood pressure. Hypoperfusion may be systemic;
not reduce poststenotic perfusion pressure enough for example, in acute myocardial infarction or
to impair glomerular filtration, but lowering acute pulmonary edema, low cardiac output may
blood pressure with antihypertensive therapy can be accompanied by stable or even elevated blood
worsen ischemia and reduce renal function.32 pressure mediated by sympathetic discharge.33,34
Hypercalcemia may increase afferent glomerular
arteriolar resistance, as mentioned above, and
L ow-Per f usion S tate s in
Nor mo tensi v e R ena l Fa ilur e may lead to marked hypovolemia because it may
increase renal fluid losses. Hypercalcemia may
The cause of classic ischemic acute renal failure simultaneously cause systemic vasoconstriction,
is hypovolemic, cardiogenic, or distributive shock, which may paradoxically maintain or increase
with systolic blood pressure typically dropping blood pressure.23,24
below 90 mm Hg. Normotensive renal failure Conversely, the severe hypoperfusion may be
usually involves milder degrees of these low-per­ local: in renal-artery stenosis, the poststenotic
fusion processes (Table 2), but azotemia occurs drop in blood pressure causes intrarenal ischemia

n engl j med 357;8  www.nejm.org  august 23, 2007 801


The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

but is accompanied by hypertension from hyper­ patients usually have one or more of the follow­
volemia or renin release into the circulation. ing signs or symptoms: hypothermia, confusion,
Small-vessel disease in malignant hypertension cool extremities, leukocytosis, “bandemia” (an
similarly leads to renal ischemia with increased elevated level of band forms of white cells), leuko­
renin secretion, which worsens the hypertension. penia, or unexplained lactic acidosis. Especially
in patients with relative hypotension and acute
renal failure, any of these clinical pictures should
Di agnosis of Nor mo tensi v e
Is che mic Acu te R ena l Fa ilur e lead the clinician to search for an occult infec­
tion, using physical examination, imaging, and
Common conditions in which normotensive is­ microbiologic studies.
chemic acute renal failure may develop include Alternatively, a patient in stable condition who
hypertension, chronic kidney disease, and old age, is receiving diuretics for hypertension or conges­
all of which are associated with narrowing and tive heart failure may have anorexia and stop eat­
blunted vasodilatory capacity of renal vessels. ing for some reason or may otherwise have de­
Other conditions include cirrhosis, infection, myo­ creased salt intake. Progressive negative sodium
cardial infarction, and congestive heart failure, balance results in volume depletion and a down­
as well as decreased intake of food, which can ward drift in blood pressure. Patients may not be
reduce renal perfusion. In addition, diuretics re­ aware of decreases in oral intake or may not
duce extracellular volume, which can compromise volunteer the information spontaneously. Thus,
cardiac output, and medications such as ACE in­ the clinician needs to question the patient, family,
hibitors and NSAIDs interfere with autoregula­ and caregivers about recent weight loss or changes
tion. In classic ischemic acute renal failure, when in diet.
a patient goes into shock, the physician should In addition to watching for low-perfusion
be vigilant in looking for a decrease in urinary states, clinicians should try to identify suscepti­
output and an increase in the serum creatinine bility factors for renal ischemia (Table 1). Be­
concentration. In the normotensive variant, when cause normotensive renal failure is multifactorial,
the urinary output decreases or the creatinine several low-perfusion states and susceptibility fac­
concentration increases, the clinician should look tors may be present. It is important to ask about
for the presence of a low-perfusion state that may the patient’s use of over-the-counter NSAIDs that
not have been readily apparent. On the first day might change renal perfusion and to obtain a
during which the creatinine concentration in­ measurement of the serum calcium concentra­
creases, the blood pressure is usually noted to be tion, corrected for any degree of hypoalbumin­
below its usual level. In persons who have under­ emia that is present. Sepsis, hypercalcemia, and
lying hypertension, blood pressure decreases from the hepatorenal syndrome may all cause both low-
high to normal (e.g., a drop in systolic blood pres­ perfusion states and increased afferent arteriolar
sure from 160 to 118 mm Hg). Since patients are resistance.21-26
normotensive when renal failure occurs, the de­ Susceptibility factors may sometimes result in
crease in blood pressure may be overlooked. Fre­ ischemic acute renal failure in the absence of a
quently the patient’s usual blood pressure is not low-perfusion state. In these cases, the factors
recorded alongside current blood pressures; the result in severe enough renal vasoconstriction to
usual blood pressure must be ascertained from cause a critical reduction in renal perfusion. Ex­
medical records and compared with the blood amples are radiocontrast agents given to patients
pressures when the serum creatinine concentra­ with chronic renal failure28 and cyclosporine, ta­
tion began to rise. In patients with hypovolemia, crolimus,27 NSAIDs,35 or COX-2 inhibitors36 given
blood pressure must often be measured in the up­ in high or supratherapeutic doses.
right position to confirm the degree and ortho­ Patients with malignant hypertension, renal-
static nature of the decline in blood pressure. artery stenosis in a solitary kidney, or bilateral
The cause of the decline in blood pressure may renal-artery stenosis have severe hypertension on
not be apparent. The following two scenarios are presentation but also have compromised renal
not uncommon. In the first, a patient has what perfusion. As high blood pressure is controlled,
turns out to be early sepsis but at the onset does renal function may worsen because of a critical
not have fever or any localizing symptoms. Such drop in glomerular capillary pressure.16,30-32

802 n engl j med 357;8  www.nejm.org  august 23, 2007

The New England Journal of Medicine


Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
current concepts

Once low-perfusion states and susceptibility mixed findings — for example, a ratio of blood
factors are recognized, a tentative diagnosis of urea nitrogen to creatinine of 25:1 and urinary
normotensive renal failure can be made, supported sodium excretion of 15 mmol per liter, suggesting
by laboratory findings and a response to therapy. prerenal azotemia, but with renal tubular epi­
thelial cells in the urinary sediment, a finding
L a bor at or y Findings that is consistent with acute tubular necrosis.

Low-perfusion states trigger the body’s water- and Ther a py a nd R e sp onse


sodium-conserving mechanisms. Thus, by the
time glomerular capillary pressure and glomer­ Low-perfusion states and risk factors for renal
ular filtration rate drop, the renal tubule is re­ ischemia are often treatable and thus should be
absorbing more water and sodium, which also identified and dealt with promptly. If possible,
increases passive reabsorption of urea. There is blood pressure that is on the lower end of the
experimental evidence that this increase in reab­ normal range should be increased by correction
sorption is facilitated by increased expression of of any hypovolemia and by dose reduction or dis­
urea transporters in the collecting duct, mediated continuation of antihypertensive medication and
by high plasma vasopressin concentrations.37 As other medications that may lower blood pressure
a result of these changes, the specific gravity of (e.g., narcotics). The patient should be evaluated
the urine often increases to 1.015 or higher, for occult infection, and any such infection should
while urinary sodium and urea excretion decrease be treated. If acute tubular necrosis has not yet
(urinary sodium, <20 mmol per liter; fractional occurred, effective therapy can reverse the in­
excretion of sodium, <1%; and fractional excre­ crease in the creatinine concentration within 24
tion of urea, <35%), and the ratio of blood urea to 48 hours. Improvement may even occur if only
nitrogen to creatinine rises from the usual value of one of the causes is reversed — for example,
10:1 to 20:1 or higher.3,9,10,38 These laboratory stopping treatment with NSAIDs may be benefi­
findings strongly suggest the presence of renal cial in a patient with severe cardiomyopathy. How­
hypoperfusion, even with a blood pressure in the ever, if acute tubular necrosis has occurred, sev­
normal range. eral days are usually required before improvement
If further ischemia causes acute tubular necro­ is seen, even after the underlying causes have been
sis, the injured tubules are no longer able to in­ treated. If the cause of acute renal failure is not
crease reabsorption of water, sodium, and urea. apparent or if the patient’s condition does not im­
The specific gravity of the urine becomes isos­ prove, consultation with a nephrologist may help
thenuric, similar to plasma, and urinary sodium to ensure complete assessment and appropriate
and fractional excretion of sodium and urea in­ management.40,41
crease to more than 20 mmol per liter, greater There are a number of considerations in treat­
than 1%, and greater than 35%, respectively, while ing patients with ischemic acute tubular necrosis.
the ratio of blood urea nitrogen to creatinine falls Since these persons are highly susceptible to re­
back to 10:1.3,9,10,38 The urinary sediment will current renal damage, volume depletion, and hy­
show sloughed renal tubular epithelial cells and potension, administration of NSAIDs and nephro­
debris-filled, muddy-brown, granular casts. Al­ toxic drugs, unnecessary anesthesia, surgery, and
though these elements are easily recognized by radiocontrast medium should be avoided.41 How­
the nephrologist examining the urinary sediment, ever, critical surgical, imaging, and percutaneous
clinical laboratories may fail to identify them.39 procedures, such as transluminal coronary angio­
Laboratory personnel may not distinguish these plasty in patients with acute myocardial infarc­
pigmented casts from other granular casts and tion, should not be withheld or delayed if clini­
may not examine a sufficient number of fields cally important harm might result. Prophylactic
on the slide to find them. Furthermore, the dip­ measures, such as administration of N-acetylcys­
stick often shows no red cells, white cells, or teine, hydration, and low volumes of iso-osmolol
proteinuria, so the clinical laboratory may not contrast agents, may reduce the risk of nephrop­
examine the sediment as part of the urinalysis. athy induced by contrast medium.42 Doses of
Many patients appear to be “between” prerenal renally excreted drugs need to be adjusted for
azotemia and acute tubular necrosis because of kidney failure and their plasma concentrations

n engl j med 357;8  www.nejm.org  august 23, 2007 803


The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

monitored. Fluid, electrolyte, and acid–base bal­ toms, volume overload, hyperkalemia, and meta­
ance must be maintained, and volume overload bolic acidosis that cannot be managed by conser­
and hyperkalemia avoided. In certain cases, sodi­ vative means. How severe these problems need to
um bicarbonate intravenous solutions (150 mmol be before treatment is initiated is controversial.
of sodium bicarbonate added to 1 liter of a 5% Arguments in favor of starting renal-replacement
dextrose solution in water) can be used for vol­ therapy early may be based on the desire to avoid
ume resuscitation, rather than sodium chloride, the dangerous metabolic, fluid, and electrolyte
to mitigate acidosis from uremia or diarrhea. derangements of uremia47; arguments in favor
Furosemide and low-dose dopamine have long of withholding renal-replacement therapy until
been used to treat acute renal failure, but there definite indications are present are based on the
is no evidence of their effectiveness.41,43,44 Simi­ risk of hemorrhage during vascular access, hypo­
larly, atrial natriuretic peptide and mannitol do tension and arrhythmia during dialysis, and pos­
not appear to ameliorate acute renal failure.41 sible dialysis-induced recurrent renal injury or
Adequate nutrition should be provided; mal­ delayed renal recovery.48 Existing data on the tim­
nutrition is associated with increased complica­ ing of renal-replacement therapy are inadequate
tions and death in patients with acute renal fail­ to provide clear guidance.47,49 Unfortunately, many
ure.41,45 Such patients frequently have accelerated patients with normotensive renal failure have
protein breakdown and increased caloric needs, important complicating factors, such as old age,
especially if they are critically ill or receiving sepsis, and heart disease, and progression to
renal-replacement therapy (hemodialysis). In gen­ frank shock and death is not unusual.
eral, daily intake should include 25 to 30 kcal per
kilogram of body weight; catabolic patients should C onclusions
receive up to 1.5 g of protein per kilogram daily.
Enteral nutrition with food or formula designed An acute increase in the serum creatinine con­
for renal failure is preferred over parenteral nu­ centration is usually due to renal ischemia. If the
trition, when possible, because it maintains the patient does not have frank hypotension, normo­
integrity of the gut, requires less fluid intake, tensive ischemic acute renal failure must be con­
and is less expensive.41,45,46 Advice from a dieti­ sidered. A drop in systolic blood pressure to the
tian or a nutrition consultant may be helpful. low-normal range (100 to 115 mm Hg) must not
Although intake of potassium and phosphate is be overlooked. Many cases can be treated quickly
typically restricted because of impaired renal ex­ by replacing volume, treating infection, or stop­
cretion, hypokalemia and hypophosphatemia due ping medications such as NSAIDs, diuretics, and
to cell uptake or external losses can occur, and antihypertensive agents, especially ACE inhibitors
supplements may be required. or angiotensin-receptor blockers.
Indications for intermittent or continuous No potential conflict of interest relevant to this article was
renal-replacement therapy are florid uremic symp­ reported.

References
1. American Society of Nephrology Re­ jury, mortality, length of stay, and costs in Acute renal failure. In: Brenner BM, ed.
nal Research Report. J Am Soc Nephrol hospitalized patients. J Am Soc Nephrol Brenner and Rector’s the kidney. 7th ed.
2005;16:1886-903. 2005;16:3365-70. Philadelphia: Saunders, 2004:1215-92.
2. Liano F, Pascual J. Epidemiology of 6. Dworkin LD, Brenner BM. The renal 10. Schrier RW, Wang W, Poole B, Mitra A.
acute renal failure: a prospective, multi­ circulations. In: Brenner BM, ed. Brenner Acute renal failure: definitions, diagnosis,
center, community-based study. Kidney Int and Rector’s the kidney. 7th ed. Philadel­ pathogenesis, and therapy. J Clin Invest
1996;50:811-8. phia: Saunders, 2004:307-52. 2004;114:5-14. [Erratum, J Clin Invest 2004;
3. Thadhani R, Pascual M, Bonventre JV. 7. Badr KF, Ichikawa I. Prerenal failure: 114:598.]
Acute renal failure. N Engl J Med 1996; a deleterious shift from renal compensa­ 11. Bonventre JV, Weinberg JM. Recent
334:1448-60. tion to decompensation. N Engl J Med advances in the pathogenesis of ischemic
4. Kohli HS, Bhaskaran MC, Muthuku­ 1988;319:623-9. acute renal failure. J Am Soc Nephrol 2003;
mar T, et al. Treatment-related acute renal 8. Schnermann J, Briggs JP, Weber PC. 14:2199-210.
failure in the elderly: a hospital-based Tubuloglomerular feedback, prostaglan­ 12. Wangsiripaisan A, Gengaro PE, Edel­
prospective study. Nephrol Dial Transplant dins, and angiotensin in the autoregula­ stein CL, Schrier RW. Role of polymeric
2000;15:212-7. tion of glomerular filtration rate. Kidney Tamm-Horsfall protein in cast formation:
5. Chertow GM, Burdick E, Honour M, Int 1984;25:53-64. oligosaccharide and tubular fluid ions.
Bonventre JW, Bates DW. Acute kidney in­ 9. Brady HR, Clarkson MR, Lieberthal W. Kidney Int 2001;59:932-40.

804 n engl j med 357;8  www.nejm.org  august 23, 2007

The New England Journal of Medicine


Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
current concepts

13. Lameire N, Van Biesen W, Vanholder R. Aliment Pharmacol Ther 2004;20:Suppl 3: 36. Graham MG. Acute renal failure re­
Acute renal failure. Lancet 2005;365:417- 31-41. lated to high-dose celecoxib. Ann Intern
30. 27. Klein IH, Abrahams A, van Ede T, Med 2001;135:69-70.
14. Devarajan P. Update on mechanisms Hene RJ, Koomans HA, Ligtenberg G. 37. Bankir L, Trinh-Trang-Tan MM. Urea
of ischemic acute kidney injury. J Am Soc Different effects of tacrolimus and cyclo­ and the kidney. In: Brenner BM, ed. Bren­
Nephrol 2006;17:1503-20. sporine on renal hemodynamics and blood ner and Rector’s the kidney. 6th ed. Phila­
15. Palmer BF. Renal dysfunction com­ pressure in healthy subjects. Transplanta­ delphia: W.B. Saunders, 2000:637-79.
plicating the treatment of hypertension. tion 2002;73:732-6. 38. Carvounis CP, Nisar S, Guro-Razu­
N Engl J Med 2002;347:1256-61. 28. Oudemans-van Straaten HM. Contrast man S. Significance of the fractional ex­
16. Abuelo JG. Diagnosing vascular causes nephropathy, pathophysiology and preven­ cretion of urea in the differential diagno­
of renal failure. Ann Intern Med 1995; tion. Int J Artif Organs 2004;27:1054-65. sis of acute renal failure. Kidney Int 2002;
123:601-14. [Erratum, Ann Intern Med 29. Lee H-Y, Kim C-H. Acute oliguric re­ 62:2223-9.
1995;124:78.] nal failure associated with angiotensin II 39. Tsai JJ, Yeun JY, Kumar VA, Don BR.
17. Taal MW, Luyckx VA, Brenner BM. Ad­ receptor antagonists. Am J Med 2001;111: Comparison and interpretation of urinal­
aptation to nephron loss. In: Brenner BM, 162-3. ysis performed by a nephrologist versus
ed. Brenner and Rector’s the kidney. 7th ed. 30. Johansen TL, Kjaer A. Reversible renal a hospital-based clinical laboratory. Am J
Philadelphia: Saunders, 2004:1955-97. impairment induced by treatment with the Kidney Dis 2005;46:820-9.
18. Schlondorff D. Renal complications angiotensin II receptor antagonist cande­ 40. Hilton R. Acute renal failure. BMJ
of nonsteroidal anti-inflammatory drugs. sartan in a patient with bilateral renal 2006;333:786-90.
Kidney Int 1993;44:643-53. artery stenosis. BMC Nephrol 2001;2:1. 41. Gill N, Nally JV Jr, Fatica RA. Renal
19. Perazella MA, Eras J. Are selective 31. Toto RD. Renal insufficiency due to failure secondary to acute tubular necro­
COX-2 inhibitors nephrotoxic? Am J Kid­ angiotensin-converting enzyme inhibitors. sis: epidemiology, diagnosis, and manage­
ney Dis 2000;35:937-40. Miner Electrolyte Metab 1994;20:193-200. ment. Chest 2005;128:2847-63.
20. Braden GL, O’Shea MH, Mulhern JG, 32. Textor SC, Novick AC, Tarazi RC, Kli­ 42. Pannu N, Wiebe N, Tonelli M. Prophy­
Germain MJ. Acute renal failure and hyper­ mas V, Vidt DG, Pohl M. Critical perfusion laxis strategies for contrast-induced ne­
kalaemia associated with cyclooxygenase-2 pressure for renal function in patients with phropathy. JAMA 2006;295:2765-79.
inhibitors. Nephrol Dial Transplant 2004; bilateral atherosclerotic renal vascular dis­ 43. Ho KM, Sheridan DJ. Meta-analysis of
19:1149-53. ease. Ann Intern Med 1985;102:308-14. frusemide to prevent or treat acute renal
21. Schor N. Acute renal failure and the 33. Givertz MM, Colucci WS, Braunwald E. failure. BMJ 2006;333:420-5.
sepsis syndrome. Kidney Int 2002;61:764- Clinical aspects of heart failure: pulmo­ 44. Karthik S, Lisbon A. Low-dose dopa­
76. nary edema, high-output heart failure. In: mine in the intensive care unit. Semin
22. Schrier RW, Wang W. Acute renal fail­ Zipes DP, Libby P, Bonow RO, Braunwald Dial 2006;19:465-71.
ure and sepsis. N Engl J Med 2004;351:159- E, eds. Braunwald’s heart disease: a text­ 45. Fiaccadori E, Maggiore U, Giacosa R,
69. book of cardiovascular medicine. 7th ed. et al. Enteral nutrition in patients with
23. Sowers JR, Barrett JD. Hormonal Philadelphia: Elsevier Saunders, 2005:539- acute renal failure. Kidney Int 2004;65:999-
changes associated with hypertension in 68. 1008.
neoplasia-induced hypercalcemia. Am J 34. Antman EM, Braunwald E. ST-eleva­ 46. Strejc JM. Considerations in the nutri­
Physiol Endocrinol Metab 1982;242:E330- tion myocardial infarction: pathology, tional management of patients with acute
E334. pathophysiology, and clinical features. In: renal failure. Hemodial Int 2005;9:135-
24. Mahnensmith RL. Hypercalcemia, hy­ Zipes DP, Libby P, Bonow RO, Braunwald 42.
pernatremia, and reversible renal insuf­ E, eds. Braunwald’s heart disease: a text­ 47. John S, Eckardt K-U. Renal replace­
ficiency. Am J Kidney Dis 1992;19:604- book of cardiovascular medicine. 7th ed. ment therapy in the treatment of acute
8. Philadelphia: Elsevier Saunders, 2005:1141- renal failure — intermittant and continu­
25. Cardenas A. Hepatorenal syndrome: 65. ous. Semin Dial 2006;19:455-64.
a dreaded complication of end-stage liver 35. Kulling PEJ, Backman EA, Skagius 48. Star RA. Treatment of acute renal fail­
disease. Am J Gastroenterol 2005;100: ASM, Beckman EA. Renal impairment ure. Kidney Int 1998;54:1817-31.
460-7. after acute diclofenac, naproxen, and sulin­ 49. Palevsky PM. Dialysis modality and
26. Moller S, Henriksen JH. Pathogenesis dac overdoses. J Toxicol Clin Toxicol 1995; dosing strategy in acute renal failure.
and pathophysiology of hepatorenal syn­ 33:173-7. [Erratum, J Toxicol Clin Toxicol Semin Dial 2006;19:165-70.
drome — is there scope for prevention? 1995;33:293.] Copyright © 2007 Massachusetts Medical Society.

powerpoint slides of journal figures and tables


At the Journal’s Web site, subscribers can automatically create PowerPoint slides.
In a figure or table in the full-text version of any article at www.nejm.org, click
on Get PowerPoint Slide. A PowerPoint slide containing the image, with its title
and reference citation, can then be downloaded and saved.

n engl j med 357;8  www.nejm.org  august 23, 2007 805


The New England Journal of Medicine
Downloaded from nejm.org on October 3, 2018. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.