Professional Documents
Culture Documents
Series Editors
John Knighton, MBBS, MRCP, FRCA Paul Sadler, MBChB, FRCA
Consultant Consultant
Intensive Care Medicine & Anaesthesia Critical Care Medicine & Anaesthesia
Portsmouth Hospitals NHS Trust Queen Alexandra Hospital
Portsmouth Portsmouth
UK UK
Founding Editor
John SP Lumley
Emeritus Professor of Vascular Surgery
University of London
London
UK
and
Honorary Consultant Surgeon
Great Ormond Street Hospital for Children NHS Trust (GOSH)
London
UK
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as
permitted under the Copyright, Designs and Patents Act 1988, this publication may only be repro-
duced, stored or transmitted, in any form or by any means, with the prior permission in writing
of the publishers, or in the case of reprographic reproduction in accordance with the terms of
licences issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside
those terms should be sent to the publishers.
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence
of a specific
fi statement, that such names are exempt from the relevant laws and regulations and
therefore free for general use.
Product liability: The publisher can give no guarantee for information about drug dosage and
application thereof contained in this book. In every individual case the respective user must check
its accuracy by consulting other pharmaceutical literature.
9 8 7 6 5 4 3 2 1
springer.com
Contents
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
v
vi Contents
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Contributors
vii
1
Assessment of Renal Function
Mohan Arkanath
Normal Functions of the Kidney sure are dependent on optimum plasma volume,
and most enzymes function best in narrow ranges
To be able to assess a degree of renal function or of pH or ion concentration.
dysfunction, it is important to first
fi consider the The kidneys take up the role of correcting any
normal functions of the kidney: alterations in the composition and volume of body
fluids that occur as a consequence of food intake,
fl
1. Maintenance of body composition: The metabolism, environmental factors, and exercise.
volume of fl fluid in the body, its osmolarity, elec- In healthy individuals, these corrections occur in
trolyte content, concentration, and acidity are all a matter of hours, and body fluid
fl volume and the
regulated by the kidney via variation in urine concentration of most ions return to normal set
excretion of water and ions. Electrolytes regulated points. In many disease states, however, these
by changes in urinary excretion include sodium, regulatory processes are disturbed, resulting in
potassium, chloride, calcium, magnesium, and persistent deviations in body fl fluid volume and
phosphate. composition.
2. Excretion of metabolic end products and
foreign substances: The most notable are urea and
a number of toxins and drugs. Body Fluid Composition
3. Production and secretion of enzymes and A large proportion of the human body is com-
hormones: posed of water (Table 1.1). Adipose tissue is low in
a. Renin: See Figure 1.1. water content and, hence, obese individuals have
b. Erythropoietin: A glycosylated, 165-amino a lower body water fraction than lean individuals.
acid protein produced by renal cortical Because of slightly greater fat content, women
interstitial cells that stimulates maturation generally contain less water than men.
of erythrocytes in the bone marrow.
c. 1, 25-Dihydroxyvitamin D3: The most active
form of vitamin D3, it is formed by proximal Clinical Evaluation of Renal Function
tubule cells. This steroid hormone plays an
important role in the regulation of body Glomerular filtration
fi rate (GFR) is generally con-
calcium and phosphate balance. sidered the best measure of renal function; it is the
sum of filtration
fi rates of all of the functioning
nephrons. It is defined
fi as the renal clearance of a
The Role of the Kidney in Homeostasis particular substance from plasma, and is expressed
Maintenance of body fluid fl composition and as the volume of plasma that can be completely
volume are important for many functions of the cleared of that substance in a unit of time. In the
body. For example, cardiac output and blood pres- following sections, we will compare the advantages
1
2 M. Arkanath
Renin TABLE 1.2. Other factors altering blood urea or serum creatininea
Angiotensinogen Angiotensin Salt balance Increased level Decreased level
(Plasma globulin) (Vasoconstrictor) and blood
Urea Prerenal causes (e.g., Cirrhosis
pressure
control congestive heart failure, Protein malnutrition
volume contraction) lead Water excess (e.g.,
FIGURE 1.1. Production and secretion of enzymes and hormones to increased tubular SIADH, saline
for renin. reabsorption infusion) leads to
• Gastrointestinal bleeding reduced tubular
• Catabolic state reabsorption
• Corticosteroids
• Hyperalimentation
and disadvantages of the various methods avail- • Tetracyclines
able for GFR estimation or quantification.
fi Creatinine Overproduction (e.g., Decreased muscle
rhabdomyolysis, vigorous mass
Clearance Urine mg/dL Volume mL/min ) sustained exercise,
= anabolic steroids, dietary
(mL/min
/ i ) Plasma mg/dL ) supplements such as
creatine)
Where x is the substance being cleared. Blocked tubular secretion
Features of an ideal marker for GFR (e.g., drugs such as
trimethoprim and
determination:
cimetidine)
• Appears endogenously in the plasma at a con- Assay interference (e.g.,
ketosis and drugs such as
stant rate
cephalosporins, flucytosine,
• Is freely filtered by the glomerulus methyldopa, levodopa,
• Does not undergo reabsorption or secretion by and ascorbic acid)
the renal tubule a
SIADH, syndrome of inappropriate ADH secretion.
• Is not eliminated by extrarenal routes
Urea
Creatinine
Urea was first isolated in 1773, and urea clearance
as a surrogate marker for GFR introduced in 1929. Creatinine is a metabolite of creatine and phos-
Although urea measurement is performed fre- phocreatine found in skeletal muscle. It is a
quently, it is well recognized that its many draw- small molecule that is not protein bound and,
backs make it a poor measure of renal function hence, freely filtered by the glomerulus. It does,
(see Table 1.2). For example, the rate of production however, undergo tubular secretion which is
is influenced
fl by factors such as the availability variable (see Table 1.2). Creatinine production
of nitrogenous substrates, and the rate of reab- can vary in an individual over time with
sorption in the tubules can be affected by volume muscle mass changes or acutely with massive
status. myocyte turnover. There are also age- and sex-
associated differences in serum creatinine (Sα)
concentration; it is lower in the elderly and in
women.
TABLE 1.1. Body fluid compartment volumesa The ratio between serum urea and creatinine
Example for a 60-kg patient can be useful when assessing the patient with
TBW = 60% × body weight 60% × 60 kg = 36 L
acute renal failure. Under normal circumstances,
ICW = 2/3 TBW / × 36 L = 24 L
23 the ratio between urea and creatinine is 10 : 1 but
ECW = 1/3 TBW 1/3× 36 L = 12 L this value can rise to greater than 20 : 1 when the
Plasma water = 1/4 ECW 1/4 × 12 L = 3 L
extracellular volume is contracted. A volume-con-
Blood volume = Plasma water ÷ 3 L ÷ (1 − 0.40) = 6.6 L tracted state ((prerenal) is a “sodium avid” state
(1 − hematocrit)
and promotes proximal tubular and distal nephron
a
TBW, total body water; ICW, intracellular water; ECW, extracellular water. reabsorption of urea but not creatinine. Acute
1. Assessment of Renal Function 3
determined by the amount of solute being excreted TABLE 1.3. Causes of hematuriaa
(mainly urea and electrolytes), and the concen- Glomerular disease Mesangial IgA nephropathy
trating ability of the kidneys. In disease, impair- Thin basement membrane disease
ment of concentrating ability requires increased Mesangial proliferative GN
volumes of urine to be passed for the same solute Membranoproliferative GN
Crescentic GN
output.
Systemic lupus erythematosus
Post streptococcal GN
• Oliguria is defifined as the excretion of less than Infective endocarditis
300 mL/d of urine, and is often caused by intrin- Alport’s syndrome
sic renal disease or obstructive uropathy. Vascular and Acute hypersensitivity interstitial nephritis
tubulointerstitial Tumors (renal cell carcinoma, Wilm’s tumor,
• Anuria suggests urinary tract obstruction until
disease leukemia)
proven otherwise. Polycystic kidney disease
• Polyuria is a persistent, large increase in urine Malignant hypertension
output, usually associated with nocturia. Poly- Analgesic nephropathy
uria is a result of excessive intake of water (e.g., Diabetes mellitus
Obstructive uropathy
compulsive water drinking), increased excretion
Urinary tract diseases Carcinoma (renal pelvis, ureter, bladder,
of solute (hyperglycemia and glycosuria), a urethra, prostate)
defective renal concentrating ability, or failure Calculi
of production of antidiuretic hormone (ADH). Retroperitoneal fibrosis
Tuberculosis
Cystitis
Chemical Testing Drugs (e.g., cyclophosphamide)
Trauma
Blood Benign prostatic hypertrophy
Urethritis
Hematuria may be macroscopic or microscopic Platelet defects (e.g., idiopathic or drug
(Table 1.3). Currently used dipstick tests for blood induced thrombocytopenic purpura)
are very sensitive, being positive if two or more a
GN, glomerulonephritis.
red cells are visible under the high-power fi field
(HPF) of a light microscope. A disadvantage is that
dipstick testing cannot distinguish between blood
and free hemoglobin. A positive dipstick has to be
concentration of 150 mg/L or more in urine. They
followed by microscopy of fresh urine to confirm fi
react primarily to albumin and are insensitive to
the presence of red cells and to exclude rare condi-
globulin or Bence-Jones protein. False-positive
tions such as hemoglobinuria and myoglobinuria.
results are common with iodinated contrast
In female patients, it is essential to enquire whether
agents; hence, urine testing should be repeated
the patient is menstruating.
after 24 hours. The normal rate of excretion of
Abnormal numbers of erythrocytes in the urine
protein in urine is 80 ± 24 mg/24 h in healthy indi-
may arise from anywhere from the glomerular
viduals, but protein excretion rates are somewhat
capillaries to the tip of the distal urethra. Dysmor-
higher in children, adolescents, and in pregnancy.
phic erythrocytes tend strongly to be associated
Fever, severe exercise, and the acute infusion of
with a glomerular source. Abnormal proteinuria
hyperoncotic solutions or certain pressor agents
along with dysmorphic erythrocytes is a reliable
(e.g., angiotensin II or norepinephrine) may
sign of glomerular disease. Urinary tract abnor-
transiently cause abnormal protein excretion
malities lead to microscopic or macroscopic
in normal individuals.
hematuria but the erythrocytes exhibit normal
The protein in normal and abnormal urine is
morphology.
derived from three sources:
Protein
1. Plasma proteins filtered
fi at the glomerulus and
Proteinuria is one the most common signs of renal escaping proximal tubular reabsorption.
disease (Table 1.4). Most reagent strips detect a 2. Proteins normally secreted by renal tubules.
1. Assessment of Renal Function 5
infl
flammatory reaction within the urinary tract. casts are indicative of stasis in the collecting
Most commonly this represents a urinary tract tubules and are seen in chronic renal failure.
infection but it may also be found with a sterile
sample in patients on antibiotic therapy, or with
References
kidney stones, tubulointerstitial nephritis, papil-
lary necrosis, or tuberculosis. 1. Brown SC, O’Reilly PH. Iohexol clearance for the
determination of glomerular filtration
fi rate in clinical
practice: evidence for a new gold standard. J Urol.
Red Blood Cells 1991; 146(3): 675–679.
As mentioned previously, erythrocytes can find fi 2. Cockroft DW, Gault MH. Prediction of creatinine
their way into the urine from any source between clearance from serum creatinine. Nephron. 1976; 16:
the glomerulus to the urethral meatus. The pres- 31–41.
3. Hoste EA, Damen J, Vanholder RC, et al. Assessment
ence of more than two to three red blood cells per
of renal function in recently admitted critically ill
HPF is usually pathological. Erythrocytes origi-
patients with normal serum creatinine. Nephrol Dial
nating in the renal parenchyma are dysmorphic, Transplant. 2005; 20(4): 747–753.
whereas those originating in the collecting system 4. Henry JB, Lauzon RB, Schaumann GB. Basic exami-
retain their uniform biconcave shape. nation of the urine. Clinical Diagnosis and Manage-
ment by Laboratory Methods. 19th ed. Philadelphia:
Casts Saunders. 1991.
5. Graff L. A Handbook of Routine Urinalysis. Philadel-
Based on their shape and origin, casts are appro- phia: Lippincott. 1983.
priately named. Hyaline casts are found in con-
centrated urine, during febrile illnesses, after
strenuous exercise, and with diuretic therapy. Suggested Reading
They are not indicative of renal disease. Red cell
Davison AM, Cameron S, Grunfeld J-P, et al. Oxford
casts indicate acute glomerulonephritis. White cell
Textbook of Clinical Nephrology. 3rd ed. Oxford:
casts indicate infection or infl
flammation, and are Oxford University Press. 2005.
seen in pyelonephritis and interstitial nephritis. Greenberg A, Cheung AK, Coffman TM, et al. Primer on
Renal tubular casts are found in cases of ATN and Kidney Diseases. 3rd Ed. London: Academic Press.
interstitial nephritis. Coarse granular casts are 2001.
nonspecificfi and represent the degeneration of a Guyton AC. Textbook of Physiology. 8th Ed. New York:
cast with a cellular element. Finally, broad waxy Saunders. 1991.
2
Imaging of Acute Renal Failure—A
Problem-Solving Approach for Intensive
Care Unit Physicians
Tom Sutherland
Acute renal failure (ARF) is a common problem in (MRI) scans analyze renal structure and renal
hospitalized patients. It has a variety of causes, artery calcifification, and dynamic gadolinium-
traditionally divided into prerenal, renal, and enhanced MRI renal studies allow functional
postrenal. A further classification
fi can be made assessment. Other functional studies, such as mer-
into medical and surgical causes, with the later captoacetyltriglycine (MAG3) and diethylene tri-
defined
fi as patients who will benefifit from mechan- amine pentaacetic acid (DTPA) show reduced renal
ical intervention. uptake and delayed excretion of tracer.
Acute tubular necrosis (ATN) is the most A further role of imaging is to determine the
common cause of ARF (approximately 45%) and number of present and functioning kidneys. For
postrenal obstruction accounts for roughly 10% of ARF to occur in previously normal kidneys, the
presentations (1). A variety of imaging modalities underlying cause must be a bilateral process, or
may be used to help diagnose the cause, or, if this else a single functioning kidney must be compro-
is not possible, to differentiate medical from sur- mised (Figure 2.1).
gical causes. ARF in renal transplants will not be
addressed.
Acute Tubular Necrosis
Acute or Chronic Renal Failure? Ultrasound will usually show enlarged kidneys
with a smooth contour caused by interstitial
This is best answered by clinical assessment rather edema, no hydronephrosis, and renal arterial and
than with imaging. Imaging findingsfi in chronic venous flow. The examination of choice in sus-
renal failure are nonspecific fi and essentially char- pected ATN is a MAG3 nuclear medicine study.
acterized by small kidneys. X-rays can show the Scintigraphic examinations in ATN using Tc-
renal outline, calcifi fication, renal bone disease and 99m-MAG3 demonstrate relatively well-preserved
effects of hyperparathyroidism. Ultrasound is an on-time renal perfusion, and delayed tracer
excellent modality for structural imaging because uptake, often with a continuing activity accumula-
it is able to detect reduced renal parenchyma, scar- tion curve. If the activity curve does have a
ring (usually secondary to previous refluxfl nephro- maximum, the time to maximum is delayed (2).
pathy), calcification,
fi and polycystic kidneys. The Excretion of tracer into the collecting system is
echogenicity of the cortex can be assessed with a delayed and reduced, but there is no obstruction
hyperechoic cortex (normal cortex is hypoechoic to drainage of the collecting systems. If excre-
to liver), present in most causes of chronic renal tion and drainage occur, the time from maximal
failure; adult polycystic kidney disease being the activity to half-maximal activity (or another
notable exception. Noncontrast computed tomo- quantitative measure of tracer excretion) is pro-
graphic (CT) and magnetic resonance imaging longed. What underlies this scintigraphic pattern
7
8 T. Sutherland
A B
FIGURE 2.1. A, ultrasound of a normal kidney. Smooth cortex, hypoechoic to liver. B, chronic renal failure with a small irregular kidney
with hyperechoic cortex compared with liver (anechoic area around the liver is peritoneal dialysis fluid).
FIGURE 2.2. MAG3 study showing progressive accumulation of peak as the kidneys are perfused and then activity declines as
tracer in the renal cortex indicating normal perfusion but no excre- tracer is excreted and passes into the bladder. Lt, left; Rt, right;
tion, consistent with ATN. A normal activity curve should initially Bkg, background.
2. Imaging of Acute Renal Failure 9
Ureteric Obstruction
For obstruction to produce ARF it must be bilat-
eral or affecting a single functioning kidney. This FIGURE 2.3. Ultrasound of a hydronephrotic kidney. Size is normal
is the classical cause of surgical ARF. with dilated renal pelvis and calyces (the anechoic central part).
10 T. Sutherland
perirenal stranding. The administration of con- the degree of obstruction. Once again, it must be
trast is contraindicated in ARF because of its stressed that the only way mechanical obstruction
potential nephrotoxicity. Although not as helpful can produce ARF in previously normal kidneys is
as a contrast-enhanced CT scanning, a non- by affecting a solitary functioning kidney, or by
contrast study can usually detect many extrinsic blocking both kidneys simultaneously.
compressing masses, such as retroperitoneal
tumors, or cervical or colon carcinomas, that
may produce bilateral obstruction. Complications Renal Artery Dissection or Occlusion
related to trauma, such as urinoma or renal pedicle
avulsions, are also visible even without contrast Each kidney is normally supplied by a single renal
(Figure 2.4). artery that arises from the aorta before dividing
Scintigraphic imaging with either Tc-99m- into an average of five segmental end-arterial
MAG3 or Tc-99m-DTPA can detect ureteric branches. Arterial dissection and occlusion is
obstruction by showing a dilated collecting system another surgical cause of ARF. Renal infarction
with delayed drainage of tracer. The collecting may be caused by blunt trauma with avulsion of
system is outlined down to the level of the obstruc- the renal pedicle or penetrating trauma transect-
tion with no or limited tracer draining beyond ing the renal artery. Renal artery dissection may
this. The negative predictive value of nuclear med- be iatrogenic, occurring during endovascular
icine scanning is extremely high, with a normal intervention. Bilateral dissection or exclusion
study virtually excluding obstruction. Of course, from the circulation can occur secondary to an
sufficient
fi tracer must be present in the collecting aortic dissection. Suspicion of a dissection or
system to reach the obstruction point. Reduced occlusion is virtually the only differential diagno-
urine production associated with renal failure sis that warrants the administration of intra-
may deliver so little tracer into the collecting venous CT contrast in the setting of ARF in which
system that obstruction cannot be excluded. The MRI scanning is unavailable. An arterial phase CT
lower the patients glomerular fi filtration rate (GFR), scan highlights the renal artery anatomy, demon-
the higher the probability of having an indetermi- strating the intimal fl
flap of dissection and can also
nate study (5). Further, an enlarged but non- detect signs of renal infarction and lacerations (6)
obstructed collecting system may mimic delayed (Figure 2.5).
drainage. In this latter situation, the specifificity of Magnetic resonance angiography (MRA) has
the examination is increased by administering excellent sensitivity and specifi ficity for detecting
20 mg (or sometimes 40 mg) of frusemide IV, and dissection and occlusion of proximal renal arter-
scanning for a further 20 minutes. A baggy non- ies. Segmental arteries are less well visualized,
obstructed system will promptly wash out, and the however, unilateral or isolated segmental pathol-
delay in washout after frusemide correlates with ogy will not produce ARF. MR contrast is much
Arrow heads are fatty stranding. Right ureteric calculus Inferior pole left kidney
FIGURE 2.5. Contrast enhanced CT scan Contrast in the aorta left kidney
(magnified view of the left kidney) with
no renal parenchymal enhancement or
renal vascular enhancement secondary
to an occlusive renal artery embolus.
less nephrotoxic than CT contrast. MRI scanning of blood flowing around the thrombus is shown
is logistically difficult
fi for most intensive care unit on color Doppler imaging (7). The kidney itself
(ICU) patients, and has a considerably longer scan may have reduced corticomedullary differentia-
time and, therefore, CT scanning remains a more tion, be enlarged, and contain focal areas of
practical modality. increased echogenicity related to edema and hem-
MAG3 examinations will reveal an absence of orrhage. Contrast-enhanced CT scanning has a
perfusion (segmental or unilateral or bilateral, as similar appearance in the acute setting, with a
the case may be) but not the cause. dilated renal vein with hypodense thrombus
within. Secondary findings,
fi such as perirenal
stranding, dilated gonadal vein, prolonged nephro-
Renal Vein Thrombosis gram, and complications such as retroperitoneal
hemorrhage may also be present. As the chronicity
Bilateral renal vein thrombosis (RVT) will present increases, the thrombus contracts and multiple
with ARF, but is very rare. If discovered, it should collateral vessels develop.
trigger a search for a mass compressing both renal Renal veins are well demonstrated by MRI scan-
veins or the suprarenal inferior vena cava (IVC) ning, with the thrombus appearing hyperintense
and a coagulopathy screen. Unilateral RVT is more on both T1- and T2-weighted images. The kidney
frequent in contrast, but will not produce ARF if appears swollen with loss of corticomedullary dif-
both kidneys are initially normal. ferentiation on T1-weighted images, with both
Usually, a single renal vein exists on each side, cortex and medulla having low signal secondary
but up to 30% of patients will have multiple renal to prolonged T1 and T2 relaxation times.
veins. RVT is most common on the left because Venography is the gold standard and can accu-
this vein is three times longer than the right vein rately determine the presence of thrombus and its
and is more liable to extrinsic compression as it extent. A catheter inserted in the common femoral
passes between the superior mesenteric artery vein allows selection of the renal vein as it enters
and the aorta. RVT may be caused by hypercoagu- the IVC and, with contrast injection, a fi
filling defect
lable states, dehydration (particularly in infants), or complete occlusion can be demonstrated.
sepsis, or trauma and occurs in up to 40% of Therapeutic measures, such as the instillation
patients with nephrotic syndrome. of localized thrombolytics and the placement of
Ultrasound can show a distended renal vein an IVC filter if the thrombus is extensive, can be
with echogenic thrombus within that vein that performed during the diagnostic procedure.
may extend into the IVC. Absent fl flow or a thin cuff Complications include damage to vessels and
12
Surgical
Medical
Medical Surgical
REGION
ANATOMICAL
US US
dissection/occlusion
CLINICAL
MRI or
Contrast enhanced Non contrast CT for
CT* obstructing mass or
MAG3 Non contrast CT for stone
? obstructing mass
TEST TO ANSWER
CLINICAL QUESTION
Nonsteroidal Anti-Inflammatory
Risk Factors for Nephrotoxicity Drugs
Many patients may be taking a potentially nephro- NSAIDs are widely used and overall renal com-
toxic drug. For example, 55% of patients admitted plications are uncommon, however, their use
to a general medical ward were taking one or in the community has been shown to increase
more of either an angiotensin-converting enzyme the risk of hospitalization with AKI by up to
inhibitor (ACEI) or angiotensin receptor antago- four-fold (4, 5). Renal effects are predominately
nist (ARB), a diuretic, or a nonsteroidal anti- caused by their effect on the production of renal
infl
flammatory drug (NSAID) (2). More than 25% of prostaglandins.
patients were taking two or three of these drugs.
Not all patients will develop renal failure; whether
a drug is nephrotoxic or not depends on patient- Effects on Renal Prostaglandins
and drug-related factors (Tables 3.1 and 3.2). Membrane-bound phospholipids are converted to
arachidonic acid by phospholipase A. Arachidonic
acid then enters the cyclooxygenase pathway
Methods of Prevention under the action of cyclooxygenase enzyme (COX)
forming prostaglandins, or enters the lipoxygen-
The following general principles should be
ase pathway, forming leukotrienes.
adhered to:
In health, renal prostaglandins have a number
1. Appropriate dose alteration in patients depending of effects: control of renin release, regulation of
on renal function, age, and ideal body weight. renal blood flow
fl (RBF) through vascular tone, and
2. Awareness of drug interactions. control of salt and water transport in the renal
3. Close attention to flfluid balance and hemody- tubules. They have both vasodilatory (prostaglan-
namic status. din [PG]-E2, PGI2) and vasoconstrictor (throm-
14
3. Drug-Induced Renal Injury 15
TABLE 3.1. Mechanism of drug-induced AKI are crucial to maintain RBF and GFR. Patients
Mechanism Drug examples who fall into this high-risk category are those with
parenchymal renal disease or renal impairment,
Altered intraglomerular Calcineurin inhibitors
hemodynamics: drugs that Vasopressors hypovolemic patients, those with congestive
interfere with the normal Amphotericin cardiac failure or liver disease (because of activa-
regulatory alterations in Contrast agents tion of the renin-angiotensin system), and the
preglomerular and postglomerular NSAIDs, ACEI/ARBs elderly. There are two isomers of the COX enzyme
arteriolar resistance can
(6). “Traditional” NSAIDs are nonselective COX
compromise renal blood flow and
glomerular filtration rate, especially inhibitors, whereas newer “coxibs” selectively
in times of hemodynamic instabilit target the production of proinfl flammatory prosta-
Drug-induced glomerulopathy: this Gold, D-penicillamine glandins by COX-2. They have fewer gastric and
usually presents with nephrotic NSAIDs platelet side effects, but it is now clear that they
syndrome or proteinuria. Renal Mesalazine
have the same effects on renal prostaglandins (7);
function need not be impaired
Drug-induced thrombotic Cyclosporin A, OKT3, therefore, caution should be still be taken in high-
microangiopathy: rare tacrolimus risk individuals.
Clopidogrel, ticlopidine NSAID-induced salt and water retention occurs
Cocaine and the effect of loop diuretics is also diminished.
Quinine
This may lead to pedal edema or hypertension, but
Direct tubular cell toxicity Aminoglycosides
Amphotericin can precipitate overt cardiac failure in at-risk
Calcineurin inhibitors patients.
Cisplatin
Methotrexate
Cocaine Other Renal Effects
Contrast media
Palmidronate 1. Acute tubulointerstitial nephritis (TIN) is
Tubular damaged caused by osmotic Mannitol less common, occurring after 3 to 5 days, or
nephrosis: tubule cells take up Dextrans even after years, of drug use. It occurs in an
nonmetabolizable molecules, Intravenous
creating an osmotic gradient. immunoglobulin
Water rapidly accumulates in Hydroxyethyl starch TABLE 3.2. Risk factors for development of drug-induced AKI
the cell, causing swelling and
vacuolation, thereby disrupting cell Patient-related factors Drug-related factors
integrity. Cellular debris causes Increased age Inherent nephrotoxic potential
tubular obstruction Preexisting chronic kidney Dose (important for drugs
Tubular damage caused by cast Statins disease inducing crystal deposition,
deposition: rhabdomyolysis Diabetes mellitus causing tubular damage,
Interstitial nephritis: drugs can cause β-lactams, quinolones, Intravascular volume depletion: and drugs altering renal
an acute allergic interstitial rifampicin, macrolides, absolute (e.g., dehydration) hemodynamics)
nephritis or chronic interstitial sulfonamides or effective (e.g., massive Prolonged duration of
damage NSAIDs ascites, nephrotic syndrome) treatment
Thiazide and loop diuretics Peripheral vascular disease Frequency of administration
Phenytoin (this increasesthe risk of Time of administration
Cimetidine, ranitidine renovascular disease)
Allopurinol Sepsis Rate of administration
Antivirals Concomitant use of diuretics (important fordrugs causing
Cocaine and other nephrotoxic drugs crystal nephropathy)
Metabolic disturbances: sodium Route of administration (e.g.,
Note: This list is not exhaustive.
depletion, hypoalbuminemia, intravenous versus oral
Source: Schetz et al., 2005; and Perazella, 2003.
acid-base disturbances (which contrast)
may exacerbate intrarenal Combination of drugs causing
crystal deposition), multiple nephrotoxic synergy (e.g.,
boxane A2) effects. In certain situations associated myeloma cephalosporins and
with high levels of circulating vasoconstrictors aminoglycosides,
aminoglycosides
vancomycin and
vancomycin,
(such as angiotensin II [ATII], endothelin, cate-
aminoglycosides)
cholamines), vasodilatory renal prostaglandins
16 S. Blakeley
idiosyncratic, non-dose-dependent manner. The ability. There is then a slow rise in serum creati-
classic features of fever, rash, arthralgia, and nine over several days.
eosinophilia are often not present. The diagnosis The risks of toxicity are increased with high
is made on renal biopsy. Treatment is to stop the initial peak serum levels, prolonged treatment,
drug and support the patient. Corticosteroids are hypovolemia, increasing age, liver dysfunction,
sometimes administered, but the evidence is hypoalbuminemia, and in combination with other
lacking. drugs (e.g., diuretics, NSAIDs, ACEIs, cephalospo-
2. Renal papillary necrosis. rins) (9).
3. Nephrotic syndrome. Once daily dosing is associated with less neph-
4. Hyperkalemia caused by hyporeninemic rotoxicity than dosing multiple times daily, but
hypoaldosteronism generally occurs in patients may not be suitable for all patient groups (e.g.,
with CRF, diabetes mellitus, and type IV renal controversial in the treatment of bacterial endo-
tubular acidosis. carditis). The risk of nephrotoxicity is reduced
when the once daily dose is given during periods
of activity (daytime) or when taking food, possi-
ACEIs and ARBs bly related to changes in urinary pH. Levels should
still be monitored carefully according to local
ATII is a potent vasoconstrictor causing constric- practice.
tion of the efferent arteriole, increasing trans-
glomerular pressure, and, thus, GFR. It also causes
systemic vasoconstriction, increasing systemic
blood pressure and improving renal perfusion. In Contrast Nephropathy
certain situations, GFR is very dependent on this
postglomerular constriction and, consequently, Incidence and Outcome
blocking this mechanism by the use of ACEI/ARBs
Contrast nephropathy (CN) is the third leading
may lead to a marked reduction in GFR (3).
cause of AKI in hospitalized patients (10), account-
If, after starting an ACEI/ARB, there is a small,
ing for 12% of cases of AKI. It is defined
fi as an
but nonprogressive rise in the serum creatinine,
acute decline in renal function, with a rise in
this is generally related to alterations in intrarenal
serum creatinine of greater than 25% from base-
hemodynamics rather than structural injury (8).
line (or an absolute rise of 44 μmol/L), occurring
If the initial serum creatinine rises by more than
24 to 48 hours after contrast. Creatinine usually
30% or if there is a progressive increase in creati-
peaks at 5 days and returns to baseline by 10 days.
nine after starting the drug, the drug should be
Incidence is varied depending on definitions
fi used
stopped and a reason sought. This situation may
and patient populations studied. Overall incidence
arise in patients with bilateral renal artery steno-
is less than 3% (11, 12), but rises in the setting
sis (or stenosis of a single functioning kidney), but
of increasing number of risk factors (13), up to
is also seen when there is a reduction in the abso-
50% with the combination of diabetes and renal
lute or effective circulating volume, in the pres-
failure (14).
ence of sepsis, and with the concomitant use of
Risk factors for development of CN:
other drugs (e.g., NSAIDs). These factors should
be addressed before restarting the drug. • Preexisting renal impairment
• Diabetes mellitus
• Absolute (e.g., dehydration) or effective volume
Aminoglycosides depletion (e.g., congestive cardiac failure,
nephrotic syndrome, liver disease)
Aminoglycoside toxicity is caused by partial • Left ventricular ejection fraction less than 40%
reabsorption of the drug by the proximal renal • Concurrent use of nephrotoxic drugs (e.g.,
tubular cells, and the first
fi indication of renal NSAIDs, aminoglycosides)
involvement is the development of polyuria • High volume, high osmolar, ionic contrast given
because of a defect in the urinary concentrating intravenously
3. Drug-Induced Renal Injury 17
• Pre procedure shock (e.g., hypotension, intra- reduce the potential for ischemic injury by inter-
aortic balloon pump) fering with active transport and decreasing the
• Increasing age oxygen demands of tubules, however, evidence is
lacking. There is no evidence that mannitol, atrial
Renal failure is normally nonoliguric and
natriuretic peptide, dopamine, or fenoldopam
resolves with supportive care. Up to 12% of
(a dopamine 1 agonist) are better than standard
patients need renal replacement therapy (11).
hydration. Calcium channel antagonists gained
The development of CN increases mortality. It
some interest but were not supported by large
is unclear whether this is because of renal failure
trials. Despite its properties as an adenosine
itself, or is simply a marker of increased disease
antagonist, there is no evidence that theophylline
severity and underlying comorbidity. In one study,
is more effective, and work continues on the anti-
the development of AKI increased mortality from
oxidant, ascorbic acid.
7 to 34% (11).
6. Barkin RL, Buvanendran A. Focus on the COX-1 and ing coronary angiography. Am J Med. 1990; 89(5):
COX-2 agents: Renal effects of nonsteroidal and 615–620.
anti-infl
flammatory drugs—NSAIDs. Am J Thera- 15. Merten GJ, Burgess WP, Gray LV, et al. Prevention of
peutics. 2004; 11: 124–129. contrast induced nephropathy with sodium bicar-
7. Gambaro G, Perazella MA. Adverse renal effects of bonate: a randomised controlled trial. JAMA. 2004;
anti inflflammatory agents: Evaluation of selective 291: 2328–2334.
and non selective cyclooxygenase inhibitors. J Int 16. Pannu N, Manns B, Lee H, Tonelli M. Systematic
Med. 2003; 253: 643–652. review of the impact of N-acetylcysteine on
8. Palmer BF. Renal dysfunction complicating treat- contrast nephropathy. Kidney Int. 2004; 65: 1366–
ment of hypertension. N Engl J Med. 2002; 347: 1374.
1256–1261. 17. Aspelin P, Aubry P, Fransson SG, Strasser R, Willen-
9. Beauchamp D, Labrecque G. Aminoglycoside neph- brock R, Berg KJ. Nephrotoxicity in high-risk
rotoxicity: Do time and frequency of administra- patients study of iso-osmolar and low-osmolar
tion matter? Curr Opin Crit Care. 2001; 7: 401–408. non-ionic contrast media study investigators. Neph-
10. Nash K, Hafeez A, Hou S. Hospital-acquired renal rotoxic effects in high-risk patients undergoing
insuffi
ficiency. Am J Kidney Dis. 2002; 39(5): 930–936. angiography. N Engl J Med. 2003; 348(6): 491–
11. Levy EM, Viscoli CM, Horwitz RI. The effect of acute 499.
renal failure on mortality. A cohort analysis. JAMA. 18. Maeder M, Klein M, Fehr T, Rickli H. Contrast
1996; 275(19): 1489–1494. nephropathy: Review focusing on prevention. J Am
12. Rudnick MR, Berns JS, Cohen RM, Goldfarb S. Coll Cardiol. 2004; 44: 1763–1771.
Contrast-media associated nephrotoxicity. Semin 19. Pannu N, Wiebe N, Tonelli M. Prophylaxis strategies
Nephrol. 1997; 17: 15–26. for contrast nephropathy. JAMA. 2006; 295: 2765–
13. Rich MW, Crecelius CA. Incidence, risk factors, and 2779.
clinical course of acute renal insuffi ficiency after 20. Cruz DN, Perazella MA, Bellomo R, Corradi V, de
cardiac catheterization in patients 70 years of age or Cal M, Kuang D, Ocampo C, Nalesso F, Ronco C.
older. A prospective study. Arch Intern Med. 1990; Extracorporeal blood purifi fication therapies for pre-
150(6): 1237–1242. vention of radiocontrast-induced nephropathy: a
14. Manske CL, Sprafka JM, Strony JT, Wang Y. Contrast systematic review. Am J Kidney Dis. 2006; 48(3):
nephropathy in azotemic diabetic patients undergo- 361–367.
4
Acute Kidney Injury
Sara Blakeley
Patients may be admitted to the intensive care unit three levels of renal dysfunction and two renal
(ICU) with acute kidney injury (AKI) or it may outcomes. The levels of renal dysfunction can be
develop during their stay. This chapter gives an over- defined
fi by changes in serum creatinine, GFR, or
view of the definition
fi and epidemiology of AKI, along urine output.
with clinical features and initial investigations.
Risk of Renal Dysfunction
• Serum creatinine increased 1.5 fold or
Definition of AKI • GFR decreased by more than 25% or
• Less than 0.5 mL/kg/h of urine for 6 hours
AKI is an abrupt (<7 d) and sustained decrease in Injury to the Kidney
kidney function (1). It is accompanied by changes • Serum creatinine doubled or
in blood biochemistry (e.g., a rise in serum creati- • GFR decreased greater than 50% or
nine), in urine output, or both. There is a spec- • Less than 0.5 mL/kg/h of urine for 12 hours
trum ranging from a mild transient rise in serum Failure of Kidney Function
creatinine, to overt renal failure needing renal • Serum creatinine increased 3 fold or
replacement therapy (RRT); hence, the term acute • An acute rise in creatinine of greater than
kidney injury (AKI) is more precise than the term 44 μmol/L so that new creatinine is greater
“acute renal failure”. than 350 μmol/L or
Multiple definitions
fi of ARF exist, and the reader • GFR decreased more than 75% or
is guided to a series of excellent reviews that high- • Less than 0.3 mL/kg/h of urine for 24 hours or
light this problem (1–5). A rise in serum creatinine anuria for 12 hours
is often used as a marker of renal dysfunction, but • Note: This takes into consideration acute-on-
it is affected by extrarenal factors, such as age, sex, chronic renal failure
race, and muscle bulk. It may lag behind changes Loss of Kidney Function
in glomerular filtration
fi rate (GFR), either in • Complete loss of kidney function for longer
decline or during recovery, and, therefore, does than 4 weeks
not always give a true reflection
fl of the GFR. Urine End-Stage Renal Disease
output can be used to definefi renal failure, but this • The need for dialysis for longer than 3 months
can be confounded by the use of diuretics, and
not all cases of renal failure are associated with
oliguria. Incidence and Outcome
Efforts have been made to develop a universal
and practical way of defining
fi AKI via either serum AKI develops in 5 to 7% of hospitalized patients
creatinine or urine output. One such recent pro- (6, 7). Six to 25% of patients on the ICU develop
posal is the RIFLE (5) system; an acronym for AKI (2, 8); overall, 4% of admissions require RRT.
19
20 S. Blakeley
This may underestimate the scale of the problem, nall failure) but intrarenal vasoconstriction, which
however, because when all degrees of kidney could progress to tubular damage (intrinsicc renal
dysfunction are considered using the RIFLE failure). Glomerular microthrombi are associated
criteria, 20% (9) of hospital patients and 67% of with disseminated intravascular coagulation, and
ICU patients developed some form of kidney can cause intrinsic AKI (20).
injury (10).
The incidence and progression of AKI varies
depending on the patient group studied. For Prerenal Failure
example, up to 20% of cardiac surgery patients
will develop some evidence of renal injury (11), Prerenal failure (Figure 4.1) accounts for 15 to
but only 1% will need RRT (12). 20% of cases of AKI on the ICU (13, 15). For the
AKI on the ICU is associated with a hospital kidneys to be perfused and, therefore, function,
mortality of 13 to 80% (2, 8, 10–17) and 57 to 80% adequate pressure, fl
flow, volume, and patent vessels
if RRT is needed. Renal failure rarely occurs on are needed.
its own, with up to 80% of patients with renal The kidney autoregulates to maintain a con-
failure on the ICU having another organ system stant renal blood flow (RBF) through a mean arte-
failure (8). Various factors have been associated rial pressure range of 65 to 180 mmHg. “Prerenal
with a worse outcome; including comorbidity, failure” is an appropriate, albeit exaggerated,
increased severity of illness, presence of sepsis, physiological response to renal hypoperfusion.
need for mechanical ventilation, oliguria, hospi- Stimulation of the renin-angiotensin-aldosterone
talization before ICU, and delayed occurrence of system attempts to retain salt and water and,
AKI (13–15). therefore, maintain RBF. Because renal tissue is
The development of AKI dramatically increases still preserved, once renal perfusion is restored,
mortality across all patient populations studied function should improve. A profound or prolonged
(8–10, 12, 14). Worsening levels of renal dysfunc- reduction in perfusion can, however, lead to isch-
tion, as described by the RIFLE criteria, correlate emic acute tubular necrosis (ATN) (intrinsic renal
well with increasing hospital mortality, with up to failure).
a 10-fold risk of death with “failure.” AKI carries Conditions leading to reduced renal perfusion
an independent risk of death, but it is unclear and, therefore, causing prerenal failure are:
whether this is related to the systemic effects of • Hypotension (relative or absolute) secondary to
renal failure itself, the effects of its treatment, or vasodilation (e.g., certain drugs, loss of vascular
is simply a refl
flection of the severity of the under- tone, and sepsis)
lying condition. • Compromised cardiac function
After AKI needing RRT, 10 to 32% of patients • Intravascular volume depletion (absolute or
are discharged from hospital still needing RRT effective)
(2, 16, 18). • Increased intra-abdominal pressure (abdominal
compartment syndrome)
Causes of AKI
Intrinsic Renal Failure
Causes of AKI can be divided into prerenal, intrin-
sic, and obstructive causes. One disease may be The commonest cause of intrinsic renal failure on
associated with different causes of ARF, for the ICU is ischemic ATN developing after pro-
example, sepsis is a common cause of renal dys- found or prolonged prerenal failure (Figure 4.2).
function on the ICU, accounting for up to 50% of Up to 80% of cases of AKI on the ICU are attrib-
cases of AKI. AKI occurs in 23% of patients with uted to ATN (2, 13, 15). Although ATN is a histo-
severe sepsis, and in 51% of patients with septic logical diagnosis, its development is suggested
shock when blood cultures are positive (19). Sepsis by the persistence of renal failure following the
is characterized by systemic vasodilation (prere-
( restoration of adequate renal perfusion. The
4. Acute Kidney Injury 21
Pre re
Pre
Pr rena
nall fa
na faililur
fail ure
ur e
Hypottens
Hypo tensiion
i on Hypovo
Hypo volla
laemiia
laem ia Low card
Low
Lo rdi
dia
iac out
iac outp
tputt
(Reduced intravascular volume)
Tota
Tot
Total lo
tal loss
loss Volu
Vol
Volume red
lume edi
dis
istr
ist
tribut
tributitio
ion
ion
GI los
los
osss Redu
Red
Re duce
duce
cedd ef
eff
ffe
fect
cti
tiv
ive ci
ive circ
ircullat
ati
tin
ing
ing
(Vomiti
(Vom
(V iting,
g ddiiarr
iarrhoea
iarrhhoea, surgic
i all ffiistu
istul
t lae))
vollu
vo lume
lume
(Ascit
(Asc
(A ites, oed
oedema 3rdd spacing
dema, “3 cing”
i ”, conge
congestiv
stive
ti e cardia
rdiac
di c
failure)
Haem
Haem
Ha emor
orrh
rhag
hag
ge
(Visibl
(Vis
(Vi ible and
d occult)
lt)
Altte
Al tere
tere
redd va
vasc
scul
ular
lar cap
cap
pac
acit
itan
itance
ance
( psis:
(Sep
(S i shu
h ntin
nting
ti g
g, vasod
asodil
dilattatio
ti n.
n Hepat
p torenall
synd
y drome HRS
HRS))
Rena
Rena
Re nall lo
loss
loss
(Diureti
(Di retics
tics
cs, poly
l uria
uria)
i )
Skiin
Sk in los
los
osss
(
(Exc
(E essive
essi
ive sweat
weati
ting
ing
ing,
g, bur
burns))
Intr
Int
Intrin
tri
insi
sic
ic re
rena
nall fa
na fail
fail
ilur
lur
ure
e
Glom
Glomer
omerul
ular
lar Tubular Inte
Int
In ters
ters
rsti
titi
ti tiall Vascul
Vasc
Va ular
lar
(Glo
(Glome
lo meru
merulo
rulone
lo neph
ne phri
ph riti
ri tis)
tis) (Aut
(A utoi
ut oimm
oi mmun
mm une
un e, tox
tox
oxic
ic, in
ic infe
fect
fe ctio
ct ious
io us))
us
Intrinsic toxins
( ha
(Rhabd
bdom
bdomyo
om y lyysi
yo siss, masassi
sive
sive hae
hae
aemomoly
mo lyysi
siss
tumo
tumo
tu mour
ur llys
ysiis
ys is myel
is, elom
loma))
Extrinsic toxins
(Rad
(Radio
io
o ccon
ontr
on tras
trastt,
as t, dru
dru
rugs
gs, an
gs antititibi
biot
bi otic
ot
tic
ics)
s))
metabolites also falls. Renal failure may be exac- be interpreted in light of the clinical setting, but
erbated by drug accumulation, or other side effects can act as another tool in the assessment of intra-
can develop, such as morphine metabolites leading vascular volume status.
to respiratory depression.
Radiological Investigations
A chest x-ray will help assess volume status, but
Investigation of the Cause of Renal patchy infifiltrates may also represent pulmonary
Failure (Table 4.1) hemorrhage, as seen in certain forms of vasculitis.
A renal ultrasound scan should be performed;
Laboratory Tests (Table 4.2) the timing will depend on the likely cause of renal
failure and the patient’s clinical state. Further
Urinalysis
imaging should be guided by the clinical scenario.
A standard dipstick for blood and protein should
be preformed and a fresh sample spun for casts:
hyaline casts (nonspecific
fi markers of renal injury), Conclusion
brown/cellular casts (ATN), and red cell casts
(acute glomerulonephritis). An estimation of AKI is a significant
fi condition affecting critically
protein excretion may be needed, either a 24-hour ill patients on the ICU. It is a systemic process
urine collection or a spot urine protein-creatinine affecting all organs, and has a major impact on
ratio, depending on local resources. patient morbidity and mortality. It is, therefore,
Urine osmolality and urinary electrolytes can important to be able to promptly recognize its
be used to help distinguish prerenal failure from development and institute the appropriate inves-
intrinsic kidney disease (Table 4.3). They should tigations to guide treatment.
7. Hou SH, Bushinsky DA, Wish JB et al. Hospital- therapy on outcome in critically ill patients. Crit
acquired renal insuffi ficiency: a prospective study. Care Med. 2002; 30: 2051–2058.
Am J Med. 1983; 74(2): 243–248. 15. Brivet FG, Kleinknecht DJ, Loirat P, Landais PJ.
8. Uchino S, Kellum JA, Bellomo R et al. Beginning and Acute renal failure in intensive care units-causes,
Ending Supportive Therapy for the Kidney (BEST outcome, and prognostic factors of hospital mortal-
Kidney) Investigators. Acute renal failure in criti- ity; a prospective, multicenter study. Crit Care Med.
cally ill patients: a multinational, multicenter study. 1996; 24(2): 192–198.
JAMA. 2005; 294(7): 813–818. 16. Silvester W, Bellomo R, Cole L. Epidemiology, man-
9. Uchino S, Bellomo R, Goldsmith D et al. An assess- agement, and outcome of severe acute renal failure
ment of the RIFLE criteria for acute renal failure in of critical illness in Australia. Crit Care Med. 2001;
hospitalized patients. Crit Care Med. 2006; 34(7): 29: 1910–1915.
1913–1917. 17. Korkeila M, Ruokonen E, Takala J. Cost of care, long
10. Hoste EA, Clermont G, Kersten A et al. RIFLE crite- term prognosis and quality of life in patients requir-
ria for acute kidney injury are associated with hos- ing renal replacement therapy during intensive
pital mortality in critically ill patients: a cohort care. Int Care Med. 2000; 26: 1824–1831.
analysis. Crit Care. 2006; 10(3): R73. 18. Bagshaw SM. Epidemiology of renal recovery after
11. Kuitunen A, Vento A, Suojaranta-Ylinen R, Pettila V. acute renal failure. Curr Opin Crit Care. 2006; 12:
Acute renal failure after cardiac surgery: evaluation 544–550.
of the RIFLE classifi fication. Ann Thoracic Surg. 2006; 19. Rangel-Frausto MS, Pittet D, Costigan M et al. The
81(2): 542–546. natural history of the systemic infl flammatory
12. Chertow GM, Levy EM, Hammermeister KE et al. response syndrome (SIRS). A prospective study.
Independent association between acute renal failure JAMA. 1995; 273(2): 117–123.
and mortality following cardiac surgery. Am J Med. 20. Schrier RW, Wang W. Acute renal failure and sepsis.
1998; 104(4): 343–348. N Engl J Med. 2004; 315: 159–169.
13. Bagshaw SM, Laupland KB, Doig CJ et al. Prognosis 21. Schrier RW, Wang W, Poole B Mitra A. Acute renal
for longterm survival and renal revoery in critically failure: definitions,
fi diagnosis, pathogenesis and
ill patients with severe acute renal failure: A popu- therapy. J Clin Invest. 2004; 114: 5–14.
lation based study. Crit Care. 2005; 9: R700– 22. Lameire N, Van Biesen W, Vanholder R. Acute renal
709. failure. Lancet. 2005; 365: 417–430.
14. Metnitz PG, Krenn CG, Steltzer H et al. Effect of 23. Druml W. Acute renal failure is not a “cute” renal
acute renal failure requiring renal replacement failure! Intensive Care Med. 2004; 30: 1886–1890.
5
Medical Management of Acute Renal Failure
Nerina Harley
Acute renal failure (ARF) is both common and Patients with renal disease have a variety of
associated with signifificant mortality in the inten- clinical presentations:
sive care unit (ICU) setting (1). With an impact on
• Asymptomatic
length of stay, likelihood of survival until dis-
• Symptoms directly referable to the kidney,
charge, and cost of care, it is vital that the increas-
e.g., hematuria, flank
fl pain
ing body of evidence in critical care nephrology is
• Extrarenal symptoms, e.g., edema, hyperten-
used to refine
fi defi finitions, diagnosis, preventive
sion, uremic symptoms, consequences of
strategies, investigations, and management of
hyperkalemia
these patients (2).
• Manifestations of the underlying pathology
The medical management of ARF relies on the
or etiology, e.g., sepsis, hypotension, rhabdo-
basic tenets of diagnosis, elimination of reversi-
myolysis, systemic vasculitis
ble factors, amelioration of exacerbating factors,
treatment of complications, and optimization of
the “kidney’s environment” to provide maximal
recovery.
Investigations
Assessment of Renal Function
Diagnosis of ARF Although they lack sensitivity and specificity fi
serum creatinine and urine output are the most
Major causes of renal disease are divided, for useful parameters in clinical practice. Glomerular
simplicity, into three areas: filtration rate (GFR) assessment may be diffi
fi ficult
in nonsteady-state conditions. Comparison with
• Prerenal causes, caused by volume depletion
previous results is vital in determination of base-
with or without relative hypotension (reduced
line function, time course, and progression.
renal perfusion)
• Intrinsic renal causes, caused by vasculitis, glo-
merular disease, and tubulointerstitial disease Urinalysis
• Postrenal or obstructive causes
Urine examination may demonstrate granular and
The most common causes of ARF have been epithelial casts in ATN, eosinophils in acute inter-
found to be acute tubular necrosis (ATN) (45%), stitial nephritis, and red cell casts in acute vascu-
prerenal (21%), acute on chronic (13%), obstruc- litis or glomerulonephritis and/or proteinuria.
tion (10%), glomerulonephritis (GN), or vasculitis There are no prospective studies of the predictive
(4%) (3). value of urine sediment in ATN. Urinary elec-
Risk factors for the development of ARF are trolytes are of limited value in most clinical
given in Table 5.1. situations because of the confounding effects of
26
5. Medical Management of Acute Renal Failure 27
TABLE 5.1. Risk factors for ARF inherent risks of bleeding should be weighed up
Risk factor Selected reference(s) in the setting of the risk-to-benefit
fi ratio.
Note: The “gold standard” for the diagnosis of
Older age Nash et al., 2002 (43)
Diabetes Parfrey et al., 1989 (44). a prerenal cause of ARF is resolution of renal
Underlying renal insufficiency impairment in response to fluid challenge. This is
Cardiac failure in contrast to significant
fi ATN, in which there is
Sepsis Brun-Buisson et al., 2004 (45) prolonged time to resolution.
Ricci et al., (46)
Absolute or relative hypovolemia
Hepatic failure Han and Hyzy, 2006 (47)
Nephrotoxins (including Barrett et al., 1993 (48) Primary Prevention
high-osmolality Aspelin et al., 2003 (49).
radiocontrast agents) McCullogh et al., 2006 (50) Primary prevention of ARF in the critically ill with
Cardiopulmonary bypass with Mangano et al., 1998 (51)
or without baseline risk factors consists of avoid-
aortic cross-clamping Chertow et al., 1998 (52)
(particularly valve surgery) ance, amelioration, and treatment of these factors
Nonrenal organ transplantation Lima et al., 2003 (53) wherever possible. Strategies can be divided into
Abdominal compartment syndrome McNelis et al., 2003 (54) nonpharmacological, for example, fluid
fl adminis-
tration to reduce the risk of contrast induced
nephropathy (6), and pharmacological. To date, no
treatments such as diuretics. Urine volume is of pharmacological strategies have conclusively dem-
little diagnostic value, although little or no output onstrated prevention of ARF from any insult (7).
is acutely useful with causes of anuria, including
shock, bilateral urinary obstruction, renal cortical
necrosis, and bilateral vascular occlusion.
Loop Diuretics
Although oliguria (<400 mL/24 h) is common in
ATN, anuria is rare and other etiologies must be
Other Markers of Renal Injury
considered. Nonoliguric patients have a better
These have been described but are not currently prognosis than oliguric patients in terms of a
generally available in the acute setting. For greater residual GFR (8), lower peak serum creati-
example, serum cystatin C has been recently nine, and dialysis-free survival at 21 days (9, 10),
proposed as a marker of ARF (4), predicting ARF possibly reflecting
fl less severe renal injury. A clini-
by at least 24 hours. cal issue of the use of loop diuretics often arises in
increasing the urine output of oliguric patients.
Experimentally, loop diuretics reduce active
Radiology
sodium chloride transport in the thick ascending
Renal ultrasound is the modality of choice to limb of the loop of Henle, decreasing energy
exclude obstruction. Reduced renal size and corti- requirements and, thus, protecting the cell in
cal thinning (although preserved in diabetic the setting of decreased energy availability. Only
nephropathy) is indicative of chronic renal impair- anecdotal human evidence suggests that the use
ment. A helical computed tomographic (CT) scan of loop diuretics may be beneficialfi in the first 24
may be useful in urolithiasis, but there are risks of hours in flushing tubular casts. There is no evi-
secondary injury with radiocontrast. dence of benefi fit in established ATN on duration
of renal failure, requirement for dialysis, or sur-
vival (11). The increase in urine output in this
Renal Biopsy
setting is caused by decreased tubular reabsorp-
Renal biopsy is considered when noninvasive eval- tion in residual functioning nephrons (not re-
uation has not established the diagnosis (5); the cruitment of nonfunctioning nephrons), volume
major indications include isolated hematuria with expansion (initial sodium retention), and urea
proteinuria, nephrotic syndrome, acute nephritic osmotic diuresis.
syndrome, and unexplained ARF. Percutaneous A number of studies have suggested worsened
biopsy is most commonly performed and the outcomes in ATN with the use of loop diuretics in
28 N. Harley
the setting of contrast media (9, 12) and cardiac The incidence of ARF and requirement for
surgery. A systematic review (13, 14) comparing RRT was not significantfi different between
fluids alone with diuretics found no evidence of
fl groups (20).
improvement in survival, incidence of ARF, or • A large randomized controlled trial of low-dose
need for renal replacement therapy (RRT). Deaf- dopamine in 328 critically ill patients with
ness, possibly permanent, is a known complica- impaired creatinine, oliguria, and at least two
tion of high-dose loop diuretics. systemic infl flammatory response syndrome
(SIRS) criteria failed to show any benefit fi in
progression, need for RRT, or death (21).
Mannitol • Studies of fenoldopam, a relatively selective
Mannitol may preserve mitochondrial function by dopamine A1 receptor agonist, although shown
minimizing postinjury edema. Human trials have to increase sodium excretion and renal blood
failed to show benefifit in reducing ARF with man- flow in healthy and hypertensive patients, has
fl
nitol versus fluid alone in rhabdomyolysis (15), not shown benefi fit in ARF in the critically ill
cardiac surgery (16), and vascular and biliary tract (22).
surgery (17). A trend toward harm was noted in
the prevention of contrast nephropathy (12).
N-Acetylcysteine
N-Acetylcysteine (NAC) has been shown in a
N
Dopamine Agonists number of studies to decrease the incidence of
Dopamine has a number of effects in the kidney contrast nephropathy in high-risk patients (23,
via dopamine A1 and A2 receptors. In the proxi- 24), but without improvement in RRT require-
mal tubule, dopamine, via the generation of cyclic ment or survival. Importantly, NAC may decrease
AMP, decreases Na+-H+ exchange and the Na+-H+- creatinine via activation of creatinine kinase (25)
ATPase pump, thus, decreasing sodium reabsorp- but not GFR. Promising studies have led to the
tion. In the collecting tubules, this effect on introduction of protocols in many institutions for
Na+-H+-ATPase and decreased aldosterone secre- the prophylactic use of NAC in the prevention
tion reduces sodium reabsorption (18). of contrast-induced nephropathy. With few side
When infused in doses of 0.5 to 3 μg/kg/min, effects, low cost, and ease of administration (oral
dopamine causes afferent and efferent glomerular or intravenous), many centers have erred on the
arteriolar dilatation, increasing blood flow
fl with side of possible benefit
fi in the face of lack of hard
little or no increase in GFR. At higher concentra- evidence of long-term benefit fi (26).
tions, dopamine causes vasoconstriction via α-
adrenergic receptors.
There is no evidence in human studies for a Others
“renal protective effect” of dopamine.
Trials have failed to demonstrate benefi fit of natri-
• In 1994, Baldwin et al. studied the effect of post- uretic peptides (10, 27) or adenosine agonists (28).
operative low-dose dopamine on renal function Experimental therapies, such as antioxidants and
after major elective vascular surgery. Patients erythropoietin are unproven in humans.
were administered saline or saline plus dopa- In the absence of further evidence, flfluids, and
mine as fluid replacement. No difference in renal possibly NAC, are the gold standards of interven-
function was demonstrated between the two tional strategies.
groups (19).
• In the North American Study of the Safety
and Efficacy
fi of Murine Monoclonal Antibody Supportive Strategies
to Tumor Necrosis Factor for the Treatment
of Septic Shock (NORASEPT) II study, 400 Fluid resuscitation and correction of hypotension
patients with septic shock and oliguria nonran- are clearly essential. There is no evidence of
domly received no, low-, or high-dose dopamine. advantage of one particular inotrope over another.
5. Medical Management of Acute Renal Failure 29
Other
Constantly review diagnosis, which may Consider further investigations, e.g., renal biopsy and radiological investigations, as appropriate
multifactorial in nature
Appropriate medical review E.g., MET resources, nephrology input
10. Allgren RL, Marbury TC, Rahman SN, et al. Anarit- 23. Birck R, Krzossok S, Markowetz F, et al. Acetylcys-
ide in acute tubular necrosis. Auriculin Anaritide teine for prevention of contrast nephropathy: meta-
Acute Renal Failure Study Group. N Engl J Med. analysis. Lancet. 2003; 362(9384): 598–603.
1997; 336(12): 828–834. 24. Pannu N, Manns B, Lee H, Tonelli M. Systematic
11. Cantarovich F, Rangoonwala B, Lorenz H, et al. review of the impact of N-acetylcysteine on contrast
High-dose furosemide for established ARF: a nephropathy. Kidney Int. 2004; 65(4): 1366–1374.
prospective, randomized, double-blind, placebo- 25. Genet S, Kale RK, Baquer NZ. Effects of free radicals
controlled, multicenter trial. Am J Kidney Dis. 2004; on cytosolic creatine kinase activities and protec-
44(3): 402–409. tion by antioxidant enzymes and sulfhydryl com-
12. Solomon R, Werner C, Mann D, et al. Effects of pounds. Mol Cell Biochem. 2000; 210(1–2): 23–28.
saline, mannitol, and furosemide to prevent acute 26. Bagshaw SM, McAlister FA, Manns BJ, Ghali WA.
decreases in renal function induced by radiocon- Acetyl-cysteine used in the prevention of contrast-
trast agents. N Engl J Med. 1994; 331(21): 1416– induced nephropathy: a case study in the pitfalls in
1420. the evolution of evidence. Arch Intern Med. 2006 Jan
13. Kellum JA. Diuretics in acute renal failure: pro- 23; 166(2): 161–166.
tective or deleterious. Blood Purif. f 1997; 15(4–6): 27. Lewis J, Salem MM, Chertow GM, et al. Atrial natri-
319–322. uretic factor in oliguric acute renal failure. Anarit-
14. Kellum JA. The use of diuretics and dopamine in ide Acute Renal Failure Study Group. Am J Kidney
acute renal failure: a systematic review of the evi- Dis. 2000; 36(4): 767–774.
dence. Crit Care (Lond). 1997; 1(2): 53–59. 28. Kramer BK, Preuner J, Ebenburger A, et al. Lack of
15. Homsi E, Barreiro MF, Orlando JM, Higa EM. renoprotective effect of theophylline during aorto-
Prophylaxis of acute renal failure in patients coronary bypass surgery. Nephrol Dial Transplant.
with rhabdomyolysis. Ren Fail. 1997; 19(2): 283– 2002; 17(5): 910–915.
288. 29. Levy B, Bollaert PE, Charpentier C, et al. Compari-
16. Ip-Yam PC, Murphy S, Baines M, et al. Renal func- son of norepinephrine and dobutamine to epineph-
tion and proteinuria after cardiopulmonary bypass: rine for hemodynamics, lactate metabolism, and
the effects of temperature and mannitol. Anesth gastric tonometric variables in septic shock: a pro-
Analg. 1994; 78(5): 842–847. spective, randomized study. Intensive Care Med.
17. Gubern JM, Sancho JJ, Simo J, Sitges-Serra A. A ran- 1997; 23(3): 282–287.
domized trial on the effect of mannitol on postop- 30. Van den Berghe G, Wouters P, Weekers F, et al. Inten-
erative renal function in patients with obstructive sive insulin therapy in the critically ill patients.
jaundice. Surgery. 1988; 103(1): 39–44. N Engl J Med. 2001; 345(19): 1359.
18. Denton MD, Chertow GM, Brady HR. “Renal-dose” 31. Van den Berghe G, Wouters PJ, Bouillon R, et al.
dopamine for the treatment of acute renal failure: Outcome benefi fit of intensive insulin therapy in the
scientifific rationale, experimental studies and clini- critically ill: Insulin dose versus glycemic control.
cal trials. Kidney Int. 1996; 50(1): 4–14. Crit Care Med. 2003; 31(2): 359–366.
19. Baldwin L, Henderson A, Hickman P. Effect of post- 32. Doig GS, Simpson F. Evidence-based guidelines for
operative low-dose dopamine on renal function nutritional support of the critically ill: results of
after elective major vascular surgery. Ann Intern a bi-national guideline development conference.
Med. 1994; 120(9): 744–747. Sydney: EvidenceBased.net; 2005. Download www.
20. Marik PE, Iglesias J. Low-dose dopamine does not EvidenceBased.net.
prevent acute renal failure in patients with septic 33. Dhaliwal R, Heyland DK. Nutrition and infection in
shock and oliguria. NORASEPT II Study Investiga- the intensive care unit: what does the evidence show?
tors. Am J Med. 1999; 107(4): 387–390. Curr Opin Crit Care. 2005; 11(5): 461–467.
21. Bellomo R, Chapman M, Finfer S, et al. Low-dose 34. Doig GS, Simpson F. Early enteral nutrition in the
dopamine in patients with early renal dysfunction: critically ill: do we need more evidence or better
a placebo-controlled randomised trial. Australian evidence? Curr Opin Crit Care. 2006; 12(2): 126–
and New Zealand Intensive Care Society (ANZICS) 130.
Clinical Trials Group. Lancet. 2000; 356(9248): 35. Bellomo R, Goldsmith D, Uchino S, Buckmaster J,
2139–2143. Hart G, Opdam H, Silvester W, Doolan L, Gutteridge
22. Tumlin JA, Finkel KW, Murray PT, et al. Fenoldopam G. Prospective controlled trial on the effect of
mesylate in early acute tubular necrosis: a random- medical emergency team on postoperative morbid-
ized, double-blind, placebo-controlled clinical trial. ity and mortality rates. Crit Care Med. 2004; 32(4):
Am J Kidney Dis. 2005; 46(1): 26–34. 916–921.
32 N. Harley
36. Hillman K, Chen J, Cretikos M, Bellomo R, Brown 45. Brun-Buisson C, et al. EPISEPSIS: a reappraisal of
D, Doig G, Finfer S, Flabouris A; MERIT study inves- the epidemiology and outcome of severe sepsis in
tigators. Introduction of the medical emergency French intensive care units. Intensive Care Med.
team (MET) system: a cluster-randomised con- 2004; 30(4): 580–588.
trolled trial. Lancet. 2005; 365 (9477): 2091–2097. 46. Ricci Z, et al. Practice patterns in the management
37. Bellomo R, Bonventre J, Macias W, Pinsky M. Man- of acute renal failure in the critically ill patient: an
agement of early acute renal failure: focus on post- international survey. Nephrol Dial Transplant. 2006;
injury prevention. Curr Opin Crit Care. 2005; 11(6): 21(3): 690–696.
542–547. 47. Han MK, Hyzy R. Advances in critical care manage-
38. Vincent JL, Weil MH. Fluid challenge revisited. Crit ment of hepatic failure and insufficiency.
fi Crit Care
Care Med. 2006; 34(5): 1333–1337. Med. 2006; 34(9 Suppl): S225–231.
39. Kohli HS, Bhaskaran MC, Muthukumar T, et al. 48. Barrett B, et al. Meta analysis of the relative nephro-
Treatment-related acute renal failure in the elderly: toxicity of high- and low-osmolality iodinated con-
a hospital-based prospective study. Nephrol Dial trast media. Radiology. 1993; 188(1): 171–178.
Transplant. 2000; 15(2): 212–217. 49. Aspelin P, et al. Nephrotoxic effects in high-risk
40. Suen WS, Mok CK, Chiu SW, et al. Risk factors for patients undergoing angiography. N Engl J Med.
development of acute renal failure (ARF) requiring 2003; 348(6): 491–499.
dialysis in patients undergoing cardiac surgery. 50. McCullogh PA, et al. Risk prediction of contrast-
Angiology. 1998; 49(10): 789–800. induced nephropathy. Am J Cardiol. 2006; 98(68):
41. Uchino S, Bellomo R, Morimatsu H, et al. External 27K–36K.
validation of severity scoring systems for acute 51. Mangano C, et al. Renal dysfunction after myocar-
renal failure using a multinational database. Crit dial revascularization: risk factors, adverse out-
Care Med. 2005; 33(9): 1961–1967. comes, and hospital resource utilization. Ann Intern
42. Murray PT, Le Gall JR, Dos Reis Miranda D, et al. Med. 1998; 128(3): 194–203.
Physiologic endpoints (effi ficacy) for acute renal 52. Chertow G, et al. Independent association between
failure studies. Curr Opin Crit Care. 2002; 8(6): acute renal failure and mortality following cardiac
519–525. surgery. Am J Med. 1998; 104(4): 343–348.
43. Nash K, et al. Hospital-acquired renal insufficiency.
fi 53. Lima E, et al. Risk factors for development of acute
Am J Kidney Dis. 2002; 39(5): 930–936. renal failure after liver transplantation. Ren Fail.
44. Parfrey P, et al. Contrast material-induced renal 2003; 25(4): 553–560.
failure in patients with diabetes mellitus, renal 54. McNelis J, et al. Abdominal compartment syndrome:
insuffificiency, or both. A prospective controlled clinical manifestations and predictive factors. Curr
study. N Engl J Med. 1989; 320(3): 143–149. Opin Crit Care. 2003; 9(2): 133–136.
6
Acute Renal Failure in the Surgical Patient
Marlies Ostermann
Acute renal failure (ARF) is a potential complica- outer medulla is metabolically very active, despite
tion of any surgical procedure. In general, the risk a relatively low oxygen delivery. As a consequence,
of ARF is increased in patients with underlying oxygen extraction in this region is approximately
vascular disease, diabetes mellitus, or chronic 80% and, in the event of ischemia, this area is often
kidney disease. High-risk situations include car- the first part of the kidney to suffer injury.
diovascular, hepatobiliary, and trauma surgery,
especially if performed as an emergency.
Risk Factors for ARF in the Surgical
Pathophysiology of ARF Patient (Table 6.1)
33
34 M. Ostermann
TABLE 6.1. Risk factors for ARF developing in surgical patients 25% or a fall in glomerular filtration rate by 25%.
Preoperative factors One to 5% of patients need renal replacement
State of hydration/adequate resuscitation preoperatively therapy. A combination of patient-specific fic comor-
Contrast media bidities and factors related to surgery usually con-
Preexisting sepsis tribute to the development of ARF.
Patient comorbidity
Patient-specific
fic risk factors may not be imme-
• Advanced age
• Diabetes mellitus diately obvious. A large study on 7310 patients
• Hypertension undergoing coronary artery bypass grafting
• Cardiovascular disease (CABG) demonstrated that 29.6% of patients
• Preexisting renal impairment reported being diabetic, but an additional 5.3% of
Operative factors patients were found to have previously undiag-
Related to anesthesia (see text) nosed diabetes. Similarly, among patients under-
Related to surgery going coronary angiography, 12% of patients were
• Emergency surgery
• Duration of cardiopulmonary bypass
found to have an undiagnosed renal artery steno-
• Clamping of renal arteries sis of greater than 75%.
• Suprarenal aortic cross clamping Risk factors related to surgery tend to be
Postoperative factors less predictable, but the risk of renal failure is
• Sepsis generally higher in patients undergoing combined
• Bleeding CABG and valve replacement compared with
• Nephrotoxic drugs patients who only need one procedure.
• Contrast media
Very recently, aprotinin was identified
fi as a risk
• Cardiovascular complications associated with a fall in cardiac
output
t t ((e.g., myocardial
di l iinfarction,
f ti pulmonary
l embolus,
b l factor for ARF. An observational study on 4374
pericardial effusion) patients undergoing revascularization showed
• Development of intra-abdominal compartment syndrome that the use of aprotinin was associated with a
doubling of the risk of renal failure requiring
dialysis when compared with aminocaprionic
Treatment consists of discontinuation of the acid, tranexamic acid, or no antifibrinolytic.
fi
offending drug and possibly steroid therapy.
TABLE 6.3. Specific tests to consider in determining the cause of ARF in surgical patientsa
Blood tests Creatinine kinase To exclude rhabdomyolysis especially after trauma
Full blood count Eosinophilia is seen in 80% of patients with drug-induced interstitial
nephritis
Urinalysis Dipstick Significant proteinuria/hematuria/casts suggest intrinsic renal
pathology
Biochemistry Urinary sodium and osmolality to help differentiate prerenal failure
from ATN
Culture To exclude urinary tract infection
Diagnostic imaging Ultrasound scan To exclude obstruction and to determine renal size
Imaging of renal perfusion (e.g., renal Renal vascular supply may be of concern after major abdominal aortic
Dopplers, computed tomographic surgery. Investigation will depend on patient stability and local
angiogram, MAG3, DTPA) resources
Measurement of intravesical To exclude intra-abdominal hypertension and compartment syndrome
pressure
a
MAG3, mercaptoacetyltriglycine; DTPA, diethylene triamine pentaacetic acid.
Future Advances ing open heart surgery: risk factors and prognosis.
Perfusion 2005; 20: 317–322.
2. Barrett BJ, Parfrey PS. Preventing nephropathy
At present, no agents have conclusively demon-
induced by contrast medium. N Engl J Med d 2006; 354:
strated a protective effect against ARF or altera- 379–386.
tion of the course of ARF. Strategies aimed at 3. Chertow GM, Levy EM, Hammermeister KE, Grover
modulating renal function and renal recovery F, Daley J. Independent association between acute
have focused on several mechanisms: renal failure and mortality following cardiac surgery.
Am J Med d 1998; 104: 343–348.
1. Reduction of renal metabolism and energy
4. Friedrich JO, Adhikari N, Herridge MS, Beyene J.
consumption of the kidneys (i.e., induction of Meta-analysis: low-dose dopamine increases urine
hypothermia, use of insulin-like factor I). output but does not prevent renal dysfunction or
2. Modulation of the inflammatory
fl system (i.e., death. Ann Intern Med d 2005; 142: 510–524.
up-regulation of the acute stress response, 5. Lassnigg A, Donner E, Grubhofer G, Presterl E, Druml
manipulation of complement system, blockade W, Hiesmayr M. Lack of renoprotective effects of
of adhesion molecules). dopamine and furosemide during cardiac surgery.
3. Ischemic preconditioning. J Am Soc Nephroll 2000; 11: 97–104.
6. Mangano DT, Tudor IC, Dietzel C. Multicenter Study
These strategies are clearly important areas of of Perioperative Ischemia Research Group; Ischemia
research but not ready for clinical application. Research and Education Foundation. The risk associ-
ated with aprotinin in cardiac surgery. N Engl J Med
References 2006; 354: 353–365.
7. Van den Berghe G, Wouters PJ, Bouillon R, et al.
1. Bahar I, Akgul A, Ozatik MA, Vural KM, Demirbag Intensive insulin therapy in critically ill patients.
AE, Boran M, Tasdemir O. Acute renal failure follow- N Engl J Med d 2001; 345: 1359–1367.
7
Rhabdomyolysis and
Compartment Syndrome
Laurie Tomlinson and Stephen Holt
Rhabdomyolysis occurs when an insult causing necrosis. There may be severe pain, with loss of
myocyte necrosis results in release of intra- muscle function and loss of distal pulses. The
cellular contents into the circulation. Renal dys- diagnosis may be occult, especially in the uncon-
function is caused by a combination of renal scious patient. Direct pressure measurements
vasoconstriction, tubular damage, and tubular can be made by passing a needle connected to a
obstruction. pressure manometer (e.g., central venous pressure
[CVP] transducer) into the affected muscle
compartment.
Causes A fasciotomy should be considered if the pressure
exceeds 40 mmHg or greater than 30 mmHg above
Rhabdomyolysis accounts for approximately 7% diastolic pressure.
of all causes of acute renal failure (ARF) during
peacetime. This figure is much higher after natural
disasters and in wartime. For example, after the Diagnosis
1998 Turkish earthquake, 12% of the hospitalized
population developed significant
fi renal dysfunc- Serum changes consequent on rhabdomyolysis:
tion and 477 patients required dialysis.
Direct crush or compression injury and drugs
are the most important causes in clinical practice, Creatine Kinase
see Table 7.1. There are often predisposing factors,
for example, alcohol, which can presensitize myo- Very high levels of the muscle enzyme CK are
cytes so they may be damaged by a more trivial pathognomic of this condition. The degree of eleva-
insult. A clinical scoring system exists (not widely tion is proportional to the degree of muscle injury.
used) based on levels of phosphate, potassium, Other muscle enzymes, such as aspartate transami-
albumin, creatine kinase (CK), and presence of nase (AST) and lactate dehydrogenase (LDH) are
dehydration and sepsis. also elevated. CK levels should decline by approxi-
mately 40% per day, a plateau or an increase should
prompt a search for ongoing muscle damage.
Compartment Syndrome
Hyperkalemia
After an appropriate precipitant, inflammation
fl
within a muscular compartment causes a vicious Hyperkalemia caused by efflux
fl of potassium from
cycle of increasing pressure. This leads to further damaged cells is an early and life-threatening
infl
flammation and damage, eventually compro- consequence of rhabdomyolysis. It should be
mising blood supply, leading to further muscle aggressively treated.
38
7. Rhabdomyolysis and Compartment Syndrome 39
Pathophysiology of ARF
Suggested causes:
1. A reduction in renal blood flow
fl . There is a
FIGURE 7.1. “Brown sugar” Mb casts under light microscope are reduction in the effective blood volume caused
similar to granular casts but have a brown/rusty tinge. Additional by fluid
fl shifts from the intravascular to extracel-
red cells, tubular cells, and other debris are also present within lular fluid
fl compartments. Mb binds to nitric
the urine. oxide (NO), preventing intrarenal vasodilation
40 L. Tomlinson and S. Holt
(especially in the medulla) and, in addition, vaso- • Bicarbonate solutions that are more concentrated
dilators (e.g., endothelin) are increased. can be administered in small aliquots, e.g., 50 mL
2. Direct heme protein tubulotoxicityy occurs, of 8.4% NaHC03 via central access in patients
probably by free radical-mediated mechanisms. who are intravascularly full—remembering that
3. Tubular cast formation. Urinary Mb and this is 1 mmol of sodium per milliliter of fluid
fl
Tamm-Horsfall protein (THP) complex and precip- and may cause sodium/fluid
fl overload
itate as tubular casts. These casts are less soluble in
acidic conditions. Although there is some evidence
Mannitol
that these complexes cause tubular obstruction,
micropuncture studies have shown relatively low Mannitol promotes an osmotic diuresis and may
intratubular pressures, suggesting that these casts reduce pressure in a swollen muscle compartment,
are as a result of reduced tubular flflow and reduced but it also causes osmotically induced tubular
washout rather than by obstruction per se. damage with vacuolation. There is no good evi-
dence that it is more effective than saline alone
and it has little scientifi
fic rationale to recommend
Treatment its routine use.
This chapter covers some of the more specialized Normally, the immune system surveys cells and
causes of acute renal failure, which, although more tissues within the body, recognizing and ignoring
likely to present to the nephrologist, could be cells expressing “self” antigens while attacking
admitted to the intensive care unit as a conse- cells without these protective epitopes. In autoim-
quence of their illness or because of complications mune conditions, the immune system does not
of their treatment. protect cells with “self” expression. Cells are
attacked and either inflflamed or killed, or circulat-
ing self-antigens are bound with antibody-forming
Systemic Vasculitis immune complexes. Immune complexes are very
large molecules that are often unable to pass
Vasculitis is the term given to inflammation
fl of through capillaries because of their size. They can
blood vessels. Vasculitis is a rare condition, with induce local inflflammation and activate the com-
an incidence of approximately 6 people per million plement cascade.
(Western) population per year (1). Vessels can be
classified
fi according to their size (Table 8.1) (2).
Renal failure can occur in any vasculitis, but this
chapter focuses on those conditions that affect
Presentation
renal function directly through inflammation
fl
Symptoms
within the glomeruli (small vessels), rather than
affecting the kidneys indirectly through a reduc- Renal failure as a result of fulminant small vessel
tion of blood supply to the kidneys (large and vasculitis will present acutely, and patients may
medium vessel diseases). be systemically unwell, requiring organ support.
More often, however, there is an indolent presenta-
tion with several months of nonspecificfi symptoms
Etiology and signs (Table 8.2).
42
8. Multisystem Causes of Acute Renal Failure 43
Outcome
Renal vasculitis is a serious illness with significant
fi
morbidity and mortality. Before treatment, it was
universally fatal; with treatment, mortality is 10%
at 18 months. Up to 50% of patients require dialy-
sis with no renal recovery. If the patient did not
require dialysis at presentation, there is a 91%
FIGURE 8.2. Photomicrograph of renal biopsy from patient with
Wegener’s (cANCA-positive) renal vasculitis. Hematoxylin and
chance of renal survival. If dialysis dependent at
eosin stain with silver counterstain, showing features consistent presentation, 30% of patients may regain renal
with the disease. The glomerulus (black arrow)
w shows segmental function with cyclophosphamide and corticoste-
fibrinoid change (red arrows) and crescent formation (yellow
( roids alone, but 90% of patients may develop renal
arrow).
w Some tubules contain red cells (white arrows). (Photo recovery with the addition of therapeutic plasma
reproduced with the kind permission of Dr. Nicholas Marley.) exchange (3, 4).
8. Multisystem Causes of Acute Renal Failure 45
complications and will have consumption of all patients can die in the acute phase. Cerebrovascular
of the clotting factors. accidents, seizures, and coma occur in 25% of
• Malignant hypertension will have classical patients, and residual impairment of renal
retinal changes, signifificant high blood pressure excretory function is present in up to 40% of
(usually >210/130), and usually a history of patients.
hypertension. HUS/TTP in adults or children older than 14
years old usually requires treatment. In those with
an apparent cause (pregnancy, malignancy, infec-
Laboratory Investigations tion, or drugs), removal of the cause is essential.
• Low hemoglobin (<7 g/dL) Supportive treatment is required. Specific fi treat-
• Thrombocytopenia (<80 × 109 cells/L) ment of the condition revolves around therapeutic
• The blood filmfi will show red cell fragments plasma exchange or plasma infusion if TPE is
(schistocytes) and increased reticulocyte not available. There is no evidence of benefi fit of
counts immunosuppression with corticosteroids, immu-
• Elevated lactate dehydrogenase and indirect noglobulins, or vincristine, or of benefit fi from
bilirubin (caused by hemolysis) antithrombotic or antiplatelet agents.
• Haptoglobin levels will usually be low because Rescue therapies from severe refractory or
of consumption relapsing disease include bilateral nephrectomy
• Coombs test is negative and/or splenectomy.
• Moderate proteinuria (1–2 g/d) with few red
cells and casts (ARF is secondary to occlusion of
capillaries rather than inflammation)
fl References
1. Hedger N, Stevens J, Drey N, Walker S, Roderick P.
Treatment and Outcome Incidence and outcome of pauci-immune rapidly
progressive glomerulonephritis in Wessex, UK: a 10-
The epidemic or sporadic diarrhea-associated year retrospective study. Nephrol Dial Transplant
HUS/TTP of young children is usually self-limiting 2000; 15(10): 1539–1539.
and mild. Renal failure requires dialysis in approxi- 2. Jennette JC, Falk RJ, Andrassy K. Nomenclature of
mately 50% of patients, but otherwise supportive systemic vasculitides. Proposal of an international
treatment is all that is required. Ninety percent of consensus conference. Arthritis Rheum 1994; 37:
patients should recover completely. Up to 5% of 187–192.
48 T. Leach
3. Andrassy K, Kuster S, Waldherr R, Ritz E. Rapidly pro- lonephritis without anti-GBM antibodies. Kidney Int
gressive glomerulonephritis: analysis of prevalence 1991; 40(4): 757–763.
and clinical course. Nephron 1991; 59(2): 206–212. 5. Hoffman GS, Kerr GS, Leavitt RY, et al. Wegener gran-
4. Pusey CD, Rees AJ, Evans DJ, Peter DK, Lockwood ulomatosis: an analysis of 158 patients. Ann Intern
CM. Plasma exchange in focal necrotizing glomeru- Medd 1992; 116: 488–498.
9
Therapeutic Plasma Exchange
Tim Leach
Therapeutic plasma exchange (TPE) is an extra- • Of molecular weight greater than 15,000 kDa
corporeal blood purifi fication technique designed so it cannot be removed in any other way,
for the removal of large molecular weight sub- and/or
stances from plasma. Large molecular weight • Of suffi
ficient half-life that TPE is quicker than
substances equilibrate slowly between the vascu- endogenous removal, and/or
lar space and the interstitium. Calculations of • Acutely toxic and resistant to conventional
the rate of their removal by TPE follows fi first- therapy
order kinetics, i.e., approximately 60% is removed
by a single plasma volume exchange, and 75%
by an exchange equal to 1.4 times the plasma
volume.
Prescription (Table 9.1)
Blood is pumped through a highly permeable
Calculation of plasma volume:
filter, replacing the filtrate with fluid as indicated
fi
(Table 9.1). Venous access on the intensive care Estimated plasma volume (L) = 0.07 × Weight
unit (ICU) is via a double-lumen dialysis catheter, (kg) × (1 − hematocrit)
but can be via two wide-gauge peripheral venous
1. Before each treatment, measure serum potas-
cannulae. If chronic therapy is indicated, an arte-
sium calcium and clotting screen.
riovenous fifistula is used. The patient and filter are
2. Calculate the estimated plasma volume: Volume
anticoagulated during the procedure.
of exchange is measured in Patient Plasma
Volumes (∼3 L).
3. Prescribe the plasma exchange:
Indications (Table 9.2) a. Number and spacing of treatments
b. Volume and type of fl fluid
The basic premise of TPE is that removal of large
4. Electrolyte supplementation as needed (potas-
molecular weight substances from the circulation
sium and calcium).
will reduce further damage and may permit rever-
5. If coagulopathic consider, fresh-frozen plasma
sal of the pathological process.
(FFP) as the final exchange volume.
Other benefits
fi include unloading the reticulo-
endothelial system to permit further endogenous
removal of circulating toxins, stimulation of lym-
After Procedure
phocyte clones, and allowing re-infusion of large
volumes of plasma without the risk of volume • Repeat electrolytes and clotting after 2 hours
overload. and increase supplementation as needed
For TPE to be appropriate, the substance to be • FFP can be given as the final exchange volume
removed should be: if coagulopathic
49
50 T. Leach
Acute renal failure (ARF) occurs in 7% of patients maintain a concentration gradient favoring the
admitted to the intensive care unit (ICU) (1). Pre- out-diffusion of unwanted solutes. Both dialysate
viously, mortality exceeded 91%, but with the and replacement solutions contain electrolytes in
introduction of dialysis, this quickly fell to approx- physiological concentrations and some form of
imately 50% (2). Overall mortality for ARF has buffer base.
remained approximately 50%, associated with The hemofilter
fi r or dialysis membrane’s permea-
increasing comorbidity (3). bility to water is determined by its surface area
There is no specifi
fic therapy for ARF other than (typically 0.5 to 2.0 m2) and the number and size
removal of the cause and ongoing supportive care of its pores (typically 0.0055-μm diameter in a
awaiting spontaneous recovery. Renal replacement high-flux
fl hemofi filter). The pore size determines
therapy (RRT) is the cornerstone of that support- the size of molecules that are freely fi
filtered along
ive care. with water. Current hemofilters
fi freely filter sub-
stances up to approximately 5000-D molecular
weight and then in decreasing amounts up to a cut
The Basics off of approximately 20,000 Da. This minimizes
loss of important larger molecules, such as
Extracorporeal RRT involves the passage of a albumin (molecular weight, 57,000 Da).
patient’s blood outside his/her body through a The ratio between the concentration of solute
dialysis or hemofilter
fi machine, where the removal in filtrate and that in plasma water is called the
of unwanted solutes and excess water and the sieving coefficient
fi , and this concept becomes
replacement of lost bicarbonate (or buffer base) important in calculating the clearance of interme-
take place. The “purified”
fi blood is returned to the diate-sized molecules. The sieving coefficient
fi for
patient. a small unbound solutes is one, decreasing to zero
Clearance can be definedfi as that volume of as molecular size and or plasma protein binding
plasma completely cleared of a substance in a increases.
given time. During extracorporeal RRT, net solute Theoretical small solute clearance can be pre-
clearance can be achieved by ultrafiltration
fi across dicted by knowledge of the blood flow
fl through the
a porous membrane down a pressure gradient (fi (fil- extracorporeal circuit (QB) and the rate of ultrafil-
fi
tration), or by diffusion across a semipermeable tration (QF) and dialysate flow
fl (QD) (Figure 10.1).
membrane down a concentration gradient (dialy- During continuous RRT (CRRT), when blood
sis), or both. During fifiltration, the filtrate is dis- flow is signifi
fl ficantly higher than dialysate or ultra-
carded and a replacement solution is added to the filtration rates, small solute clearance is deter-
fi
blood to maintain fluid
fl and electrolyte equilib- mined by dialysate and ultrafiltrate
fi flow. Assuming
rium. During dialysis, a continuous stream of complete concentration equilibrium between dial-
dialysate is passed in the opposite direction to ysate and plasma water and a sieving coefficient
fi
blood, on the nonblood side of the membrane, to equal to one:
51
52 J.H. Reeves
In 1977, when Peter Kramer firstfi described con- Mechanism of Solute Removal
tinuous arteriovenous hemofi filtration (CAVH)
as a therapy for diuretic resistant flfluid overload Convection: when solute is cleared by ultrafiltra-
fi
(5), the only other types of RRT were IHD and tion through a porous membrane, we say that
peritoneal dialysis (PD). Since then, the classifica-
fi the clearance of the solute is convective—carried
tion of RRT has expanded (6). See the glossary in by the bulk flow of plasma water. During CRRT,
Table 10.1. the process is called hemofiltration
fi .
Diffusion: when solute is cleared by diffusion
down a concentration gradient across a semi-
Duration or Timing of Therapy
permeable membrane, we say that the clearance
Continuous: CRRT aims to provide support 24 h/d, of the solute is diffusive. The process is called
but, in practice, is interrupted by factors such as hemodialysis.
patient transfer out of the ICU or circuit failure. Combinations: diafiltration
fi is the term applied
Intermittent: IHD requires approximately 3 hours when both convection and diffusion are operat-
per treatment. Patients with end-stage renal ing to remove solute.
failure (ESRF) may be maintained in the com- Adsorption: adsorption is the binding of sub-
munity with as few as three dialysis sessions per stances to the membrane under molecular
10. Renal Replacement Therapy 53
attraction. Adsorption is used specifi fically for There is little information regarding what spe-
toxin removal using activated charcoal car- cific
fi threshold plasma concentrations of potas-
tridges (10) and it is being tested as a means of sium, bicarbonate, urea, or creatinine should be
blood purifification in sepsis (11). used for the institution of RRT. In a retrospective
comparison of early versus late CRRT in trauma
associated ARF, Gettings et al. (23) found that
Intensity of Therapy patients in whom CRRT was commenced at a
mean blood urea nitrogen (BUN) of 42.6 mg/dL
For example, slow continuous ultrafiltration
fi
(15.2 mmol/L) had a survival rate of 39% com-
(SCUF) is performed to simply remove excess
pared with 20% in patients in whom the
extracellular flfluid. High-volume hemofi filtration
mean BUN at commencement was 94.5 mg/dL
(HVHF) is used to intentionally accelerate the
(33.7 mmol/L).
clearance of target mediators, and high-fl
flux dialy-
sis (HFD) (dialysis performed with highly perme-
able membranes) is designed to accelerate the “Nonrenal” Indications
clearance of urea and larger molecules.
• Drug and toxin removal
• Sepsis and septic shock
• Inborn errors of metabolism
Indications for RRT • Congestive cardiac failure
• Cerebral edema
“Traditional” indications (19)
IHD can accelerate the elimination of small
• Diuretic resistant fluid overload
(<500 mw) unbound toxins with a low volume of
• Life-threatening hyperkalemia
distribution and minimal plasma protein binding
• Severe metabolic acidosis
(11), e.g., lithium, methanol, ethylene glycol, and
• Symptomatic uremia
salicylates. Continuous hemofiltration
fi has been
Kramer’s original description of CAVH involved used in lithium toxicity (24), with better hemody-
the treatment of diuretic resistant fl
fluid overload namic stability (25) but lower solute clearance
(5). During the intervening years, there has been than IHD.
controversy surrounding the use of diuretics in The use of extracorporeal blood purification
fi
ARF (20), but they may be helpful in the fl fluid techniques in sepsis and septic shock is attractive
management of ARF before the institution of RRT but unproven. There were early observations of
(21, 22). improved cardiovascular and respiratory function
54 J.H. Reeves
in patients with severe sepsis after commence- of distribution of urea was cleared during the
ment of continuous hemofiltration
fi (26). This, session. Using a single compartment exponential
together with the identification
fi of inflflammatory washout model, we can predict that the final fi
mediators in filtrate
fi (27), led to efforts to increase concentration of the solute is approximately
infl
flammatory mediator removal during RRT. 37% of the starting concentration when the Kt/V
High-volume conventional hemofiltration
fi (14, 28), is 1.0.
large-pore hemofiltration
fi (29), and plasma filtra- In chronic renal failure, a minimum Kt/V of 1.2
tion with (30) or without (15) coupled adsorption should be delivered three times per week (36).
have been examined in small clinical trials. There There has been one randomized controlled trial
is currently no Level I evidence for the use of assessing dose of dialysis in ARF (7). Schiffl fl com-
extracorporeal blood purifi fication therapy in pared daily dialysis with alternate daily dialysis in
sepsis. 160 patients with ARF. The 28-day mortality using
End-stage cardiac failure is characterized by intention-to-treat analysis was 28% in the daily
progressive fl fluid retention, renal impairment, treated patients and 46% in the patients treated
neurohumoral stimulation, and diuretic resistance every other day. Multiple other outcomes were
(31). It was shown that patients with advanced improved in the daily dialysis group. Although this
cardiac failure can tolerate substantial fluid fl trial was controversial (37), it suggests that dialy-
removal during ultrafiltration
fi (12), with salutary sis every second day is insufficient
fi in critically ill
effects that persist beyond the time of fl fluid patients with ARF.
removal. Improved renal function, decreased heart Quantifi fication of clearance during CRRT can be
failure scores, lowered B natriuretic peptide levels, likened to the calculation of creatinine clearance,
decreased hospital length of stay, and fewer read- which is described by the formula UV/P. U is the
missions have all been observed in case-controlled urine concentration of the solute, V is the volume
studies (32). Most recently, a small randomized collected in a given time, and P is the plasma con-
controlled study showed that early ultrafiltration
fi centration. During CRRT, if we know the volume
results in increased weight loss at 24 h compared of efflfluent produced in a given time and the con-
with diuretics alone (33). Larger studies are war- centration of solute (e.g., urea) in both effluent
fl
ranted for this indication. and plasma, we can calculate its clearance (38).
Cerebral edema can complicate IHD (34) and The result is a value in milliliters per minute. To
contribute to the clinical picture of disequilib- simplify the estimate further, let us assume that
rium. In patients with hepatic encephalopathy, the sieving coefficient
fi is 1.0, and that there is full
early studies compared the effects of IHD and concentration equilibrium between plasma water
CRRT (35). CRRT was associated with less decrease and dialysate. Then the effluent
fl concentration will
in mean arterial pressure, less increase in intracra- equal the plasma concentration, and clearance is
nial pressure, and less change in cerebral perfu- simply equal to the rate of production of effluent
fl
sion pressure. from the hemofi filter.
Ronco, in 2000 (39), randomized 425 critically
ill patients with ARF to three different doses
Choosing the Dose and Mode of RRT (ultrafiltration
fi rates) of CRRT: 20, 35, or 45 mL/h/
kg. Fifteen-day survivals were 41%, 57%, and 58%,
Dialysis dose is a concept familiar to nephrologist respectively. This landmark study was one of the
in the setting of ESRF: Kt/V. first to formally adjust dose of CRRT based on
fi
K is clearance (the volume of solute, usually patient weight. Importantly, it suggests that there
urea, cleared in a given time), t is duration of treat- is a threshold minimum level of clearance required
ment, and V is volume of distribution of the solute. for adequate CRRT of approximately 35 mL/h/kg.
Kt is the volume of plasma water cleared of solute Debate regarding the relative merits of CRRT
during the session, and Kt/V is Kt as a proportion and IHD has continued for nearly 30 years. From
of its volume of distribution. the outset (5), the attraction of CRRT was its sim-
For example, a Kt/V of 1.0 for urea means that plicity and cardiorespiratory stability compared
a total volume of plasma water equal to the volume with IHD. Now that CRRT is as complex as IHD
10. Renal Replacement Therapy 55
and potentially more expensive to perform for of an artifi ficial kidney, prognosis. Am J Med d 1955;
long periods, there is a paucity of good compara- 18:187.
tive studies that report outcomes such as mortal- 3. Mehta RL, Pascual MT, Soroko S, et al. Spectrum of
ity or recovery of renal function. In 2001, Mehta acute renal failure in the intensive care unit: the
PICARD experience. Kidney Intt 2004; 66(4):
randomized 166 critically ill patients to CRRT or
1613–1621.
IHD (40). The observed mortality was higher in
4. Meyer TW, Walther JL, Pagtalunan ME, et al. The
the group receiving CRRT, but this group had a clearance of protein-bound solutes by hemofiltra- fi
higher severity of illness. In 2002, Kellum pub- tion and hemodiafi filtration. Kidney Intt 2005; 68(2):
lished a meta-analysis of 13 trials, 3 randomized 867–877.
and 10 observational, comparing IHD and CRRT 5. Kramer P, Wigger W, Rieger J, et al. Arteriovenous
(41). Overall, there was no difference in mortality, haemofi filtration: A new and simple method for
but only six studies compared groups of equal treatment of over-hydrated patients resistant to
severity. In these six studies, the mortality was diuretics. Klin Wschr 1977; 55: 1121–1122.
lower in patients treated with CRRT. Since then, a 6. Ronco C, Bellomo R. Continuous renal replacement
well-designed prospective randomized controlled therapy: evolution in technology and current nomen-
clature. Kidney Int Suppl 1998; 66: S160–164.
trial has been published comparing CRRT with
7. Schifflfl H, Lang SM, Fischer R. Daily hemodialysis
IHD in 80 critically ill patients with ARF. There
and the outcome of acute renal failure. N Engl J Med
was no difference in survival or renal recovery 2002; 346(5): 305–310.
between the two groups. Although there was more 8. Marshall MR, Golper TA, Shaver MJ, Chatoth DK.
hemodynamic disturbance during IHD, this did Hybrid renal replacement modalities for the criti-
not translate to a survival benefifit for CRRT (42). cally ill. Contrib Nephrol 2001(132): 252–257.
9. Naka T, Baldwin I, Bellomo R, Fealy N, Wan L. Pro-
longed daily intermittent renal replacement therapy
Summary in ICU patients by ICU nurses and ICU physicians.
Int J Artif Organs 2004; 27(5): 380–387.
• RRT reduces the mortality of ARF from more 10. Zimmerman JL. Poisonings and overdoses in the
intensive care unit: general and specifi fic manage-
than 90% to approximately 50%.
ment issues. Crit Care Med 2003; 31(12): 2794–
• There is no proven difference in outcome
2801.
between intermittent and CRRTs, as long as a 11. Nalesso F. Plasma filtration adsorption dialysis
minimum dose of Kt/V of 6 to 8 per week is (PFAD): a new technology for blood purification. fi
achieved during hemodialysis or a clearance Int J Artif Organs 2005; 28(7): 731–738.
greater than 35 mL/kg/h is achieved during con- 12. Silverstein ME, Ford C, Lysaght MJ, Henderson LW.
tinuous hemofi filtration. Treatment of severe fl fluid overload by ultrafifiltration.
• New hybrid therapies (slow long extended daily New Eng J Med d 1974; 291(15): 747–751.
dialysis [SLEDD], extended daily dialysis [EDD], 13. Ronco C, Bellomo R, eds. Continuous high fl flux
and prolonged daily intermittent RRT [PDIRRT]) dialysis: an effificient renal replacement. Heidelberg:
may avoid the destabilization associated with Springer Verlag; 1996.
14. Cole L, Bellomo R, Journois D, et al. High-volume
IHD, and mitigate the cost and inconvenience of
haemofi filtration in human septic shock. Intensive
continuous hemofiltration.
fi
Care Med 2001; 27(6): 978–986.
15. Reeves JH, Butt WW, Shann F, et al. Continuous plas-
References mafi filtration in sepsis syndrome. Plasmafi filtration in
Sepsis Study Group. Crit Care Med d 1999; 27(10):
1. Brivet FG, Kleinknecht DJ, Loirat P, Landais PJ. 2096–2104.
Acute renal failure in intensive care units-causes, 16. Ronco C, Brendolan A, d’Intini V, et al. Coupled
outcome, and prognostic factors of hospital mortal- plasma filtration
fi adsorption: rationale, technical
ity; a prospective, multicenter study. French Study development and early clinical experience. Blood
Group on Acute Renal Failure. Crit Care Med 1996; Purif 2003; 21(6): 409–416.
24(2): 192–198. 17. Marshall MR, Ma T, Galler D, et al. Sustained low-
2. Smith LH, Post RS, et al. Post traumatic renal insuf- effificiency daily diafifiltration (SLEDD-f) for critically
ficiency in military casualties. II. Management, use ill patients requiring renal replacement therapy:
56 J.H. Reeves
towards an adequate therapy. Nephrol Dial Trans- in septic shock. Crit Care Med 2002; 30(6): 1250–
plantt 2004; 19(4): 877–884. 1255.
18. Kumar VA, Craig M, Depner TA, Yeun JY. Extended 31. Ellison DH. Diuretic therapy and resistance in con-
daily dialysis: A new approach to renal replacement gestive heart failure. Cardiology 2001; 96(3-4): 132–
for acute renal failure in the intensive care unit. Am 143.
J Kidney Dis 2000; 36(2): 294–300. 32. Costanzo MR, Saltzberg M, O’Sullivan J, Sobotka P.
19. Palevsky PM. Renal replacement therapy I: indica- Early ultrafiltration
fi in patients with decompensated
tions and timing. Crit Care Clin 2005; 21(2): 347– heart failure and diuretic resistance. J Am Coll
356. Cardiol 2005; 46(11): 2047–2051.
20. Mehta RL, Pascual MT, Soroko S, Chertow GM. 33. Bart BA, Boyle A, Bank AJ, et al. Ultrafiltration
fi
Diuretics, mortality, and nonrecovery of renal func- versus usual care for hospitalized patients with
tion in acute renal failure. JAMA 2002; 288(20): heart failure: the Relief for Acutely Fluid-
2547–2553. Overloaded Patients With Decompensated Conges-
21. Cantarovich F, Rangoonwala B, Lorenz H, et al. tive Heart Failure (RAPID-CHF) trial. J Am Coll
High-dose furosemide for established ARF: a pro- Cardiol 2005; 46(11): 2043–2046.
spective, randomized, double-blind, placebo- 34. Walters RJ, Fox NC, Crum WR, et al. Haemodialysis
controlled, multicenter trial. Am J Kidney Dis 2004; and cerebral oedema. Nephron 2001; 87(2): 143–147.
44(3): 402–409. 35. Davenport A, Will EJ, Davison AM, et al. Changes in
22. Uchino S, Doig GS, Bellomo R, et al. Diuretics and intracranial pressure during machine and continu-
mortality in acute renal failure. Crit Care Med d 2004; ous haemofi filtration. Int J Artif Organs 1989; 12(7):
32(8): 1669–1677. 439–444.
23. Gettings LG, Reynolds HN, Scalea T. Outcome in 36. Eknoyan G, Levin N. NKF-K/DOQI Clinical Practice
post-traumatic acute renal failure when continuous Guidelines: Update 2000. Foreword. Am J Kidney Dis
renal replacement therapy is applied early vs. late. 2001; 37(1 Suppl 1): S5–6.
Intensive Care Med d 1999; 258: 805–813. 37. Drazen JM, Ingelfinger
fi JR, Curfman GD. Removal of
24. van Bommel EF, Kalmeijer MD, Ponssen HH. Treat- expression of concern: Schiffl fl H, et al. Daily hemo-
ment of life-threatening lithium toxicity with high- dialysis and the outcome of acute renal failure. N
volume continuous venovenous hemofi filtration. Am Engl J Med 2002; 346: 305–310. N Engl J Med 2003;
J Nephrol 2000; 20(5): 408–411. 349(20): 1965.
25. Maggiore Q, Pizzarelli F, Dattolo P, et al. Cardio- 38. Reeves JH, Butt WW. A comparison of solute clear-
vascular stability during haemodialysis, haemo- ance during continuous hemofiltration,
fi hemodiafi fil-
filtration and haemodiafi filtration. Nephrol Dial tration, and hemodialysis using a polysulfone
Transplant 2000; 15 Suppl 1: 68–73. hemofi filter. ASAIO J 1995; 41(1): 100–104.
26. Gotloib L, Barzilay E, Shustak A, et al. Hemofiltra-
fi 39. Ronco C, Bellomo R, Homel P, et al. Effects of differ-
tion in septic ARDS. The artificial fi kidney as an ent doses in continuous veno-venous haemofiltration
fi
artifi
ficial endocrine lung. Resuscitation 1986; 13(2): on outcomes of acute renal failure: a prospective
123–132. randomised trial. Lancett 2000; 356(9223): 26–30.
27. Tonnesen E, Hansen MB, Hohndorf K, et al. Cyto- 40. Mehta RL, McDonald B, Gabbai FB, et al. A random-
kines in plasma and ultrafi filtrate during continuous ized clinical trial of continuous versus intermittent
arteriovenous haemofi filtration. Anaesth Intensive dialysis for acute renal failure. Kidney Intt 2001;
Care 1993; 21(6): 752–758. 60(3): 1154–1163.
28. Cole L, Bellomo R, Hart G, et al. A phase II random- 41. Kellum JA, Angus DC, Johnson JP, et al. Continuous
ized, controlled trial of continuous hemofiltration
fi versus intermittent renal replacement therapy: a
in sepsis. Crit Care Med d 2002; 30(1): 100–106. meta-analysis. Intensive Care Med d 2002; 28(1):
29. Uchino S, Bellomo R, Goldsmith D, et al. Super high 29–37.
flux hemofi filtration: a new technique for cytokine 42. Augustine JJ, Sandy D, Seifert TH, Paganini EP. A
removal. Intensive Care Med d 2002; 28(5): 651–655. randomized controlled trial comparing intermit-
30. Ronco C, Brendolan A, Lonnemann G, et al. A pilot tent with continuous dialysis in patients with ARF.
study of coupled plasma filtration
fi with adsorption Am J Kidney Dis 2004; 44(6): 1000–1007.
11
Technical Aspects of Renal
Replacement Therapy
Sara Blakeley
Since its inception in 1977 (1), methods and equip- CRRT on the ICU nowadays is almost exclu-
ment used for the delivery of continuous renal sively venovenous in nature; in other words, blood
replacement therapy (CRRT) has undergone many is removed from a large vein and returned to a
changes. Intermittent hemodialysis (IHD) is per- large vein. The different modes of CRRT differ
formed on some intensive care units (ICU), but this mainly in their method of solute removal, and
chapter will concentrate on continuous therapies. most modern machines are capable of providing
the full range of modalities. Newer machines offer
greater ease in switching between modes.
Mode of CRRT There is much discussion regarding the rela-
tionship between dialysis dose delivered and sur-
There is a spectrum of treatment available and vival (6). A prospective randomized study of post
with choice comes debate; continuous versus dilution CVVH in critically ill patients found
intermittent, convective versus diffusive therapy improved survival with an ultrafiltration
fi dose of
(2, 3). Continuous therapies have been associated 35 ml/kg/hour (7). Ultrafiltration
fi rate is used as a
with better renal survival compared with IHD, dose surrogate. This finding was not repeated in a
although no compelling effect on overall mortality subsequent study (8) but a more recent study
has been seen (4). found did find an improvement in survival when
Diffusion is the movement of molecules from adding a dialysis dose to CVVH (9). It is unclear
an area of high concentration (blood) to one of whether this survival advantage was due to a
a lower concentration (dialysis fl fluid circulating higher dose of CRRT overall or due to the addition
in a counter current direction) across a semi- of a diffusive therapy to a convective one. These
permeable membrane. With convection a pressure findings are being investigated further in other
fi
(transmembrane pressure) is applied across the studies, to clarify what is the optimal dose and
membrane, this drives water out and carries with means of delivery. In the mean time 35 ml/kg/hour
it dissolves solutes (solvent drag). The “waste” fl
fluid is often recommended as minimum that should be
produced is termed ultrafiltrate.
fi Both diffusive provided (10–12).
and convective therapies are good at removing Standard therapies in use are:
small molecular weight molecules (<5000 Da),
such as urea and creatinine, but “middle mole- Continuous venovenous hemofi filtration (CVVH),
cules” (10,000–50,000 Da) such as β2 microglobu- characterized by predominantly convective
lin and cytokines (5) may be better removed solute clearance (Figure 11.1A).
by convection. Currently, there is no definitive
fi Continuous venovenous hemodialysis (CVVHD),
evidence to suggest one particular mode over characterized by predominantly diffusive solute
another, and choice is often guided by local clearance but with some convection occurring
expertise. because of ultrafi
filtration.
57
58 S. Blakeley
Effluent pump
Replacement fluid
entering pre filter (pre
dilution)
Haemofilter
Pressure monitor
Air detector
A
Anticoagulation syringe
Effluent pump
Haemofilter
Pressure monitor
Air detector
B
FIGURE 11.1. A, CVVH with prefilter replacement fluid delivery (predilution). B, CVVHDF with postfilter replacement fluid delivery
(postdilution).
11. Technical Aspects of Renal Replacement Therapy 59
The fluid contains a buffer, either lactate or molecular weight less than 50,000 Daltons (Da),
bicarbonate, and electrolytes (sodium, chloride, i.e., smaller than albumin. Membranes are com-
magnesium, and calcium). Typically, replacement posed of two substances, cellulose (e.g., cupro-
fluids are potassium- and phosphate-free, and
fl phan) and synthetic fibers (e.g., polysulphone,
these may need to be replaced as clinically indi- polyamide, or polyacrylonitrile).
cated. Glucose is often not present.
Membrane Characteristics
Prefilter Versus Postfilter Infusion
Biocompatibility
Replacement fluid
fl is infused into the circuit either
before the filter (predilution) or after the filter Contact of blood with the fi
filter surface can lead to
(postdilution). Predilution lowers the hematocrit complement and leucocyte activation, triggering
of the blood passing through the filter,
fi potentially the coagulation cascade and infl flammatory
reducing anticoagulation needs and allowing pathways. More “biocompatible” indicates less
increased ultrafifiltration rates. This is at the expense complement/leucocyte activation. Activation of
of less-effective solute clearance, therefore, an infl
flammatory mediators has been suggested as
increased ultrafifiltration rate is needed to achieve one mechanism leading to ongoing renal injury,
similar solute clearances. Newer machines allow a and, therefore, delay or nonreturn of renal func-
combination of predilution and postdilution. tion (4). Kidneys that have already been injured
because of reduced renal perfusion lose their
ability to autoregulate pressure changes and the
Lactate Versus Bicarbonate Buffer kidney becomes very sensitive to even small
changes in renal perfusion. Early reports that bio-
Both lactate- and bicarbonate-buffered solutions
incompatible cellulose-based membranes led to a
have been shown to be effective in correcting
worse outcome have been debated, but, currently,
metabolic acidosis (11, 13). Bicarbonate is pre-
the evidence is not robust enough to defi finitively
ferred in patients with a preexisting lactic acidosis
recommend a synthetic membrane over cellulose
(e.g., septic shock) or with liver failure (11) because
or modifiedfi cellulose membrane (15). As it is,
these patients may be unable to metabolize an
most membranes used in CRRT are synthetic
exogenous lactate load normally, thus, worsening
because they have a greater degree of flflux.
the acidosis. Lactate levels are often seen to rise in
other patient groups who have a lactate buffer, but
the signifificance of this is unclear because hyper- Flux
lactatemia is not always associated with an acido-
This is a measure of ultrafifiltration capacity and is
sis. Bicarbonate is a more physiological buffer, but,
based on the membrane ultrafiltration
fi coeffi
ficient.
because it is unstable in solution, it needs to be
A filter with a high permeability coeffi ficient to
added just before use. Studies have compared out-
water will allow more ultrafi filtration (high flux)
comes of bicarbonate versus lactate buffers but
and, hence, more convective transport. Permeabil-
the evidence is inconclusive. Lactate intolerance
ity is a measure of the clearance of middle molec-
has been arbitrarily defi fined as a rise of greater
ular weight molecules and high permeability is
than 5 mmol/L during CRRT (14), and a change to
seen with high flflux membranes (synthetic mem-
a bicarbonate buffer should be considered.
branes). It should be noted that high permeability
does not always equate to high urea clearance
(efficiency).
fi
Hemofilters
Structure Vascular Access
Thousands of hollow fibers (membrane) are
bundled together forming a hemofilter fi with a When initially developed, CRRT used an arterial
large surface area of 0.6 to 1.2 m2. Pores in the and a venous catheter (arteriovenous) (1). A wide-
membrane allow the passage of molecules with a bore (11.5–13.5 French) dual-lumen vascular dial-
11. Technical Aspects of Renal Replacement Therapy 61
ysis (venovenous) catheter is now generally used. through the circuit can also result in the formation
Mostly composed of polyurethane, they can have of platelet microthrombi, which can occlude the
an antibiotic/antimicrobial coating. filter. Loss of the filter through clotting can lead to
fi
Blood is pumped from the patient (arterial ineffective dialysis and patient blood loss, as well
side) through proximal side holes into one lumen, as being a drain on resources, both nursing and
and is then returned though a port at the distal tip financial. Methods such as predilution and ensur-
fi
of the second lumen (venous side). High blood ing adequate vascular access can be used, but some
flows without high pressures are ideal catheter
fl form of anticoagulation for the extracorporeal
design requirements, and the dual-lumen design circuit is often needed. Remember: circuit failure is
allows continuity and reduces recirculation. more often caused by inadequate vascular access
Vascular catheters differ in length, diameter of rather than inadequate anticoagulation.
lumen, and positioning of ports, and some have an
extra lumen added for drug infusions. Remember:
No Anticoagulation
always assume that EACH catheter limb contains
heparin and aspirate at least 5 mL of blood before In the setting of deranged clotting (e.g., interna-
using. Cuffed dialysis catheters are generally not tional normalized ratio [INR] >2, activated partial
used on the ICU, but may be indicated in stable thromboplastin time [aPTT] >60 s) and/or throm-
patients who are free of infection, and who require bocytopenia (e.g., platelet count <50,000) or a
ongoing renal replacement therapy (RRT) (Table high risk of bleeding, further anticoagulation is
11.1). often not necessary or carries the risk of bleeding.
With adequate access and predilution, it is possi-
ble to run the circuit without any anticoagulation
Positioning for an acceptable period of time and achieve good
A correctly positioned catheter will have a better solute clearance.
blood flow; good access is the key to good dialysis.
The tip of a jugular or subclavian catheter should
Unfractionated Heparin
extend to the superior vena cava and rest 1 to 2 cm
above the right atrium. Too short, and there is Unfractionated heparin (UFH) is the most com-
the possibility of recirculation, whereas a catheter monly used extracorporeal anticoagulant. The
that is too long risks atrial perforation. Femoral circuit is often primed with heparin (e.g., 1000–
catheters should be longer (>20 cm) to reach the 10,000 IU) because it is highly negatively charged
inferior vena cava, therefore, minimizing recircu- and is absorbed onto the plastic circuit. Depend-
lation and achieving better fl flow rates. ing on the risk of bleeding, a bolus dose (e.g., 10–
20 IU/kg) can be administered and a continuous
infusion started. A low-dose infusion (<5 IU/kg/h)
Site of Catheter aims NOT to prolong the aPTT. If clotting occurs,
Debate continues regarding the ideal site for place- a medium dose can be considered (5–10 IU/kg/h),
ment of dialysis catheters in terms of safety of aiming for mild prolongation of the aPTT (1–1.4
insertion and infection risk. Where long term times normal) (16). Prefilter
fi heparin can be neu-
dialysis is a possibility, there is concern that tralized with postfilter
fi protamine (e.g., 1000 IU/h
subclavian catheters may be associated with an to 10 mg/h), called regional heparinization. If
increased incidence of subclavian stenosis/throm- patients require formal heparinization for condi-
bosis, creating long-term problems for arteriove- tions such as a pulmonary embolus, this should be
nous fistula formation. continued, no extra “fi filter” heparin is needed.
The methods, site of sampling, and frequency
of anticoagulation monitoring vary depending
Anticoagulation on local protocols. Commonly used methods are
the activated coagulation time (ACT) and aPTT.
As blood flows
fl through the filter and fluid is However, it should be remembered that there is
removed, viscosity increases and there is a ten- not always a linear correlation between dose of
dency toward filter clotting. Passage of blood heparin or degree of anticoagulation and filter
fi life.
62 S. Blakeley
Dialysis related
Hypotension Hypovolemia secondary to total volume removal or speed of removal, i.e., not allowing body
compartments to equilibrate (commonest) t
High pump speeds may be enough to precipitate hypotension in unstable patients
Life-threatening anaphylactoid hypersensitivity reactions have been described with the use of certain
membranes (e.g., polyacrylonitrile membranes, such as AN69) with concurrent ACEI therapy
Activation of inflammatory and vasodilatory mediators (e.g., bradykinin) related to membrane
bioincompatibility
Anticoagulation-related complications Local or systemic bleeding
Related to specific type of anticoagulant: e.g., HITS (heparin) and hypocalcemia (RCA)
Electrolyte disturbances Including hypokalemia, hypophosphatemia, and hypoglycemia
Acid-base disturbances Metabolic acidosis related to lactate buffer in replacement fluid if unable to handle a large exogenous
lactate load (e.g., liver failure, septic shock)
Metabolic alkalosis related to RCA
Temperature disturbances A degree of cooling always occurs, this may lead to “normothermia” in febrile patients (i.e., masking a
pyrexia) or cause marked hypothermia
Vitamin and micronutrient depletion Water-soluble vitamins, trace minerals, certain hormones (e.g., glucocorticoids), amino acids
Inappropriate prescribing of drugs Generally leads to underdosing while a patient is on the filter
Further renal injury Systemic hypotension (see above) reducing already compromised renal perfusion
Because of release of inflammatory mediators triggered by blood coming into contact with the filter
and tubing
a
ACEI, angiotensin converting enzyme inhibitor.
11. Technical Aspects of Renal Replacement Therapy 63
Prostacyclin ery of acute renal failure. Cur Opin Crit Care. 2005;
11: 548–554.
Prostaglandin (PG)-I2 is a natural anticoagulant 5. Ricci Z, Ronco C, Bachetoni A, et al. Solute removal
that is a potent antiplatelet agent and has been during continuous renal replacement therapy in
shown to reduce platelet microthrombi during critically ill patients; convection versus diffusion.
dialysis. It is often used in patients with a high risk Crit Care. 2006; 10: R67–R74.
of bleeding, but because it is a potent arterial 6. Clark WR, Turk JE, Kraus MA, Gao D. Dose deter-
vasodilator, some patients develop symptomatic minants in continuous renal replacement therapy.
hypotension. It has been used on its own and in Artif Organs. 2003; 27: 815–820.
7. Ronco C, Bellomo R, Homal P, et al. Effects of dif-
combination with low-dose heparin.
ferent dose in continuous veno-venous haemofiltra-fi
tion on outcomes of acute renal failure: a prospective
randomised trial. Lancet. 2000; 356: 26–30.
Other Anticoagulants 8. Bouman C, et al. Effects of early high-volume con-
Heparinoids (e.g., danaparoid) have minimal tinuous Venovenous hemofiltratin
fi of survival and
effects on platelets and can be used in HITS (but recovery of renal function in intensive care patients
remembering the potential cross reactivity in with acute renal failure: a prospective randomized
5–10% of patients). However, standard markers of trial. Crit Care Med. 2000; 30: 2205–2211.
9. Saudan P, et al. Adding a dialysis dose to continuous
anticoagulation are not reliable and its effect is
hemofi filtration increases survival in patients with
prolonged in renal failure. Factor Xa inhibitors acute renal failure. Kidney Int. 2006; 70: 1312–
(e.g., fondaparinux) and direct thrombin inhibi- 1317.
tors (e.g., recombinant hirudin) can be safely used 10. Cariou A, Vinsonneau C, Dhainaut JF. Adjunctive
in HITS, but, to date, have limited use in CRRT. therapies in sepsis: an evidence-based review. Crit
Care Med. 2004; 32: S562–S570.
11. www.adqi.net
Comment 12. Ronco C. Renal replacement therapy for acute
kidney injury: let’s follow the evidence. Int J Artif
A recent systematic review (16) found that there
Organs. 2007; 30: 89–94.
was no conclusive evidence to suggest one strat- 13. Naka T, Bellomo R. Bench-to-bedside review: treat-
egy over another, but the chosen method should ing acid-base abnormalities in the intensive care
take into account patient characteristics and local unit–the role of renal replacement therapy. Crit
facilities. Heparin (UFN and LMWH) has the Care. 2004; 8: 108–114.
greatest evidence and experience behind it, but 14. Hilton PJ, Taylor J, Forni LG, Treacher DF. Bicarbo-
RCA is increasing in popularity and ease of use. nate-based haemofiltration
fi in the management of
acute renal failure with lactic acidosis. QJM. 1998; 4:
279–283.
References 15. Teehan GS, Liangos O, Lau J, et al. Dialysis mem-
1. Kramer P, Wigger W, Rieger J, et al. Arteriovenous brane and modality in acute renal failure: under-
haemofi filtration: a new and simple method for treat- standing discordant meta-analyses. Sem Dialysis.
ment of over-hydrated patients resistant to diuret- 2003; 16: 356–360.
ics. Klin Wochenschr. 1977; 55: 1121–1122. 16. Oudemans-van Straaten HM, Wester JPJ, de Pont
2. Palevsky PM. Dialysis Modality and Dosing Strat- ACJM, Schetz MRC. Anticoagulation strategies in
egy in Acute Renal Failure. Sem Dialysis. 2006; 19: continuous renal replacement therapy: can the
165–170. choice be evidence based? Intensive Care Med. 2006;
3. Van Biesen W, Vanholder R, Lameire N. Dialysis 32: 188–202.
strategies in critically ill acute renal failure patients.
Curr Opin Crit Care. 2003; 9: 491–495. Suggested Reading
4. Palvesky PM, Baldwin I, Davenport A, et al. Renal
replacement therapy and the kidney: minimising Bellomo R, Baldwin I, Ronco C, Golper T. Atlas of Hemo-
the impact of renal replacement therapy on recov- fi
filtration. WB Saunders. 2002.
12
End-Stage Renal Disease
Emile Mohammed
64
12. End-Stage Renal Disease 65
Semi
em
mi-pe
-pe
-perme
perm
rmeab
rmeablle
memb
em
mbbran
ranee
Glucose
Capillary
3
“Bag in”
2.5
PDF volume (litres)
1.5
0.5
0
Bag drained out Time (during day)
2.5
“Bag in”
PDF volume (litres)
1.5
0.5
0
wet day overnight exchanges FIGURE 12.3. PD regimes.
12. End-Stage Renal Disease 67
The following clinical parameters act as guide- • Anaphylaxis. Anaphylaxis can occur by comple-
lines to achieve this: ment activation with the use of a bioincompati-
ble membrane and normally occurs within the
• Target weight and blood pressure control. Target first 20 minutes of treatment.
weight is defifined as the patient’s weight in which • Catheter related-sepsis. Catheter related-sepsis
all the fluid compartments are physiologically requires aggressive antibiotic treatment and
normal. Excess weight (which will be essentially catheter removal. If temporary catheters are
salt and water) results in hypertension. The being used, once weekly catheter changes are
target weight is achieved by gradual weight recommended.
reduction on successive dialyses until the • Pyrogenic reactions. Uncommon if ultrapure
patient is free from both pulmonary and peri- water is used.
pheral edema, but, below which, hypotension • Dialysis equilibrium syndrome. Rare in estab-
occurs. lished dialysis patients. It can occur from overag-
• Acid-base balance. Dialysis must be performed gressive dialysis causing a rapid reduction in
frequently and long enough to maintain normal serum osmolality and resulting in cerebral
acid-base balance. edema.
• Bone biochemistry. Along with vitamin D sup- • Modern fail-safe machines minimize other com-
plementation, serum calcium and phosphate plications such as air embolism and accidental
levels should be maintained within normal circuit disconnection.
limits.
• Nutritional state. It is important to remember
that a high proportion of ESRD patients within Peritoneal Dialysis
the ICU will have a low serum albumin, low Although PD is a technically safe procedure, there
body mass index, an infl flammatory and/or may be clinical reasons to convert to temporary
hypercatabolic state, and a low dietary intake. It HD.
is, therefore, necessary to obtain dietary advice, These are as follows:
treat correctable factors, give dietary supple-
ments, and have a low threshold for nasogastric • Abdominal surgery
(NG), percutaneous endoscopic gastrostomy • Diaphragmatic fluid leak resulting in effusions
(PEG), or even parenteral nutrition, if indicated. • Respiratory compromise from splinting of dia-
Dialysis prescriptions must accommodate these phragm by PDF
nutritional requirements. • Severe hypoalbuminemic state
• Peritonitis or catheter-related sepsis
A good starting point for prescribing dialysis • Inadequate ultrafiltration
fi in the context of
would be to continue the patient’s regular dialysis aggressive flfluid management and/or hypercata-
regime and adjust the dose of dialysis, in conjunc- bolic state of the patient
tion with the nephrologists, to achieve the above
parameters.
Renal Transplantation
Dialysis-Related Complications Renal transplantation represents the best mode of
therapy for ESRD patients, both in cost effective-
Hemodialysis ness and quality of life (3). There have been many
• Hypotension. Hypotension can be minimized improvements in renal transplantation, such as the
with an accurate assessment of target weight, refi
finement of immunosuppression regimens, and
judicious use of antihypertensive medications, patient-donor selection and work-up as well as
sodium restriction, increasing treatment dura- their compatibilities. The major challenge facing
tion, and careful choice of dialysis modality, e.g., transplantation is that its demand far outstrips the
hemodiafiltration
fi in the cardiovascularly unsta- availability. Every effort should be made to increase
ble patient. the number of donors. In parallel to this, there is
68 E. Mohammed
much research in the development of stem cell allografts. Such sensitization can cause severe
transplantation and xenotransplantation. hyperacute rejection.
Other cancers that should be screened for include membrane flux in maintenance hemodialysis. N Engl
renal, cervical, and vaginal cancers. J Med 2002; 347(25): 2010–2019.
2. Paniagua R, Amato D, Vonesh E, et al. Health-related
quality of life predicts outcomes but is not affected
by peritoneal clearance: The ADEMEX trial. Kidney
Summary Intt 2005; 67(3): 1093–1104.
3. The EBPG Expert Group on Renal Transplantation.
ESRD and its complications are becoming more European Best Practice Guidelines for Renal Trans-
commonplace, particularly in the ICU environ- plantation (Part 1). Nephrol Dial Transplantt 2000;
ment. The challenges associated with this group 15(Supp 7): 1–85.
of patients also continue to rise and requires a
multidisciplinary approach, which includes the
intensivist and the nephrologist. Suggested Reading
Davison AM, Cameron S, Grunfeld J-P, et al. Oxford
References Textbook of Clinical Nephrology. 3rd edition. Oxford:
Oxford University Press. 2005.
1. Eknoyan G, Beck GJ, Cheung AK, et al. Hemodialysis Levy J, Brown E, Morgan J. Oxford Handbook of Dialy-
(HEMO) Study Group. Effect of dialysis dose and sis. Oxford: Oxford University Press. 2001.
13
Clinical Hyperkalemia and Hypokalemia
Harn-Yih Ong
71
72 H.-Y. Ong
the patient) and can produce a 1 to 2 mmol/L arti- hyperglycemia is effective and rapid in onset of
factual increase in potassium levels. The major action. For example, 10 U intravenous insulin
causes are: can be expected to lower the plasma potassium
concentration by 0.5 to 1.5 mmol/L within 15
• Contamination by drip site or laboratory error
minutes, lasting 2 to 4 hours.
• Hemolysis caused by mechanical trauma during
• Sodium bicarbonate: Increasing the pH of the
venipuncture
ECF compartment with sodium bicarbonate
• Thrombocytosis (platelet count >900 × 1010
(e.g., 100 mmol over 1–2 hours) will shift potas-
cells/L)
sium into cells as an acidosis is corrected. Its
• Leukocytosis (white cell count >70 × 109 cells/L)
effectiveness is debated, but it is still recom-
or mononucleosis (“leaky red blood cells”)
mended when hyperkalemia is associated with
• Familial pseudohyperkalemia or other uncom-
severe inorganic metabolic acidosis. Ensur-
mon genetic syndromes
ing effective diuretic therapy fi
first lessens the
likelihood of developing volume overload as a
complication.
Management • Salbutamol: Meter-dose inhalers, nebulized, or
intravenous salbutamol seem equally effective
Severe hyperkalemia (plasma potassium
in reducing potassium levels.
>6.5 mmol/L or with ECG changes) is a medical
emergency and therapy should begin
immediately. Removal of Excess Potassium
Treatments that shift potassium into the cells have
Stabilization of Cardiac Membranes: no effect on total body potassium. Excess total
Intravenous Calcium body potassium may be removed by gastrointes-
tinal elimination, renal excretion, or renal replace-
Although no randomized studies exist to support ment therapy.
the use of calcium salts administered intrave-
nously (e.g., 10 mL 10% calcium gluconate over • Renal excretion of potassium may be enhanced
2–3 min into a large vein), it is still recommended with the use of diuretics, especially loop diuret-
as first-line therapy in the presence of ECG changes ics, which increase the delivery of sodium and
or arrhythmia. The protective effect of calcium is the urine flow rate to the cortical collecting duct,
evident within minutes and lasts for 30 to 60 increasing tubular secretion of potassium. If the
minutes. Caution should be used in patients who patient is volume depleted, hydration with saline
take digoxin because calcium has been reported can be administered with the diuretic.
to worsen the myocardial effects of digoxin toxic- • Although widely used clinically in the treatment
ity. An alternative is to consider using magnesium of hyperkalemia, cation exchange resins do not
instead of calcium to stabilize the myocardium. seem to be effective at 4 hours and should not
Calcium may reduce the immediate risk of cardiac be relied on for rapid effects.
arrest, but represents a temporizing measure only. • Hemodialysis or continuous renal replacement
Plasma potassium concentration is unaltered. therapies are the treatments of last resort, with
the exception of patients already receiving these
therapies.
Transcellular Redistribution
• Insulin and dextrose: Administration of insulin
always causes a transient reduction in plasma Hypokalemia
potassium because the activated insulin recep-
tor stimulates Na+/K+-ATPase, driving cellular Hypokalemia is usually defined
fi as a plasma potas-
uptake of potassium. The use of insulin and sium concentration of less than 3.5 mmol/L, and
glucose (e.g., 50 mL of 50% glucose to prevent refers to a deficit
fi of potassium ions in the ECF
hypoglycemia) for the emergency treatment of compartment.
74 H.-Y. Ong
Mild Hypokalemia
Worsening of Cardiac Function
Mild hypokalemia (plasma potassium, 3.0–
Chronic hypokalemia exacerbates CCF by impair-
3.5 mmol/L) is often asymptomatic. Skeletal
ing the function and contractility of myocardial
muscle weakness is not usually seen with potas-
muscle. This reduces stroke volume and cardiac
sium greater than 3.0 mmol/L.
output. Cardiac necrosis has been described.
Moderate Hypokalemia
Symptoms of moderate hypokalemia (plasma Renal and Metabolic
potassium, 2.5–3.0 mmol/L) are nonspecific:fi lassi- Ammoniagenesis and Metabolic Alkalosis
tude, fatigue, constipation, weakness in the lower
extremities, and myalgia. A paralytic ileus can Chronic hypokalemia results in ICF acidification,
fi
occur. which stimulates ammoniagenesis in the proximal
tubule. Increased ammonia trapping and net
acid excretion coupled to enhanced bicarbonate
Severe Hypokalemia
production perpetuates metabolic alkalosis.
Ascending paralysis can occur in severe hypoka- Increased ammoniagenesis may be clinically
lemia (plasma potassium, ≤2.0–2.5 mmol/L). The important in patients with severe liver disease,
lower extremities are typically involved fi
first (par- in whom hypokalemia may precipitate hepatic
ticularly the quadriceps), followed by the trunk encephalopathy.
76 H.-Y. Ong
77
78 H. Gangopadhyay
Treatment
Classification (Table 14.3)
Much has been mentioned regarding treatment of
hyponatremia and the potential complication of 1. Decreased total body sodium (extracellular
central pontine myelinolysis (CPM). This is a water and sodium loss with excess water
potentially fatal condition that can occur when loss)
the serum sodium is corrected too quickly. CPM 2. Normal total body sodium (extracellular water
can lead to mutism, dysphasia, spastic quadriple- deficiency
fi associated with minimal sodium
gia, pseudobulbar palsy, delirium coma, and even loss)
death. Risk factors for its development are alco- 3. Increased total body sodium (extracellular
holism, burns, use of thiazide diuretics, a mal- water and sodium gain with relatively excess
nourished state, and hypokalemia. sodium gain)
80 H. Gangopadhyay
Acid-base disturbances are common on the inten- if one or more of these three variables changes,
sive care unit (ICU) and should always prompt a rather than bicarbonate itself being the main deter-
search for the underlying cause (Tables 15.1–15.3). minant of change, as in the traditional approach.
Recently another, although not all that new (1), “Strong ions” are ions that exist completely dis-
way of understanding the mechanism behind sociated (charged) in water. There are more strong
acid-base disorders has gained popularity. This cations than anions and this difference is called
chapter starts with just a brief discussion of this the SID. SID has an electrochemical effect on water
approach, because there are several excellent over- dissociation; as SID becomes more negative, H+
views in the literature (2–6). increases and pH falls.
“Weak ions” exist as dissociated (A−) or associ-
ated forms (HA), and together they make up ATOT.
Traditional Versus Physiochemical Nonvolatile weak acids, mainly albumin but also
Approach to Acid-Base Disturbances inorganic phosphate and bicarbonate, dissociate
in body fluids forming weak anions.
Most “traditional” views of acid-base balance rely To maintain electrical neutrality, the SID must
on the Henderson-Hesselbach equation, but there be balanced by an equal and opposite charge,
has always been debate regarding quantification
fi the effective SID (SIDe). The SIDe is made up of
of the metabolic component. It can be described predominately weak acids (A−) and CO2 (Equation
using plasma bicarbonate and anion gap (AG) or 15.1). If the SID is calculatedd from the measured
by using standard base excess (SBE). However, plasma strong ion concentrations (Equation 15.2),
SBE is affected by changes in sodium, chloride, it is termed the apparent SID (SIDa). However, this
and albumin, and a significant
fi number of abnor- will not be precise if there are unmeasured anions
mal AGs will be missed unless the effect of albumin or cations present. If the SIDe and the SIDa do not
is considered (7–9). The third way of quantifying match (strong ion gap [SIG]) then unmeasured
the metabolic component is the strong ion differ- ions mustt be present. SIG should be zero, although,
ence (SID) (1), taking into consideration nonbi- in practice, there is a degree of normal variability.
carbonate buffers, such as inorganic phosphate If SIG is positive, there must be more unmeasured
and serum albumin, both of which may be abnor- strong anions (e.g., ketones, sulfates), and if it is
mal in critically ill patients. negative, more unmeasured cations.
Stewart’s (physiochemical) approach relies on
Equation 15.1:
the principles of physical chemistry; electroneu-
SIDe = A− + HCO3−
trality, conservation of mass, and dissociation of
electrolytes (1). It considers three independent Equation 15.2:
variables for pH: the SID, pCO2 and total weak acid SIDa = ([Na+] + [K+] + [Ca2+] + [Mg2+]) − ([Cl−] +
concentration (ATOT). H+ or HCO3− can only change [lactate])
81
82 S. Blakeley
Note: the effects of severe alkalemia (pH > 7.60) appear more pronounced with respiratory rather than metabolic alkalosis.
a benefifit on outcome in patients with lactic acido- cacy of renal replacement therapy (RRT) tech-
sis, but it is diffi
ficult to separate the effects of the niques, therefore, the acidosis can be cleared
acidosis from the effects of the underlying patho- within 48 hours (16). Despite the fact that lactate
logical condition. has a low molecular weight and it is easily removed
Sodium bicarbonate is not without risks, by continuous therapies, RRT has been shown to
including: contribute to only less than 3% of total lactate
clearance (17). It may be that RRT plays a role in
• Acute fluid overload. improving the overall clinical situation (e.g.,
• Alkaline “overshoot” caused by subsequent hemodynamic and fluid fl status), therefore, pro-
metabolism of organic salts increasing the bicar- moting the clearance of lactate, rather than actu-
bonate concentration. ally removing it.
• Paradoxical intracellular acidosis. CO2 released
diffuses more quickly than bicarbonate ions into
cells, therefore, initially lowering the intracellu-
lar pH. There is debate regarding whether this is Metabolic Alkalosis
clinically significant.
fi
Metabolic alkalosis is common, but carries an
associated mortality that increases with increas-
Use of Renal Replacement Therapy ing pH, especially pH higher than 7.55 (18). The
In acute renal failure, the daily production of cause, rather than the alkalosis itself, may be the
mineral acids is adequately matched by the effi-
fi reason for the increased mortality.
15. Clinical Metabolic Acidosis and Alkalosis 85
13. Bellomo R, Ronco C. The pathogenesis of lactic aci- 16. Levraut J, Grimaud D. Treatment of metabolic
dosis in sepsis. Curr Opin Crit Care. 1999; 5: 452–457. acidosis. Curr Opin Crit Care. 2003; 9: 260–
14. Stacpoole PW, Wright EC, Baumgartner TG, et al. for 265.
the DCA-Lactic acidosis Study Group: Natural 17. Levraut J, Ciebiera JP, Jambou P, et al. Effect of con-
history and course of acquired lactic acidosis in tinuous venovenous hemofiltration
fi with dialysis on
adults. Am J Med. 1994; 97: 47–54. lactate clearance in critically ill patients. Crit Care
15. Husain FA, Martin MJ, Mullenix PS, et al. Serum Med. 1997; 25(1): 58–62.
lactate and base defi ficit as predictors of mortality 18. Galla JH. Metabolic alkalosis. J Am Soc Nephrol.
and morbidity. Am J Surg. 2003; 185(5): 485–491. 2000; 11: 369–375.
Index
87
88 Index
D F Hypocalcemia, 39, 50
Daclizumab, 69 Familial hypokalemic periodic Hypokalemia, 50, 74–76, 82, 85,
Dextrose, 17, 73 paralysis, 74 8575, 73076
Diabetes insipidus, nephrogenic, 76 Fasciotomy, 38 Hypomagnesemia, 75
Diabetes mellitus, 6, 33, 34 Fluid composition, of the body, 1, Hyponatremia, 22, 77–79, 79
Dialysis 2 Hypoperfusion, renal, 9, 20.6
optimal dose in, 65–67 Fluid resuscitation, in acute renal Hypotension
peritoneal, 64–67 failure patients, 28–29 acute renal failure-related, 26, 29
principles of, 64 anesthesia-related, 33
as renal vasculitis treatment, 44 G dialysis-related, 67
for urinary myoglobin Glomerular filtration,
fi cessation of, as postoperative renal failure
reduction, 39–40 33 cause, 33
Dialysis equilibrium syndrome, 67 Glomerular filtration
fi rate (GFR), 2 prerenal failure-related, 20
Diethylene triamine pentaacetic in acute kidney injury, 19 relative, 36
acid (DTPA) studies, 7, 10 assessment of, 3 in surgical patients, 33
Digoxin, toxicity of, 73 in metabolic alkalosis, 85 therapeutic plasma exchange-
1,25-Dihydroxyvitamin D3, 1 Glomerulonephritis, 6, 9, 26 related, 50
Disseminated intravascular Glucose control, tight, 29, 36 Hypovolemia, 21, 49, 85
coagulation, 46–47 Glycosuria, renal, 5
Diuretics Goodpasture’s syndrome, 9 I
as acute renal failure Imaging, of acute renal failure,
prophylaxis, 34, 36 H 7–13. See also Computed
as acute renal failure treatment, Hematuria, 4, 26, 42 tomography (CT); Magnetic
36 Hemodiafiltration,
fi 64 resonance angiography
loop, 15, 17, 27–28 continuous venovenous, 59 (MRA); Magnetic resonance
nephrotoxicity of, 16 Hemodialysis, 64 imaging (MRI);
Dopamine, as renal failure complications of, 67 Scintigraphic imaging
prophylaxis, 36 continuous, 57–63 diagnostic algorithm for, 12
Dopamine agonists, 28 as hyperkalemia treatment, 74 Immunosuppression, renal failure-
Drugs. See also names of specificfi intermittent, 52, 54, 60 related, 22
drugs Hemofi filters, 50 Immunosuppressive therapy, in
as hypokalemia cause, 74 Hemofi filtration, 52, 64 renal transplant recipients,
impaired renal clearance of, 23–24 continuous venovenous, 52, 57, 68–69
nephrotoxicity of, 14–18, 23, 59 Infection
33–34, 38, 39 high-volume, 53 in renal transplant recipients, 69
as rhabdomyolysis cause, 38, 39 prophylactic, 29 as rhabdomyolysis cause, 38
for urinary myoglobin Inotrope therapy, for acute renal
E reduction, 39–40 failure, 29
Edema Hemofi filtration machines, 58, Insulin therapy, 36, 73
pulmonary, 43 59–60 Intensive care unit (ICU) patients
renal, 9 Hemolytic uremic syndrome, 46, acute kidney injury in, 19
Electrocardiography 47 acute renal failure in, 26
in hyperkalemia, 71–72 Hemorrhage, gastrointestinal, 23 intrinsic renal failure in, 20, 21
in hypokalemia, 75 Henoch-Schönlein purpura, 43 obstructive renal failure in, 22
Electrolyte disturbances. See also Heparin, 61–62 Inulin clearance, 3
specific
fi electrolyte Homeostasis, role of the kidney in, Ischemia, renal, 33
disturbances 1
acute kidney injury-related, 22 Hypercalcemia, 24 K
Encephalopathy, uremic, 22 Hyperglycemia, 22 Kidney
Enteral support, in critically ill Hyperkalemia, 16, 22, 71–73 hydronephrotic, 9
patients, 29 Hypernatremia, 79–80 normal, ultrasound imaging of,
Eosinophilia, 24 Hyperphosphatemia, 22, 24, 39 7, 8
Erythropoietin, 1, 22 Hypertension, 13, 42 normal functions of, 1
Excretion, urinary, 1 malignant, 47 polycystic, 7
Index 89