Competency-Based Critical Care

Series Editors

John Knighton, MBBS, MRCP, FRCA Paul Sadler, MBChB, FRCAConsultant ConsultantIntensive Care Medicine & Anaesthesia Critical Care Medicine & AnaesthesiaPortsmouth Hospitals NHS Trust Queen Alexandra HospitalPortsmouth PortsmouthUK UK

Founding Editor

John SP LumleyEmeritus Professor of Vascular SurgeryUniversity of LondonLondonUK

and

Honorary Consultant SurgeonGreat Ormond Street Hospital for Children NHS Trust (GOSH)LondonUK

Other titles in this series

SepsisSimon Baudouin (Ed.)

Sara Blakeley (Ed.)

Renal Failure andReplacement Therapies

Sara Blakeley, BM, MRCP, EDICQueen Alexandra HospitalPortsmouthHampshire, UK

British Library Cataloguing in Publication DataRenal Failure and Replacement Therapies.—(Competency-based critical care) 1. Kidneys—Diseases 2. Kidneys—Diseases—Treatment 3. Acute renal failure 4. Acute renal failure—Treatment I. Blakeley, Sara 616.6′1 ISBN-13: 9781846289361

Library of Congress Control Number: 2007927934

Competency-Based Critical Care Series ISSN 1864-9998

ISBN: 978-1-84628-936-1 e-ISBN: 978-1-84628-937-8

© Springer-Verlag London Limited 2008

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 writingof 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 statement, that such names are exempt from the relevant laws and regulations and fitherefore 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 checkits accuracy by consulting other pharmaceutical literature.

9 8 7 6 5 4 3 2 1

springer.com

v

Contents

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

Chapter 1 Assessment of Renal Function . . . . . . . . . . . . . . . . . . . . . 1Mohan Arkanath

Chapter 2 Imaging of Acute Renal Failure—A Problem-SolvingApproach for Intensive Care Unit Physicians . . . . . . . . 7Tom Sutherland

Chapter 3 Drug-Induced Renal Injury . . . . . . . . . . . . . . . . . . . . . . . . 14Sara Blakeley

Chapter 4 Acute Kidney Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19Sara Blakeley

Chapter 5 Medical Management of Acute Renal Failure . . . . . . . . 26Nerina Harley

Chapter 6 Acute Renal Failure in the Surgical Patient . . . . . . . . . . 33Marlies Ostermann

Chapter 7 Rhabdomyolysis and Compartment Syndrome . . . . . . . 38Laurie Tomlinson and Stephen Holt

Chapter 8 Multisystem Causes of Acute Renal Failure . . . . . . . . . . 42Tim Leach

Chapter 9 Therapeutic Plasma Exchange . . . . . . . . . . . . . . . . . . . . . 49Tim Leach

Chapter 10 Renal Replacement Therapy . . . . . . . . . . . . . . . . . . . . . . . 51John H. Reeves

Chapter 11 Technical Aspects of Renal Replacement Therapy . . . . 57Sara Blakeley

vi Contents

Chapter 12 End-Stage Renal Disease . . . . . . . . . . . . . . . . . . . . . . . . . . 64Emile Mohammed

Chapter 13 Clinical Hyperkalemia and Hypokalemia . . . . . . . . . . . . 71Harn-Yih Ong

Chapter 14 Clinical Hyponatremia and Hypernatremia . . . . . . . . . . 77Himangsu Gangopadhyay

Chapter 15 Clinical Metabolic Acidosis and Alkalosis . . . . . . . . . . . 81Sara Blakeley

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

Contributors

Mohan Arkanath, MRCPConsultant NephrologistDoncaster Royal Infi rmaryfiDoncaster, UK

Sara Blakeley, BM, MRCP, EDICQueen Alexandra HospitalPortsmouthHampshire, UK

Nerina Harley, MBBS, MD, PGDIPEcho,FRACP, FJFICM

Intensive Care ConsultantThe Royal Melbourne HospitalVictoria, Australia

Stephen Holt, PhD, FRCPConsultant Nephrologist and Honorary

Senior LecturerSussex Kidney UnitBrighton and Sussex University HospitalsRoyal Sussex County HospitalBrighton, UK

Himangsu Gangopadhyay, MD, MBBS,FFARCS(Ireland), FJFICM

Consultant in Intensive CareFrankston HospitalFrankston, Victoria, Australia

Tim Leach, BM, FRCPConsultant NephrologistWessex Renal and Transplant UnitQueen Alexandra HospitalPortsmouth, UK

Emile Mohammed, MB, ChB, MRCP (UK)Lecturer in Medicine (University of the West

Indies) and Consultant Nephrologist

vii

School of Clinical Medicine and ResearchQueen Elizabeth Hospital & Cavehill CampusBridgetown, Barbados

Harn-Yih OngRegistrar Intensive CareSt Vincent’s HospitalMelbourne, VictoriaAustralia

Marlies OstermannConsultant Nephrology and Critical CareIntensive Care UnitSt Thomas’ HospitalLondon, UK

John H. Reeves, MD, MBBS, FANZCA, FJFICM,EDIC

Consultant in Intensive Care and AnaestheticsDepartment of Anaesthesia and Pain

ManagementAlfred HospitalMelbourne, VictoriaAustralia

Tom Sutherland, MBBS (Hons)Radiology RegistrarSt Vincent’s HospitalMelbourne, VictoriaAustralia

Laurie Tomlinson, MRCPSpecialist Registrar NephrologySussex Kidney UnitBrighton and Sussex University HospitalsRoyal Sussex County HospitalBrighton, UK

1

1Assessment of Renal Function

Mohan Arkanath

sure are dependent on optimum plasma volume,and most enzymes function best in narrow ranges of pH or ion concentration.

The kidneys take up the role of correcting any alterations in the composition and volume of body fluids that occur as a consequence of food intake, flmetabolism, environmental factors, and exercise.In healthy individuals, these corrections occur ina matter of hours, and body fluid volume and the flconcentration of most ions return to normal set points. In many disease states, however, these regulatory processes are disturbed, resulting inpersistent deviations in body fluid volume and flcomposition.

Body Fluid Composition

A large proportion of the human body is com-posed of water (Table 1.1). Adipose tissue is low inwater content and, hence, obese individuals have a lower body water fraction than lean individuals. Because of slightly greater fat content, women generally contain less water than men.

Clinical Evaluation of Renal Function

Glomerular filtration rate (GFR) is generally con-fisidered the best measure of renal function; it is the sum of filtration rates of all of the functioning finephrons. It is defined as the renal clearance of a fiparticular substance from plasma, and is expressed as the volume of plasma that can be completely cleared of that substance in a unit of time. In the following sections, we will compare the advantages

Normal Functions of the Kidney

To be able to assess a degree of renal function or dysfunction, it is important to first consider the finormal functions of the kidney:

1. Maintenance of body composition: The volume of fluid in the body, its osmolarity, elec-fltrolyte content, concentration, and acidity are allregulated by the kidney via variation in urine excretion of water and ions. Electrolytes regulatedby changes in urinary excretion include sodium,potassium, chloride, calcium, magnesium, and phosphate.

2. Excretion of metabolic end products and foreign substances: The most notable are urea and a number of toxins and drugs.

3. Production and secretion of enzymes and hormones: a. Renin: See Figure 1.1. b. Erythropoietin: A glycosylated, 165-amino

acid protein produced by renal corticalinterstitial cells that stimulates maturation of erythrocytes in the bone marrow.

c. 1, 25-Dihydroxyvitamin D3: The most active form of vitamin D3, it is formed by proximal tubule cells. This steroid hormone plays animportant role in the regulation of body calcium and phosphate balance.

The Role of the Kidney in Homeostasis

Maintenance of body fluid composition and flvolume are important for many functions of the body. For example, cardiac output and blood pres-

2 M. Arkanath

and disadvantages of the various methods avail-able for GFR estimation or quantification.fi

Clearance Urine mg/dL Volume mL/minPlasma mg/d( )mL/min

=)mL/min

)

Where x is the substance being cleared.Features of an ideal marker for GFR

determination:

• Appears endogenously in the plasma at a con-stant rate

• Is freely fi ltered by the glomerulusfi• Does not undergo reabsorption or secretion by

the renal tubule• Is not eliminated by extrarenal routes

Urea

Urea was fi rst isolated in 1773, and urea clearancefias a surrogate marker for GFR introduced in 1929.Although urea measurement is performed fre-quently, it is well recognized that its many draw-backs make it a poor measure of renal function(see Table 1.2). For example, the rate of production is influenced by factors such as the availability flof nitrogenous substrates, and the rate of reab-sorption in the tubules can be affected by volumestatus.

Creatinine

Creatinine is a metabolite of creatine and phos-phocreatine found in skeletal muscle. It is a small molecule that is not protein bound and, hence, freely fi ltered by the glomerulus. It does, fihowever, undergo tubular secretion which is variable (see Table 1.2). Creatinine productioncan vary in an individual over time withmuscle mass changes or acutely with massivemyocyte turnover. There are also age- and sex-associated differences in serum creatinine (Sα) concentration; it is lower in the elderly and inwomen.

The ratio between serum urea and creatinine can be useful when assessing the patient with acute renal failure. Under normal circumstances, the ratio between urea and creatinine is 10 : 1 but this value can rise to greater than 20 : 1 when theextracellular volume is contracted. A volume-con-tracted state (prerenal( ) is a “sodium avid” state and promotes proximal tubular and distal nephronreabsorption of urea but not creatinine. Acute

Renin

Angiotensinogen(Plasma globulin)

Angiotensin(Vasoconstrictor)

Salt balanceand blood pressurecontrol

FIGURE 1.1. Production and secretion of enzymes and hormonesfor renin.

TABLE 1.1. Body fluid compartment volumesa

Example for a 60-kg patient

TBW = 60% × body weight 60% × 60 kg = 36 LICW = 2/2// TBW 2/2 3// × 36 L = 24 LECW = 1/1// TBW 1/1// × 36 L = 12 LPlasma water = 1/1// ECW 1/1 4// × 12 L = 3 LBlood volume = Plasma water ÷ 3 L ÷ (1 − 0.40) = 6.6 L (1 − hematocrit)

aTBW, total body water; ICW, intracellular water; ECW, extracellular water.

TABLE 1.2. Other factors altering blood urea or serum creatininea

Increased level Decreased level

Urea Prerenal causes (e.g., Cirrhosis congestive heart failure, Protein malnutrition volume contraction) lead Water excess (e.g., to increased tubular SIADH, saline reabsorption infusion) leads to • Gastrointestinal bleeding reduced tubular • Catabolic state reabsorption • Corticosteroids • Hyperalimentation • TetracyclinesCreatinine Overproduction (e.g., Decreased muscle rhabdomyolysis, vigorous mass sustained exercise, anabolic steroids, dietary supplements such as creatine) Blocked tubular secretion (e.g., drugs such as trimethoprim and cimetidine) Assay interference (e.g., ketosis and drugs such as cephalosporins, flucytosine, methyldopa, levodopa, and ascorbic acid)

aSIADH, syndrome of inappropriate ADH secretion.

1. Assessment of Renal Function 3

tubular necrosis (ATN) will have a urea-to-creati-nine ratio of 10 : 1 because tubular reabsorption of urea is not preferentially increased.

Assessment of GFR

Sα has become a standard measure of renal func-tion because of its convenience and low costs; however, it is crude marker of GFR. A 24-hourcreatinine clearance (Cα) is often used in practice as a measure of the GFR because creatinine is freely filtered and not reabsorbed by the tubule.fiHowever, approximately 15% of urinary creati-nine is a result of tubular secretion and, thus, this method overestimates GFR. The other problemswith Cα are incomplete urine collection andincreasing creatinine secretion. Inulin clearance is traditionally considered the “gold standard” for the measurement of GFR (1). It is one of the most accurate measures of renal function, but the inconvenience of administration, cost, andlimited supply of inulin preclude its use in routine practice.

Estimated Ca: The Cockroft and Gault

Formula (2)

This formula is used to estimate Cα at the bedside using age of the patient and Sα value (both corre-late inversely with GFR), and the ideal body weight (IBW) of the patient.

This formula can used only when the Sα value is in a steady state and not when it is rapidly changing, as in acute renal failure. It has also been shown that Sα and estimations of GFR using various formulae are often inaccurate in critically ill patients (3). Although they may give an esti-mate, their limitations should be rememberedwhen assessing renal function in patients on the intensive care unit. Other factors, such as urine output, clearance of acid and electrolytes, and rate of change of serum urea and creatinine should allbe considered together.

The Cockroft and Gault formula is:

Estimated Cpatient age weight in kg

S in mol/Lα

α=

−( )( )

140

Multiply by 0.85 for women, 1.23 if male.

Urinalysis (4, 5)

The microscopic examination of the urinary sedi-ment is an indispensable part of the work-up of patients with renal insuffi ciency, proteinuria, hema-fituria, urinary tract infections, or kidney stones. Acareful urinalysis has been referred to as a “poorman’s renal biopsy.” The urine should be collectedas a midstream catch or fresh catheter specimen,and because the urine sediment can degeneratewith time, it should be examined soon after collec-tion. Urinalysis should include a dipstick examina-tion for specifi c gravity, pH, protein, hemoglobin, figlucose, ketones, nitrites, and leucocytes. Thisshould be followed by microscopic examination if there are positive findings. Microscopic examina-fition should check for all formed elements: crystals,cells, casts, and infecting organisms.

Appearance

Normal urine is clear, with a faint yellow tinge caused by the presence of urochromes. As it be-comes more concentrated, its color deepens. Biliru-bin, other pathologic metabolites, and a variety of drugs may discolor the urine or change its smell.

Specific Gravity

The urine specific gravity is a conveniently deter-fimined but inaccurate surrogate of osmolality. Spe-cific gravities of 1.001 to 1.035 correspond to anfiosmolality range of 50 to 1000 mOsm/kg. A spe-cific gravity near 1.010 connotes isosthenuria fi(urine osmolality matching plasma).

The specific gravity is used to determine fiwhether the urine is, or can be, concentrated. During a solute diuresis accompanying hypergly-cemia, diuretic therapy, or relief of obstruction, the urine is isosthenuric. In contrast, with a waterdiuresis caused by overhydration or diabetes insipidus, the specifi c gravity is low. It is also useful fiin differentiating between prerenal cause of renalfailure (high) and ATN.

Volume

In health, the volume of urine passed is primarily determined by diet and fluid intake. The mini-flmum amount passed to stay in fluid balance is fl

4 M. Arkanath

determined by the amount of solute being excreted(mainly urea and electrolytes), and the concen-trating ability of the kidneys. In disease, impair-ment of concentrating ability requires increasedvolumes of urine to be passed for the same soluteoutput.

• Oliguria is defi ned as the excretion of less than fi300 mL/d of urine, and is often caused by intrin-sic renal disease or obstructive uropathy.

• Anuria suggests urinary tract obstruction until proven otherwise.

• Polyuria is a persistent, large increase in urineoutput, usually associated with nocturia. Poly-uria is a result of excessive intake of water (e.g., compulsive water drinking), increased excretionof solute (hyperglycemia and glycosuria), a defective renal concentrating ability, or failure of production of antidiuretic hormone (ADH).

Chemical Testing

Blood

Hematuria may be macroscopic or microscopic (Table 1.3). Currently used dipstick tests for bloodare very sensitive, being positive if two or morered cells are visible under the high-power fieldfi(HPF) of a light microscope. A disadvantage is thatdipstick testing cannot distinguish between blood and free hemoglobin. A positive dipstick has to befollowed by microscopy of fresh urine to confirm fithe presence of red cells and to exclude rare condi-tions such as hemoglobinuria and myoglobinuria. In female patients, it is essential to enquire whetherthe patient is menstruating.

Abnormal numbers of erythrocytes in the urinemay arise from anywhere from the glomerularcapillaries to the tip of the distal urethra. Dysmor-phic erythrocytes tend strongly to be associatedwith a glomerular source. Abnormal proteinuria along with dysmorphic erythrocytes is a reliable sign of glomerular disease. Urinary tract abnor-malities lead to microscopic or macroscopic hematuria but the erythrocytes exhibit normalmorphology.

Protein

Proteinuria is one the most common signs of renal disease (Table 1.4). Most reagent strips detect a

concentration of 150 mg/L or more in urine. They react primarily to albumin and are insensitive toglobulin or Bence-Jones protein. False-positiveresults are common with iodinated contrast agents; hence, urine testing should be repeatedafter 24 hours. The normal rate of excretion of protein in urine is 80 ± 24 mg/24 h in healthy indi-viduals, but protein excretion rates are somewhathigher in children, adolescents, and in pregnancy.Fever, severe exercise, and the acute infusion of hyperoncotic solutions or certain pressor agents (e.g., angiotensin II or norepinephrine) may transiently cause abnormal protein excretionin normal individuals.

The protein in normal and abnormal urine isderived from three sources:

1. Plasma proteins filtered at the glomerulus andfiescaping proximal tubular reabsorption.

2. Proteins normally secreted by renal tubules.

TABLE 1.3. Causes of hematuriaa

Glomerular disease Mesangial IgA nephropathyThin basement membrane disease Thin basement membrane disease

Mesangial proliferative GN Membranoproliferative GN Crescentic GN Systemic lupus erythematosus Post streptococcal GN Infective endocarditis Alport’s syndromeVascular and Acute hypersensitivity interstitial nephritis tubulointerstitial Tumors (renal cell carcinoma, Wilm’s tumor, disease leukemia) Polycystic kidney disease Malignant hypertension Analgesic nephropathy Diabetes mellitus Obstructive uropathyUrinary tract diseases Carcinoma (renal pelvis, ureter, bladder, urethra, prostate) Calculi Retroperitoneal fibrosis Tuberculosis Cystitis Drugs (e.g., cyclophosphamide) Trauma Benign prostatic hypertrophy Urethritis Platelet defects (e.g., idiopathic or drug induced thrombocytopenic purpura)

aGN, glomerulonephritis.

1. Assessment of Renal Function 5

3. Proteins derived from the lower urinary tractor leaking into the urine as a result of tissue injury or inflammation.fl

Most patients with persistent proteinuria shouldundergo a quantitative measurement of proteinexcretion, with a 24-hour urine measurement.A protein excretion rate greater than 3.5 g/d(nephrotic range proteinuria) should prompt further investigations to ascertain the exact cause of the proteinuria with measurements of urea, creatinine, liver function tests (most impor-tantly, serum albumin), and a full blood count.A defi nitive diagnosis has to be achieved with afirenal biopsy. Nephrotic syndrome is defined as a ficombination of proteinuria in excess of 3 g/d, hypoalbuminemia, edema, and hyperlipidemia.

Glucose

Renal glycosuria is uncommon and any positivetest requires evaluation of diabetes mellitus.

Bacteriuria

Dipsticks detect nitrites produced from the reduc-tion of urinary nitrate by bacteria and also leuco-cyte esterase, an enzyme specific for neutrophils. fiDetection of both nitrites and leucocytes on dip-stick has a high predictive value for urinary tractinfection.

Urine Microscopy

An unspun sample of urine may be examined under low- or high-power microscopy, however, a spun sample is a more accurate. Urine is centri-fuged, the supernatant is discarded, and an aliquotof the residue is placed on a glass slide using a Pasteur pipet. All patients suspected of havingrenal disease should have urine microscopy.

White Blood Cells

The presence of 10 or more white blood cells percubic millimeter is abnormal and indicates an

TABLE 1.4. Causes of proteinuria based on pathophysiologic mechanisma

Glomerular proteinuria Primary glomerular disease Secondary glomerular disease• Minimal change disease • Drugs: e.g., mercurials, gold compounds, heroin,• Mesangial proliferative GN penicillamine, probenecid, captopril, lithium, NSAID• Focal and segmental glomerulosclerosis (FSGS) • Allergens: bee sting, pollen, milk• Membranous GN • Infections: bacterial, viral, protozoal, fungal, helminthic• Mesangiocapillary GN • Neoplastic: solid tumors, leukemia• Fibrillary GN • Multisystem: SLE, Henoch-Schonlein purpura, amyloidosis• Crescentic GN • Heredofamilial: diabetes mellitus, congenital nephritic syndrome, Fabry’s disease, Alport’s syndrome • Others: febrile proteinuria, postexercise proteinuria, benign orthostatic proteinuria

Tubular proteinuria Endogenous toxins: light chain damage to proximal tubular, lysozyme (leukemia)Exogenous toxins and drugs: mercury, lead, cadmium, outdated tetracycline

Tubulointerstitial disease SLEAcute hypersensitivity interstitial nephritisAcute bacterial pyelonephritisObstructive uropathyChronic interstitial nephritis

Overflow proteinuria Multiple myelomaLight chain diseaseAmyloidosisHemoglobinuriaMyoglobinuriaCertain colonic or pancreatic carcinomas

Tissue proteinuria Acute inflammation of urinary tractUroepithelial tumors

aGN, glomerulonephritis; NSAID, nonsteroidal anti-inflammatory drugs; SLE, systemic lupus erythematosus.

6 M. Arkanath

infl ammatory reaction within the urinary tract. flMost commonly this represents a urinary tract infection but it may also be found with a sterile sample in patients on antibiotic therapy, or with kidney stones, tubulointerstitial nephritis, papil-lary necrosis, or tuberculosis.

Red Blood Cells

As mentioned previously, erythrocytes can findfitheir way into the urine from any source betweenthe glomerulus to the urethral meatus. The pres-ence of more than two to three red blood cells per HPF is usually pathological. Erythrocytes origi-nating in the renal parenchyma are dysmorphic,whereas those originating in the collecting systemretain their uniform biconcave shape.

Casts

Based on their shape and origin, casts are appro-priately named. Hyaline casts are found in con-centrated urine, during febrile illnesses, afterstrenuous exercise, and with diuretic therapy. They are not indicative of renal disease. Red cell casts indicate acute glomerulonephritis. White cellcasts indicate infection or infl ammation, and areflseen in pyelonephritis and interstitial nephritis.Renal tubular casts are found in cases of ATN andinterstitial nephritis. Coarse granular casts arenonspecific and represent the degeneration of aficast with a cellular element. Finally, broad waxy

casts are indicative of stasis in the collectingtubules and are seen in chronic renal failure.

References

1. Brown SC, O’Reilly PH. Iohexol clearance for thedetermination of glomerular filtration rate in clinicalfipractice: evidence for a new gold standard. J Urol. 1991; 146(3): 675–679.

2. Cockroft DW, Gault MH. Prediction of creatinineclearance from serum creatinine. Nephron. 1976; 16:31–41.

3. Hoste EA, Damen J, Vanholder RC, et al. Assessment of renal function in recently admitted critically illpatients with normal serum creatinine. Nephrol Dial Transplant. 2005; 20(4): 747–753.

4. Henry JB, Lauzon RB, Schaumann GB. Basic exami-nation of the urine. Clinical Diagnosis and Manage-ment by Laboratory Methods. 19th ed. Philadelphia: Saunders. 1991.

5. Graff L. A Handbook of Routine Urinalysis. Philadel-phia: Lippincott. 1983.

Suggested Reading

Davison AM, Cameron S, Grunfeld J-P, et al. OxfordTextbook of Clinical Nephrology. 3rd ed. Oxford: Oxford University Press. 2005.

Greenberg A, Cheung AK, Coffman TM, et al. Primer onKidney Diseases. 3rd Ed. London: Academic Press.2001.

Guyton AC. Textbook of Physiology. 8th Ed. New York:Saunders. 1991.

7

2Imaging of Acute Renal Failure—A Problem-Solving Approach for IntensiveCare Unit Physicians

Tom Sutherland

(MRI) scans analyze renal structure and renal artery calcifi cation, and dynamic gadolinium-fienhanced MRI renal studies allow functional assessment. Other functional studies, such as mer-captoacetyltriglycine (MAG3) and diethylene tri-amine pentaacetic acid (DTPA) show reduced renal uptake and delayed excretion of tracer.

A further role of imaging is to determine thenumber of present and functioning kidneys. ForARF to occur in previously normal kidneys, the underlying cause must be a bilateral process, or else a single functioning kidney must be compro-mised (Figure 2.1).

Acute Tubular Necrosis

Ultrasound will usually show enlarged kidneyswith a smooth contour caused by interstitialedema, no hydronephrosis, and renal arterial and venous fl ow. The examination of choice in sus-flpected ATN is a MAG3 nuclear medicine study.

Scintigraphic examinations in ATN using Tc-99m-MAG3 demonstrate relatively well-preservedon-time renal perfusion, and delayed traceruptake, often with a continuing activity accumula-tion curve. If the activity curve does have a maximum, the time to maximum is delayed (2). Excretion of tracer into the collecting system is delayed and reduced, but there is no obstructionto drainage of the collecting systems. If excre-tion and drainage occur, the time from maximal activity to half-maximal activity (or anotherquantitative measure of tracer excretion) is pro-longed. What underlies this scintigraphic pattern

Acute renal failure (ARF) is a common problem in hospitalized patients. It has a variety of causes,traditionally divided into prerenal, renal, andpostrenal. A further classification can be made fiinto medical and surgical causes, with the later defined as patients who will benefifi t from mechan-fiical intervention.

Acute tubular necrosis (ATN) is the mostcommon cause of ARF (approximately 45%) andpostrenal obstruction accounts for roughly 10% of presentations (1). A variety of imaging modalitiesmay be used to help diagnose the cause, or, if this is not possible, to differentiate medical from sur-gical causes. ARF in renal transplants will not be addressed.

Acute or Chronic Renal Failure?

This is best answered by clinical assessment rather than with imaging. Imaging findings in chronicfirenal failure are nonspecific and essentially char-fiacterized by small kidneys. X-rays can show the renal outline, calcifi cation, renal bone disease and fieffects of hyperparathyroidism. Ultrasound is anexcellent modality for structural imaging becauseit is able to detect reduced renal parenchyma, scar-ring (usually secondary to previous reflux nephro-flpathy), calcification, and polycystic kidneys. Thefiechogenicity of the cortex can be assessed with a hyperechoic cortex (normal cortex is hypoechoicto liver), present in most causes of chronic renal failure; adult polycystic kidney disease being the notable exception. Noncontrast computed tomo-graphic (CT) and magnetic resonance imaging

8 T. Sutherland

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 tracer in the renal cortex indicating normal perfusion but no excre-tion, consistent with ATN. A normal activity curve should initially

peak as the kidneys are perfused and then activity declines astracer is excreted and passes into the bladder. Lt, left; Rt, right; Bkg, background.

Liver LiverLiver Dialysis fluid Renal cortexLiver

A B

2. Imaging of Acute Renal Failure 9

is parenchymal retention of MAG3 by the remain-ing viable tubular cells, whereas tubular obstruc-tion prevents drainage of tracer into the collectingsystem. The tubular cells continue to take up tracerwhile they are viable and, thus, the cortical activity can be used to monitor disease progress, with pro-gressive loss of cortical activity being a poor prog-nostic indicator (Figure 2.2).

If a MAG3 study cannot be performed, ultra-sound will demonstrate a cortex of normal echo-genicity with either a normal or hypoechoic medulla. The renal arteries can also be interro-gated for the renal index (RI), which is an objec-tive measure of the resistance to renal perfusion.RI is defi ned as (systolic velocity minus diastolic fivelocity) divided by systolic velocity, and has been heavily investigated to determine whetherelevation in RI can differentiate ATN from renal hypoperfusion not yet complicated by ATN. Unfor-tunately, there have been mixed results and, gener-ally, RI has inadequate specifi city for routine ficlinical use.

Glomerulonephritis and Acute

Interstitial Nephritis

Clinical history and urinalysis plays a vital role indifferentiating GN and acute interstitial nephritis(AIN) with the “gold standard” diagnostic test being a renal biopsy. The main role of imaging is to detect structural signs of chronic renal diseaseand to exclude other causes of ARF. MAG3 studieswill show poorly functioning kidneys, but noaccumulation pattern and also no obstructionto drainage. Edema can sometimes be demon-strated with ultrasound, manifest as hypoechoiclarge kidneys. If there is a clinical suspicion to direct imaging, a careful search may also find fiother signs of the underlying cause, for example, pulmonary hemorrhage in Goodpasture’s syn-drome and pulmonary nodules in Wegener’sgranulomatosis.

Ureteric Obstruction

For obstruction to produce ARF it must be bilat-eral or affecting a single functioning kidney. This is the classical cause of surgical ARF.

Ultrasound is the first-line test for obstruction.fiIt is radiation free, portable, is not nephrotoxic, and can simultaneously gather other structuralinformation. Obstructed kidneys are typically normal sized with dilated ureters, renal pelvis, andcalyceal systems. These urine-filled structures fiappear as anechoic areas with posterior acoustic enhancement. Caution is required in interpreta-tion because a ureter and renal pelvis may bedilated without being obstructed (a false positive). This can occur after previous obstruction thatleaves a residual “baggy” collecting system, or asan anatomical variant (enlarged extrarenal pelvis). Differentiation can often be made by examiningthe bladder for ureteric jets, which are the periodic expulsion of urine from the ureter into the bladder,which can be detected by Doppler imaging. If ure-teric jets are present, then there is not a completeobstruction on that side (3). False negatives canoccur in the hyperacute setting if the renal collect-ing system has not had time to dilate, or if the system has spontaneously decompressed by forni-ceal or renal pelvis rupture (Figure 2.3).

Noncontrast CT scan is the gold standard fordetecting ureteric calculi (4). The ureters can usually be traced between the kidney and bladder, and a hyperdense stone can be seen at the distalsite of hydroureter. More than 99% of renal calculi are radiopaque on CT scan, however, xanthine calculi may be radiolucent and stones associated with indinavir are radiolucent. The obstructed kidney is typically edematous (i.e., swollen) with

Cortex CalyxRenal pelvis

FIGURE 2.3. Ultrasound of a hydronephrotic kidney. Size is normal with dilated renal pelvis and calyces (the anechoic central part).

10 T. Sutherland

perirenal stranding. The administration of con-trast is contraindicated in ARF because of itspotential nephrotoxicity. Although not as helpful as a contrast-enhanced CT scanning, a non-contrast study can usually detect many extrinsiccompressing masses, such as retroperitonealtumors, or cervical or colon carcinomas, thatmay produce bilateral obstruction. Complicationsrelated to trauma, such as urinoma or renal pedicle avulsions, are also visible even without contrast (Figure 2.4).

Scintigraphic imaging with either Tc-99m-MAG3 or Tc-99m-DTPA can detect ureteric obstruction by showing a dilated collecting system with delayed drainage of tracer. The collecting system is outlined down to the level of the obstruc-tion with no or limited tracer draining beyondthis. The negative predictive value of nuclear med-icine scanning is extremely high, with a normalstudy virtually excluding obstruction. Of course, sufficient tracer must be present in the collectingfisystem to reach the obstruction point. Reducedurine production associated with renal failure may deliver so little tracer into the collectingsystem that obstruction cannot be excluded. Thelower the patients glomerular filtration rate (GFR),fithe higher the probability of having an indetermi-nate study (5). Further, an enlarged but non-obstructed collecting system may mimic delayed drainage. In this latter situation, the specifi city of fithe examination is increased by administering 20 mg (or sometimes 40 mg) of frusemide IV, andscanning for a further 20 minutes. A baggy non-obstructed system will promptly wash out, and thedelay in washout after frusemide correlates with

the degree of obstruction. Once again, it must bestressed that the only way mechanical obstructioncan produce ARF in previously normal kidneys isby affecting a solitary functioning kidney, or by blocking both kidneys simultaneously.

Renal Artery Dissection or Occlusion

Each kidney is normally supplied by a single renalartery that arises from the aorta before dividinginto an average of fi ve segmental end-arterial fibranches. Arterial dissection and occlusion is another surgical cause of ARF. Renal infarction may be caused by blunt trauma with avulsion of the renal pedicle or penetrating trauma transect-ing the renal artery. Renal artery dissection may be iatrogenic, occurring during endovascularintervention. Bilateral dissection or exclusionfrom the circulation can occur secondary to an aortic dissection. Suspicion of a dissection or occlusion is virtually the only differential diagno-sis that warrants the administration of intra-venous CT contrast in the setting of ARF in whichMRI scanning is unavailable. An arterial phase CT scan highlights the renal artery anatomy, demon-strating the intimal flap of dissection and can alsofldetect signs of renal infarction and lacerations (6)(Figure 2.5).

Magnetic resonance angiography (MRA) hasexcellent sensitivity and specifi city for detectingfidissection and occlusion of proximal renal arter-ies. Segmental arteries are less well visualized, however, unilateral or isolated segmental pathol-ogy will not produce ARF. MR contrast is much

Arrow heads are fatty stranding. Inferior pole left kidneyRight ureteric calculus

FIGURE 2.4. Noncontrast CT scan withcomplete right-sided and moderate left-sided obstruction secondary to bilateralureteric calculi (left calculus not shown). Edematous fatty stranding around eachkidney with the right calculus in the dilated ureter. Arrowheads indicate fattystranding.

2. Imaging of Acute Renal Failure 11

less nephrotoxic than CT contrast. MRI scanningis logistically difficult for most intensive care unit fi(ICU) patients, and has a considerably longer scantime and, therefore, CT scanning remains a morepractical modality.

MAG3 examinations will reveal an absence of perfusion (segmental or unilateral or bilateral, asthe case may be) but not the cause.

Renal Vein Thrombosis

Bilateral renal vein thrombosis (RVT) will present with ARF, but is very rare. If discovered, it shouldtrigger a search for a mass compressing both renalveins or the suprarenal inferior vena cava (IVC)and a coagulopathy screen. Unilateral RVT is more frequent in contrast, but will not produce ARF if both kidneys are initially normal.

Usually, a single renal vein exists on each side,but up to 30% of patients will have multiple renalveins. RVT is most common on the left becausethis vein is three times longer than the right veinand is more liable to extrinsic compression as itpasses between the superior mesenteric artery and the aorta. RVT may be caused by hypercoagu-lable states, dehydration (particularly in infants),sepsis, or trauma and occurs in up to 40% of patients with nephrotic syndrome.

Ultrasound can show a distended renal veinwith echogenic thrombus within that vein thatmay extend into the IVC. Absent flow or a thin cuff fl

of blood fl owing around the thrombus is shown flon color Doppler imaging (7). The kidney itself may have reduced corticomedullary differentia-tion, be enlarged, and contain focal areas of increased echogenicity related to edema and hem-orrhage. Contrast-enhanced CT scanning has asimilar appearance in the acute setting, with a dilated renal vein with hypodense thrombuswithin. Secondary findings, such as perirenal fistranding, dilated gonadal vein, prolonged nephro-gram, and complications such as retroperitonealhemorrhage may also be present. As the chronicity increases, the thrombus contracts and multiplecollateral vessels develop.

Renal veins are well demonstrated by MRI scan-ning, with the thrombus appearing hyperintenseon both T1- and T2-weighted images. The kidney appears swollen with loss of corticomedullary dif-ferentiation on T1-weighted images, with bothcortex and medulla having low signal secondary to prolonged T1 and T2 relaxation times.

Venography is the gold standard and can accu-rately determine the presence of thrombus and itsextent. A catheter inserted in the common femoral vein allows selection of the renal vein as it enters the IVC and, with contrast injection, a filling defect fior complete occlusion can be demonstrated.Therapeutic measures, such as the instillationof localized thrombolytics and the placement of an IVC fi lter if the thrombus is extensive, can be fiperformed during the diagnostic procedure. Complications include damage to vessels and

FIGURE 2.5. Contrast enhanced CT scan (magnified view of the left kidney) withno renal parenchymal enhancement or renal vascular enhancement secondaryto an occlusive renal artery embolus.

Contrast in the aorta left kidney

12 T. Sutherland

? A

RF

wit

h pr

evio

usly

nor

mal

kid

neys

or

acut

e on

chr

onic

ren

al

fail

ure

Ans

wer

wit

h hi

stor

y cl

inic

al f

indi

ngs

and

Lab

orat

ory

test

s

Acu

te o

n ch

roni

c re

nal

fail

ure

Pre

rena

l

Med

ical

?AT

N

Med

ical

Sur

gica

l

US

US

MR

I or

ang

iogr

aphy

if

hig

h pr

etes

t pr

obab

ilit

y

? R

enal

vei

nth

rom

bosi

s ?

Aor

tic

or r

enal

ar

tery

di

ssec

tion

/occ

lusi

on

Ren

al

AR

F w

ith

Pre

viou

sly

norm

al k

idne

ys

Pos

tren

al

ANATOMICAL REGION

Sur

gica

l

? G

lom

erul

onep

hrit

is o

r ac

ute

inte

rsti

tial

nep

hrit

is

? O

bstr

ucti

on

CLINICALDDx

US

– la

rgel

y to

exc

lude

ot

her

diag

nose

s

US

US

to c

onfi

rm c

hron

ic

chan

ges,

sca

rs e

tc. +

/-

MA

G3

for

fun

ctio

n.

TEST TO ANSWER CLINICAL QUESTION

MR

I or

C

ontr

ast e

nhan

ced

CT

*N

on c

ontr

ast C

T f

or

obst

ruct

ing

mas

s or

st

one

MA

G3

Non

con

tras

t CT

for

?

obst

ruct

ing

mas

s

FIG

UR

E 2

.6.

Diag

nost

ic a

lgor

ithm

for i

mag

ing

in A

RF. U

S, u

ltras

ound

.

2. Imaging of Acute Renal Failure 13

dislodgement of thrombus resulting in pulmonary embolism, and require the administration of potentially nephrotoxic intravenous contrast.

Renal Artery Stenosis

Renal artery stenosis (RAS) is most commonly associated with hypertension and is usually causedby atherosclerotic plaques and less commonly by fibromuscular dysplasia. It can exacerbate renalfihypoperfusion and is a cause of chronic renalimpairment. In these cases, signs of chronic renalimpairment, as previously mentioned, will usually be present. It is extremely unlikely to be the caus-ative factor for ARF presenting de novo. In patientspresenting with acute-on-chronic renal failure, itis one of the many possible causes of the preexist-ing renal failure and enters the differential diag-nosis in this fashion.

Conclusion

Clinical assessment in ARF is vital to construct a differential diagnosis list. Imaging can then havea targeted role to answer questions and enable the

final diagnosis to be made. Using the above dis-ficussion, a potential diagnostic algorithm is given in Figure 2.6.

I thank Professor Alexander Pitman for hisassistance and guidance with this work.

References

1. Liano F, Pascual J. Epidemiology of acute renal failure:a prospective, multicenter, community-based study. Kidney Int 1996; 50: 811–818.

2. Dahnert W. Radiology review manual. 5th ed.Philadelphia: Lippincott Williams & Wilkins. 2003.

3. Middleton WD, Kurtz AB, Hertzberg BS. Ultrasound—The requisites. 2nd ed. St. Louis: Mosby. 2004.

4. Sheafor DH et al. Non-enhanced helical CT and USin the emergency evaluation of patients with renal colic: Prospective comparison. Radiology 2000; 217: 792–797.

5. Brant WE, Helms CA. Fundamentals of diagnostic radiology. 2nd ed. Philadelphia: Lippincott Williams & Wilkins. 1999.

6. Urban BA, Ratner LE, Fishman EK. Three-dimensional volume-rendered CT angiography of the renal arter-ies and veins: Normal anatomy, variants, and clinical applications. Radiographics 2001; 21: 373–386.

7. Pitman AG, Major NM, Tello R. Radiology core review. Edinburgh: Saunders. 2003.

14

3Drug-Induced Renal Injury

Sara Blakeley

4. Monitoring of drug levels (where appropriate).5. Discontinuation or change of mediation when

possible.6. Consider speci c therapies, e.g., fi N-acetylcysteineNN

(NAC; contrast), rasburicase, and allopurinol(tumor lysis syndrome), urinary alkalinization(drug-induced rhabdomyolysis).

7. Close liaison with ICU pharmacist.

Nonsteroidal Anti-Inflammatory

Drugs

NSAIDs are widely used and overall renal com-plications are uncommon, however, their use in the community has been shown to increasethe risk of hospitalization with AKI by up tofour-fold (4, 5). Renal effects are predominately caused by their effect on the production of renalprostaglandins.

Effects on Renal Prostaglandins

Membrane-bound phospholipids are converted toarachidonic acid by phospholipase A. Arachidonic acid then enters the cyclooxygenase pathway under the action of cyclooxygenase enzyme (COX) forming prostaglandins, or enters the lipoxygen-ase pathway, forming leukotrienes.

In health, renal prostaglandins have a numberof effects: control of renin release, regulation of renal blood flow (RBF) through vascular tone, andflcontrol of salt and water transport in the renal tubules. They have both vasodilatory (prostaglan-din [PG]-E2, PGI2) and vasoconstrictor (throm-

Drug-induced nephrotoxicity contributes to 8 to60% of all cases of in-hospital acute kidney injury (AKI) and 1 to 23% of cases of AKI seen on theintensive care unit (ICU) (1). The wide variationsare caused by different definitions of AKI and dif-fiferent patient populations. Drug-induced nephrop-athies are often underdiagnosed, and should beconsidered in every type of renal failure, both acute renal failure and chronic renal failure (CRF).

Risk Factors for Nephrotoxicity

Many patients may be taking a potentially nephro-toxic drug. For example, 55% of patients admitted to a general medical ward were taking one or more of either an angiotensin-converting enzyme inhibitor (ACEI) or angiotensin receptor antago-nist (ARB), a diuretic, or a nonsteroidal anti-infl ammatory drug (NSAID) (2). More than 25% of flpatients 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-and drug-related factors (Tables 3.1 and 3.2).

Methods of Prevention

The following general principles should be adhered to:

1. Appropriate dose alteration in patients dependingon renal function, age, and ideal body weight.

2. Awareness of drug interactions.3. Close attention to fluid balance and hemody-fl

namic status.

3. Drug-Induced Renal Injury 15

are crucial to maintain RBF and GFR. Patientswho fall into this high-risk category are those withparenchymal renal disease or renal impairment,hypovolemic patients, those with congestive cardiac failure or liver disease (because of activa-tion of the renin-angiotensin system), and the elderly. There are two isomers of the COX enzyme(6). “Traditional” NSAIDs are nonselective COXinhibitors, whereas newer “coxibs” selectively target the production of proinfl ammatory prosta-flglandins by COX-2. They have fewer gastric and platelet side effects, but it is now clear that they have the same effects on renal prostaglandins (7);therefore, caution should be still be taken in high-risk individuals.

NSAID-induced salt and water retention occursand the effect of loop diuretics is also diminished.This may lead to pedal edema or hypertension, but can precipitate overt cardiac failure in at-risk patients.

Other Renal Effects

1. Acute tubulointerstitial nephritis (TIN) is less common, occurring after 3 to 5 days, or even after years, of drug use. It occurs in an

TABLE 3.1. Mechanism of drug-induced AKI

Mechanism Drug examples

Altered intraglomerular Calcineurin inhibitorshemodynamics: drugs that Vasopressorsinterfere with the normal Amphotericinregulatory alterations in Contrast agentspreglomerular and postglomerular NSAIDs, ACEI/ARBsarteriolar resistance cancompromise renal blood flow andglomerular filtration rate, especiallyin times of hemodynamic instabilit

Drug-induced glomerulopathy: this Gold, D-penicillamineusually presents with nephrotic NSAIDssyndrome or proteinuria. Renal Mesalazinefunction need not be impaired

Drug-induced thrombotic Cyclosporin A, OKT3,microangiopathy: rare tacrolimus

Clopidogrel, ticlopidineCocaineQuinine

Direct tubular cell toxicity AminoglycosidesAmphotericinCalcineurin inhibitorsCisplatinMethotrexateCocaineContrast mediaPalmidronate

Tubular damaged caused by osmotic Mannitolnephrosis: tubule cells take up Dextransnonmetabolizable molecules, Intravenouscreating an osmotic gradient. immunoglobulinWater rapidly accumulates in Hydroxyethyl starchthe cell, causing swelling andvacuolation, thereby disrupting cellintegrity. Cellular debris causestubular obstruction

Tubular damage caused by cast Statinsdeposition: rhabdomyolysis

Interstitial nephritis: drugs can cause β-lactams, quinolones,an acute allergic interstitial rifampicin, macrolides,nephritis or chronic interstitial sulfonamidesdamage NSAIDs

Thiazide and loop diureticsPhenytoinCimetidine, ranitidineAllopurinolAntiviralsCocaine

Note: This list is not exhaustive.Source: Schetz et al., 2005; and Perazella, 2003.

boxane A2) effects. In certain situations associated with high levels of circulating vasoconstrictors(such as angiotensin II [ATII], endothelin, cate-cholamines), vasodilatory renal prostaglandins

TABLE 3.2. Risk factors for development of drug-induced AKI

Patient-related factors Drug-related factors

Increased age Inherent nephrotoxic potentialPreexisting chronic kidney Dose (important for drugs

disease inducing crystal deposition,Diabetes mellitus causing tubular damage,Intravascular volume depletion: and drugs altering renal

absolute (e.g., dehydration) hemodynamics)or effective (e.g., massive Prolonged duration ofascites, nephrotic syndrome) treatment

Peripheral vascular disease Frequency of administration(this increasesthe risk of Time of administrationrenovascular disease)

Sepsis Rate of administrationConcomitant use of diuretics (important fordrugs causing

and other nephrotoxic drugs crystal nephropathy)Metabolic disturbances: sodium Route of administration (e.g.,

depletion, hypoalbuminemia, intravenous versus oralacid-base disturbances (which contrast)may exacerbate intrarenal Combination of drugs causingcrystal deposition), multiple nephrotoxic synergy (e.g.,myeloma cephalosporins and

aminoglycosides aminoglycosides,vancomycin and vancomycin, and

aminoglycosides)

16 S. Blakeley

idiosyncratic, non-dose-dependent manner. Theclassic features of fever, rash, arthralgia, and eosinophilia are often not present. The diagnosisis made on renal biopsy. Treatment is to stop thedrug and support the patient. Corticosteroids are sometimes administered, but the evidence is lacking.

2. Renal papillary necrosis.3. Nephrotic syndrome.4. Hyperkalemia caused by hyporeninemic

hypoaldosteronism generally occurs in patientswith CRF, diabetes mellitus, and type IV renaltubular acidosis.

ACEIs and ARBs

ATII is a potent vasoconstrictor causing constric-tion of the efferent arteriole, increasing trans-glomerular pressure, and, thus, GFR. It also causessystemic vasoconstriction, increasing systemicblood pressure and improving renal perfusion. Incertain situations, GFR is very dependent on thispostglomerular constriction and, consequently,blocking this mechanism by the use of ACEI/ARBsmay lead to a marked reduction in GFR (3).

If, after starting an ACEI/ARB, there is a small,but nonprogressive rise in the serum creatinine,this is generally related to alterations in intrarenalhemodynamics rather than structural injury (8). If the initial serum creatinine rises by more than 30% or if there is a progressive increase in creati-nine after starting the drug, the drug should bestopped and a reason sought. This situation may arise in patients with bilateral renal artery steno-sis (or stenosis of a single functioning kidney), but is also seen when there is a reduction in the abso-lute or effective circulating volume, in the pres-ence of sepsis, and with the concomitant use of other drugs (e.g., NSAIDs). These factors shouldbe addressed before restarting the drug.

Aminoglycosides

Aminoglycoside toxicity is caused by partial reabsorption of the drug by the proximal renaltubular cells, and the first indication of renalfiinvolvement is the development of polyuria because of a defect in the urinary concentrating

ability. There is then a slow rise in serum creati-nine over several days.

The risks of toxicity are increased with high initial peak serum levels, prolonged treatment, hypovolemia, increasing age, liver dysfunction,hypoalbuminemia, and in combination with other drugs (e.g., diuretics, NSAIDs, ACEIs, cephalospo-rins) (9).

Once daily dosing is associated with less neph-rotoxicity than dosing multiple times daily, butmay not be suitable for all patient groups (e.g.,controversial in the treatment of bacterial endo-carditis). The risk of nephrotoxicity is reducedwhen the once daily dose is given during periodsof activity (daytime) or when taking food, possi-bly related to changes in urinary pH. Levels shouldstill be monitored carefully according to localpractice.

Contrast Nephropathy

Incidence and Outcome

Contrast nephropathy (CN) is the third leading cause of AKI in hospitalized patients (10), account-ing for 12% of cases of AKI. It is defined as anfiacute decline in renal function, with a rise inserum creatinine of greater than 25% from base-line (or an absolute rise of 44 μmol/L), occurring24 to 48 hours after contrast. Creatinine usually peaks at 5 days and returns to baseline by 10 days.Incidence is varied depending on definitions usedfiand patient populations studied. Overall incidence is less than 3% (11, 12), but rises in the setting of increasing number of risk factors (13), up to50% with the combination of diabetes and renalfailure (14).

Risk factors for development of CN:

• Preexisting renal impairment• Diabetes mellitus• Absolute (e.g., dehydration) or effective volume

depletion (e.g., congestive cardiac failure,nephrotic syndrome, liver disease)

• Left ventricular ejection fraction less than 40%• Concurrent use of nephrotoxic drugs (e.g.,

NSAIDs, aminoglycosides)• High volume, high osmolar, ionic contrast given

intravenously

3. Drug-Induced Renal Injury 17

• Pre procedure shock (e.g., hypotension, intra-aortic balloon pump)

• Increasing age

Renal failure is normally nonoliguric andresolves with supportive care. Up to 12% of patients need renal replacement therapy (11).

The development of CN increases mortality. Itis unclear whether this is because of renal failureitself, or is simply a marker of increased diseaseseverity and underlying comorbidity. In one study,the development of AKI increased mortality from 7 to 34% (11).

Methods of Prevention

Intravenous Hydration

Volume expansion with intravenous fluid before fland after contrast is proven to be beneficial. fiTiming, dose, and type of fluid administered are flless well defined. Isotonic flfi uid (using sodium flbicarbonate) may be better than hypotonic fluid flbecause of better volume expansion, urinary alka-linization, and reduction of free radical-mediatedinjury (15).

N-Acetylcysteine

NAC counteracts renal vasoconstriction but alsoscavenges oxygen free radicals, thereby preventingdirect oxidative tissue damage after exposure to contrast media (16).

Data from a series of meta-analyses are mixed because of the lack of a standardized definition of fiCN, study heterogeneity, and publication bias.Despite this, NAC seems to reduce the risk of renalinjury in high-risk patients.

Type of Contrast Used

Studies compared high, low and iso-osmolar prep-arations. In high-risk patients, iso-osmolar com-pounds are associated with a reduced incidence of CN (17). Nonionic rather than ionic compounds have also been found to be less nephrotoxic.

Other Therapies (18, 19)

A recent meta-analysis found no evidence thatperiprocedural renal replacement therapy reducedthe incidence of CN (20). Loop diuretics may

reduce the potential for ischemic injury by inter-fering with active transport and decreasing the oxygen demands of tubules, however, evidence islacking. There is no evidence that mannitol, atrial natriuretic peptide, dopamine, or fenoldopam (a dopamine 1 agonist) are better than standard hydration. Calcium channel antagonists gainedsome interest but were not supported by largetrials. Despite its properties as an adenosine antagonist, there is no evidence that theophyllineis more effective, and work continues on the anti-oxidant, ascorbic acid.

Suggestions for High-Risk Patients

• Identify high-risk patients, correct modifiable firisks, and consider the risk/benefit to the patient fiof the procedure

• 600 to 1200 mg NAC twice daily for 2 days before procedure and two doses after the procedure

• Intravenous hydration with saline at 1 mL/kg/hfor 6 to 12 hours before and after theprocedure

Or

5% dextrose and H2O + 154 mEq/L sodium bicarbonate at 3 mL/kg/h for 1 hour before the procedure and 6 hours after the procedure (add154 mL of 1000 mEq/L sodium bicarbonate to 846 mL of 5% dextrose in H2O)

• Use the minimum volume of iso-osmolar or low osmolar contrast

References

1. Schetz M, Dasta J, Goldstein S, Golper T. Drug-induced acute kidney injury. Curr Opin Crit Care. 2005; 11: 555–565.

2. Loboz KK, Shenfield GM. Drug combinations and fiimpaired renal function—the “triple whammy”.Br J Clin Pharm. 2005; 59(2): 239–243.

3. Perazella MA. Drug induced renal failure: Update onnew medications and unique mechanisms of neph-rotoxicity. Am J Med Sci. 2003; 325(6): 349–362.

4. Perez Gutthann S, Garcia Rodriguez LA, Raiford DS,et al. Nonsteroidal anti-infl ammatory drugs and the flrisk of hospitalization for acute renal failure. Arch Intern Med. 1996; 156: 2433–2439.

5. Griffi n MR, Yared A, Ray WA. Nonsteroidal anti-fiinfl ammatory drugs and acute renal failure in elderly flpersons. Am J Epidemiol. 2000; 151: 488–496.

18 S. Blakeley

6. Barkin RL, Buvanendran A. Focus on the COX-1 andCOX-2 agents: Renal effects of nonsteroidal andanti-infl ammatory drugs—NSAIDs. fl Am J Thera-peutics. 2004; 11: 124–129.

7. Gambaro G, Perazella MA. Adverse renal effects of anti infl ammatory agents: Evaluation of selectivefland non selective cyclooxygenase inhibitors. J Int Med. 2003; 253: 643–652.

8. Palmer BF. Renal dysfunction complicating treat-ment of hypertension. N Engl J Med. 2002; 347:1256–1261.

9. Beauchamp D, Labrecque G. Aminoglycoside neph-rotoxicity: Do time and frequency of administra-tion matter? Curr Opin Crit Care. 2001; 7: 401–408.

10. Nash K, Hafeez A, Hou S. Hospital-acquired renal insuf ciency.fi Am J Kidney Dis. 2002; 39(5): 930–936.

11. Levy EM, Viscoli CM, Horwitz RI. The effect of acuterenal failure on mortality. A cohort analysis. JAMA. 1996; 275(19): 1489–1494.

12. Rudnick MR, Berns JS, Cohen RM, Goldfarb S.Contrast-media associated nephrotoxicity. Semin Nephrol. 1997; 17: 15–26.

13. Rich MW, Crecelius CA. Incidence, risk factors, andclinical course of acute renal insuffi ciency after ficardiac catheterization in patients 70 years of age orolder. A prospective study. Arch Intern Med. 1990; 150(6): 1237–1242.

14. Manske CL, Sprafka JM, Strony JT, Wang Y. Contrast nephropathy in azotemic diabetic patients undergo-

ing coronary angiography. Am J Med. 1990; 89(5):615–620.

15. Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrast induced nephropathy with sodium bicar-bonate: a randomised controlled trial. JAMA. 2004;291: 2328–2334.

16. Pannu N, Manns B, Lee H, Tonelli M. Systematic review of the impact of N-acetylcysteine on contrast nephropathy. Kidney Int. 2004; 65: 1366–1374.

17. Aspelin P, Aubry P, Fransson SG, Strasser R, Willen-brock R, Berg KJ. Nephrotoxicity in high-riskpatients study of iso-osmolar and low-osmolarnon-ionic contrast media study investigators. Neph-rotoxic effects in high-risk patients undergoingangiography. N Engl J Med. 2003; 348(6): 491–499.

18. Maeder M, Klein M, Fehr T, Rickli H. Contrast nephropathy: Review focusing on prevention. J AmColl Cardiol. 2004; 44: 1763–1771.

19. Pannu N, Wiebe N, Tonelli M. Prophylaxis strategiesfor contrast nephropathy. JAMA. 2006; 295: 2765–2779.

20. Cruz DN, Perazella MA, Bellomo R, Corradi V, de Cal M, Kuang D, Ocampo C, Nalesso F, Ronco C. Extracorporeal blood purifi cation therapies for pre-fivention of radiocontrast-induced nephropathy: asystematic review. Am J Kidney Dis. 2006; 48(3):361–367.

19

4Acute Kidney Injury

Sara Blakeley

three levels of renal dysfunction and two renaloutcomes. The levels of renal dysfunction can bedefined by changes in serum creatinine, GFR, or fiurine output.

Risk of Renal Dysfunction • Serum creatinine increased 1.5 fold or • GFR decreased by more than 25% or • Less than 0.5 mL/kg/h of urine for 6 hoursInjury to the Kidney • Serum creatinine doubled or

• GFR decreased greater than 50% or • Less than 0.5 mL/kg/h of urine for 12 hoursFailure of Kidney Function • Serum creatinine increased 3 fold or • An acute rise in creatinine of greater than

44 μmol/L so that new creatinine is greaterthan 350 μmol/L or

• GFR decreased more than 75% or • Less than 0.3 mL/kg/h of urine for 24 hours or

anuria for 12 hours• Note: This takes into consideration acute-on-

chronic renal failureLoss of Kidney Function • Complete loss of kidney function for longer

than 4 weeksEnd-Stage Renal Disease • The need for dialysis for longer than 3 months

Incidence and Outcome

AKI develops in 5 to 7% of hospitalized patients (6, 7). Six to 25% of patients on the ICU develop AKI (2, 8); overall, 4% of admissions require RRT.

Patients may be admitted to the intensive care unit(ICU) with acute kidney injury (AKI) or it may develop during their stay. This chapter gives an over-view of the de nition and epidemiology of AKI, along fiwith clinical features and initial investigations.

Definition of AKI

AKI is an abrupt (<7 d) and sustained decrease in kidney function (1). It is accompanied by changesin blood biochemistry (e.g., a rise in serum creati-nine), in urine output, or both. There is a spec-trum ranging from a mild transient rise in serum creatinine, to overt renal failure needing renalreplacement therapy (RRT); hence, the term acutekidney injury (AKI) is more precise than the term“acute renal failure”.

Multiple definitions of ARF exist, and the readerfiis guided to a series of excellent reviews that high-light this problem (1–5). A rise in serum creatinine is often used as a marker of renal dysfunction, but it is affected by extrarenal factors, such as age, sex,race, and muscle bulk. It may lag behind changesin glomerular filtration rate (GFR), either in fidecline or during recovery, and, therefore, doesnot always give a true reflection of the GFR. Urine floutput can be used to define renal failure, but this fican be confounded by the use of diuretics, andnot all cases of renal failure are associated with oliguria.

Efforts have been made to develop a universal and practical way of defining AKI via either serum ficreatinine or urine output. One such recent pro-posal is the RIFLE (5) system; an acronym for

20 S. Blakeley

This may underestimate the scale of the problem,however, because when all degrees of kidney dysfunction are considered using the RIFLE criteria, 20% (9) of hospital patients and 67% of ICU patients developed some form of kidney injury (10).

The incidence and progression of AKI varies depending on the patient group studied. For example, up to 20% of cardiac surgery patientswill develop some evidence of renal injury (11),but only 1% will need RRT (12).

AKI on the ICU is associated with a hospital mortality of 13 to 80% (2, 8, 10–17) and 57 to 80%if RRT is needed. Renal failure rarely occurs onits own, with up to 80% of patients with renal failure on the ICU having another organ systemfailure (8). Various factors have been associatedwith a worse outcome; including comorbidity, increased severity of illness, presence of sepsis,need for mechanical ventilation, oliguria, hospi-talization before ICU, and delayed occurrence of AKI (13–15).

The development of AKI dramatically increases mortality across all patient populations studied (8–10, 12, 14). Worsening levels of renal dysfunc-tion, as described by the RIFLE criteria, correlate well with increasing hospital mortality, with up toa 10-fold risk of death with “failure.” AKI carriesan independent risk of death, but it is unclear whether this is related to the systemic effects of renal failure itself, the effects of its treatment, oris simply a refl ection of the severity of the under-fllying condition.

After AKI needing RRT, 10 to 32% of patientsare discharged from hospital still needing RRT (2, 16, 18).

Causes of AKI

Causes of AKI can be divided into prerenal, intrin-sic, and obstructive causes. One disease may beassociated with different causes of ARF, for example, sepsis is a common cause of renal dys-function on the ICU, accounting for up to 50% of cases of AKI. AKI occurs in 23% of patients withsevere sepsis, and in 51% of patients with septic shock when blood cultures are positive (19). Sepsis is characterized by systemic vasodilation (prere-(

nal failure) but intrarenal vasoconstriction, whichlcould progress to tubular damage (intrinsic renal cfailure). Glomerular microthrombi are associatedwith disseminated intravascular coagulation, andcan cause intrinsic AKI (20).

Prerenal Failure

Prerenal failure (Figure 4.1) accounts for 15 to20% of cases of AKI on the ICU (13, 15). For the kidneys to be perfused and, therefore, function,adequate pressure, flow, volume, and patent vesselsflare needed.

The kidney autoregulates to maintain a con-stant renal blood fl ow (RBF) through a mean arte-flrial pressure range of 65 to 180 mmHg. “Prerenalfailure” is an appropriate, albeit exaggerated,physiological response to renal hypoperfusion.Stimulation of the renin-angiotensin-aldosterone system attempts to retain salt and water and,therefore, maintain RBF. Because renal tissue is still preserved, once renal perfusion is restored,function should improve. A profound or prolongedreduction in perfusion can, however, lead to isch-emic acute tubular necrosis (ATN) (intrinsic renal failure).

Conditions leading to reduced renal perfusion and, therefore, causing prerenal failure are:

• Hypotension (relative or absolute) secondary to vasodilation (e.g., certain drugs, loss of vascular tone, and sepsis)

• Compromised cardiac function• Intravascular volume depletion (absolute or

effective)• Increased intra-abdominal pressure (abdominal

compartment syndrome)

Intrinsic Renal Failure

The commonest cause of intrinsic renal failure on the ICU is ischemic ATN developing after pro-found or prolonged prerenal failure (Figure 4.2).Up to 80% of cases of AKI on the ICU are attrib-uted to ATN (2, 13, 15). Although ATN is a histo-logical diagnosis, its development is suggested by the persistence of renal failure following therestoration of adequate renal perfusion. The

4. Acute Kidney Injury 21

FIGURE 4.1. Causes of prerenal failure. GI, gastrointestinal.

PPPrPrPrPrPreeeee rererererenananananalllllll fffafafafafailililililililururururureeeee

HHHyHypopovovollalalaememiiaiaia(Reduced intravascular volume)

LLoLoLoww cacaca drdrdrdiiaiaiaccc oououttptptp tututut

TToToToToToTottatatatatatalllllll llolololololossssssssssss VVoVoVoVoVoVollulululululumemememememe rrrrr r dedededededediisisisisisisttrtrtrtrtrtribibibibibibib tututututututiioioioioioionnnnnn

GIGIGIGIGIGIGI llllll lososososososssssss(V(Vom(Vom(Vom(Vomitiitinitinitinitin dg, dg, diiarriarriarriarrhhoeahoeahoeahoea susu, su, irgicrgicrgicrgical fal fal fal fistuistuistuistulae)lae)lae)lae)

HHaHaHaHaHaHaemememememororororor hrhrhrhrhrhrhagagagagagggeeeee (Vi(Vis(Vis(Vis(Visiblibleibleibleible dandandand and occoccocc occ lt)ult)ult)ult)ult)

RReReReReReRenananananalllllll llolololololossssssssss(Di(Diu(Diu(Diu tiretiretireticscscscs, cs, lpolypolypoly iuriauriauria))))

SkSkSkSkSkSkSkiinininininin llllll lososososossssss(E(Exc(Exc(Exc(Exc( iessiessiessiessive sve sve sve s tweatweatweatweatiinginging,ing,g, bburburbur bur )ns)ns)ns)ns) )

RReReReReReddududududucececececedddddd fefefefefefffefefefefe tctctctctctiiviviviviveeeee icicicicicircrcrcrcrc lululululul tatatatatatiinininininggggg gvovovovovollululululumememememe

(A(Asc(Asc(Ascititesitesites oe, oeddemademadema “3“3, 3rdrd spaspa spa icingcingcing”, congeongeonge tistivstivstive ca dirdiardiardiaccc dd

failure)

AlAlAlAlAlAlAlttetetetetetererererereddddddd vavavavavascscscscsc lululululululararararar cccc capapapapapppacacacacacitititititititananananancecececece SSeSeSeSe( p isis:sis:sis:sis: hshushushu shu tintinntinnting, vg dasodasodasodasodil tilatilatilatilat tiatioatioatioatio Hn. H tepatepatepatepatp orenorenorenoren lalalalal

dsyndy rome HRS HRS)))

HHyHypopottetensnsiioionn

IInInInIntttrtrtrtriinininin iisisisisicccc rerererenanananallllll fffafafafaillililililurururureeee

GlGlGlGlomomomomerererer lululularararar(G(G(G(Glolololomemememerurururulolololonenenenephphphphriririrititititis)s)s)s)

Tubular IInInInttetetetersrsrsrstitititititititi lalalal(A(A(A(Aututututoioioioimmmmmmmmununununeeee, ttt toxoxoxoxicicicic, ininininfefefefectctctctioioioiousususus))))

VVaVaVascscscsc lululularararar

Ischaemic(E(EE(E(E txtxxtxtxtrerememe p pprere r renen lalaalalala , ,

sepsis, , papapapap ncncncncn rererereatatatatatititititi isisisisis)))))

Toxic

Intrinsic toxins (R(R(R(R( hahahahabdbdbdbdomomomyoyoyoy lylylylyysisisisiss,,, m mmasasassisisisiveveve hhh haeaeaemomomolylylylyysisisisisss

ttututumomomoururur lll lysysysiisisis, mymymy lelelelomomom )a)a)a)

Extrinsic toxins (RR(R(R(R( adadadadadiooioioio ccc cononontrtrtrtrtrasasasttt,t,, dddd drurururugsgssgsg ,, , anananantititititibibibibibiottotototicicicicics)s)s)s)s))

LLaLaLaLaLaLargrgrgrgrgrggeeeeee vevevevevevessssssssssss lelelelelelel(R(R(R(R(Renenovovasascucullalalalarr didididididiseseasasee,,

atatatatheheheherorororoememememboboboolililic)c)c)c))

SSmSmSmSmSm lalalalalalllllll vevevevevessssssssss lelelelelel (V((V(Vasasasascucucuculilililititititissss,, HHH HRSRSRSRS,

rerererenononovavvavascsscscsc lulululularrarar, , mamamamalililililignngngnannanantttt tt hypertension)

FIGURE 4.2. Causes of intrinsic renal failure (red, commoner causes on the ICU). HRS, hepatorenal syndrome.d

22 S. Blakeley

pathophysiology of ischemic ATN is reviewed elsewhere (21, 22), but an alteration in glomerular hemo dynamics with marked afferent arteriolarrenal vasoconstriction causes a fall in glomerularfiltration pressure and subsequently causes isch-fiemia. This particularly affects the outer medulla.Tubular damage leads to loss of normal cell-to-celladhesion and allows back leakage of filtrate intofithe interstitium. Shed cells precipitate, withprotein obstructing the tubules and further com-promising tubular function. Local infl ammatory flmediators respond to cell injury, perpetuating theprocess.

ATN can also develop secondary to a variety of intrinsic or extrinsic renal toxins. Vascular causes of renal failure may be present at a prerenal or intrarenal level, and should be considered in vas-culopaths. The more classic glomerular causes forintrinsic renal failure are seen less frequently on the ICU, but are important to recognize becausethey require specific treatment.fi

Postrenal or Obstructive Renal Failure

Obstruction can occur at any level of the urinary collecting system and can be caused by intrinsic (e.g., stones, tumor) or extrinsic causes (e.g., sur-rounding or infi ltrating tumor, large inflfi ammatory flabdominal aortic aneurysms). Obstruction is an infrequent cause of AKI on the ICU but is impor-tant to be excluded in all cases.

Complications of AKI

AKI is a systemic disease, having effects on practi-cally all organ systems (23). It is becoming increas-ingly recognized that there is “cross talk” betweenthe injured kidney and other organs through the release of proinfl ammatory cytokines. Complica-fltions related to other failing organs may be seen, but complications specific to the failing kidney arefias follows.

Retention of Uremic Toxins

Accumulation of toxins, including urea, can leadto nausea, vomiting, drowsiness, a bleeding ten-dency, uremic fl ap, and, rarely, coma (uremicflencephalopathy) and a pericardial rub.

Volume Overload

Salt and water retention occurs early, and is acommon reason for initiating RRT on the ICU (16). Volume overload may have deleterious effects on cardiac and respiratory function, with thedevelopment of peripheral edema affecting wound healing and pressure areas.

Acidosis

There is retention of organic anions (e.g., phos-phate) and reduced production of bicarbonate by the failing tubules. In critically ill patients,this may be aggravated by the presence of a non-renal acidosis, for example, lactic acidosis fromsepsis and respiratory acidosis from respiratory failure.

Electrolyte and Mineral Disturbances

Hyponatremia, hyperkalemia, and hyperphospha-temia are commonly seen.

Anemia

Anemia can develop because of inappropriate levels of erythropoietin (decreased synthesis) orincreased red cell fragility, causing premature red cell destruction. Uremia is also associated withplatelet dysfunction and increased risk of gastro-intestinal bleeding.

Immunosuppression

Renal failure itself can impair humoral and cellu-lar immunity, putting the patient at risk of infec-tious complications.

Metabolic Consequences

Hyperglycemia occurs because of peripheral in-sulin resistance and increased hepatic gluconeo-genesis. Protein catabolism is also activated.

Drug Accumulation

Renal failure may be secondary to drugs, but, as GFR falls, renal clearance of drugs and their

4. Acute Kidney Injury 23

metabolites also falls. Renal failure may be exac-erbated by drug accumulation, or other side effectscan develop, such as morphine metabolites leadingto respiratory depression.

Investigation of the Cause of Renal

Failure (Table 4.1)

Laboratory Tests (Table 4.2)

Urinalysis

A standard dipstick for blood and protein should be preformed and a fresh sample spun for casts:hyaline casts (nonspecific markers of renal injury), fibrown/cellular casts (ATN), and red cell casts(acute glomerulonephritis). An estimation of protein excretion may be needed, either a 24-hoururine collection or a spot urine protein-creatinineratio, depending on local resources.

Urine osmolality and urinary electrolytes canbe used to help distinguish prerenal failure fromintrinsic kidney disease (Table 4.3). They should

be interpreted in light of the clinical setting, but can act as another tool in the assessment of intra-vascular volume status.

Radiological Investigations

A chest x-ray will help assess volume status, butpatchy infi ltrates may also represent pulmonary fihemorrhage, as seen in certain forms of vasculitis. A renal ultrasound scan should be performed;the timing will depend on the likely cause of renal failure and the patient’s clinical state. Furtherimaging should be guided by the clinical scenario.

Conclusion

AKI is a significant condition affecting critically fiill patients on the ICU. It is a systemic processaffecting all organs, and has a major impact on patient morbidity and mortality. It is, therefore, important to be able to promptly recognize its development and institute the appropriate inves-tigations to guide treatment.

TABLE 4.1. History and examination in AKI

A full history should be take with reference to the following:• Presence of risk factors: known chronic renal disease, advanced age, diabetes mellitus, ischemic heart disease, hypertension, peripheral vascular

di li di t hi h i kdisease, liver disease, recent high-risk surgery• Previous episodes of renal failure Previous episodes of renal failure• Family history of renal disease• Rashes, joint aches, sinusitis, and hemoptysis suggesting a systemic condition• Review blood pressure and anesthetic charts for periods of profound or prolonged hypotension in relation to the patients usual blood pressure• Review fluid balance charts considering hidden losses, such as sweating or “third spacing.” The sudden onset of anuria suggests obstruction or a

t t hi l t R b th t di ti k th i t t l k “ tifi i ll d”catastrophic vascular event. Remember that diuretics can make the urine output look “artificially good”• Drug charts should be reviewed for intravenous contrast, chemotherapeutic agents, analgesics, antibiotics, and herbal remedies, including any

new medications taken in the past month• Determine baseline creatine and pattern of change (i.e., sudden jump or gradual decline). The rate of change may be more important than the

absolute value• Review any previous urinalyses for previous hematuria or proteinuria that may suggest chronic disease

A full examination should be performed with reference to the following:• Pressure: mean arterial pressure in relation to the patients usual readings• Flow: cardiac output studies and/or markers of end organ perfusion (e.g., lactate)• Volume: overall volume status as well as intravascular volume status• Patent vessels: evidence of generalized vascular disease• Rashes or splinter hemorrhages suggesting vasculitis, cholesterol emboli, or infective endocarditis• Palpable bladder or kidney, suggesting obstruction• Raised intra-abdominal pressure or tense limbs, suggesting compartment syndrome

24 S. Blakeley

TABLE 4.2. Laboratory investigations for AKI

Finding Comment

Anemia Normochromic, normocytic suggests chronicityThrombocytopenia + microangiopathic hemolytic anemia Consider hemolytic uremic syndrome, thrombotic thrombocytopenic purpura, disseminated intravascular coagulation (DIC)Neutrophilia, “left shift” + thrombocytopenia Consider sepsisEosinophilia Consider vasculitis, allergic interstitial nephritisAbnormal coagulation profile Sepsis, DIC, hepatorenal syndrome, systemic lupus erythematosus (SLE)Elevated urea and creatinine Will rise with any cause of renal failure. A normal serum creatinine does not exclude the presence of renal dysfunction, and conversely an elevated creatinine may underestimate the degree of renal dysfunctionElevated serum Cystatin C A newer marker of renal dysfunction, but needs further evaluation on ICU patients. Freely filtered at the glomerulus, and fully metabolized by proximal tubular cells, if GFR falls, levels riseHypercalcemia Elevated in malignancy (including hematological). Calcium may be high, low, or normal in chronic kidney disease (CKD)Hyperphosphatemia Extremely elevated in rhabdomyolysis and tumor lysis syndrome, but will rise with any cause of low GFR. May be normal or high in CKDElevated creatinine kinase RhabdomyolysisAbnormal liver function tests Consider hepatorenal syndrome; sepsis; vasculitis; and hemolysis, elevated liver enzymes, and low platelet count (HELLP) syndromeElevated uric acid Seen in preeclampsia but will rise with any fall in GFRAbnormal immunoglobulins Look for myeloma and other hematological malignanciesElevated antinuclear factor, double-stranded DNA, Investigate further for systemic diseases, such as SLE and vasculitis antineutrophil cytoplasmic antibodies (ANCA), antiglomerular basement membrane antibodies, and low complementPositive virology/serology/microbiology Certain forms of renal injury seen with specific infections, e.g., hepatitis B/C, HIV, leptospirosis, verotoxin-producing Escherichia coli

TABLE 4.3. Urinary findings in prerenal failure and ATNa

Prerenal Intrinsic

Urine Na <20 mmol/L >40 mmol/LUrine : plasma (U : P) urea >20 <10 ratioU : P creatinine ratio >40 <10–20U : P osmolality >2.1 <1.2Specific gravity >1.020 <1.010Urine osmolality >500 <400Urine osmolality High (>serum + Low (<serum +

100 mOsm/L) 100 mOsm/L)FE sodiumb <1% >1–2%

aThese results should be interpreted with caution in patients who have had diuretics, large volume resuscitation, the elderly, or patients with chronicrenal failure.bFE, fractional excretion of sodium: in a prerenal state, sodium is actively reabsorbed by working tubules to maintain intravascular volume. The kidney will, therefore, produce concentrated urine with a low concentra-tion of sodium. Creatinine is still excreted, but relatively less sodiumappears in the urine. Hence, if the tubules are intact, the amount of sodiumexcreted compared with creatinine (fractional excretion) falls.

FE NaUrine Na plasma creatinine

Plasma Na Urine creatinine= 100

References

1. Kellum JA, Levin N, Bouman C, Lameire N. Devleop-ing a concensus classification for acute renal failure.fiCurr Opin Crit Care. 2002; 8: 509–514.

2. Tillyard A, Keays R, Soni N. The diagnosis of acuterenal failure in intensive care: mongrel or pedigree?Anaesthesia. 2005; 60: 903–914.

3. Hoste EA, Kellum JA. Acute kidney injury: epidemi-ology and diagnostic criteria. Curr Opin Crit Care. 2006; 12: 531–537.

4. Mehta RL, Chertow GM. Acute renal failure defini-fitions and classifi cation: Time for a change? fi J Am Soc Nephrol. 2003; 14: 2178–2187.

5. Bellomo R, Ronco C, Kellum JA et al. Acute renalfailure—Definition, outcome measures, animalfimodels, fl uid therapy and information technology flneeds. The Second International Consensus Con-ference of the Acute Dialysis Quality Initiative(ADQI) group. Crit Care. 2004; 8: R204–R212.

6. Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. fi Am J Kid Dis. 2002; 39(5): 930–936.

4. Acute Kidney Injury 25

7. Hou SH, Bushinsky DA, Wish JB et al. Hospital-acquired renal insuffi ciency: a prospective study. fiAm J Med. 1983; 74(2): 243–248.

8. Uchino S, Kellum JA, Bellomo R et al. Beginning andEnding Supportive Therapy for the Kidney (BESTKidney) Investigators. Acute renal failure in criti-cally ill patients: a multinational, multicenter study.JAMA. 2005; 294(7): 813–818.

9. Uchino S, Bellomo R, Goldsmith D et al. An assess-ment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit Care Med. 2006; 34(7): 1913–1917.

10. Hoste EA, Clermont G, Kersten A et al. RIFLE crite-ria for acute kidney injury are associated with hos-pital mortality in critically ill patients: a cohort analysis. Crit Care. 2006; 10(3): R73.

11. Kuitunen A, Vento A, Suojaranta-Ylinen R, Pettila V. Acute renal failure after cardiac surgery: evaluation of the RIFLE classifi cation.fi Ann Thoracic Surg. 2006;81(2): 542–546.

12. Chertow GM, Levy EM, Hammermeister KE et al. Independent association between acute renal failureand mortality following cardiac surgery. Am J Med.1998; 104(4): 343–348.

13. Bagshaw SM, Laupland KB, Doig CJ et al. Prognosis for longterm survival and renal revoery in critically ill patients with severe acute renal failure: A popu-lation based study. Crit Care. 2005; 9: R700–709.

14. Metnitz PG, Krenn CG, Steltzer H et al. Effect of acute renal failure requiring renal replacement

therapy on outcome in critically ill patients. Crit Care Med. 2002; 30: 2051–2058.

15. Brivet FG, Kleinknecht DJ, Loirat P, Landais PJ.Acute renal failure in intensive care units-causes, outcome, and prognostic factors of hospital mortal-ity; a prospective, multicenter study. Crit Care Med.1996; 24(2): 192–198.

16. Silvester W, Bellomo R, Cole L. Epidemiology, man-agement, and outcome of severe acute renal failure of critical illness in Australia. Crit Care Med. 2001;29: 1910–1915.

17. Korkeila M, Ruokonen E, Takala J. Cost of care, long term prognosis and quality of life in patients requir-ing renal replacement therapy during intensivecare. Int Care Med. 2000; 26: 1824–1831.

18. Bagshaw SM. Epidemiology of renal recovery afteracute renal failure. Curr Opin Crit Care. 2006; 12:544–550.

19. Rangel-Frausto MS, Pittet D, Costigan M et al. Thenatural history of the systemic infl ammatory flresponse syndrome (SIRS). A prospective study.JAMA. 1995; 273(2): 117–123.

20. Schrier RW, Wang W. Acute renal failure and sepsis.N Engl J Med. 2004; 315: 159–169.

21. Schrier RW, Wang W, Poole B Mitra A. Acute renalfailure: definitions, diagnosis, pathogenesis and fitherapy. J Clin Invest. 2004; 114: 5–14.

22. Lameire N, Van Biesen W, Vanholder R. Acute renal failure. Lancet. 2005; 365: 417–430.

23. Druml W. Acute renal failure is not a “cute” renal failure! Intensive Care Med. 2004; 30: 1886–1890.

26

5Medical Management of Acute Renal Failure

Nerina Harley

Patients with renal disease have a variety of clinical presentations:

• Asymptomatic• Symptoms directly referable to the kidney,

e.g., hematuria, flank painfl• Extrarenal symptoms, e.g., edema, hyperten-

sion, uremic symptoms, consequences of hyperkalemia

• Manifestations of the underlying pathology or etiology, e.g., sepsis, hypotension, rhabdo-myolysis, systemic vasculitis

Investigations

Assessment of Renal Function

Although they lack sensitivity and specificity fiserum creatinine and urine output are the most useful parameters in clinical practice. Glomerular filtration rate (GFR) assessment may be diffifi cult fiin nonsteady-state conditions. Comparison withprevious results is vital in determination of base-line function, time course, and progression.

Urinalysis

Urine examination may demonstrate granular and epithelial casts in ATN, eosinophils in acute inter-stitial nephritis, and red cell casts in acute vascu-litis or glomerulonephritis and/or proteinuria.There are no prospective studies of the predictivevalue of urine sediment in ATN. Urinary elec-trolytes are of limited value in most clinical situations because of the confounding effects of

Acute renal failure (ARF) is both common and associated with signifi cant mortality in the inten-fisive care unit (ICU) setting (1). With an impact on length of stay, likelihood of survival until dis-charge, and cost of care, it is vital that the increas-ing body of evidence in critical care nephrology is used to refine defifi nitions, diagnosis, preventivefistrategies, investigations, and management of these patients (2).

The medical management of ARF relies on thebasic tenets of diagnosis, elimination of reversi-ble factors, amelioration of exacerbating factors,treatment of complications, and optimization of the “kidney’s environment” to provide maximalrecovery.

Diagnosis of ARF

Major causes of renal disease are divided, forsimplicity, into three areas:

• Prerenal causes, caused by volume depletion with or without relative hypotension (reduced renal perfusion)

• Intrinsic renal causes, caused by vasculitis, glo-merular disease, and tubulointerstitial disease

• Postrenal or obstructive causes

The most common causes of ARF have beenfound to be acute tubular necrosis (ATN) (45%),prerenal (21%), acute on chronic (13%), obstruc-tion (10%), glomerulonephritis (GN), or vasculitis (4%) (3).

Risk factors for the development of ARF are given in Table 5.1.

5. Medical Management of Acute Renal Failure 27

treatments such as diuretics. Urine volume is of little diagnostic value, although little or no outputis acutely useful with causes of anuria, including shock, bilateral urinary obstruction, renal corticalnecrosis, and bilateral vascular occlusion.

Other Markers of Renal Injury

These have been described but are not currently generally available in the acute setting. For example, serum cystatin C has been recently proposed as a marker of ARF (4), predicting ARF by at least 24 hours.

Radiology

Renal ultrasound is the modality of choice toexclude obstruction. Reduced renal size and corti-cal thinning (although preserved in diabeticnephropathy) is indicative of chronic renal impair-ment. A helical computed tomographic (CT) scan may be useful in urolithiasis, but there are risks of secondary injury with radiocontrast.

Renal Biopsy

Renal biopsy is considered when noninvasive eval-uation has not established the diagnosis (5); themajor indications include isolated hematuria withproteinuria, nephrotic syndrome, acute nephritic syndrome, and unexplained ARF. Percutaneous biopsy is most commonly performed and the

inherent risks of bleeding should be weighed upin the setting of the risk-to-benefit ratio.fi

Note: The “gold standard” for the diagnosis of a prerenal cause of ARF is resolution of renalimpairment in response to fl uid challenge. This is flin contrast to significant ATN, in which there is fiprolonged time to resolution.

Primary Prevention

Primary prevention of ARF in the critically ill withor without baseline risk factors consists of avoid-ance, amelioration, and treatment of these factorswherever possible. Strategies can be divided intononpharmacological, for example, fluid adminis-fltration to reduce the risk of contrast inducednephropathy (6), and pharmacological. To date, nopharmacological strategies have conclusively dem-onstrated prevention of ARF from any insult (7).

Loop Diuretics

Although oliguria (<400 mL/24 h) is common in ATN, anuria is rare and other etiologies must be considered. Nonoliguric patients have a betterprognosis than oliguric patients in terms of a greater residual GFR (8), lower peak serum creati-nine, and dialysis-free survival at 21 days (9, 10),possibly reflecting less severe renal injury. A clini-flcal issue of the use of loop diuretics often arises in increasing the urine output of oliguric patients.

Experimentally, loop diuretics reduce activesodium chloride transport in the thick ascending limb of the loop of Henle, decreasing energy requirements and, thus, protecting the cell in the setting of decreased energy availability. Only anecdotal human evidence suggests that the use of loop diuretics may be beneficial in the fifi rst 24 fihours in fl ushing tubular casts. There is no evi-fldence of benefi t in established ATN on duration fiof renal failure, requirement for dialysis, or sur-vival (11). The increase in urine output in this setting is caused by decreased tubular reabsorp-tion in residual functioning nephrons (not re-cruitment of nonfunctioning nephrons), volume expansion (initial sodium retention), and ureaosmotic diuresis.

A number of studies have suggested worsened outcomes in ATN with the use of loop diuretics in

TABLE 5.1. Risk factors for ARF

Risk factor Selected reference(s)

Older age Nash et al., 2002 (43)Diabetes Parfrey et al., 1989 (44).Underlying renal insufficiencyCardiac failureSepsis Brun-Buisson et al., 2004 (45)

Ricci et al., (46)Absolute or relative hypovolemiaHepatic failure Han and Hyzy, 2006 (47)Nephrotoxins (including Barrett et al., 1993 (48)

high-osmolality Aspelin et al., 2003 (49).radiocontrast agents) McCullogh et al., 2006 (50)

Cardiopulmonary bypass with Mangano et al., 1998 (51)aortic cross-clamping Chertow et al., 1998 (52)(particularly valve surgery)

Nonrenal organ transplantation Lima et al., 2003 (53)Abdominal compartment syndrome McNelis et al., 2003 (54)

28 N. Harley

the setting of contrast media (9, 12) and cardiac surgery. A systematic review (13, 14) comparingfluids alone with diuretics found no evidence of flimprovement in survival, incidence of ARF, orneed for renal replacement therapy (RRT). Deaf-ness, possibly permanent, is a known complica-tion of high-dose loop diuretics.

Mannitol

Mannitol may preserve mitochondrial function by minimizing postinjury edema. Human trials have failed to show benefi t in reducing ARF with man-finitol versus fl uid alone in rhabdomyolysis (15),flcardiac surgery (16), and vascular and biliary tractsurgery (17). A trend toward harm was noted inthe prevention of contrast nephropathy (12).

Dopamine Agonists

Dopamine has a number of effects in the kidney via dopamine A1 and A2 receptors. In the proxi-mal tubule, dopamine, via the generation of cyclicAMP, decreases Na+-H+ exchange and the Na+-H+-ATPase pump, thus, decreasing sodium reabsorp-tion. In the collecting tubules, this effect onNa+-H+-ATPase and decreased aldosterone secre-tion reduces sodium reabsorption (18).

When infused in doses of 0.5 to 3 μg/kg/min,dopamine causes afferent and efferent glomerulararteriolar dilatation, increasing blood flow withfllittle or no increase in GFR. At higher concentra-tions, dopamine causes vasoconstriction via α-adrenergic receptors.

There is no evidence in human studies for a “renal protective effect” of dopamine.

• In 1994, Baldwin et al. studied the effect of post-operative low-dose dopamine on renal function after major elective vascular surgery. Patientswere administered saline or saline plus dopa-mine as fl uid replacement. No difference in renalflfunction was demonstrated between the twogroups (19).

• In the North American Study of the Safety and Efficacy of Murine Monoclonal Antibody fito Tumor Necrosis Factor for the Treatmentof Septic Shock (NORASEPT) II study, 400patients with septic shock and oliguria nonran-domly received no, low-, or high-dose dopamine.

The incidence of ARF and requirement for RRT was not significant different betweenfigroups (20).

• A large randomized controlled trial of low-dosedopamine in 328 critically ill patients withimpaired creatinine, oliguria, and at least twosystemic infl ammatory response syndromefl(SIRS) criteria failed to show any benefit in fiprogression, need for RRT, or death (21).

• Studies of fenoldopam, a relatively selective dopamine A1 receptor agonist, although shown to increase sodium excretion and renal bloodflow in healthy and hypertensive patients, has flnot shown benefi t in ARF in the critically illfi(22).

N-Acetylcysteine

N-Acetylcysteine (NAC) has been shown in a NNnumber of studies to decrease the incidence of contrast nephropathy in high-risk patients (23, 24), but without improvement in RRT require-ment or survival. Importantly, NAC may decrease creatinine via activation of creatinine kinase (25)but not GFR. Promising studies have led to theintroduction of protocols in many institutions forthe prophylactic use of NAC in the prevention of contrast-induced nephropathy. With few side effects, low cost, and ease of administration (oral or intravenous), many centers have erred on theside of possible benefit in the face of lack of hard fievidence of long-term benefit (26).fi

Others

Trials have failed to demonstrate benefi t of fi natri-uretic peptides (10, 27) or adenosine agonists (28). Experimental therapies, such as antioxidants and erythropoietin are unproven in humans.

In the absence of further evidence, fluids, andflpossibly NAC, are the gold standards of interven-tional strategies.

Supportive Strategies

Fluid resuscitation and correction of hypotension are clearly essential. There is no evidence of advantage of one particular inotrope over another.

5. Medical Management of Acute Renal Failure 29

No evidence exists that reversing hypotensionwith noradrenaline compromises mesenteric or renal blood fl ow (29).fl

Van den Berghe demonstrated that tight glucose control with intravenous therapy improved out-comes in critically ill patients, including decreasedincidence of ARF (30, 31).

Nutritional support of the critically ill is the tbasic standard of care, although not alwaysachieved. Early enteral nutrition is supported by meta-analyses of Level II trials; benefits include fipreservation of muscle mass, the maintenance of the gastrointestinal mucosal barrier and immunestatus, and a possible reduction in multiorgandysfunction (32–34). A Level I trial is currently planned by the Australian and New Zealand Inten-sive Care Society (ANZICS) clinical trials group comparing early enteral nutrition with standardcare in 1470 critically ill patients intolerant of early enteral nutrition (www.actr.org.au).

There is no evidence to support prophylactic hemofiltrationfi to prevent contrast nephropathy, despite filtration removal of contrast.fi

Early recognition and adequate management of the deteriorating clinical condition of a patient is fundamental in the prevention of morbidity andmortality, including ARF. In a hospital-basedstudy of a medical emergency team (MET) system,Bellomo et al. demonstrated reduced incidence of postoperative adverse outcomes, mortality rate, and mean hospital length of stay (35). This wasnot validated in a larger cluster-randomized con-trolled trial of 23 hospitals with no effect on inci-dence of cardiac arrest, unplanned admissions toICU, or unexpected death (36). Nevertheless, the principles of early recognition, monitoring, andadequate response remain fundamental.

Postinjury Prevention of ARF

Secondary renal injury occurs after the primary insult has triggered the initial injury to thekidneys.

Strategies for postinjury prevention of ARFoverlap with those of primary prevention (37) as described above. They include maintenance of adequate intravascular volume, cardiac output,mean arterial blood pressure (MAP), avoidance of further insult, and supportive strategies.

Assessment and Correction of

Volume Depletion

Clinical signs of relative or absolute volume deple-tion include loss of tissue turgor, hypotension, postural hypotension, and decreased venous pres-sure (reduced jugular venous pressure). Althoughleft ventricular end diastolic pressure (LVEDP)is the most important determinant of left ventricu-lar output and, thus, tissue perfusion, centralvenous pressure is useful because it has a directrelationship to LVEDP, with the exceptions of pure left-sided or pure right-sided failure (cor pulmonale). Other clinical manifestations may be specifi c to the source of depletion (lossesfior third space sequestration), type of fluid lost, fland to the associated electrolyte and acid-base abnormalities.

Recently, Vincent proposed a protocol forroutine fluid challenge with defifl ned rules based fion clinical response to the volumes infused, allowing for prompt deficit correction while fiminimizing risks of fluid overload (38).fl

Maintenance of Adequate MAP and

Cardiac Output

There is currently insufficient data to recommend fifirm therapeutic targets, suffifi ce to say, Level III fievidence suggests failure to maintain systolic blood pressure greater than 80 mmHg or MAP greater than 50 mmHg is associated with increased risk of developing ARF (39). A low cardiac outputis a major risk factor for ARF after cardiac surgery (40), but supranormal cardiac output has nobeneficial effect.fi

Avoidance of Further Insults

Appropriate dosage of medication and avoidance(or protective strategies) of nephrotoxins is advised.

Renal Replacement Therapy

Intermittent versus continuous renal replacement strategies and dose delivery are discussed in a later chapter.

30 N. Harley

Summary

Measures of severity of illness are limited whenapplied to patients with ARF and, at present, noneare currently adequate to predict mortality of ARF(41). In the light of available evidence, guidelines for the management of ARF have been formulatedand given in Table 5.2. Admiral efforts to clearly defi ne levels of injury in ARF and physiologicalfiend points (42) will enable further research into prevention, amelioration, and management of ARF and, thus, impact on the significant associ-fiated morbidity and mortality.

References

1. Levy EM, Viscoli CM, Horwitz RI. The effect of acute renal failure on mortality. A cohort analysis. JAMA. 1996; 275(19): 1489–1494.

2. Kellum JA. Acute renal failure, interdisciplinary knowledge and the need for standardization. Curr Opin Crit Care. 2005; 11(6): 525–526.

3. Liano G, Pascual J. Acute renal failure. Madrid AcuteRenal Failure Study Group. Lancet. 1996; 347(8999):479.

4. Herget-Rosenthal S, Marggraf G, Husing J, et al.Early detection of acute renal failure by serum cys-tatin C. Kidney Int. 2004; 66(3): 1115–1122.

5. Madaio MP. Renal biopsy. Kidney Int. 1990; 38(3):529–543.

6. Mueller C, Buerkle G, Buettner HJ, et al. Prevention of contrast media-associated nephropathy: ran-domized comparison of 2 hydration regimens in 1620 patients undergoing coronary angioplasty.Arch Intern Med. 2002; 162(3): 329–336.

7. Kellum JA, Leblanc M, Gibney RT, et al. Primary prevention of acute renal failure in the critically ill. Curr Opin Crit Care. 2005; 11(6): 537–541.

8. Rahman SN, Conger JD. Glomerular and tubularfactors in urine fl ow rates of acute renal failure flpatients. Am J Kidney Dis. 1994; 23(6): 788–793.

9. Lassnigg A, Donner E, Grubhofer G, et al. Lack of renoprotective effects of dopamine and furosemide during cardiac surgery. J Am Soc Nephrol. 2000; 11(1): 97–104.

TABLE 5.2. Summary of guidelines for management of ARF

Make diagnosis

Exclusion of prerenal causes E.g., volume depletion, cardiac and liver disease, nephrotoxinsExclusion of postrenal causes Renal ultrasoundExclusion of intrinsic renal causes E.g., review urinary sediment, consider renal biopsyEvaluation of urinary electrolytes Only in the absence of diuretics

Treat reversible causes

Volume resuscitation Maintain fluid balance but avoid overloadBlood pressure support Inotropic if necessary, however no validated physiological targets or end points have been establishedTreat electrolyte complications E.g., hyperkalemiaNo dopamine

Address/avoid exacerbating factors

Treatment of underlying etiology E.g., treat sepsis, further investigation if diagnosis unclear, surgical referral if appropriateAdjust medication dosage Discontinue any nephrotoxic drugsAvoid further renal insult E.g., minimize contrast-induced injury with fluids and consider NAC

Optimize “kidney’s environment”

Maintain renal perfusion Maintain fluid balance, blood pressure, and cardiac outputAdequate nutritionGlucose controlAppropriate RRT Timely introduction of RRT, avoidance of complications (e.g., hypotension, line-related sepsis, biocompatible dialysis membranes), and appropriate dose of dialysis

Other

Constantly review diagnosis, which may Consider further investigations, e.g., renal biopsy and radiological investigations, as appropriate multifactorial in natureAppropriate medical review E.g., MET resources, nephrology input

5. Medical Management of Acute Renal Failure 31

10. Allgren RL, Marbury TC, Rahman SN, et al. Anarit-ide in acute tubular necrosis. Auriculin AnaritideAcute Renal Failure Study Group. N Engl J Med. 1997; 336(12): 828–834.

11. Cantarovich F, Rangoonwala B, Lorenz H, et al. High-dose furosemide for established ARF: aprospective, randomized, double-blind, placebo-controlled, multicenter trial. Am J Kidney Dis. 2004; 44(3): 402–409.

12. Solomon R, Werner C, Mann D, et al. Effects of saline, mannitol, and furosemide to prevent acutedecreases in renal function induced by radiocon-trast agents. N Engl J Med. 1994; 331(21): 1416–1420.

13. Kellum JA. Diuretics in acute renal failure: pro-tective or deleterious. Blood Purif 1997; 15(4–6): ff319–322.

14. Kellum JA. The use of diuretics and dopamine in acute renal failure: a systematic review of the evi-dence. Crit Care (Lond). 1997; 1(2): 53–59.

15. Homsi E, Barreiro MF, Orlando JM, Higa EM.Prophylaxis of acute renal failure in patientswith rhabdomyolysis. Ren Fail. 1997; 19(2): 283–288.

16. Ip-Yam PC, Murphy S, Baines M, et al. Renal func-tion and proteinuria after cardiopulmonary bypass:the effects of temperature and mannitol. Anesth Analg. 1994; 78(5): 842–847.

17. Gubern JM, Sancho JJ, Simo J, Sitges-Serra A. A ran-domized trial on the effect of mannitol on postop-erative renal function in patients with obstructivejaundice. Surgery. 1988; 103(1): 39–44.

18. Denton MD, Chertow GM, Brady HR. “Renal-dose” dopamine for the treatment of acute renal failure: scientifi c rationale, experimental studies and clini-fical trials. Kidney Int. 1996; 50(1): 4–14.

19. Baldwin L, Henderson A, Hickman P. Effect of post-operative low-dose dopamine on renal functionafter elective major vascular surgery. Ann Intern Med. 1994; 120(9): 744–747.

20. Marik PE, Iglesias J. Low-dose dopamine does notprevent acute renal failure in patients with septicshock and oliguria. NORASEPT II Study Investiga-tors. Am J Med. 1999; 107(4): 387–390.

21. Bellomo R, Chapman M, Finfer S, et al. Low-dosedopamine in patients with early renal dysfunction:a placebo-controlled randomised trial. Australian and New Zealand Intensive Care Society (ANZICS) Clinical Trials Group. Lancet. 2000; 356(9248): 2139–2143.

22. Tumlin JA, Finkel KW, Murray PT, et al. Fenoldopammesylate in early acute tubular necrosis: a random-ized, double-blind, placebo-controlled clinical trial. Am J Kidney Dis. 2005; 46(1): 26–34.

23. Birck R, Krzossok S, Markowetz F, et al. Acetylcys-teine for prevention of contrast nephropathy: meta-analysis. Lancet. 2003; 362(9384): 598–603.

24. Pannu N, Manns B, Lee H, Tonelli M. Systematicreview of the impact of N-acetylcysteine on contrast nephropathy. Kidney Int. 2004; 65(4): 1366–1374.

25. Genet S, Kale RK, Baquer NZ. Effects of free radicalson cytosolic creatine kinase activities and protec-tion by antioxidant enzymes and sulfhydryl com-pounds. Mol Cell Biochem. 2000; 210(1–2): 23–28.

26. Bagshaw SM, McAlister FA, Manns BJ, Ghali WA.Acetyl-cysteine used in the prevention of contrast-induced nephropathy: a case study in the pitfalls inthe evolution of evidence. Arch Intern Med. 2006 Jan 23; 166(2): 161–166.

27. Lewis J, Salem MM, Chertow GM, et al. Atrial natri-uretic factor in oliguric acute renal failure. Anarit-ide Acute Renal Failure Study Group. Am J Kidney Dis. 2000; 36(4): 767–774.

28. Kramer BK, Preuner J, Ebenburger A, et al. Lack of renoprotective effect of theophylline during aorto-coronary bypass surgery. Nephrol Dial Transplant.2002; 17(5): 910–915.

29. Levy B, Bollaert PE, Charpentier C, et al. Compari-son of norepinephrine and dobutamine to epineph-rine for hemodynamics, lactate metabolism, and gastric tonometric variables in septic shock: a pro-spective, randomized study. Intensive Care Med.1997; 23(3): 282–287.

30. Van den Berghe G, Wouters P, Weekers F, et al. Inten-sive insulin therapy in the critically ill patients. N Engl J Med. 2001; 345(19): 1359.

31. Van den Berghe G, Wouters PJ, Bouillon R, et al. Outcome benefi t of intensive insulin therapy in theficritically ill: Insulin dose versus glycemic control.Crit Care Med. 2003; 31(2): 359–366.

32. Doig GS, Simpson F. Evidence-based guidelines for nutritional support of the critically ill: results of a bi-national guideline development conference.Sydney: EvidenceBased.net; 2005. Download www.EvidenceBased.net.

33. Dhaliwal R, Heyland DK. Nutrition and infection in the intensive care unit: what does the evidence show?Curr Opin Crit Care. 2005; 11(5): 461–467.

34. Doig GS, Simpson F. Early enteral nutrition in the critically ill: do we need more evidence or better evidence? Curr Opin Crit Care. 2006; 12(2): 126–130.

35. Bellomo R, Goldsmith D, Uchino S, Buckmaster J, Hart G, Opdam H, Silvester W, Doolan L, Gutteridge G. Prospective controlled trial on the effect of medical emergency team on postoperative morbid-ity and mortality rates. Crit Care Med. 2004; 32(4): 916–921.

32 N. Harley

36. Hillman K, Chen J, Cretikos M, Bellomo R, Brown D, Doig G, Finfer S, Flabouris A; MERIT study inves-tigators. Introduction of the medical emergency team (MET) system: a cluster-randomised con-trolled trial. Lancet. 2005; 365 (9477): 2091–2097.

37. Bellomo R, Bonventre J, Macias W, Pinsky M. Man-agement of early acute renal failure: focus on post-injury prevention. Curr Opin Crit Care. 2005; 11(6):542–547.

38. Vincent JL, Weil MH. Fluid challenge revisited. Crit Care Med. 2006; 34(5): 1333–1337.

39. Kohli HS, Bhaskaran MC, Muthukumar T, et al. Treatment-related acute renal failure in the elderly:a hospital-based prospective study. Nephrol Dial Transplant. 2000; 15(2): 212–217.

40. Suen WS, Mok CK, Chiu SW, et al. Risk factors fordevelopment of acute renal failure (ARF) requiringdialysis in patients undergoing cardiac surgery.Angiology. 1998; 49(10): 789–800.

41. Uchino S, Bellomo R, Morimatsu H, et al. Externalvalidation of severity scoring systems for acuterenal failure using a multinational database. Crit Care Med. 2005; 33(9): 1961–1967.

42. Murray PT, Le Gall JR, Dos Reis Miranda D, et al.Physiologic endpoints (effi cacy) for acute renalfifailure studies. Curr Opin Crit Care. 2002; 8(6): 519–525.

43. Nash K, et al. Hospital-acquired renal insufficiency. fiAm J Kidney Dis. 2002; 39(5): 930–936.

44. Parfrey P, et al. Contrast material-induced renal failure in patients with diabetes mellitus, renal insuffi ciency, or both. A prospective controlled fistudy. N Engl J Med. 1989; 320(3): 143–149.

45. Brun-Buisson C, et al. EPISEPSIS: a reappraisal of the epidemiology and outcome of severe sepsis in French intensive care units. Intensive Care Med. 2004; 30(4): 580–588.

46. Ricci Z, et al. Practice patterns in the management of acute renal failure in the critically ill patient: an international survey. Nephrol Dial Transplant. 2006; 21(3): 690–696.

47. Han MK, Hyzy R. Advances in critical care manage-ment of hepatic failure and insufficiency. fi Crit Care Med. 2006; 34(9 Suppl): S225–231.

48. Barrett B, et al. Meta analysis of the relative nephro-toxicity of high- and low-osmolality iodinated con-trast media. Radiology. 1993; 188(1): 171–178.

49. Aspelin P, et al. Nephrotoxic effects in high-riskpatients undergoing angiography. N Engl J Med. 2003; 348(6): 491–499.

50. McCullogh PA, et al. Risk prediction of contrast-induced nephropathy. Am J Cardiol. 2006; 98(68): 27K–36K.

51. Mangano C, et al. Renal dysfunction after myocar-dial revascularization: risk factors, adverse out-comes, and hospital resource utilization. Ann Intern Med. 1998; 128(3): 194–203.

52. Chertow G, et al. Independent association between acute renal failure and mortality following cardiac surgery. Am J Med. 1998; 104(4): 343–348.

53. Lima E, et al. Risk factors for development of acuterenal failure after liver transplantation. Ren Fail. 2003; 25(4): 553–560.

54. McNelis J, et al. Abdominal compartment syndrome: clinical manifestations and predictive factors. Curr Opin Crit Care. 2003; 9(2): 133–136.

33

6Acute Renal Failure in the Surgical Patient

Marlies Ostermann

outer medulla is metabolically very active, despite a relatively low oxygen delivery. As a consequence,oxygen extraction in this region is approximately 80% and, in the event of ischemia, this area is often the fi rst part of the kidney to suffer injury.fi

Risk Factors for ARF in the Surgical

Patient (Table 6.1)

Effect of Anesthesia

Most inhalation and intravenous anesthetics cause some degree of cardiac depression and/or vasodi-lation, resulting in lowering of the blood pressure.Similarly, spinal or epidural anesthesia can causehypotension as a result of increased venous ca-pacitance and arterial vasodilation. In response, compensatory changes lead to renal arterial vaso-constriction and increased retention of sodium and water. Maintenance of an adequate intravas-cular volume and blood pressure in the range of patient’s baseline blood pressure is essential toprevent anesthesia-related renal impairment.

Effect of Drugs

Certain medication may be directly nephrotoxic(e.g., aminoglycosides, amphotericin, nonsteroi-dal anti-infl ammatory drugs) or may alter intra-flrenal hemodynamics (e.g., angiotensin-convertingenzyme inhibitors, nonsteroidal anti-inflamma-fltory drugs). In rare situations, a drug-induced interstitial nephritis may occur. This is most com-monly caused by antibiotics or nonsteroidal anal-gesics, but any drug can cause interstitial nephritis.

Acute renal failure (ARF) is a potential complica-tion of any surgical procedure. In general, the risk of ARF is increased in patients with underlying vascular disease, diabetes mellitus, or chronickidney disease. High-risk situations include car-diovascular, hepatobiliary, and trauma surgery, especially if performed as an emergency.

Pathophysiology of ARF

The most common causes of postoperative renal failure are hypotension (relative and absolute)and/or volume depletion. Renal oxygen consump-tion is determined by blood fl ow, making the flkidneys particularly vulnerable to ischemic injury when fl ow is reduced. The kidneys receive 20 to fl25% of the cardiac output. In healthy individuals,renal blood flow is regulated by a complex inter-flplay between intrinsic autoregulation, and hor-monal and neuronal influences. This results in a flrelatively constant renal blood flow when mean flarterial blood pressure (MAP) is 80 to 180 mmHg. Outside of these limits, renal blood flow becomes flpressure dependent, and glomerular filtration ficeases when MAP is less than 40 to 50 mmHg. Inpatients with long-standing hypertension, thismechanism of autoregulation is lost, and renalfunction is pressure dependent, often needing a MAP of greater than 80 mmHg. As a result, these patients need higher blood pressures to maintainglomerular filtration and are particularly vulner-fiable to hypotension or volume depletion.

The whole kidney only extracts less than 10%of the oxygen carried through the kidney. The thick ascending limb of the loop of Henle in the

34 M. Ostermann

Treatment consists of discontinuation of the offending drug and possibly steroid therapy.

Development of Intra-abdominal

Compartment Syndrome

Patients with severe ileus, bowel obstruction,pancreatitis, or after trauma can develop increasedabdominal pressure leading to increased pressure in renal veins and renal parenchyma resulting indecreased renal perfusion. There is a strong cor-relation between intra-abdominal pressure andARF, with oliguria developing when the intra-abdominal pressure is greater than 15 mmHg andanuria developing when pressure is greater than30 mmHg. Intra-abdominal decompression is thetreatment of choice.

Type of Surgery

Cardiac Surgery (Table 6.2)

Twenty to 40% of patients undergoing cardiacsurgery experience a rise in serum creatinine by

25% or a fall in glomerular fi ltration rate by 25%.fiOne to 5% of patients need renal replacement therapy. A combination of patient-specificfi comor-cbidities and factors related to surgery usually con-tribute to the development of ARF.

Patient-specificfi risk factors may not be imme-cdiately obvious. A large study on 7310 patientsundergoing coronary artery bypass grafting(CABG) demonstrated that 29.6% of patients reported being diabetic, but an additional 5.3% of patients were found to have previously undiag-nosed diabetes. Similarly, among patients under-going coronary angiography, 12% of patients werefound to have an undiagnosed renal artery steno-sis of greater than 75%.

Risk factors related to surgery tend to beless predictable, but the risk of renal failure is generally higher in patients undergoing combined CABG and valve replacement compared with patients who only need one procedure.

Very recently, aprotinin was identified as a riskfifactor for ARF. An observational study on 4374patients undergoing revascularization showedthat the use of aprotinin was associated with a doubling of the risk of renal failure requiringdialysis when compared with aminocaprionicacid, tranexamic acid, or no antifibrinolytic.fi

Methods of Prevention of ARF

Several studies have focussed on prevention of ARF after cardiac surgery. To date, no magic bullethas been identifi ed. Prophylactic therapy withfidopamine, mannitol, theophylline, and diureticshas not been effective. There is some evidence thatdiuretic use after cardiac surgery increases therisk of ARF. Data regarding the benefits of off-fipump surgery in patients at high risk of renal

TABLE 6.1. Risk factors for ARF developing in surgical patients

Preoperative factors State of hydration/adequate resuscitation preoperatively Contrast media Preexisting sepsis Patient comorbidity • Advanced age • Diabetes mellitus • Hypertension • Cardiovascular disease • Preexisting renal impairment

Operative factors Related to anesthesia (see text) Related to surgery • Emergency surgery • Duration of cardiopulmonary bypass • Clamping of renal arteries • Suprarenal aortic cross clamping

Postoperative factors • Sepsis • Bleeding • Nephrotoxic drugs • Contrast media • Cardiovascular complications associated with a fall in cardiac

t t ( di l i f ti l b loutput (e.g., myocardial infarction, pulmonary embolus,pericardial effusion)

• Development of intra-abdominal compartment syndrome

TABLE 6.2. Risk factors after cardiac surgery

Patient specific Surgery related

• Advanced age • Emergency surgery• Diabetes mellitus • Reoperation• Hypertension • Prolonged duration of• Preexisting renal impairment cardiopulmonary bypass• Impaired left ventricular function • Reexploration for bleeding • Pericardial tamponade • Deep sternal or systemic infection

6. Acute Renal Failure in the Surgical Patient 35

failure is still confl icting. Although low cardiac floutput has been shown to be a strong risk factorfor ARF after cardiac surgery, there is no evidencethat increasing cardiac output from adequate tosupranormal has beneficial renal effects.fi

Vascular Surgery

Patients undergoing vascular surgery generally represent a high-risk group for renal failure. Theexact incidence of renal injury in this context isunknown but, again, depends on the definition of firenal injury, patient characteristics, and type of surgery. Open surgical repair of abdominal aneu-rysms is associated with a high risk of renal failure,especially with suprarenal aortic cross clamping,massive bleeding, or cholesterol embolization. The emergence of endovascular repair has led to a reduced incidence of ARF, however, the risk isnot completely abolished by vascular stents. Inrare instances, endovascular stents have beenfound to migrate, resulting in occlusion of arterial orifi ces, including renal arteries.fi

Urological Surgery

Obstruction is a common cause of ARF in patientswith urological problems. Although this diagnosisis usually made before surgery, it may occurpostoperatively (e.g., after renal transplantationor ureteric surgery). In general, the majority of ARF posturological surgery is because of acutetubular necrosis precipitated by hypotension,

volume depletion, bleeding complications, and/or urosepsis.

Diagnosis (Table 6.3)

At present, there are no universally accepted cri-teria for the definition of ARF. Most arbitrary defifi -finitions are based on a rise in serum creatinine or fall in calculated creatinine clearance. It is impor-tant to remember that the glomerular filtration firate has to fall to less than 50% before serumcreatinine rises, which means that any rise in serum creatinine always implies significant renal fiinjury.

Treatment

General Measures

• Restoration of renal perfusion pressure• Correction of volume depletion• Avoidance of hypotension (including relative

hypotension). Be guided by the patient’s preex-isting blood pressure

• Fluid resuscitation, as appropriate, and use of vasopressor agents if perfusing pressure stillinadequate

• Optimal treatment of sepsis/septic shock• Avoidance of nephrotoxic medication if possi-

ble. Close monitoring of drug levels when ami-noglycosides are necessary

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 traumaFull blood count Eosinophilia is seen in 80% of patients with drug-induced interstitial nephritis

Urinalysis Dipstick Significant proteinuria/hematuria/casts suggest intrinsic renal pathologyBiochemistry Urinary sodium and osmolality to help differentiate prerenal failure from ATNCulture To exclude urinary tract infection

Diagnostic imaging Ultrasound scan To exclude obstruction and to determine renal sizeImaging 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 syndromepressure

aMAG3, mercaptoacetyltriglycine; DTPA, diethylene triamine pentaacetic acid.

36 M. Ostermann

• Renal replacement therapy in case of severemetabolic acidosis, unresponsive fluid overload,flresistant hyperkalemia, or pericarditis

• Early involvement of nephrologists if an intrinsic cause of renal failure is a possibility or if thepatient is likely to require ongoing renal support

Specific Measures

• Removal of septic focus if possible (e.g. drainageof intra-abdominal abscess)

• Relief of obstruction• Management of abdominal compartment syn-

drome (including consideration of abdominaldecompression)

• Revascularization of kidneys, if appropriate

Treatments that Have Not Been Shown to

Alter the Course of ARF

• Diuretics (unless patient is fluid overloaded)fl• Low-dose dopamine

Vasopressor Agents and the Kidney

Vasopressor agents are often needed to manage septic shock, and noradrenaline and dopamine aregood fi rst-line drugs. Although there are few directficomparison studies, patients with septic shock tend to respond better to noradrenaline. Concernregarding the potential for noradrenaline to impairrenal and mesenteric perfusion has been reducedby studies showing that reversal of hypotension with noradrenaline outweighed this effect and increased renal and mesenteric perfusion.

Prevention

General Vigilance

Meticulous attention to fl uid balance, blood pres-flsure, prescribed drugs, and treatment of sepsis arethe most important preventive measures. There isinsuffi cient evidence to recommend specififi c phys-fiiological targets (MAP, cardiac output, fi lling pres-fisures) that will ensure adequate renal perfusion. Instead, therapy needs to be individualized basedon the baseline physiological condition of theindividual patient.

Early Resuscitation

Early recognition of patients at risk and timely resuscitation has been shown to result in signifi-ficant reduction of the development of ARF.

Prevention of Contrast-Induced

Nephropathy

The administration of fl uids has been shown to beflthe most important factor in prevention of con-trast-induced renal injury. Although the optimal fluid regimen is uncertain, available data supportfla regimen of 0.9% saline at 1 mL/kg/h intrave-nously from up to 12 hours before administrationof contrast medium and for up to 12 hours after. Studies on the prophylactic role of N-acetylcyste-NNine have had confl icting results. Meta-analysesflhave concluded that prophylactic N-acetylcysteineNNwas harmless and may prevent an acute rise of serum creatinine after intravenous contrast, butsurvival and need for dialysis were not affected.

Tight Glucose Control

A single-center randomized controlled trial on intensive insulin therapy in postoperativeventilated patients showed a 41% decrease in the incidence of ARF requiring dialysis in the group of patients whose blood sugars were tightly con-trolled between 4.4 and 6.1 mmol/L compared with patients whose blood sugar was allowed to rise to 12 mmol/L before insulin was initiated.Further studies are necessary to confirm these firesults.

Prophylactic Dopamine or Diuretics

A recent meta-analysis of 61 trials showed thatlow-dose dopamine (<5 μg/kg/min) often increasedurine output but had no effect on renal functionor prevention of ARF.

Natriuretic Peptides

Urodilatin (renal natriuretic peptide) or anaritide(synthetic form of atrial natriuretic peptide)have failed to show any protective effect on thekidney.

6. Acute Renal Failure in the Surgical Patient 37

Future Advances

At present, no agents have conclusively demon-strated a protective effect against ARF or altera-tion of the course of ARF. Strategies aimed at modulating renal function and renal recovery have focused on several mechanisms:

1. Reduction of renal metabolism and energy consumption of the kidneys (i.e., induction of hypothermia, use of insulin-like factor I).

2. Modulation of the inflammatory system (i.e., flup-regulation of the acute stress response,manipulation of complement system, blockade of adhesion molecules).

3. Ischemic preconditioning.

These strategies are clearly important areas of research but not ready for clinical application.

References

1. Bahar I, Akgul A, Ozatik MA, Vural KM, DemirbagAE, Boran M, Tasdemir O. Acute renal failure follow-

ing open heart surgery: risk factors and prognosis. Perfusion 2005; 20: 317–322.

2. Barrett BJ, Parfrey PS. Preventing nephropathy induced by contrast medium. N Engl J Med 2006; 354: d379–386.

3. Chertow GM, Levy EM, Hammermeister KE, GroverF, Daley J. Independent association between acuterenal failure and mortality following cardiac surgery.Am J Med 1998; 104: 343–348.d

4. Friedrich JO, Adhikari N, Herridge MS, Beyene J. Meta-analysis: low-dose dopamine increases urine output but does not prevent renal dysfunction ordeath. Ann Intern Med 2005; 142: 510–524.d

5. Lassnigg A, Donner E, Grubhofer G, Presterl E, Druml W, Hiesmayr M. Lack of renoprotective effects of dopamine and furosemide during cardiac surgery. J Am Soc Nephrol 2000; 11: 97–104.l

6. Mangano DT, Tudor IC, Dietzel C. Multicenter Study of Perioperative Ischemia Research Group; Ischemia Research and Education Foundation. The risk associ-ated with aprotinin in cardiac surgery. N Engl J Med2006; 354: 353–365.

7. Van den Berghe G, Wouters PJ, Bouillon R, et al.Intensive insulin therapy in critically ill patients. N Engl J Med 2001; 345: 1359–1367.d

38

7Rhabdomyolysis andCompartment Syndrome

Laurie Tomlinson and Stephen Holt

necrosis. There may be severe pain, with loss of muscle function and loss of distal pulses. The diagnosis may be occult, especially in the uncon-scious patient. Direct pressure measurementscan be made by passing a needle connected to apressure manometer (e.g., central venous pressure[CVP] transducer) into the affected muscle compartment.

A fasciotomy should be considered if the pressureexceeds 40 mmHg or greater than 30 mmHg abovediastolic pressure.

Diagnosis

Serum changes consequent on rhabdomyolysis:

Creatine Kinase

Very high levels of the muscle enzyme CK arepathognomic of this condition. The degree of eleva-tion is proportional to the degree of muscle injury.Other muscle enzymes, such as aspartate transami-nase (AST) and lactate dehydrogenase (LDH) arealso elevated. CK levels should decline by approxi-mately 40% per day, a plateau or an increase should prompt a search for ongoing muscle damage.

Hyperkalemia

Hyperkalemia caused by efflux of potassium fromfldamaged cells is an early and life-threateningconsequence of rhabdomyolysis. It should be aggressively treated.

Rhabdomyolysis occurs when an insult causing myocyte necrosis results in release of intra-cellular contents into the circulation. Renal dys-function is caused by a combination of renalvasoconstriction, tubular damage, and tubularobstruction.

Causes

Rhabdomyolysis accounts for approximately 7%of all causes of acute renal failure (ARF) during peacetime. This fi gure is much higher after naturalfidisasters and in wartime. For example, after the1998 Turkish earthquake, 12% of the hospitalized population developed significant renal dysfunc-fition and 477 patients required dialysis.

Direct crush or compression injury and drugsare the most important causes in clinical practice, see Table 7.1. There are often predisposing factors, for example, alcohol, which can presensitize myo-cytes so they may be damaged by a more trivial insult. A clinical scoring system exists (not widely used) based on levels of phosphate, potassium,albumin, creatine kinase (CK), and presence of dehydration and sepsis.

Compartment Syndrome

After an appropriate precipitant, inflammationflwithin a muscular compartment causes a viciouscycle of increasing pressure. This leads to furtherinfl ammation and damage, eventually compro-flmising blood supply, leading to further muscle

7. Rhabdomyolysis and Compartment Syndrome 39

Acidosis

Metabolic acidosis may be caused by increasedlactate production and lactate release by damagedmuscle. Myoglobin (Mb) is considerably moretoxic in an acid milieu.

Early Hypocalcemia and Late Hypercalcemia

Serum calcium levels often fall dramatically,with total calcium less than 1.7 mmol/L in the early stages. This is caused by sequestration into damaged muscle and reperfusion-induced cellu-lar calcium uptake. In contrast, intracellularcalcium concentrations in damaged muscle may rise by up to 10-fold. In the recovery phase of rhabdomyolysis, serum calcium levels normalize and may even “overshoot,” secondary to calciumrelease by recovering myocytes and a transient

rise in parathyroid hormone (a reflex to the initial flhypocalcemia).

Symptoms of hypocalcemia are rare and treat-ment with intravenous calcium should be avoidedunless tetany or cardiac dysfunction is present.Pharmacologically administered calcium is takenup avidly by the damaged muscle. It may be depos-ited as inorganic complexes causing “heterotopiccalcification,”fi which delays recovery and can leadto long-term muscle dysfunction.

Hyperphosphatemia

Serum phosphate often exceeds 3 mmol/L.

Urinary Abnormalities

Urine dipsticks are usually positive for blood because they detect the heme moiety present in both hemoglobin (Hb) and Mb. On microscopy,few red cells are seen (unless there is coexistent trauma), instead, the characteristic “brown sugarcasts” of Mb are seen (Figure 7.1). When there is uncertainty, Mb can be specifically assayed in the fiurine, although it has a short half-life. An assay exists for myosin heavy chain, which remains pos-itive for up to 12 days after the initial insult.

Pathophysiology of ARF

Suggested causes:

1. A reduction in renal blood flowfl . There is a reduction in the effective blood volume causedby fluid shifts from the intravascular to extracel-fllular fluid compartments. Mb binds to nitric floxide (NO), preventing intrarenal vasodilation

TABLE 7.1. Causes of rhabdomyolysis

Physical Trauma, hyperthermia, hypothermia, exercise, electric shock, seizures, delirium tremensToxins and drugs Alcohol, statins, amphetamines, aspirin (overdose), barium, barbiturates, caffeine, carbon monoxide, ecstasy, ethylene

glycol, LSD, malignant hyperpyrexia, neuroleptic malignant syndrome, opiates, toluene, snake/insect bites, vasopressinMuscle ischemia Vascular ischemia, coma, sickle cell disease, surgery, vasoconstrictors, CO2 angiographyInfection Virtually any viral or bacterial infection (e.g., influenza, HIV, Epstein-Barr virus, Legionella, tetanus, malaria, Bacillus cereus)Metabolic Hypernatremia/hyponatremia, hypokalemia, hypophosphatemia, diabetic ketoacidosis, diabetic hyperosmolar coma,

water intoxication, myxedemaInherited Deficiency of carnitine palmityl transferase II, phosphofructokinase, myophosphorylase (McArdles), myoadenylate deaminase,

cytochrome oxidase, succinic dehydrogenase, coenzyme Q10 deficiency, King-Denborough Syndrome, Wilson’s diseaseImmune Polymyositis, dermatomyositis

FIGURE 7.1. “Brown sugar” Mb casts under light microscope are similar to granular casts but have a brown/rusty tinge. Additionalred cells, tubular cells, and other debris are also present withinthe urine.

40 L. Tomlinson and S. Holt

(especially in the medulla) and, in addition, vaso-dilators (e.g., endothelin) are increased.

2. Direct heme protein tubulotoxicity occurs,yprobably by free radical-mediated mechanisms.

3. Tubular cast formation. Urinary Mb andTamm-Horsfall protein (THP) complex and precip-itate as tubular casts. These casts are less soluble inacidic conditions. Although there is some evidencethat these complexes cause tubular obstruction,micropuncture studies have shown relatively low intratubular pressures, suggesting that these castsare as a result of reduced tubular flow and reducedflwashout rather than by obstruction per se.

Treatment

Intravascular Volume Expansion

Intravascular volume expansion at the first possi-fible opportunity after the insult is the single most effective therapeutic maneuver in rhabdomyoly-sis. This not only prevents or limits renal damage but may play a role in preventing acidosis andlimiting ongoing damage caused by hypoperfu-sion. Very large volumes of fluid can be lost intoflareas of muscle injury. In trauma situations inwhich there is a risk of crush injury, fluid resusci-fltation should be commenced before the victim is extricated.

Alkalinization

There is much compelling evidence to suggest that urinary alkalinization greatly reduces the nephro-toxicity of Mb. However, there are no large humantrials that confirm this consistent fifi nding fromfianimal research. The potential benefits in this fisetting include reduced renal vasoconstriction, a dramatic reduction in the ability of Mb to cause oxidant damage, and increased solubility of Mb-THP complex.

Alkalinization can lower ionized calcium still further and, if administered, it is wise to periodi-cally check the ionized calcium.

A suggested fluid replacement regime wouldflbe:

• Isotonic bicarbonate (1.26% sodium bicarbonate)until urine pH is greater than 7 if the patient is intravascularly volume deplete (which is usual)

• Bicarbonate solutions that are more concentrated can be administered in small aliquots, e.g., 50 mLof 8.4% NaHC03 via central access in patientswho are intravascularly full—remembering that this is 1 mmol of sodium per milliliter of fluid fland may cause sodium/fluid overloadfl

Mannitol

Mannitol promotes an osmotic diuresis and may reduce pressure in a swollen muscle compartment,but it also causes osmotically induced tubulardamage with vacuolation. There is no good evi-dence that it is more effective than saline alone and it has little scientifi c rationale to recommend fiits routine use.

Dialysis/Hemofiltration

The circulating concentration of Mb can be reduced by hemofi ltration, plasma exchange, andfihemodialysis, with dialysis being somewhat less successful. It has not been shown that any physical therapy materially reduces renal Mb burden or shortens the duration of renal replacementtherapy. Physical therapies may have a role, if commenced early, or if Mb release can be antici-pated, such as during arterial surgery.

Prognosis

There is an unquantifiable early mortality, mainly ficaused by hyperkalemia or the insult. Mortality after diagnosis is up to 20%, usually caused by other associated conditions, e.g., sepsis. Survivors of the Japanese earthquake in 1995 arriving inhospital within 6 hours had an approximately 20%chance of developing ARF, whereas all of thosearriving after 40 hours developed renal failure. If the patient recovers from the initial insult, therenal dysfunction almost always resolves, but cantake up to 3 months.

Summary

• Rhabdomyolysis is a common cause of ARF, with important early biochemical changes that may be fatal if treatment is not instituted quickly

7. Rhabdomyolysis and Compartment Syndrome 41

• Compartment syndromes can be occult and areimportant to detect and monitor to protect against further renal injury

• Early treatment with volume replacement (±alkalinization) will reduce the risk of renalfailure and the need for renal replacement therapy

Suggested Reading

Holt SG, Moore KP. Pathogenesis and treatment of renal dysfunction in rhabdomyolysis. Intensive Care Med. 2001 May; 27(5): 803–811.

Zager R. Rhabdomyolysis and myohemoglobinuric acute renal failure. Kidney Int. 1996 Feb; 49(2): 314–326.

42

8Multisystem Causes of AcuteRenal Failure

Tim Leach

Normally, the immune system surveys cells andtissues within the body, recognizing and ignoringcells expressing “self” antigens while attackingcells without these protective epitopes. In autoim-mune conditions, the immune system does not protect cells with “self” expression. Cells are attacked and either infl amed or killed, or circulat-fling self-antigens are bound with antibody-forming immune complexes. Immune complexes are very large molecules that are often unable to pass through capillaries because of their size. They can induce local infl ammation and activate the com-flplement cascade.

Presentation

Symptoms

Renal failure as a result of fulminant small vessel vasculitis will present acutely, and patients may be systemically unwell, requiring organ support. More often, however, there is an indolent presenta-tion with several months of nonspecific symptomsfiand signs (Table 8.2).

Signs

Fulminant systemic vasculitis presents with the nephritic syndrome:

• Azotemia (more often with oligoanuria)• Hypertension and edema from fluid overloadfl

(but can present with circulatory collapse causedby vasodilation and dehydration)

• Hematuria with red cell casts

This chapter covers some of the more specializedcauses of acute renal failure, which, although morelikely to present to the nephrologist, could beadmitted to the intensive care unit as a conse-quence of their illness or because of complicationsof their treatment.

Systemic Vasculitis

Vasculitis is the term given to inflammation of flblood vessels. Vasculitis is a rare condition, withan incidence of approximately 6 people per million(Western) population per year (1). Vessels can beclassified according to their size (Table 8.1) (2).fi

Renal failure can occur in any vasculitis, but thischapter focuses on those conditions that affect renal function directly through inflammationflwithin the glomeruli (small vessels), rather thanaffecting the kidneys indirectly through a reduc-tion of blood supply to the kidneys (large andmedium vessel diseases).

Etiology

Small vessel vasculitides separate into two broad groups: those in which immune complexes aredeposited within the renal glomeruli and thosewithout evidence of deposition histologically. The latter group are termed pauci-immune; lacking (literally “few”) immune complexes. This distinc-tion is useful for making a histological diagnosisand for estimating prognosis, but the underlying cause of these small vessel vasculitides is similar:autoimmunity.

8. Multisystem Causes of Acute Renal Failure 43

TABLE 8.1. Vasculitic conditions according to the vessel size affected

Vessel size Condition

Large Takayasu’s arteritisGiant cell arteritis

Medium Polyarteritis nodosaKawasaki disease

Small Pauci-immune:• Wegener’s granulomatosis (WG)• Microscopic polyangiitis (MPA)• Churg-Strauss disease (CSD) Immune-complex forming• Goodpasture syndrome (GS) Immune-complex depositing:• SLE• Henoch-Schönlein purpura (HSP)• Cryoglobulins• Rheumatoid arthritis

Source: Jeanette et al., 1994.

TABLE 8.2. Nonrenal symptoms suffered in systemic vasculitis

Relative frequency(some patients

System experienced moreaffected than one) Symptoms

Lower 63% Coughrespiratory Breathlessness (on

exertion/at rest/orthopnea) Hemoptysis

Upper 50% Sinusitisrespiratory Nasal congestion

Reduced hearing Epistaxis

Skeletal 42% Arthralgia Arthritis Synovitis

Muscular 33% Myalgia Myositis Muscle weakness

Dermatological 22% Rash (purpura/echinoses/ malar flush)

Neurological 14% Headache Lethargy Loss of concentration Coma

Source: Hedger et al., 2000.

FIGURE 8.1. Plain chest x-ray of fulminant pulmonary hemorrhageof 17-year-old male patient with systemic vasculitis caused by Henoch-Schönlein purpura. The patchy diffuse alveolar shadowingis clearly seen.

pneumonitis (Figure 8.1), or superadded pneumonia

Laboratory Investigations (Table 8.3)

Imaging

Chest x-ray may be clear or show pulmonary edema or pulmonary hemorrhage (Figure 8.1).Renal ultrasound will usually show normal-sizedkidneys with no obstruction.

Renal Biopsy

The ultimate investigation for renal vasculitis isthe biopsy. Patients need to be hemodynamically stable, normotensive, have a normal platelet countand coagulation screen and be able to lie flat (see flFigure 8.2).

Treatment

Treatments of patients with vasculitis fall into three main areas:

Resuscitation

Patients should be stabilized in terms of airway, breathing, and circulation, as in any serious illness.

Other features include:

• Fever• Raised purpuric rash• Respiratory failure secondary to pulmonary

edema, pulmonary hemorrhage from vasculitic

44 T. Leach

TABLE 8.3. Investigations for systemic vasculitis

Investigation Rationale Expected result

Urea and electrolytes Likely renal failure Elevated urea and creatinine Potentially elevated potassiumFull blood count Anemia Anemia Elevated white cell count Normal/elevated platelet count (cf HUS)fAntineutrophil cytoplasmic antibody WG and MPA associated with positive ANCA Positive cANCA in WG (ANCA) Positive pANCA in MPA Otherwise negativeAntiglomerular basement antibody Goodpasture syndrome Positive in GS (aGBM) Otherwise negativeComplement Involved in inflammation Usually mildly or significantly reduced levels of C3 and C4ESR/CRP Inflammatory markers Usually significantly elevated

ESR, erythrocyte sedimentation rate; CRP, C-reactive protein.

FIGURE 8.2. Photomicrograph of renal biopsy from patient withWegener’s (cANCA-positive) renal vasculitis. Hematoxylin andeosin stain with silver counterstain, showing features consistent with the disease. The glomerulus (black arrow) shows segmentalwfibrinoid change (red arrows) and crescent formation (yellow(arrow). Some tubules contain red cells (w white arrows). (Photoreproduced with the kind permission of Dr. Nicholas Marley.)

Maintenance

Maintenance of adequate oxygenation, circula-tion, and fl uid balance is vital to their recovery. flHemofi ltration, hemodiafifi ltration, or hemodialy-fisis depending on the patient’s hemodynamic stability should be used to prevent and treat com-plications that do not respond to standard medical therapies. Infection, bleeding, and malnutritionshould be sought and remedied.

Disease-Specific Treatment

The disease is one of an abnormal immuneresponse inducing renal and other organ inflam-flmation and treatment falls into three groups.

1. Reduction of inflammation with corticoste-flroids (e.g., intravenous methylprednisolone fol-lowed by oral prednisolone).

2. Reduction of antibody formation withimmunosuppressive agents (e.g., cyclophospha-mide administered intravenously every 2 to 4weeks or orally everyday).

3. Removal of the already-formed antibodieswith therapeutic plasma exchange if the patient haspulmonary hemorrhage (potentially life threaten-ing) or is requiring renal dialysis (severe disease).

Outcome

Renal vasculitis is a serious illness with significant fimorbidity 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% chance of renal survival. If dialysis dependent atpresentation, 30% of patients may regain renalfunction with cyclophosphamide and corticoste-roids alone, but 90% of patients may develop renalrecovery with the addition of therapeutic plasmaexchange (3, 4).

8. Multisystem Causes of Acute Renal Failure 45

Infection is the most significant side effect of fitreatment. Bacterial sepsis, viral infections such asherpes zoster and cytomegalovirus, fungal sepsis,and Pneumocystis carinii pneumonitis (PCP) occur in 40% of treated patients (5). Prophylaxisagainst fungi and PCP is recommended.

Systemic Lupus Erythematosus

Systemic lupus erythematosus (SLE) is a multisys-tem autoimmune condition of primarily young(20–30 years old) women (Figure 8.3). The Ameri-can College of Rheumatology defines the diagno-fisis of lupus for a patient presenting with four ormore of the following features at the same time, orindividually during a period of time:

1. Malar rash: redness or rash that may appear in a butterfl y confifl guration across the nose fiand cheeks. It can appear on one or both sides of the face and is usually flat.fl

2. Discoid rash: thick raised patches that can occur on any part of the body and may resultin scarring.

3. Sun sensitivity: a reaction to sunlight that is more severe than just sunburn.

4. Oral ulcers: frequent development of mouthor nose ulcers.

5. Arthritis: pain, tenderness, or swelling in twoor more joints.

6. Pleurisy or pericarditis. 7. Nephritis: proteinuria or cellular casts in the

urine. 8. Nervous system disorder: seizures or psy-

chotic behavior that cannot be attributed todrugs or metabolic dysfunction.

9. Blood system disorder: hemolytic anemia, leu-kopenia, lymphopenia, or thrombocytopenia.

10. Immunologic disorder: the presence of the lupus erythematosus (LE) cell, a false-positivereaction to the tests for syphilis, or the pre-sence of autoantibodies.

11. Positive antinuclear antibodies (ANA): anti-bodies against the nucleus of cells, particu-larly against double-stranded DNA.

Lupus should also be suspected in young women presenting with purpura, easy bruising,diffuse lymphadenopathy, hepatosplenomegaly,peripheral neuropathy, endocarditis, myocarditis,

interstitial pneumonitis, or aseptic meningitis. Apositive Coombs test, low complement levels, andimmune deposits at the dermal-epidermal junc-tion on skin biopsy are also suggestive of lupus.

Patients with SLE can present in extremis withany system involvement. The presence of a multi-system disease such as SLE should always beentertained in such patients.

Laboratory Testing

Tests that provide potentially diagnostically usefulinformation when SLE is suspected include:

• Complete blood count and differential• Serum creatinine• Serum albumin• Serologic test for syphilis (falsely positive

because of cross-reactivity)• Urinalysis• 24-hour urine collection for calculation of creati-

nine clearance and quantifi cation of proteinuriafi• Autoantibody testing: ANA, antibodies to phos-

pholipids, antibodies to double-stranded DNA,and antibodies to Smith (Sm)

FIGURE 8.3. Magnetic resonance imaging scan of brain of 21-year-old woman with cerebral lupus. T2-weighted sagittal sectionshows increased white and grey matter signal, particularly in thefrontal and parietal regions, in keeping with cerebral vasculitis.

46 T. Leach

• Complement levels (total hemolytic comple-ment [CH50], C3 and C4)

Imaging

This may be valuable but is not routinely obtainedunless indicated by the presence of symptoms,clinical fi ndings, or laboratory abnormalities.fi

Examples include:

• Plain radiographs of involved joints• Renal ultrasound to assess kidney size and rule

out an obstructive post-renal cause when thereis evidence of acute renal failure

• Chest X ray• Echocardiography (for suspected pericardial

involvement or to seek a source of emboli)• CT scanning (for abdominal pain, suspected

pancreatitis, lymphadenopathy)• Magnetic resonance imaging (for seizure activ-

ity, focal neurologic defi cits or cognitive dys-fifunction/personality changes—see Figure 8.3)

• Contrast angiography may be valuable if vascu-litis affecting medium-sized arteries is suspected(mesenteric or limb-threatening ischemia).

Treatment

Treatment is broad and depends on the system involved. General principles are:

• Reduce inflammation with corticosteroidsfl• Prevent further autoantibody production with

immunosuppressant medications such as cyclo-phosphamide or mycophenolate mofetil

• Remove circulating autoantibodies with thera-peutic plasma exchange

Antiphospholipid antibodies and the presenceof the lupus anticoagulant increase coagulation and lead to arterial and venous thromboses. Use of antiplatelet drugs, thrombolytics, and antico-agulation in these patients often prevents further problems.

Fertility and pregnancy in lupus patients areoften problematic. Pregnancy alters the immunestate, often leading to flares of lupus postpartum.flCoagulation abnormalities often lead to difficul-fities with conception and spontaneous abortion. Primary infertility or recurrent miscarriages arerelatively common presenting features of SLE.

Thrombotic Microangiopathy

Thrombotic microangiopathy (TMA) covers theacute syndrome of microangiopathic hemolyticanemia, thrombocytopenia, and variable signs of organ injury caused by platelet thromboses inthe microcirculation. Two clinically distinct but pathologically identical syndromes are described:hemolytic uremic syndrome (HUS) and throm-botic thrombocytopenic purpura (TTP). HUSusually affects children with renal failure butminimal neurological involvement. TTP is a disease of adults, with predominantly neurologi-cal involvement and variable other organ disease.The two syndromes often overlap and, thus, aretermed HUS/TTP.

Etiology

The mechanisms behind the development of TMA are poorly understood but seem to involve endothelial injury, which triggers events leading to microvascular thrombosis, microangiopathichemolytic anemia, and platelet consumption.Most causative factors lead to toxicity of theendothelial cells: autoantibodies, exotoxins andendotoxins, immune complexes, and certaindrugs.

The pathology of TMA consists of capillary andarteriolar wall widening, with swelling and detach-ment of the endothelium, and occlusion or severe restriction of the lumen and platelet micro-thrombi. In HUS, this occurs mainly in the kidney,in TTP, mainly in the brain.

Presentation

HUS/TTP rarely causes specifi c symptoms. In thefisituations shown in Table 8.4, vomiting, pallor,purpura, anuria, and/or neurological signs should alert the clinician to the possibility of TMA.

Differential Diagnosis

• Systemic vasculitis will usually present with arthralgia/arthritis, and the platelet count will be normal and will rarely have central neuro-logical involvement.

• Disseminated intravascular coagulation (DIC)is usually associated with shock or obstetric

8. Multisystem Causes of Acute Renal Failure 47

complications and will have consumption of allof the clotting factors.

• Malignant hypertension will have classical retinal changes, signifi cant high blood pressure fi(usually >210/130), and usually a history of hypertension.

Laboratory Investigations

• Low hemoglobin (<7 g/dL)• Thrombocytopenia (<80 × 109 cells/L)• The blood film will show red cell fragments fi

(schistocytes) and increased reticulocyte counts

• Elevated lactate dehydrogenase and indirectbilirubin (caused by hemolysis)

• Haptoglobin levels will usually be low because of consumption

• Coombs test is negative• Moderate proteinuria (1–2 g/d) with few red

cells and casts (ARF is secondary to occlusion of capillaries rather than inflammation)fl

Treatment and Outcome

The epidemic or sporadic diarrhea-associated HUS/TTP of young children is usually self-limiting and mild. Renal failure requires dialysis in approxi-mately 50% of patients, but otherwise supportive treatment is all that is required. Ninety percent of patients should recover completely. Up to 5% of

patients can die in the acute phase. Cerebrovascularaccidents, seizures, and coma occur in 25% of patients, and residual impairment of renal excretory function is present in up to 40% of patients.

HUS/TTP in adults or children older than 14 years old usually requires treatment. In those withan apparent cause (pregnancy, malignancy, infec-tion, or drugs), removal of the cause is essential. Supportive treatment is required. Specific treat-fiment of the condition revolves around therapeuticplasma exchange or plasma infusion if TPE is not available. There is no evidence of benefi t of fiimmunosuppression with corticosteroids, immu-noglobulins, or vincristine, or of benefit from fiantithrombotic or antiplatelet agents.

Rescue therapies from severe refractory or relapsing disease include bilateral nephrectomy and/or splenectomy.

References

1. Hedger N, Stevens J, Drey N, Walker S, Roderick P. Incidence and outcome of pauci-immune rapidly progressive glomerulonephritis in Wessex, UK: a 10-year retrospective study. Nephrol Dial Transplant 2000; 15(10): 1539–1539.

2. Jennette JC, Falk RJ, Andrassy K. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 1994; 37: 187–192.

TABLE 8.4. Conditions and situations associated with the development of HUS/TTP

Acquired Shigatoxin (Escherichia coli 0157)iPregnancyPneumococcal infectionSystemic disease HIV infection SLE Scleroderma MalignancyDrug associated Mitomycin C, tamoxifen, bleomycin, cisplatin, clopidogrel, quinine, interferon, OKT3, cyclosporine, tacrolimus (among others)Organ transplant De novo (usually drugs) Recurrent posttransplant HUS

Genetic and familial formsIdiopathic and atypical forms

48 T. Leach

3. Andrassy K, Kuster S, Waldherr R, Ritz E. Rapidly pro-gressive glomerulonephritis: analysis of prevalence and clinical course. Nephron 1991; 59(2): 206–212.

4. Pusey CD, Rees AJ, Evans DJ, Peter DK, Lockwood CM. Plasma exchange in focal necrotizing glomeru-

lonephritis without anti-GBM antibodies. Kidney Int1991; 40(4): 757–763.

5. Hoffman GS, Kerr GS, Leavitt RY, et al. Wegener gran-ulomatosis: an analysis of 158 patients. Ann InternMed 1992; 116: 488–498.d

49

9Therapeutic Plasma Exchange

Tim Leach

• Of molecular weight greater than 15,000 kDa so it cannot be removed in any other way, and/or

• Of suffi cient half-life that TPE is quicker than fiendogenous removal, and/or

• Acutely toxic and resistant to conventionaltherapy

Prescription (Table 9.1)

Calculation of plasma volume:

Estimated plasma volume (L) = 0.07 × Weight (kg) × (1 − hematocrit)

1. Before each treatment, measure serum potas-sium calcium and clotting screen.

2. Calculate the estimated plasma volume: Volumeof exchange is measured in Patient Plasma Volumes (∼3 L).

3. Prescribe the plasma exchange:a. Number and spacing of treatmentsb. Volume and type of fluidfl

4. Electrolyte supplementation as needed (potas-sium and calcium).

5. If coagulopathic consider, fresh-frozen plasma(FFP) as the fi nal exchange volume.fi

After Procedure

• Repeat electrolytes and clotting after 2 hours and increase supplementation as needed

• FFP can be given as the fi nal exchange volume fiif coagulopathic

Therapeutic plasma exchange (TPE) is an extra-corporeal blood purifi cation technique designed fifor the removal of large molecular weight sub-stances from plasma. Large molecular weightsubstances equilibrate slowly between the vascu-lar space and the interstitium. Calculations of the rate of their removal by TPE follows first-fiorder kinetics, i.e., approximately 60% is removedby a single plasma volume exchange, and 75% by an exchange equal to 1.4 times the plasma volume.

Blood is pumped through a highly permeablefilter, replacing the fifi ltrate with flfi uid as indicatedfl(Table 9.1). Venous access on the intensive careunit (ICU) is via a double-lumen dialysis catheter, but can be via two wide-gauge peripheral venouscannulae. If chronic therapy is indicated, an arte-riovenous fistula is used. The patient and fifi lter are fianticoagulated during the procedure.

Indications (Table 9.2)

The basic premise of TPE is that removal of largemolecular weight substances from the circulationwill reduce further damage and may permit rever-sal of the pathological process.

Other benefits include unloading the reticulo-fiendothelial system to permit further endogenousremoval of circulating toxins, stimulation of lym-phocyte clones, and allowing re-infusion of largevolumes of plasma without the risk of volume overload.

For TPE to be appropriate, the substance to be removed should be:

50 T. Leach

TABLE 9.1. Example of TPE prescriptiona

Indication Plasma volumes Replacement fluid Number of exchanges Exchange frequency

Rapidly progressive 1 to 1.5 Albumin, with FFP if 7 to 10 Daily or alternate days glomerulonephritis pulmonary hemorrhageAcute renal failure caused by myeloma kidneyHyperviscosity 1 Albumin/saline mixtureb Until symptoms subside or DailySyndrome plasma viscosity

normalAnti-glomerular basement 1.5 Albumin, with FFP if 7 to 10 Daily

membrane (GBM) disease pulmonary hemorrhageHemolytic uremic 1 to 2 All FFP Until platelets normal/no Once or twice daily

syndrome/thrombotic red cell fragmentsthrombocytopenic purpura (usually 7 to 16)

Guillain-Barré Syndrome 2 Albumin 4 Alternate days

aFrom: Wessex Renal and Transplant Unit, United Kingdom.bNo more than one part saline to two parts albumin (i.e., ≤1 L saline for 2 L albumin).

TABLE 9.2. American Association of Blood Banks indications for TPE

1. Standard and acceptable 2. Sufficient evidence to suggest • Chronic inflammatory efficacy/acceptable adjunct demyelinating • Cold agglutinin disease polyneuropathy • Protein-bound toxins • Cryoglobulinemia (drug overdose/ • Anti-GBM disease poisonings) • Guillain-Barré syndrome • Hemolytic uremic • Familial hypercholesterolemia syndrome • Myasthenia gravis • Rapidly progressive • Posttransfusion purpura glomerulonephritis • Thrombotic • Systemic vasculitis thrombocytopenic purpura • Acute renal failure caused

by myeloma kidney by myeloma kidney

33. InconclusiveInconclusive evidenceevidence oror 44.rr NoNo efficacyefficacy inin trialstrialsuncertain benefit-to-risk ratio • AIDS

• ABO-incompatible organ or • Amyotrophic lateral marrow transplantation sclerosis • Coagulation factor inhibitors • Dermato/polymyositis • Idiopathic thrombocytopenic • Psoriasis purpura • Renal transplant rejection • Multiple sclerosis • Rheumatoid arthritis • Progressive systemic sclerosis • Schizophrenia • Thyroid storm • Warm autoimmune hemolytic anemia

Complications

• Hypotension: vasovagal, hypovolemia• Fluid overload• Hypocalcemia causing tetany• Hypokalemia• Coagulopathy caused by removal of clotting

factors or thrombocytopenia with heparinanticoagulation

• Protein-bound drug removal (administer drugsafter TPE)r

• ACE inhibitors may cause flushing and hypoten-flsion (stop 24 h before treatment)

51

10Renal Replacement Therapy

John H. Reeves

maintain a concentration gradient favoring theout-diffusion of unwanted solutes. Both dialysateand replacement solutions contain electrolytes inphysiological concentrations and some form of buffer base.

The hemofilterfi or dialysis membrane’s permea-rbility to water is determined by its surface area (typically 0.5 to 2.0 m2) and the number and size of its pores (typically 0.0055-μm diameter in a high-flux hemofifl lter). The pore size determines fithe size of molecules that are freely filtered along fiwith water. Current hemofilters freely fifi lter sub-fistances up to approximately 5000-D molecularweight and then in decreasing amounts up to a cutoff of approximately 20,000 Da. This minimizesloss of important larger molecules, such as albumin (molecular weight, 57,000 Da).

The ratio between the concentration of solutein fi ltrate and that in plasma water is called thefisieving coefficientfi , and this concept becomes important in calculating the clearance of interme-diate-sized molecules. The sieving coefficient for fia small unbound solutes is one, decreasing to zero as molecular size and or plasma protein binding increases.

Theoretical small solute clearance can be pre-dicted by knowledge of the blood flow through the flextracorporeal circuit (QB) and the rate of ultrafil-fitration (QF) and dialysate flow (Qfl D) (Figure 10.1).

During continuous RRT (CRRT), when blood flow is signififl cantly higher than dialysate or ultra-fifiltration rates, small solute clearance is deter-fimined by dialysate and ultrafiltrate flfi ow. Assuming flcomplete concentration equilibrium between dial-ysate and plasma water and a sieving coefficient fiequal to one:

Acute renal failure (ARF) occurs in 7% of patientsadmitted to the intensive care unit (ICU) (1). Pre-viously, mortality exceeded 91%, but with theintroduction of dialysis, this quickly fell to approx-imately 50% (2). Overall mortality for ARF hasremained approximately 50%, associated withincreasing comorbidity (3).

There is no specifi c therapy for ARF other than firemoval of the cause and ongoing supportive care awaiting spontaneous recovery. Renal replacementtherapy (RRT) is the cornerstone of that support-ive care.

The Basics

Extracorporeal RRT involves the passage of a patient’s blood outside his/her body through adialysis or hemofilter machine, where the removal fiof unwanted solutes and excess water and the replacement of lost bicarbonate (or buffer base) take place. The “purified” blood is returned to thefipatient.

Clearance can be defined as that volume of fiplasma completely cleared of a substance in a given time. During extracorporeal RRT, net solute clearance can be achieved by ultrafiltration across fia porous membrane down a pressure gradient (fi(( l-fitration), or by diffusion across a semipermeable membrane down a concentration gradient (dialy-sis), or both. During filtration, the fifi ltrate is dis-ficarded and a replacement solution is added to the blood to maintain fluid and electrolyte equilib-flrium. During dialysis, a continuous stream of dialysate is passed in the opposite direction to blood, on the nonblood side of the membrane, to

52 J.H. Reeves

Clearance during CRRT = QF + QD = QE

In contrast, during intermittent hemodialysist(IHD), dialysate fl ow is signififl cantly higher thanfiblood fl ow, and blood flfl ow becomes the limitingflfactor for solute clearance. Small solute clearance is proportional to the plasma flow through thefldialyzer:

Clearance during IHD ≅ QB × (1 − hematocrit)

This simplifi ed analysis only holds for small, fiunbound solutes, such as urea and creatinine.Increasing molecular weight decreases diffusiveclearance more signifi cantly than convective clear-fiance. Binding to macromolecules, such as albumin, decreases convective and diffusive clearance (4).

Classification of RRT

In 1977, when Peter Kramer first described con-fitinuous arteriovenous hemofi ltration (CAVH)fias a therapy for diuretic resistant fluid overloadfl(5), the only other types of RRT were IHD andperitoneal dialysis (PD). Since then, the classifica-fition of RRT has expanded (6). See the glossary in Table 10.1.

Duration or Timing of Therapy

Continuous: CRRT aims to provide support 24 h/d,but, in practice, is interrupted by factors such as patient transfer out of the ICU or circuit failure.

Intermittent: IHD requires approximately 3 hoursper treatment. Patients with end-stage renalfailure (ESRF) may be maintained in the com-munity with as few as three dialysis sessions per

week. In contrast, ICU patients with ARF may require daily treatment (7).

Hybrid: there is increasing interest in hybridapproaches for extracorporeal RRT in the ICU(8, 9). For example, during sustained low-efficiency dialysis (SLED), daily intermittentfitherapy is reduced in intensity and extended in duration up to 8 or 12 hours. The regular breaksaid staffi ng and free patients for investigationsfiand procedures. The reduced intensity reduces the cardiorespiratory destabilization associatedwith IHD.

Access

Arteriovenous: arteriovenous access involves can-nulation of a medium-sized artery and largevein—often the femoral artery and vein. Bloodfl ows passively from the artery through theflextracorporeal circuit (ECC) and back through the vein, driven by the mean arterial pressure.This method is reserved for situations in which resources are limited.

Venovenous: venovenous access involves cannula-tion of central veins, most commonly with a double-lumen cannula in the femoral, internal jugular, or subclavian vein. Blood flow is drivenflby a pump in the ECC. This increases reliability of blood flow and maximizes solute clearance,flbut introduces the need for more complex safety mechanisms to detect fault conditions, such asair embolism or circuit occlusion.

Mechanism of Solute Removal

Convection: when solute is cleared by ultrafiltra-fition through a porous membrane, we say that the clearance of the solute is convective—carried by the bulk fl ow of plasma water. During CRRT,flthe process is called hemofiltrationfi .

Diffusion: when solute is cleared by diffusion down a concentration gradient across a semi-permeable membrane, we say that the clearance of the solute is diffusive. The process is calledhemodialysis.

Combinations: diafiltrationfi is the term appliedwhen both convection and diffusion are operat-ing to remove solute.

Adsorption: adsorption is the binding of sub-stances to the membrane under molecular

DialysateQD

EffluentQE

Haemofiltration replacement ≅ QF

Blood flow QBMembrane

FIGURE 10.1. Predicting clearance during RRT.

10. Renal Replacement Therapy 53

attraction. Adsorption is used specifi cally for fitoxin removal using activated charcoal car-tridges (10) and it is being tested as a means of blood purifi cation in sepsis (11).fi

Intensity of Therapy

For example, slow continuous ultrafiltration fi(SCUF) is performed to simply remove excess extracellular fluid. High-volume hemofifl ltration fi(HVHF) is used to intentionally accelerate theclearance of target mediators, and high-fl ux dialy-flsis (HFD) (dialysis performed with highly perme-able membranes) is designed to accelerate theclearance of urea and larger molecules.

Indications for RRT

“Traditional” indications (19)

• Diuretic resistant fl uid overloadfl• Life-threatening hyperkalemia• Severe metabolic acidosis• Symptomatic uremia

Kramer’s original description of CAVH involvedthe treatment of diuretic resistant fluid overload fl(5). During the intervening years, there has beencontroversy surrounding the use of diuretics in ARF (20), but they may be helpful in the fluid flmanagement of ARF before the institution of RRT (21, 22).

There is little information regarding what spe-cific threshold plasma concentrations of potas-fisium, bicarbonate, urea, or creatinine should beused for the institution of RRT. In a retrospectivecomparison of early versus late CRRT in traumaassociated ARF, Gettings et al. (23) found thatpatients in whom CRRT was commenced at amean blood urea nitrogen (BUN) of 42.6 mg/dL(15.2 mmol/L) had a survival rate of 39% com-pared with 20% in patients in whom the mean BUN at commencement was 94.5 mg/dL(33.7 mmol/L).

“Nonrenal” Indications

• Drug and toxin removal• Sepsis and septic shock• Inborn errors of metabolism• Congestive cardiac failure• Cerebral edema

IHD can accelerate the elimination of small(<500 mw) unbound toxins with a low volume of distribution and minimal plasma protein binding(11), e.g., lithium, methanol, ethylene glycol, andsalicylates. Continuous hemofiltration has been fiused in lithium toxicity (24), with better hemody-namic stability (25) but lower solute clearancethan IHD.

The use of extracorporeal blood purification fitechniques in sepsis and septic shock is attractive but unproven. There were early observations of improved cardiovascular and respiratory function

TABLE 10.1. RRT: A glossarya

Acronym Full Name Notes Seed reference

SCUF Slow continuous ultrafiltration AV or VV Silverstein 1974 (12)CAVH Continuous arteriovenous hemofiltration Kramer 1977 (5)CAVHD Continuous arteriovenous hemodialysisCAVHDF Continuous arteriovenous hemodiafiltrationCVVH Continuous venovenous hemofiltrationCVVHD Continuous venovenous hemodialysisCVVHDF Continuous venovenous hemodiafiltrationCHFD Continuous high flux dialysis AV or VV Ronco 1996 (13)HVHF High volume hemofiltration AV or VV Cole 2001 (14)CPF Continuous plasma filtration AV or VV Reeves 1999 (15)CPFA Coupled plasma filtration adsorption AV or VV Ronco 2003 (16)SLED Sustained low efficiency dialysis Marshall 2004 (17)EDD Extended daily dialysis Kumar 2000 (18)PDIRRT Prolonged daily intermittent RRT Naka 2004 (9)

aAV, arteriovenous; VV, venovenous.

54 J.H. Reeves

in patients with severe sepsis after commence-ment of continuous hemofiltration (26). This,fitogether with the identification of inflfi ammatory flmediators in filtrate (27), led to efforts to increase fiinfl ammatory mediator removal during RRT.flHigh-volume conventional hemofiltration (14, 28), filarge-pore hemofiltration (29), and plasma fifi ltra-fition with (30) or without (15) coupled adsorption have been examined in small clinical trials. Thereis currently no Level I evidence for the use of extracorporeal blood purifi cation therapy infisepsis.

End-stage cardiac failure is characterized by progressive fluid retention, renal impairment,flneurohumoral stimulation, and diuretic resistance(31). It was shown that patients with advancedcardiac failure can tolerate substantial fluidflremoval during ultrafiltration (12), with salutary fieffects that persist beyond the time of fluidflremoval. Improved renal function, decreased heart failure scores, lowered B natriuretic peptide levels, decreased hospital length of stay, and fewer read-missions have all been observed in case-controlledstudies (32). Most recently, a small randomized controlled study showed that early ultrafiltrationfiresults in increased weight loss at 24 h compared with diuretics alone (33). Larger studies are war-ranted for this indication.

Cerebral edema can complicate IHD (34) and contribute to the clinical picture of disequilib-rium. In patients with hepatic encephalopathy, early studies compared the effects of IHD andCRRT (35). CRRT was associated with less decrease in mean arterial pressure, less increase in intracra-nial pressure, and less change in cerebral perfu-sion pressure.

Choosing the Dose and Mode of RRT

Dialysis dose is a concept familiar to nephrologistin the setting of ESRF: Kt/V.

K is clearance (the volume of solute, usually urea, cleared in a given time), t is duration of treat-ment, and V is volume of distribution of the solute.Kt is the volume of plasma water cleared of solute during the session, and Kt/V is Kt as a proportionof its volume of distribution.

For example, a Kt/V of 1.0 for urea means thata total volume of plasma water equal to the volume

of distribution of urea was cleared during the session. Using a single compartment exponential washout model, we can predict that the finalficoncentration of the solute is approximately 37% of the starting concentration when the Kt/Vis 1.0.

In chronic renal failure, a minimum Kt/V of 1.2should be delivered three times per week (36).There has been one randomized controlled trialassessing dose of dialysis in ARF (7). Schiffl com-flpared daily dialysis with alternate daily dialysis in160 patients with ARF. The 28-day mortality usingintention-to-treat analysis was 28% in the daily treated patients and 46% in the patients treated every other day. Multiple other outcomes were improved in the daily dialysis group. Although this trial was controversial (37), it suggests that dialy-sis every second day is insufficient in critically ill fipatients with ARF.

Quantifi cation of clearance during CRRT can befilikened to the calculation of creatinine clearance,which is described by the formula UV/P. U is the urine concentration of the solute, V is the volumecollected in a given time, and P is the plasma con-centration. During CRRT, if we know the volumeof effl uent produced in a given time and the con-flcentration of solute (e.g., urea) in both effluent fland plasma, we can calculate its clearance (38). The result is a value in milliliters per minute. To simplify the estimate further, let us assume that the sieving coefficient is 1.0, and that there is full ficoncentration equilibrium between plasma waterand dialysate. Then the effluent concentration willflequal the plasma concentration, and clearance issimply equal to the rate of production of effluent flfrom the hemofi lter.fi

Ronco, in 2000 (39), randomized 425 critically ill patients with ARF to three different doses(ultrafiltration rates) of CRRT: 20, 35, or 45 mL/h/fikg. Fifteen-day survivals were 41%, 57%, and 58%, respectively. This landmark study was one of the first to formally adjust dose of CRRT based onfipatient weight. Importantly, it suggests that there is a threshold minimum level of clearance requiredfor adequate CRRT of approximately 35 mL/h/kg.

Debate regarding the relative merits of CRRT and IHD has continued for nearly 30 years. Fromthe outset (5), the attraction of CRRT was its sim-plicity and cardiorespiratory stability comparedwith IHD. Now that CRRT is as complex as IHD

10. Renal Replacement Therapy 55

and potentially more expensive to perform for long periods, there is a paucity of good compara-tive studies that report outcomes such as mortal-ity or recovery of renal function. In 2001, Mehtarandomized 166 critically ill patients to CRRT orIHD (40). The observed mortality was higher inthe group receiving CRRT, but this group had a higher severity of illness. In 2002, Kellum pub-lished a meta-analysis of 13 trials, 3 randomized and 10 observational, comparing IHD and CRRT(41). Overall, there was no difference in mortality, but only six studies compared groups of equal severity. In these six studies, the mortality was lower in patients treated with CRRT. Since then, awell-designed prospective randomized controlledtrial has been published comparing CRRT with IHD in 80 critically ill patients with ARF. Therewas no difference in survival or renal recovery between the two groups. Although there was more hemodynamic disturbance during IHD, this didnot translate to a survival benefi t for CRRT (42).fi

Summary

• RRT reduces the mortality of ARF from more than 90% to approximately 50%.

• There is no proven difference in outcome between intermittent and CRRTs, as long as aminimum dose of Kt/V of 6 to 8 per week is achieved during hemodialysis or a clearancegreater than 35 mL/kg/h is achieved during con-tinuous hemofi ltration.fi

• New hybrid therapies (slow long extended daily dialysis [SLEDD], extended daily dialysis [EDD],and prolonged daily intermittent RRT [PDIRRT]) may avoid the destabilization associated withIHD, and mitigate the cost and inconvenience of continuous hemofiltration.fi

References

1. Brivet FG, Kleinknecht DJ, Loirat P, Landais PJ.Acute renal failure in intensive care units-causes,outcome, and prognostic factors of hospital mortal-ity; a prospective, multicenter study. French Study Group on Acute Renal Failure. Crit Care Med 1996; 24(2): 192–198.

2. Smith LH, Post RS, et al. Post traumatic renal insuf-fi ciency in military casualties. II. Management, usefi

of an artifi cial kidney, prognosis. fi Am J Med 1955; d18:187.

3. Mehta RL, Pascual MT, Soroko S, et al. Spectrum of acute renal failure in the intensive care unit: thePICARD experience. Kidney Int 2004; 66(4): t1613–1621.

4. Meyer TW, Walther JL, Pagtalunan ME, et al. Theclearance of protein-bound solutes by hemofiltra-fition and hemodiafi ltration.fi Kidney Int 2005; 68(2): t867–877.

5. Kramer P, Wigger W, Rieger J, et al. Arteriovenous haemofi ltration: A new and simple method forfitreatment of over-hydrated patients resistant todiuretics. Klin Wschr 1977; 55: 1121–1122.

6. Ronco C, Bellomo R. Continuous renal replacementtherapy: evolution in technology and current nomen-clature. Kidney Int Suppl 1998; 66: S160–164.

7. Schiffl H, Lang SM, Fischer R. Daily hemodialysisfland the outcome of acute renal failure. N Engl J Med2002; 346(5): 305–310.

8. Marshall MR, Golper TA, Shaver MJ, Chatoth DK. Hybrid renal replacement modalities for the criti-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 in ICU patients by ICU nurses and ICU physicians. Int J Artif Organs 2004; 27(5): 380–387.

10. Zimmerman JL. Poisonings and overdoses in theintensive care unit: general and specifi c manage-fiment issues. Crit Care Med 2003; 31(12): 2794–2801.

11. Nalesso F. Plasma fi ltration adsorption dialysisfi(PFAD): a new technology for blood purification.fiInt J Artif Organs 2005; 28(7): 731–738.

12. Silverstein ME, Ford C, Lysaght MJ, Henderson LW.Treatment of severe fluid overload by ultrafifl ltration.fiNew Eng J Med 1974; 291(15): 747–751.d

13. Ronco C, Bellomo R, eds. Continuous high flux fldialysis: an effi cient renal replacement. Heidelberg:fiSpringer Verlag; 1996.

14. Cole L, Bellomo R, Journois D, et al. High-volumehaemofi ltration in human septic shock.fi IntensiveCare Med 2001; 27(6): 978–986.

15. Reeves JH, Butt WW, Shann F, et al. Continuous plas-mafi ltration in sepsis syndrome. Plasmafifi ltration in fiSepsis Study Group. Crit Care Med 1999; 27(10):d2096–2104.

16. Ronco C, Brendolan A, d’Intini V, et al. Coupled plasma filtration adsorption: rationale, technicalfidevelopment and early clinical experience. Blood Purif 2003; 21(6): 409–416.

17. Marshall MR, Ma T, Galler D, et al. Sustained low-effi ciency daily diafifi ltration (SLEDD-f) for critically fiill patients requiring renal replacement therapy:

56 J.H. Reeves

towards an adequate therapy. Nephrol Dial Trans-plant 2004; 19(4): 877–884.t

18. Kumar VA, Craig M, Depner TA, Yeun JY. Extendeddaily dialysis: A new approach to renal replacement for acute renal failure in the intensive care unit. Am J Kidney Dis 2000; 36(2): 294–300.

19. Palevsky PM. Renal replacement therapy I: indica-tions and timing. Crit Care Clin 2005; 21(2): 347–356.

20. Mehta RL, Pascual MT, Soroko S, Chertow GM. Diuretics, mortality, and nonrecovery of renal func-tion in acute renal failure. JAMA 2002; 288(20):2547–2553.

21. Cantarovich F, Rangoonwala B, Lorenz H, et al. High-dose furosemide for established ARF: a pro-spective, randomized, double-blind, placebo-controlled, multicenter trial. Am J Kidney Dis 2004;44(3): 402–409.

22. Uchino S, Doig GS, Bellomo R, et al. Diuretics andmortality in acute renal failure. Crit Care Med 2004;d32(8): 1669–1677.

23. Gettings LG, Reynolds HN, Scalea T. Outcome in post-traumatic acute renal failure when continuousrenal replacement therapy is applied early vs. late.Intensive Care Med 1999; 258: 805–813.d

24. van Bommel EF, Kalmeijer MD, Ponssen HH. Treat-ment of life-threatening lithium toxicity with high-volume continuous venovenous hemofi ltration.fi Am J Nephrol 2000; 20(5): 408–411.

25. Maggiore Q, Pizzarelli F, Dattolo P, et al. Cardio-vascular stability during haemodialysis, haemo-fi ltration and haemodiafifi ltration. fi Nephrol Dial Transplant 2000; 15 Suppl 1: 68–73.

26. Gotloib L, Barzilay E, Shustak A, et al. Hemofiltra-fition in septic ARDS. The artificial kidney as anfiartifi cial endocrine lung. fi Resuscitation 1986; 13(2):123–132.

27. Tonnesen E, Hansen MB, Hohndorf K, et al. Cyto-kines in plasma and ultrafi ltrate during continuousfiarteriovenous haemofi ltration.fi Anaesth IntensiveCare 1993; 21(6): 752–758.

28. Cole L, Bellomo R, Hart G, et al. A phase II random-ized, controlled trial of continuous hemofiltrationfiin sepsis. Crit Care Med 2002; 30(1): 100–106.d

29. Uchino S, Bellomo R, Goldsmith D, et al. Super high fl ux hemofifl ltration: a new technique for cytokine firemoval. Intensive Care Med 2002; 28(5): 651–655.d

30. Ronco C, Brendolan A, Lonnemann G, et al. A pilot study of coupled plasma filtration with adsorptionfi

in septic shock. Crit Care Med 2002; 30(6): 1250–1255.

31. Ellison DH. Diuretic therapy and resistance in con-gestive heart failure. Cardiology 2001; 96(3-4): 132–143.

32. Costanzo MR, Saltzberg M, O’Sullivan J, Sobotka P. Early ultrafiltration in patients with decompensatedfiheart failure and diuretic resistance. J Am Coll Cardiol 2005; 46(11): 2047–2051.

33. Bart BA, Boyle A, Bank AJ, et al. Ultrafiltration fiversus usual care for hospitalized patients withheart failure: the Relief for Acutely Fluid-Overloaded Patients With Decompensated Conges-tive Heart Failure (RAPID-CHF) trial. J Am Coll Cardiol 2005; 46(11): 2043–2046.

34. Walters RJ, Fox NC, Crum WR, et al. Haemodialysisand cerebral oedema. Nephron 2001; 87(2): 143–147.

35. Davenport A, Will EJ, Davison AM, et al. Changes inintracranial pressure during machine and continu-ous haemofi ltration.fi Int J Artif Organs 1989; 12(7):439–444.

36. Eknoyan G, Levin N. NKF-K/DOQI Clinical PracticeGuidelines: Update 2000. Foreword. Am J Kidney Dis2001; 37(1 Suppl 1): S5–6.

37. Drazen JM, Ingelfinger JR, Curfman GD. Removal of fiexpression of concern: Schiffl H, et al. Daily hemo-fldialysis and the outcome of acute renal failure. NEngl J Med 2002; 346: 305–310. N Engl J Med 2003;349(20): 1965.

38. Reeves JH, Butt WW. A comparison of solute clear-ance during continuous hemofiltration, hemodiafifi l-fitration, and hemodialysis using a polysulfone hemofi lter.fi ASAIO J 1995; 41(1): 100–104.

39. Ronco C, Bellomo R, Homel P, et al. Effects of differ-ent doses in continuous veno-venous haemo filtration fion outcomes of acute renal failure: a prospective randomised trial. Lancet 2000; 356(9223): 26–30.t

40. Mehta RL, McDonald B, Gabbai FB, et al. A random-ized clinical trial of continuous versus intermittentdialysis for acute renal failure. Kidney Int 2001; t60(3): 1154–1163.

41. Kellum JA, Angus DC, Johnson JP, et al. Continuousversus intermittent renal replacement therapy: ameta-analysis. Intensive Care Med 2002; 28(1):d29–37.

42. Augustine JJ, Sandy D, Seifert TH, Paganini EP. Arandomized controlled trial comparing intermit-tent with continuous dialysis in patients with ARF.Am J Kidney Dis 2004; 44(6): 1000–1007.

57

11Technical Aspects of Renal Replacement Therapy

Sara Blakeley

CRRT on the ICU nowadays is almost exclu-sively venovenous in nature; in other words, blood is removed from a large vein and returned to a large vein. The different modes of CRRT differmainly in their method of solute removal, andmost modern machines are capable of providing the full range of modalities. Newer machines offergreater ease in switching between modes.

There is much discussion regarding the rela-tionship between dialysis dose delivered and sur-vival (6). A prospective randomized study of post dilution CVVH in critically ill patients foundimproved survival with an ultrafiltration dose of fi35 ml/kg/hour (7). Ultrafiltration rate is used as a fidose surrogate. This fi nding was not repeated in a fisubsequent study (8) but a more recent study found did fi nd an improvement in survival when fiadding a dialysis dose to CVVH (9). It is unclearwhether this survival advantage was due to ahigher dose of CRRT overall or due to the addition of a diffusive therapy to a convective one. Thesefindings are being investigated further in otherfistudies, to clarify what is the optimal dose andmeans of delivery. In the mean time 35 ml/kg/hour is often recommended as minimum that should be provided (10–12).

Standard therapies in use are:

Continuous venovenous hemofi ltration (CVVH), ficharacterized by predominantly convective solute clearance (Figure 11.1A).

Continuous venovenous hemodialysis (CVVHD),characterized by predominantly diffusive solute clearance but with some convection occurringbecause of ultrafi ltration.fi

Since its inception in 1977 (1), methods and equip-ment used for the delivery of continuous renalreplacement therapy (CRRT) has undergone many changes. Intermittent hemodialysis (IHD) is per-formed on some intensive care units (ICU), but this chapter will concentrate on continuous therapies.

Mode of CRRT

There is a spectrum of treatment available andwith choice comes debate; continuous versus intermittent, convective versus diffusive therapy (2, 3). Continuous therapies have been associated with better renal survival compared with IHD,although no compelling effect on overall mortality has been seen (4).

Diffusion is the movement of molecules from an area of high concentration (blood) to one of a lower concentration (dialysis fluid circulating flin a counter current direction) across a semi-permeable membrane. With convection a pressure(transmembrane pressure) is applied across the membrane, this drives water out and carries with it dissolves solutes (solvent drag). The “waste” fluid flproduced is termed ultrafiltrate. Both diffusive fiand convective therapies are good at removing small molecular weight molecules (<5000 Da), such as urea and creatinine, but “middle mole-cules” (10,000–50,000 Da) such as β2 microglobu-lin and cytokines (5) may be better removedby convection. Currently, there is no definitive fievidence to suggest one particular mode overanother, and choice is often guided by local expertise.

58 S. Blakeley

Blood flow from patient- ‘arterial’ side

Blood flow to patient- ‘venous’ side

Replacement fluidentering pre filter (predilution)

Ultrafiltrate andused dialysate(waste)

Blood pump

Replacement fluid pump

Effluent pump

HaemofilterPressure monitorAir detector

A

FIGURE 11.1. A, CVVH with prefilter replacement fluid delivery (predilution). B, CVVHDF with postfilter replacement fluid delivery(postdilution).

Blood flow from patient- ‘arterial’ side

Blood flow to patient- ‘venous’ side

Replacement fluidentering post filter(post dilution)

Dialysate fluid(running in acounter currentdirection to bloodflow through thehaemofilter)

Ultrafiltrate andused dialysate(waste)

Anticoagulation syringe

Blood pump

Replacement fluid pump

Effluent pump

Dialysate pump

HaemofilterPressure monitorAir detector

B

11. Technical Aspects of Renal Replacement Therapy 59

FIGURE 11.2. Two examples of hemofiltration machines. The Edwards’ Aquarius System (left) and Hospal Prisma (t right).tt

Continuous venovenous hemodiafiltration fi(CVVHDF), characterized by a mixture of diffusive and convective solute clearance (Figure 11.1B).

Slow continuous ultrafi ltration (SCUF) is removal fiof water in response to a pressure gradient, buta degree of solute removal via convection willalso occur.

Therapeutic plasma exchange (TPE). Plasma isseparated, removed, and replaced with eitheralbumin or fresh frozen plasma.

Hemofiltration Machines (Figure 11.2)

All machines have a similar set up. Newermachines differ in their degree of “user friendli-ness,” their range of pump flow rates, and the fldegree of monitoring.

A roller pump controls the flow of blood from flthe patient through the filtration circuit and then fiback to the patient. Blood flow rate is fl set by the operator but is limited by vascular access quality, blood viscosity, patency of the hemofilter, and car-fidiovascular stability of the patient. The maximalachievable blood flow rate varies between differ-flent makes of machine (50–450 mL/min).

Infusion rates of replacement fluid and/or dial-flysate are set, with newer machines offering increased fl uid removal rates (0–10 L/h) to performflhigh-volume hemofiltration (HVHF). A desired finet fluid loss (0–1000 mL/h) is set by the operator fland constantly monitored by the machine. Themachine generally calculates the ultrafiltration firate depending on the rate of replacement fluid fland fluid removal set by the operator. A bag thenflcollects the effl uent (waste), which is composed of flultrafi ltrate and spent dialysate (if being used).fi

A series of safety mechanisms are in place toprevent the inadvertent introduction of air to thepatient and to detect blood leakage. Alarms warn of pressure changes within the circuit:

• Access pressure: Pressure in the “arterial” limb removing blood from the patient is a negative pressure reflecting blood suction. It is deter-flmined by blood flow, patient’s blood pressure, fland intravascular fl uid status. Excessively flnegative pressures (e.g., >300 mmHg) risk vascu-lar injury and hemolysis and suggest occlusion somewhere in the access limb of the blood circuit, e.g., kinked line or blocked vascular access.

• Filter pressure: A rise in pressure indicates thatthe fi lter may be starting to clot.fi

• Return pressure: Pressure in the “venous” limbreturning blood to the patient is a positive pres-sure refl ecting resistance to venous return. Low flpressures may indicate disconnection but can beseen with changes in position. High pressures are seen with occlusion to fl ow, e.g., kinking or flclotting, but also if the blood flow rate is too flhigh.

Replacement and Dialysis Fluid

With convective therapies, large volumes of fluid fl(effluent) are removed from the patient per hour.flTo maximize solute removal, prevent the patientfrom becoming hypovolemic, and replace wantedelectrolytes, a replacement fluid fl is infused. With diffusive therapies, dialysate fluidfl is run in a dcounterclockwise direction through the fluid flcompartment of the filter, against the blood flfi ow. flThis fluid can be formulated in-house, or more flcommonly as commercially prepared bags of sterile fl uid.fl

60 S. Blakeley

The fl uid contains a buffer, either lactate orflbicarbonate, and electrolytes (sodium, chloride,magnesium, and calcium). Typically, replacementfluids are potassium- and phosphate-free, and flthese may need to be replaced as clinically indi-cated. Glucose is often not present.

Prefilter Versus Postfilter Infusion

Replacement fluid is infused into the circuit eitherflbefore the fi lter (predilution) or after the fifi lter fi(postdilution). Predilution lowers the hematocritof the blood passing through the filter, potentially fireducing anticoagulation needs and allowingincreased ultrafi ltration rates. This is at the expensefiof less-effective solute clearance, therefore, anincreased ultrafi ltration rate is needed to achievefisimilar solute clearances. Newer machines allow acombination of predilution and postdilution.

Lactate Versus Bicarbonate Buffer

Both lactate- and bicarbonate-buffered solutionshave been shown to be effective in correctingmetabolic acidosis (11, 13). Bicarbonate is pre-ferred in patients with a preexisting lactic acidosis(e.g., septic shock) or with liver failure (11) becausethese patients may be unable to metabolize anexogenous lactate load normally, thus, worsening the acidosis. Lactate levels are often seen to rise in other patient groups who have a lactate buffer, butthe signifi cance of this is unclear because hyper-filactatemia is not always associated with an acido-sis. Bicarbonate is a more physiological buffer, but,because it is unstable in solution, it needs to beadded just before use. Studies have compared out-comes of bicarbonate versus lactate buffers butthe evidence is inconclusive. Lactate intolerancehas been arbitrarily defi ned as a rise of greater fithan 5 mmol/L during CRRT (14), and a change toa bicarbonate buffer should be considered.

Hemofilters

Structure

Thousands of hollow fi bers (fi membrane) arebundled together forming a hemofilter with afilarge surface area of 0.6 to 1.2 m2. Pores in themembrane allow the passage of molecules with a

molecular weight less than 50,000 Daltons (Da),i.e., smaller than albumin. Membranes are com-posed of two substances, cellulose (e.g., cupro-phan) and synthetic fi bers (e.g., polysulphone,fipolyamide, or polyacrylonitrile).

Membrane Characteristics

Biocompatibility

Contact of blood with the filter surface can lead to ficomplement and leucocyte activation, triggeringthe coagulation cascade and infl ammatory flpathways. More “biocompatible” indicates lesscomplement/leucocyte activation. Activation of infl ammatory mediators has been suggested as flone mechanism leading to ongoing renal injury,and, therefore, delay or nonreturn of renal func-tion (4). Kidneys that have already been injured because of reduced renal perfusion lose theirability to autoregulate pressure changes and the kidney becomes very sensitive to even smallchanges in renal perfusion. Early reports that bio-incompatible cellulose-based membranes led to a worse outcome have been debated, but, currently,the evidence is not robust enough to defi nitively firecommend a synthetic membrane over cellulose or modified cellulose membrane (15). As it is,fimost membranes used in CRRT are syntheticbecause they have a greater degree of flux.fl

Flux

This is a measure of ultrafi ltration capacity and isfibased on the membrane ultrafiltration coeffifi cient. fiA lter with a high permeability coeffifi cient to fiwater will allow more ultrafi ltration (high flfi ux) fland, hence, more convective transport. Permeabil-ity is a measure of the clearance of middle molec-ular weight molecules and high permeability is seen with high flux membranes (synthetic mem-flbranes). It should be noted that high permeability does not always equate to high urea clearance(efficiency).fi

Vascular Access

When initially developed, CRRT used an arterialand a venous catheter (arteriovenous) (1). A wide-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.Mostly composed of polyurethane, they can havean antibiotic/antimicrobial coating.

Blood is pumped from the patient (arterial side) through proximal side holes into one lumen,and is then returned though a port at the distal tip of the second lumen (venous side). High blood flows without high pressures are ideal catheterfldesign requirements, and the dual-lumen designallows continuity and reduces recirculation.

Vascular catheters differ in length, diameter of lumen, and positioning of ports, and some have an extra lumen added for drug infusions. Remember: always assume that EACH catheter limb contains heparin and aspirate at least 5 mL of blood beforeusing. Cuffed dialysis catheters are generally not used on the ICU, but may be indicated in stable patients who are free of infection, and who requireongoing renal replacement therapy (RRT) (Table11.1).

Positioning

A correctly positioned catheter will have a betterblood fl ow;fl good access is the key to good dialysis. The tip of a jugular or subclavian catheter should extend to the superior vena cava and rest 1 to 2 cm above the right atrium. Too short, and there isthe possibility of recirculation, whereas a catheter that is too long risks atrial perforation. Femoral catheters should be longer (>20 cm) to reach the inferior vena cava, therefore, minimizing recircu-lation and achieving better flow rates.fl

Site of Catheter

Debate continues regarding the ideal site for place-ment of dialysis catheters in terms of safety of insertion and infection risk. Where long term dialysis is a possibility, there is concern that subclavian catheters may be associated with an increased incidence of subclavian stenosis/throm-bosis, creating long-term problems for arteriove-nous fi stula formation.fi

Anticoagulation

As blood flows through the fifl lter and flfi uid is flremoved, viscosity increases and there is a ten-dency toward fi lter clotting. Passage of blood fi

through the circuit can also result in the formation of platelet microthrombi, which can occlude thefilter. Loss of the fifi lter through clotting can lead to fiineffective dialysis and patient blood loss, as wellas being a drain on resources, both nursing and financial. Methods such as predilution and ensur-fiing adequate vascular access can be used, but someform of anticoagulation for the extracorporealcircuit is often needed. Remember: circuit failure is more often caused by inadequate vascular access rather than inadequate anticoagulation.

No Anticoagulation

In the setting of deranged clotting (e.g., interna-tional normalized ratio [INR] >2, activated partial thromboplastin time [aPTT] >60 s) and/or throm-bocytopenia (e.g., platelet count <50,000) or a high risk of bleeding, further anticoagulation isoften not necessary or carries the risk of bleeding. With adequate access and predilution, it is possi-ble to run the circuit without any anticoagulationfor an acceptable period of time and achieve good solute clearance.

Unfractionated Heparin

Unfractionated heparin (UFH) is the most com-monly used extracorporeal anticoagulant. Thecircuit is often primed with heparin (e.g., 1000–10,000 IU) because it is highly negatively charged and is absorbed onto the plastic circuit. Depend-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) aims NOT to prolong the aPTT. If clotting occurs,a medium dose can be considered (5–10 IU/kg/h),aiming for mild prolongation of the aPTT (1–1.4times normal) (16). Prefilter heparin can be neu-fitralized with postfilter protamine (e.g., 1000 IU/h fito 10 mg/h), called regional heparinization. If patients require formal heparinization for condi-tions such as a pulmonary embolus, this should be continued, no extra “fi lter” heparin is needed.fi

The methods, site of sampling, and frequency of anticoagulation monitoring vary depending on local protocols. Commonly used methods are the activated coagulation time (ACT) and aPTT. However, it should be remembered that there is not always a linear correlation between dose of heparin or degree of anticoagulation and filter life.fi

62 S. Blakeley

Low Molecular Weight Heparin

Compared with UFH, low molecular weight heparin (LMWH) is considered more effective inreducing fi brin deposition on dialyzer membranesfiand, thus, preventing circuit clotting, however, its superiority to UFH has not been proven in trials. There is a reduced incidence of heparin-induced thrombocytopenia syndrome (HITS) comparedwith UFH, but it is more costly. With more experi-ence LMWH use has increased in popularity. Stan-dard markers of anticoagulation are not reliableand anti-Xa levels should be monitored with pro-longed use (target, 0.25–0.35 U/mL) (16). However, there is not always a correlation between anti-Xalevels and fi lter life.fi

Regional Citrate Anticoagulation

Infused citrate complexes with calcium and pre-vents activation of the coagulation cascade andplatelets. Post dialyser, calcium is reinfused. Regional citrate anticoagulation (RCA) is an effec-tive form of anticoagulation, and because only the circuit is anticoagulated, it is safe to use in patients at risk of bleeding. However, its sideeffects and complex infusion protocol have limitedits widespread use. Side effects include citrate toxicity if citrate is not metabolized rapidly or adequately (e.g., in liver failure), hypernatre-mia, hyper/hypocalcemia, and metabolic alkalosis(each citrate molecule is metabolized to threebicarbonates).

TABLE 11.1. Complications of RRTa

Access related

Complications during insertion Bleeding, local traumaInfection Systemic or localCatheter-related thrombosisPatient immobility Particularly with femoral lines

Circuit related

Membrane bioincompatibility See textAir embolismBlood loss Caused by clotted filters (common) or disconnection (rare)Fluid balance errorsHemolysis

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 bioincompatibilityAnticoagulation-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 hypoglycemiaAcid-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 RCATemperature disturbances A degree of cooling always occurs, this may lead to “normothermia” in febrile patients (i.e., masking a pyrexia) or cause marked hypothermiaVitamin and micronutrient depletion Water-soluble vitamins, trace minerals, certain hormones (e.g., glucocorticoids), amino acidsInappropriate prescribing of drugs Generally leads to underdosing while a patient is on the filterFurther 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

aACEI, angiotensin converting enzyme inhibitor.

11. Technical Aspects of Renal Replacement Therapy 63

Prostacyclin

Prostaglandin (PG)-I2 is a natural anticoagulantthat is a potent antiplatelet agent and has beenshown to reduce platelet microthrombi duringdialysis. It is often used in patients with a high risk of bleeding, but because it is a potent arterialvasodilator, some patients develop symptomatic hypotension. It has been used on its own and in combination with low-dose heparin.

Other Anticoagulants

Heparinoids (e.g., danaparoid) have minimaleffects on platelets and can be used in HITS (butremembering the potential cross reactivity in5–10% of patients). However, standard markers of anticoagulation are not reliable and its effect is prolonged in renal failure. Factor Xa inhibitors(e.g., fondaparinux) and direct thrombin inhibi-tors (e.g., recombinant hirudin) can be safely used in HITS, but, to date, have limited use in CRRT.

Comment

A recent systematic review (16) found that there was no conclusive evidence to suggest one strat-egy over another, but the chosen method shouldtake into account patient characteristics and localfacilities. Heparin (UFN and LMWH) has the greatest evidence and experience behind it, but RCA is increasing in popularity and ease of use.

References

1. Kramer P, Wigger W, Rieger J, et al. Arteriovenous haemofi ltration: a new and simple method for treat-fiment of over-hydrated patients resistant to diuret-ics. Klin Wochenschr. 1977; 55: 1121–1122.

2. Palevsky PM. Dialysis Modality and Dosing Strat-egy in Acute Renal Failure. Sem Dialysis. 2006; 19: 165–170.

3. Van Biesen W, Vanholder R, Lameire N. Dialysisstrategies in critically ill acute renal failure patients.Curr Opin Crit Care. 2003; 9: 491–495.

4. Palvesky PM, Baldwin I, Davenport A, et al. Renalreplacement therapy and the kidney: minimisingthe impact of renal replacement therapy on recov-

ery of acute renal failure. Cur Opin Crit Care. 2005; 11: 548–554.

5. Ricci Z, Ronco C, Bachetoni A, et al. Solute removalduring continuous renal replacement therapy in critically ill patients; convection versus diffusion.Crit Care. 2006; 10: R67–R74.

6. Clark WR, Turk JE, Kraus MA, Gao D. Dose deter-minants in continuous renal replacement therapy.Artif Organs. 2003; 27: 815–820.

7. Ronco C, Bellomo R, Homal P, et al. Effects of dif-ferent dose in continuous veno-venous haemofiltra-fition on outcomes of acute renal failure: a prospective randomised trial. Lancet. 2000; 356: 26–30.

8. Bouman C, et al. Effects of early high-volume con-tinuous Venovenous hemofiltratin of survival and firecovery of renal function in intensive care patientswith acute renal failure: a prospective randomizedtrial. Crit Care Med. 2000; 30: 2205–2211.

9. Saudan P, et al. Adding a dialysis dose to continuoushemofi ltration increases survival in patients with fiacute renal failure. Kidney Int. 2006; 70: 1312–1317.

10. Cariou A, Vinsonneau C, Dhainaut JF. Adjunctivetherapies in sepsis: an evidence-based review. Crit Care Med. 2004; 32: S562–S570.

11. www.adqi.net12. Ronco C. Renal replacement therapy for acute

kidney injury: let’s follow the evidence. Int J Artif Organs. 2007; 30: 89–94.

13. Naka T, Bellomo R. Bench-to-bedside review: treat-ing acid-base abnormalities in the intensive care unit–the role of renal replacement therapy. Crit Care. 2004; 8: 108–114.

14. Hilton PJ, Taylor J, Forni LG, Treacher DF. Bicarbo-nate-based haemofiltration in the management of fiacute renal failure with lactic acidosis. QJM. 1998; 4: 279–283.

15. Teehan GS, Liangos O, Lau J, et al. Dialysis mem-brane and modality in acute renal failure: under-standing discordant meta-analyses. Sem Dialysis. 2003; 16: 356–360.

16. Oudemans-van Straaten HM, Wester JPJ, de PontACJM, Schetz MRC. Anticoagulation strategies in continuous renal replacement therapy: can the choice be evidence based? Intensive Care Med. 2006; 32: 188–202.

Suggested Reading

Bellomo R, Baldwin I, Ronco C, Golper T. Atlas of Hemo-filtration.fi WB Saunders. 2002.

64

12End-Stage Renal Disease

Emile Mohammed

Hemodialysis

Many hemodialysis (HD) techniques have been developed, particularly in the ICU setting, ranging from conventional HD, high-flux HD, hemodiafifl l-fitration, and hemofi ltration. These techniquesfisimply use varying degrees of convection or diffusion. For example, hemofiltration is a convec-fitive treatment with good clearance of mid-sizemolecules but with poor small molecule clear-ance. The converse holds for conventional inter-mittent HD.

Peritoneal Dialysis

The peritoneum acts as a natural semipermeablemembrane. Dissolved waste products and water pass from the blood, via the peritoneal capillaries,through the mesothelial cells and interstitium tothe peritoneal dialysis (PD) fluid (PDF). This flprocess is referred to as ultrafi ltration (Figurefi12.2). Water-soluble waste products pass down a concentration gradient that is generated by anosmotic gradient. This, in turn, is created by glucose or glucose polymers added to the PDF.

PD regimens are all based on repetitions of a basic cycle, which comprises inflow of PDF, a dwellfltime of the PDF within the peritoneal cavity, andthen drainage. The various types of PD are all based on this principle. They include continuous ambulatory PD (CAPD), automated PD (APD),tidal PD, and intermittent PD. Typical regimes are illustrated in Figure 12.3.

There are now approximately one million people on renal replacement therapy worldwide. In thecurrent era of chronic noncommunicable disease, this number is set to double within the next decade. Patients with end-stage renal disease (ESRD) carry a significantly higher cardiovascularfimorbidity and mortality compared with the general population. This is because of the “uremic”cardiovascular factors (Table 12.1). The result is that there will be a growing number of ESRDpatients being managed within the intensive careunit (ICU) setting.

Hemodialysis and Peritoneal Dialysis

Mechanisms of Dialysis

There are two basic principles of dialysis thatallow the body’s homeostasis to be achieved in the absence of a natural kidney. They are as follows:

• Convection, in which there is movement (inlarge volumes) of solvent, which drags dissolvedsolute across a membrane with a hydrostatic pressure gradient.

• Diffusion, in which there is passive movement of solute from a high- to a low-concentration gradient across a membrane (Figure 12.1). Dif-fusion depends not only on the transmembranegradient but the membrane characteristics (e.g.,pore size).

Diffusion is more effective in clearing smallmolecules and convection improves mid-size mol-ecule clearance.

12. End-Stage Renal Disease 65

FIGURE 12.1. Diagrammatic representa-tion of blood purification within thedialyzer.

TABLE 12.1. Cardiovascular risk factors in the uremic patient

Traditional coronary risk factors Uremic-related cardiovascular(The Framingham Study) risk factors

Hypertension Increased extracellular fluid volume

High levels of low-density Calcification and highlipoprotein calcium/phosphate product

Low levels of high-density Parathyroid hormonelipoprotein

Smoking AnemiaDiabetes Oxidant stressOlder age MalnutritionMale sex Pulse pressureWhite race TriglyceridesPhysical inactivity Lipoprotein remnantsMenopause Lipoprotein ALeft ventricular hypertrophy Homocysteine

Inflammation (C-reactive protein)Sleep disorders

S S S S SSeeeeemi-pe-pe-pepe-permermrmermermeabablablablablle ee mmmmmmembemembbembbranranranranraneeeee

Urea, creatinine

Sodium

Water

Dialysate flow (400-800ml/min)

Blood flow (250-500ml/min)

Potassium

Calcium, bicarbonate

Clinical Parameters

Much debate surrounds the “optimal” dialysisdose, although a minimal dialysis dose has beenuniversally accepted. Within the context of thrice-weekly HD, there seems to be no added benefit of fihigh-dose dialysis compared with the conven-tional dose of dialysis (the HEMO study) (1).Dialysis adequacy is measured by urea kinetic modeling (UKM) using the urea reduction ratio,or Kt/V, where K is the dialyzer urea clearance, tis the duration of dialysis, and V is the urea distri-bution volume. In a well-nourished stable HDpatient, a Kt/V of 0.8 to 1.0 is the minimum accept-able threshold per dialysis session. The Adequacy

of Peritoneal Dialysis in Mexico (ADEMEX) study (2) again reveals the same controversy of defining fithe optimal dialysis dose in PD patients. In this study, there was a neutral effect on patient survival between a control group on conventional CAPDcompared with a study group on a modified pre-fiscription, which achieved increased small soluteclearance, measured by peritoneal creatinine clearance and peritoneal Kt/V.

There are no “fi xed targets” that determine fidialysis adequacy in the ICU setting. Dialysisdose and duration must be determined by balancing the clinical condition of the patient,while achieving as normal a physiological state as possible.

66 E. Mohammed

Typcal CAPD regime

0

0.5

1

1.5

2

2.5

3

Time (during day)

“Bag in”

PD

F v

olum

e (l

itre

s)

Bag drained out

Typical APD regime

0

0.5

1

1.5

2

2.5

“Bag in”

wet day overnight exchanges

PD

F v

olum

e (l

itre

s)

FIGURE 12.3. PD regimes.

Peritoneal cavity Interstitium

Soluble waste products and water

Glucose

Capillary

Mesothelium FIGURE 12.2. Ultrafiltration in PD.

12. End-Stage Renal Disease 67

The following clinical parameters act as guide-lines to achieve this:

• Target weight and blood pressure control. Target weight is defi ned as the patient’s weight in which fiall the fl uid compartments are physiologically flnormal. Excess weight (which will be essentially salt and water) results in hypertension. Thetarget weight is achieved by gradual weight reduction on successive dialyses until the patient is free from both pulmonary and peri-pheral edema, but, below which, hypotensionoccurs.

• Acid-base balance. Dialysis must be performed frequently and long enough to maintain normalacid-base balance.

• Bone biochemistry. Along with vitamin D sup-plementation, serum calcium and phosphate levels should be maintained within normallimits.

• Nutritional state. It is important to rememberthat a high proportion of ESRD patients withinthe ICU will have a low serum albumin, low body mass index, an infl ammatory and/or flhypercatabolic state, and a low dietary intake. Itis, therefore, necessary to obtain dietary advice,treat correctable factors, give dietary supple-ments, and have a low threshold for nasogastric (NG), percutaneous endoscopic gastrostomy (PEG), or even parenteral nutrition, if indicated. Dialysis prescriptions must accommodate these nutritional requirements.

A good starting point for prescribing dialysis would be to continue the patient’s regular dialysis regime and adjust the dose of dialysis, in conjunc-tion with the nephrologists, to achieve the aboveparameters.

Dialysis-Related Complications

Hemodialysis

• Hypotension. Hypotension can be minimized with an accurate assessment of target weight, judicious use of antihypertensive medications,sodium restriction, increasing treatment dura-tion, and careful choice of dialysis modality, e.g.,hemodiafiltration in the cardiovascularly unsta-fible patient.

• Anaphylaxis. Anaphylaxis can occur by comple-ment activation with the use of a bioincompati-ble membrane and normally occurs within thefi rst 20 minutes of treatment.fi

• Catheter related-sepsis. Catheter related-sepsisrequires aggressive antibiotic treatment andcatheter removal. If temporary catheters arebeing used, once weekly catheter changes are recommended.

• Pyrogenic reactions. Uncommon if ultrapure water is used.

• Dialysis equilibrium syndrome. Rare in estab-lished dialysis patients. It can occur from overag-gressive dialysis causing a rapid reduction in serum osmolality and resulting in cerebraledema.

• Modern fail-safe machines minimize other com-plications such as air embolism and accidental circuit disconnection.

Peritoneal Dialysis

Although PD is a technically safe procedure, there may be clinical reasons to convert to temporary HD.

These are as follows:

• Abdominal surgery• Diaphragmatic fl uid leak resulting in effusionsfl• Respiratory compromise from splinting of dia-

phragm by PDF• Severe hypoalbuminemic state• Peritonitis or catheter-related sepsis• Inadequate ultrafiltration in the context of fi

aggressive fluid management and/or hypercata-flbolic state of the patient

Renal Transplantation

Renal transplantation represents the best mode of therapy for ESRD patients, both in cost effective-ness and quality of life (3). There have been many improvements in renal transplantation, such as therefi nement of immunosuppression regimens, andfipatient-donor selection and work-up as well as their compatibilities. The major challenge facingtransplantation is that its demand far outstrips the availability. Every effort should be made to increasethe number of donors. In parallel to this, there is

68 E. Mohammed

much research in the development of stem cell transplantation and xenotransplantation.

Evaluation, Selection, and

Preparation of the Potential

Transplant Recipient

General Evaluation

There are few absolute contraindications to renaltransplantation. These are uncontrolled cancer,HIV positivity, active systemic infections, and/or any condition with a life expectancy of shorterthan 2 years. Conditions increasing the risk of posttransplant morbidity and mortality includelong duration of dialysis, previous incidence of recurrent infections, cardiovascular disease, andgastrointestinal complications. Such patientsrequire a particularly careful work-up and aggres-sive management of risk factors (e.g., hyper-tension, obesity, and vascular disease) before transplantation.

Psychological Evaluation

The use of psychiatric screening is not universally adopted but may be useful in assessing compli-ance with immunosuppressive regimes. Poorcompliance signifi cantly worsens renal allograftfioutcomes.

Recurrent Renal Disease

It is important to ascertain the underlying causeof renal failure because some diseases recur in the transplanted kidney, most notably, focal and seg-mental glomerulosclerosis.

Immunological

ABO blood group must be compatible. HLA typingof donors and recipients allows assessment of compatibility. HLA DR is more important thanHLA B, which is more important than HLA A.A lymphocyte cross-match is also performed: therecipient is screened for preexisting antibodies todonor lymphocytes, which arise in response toprevious blood transfusions, pregnancies, or renal

allografts. Such sensitization can cause severe hyperacute rejection.

Evaluation and Selection of Donors

There are two sources of donors:

Cadaveric: Cadaveric kidneys may be either from patients with brainstem death and a maintainedcardiac output or from nonheart-beatingdonors. Donors with sepsis, malignancy, infec-tion with hepatitis B, hepatitis C, HIV, or tuber-culosis, or irreversible renal failure are notconsidered for donation.

Live: The use of kidneys from living donors is recommended for renal transplantation when-ever possible, in light of the growing body of evidence of favorable outcomes after transplan-tation. Before being selected as a living donor,thorough counseling, medical, physical, andpsychological evaluation is performed. Outcomestudies have revealed lower mortality rates inliving donors compared with the general popu-lation. This is probably caused in part by patient selection and the fact that this group of patients receive long-term medical follow-up.

Immunosuppression

The immunosuppression regime is tailored to each patient in an effort to minimize rejectionas well as side effects. The following is a brief summary of drugs used, usually in combination.The commonest regime is triple therapy, e.g.,cyclosporin, azathioprine, and prednisolone.

Corticosteroids have broad but potent immuno-suppressive actions. They are used in high dosesin induction therapy as well as for episodes of acute rejection. They are tapered to small mainte-nance doses or stopped completely over time.Because of the broad actions of corticosteroids,there are a large range of side effects.

Cyclosporin is a calcineurin inhibitor. Althoughthere is little myelotoxicity, cyclosporin is nephro-toxic and does contribute to chronic allograft rejection. It is an important agent, because itsintroduction improved 1-year graft survival by 15to 20%.

12. End-Stage Renal Disease 69

Tacrolimus is also a potent calcineurin inhibitor. Its side effect profi le is similar to cyclosporin butfiseems to be more diabetogenic, particularly in theblack population.

Azathioprine has been widely used for trans-plantation and it continues to be an integral partof many immunosuppression regimens. It inhibitspurine metabolism and, therefore, cellular prolif-eration. There is a significant side-effect profifi le, fimost notably its myelosuppressive effect, which isworsened by concomitant use of other drugs such as allopurinol.

Mycophenolate mofetil acts similarly to azathio-lprine but is more specifi c for lymphocytes. fiMyelosuppression must also be monitored andmycophenolate mofetil has more gastrointestinal side effects than azathioprine.

Daclizumab and basiliximab are anti-CD25 antibodies. They target activated T cells only andare used as an induction agent to prevent early rejection.

Polyclonal antibodies to T-cells, ALG and ATG, aswell as monoclonal antibody to T cells, OKT3, areused to treat refractory acute rejection and some-times used as an induction agent to prevent rejec-tion. These drugs are used cautiously because theiradministration is associated with cytokine release syndrome and pulmonary edema. Other side effects include subsequent infection and posttrans-plantation lymphoproliferative disease (PTLD).

Complications of Transplantation

Immediate

• Acute tubular necrosis is the commonest cause of early graft function and is more likely to occur with prolonged ischemic times and asys-tolic, hypotensive, or elderly donors.

• Surgical complications are now the commonest cause of early graft loss. These include renal veinand arterial thrombosis and urinary leaks. Thesecomplications are surveyed for with Dopplerultrasound and isotope scanning. Occasionally renal angiography or surgical reexploration isrequired in the anuric kidney.

• Hyperacute rejection is now very uncommon,with the more meticulous screening for thepresence of preexisting antibodies. The only available treatment is graft nephrectomy.

Early

• Acute rejection should be suspected in patients with established graft function who then experi-ence a rise in serum creatinine. A biopsy is usually required to confirm the clinical dia gnosis fiand the treatment involves high-dose steroids,usually with some modification of the immuno-fisuppression regime. Antibodies are consideredin steroid-refractory acute rejection.

• Infectious complications tend to vary with time after transplantation. Within the first month fiafter transplantation, most infections are sur-gically related, such as atelectasis and wound infection. From 1 to 6 months after transplanta-tion, opportunistic infections can emerge, suchas Pneumocystis carinii and Aspergillus fumiga-tus. Clinical infection caused by the effects of modulating viruses, cytomegalovirus (CMV) and Epstein-Barr virus (EBV) are serious com-plications for which there should be surveillancewith prophylaxis administered where appropri-ate. PTLD in the early period tends to be anEBV-related malignancy.

• Mechanical complications, such as arterial andureteric stenoses and lymphoceles exerting localpressure, sometimes evolve as a cause of deteri-orating graft function.

Late

• Chronic rejection occurs secondary to a combi-nation of immunological and nonimmunolo-gical factors. The result is an irreversibleprogressive decline in graft function and is oftenassociated with proteinuria.

• Recurrence of original disease may result in graft failure. There is a particularly high rate of recur-rence of focal and segmental glomerulosclerosisbut it is diffi cult to predict and, therefore, is not fia contraindication to transplantation.

• Cardiovascular disease is the main cause of death in the transplant population and the rate of cardiovascular disease is higher than in the general population. This is because of the addi-tional risk factors of the ESRD population (Table12.1) as well as the adverse effects of the immu-nosuppressive agents.

• Malignancy, in particular, skin cancer, is three times more common than the general population.

70 E. Mohammed

Other cancers that should be screened for include renal, cervical, and vaginal cancers.

Summary

ESRD and its complications are becoming morecommonplace, particularly in the ICU environ-ment. The challenges associated with this groupof patients also continue to rise and requires a multidisciplinary approach, which includes theintensivist and the nephrologist.

References

1. Eknoyan G, Beck GJ, Cheung AK, et al. Hemodialysis(HEMO) Study Group. Effect of dialysis dose and

membrane fl ux in maintenance hemodialysis. fl N Engl 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 affectedby peritoneal clearance: The ADEMEX trial. KidneyInt 2005; 67(3): 1093–1104.t

3. The EBPG Expert Group on Renal Transplantation. European Best Practice Guidelines for Renal Trans-plantation (Part 1). Nephrol Dial Transplant 2000;t15(Supp 7): 1–85.

Suggested Reading

Davison AM, Cameron S, Grunfeld J-P, et al. OxfordTextbook of Clinical Nephrology. 3rd edition. Oxford:Oxford University Press. 2005.

Levy J, Brown E, Morgan J. Oxford Handbook of Dialy-sis. Oxford: Oxford University Press. 2001.

71

13Clinical Hyperkalemia and Hypokalemia

Harn-Yih Ong

excreting 90 to 95% of the daily potassium load. Decreased renal excretion of potassium is, there-fore, the most important cause of hyperkalemia. Hyperkalemia may follow failure of glomerularfiltration or tubular secretion of potassium.fi

Clinical Features

Hyperkalemia is often asymptomatic, but severehyperkalemia (potassium >6.5 mmol/L) may pre-sent with neuromuscular disturbances such as distal paresthesia, generalized muscle weakness,an ascending flaccid paralysis, or ventilatory flfailure. Sudden cardiac death may occur when plasma potassium concentration exceeds 7.0 to7.5 mmol/L, and this may be the earliest and only manifestation. Acute hyperkalemia is much lesswell tolerated than chronic hyperkalemia.

Electrocardiogram Changes

Some patients may show a gradual progressionof electrocardiogram (ECG) findings but may fiprogress rapidly without warning. Cardiac arrestcaused by asystole, ventricular tachycardia, or ven-tricular fibrillation may occur at any point alongfithis progression. Hyperkalemia can be life threat-ening even if the ECG is normal. Approximately 50% of patients with potassium levels exceeding 6.0 mmol/L have a normal ECG.

• K+, 5 to 6 mmol/L: peaked T waves and short-ened QT interval

Hyperkalemia

Hyperkalemia is defined as a plasma potassium ficoncentration greater than 5.0 mmol/L and refersto an excess concentration of potassium ions inthe extracellular fluid (ECF) compartment.fl

Causes of Hyperkalemia

(Tables 13.1 and 13.2)

Because of the ability of the kidneys to excrete a large amount of potassium, hyperkalemia causedby accelerated exogenous intake usually indicatesthe presence of a subtle or overt defect in renal potassium handling or an altered transcellulardistribution.

Compartmental Shift

Potassium is the principal intracellular cation, and maintenance of the normal distribution of potas-sium between the intracellular and extracellularcompartments relies on several regulatory mecha-nisms. Under normal conditions, ingested potas-sium is absorbed into the portal circulationand rapidly shifted into cells by Na+/K+-ATPase under the influence of insulin and circulatingfl β-adrenergic catecholamines. When these mecha-nisms are perturbed, hyperkalemia may occur.

Decreased Renal Excretion

The renal system maintains external potassiumbalance in the long term and is responsible for

72 H.-Y. Ong

TABLE 13.1. Extrarenal causes of hyperkalemia

Increased intake Exogenous sources Potassium supplements (especially with advanced age, diabetes mellitus, underlying renal impairment, and/or the concurrent use of potassium-sparing diuretics) Stored packed red blood cells. Cardiac tolerance to hyperkalemia is decreased by hypocalcemia because of the presence of the anticoagulant, citrate Potassium penicillin G Salt substitutes Endogenous sources (tissue breakdown) Rhabdomyolysis. Each kilogram of lean muscle mass contains more than 100 mmol of potassium Hemolysis Tumor lysis syndrome Reperfusion syndrome. Ischemia results in acidosis in the affected area and an outflow of intracellular potassium, this potassium is “washed out” when the region is reperfused Catabolic statesCompartmental shift Inhibition of Na+/K+-ATPase Insulin resistance/deficiency

β2-adrenergic catecholamine deficiency or resistance; e.g., β-blockersDrugs and toxins; e.g., cardiac glycosides

Altered transcellular electrochemical Inorganic metabolic acidosis K+ gradient ECF hypertonicity or hyperosmolar states; e.g., hyperglycemia, hypertonic solutions such as mannitol,

hypertonic saline Increased conductance of K+ channels Drugs; e.g., depolarizing muscle relaxants (succinylcholine) Familial hyperkalemic periodic paralysis

TABLE 13.2. Renal causes of hyperkalemia

Inadequate sodium delivery to the cortical collecting duct Intravascular volume depletion Acute circulatory failure, e.g., vasodilatory or septic shock Severe cardiac failure Hepatic cirrhosis with ascites Renal failure Acute (oliguric) and chronic renal failure

Secondary hypoaldosteronism Low renin Suppressed renin secretion, e.g., cyclosporin, tacrolimus Distal (type IV) renal tubular acidosis, e.g., associated with diabetic nephropathy, tubular interstitial nephritis, inhibitors of prostaglandin production Normal/high renin Impaired aldosterone production, e.g., heparin Impaired angiotensin II production/action, e.g., angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers Normal/high renin, low cortisol Primary adrenal insufficiency, e.g., Addison’s disease (autoimmune) Impaired adrenal enzyme biosynthesis, e.g., azole antifungals (ketoconazole), congenital adrenal enzyme defects (rare)

Primary hypoaldosteronism (rare)Tubular hyperkalemia without aldosterone deficiency (aldosterone resistance) Renal distal tubular potassium Aldosterone antagonists, e.g., spironolactone secretory defect (acquired) Luminal sodium channel blockade, e.g., amiloride, cimetidine, trimethoprim, pentamidine Inhibition of basolateral Na+/K+-ATPase, e.g., cyclosporine Acute or chronic tubulointerstitial nephritis, e.g., diabetic nephropathy, obstructive uropathy, postrenal transplant, lupus nephritis, sickle cell nephropathy, amyloidosis Pseudohypoaldosteronism Type I and II. Rare

• K+, 6 to 7 mmol/L: prolonged PR interval, AVdissociation, flattening and loss of P wave, andflwidening of QRS complex

• K+, greater than 7 to 8 mmol/L: sine wave pattern, ventricular fi brillation, and asystolefi

Pseudohyperkalemia

Pseudohyperkalemia is caused by the release of potassium from damaged cells in vitro (duringor after the removal of the blood sample from

13. Clinical Hyperkalemia and Hypokalemia 73

the patient) and can produce a 1 to 2 mmol/L arti-factual increase in potassium levels. The major causes are:

• Contamination by drip site or laboratory error• Hemolysis caused by mechanical trauma during

venipuncture• Thrombocytosis (platelet count >900 × 1010

cells/L)• Leukocytosis (white cell count >70 × 109 cells/L)

or mononucleosis (“leaky red blood cells”)• Familial pseudohyperkalemia or other uncom-

mon genetic syndromes

Management

Severe hyperkalemia (plasma potassium >6.5 mmol/L or with ECG changes) is a medical emergency and therapy should begin immediately.

Stabilization of Cardiac Membranes:

Intravenous Calcium

Although no randomized studies exist to supportthe use of calcium salts administered intrave-nously (e.g., 10 mL 10% calcium gluconate over2–3 min into a large vein), it is still recommended as fi rst-line therapy in the presence of ECG changesfior arrhythmia. The protective effect of calcium isevident within minutes and lasts for 30 to 60 minutes. Caution should be used in patients whotake digoxin because calcium has been reported to worsen the myocardial effects of digoxin toxic-ity. An alternative is to consider using magnesium instead of calcium to stabilize the myocardium. Calcium may reduce the immediate risk of cardiacarrest, but represents a temporizing measure only.Plasma potassium concentration is unaltered.

Transcellular Redistribution

• Insulin and dextrose: Administration of insulinalways causes a transient reduction in plasma potassium because the activated insulin recep-tor stimulates Na+/K+-ATPase, driving cellular uptake of potassium. The use of insulin and glucose (e.g., 50 mL of 50% glucose to prevent hypoglycemia) for the emergency treatment of

hyperglycemia is effective and rapid in onset of action. For example, 10 U intravenous insulincan be expected to lower the plasma potassiumconcentration by 0.5 to 1.5 mmol/L within 15 minutes, lasting 2 to 4 hours.

• Sodium bicarbonate: Increasing the pH of the ECF compartment with sodium bicarbonate(e.g., 100 mmol over 1–2 hours) will shift potas-sium into cells as an acidosis is corrected. Its effectiveness is debated, but it is still recom-mended when hyperkalemia is associated withsevere inorganic metabolic acidosis. Ensur-ing effective diuretic therapy first lessens the filikelihood of developing volume overload as acomplication.

• Salbutamol: Meter-dose inhalers, nebulized, or intravenous salbutamol seem equally effective in reducing potassium levels.

Removal of Excess Potassium

Treatments that shift potassium into the cells haveno effect on total body potassium. Excess totalbody potassium may be removed by gastrointes-tinal elimination, renal excretion, or renal replace-ment therapy.

• Renal excretion of potassium may be enhanced with the use of diuretics, especially loop diuret-ics, which increase the delivery of sodium andthe urine fl ow rate to the cortical collecting duct,flincreasing tubular secretion of potassium. If the patient is volume depleted, hydration with salinecan be administered with the diuretic.

• Although widely used clinically in the treatmentof hyperkalemia, cation exchange resins do not seem to be effective at 4 hours and should notbe relied on for rapid effects.

• Hemodialysis or continuous renal replacement therapies are the treatments of last resort, withthe exception of patients already receiving these therapies.

Hypokalemia

Hypokalemia is usually defined as a plasma potas-fisium concentration of less than 3.5 mmol/L, and refers to a deficit of potassium ions in the ECF ficompartment.

74 H.-Y. Ong

Causes of Hypokalemia (Table 13.3)

Hypokalemia indicates a disruption of normal potassium homeostasis and is almost always caused by potassium depletion caused by increasedlosses, either through the kidneys or gut. In eithercase, drugs are the most common cause. Less fre-quently, hypokalemia occurs because of an acute shift of potassium from extracellular to intracel-lular fl uid (ICF) compartments (compartmentalflshift). For any given cause, the magnitude of thechange is relatively small (usually <1.0 mmol/L),but a combination of factors may lead to a signifi-ficant fall in plasma potassium.

Familial Hypokalemic Periodic Paralysis

Familial Hypokalemic Periodic Paralysis is anuncommon inherited (autosomal dominant) dis-order characterized by recurrent episodes of muscle paralysis that subside spontaneously after 6 to 24 hours. The primary defect is the gene encod-ing the skeletal muscle dihydropyridine receptor, avoltage-gated calcium channel. Acute attacks are precipitated by a reduction in plasma potassium, such as after carbohydrate-rich meals (increasedinsulin) or rest after exertion (increased catechol-amines). Administration of potassium is lifesaving and should be administered to treat acute attacks.

TABLE 13.3. Causes of hypokalemia

Inadequate intake of potassium (rare)Compartmental shift Stimulation of Na+/K+-ATPase-mediated Endogenous or exogenous insulin cellular potassium uptake

β2-adrenergic receptors activation or indirect β2 sympathomimetics (e.g., salbutamol, dobutamine) Increased endogenous adrenaline release (e.g., delirium tremens, hypoglycemia, coronary ischemia, after cardiopulmonary resuscitation) Phosphodiesterase inhibition (e.g., xanthines: milrinone, theophylline, and caffeine) Thyrotoxic hypokalemic periodic paralysis Rapid cell growth (e.g., after vitamin B12 treatment for pernicious anemia) Metabolic alkalosis In general, [K+]Plasma falls by 0.4 mmol/L per 0.1 U increase in ECF pH

Inhibition or blockade of K+ conductance Barium causes irreversible blockade of K+ conductance channels, preventing passive efflux of channels and unopposed Na+/K+ - ATPase K+ from cells

Chloroquine toxicity causes a dose-dependent reversible blockade of K+ channelsFamilial hypokalemic periodic paralysis: K+ channels are normal but the muscle behaves as if

channels are blocked

Accelerated loss of potassium (can be classified according to acid-base status) Chloride-responsive metabolic alkalosis Gastrointestinal loss (e.g., vomiting, gastric suctioning, colonic villous adenoma) (chloride depletion causes ECF Renal loss (e.g., loop diuretics, Bartter’s syndrome, thiazide diuretics, Gitelman syndrome) contraction with secondary hyperaldosteronism, leading to renal potassium wasting) Chloride-resistant metabolic alkalosis Primary hyperaldosteronism: high aldosterone, low renin, normal cortisol levels, e.g., adrenal (the primary disturbance is adenoma, bilateral adrenal hyperplasia mineralocorticoid excess Glucocorticoid excess: normal aldosterone, normal renin and high cortisol levels, e.g., Cushing’s resulting in sodium and water syndrome retention, ECF expansion, Syndrome of apparent mineralocorticoid excess: low aldosterone level, low renin level, and hypertension, metabolic normal cortisol, e.g., liquorice, Liddle syndrome alkalosis, and the continuous Congenital adrenal hyperplasia: low aldosterone, low renin, and low cortisol stimulation of potassium secretion in the cortical collecting duct) Hyperchloremic metabolic acidosis Gastrointestinal loss of potassium, e.g., diarrhea, urinary diversion (non anion gap) Renal tubular acidosis Organic metabolic acidosis (high anion gap) E.g., lactic acidosis, ketoacidosis Magnesium deficiency (normal acid-base E.g., alcoholism, malabsorption, drugs: diuretics, gentamicin, cisplatin, amphotericin balance)

13. Clinical Hyperkalemia and Hypokalemia 75

Magnesium Depletion

Concurrent hypomagnesemia coexists in up to40% of cases of hypokalemia because of drugs ordisease processes that cause loss of both ions.Whether hypokalemia is caused by magnesiumdepletion or is an independent effect, it is difficult fito assess. Magnesium is essential in the stabiliza-tion of cellular membranes, and hypomagnesemiais associated with increased losses of intracellularpotassium, and increased fecal and urinary potas-sium wasting. Regardless of the cause, hypokale-mia becomes refractory when hypomagnesemia isalso present, especially with plasma magnesiumconcentrations less than 0.6 mmol/L. In these cir-cumstances, correction of hypokalemia requires initial restoration of magnesium.

Clinical Effects

Hypokalemia alters the function of excitablemembranes, an effect more notable in skeletalmuscle than cardiac muscle. Renal and metaboliceffects are also seen. The likelihood of symptoms seems to correlates with the rapidity of decrease in plasma potassium.

Neuromuscular

Mild Hypokalemia

Mild hypokalemia (plasma potassium, 3.0–3.5 mmol/L) is often asymptomatic. Skeletalmuscle weakness is not usually seen with potas-sium greater than 3.0 mmol/L.

Moderate Hypokalemia

Symptoms of moderate hypokalemia (plasma potassium, 2.5–3.0 mmol/L) are nonspecific: lassi-fitude, fatigue, constipation, weakness in the lower extremities, and myalgia. A paralytic ileus canoccur.

Severe Hypokalemia

Ascending paralysis can occur in severe hypoka-lemia (plasma potassium, ≤2.0–2.5 mmol/L). The lower extremities are typically involved first (par-fiticularly the quadriceps), followed by the trunk

and upper extremities, with eventual impairmentof respiratory muscle function and ventilatory failure. Rhabdomyolysis can be seen with severehypokalemia.

Cardiovascular

Arrhythmias

The arrhythmogenic potential of hypokalemia issignificantly increased in the presence of digoxin,fimyocardial ischemia, congestive cardiac failure (CCF), or left ventricular hypertrophy. Hypokale-mia predisposes to serious ventricular ectopy andis associated with increased frequency of primary ventricular fibrillation and mortality, particularly fiin the setting of ischemic heart disease and acute myocardial infarction.

ECG Changes

• Flattened T wave• U wave (delayed ventricular and Purkinje

repolarization)• Ventricular ectopics• Note: When potassium is at ≤2.0 mmol/L, T and U

waves combine, creating a wave with greater amplitude with risk of reentry circuits arisingduring the prolonged relative refractory period,leading to the development of tachyarrhythmias

Worsening of Cardiac Function

Chronic hypokalemia exacerbates CCF by impair-ing the function and contractility of myocardialmuscle. This reduces stroke volume and cardiacoutput. Cardiac necrosis has been described.

Renal and Metabolic

Ammoniagenesis and Metabolic Alkalosis

Chronic hypokalemia results in ICF acidification, fiwhich stimulates ammoniagenesis in the proximaltubule. Increased ammonia trapping and netacid excretion coupled to enhanced bicarbonateproduction perpetuates metabolic alkalosis.Increased ammoniagenesis may be clinically important in patients with severe liver disease, in whom hypokalemia may precipitate hepaticencephalopathy.

76 H.-Y. Ong

Nephrogenic Diabetes Insipidus

Chronic hypokalemia commonly leads to irrevers-ible tubulointerstitial changes resulting in neph-rogenic diabetes insipidus (DI).

Diagnostic Approach to Hypokalemia

1. History of associated conditions, includingdrug history.

2. Examination: assessment of extracellularvolume and acid-base status.

3. Investigations: a. Urinary potassium. i. Measured on a 24-hour urine collection,

urinary potassium should be less than 10 to 15 mmol/24 hours with hypokalemiaof extrarenal origin (indicating appro-lpriate renal potassium conservation).

ii. Random urinary potassium may begreater than 20 mmol/L in cases of renal potassium wasting, but this cannot be interpreted unambiguously becauseurinary potassium reflects the Kfl + to H2O ratio (and depends mainly on the amountof water reabsorbed).

b. Urinary anion gap.4. Serum magnesium.5. Plasma aldosterone, renin, and cortisol levels.

Treatment

The immediate objective of potassium supple-mentation is to prevent life-threatening cardiacand muscular complications, with the rate and

route of administration depending on the plasmapotassium as well as the presence and serious-ness of the pathophysiological consequences of hypokalemia. The underlying disorder should be treated simultaneously.

Continuous ECG monitoring of cardiac rhythm is indicated for plasma potassium ≤2.5 mmol/L orintravenous infusion rates of at least 10 mmol/h.Maximum rate of intravenous potassium admin-istration should not exceed 20 mmol/h. Largeenteral doses of potassium are difficult to deliverfibecause potassium salts are mucosal irritants, with both tablet and liquid forms reported to cause esophageal and gastrointestinal ulceration.

Suggested Reading

Gennari FJ. Current Concepts: Hypokalemia. N Engl J Med. 1998; 339(7): 451–457.

Glover P. Hypokalaemia. Crit Care Resusc. 1999; 1:239–251.

Halperin M, Kamel KS. Potassium. Lancet 1998; 352:t135–140.

Mahoney BA, Smith WAD, Lo DS, et al. Emergency Inter-ventions for Hyperkalaemia. Cochrane Database Syst Rev. 2006. Issue 1.

Peterson LN, Levi M (Edited by Robert W. Schrier). Renal and Electrolyte disorders: Disorders of Potassium Metabolism. 6th ed. Lippincott, Williamsand Wilkins. 2003.

Rastegar A, Soleimani M. Hypokalaemia and hyperka-laemia. Postgrad Med J. 2001; 77: 759–764.

77

14Clinical Hyponatremia and Hypernatremia

Himangsu Gangopadhyay

Transurethral Resection of the Prostate

(TURP) Syndrome

• Hyponatremia• Cardiovascular changes: hypertension/

hypotension, bradycardia• Neurological changes: agitation, confusion,

fi tting, visual disturbances, fifi xed dilated fipupils

Irrigating fl uid usually containing 1.5% glycineflis absorbed during the procedure. Excess fluid flis absorbed and leads to an increase in total body water and hyponatremia. There is an in-crease in the osmolar gap. With the use of glycine, other features may be seen: hypergly-cinemia, hyperserinemia (a metabolite of glycine), hyperammonemia, metabolic acidosis, and hypocalcemia.

Syndrome of Inappropriate Antidiuretic

Hormone Secretion

Syndrome of inappropriate antidiuretic hormone (ADH) secretion (SIADH) is diagnosed on the fol-lowing features:

• Hypotonic hyponatremia• Urine osmolality greater than plasma osmolal-

ity (inappropriately concentrated urine osmolality)

• Urine sodium greater than 20 mmol/L• Other causes of hyponatremia excluded (e.g.,

cardiac, endocrine)• Normovolemia

Hyponatremia

Hyponatremia reflects an abnormal ratio of flsodium to water and is defi ned as a serum sodium filess than 135 mmol/L. Mild to moderate hypona-tremia (serum sodium 128–134 mmol/L) is seen in20 to 30% of hospital inpatients. Severe hypona-tremia (serum sodium less than 120 mmol/L) is seen in only 2 to 5% of hyponatremic patients.

Classification (Table 14.1)

1. Hyponatremia with decreased sodium andwater (hypotonic dehydration). There is gener-ally loss of sodium and water with inappropri-ate fl uid replacement.fl

2. Hyponatremia with normal body sodium andnormal or increased body water, the patient isusually euvolemic. Plasma osmolality may benormal, low, or high.

3. Hyponatremia with increased sodium and water.

Pseudohyponatremia

Isosmolar hyponatremia (serum osmolality of 275–290 mOsm/kg) is usually caused by severe hyperlipidemia or hyperproteinemia and has been called “pseudohyponatremia.” Measurement of plasma sodium by an ion-selective electrode is not affected by the volume of plasma “solids” and, therefore, will give a true sodium level.

78 H. Gangopadhyay

TABLE 14.1. Causes and classification of hyponatremia

Osmolality Volume status Other

Decreased sodium and water (hypotonic dehydration) Extra renal sodium loss Excessive sweating Low Hypovolemic Urine Na < 20 mmol/L Burns Vomiting Diarrhea Renal sodium loss Osmotic diuretics Low Hypovolemic Urine Na > 20 mmol/L Thiazide and loop diuretics Mineralocorticoid deficiency Salt-losing nephropathy Diuretic phase of acute tubular necrosis Renal tubular acidosis

Normal body sodium and normal or increased body water High plasma osmolality Hyperglycemia High Normal or hypovolemic Mannitol Glycerol Glycine Sorbitol Normal plasma osmolality Hyperlipidemia Normal Normal Hyperproteinemia (pseudohyponatremia) Low plasma osmolality Stress (physical, psychological) Low Normal Urine Na > 20 mmol/L Antidiuretic drugs Glucocorticoid deficiency Myxedema SIADH Transurethral resection of prostate Acute/chronic renal failure Severe polydipsia Low Normal Urine Na < 20 mmol/L

Increased sodium and water Edematous states Cardiac failure Low Edematous Liver failure Renal failure Nephrotic syndrome

TABLE 14.2. Causes of SIADH

Ectopic ADH production Small cell bronchial carcinomaPancreatic adenocarcinomaLeukemiaLymphomaThymoma

Central nervous system disorders TraumaBrain tumorCentral nervous system infection (meningitis, encephalitis, abscess)Subarachnoid hemorrhageAcute intermittent porphyriaGuillain-Barré syndromeSystemic lupus erythematous

Pulmonary diseases PneumoniaTuberculosis

Lung abscess

14. Clinical Hyponatremia and Hypernatremia 79

Treatment is with fl uid restriction and treat-flment of the underlying cause if possible. See Table 14.2 for a list of causes.

Clinical Presentation

Clinical presentation depends on the rapidity andseverity of the development of the hyponatremiaas well as the underlying condition. Mild hyponatremia may be symptomless or presentwith anorexia, nausea, lethargy, and apathy. Signsand symptoms of severe hyponatremia includedisorientation, agitation, seizures, depressed refl exes, focal neurological signs, and, eventually,flCheyne-Stokes respiration.

Diagnosis

Diagnosis of the cause of hyponatremia involves:

1. History: a. Medical history: cardiac disease, renal disease,

hepatic disease, endocrine disorders b. Current and recent medication: diuretics,

antidepressants2. Physical examination: volume status, evidence

of cardiac, renal, liver failure3. Investigations: a. Serum osmolality b. Urine sodium and osmolality c. Blood glucose d. Serum lipids e. Thyroid function f. Plasma proteins

Treatment

Much has been mentioned regarding treatment of hyponatremia and the potential complication of central pontine myelinolysis (CPM). This is a potentially fatal condition that can occur whenthe serum sodium is corrected too quickly. CPM can lead to mutism, dysphasia, spastic quadriple-gia, pseudobulbar palsy, delirium coma, and evendeath. Risk factors for its development are alco-holism, burns, use of thiazide diuretics, a mal-nourished state, and hypokalemia.

Asymptomatic Patients

It is important to fully assess the patient and treat the underlying cause. The patient’s fluid status flshould be assessed; hypovolemia should be treatedwith normal saline, whereas hypervolemic states(such as congestive cardiac failure) will require salt and water restriction. With patients who are euvolemic and asymptomatic, fluid restriction flmay be adequate.

Symptomatic Patients

If the onset is acute and the patient is symptomatic(e.g., fi tting, altered conscious level) then aggres-fisive correction of sodium is required. This can be performed by the cautious infusion of 3% salineat a rate of 1 mL/kg/h until the sodium rises to 120 to 125 mmol/L. The patient should be closely mon-itored during this period, and the presence of seizure or coma may require airway protection, anticonvulsants, and other supportive care beforethe serum sodium is corrected appropriately. Serum sodium should not be allowed to rise morethan 2 mmol/L/h and not more than 25 mmol in 48 hours. Frequent checking of serum sodium is,therefore, important. In patients with chronichyponatremia, a more gradual rise of serumsodium is appropriate.

Hypernatremia

Hypernatremia is defined as serum sodium greater fithan 145 mmol/L. It is always associated with a hyperosmolar state.

Classification (Table 14.3)

1. Decreased total body sodium (extracellularwater and sodium loss with excess waterloss)

2. Normal total body sodium (extracellular waterdeficiency associated with minimal sodiumfiloss)

3. Increased total body sodium (extracellular water and sodium gain with relatively excess sodium gain)

80 H. Gangopadhyay

Mechanism

The defence mechanisms against hypernatremiaare ADH release and thirst. Any clinical conditionaffecting these mechanisms will lead to hyperna-tremia if not addressed properly.

Clinical Presentation

Clinical signs and symptoms are not seen untilserum sodium reaches more than 155 mmol/L. Symptoms include fever, restlessness, irritability,drowsiness, lethargy, confusion, and coma. Con-vulsions are rarely seen.

Evaluation of Hypernatremia

Diagnosis of the cause of hypernatremiainvolves:

1. History a. Fluid loss/excessive water loss b. Poor fl uid intakefl c. Increased salt intake d. Intravenous sodium bicarbonate or hyper-

tonic saline therapy e. Clinical scenario that can lead to neuro-

genic/nephrogenic diabetes insipidus2. Physical examination: assessment of volume

status is most important3. Investigations: a. Serum osmolality b. Urine sodium and osmolality

Interpretation of Urinary Osmolality

A serum osmolality greater than 290 mOsm/kg should be accompanied by highly concentrated urine if the concentrating mechanism of the tubules is intact and ADH release is adequate. Aurinary osmolality of less than 200 mOsm/kg usually suggests some form of diabetes insipidus.

Treatment

Treating underlying cause is very important. Fluid therapy with pure water orally or nasogastrically could be suffi cient in some situations, but intrave-finous therapy with 5% dextrose or pure waterthrough a central vein is often necessary.

A rapid decrease of serum sodium could beequally detrimental, and a decrease of serumsodium by 2 mmol/L/h is acceptable.

TABLE 14.3. Classification and causes of hypernatremia

Decreased total body sodium (water greater than sodium loss) Extra renal loss Vomiting Diarrhea Excessive sweating Dialysis Renal loss Osmotic diuretics (e.g., glucose, urea, mannitol)

Normal total body sodium (water loss) Extra renal Unconscious state Thirst center dysfunction Mechanical obstruction Inappropriate intravenous therapy No access to water Renal loss Cranial diabetes insipidus Nephrogenic diabetes insipidus

Increased total body sodium (sodium greater than water gain) High sodium intake Sea water ingestion Accidental/intentional salt ingestion Hypertonic saline Sodium bicarbonate infusion Low sodium output Mineralocorticoid excess

81

15Clinical Metabolic Acidosis and Alkalosis

Sara Blakeley

if one or more of these three variables changes, rather than bicarbonate itself being the main deter-minant of change, as in the traditional approach.

“Strong ions” are ions that exist completely dis-sociated (charged) in water. There are more strong cations than anions and this difference is called the SID. SID has an electrochemical effect on waterdissociation; as SID becomes more negative, H+

increases and pH falls.“Weak ions” exist as dissociated (A−) or associ-

ated forms (HA), and together they make up ATOT. Nonvolatile weak acids, mainly albumin but alsoinorganic phosphate and bicarbonate, dissociatein body fl uids forming weak anions.fl

To maintain electrical neutrality, the SID must be balanced by an equal and opposite charge, the effective SID (SIDe). The SIDe is made up of predominately weak acids (A−) and CO2 (Equation 15.1). If the SID is calculated from the measured dplasma strong ion concentrations (Equation 15.2),it is termed the apparent SID (SIDa). However, this will not be precise if there are unmeasured anions or cations present. If the SIDe and the SIDa do not match (strong ion gap [SIG]) then unmeasuredions must be present. SIG should be zero, although,tin practice, there is a degree of normal variability. If SIG is positive, there must be more unmeasuredstrong anions (e.g., ketones, sulfates), and if it is negative, more unmeasured cations.

Equation 15.1:SIDe = A− + HCO3

−

Equation 15.2: SIDa = ([Na+] + [K+] + [Ca2+] + [Mg2+]) − ([Cl−] +

[lactate])

Acid-base disturbances are common on the inten-sive care unit (ICU) and should always prompt a search for the underlying cause (Tables 15.1–15.3).Recently another, although not all that new (1), way of understanding the mechanism behindacid-base disorders has gained popularity. Thischapter starts with just a brief discussion of this approach, because there are several excellent over-views in the literature (2–6).

Traditional Versus Physiochemical

Approach to Acid-Base Disturbances

Most “traditional” views of acid-base balance rely on the Henderson-Hesselbach equation, but therehas always been debate regarding quantification fiof the metabolic component. It can be describedusing plasma bicarbonate and anion gap (AG) orby using standard base excess (SBE). However, SBE is affected by changes in sodium, chloride, and albumin, and a significant number of abnor-fimal AGs will be missed unless the effect of albumin is considered (7–9). The third way of quantifying the metabolic component is the strong ion differ-ence (SID) (1), taking into consideration nonbi-carbonate buffers, such as inorganic phosphateand serum albumin, both of which may be abnor-mal in critically ill patients.

Stewart’s (physiochemical) approach relies on the principles of physical chemistry; electroneu-trality, conservation of mass, and dissociation of electrolytes (1). It considers three independentvariables for pH: the SID, pCO2 and total weak acidconcentration (ATOT). H+ or HCO3

− can only change

82 S. Blakeley

TABLE 15.1. Classification of metabolic acidosis according to AG

High AG (unmeasured anions) Lactate Ketones Diabetic, alcoholic, or starvation ketoacidosis Renal failure Ingested toxins E.g., salicylate, paracetamol, formaldehyde, ethylene glycol, methanol

Non-AG (hyperchloremic) Gastrointestinal bicarbonate loss E.g., diarrhea, ureterosigmoidostomy, jejunal/ileal loop, small bowel fistulae, pancreatic fistulae, biliary fistulae Renal bicarbonate loss Renal tubular acidosis Exogenous acid load Loss of potential bicarbonate E.g., ketosis with ketone excretion Cation exchange resins

Low AG Hypoalbuminemia Increased concentration of unmeasured cations E.g., magnesium, calcium, lithium Accumulation of positively charged compounds E.g., cationic antibiotics, IgG myeloma normally absent from serum Interference with measurement E.g., hyperlipidemia

AG = ([Na+] + [K+]) + ([Cl−] + [HCO3−]) (Normal, 10–18 mmol/L)

AGadjusted = AGobserved + 0.25 × ([normal albumin in g/L] − [observed albumin in g/L])

TABLE 15.2. Pathological classification of causes of metabolic alkalosis

Exogenous or endogenous bicarbonate load Acute alkali administration Milk alkali syndrome Recovery from lactic acidosis or ketoacidosis Caused by excess regeneration of HCO3

Massive blood transfusion Caused by metabolism of citrate

Chloride depletion Vomiting, high gastric drainage Gastrocystoplasty Diuretics Posthypercapnia Diarrhea secondary to villous adenoma Villous adenomas are usually associated with hyperchloremic metabolic acidosis, but some can secrete chloride Cystic fibrosis

Potassium deficiency Primary hyperaldosteronism Caused by adenoma, carcinoma; idiopathic or hyperplasia Secondary hyperaldosteronism E.g., Cushing syndrome/disease, exogenous corticosteroids, accelerated hypertension, renal artery stenosis Drugs E.g., carbenoxolone, liquorice Primary hyperaldosteronism Caused by adenoma, carcinoma; idiopathic or hyperplasia Secondary hyperaldosteronism E.g., Cushing syndrome/disease, exogenous corticosteroids, accelerated hypertension, renal artery stenosis Bartter’s syndrome Gitelman syndrome Magnesium or potassium deficiency High-dose intravenous penicillins Caused by distal delivery of nonreabsorbable anions with an absorbable cation such as sodium

Other Severe hypoalbuminemia E.g., nephrotic syndrome

15. Clinical Metabolic Acidosis and Alkalosis 83

Equation 15.3:SIDa − SIDe = SIG (equals zero unless

unmeasured anions or cationspresent)

By using this approach, a change in the balance between strong anions and cations accounts for acid-base disturbances seen on the ICU. For example, excess chloride from a high-volume saline infusion reduces the SID, causing an acido-sis. A loss of chloride from a large nasogastricoutput increases the SID, causing an alkalosis.

The modern debate now centers on which is thebest and most practical method of assessing acid-base disorders in critically ill patients, and, in par-ticular, detecting the presence of unmeasured anions:the SIG. To date, neither the SIG, corrected AG, or corrected SBE has been found to be overwhelmingly superior to the other, but all have been found to bepredictors of increased mortality (10, 11).

Lactic Acidosis (Table 15.4)

When more lactate is produced than is metabo-lized, hyperlactatemia occurs. Elevated levels aregenerally taken to indicate anaerobic metabolismcaused by generalized or regional tissue hypoxia(12). However, it is important to remember that

lactate can be produced under aerobic conditions,especially in the setting of infl ammatory pro-flcesses. In sepsis, infl ammatory mediators can flaccelerate aerobic glycolysis, leading to increasedglucose turnover and increased lactate production(12, 13). Specifi c organs are also capable of pro-fiducing excess lactate under times of stress, for example, the lungs and gut (12, 13).

Elevated lactate levels may also indicate decreased clearance (12). Lactate is mostly metab-olized by the liver (50%), and decreased clearance can occur with impaired liver function or decreased liver blood fl ow.fl

Unless lactate levels are specifi cally measured, fian elevated level may be “missed” by the AG or SBE because of factors such as hypoalbuminemia.In addition, hyperlactatemia is not always accom-panied by an acidosis.

Elevated lactate levels, particularly those thatfail to clear during the first 24 hours, have been firepeatedly shown to correlate with a poor outcome(11, 14, 15).

Use of Sodium Bicarbonate

Controversy exists regarding the use of alkali therapy (e.g., sodium bicarbonate) to treat meta-bolic acidosis (16). No clinical studies have shown

TABLE 15.3. Clinical effects of metabolic acidosis and metabolic alkalosis

Metabolic acidosis Metabolic alkalosis

Cardiac • Reduced myocardial contractility and cardiac failure • Decreased myocardial perfusion caused by arteriolar (but unclear association in clinical practice) constriction• Resistance to catecholamines • Reduced angina threshold• Increased risk of arrhythmias • Increased risk of arrhythmias

Respiratory • Stimulation of the carotid body and aortic • Depressed ventilation (causing compensatory respiratory chemoreceptors causing hyperventilation acidosis), which could lead to failure to wean a patient (Kussmaul respiration) from mechanical ventilation

Neurological • Impaired consciousness (mechanism not fully • Decreased cerebral blood flow caused by arteriolaryunderstood and not clearly correlating with the constrictiondegree of acidosis) • Headache, confusion, mental obtundation

• Intracerebral vasodilation (may be clinically • Neuromuscular excitability: seizures and tetany (y related tosignificant in the setting of already raisedd reduction in ionized calcium level)llintracranial pressure)

Metabolic and immune • Hyperkalemia (extracellular potassium shifts) • Hypokalemia• Glucose intolerance (inhibition of glycolysis and • Hypomagnesemiad

hepatic gluconeogenesis)• Immune activation • Hypophosphatemia • Stimulation of aerobic glycolysis leading production of lactic acid and ketoacids • Decreased plasma ionized calcium concentration

Note: the effects of severe alkalemia (pH > 7.60) appear more pronounced with respiratory rather than metabolic alkalosis.

84 S. Blakeley

a benefi t on outcome in patients with lactic acido-fisis, but it is diffi cult to separate the effects of thefiacidosis from the effects of the underlying patho-logical condition.

Sodium bicarbonate is not without risks, including:

• Acute fl uid overload.fl• Alkaline “overshoot” caused by subsequent

metabolism of organic salts increasing the bicar-bonate concentration.

• Paradoxical intracellular acidosis. CO2 releaseddiffuses more quickly than bicarbonate ions intocells, therefore, initially lowering the intracellu-lar pH. There is debate regarding whether this isclinically significant.fi

Use of Renal Replacement Therapy

In acute renal failure, the daily production of mineral acids is adequately matched by the effi-fi

TABLE 15.4. Classification of lactic acidosis

Type A (hypoxic) Circulatory insufficiency (e.g., shock, cardiac arrest) Regional hypoperfusion Carbon monoxide and cyanide poisoning (caused by mitochondrial enzyme inhibition) Severe hypoxia Severe exercise/prolonged seizures (caused by increased muscle activity) Severe anemia

Type B (nonhypoxic) Sepsis in general (caused by accelerated aerobic glycolysis and mitochondrial dysfunction) and with specific infections (e.g., cholera, falciparum

malaria, AIDS)Thiamine deficiency (thiamine is a cofactor of pyruvate dehydrogenase) Thiamine deficiency (thiamine is a cofactor of pyruvate dehydrogenase)

Fulminant hepatic failure, severe liver disease Alcoholism (ethanol oxidation increases the conversion of pyruvate to lactate and inhibits of pathways of pyruvate metabolism) Severe renal failure Malignancy, leukemia, and lymphoma Short bowel syndrome (D lactic acidosis) Diabetic ketoacidosis (ketones may inhibit hepatic lactate uptake)

Type B: drug induced Phenformin/metformin Ethanol, methanol, ethylene glycol Salicylate poisoning Paracetamol poisoning Intravenous fructose, sorbitol Cyanide

β-agonists (e.g., salbutamol, adrenaline, caused by increased glyconeogenesis, glycogenolysis, lipolysis, and cyclic AMP activity)

Type B: inborn errors of metabolism Glucose 6 phosphatase deficiency Fructose 1,6 diphosphatase deficiency Deficiency of enzymes of oxidative phosphorylation

cacy of renal replacement therapy (RRT) tech-niques, therefore, the acidosis can be clearedwithin 48 hours (16). Despite the fact that lactatehas a low molecular weight and it is easily removedby continuous therapies, RRT has been shown to contribute to only less than 3% of total lactateclearance (17). It may be that RRT plays a role in improving the overall clinical situation (e.g., hemodynamic and fluid status), therefore, pro-flmoting the clearance of lactate, rather than actu-ally removing it.

Metabolic Alkalosis

Metabolic alkalosis is common, but carries anassociated mortality that increases with increas-ing pH, especially pH higher than 7.55 (18). Thecause, rather than the alkalosis itself, may be the reason for the increased mortality.

15. Clinical Metabolic Acidosis and Alkalosis 85

Metabolic alkalosis can be classified clinically fi(Table 15.2) or according to the underlyingpathophysiology: initiating process or perpetuat-ing processes (18).

Initiating Processes

1. Loss of hydrogen ions through the kidney (e.g., loop diuretics, primary hyperaldosteronism) or gastrointestinal tract (e.g., vomiting, nasogas-tric suctioning).

2. Accumulation of alkali from exogenous (e.g., sodium bicarbonate in the setting of renal insuf-ficiency) or endogenous sources (e.g., metabolismfiof ketones after diabetic ketoacidosis).

The kidneys have a large capacity to excrete excess bicarbonate and are mostly able to correct the alkalosis; however, in certain situations, the kidney will retain rather than excrete alkali, there-fore, perpetuating the alkalosis. The persistence of ga metabolic alkalosis suggests there is an ongoing,untreated process.

Perpetuating Factors

1. Hypovolemia. Decreased renal perfusion stimu-lates the renin-angiotensin-aldosterone system (RAAS), therefore, increasing sodium absorp-tion and hydrogen ion secretion.

2. Chloride depletion. Stimulation of the RAASleads to increased bicarbonate reabsorption inthe absence of adequate chloride.

3. Severe hypokalemia. Caused by an intracellu-lar shift of hydrogen ions in exchange for potassium.

4. Reduced glomerular filtration ratefi (GFR). Reduction in GFR causes a decrease in filtered fibicarbonate.

Treatment

1. Correct the primary cause if possible, e.g., stop or reduce diuretics.

2. Address perpetuating factors. Most forms of metabolic alkalosis are chloride responsive, andisotonic sodium chloride will correct both chlo-ride and volume depletion. As the chloride deficit fiis corrected, there will be an alkaline diuresis andthe plasma bicarbonate will start to normalize.

3. Consider specific treatments, e.g., surgery fifor an adrenal adenoma or spironolactone forhyperaldosteronism.

4. Other: a. Hydrochloric acid: rarely used. b. Acetazolamide: increases bicarbonate loss,

but only if GFR is adequate. c. RRT: in patients with severe cardiac or renal

disease, RRT may assist volume and electro-lyte correction. Standard replacement fluids flwill need modifi cation as they contain bicar-fibonate or its precursor lactate.

References

1. Stewart PA. Modern quantitative acid-base chemis-try. Can J Pharmacol. 1983; 61: 1444–1461.

2. Story DA. Bench-to-bedside review: A brief history of clinical acid-base. Crit Care. 2004; 8: 253–258.

3. Kellum JA. Determinants of blood pH in health and disease. Crit Care. 2000; 4: 6–14.

4. Kaplan LJ, Frangos S. Clinical review: Acid-base abnormalities in the intensive care unit. Crit Care. 2005; 9: 198–203.

5. Morgan TJ. Clinical review: The meaning of acid-base abnormalities in the intensive care unit—effects of fluid administration. fl Crit Care. 2005; 9: 204–211.

6. Morgan TJ. What exactly is the strong ion gap, anddoes anybody care? Crit Care Resusc. 2004; 6: 155–166.

7. Figge J, Jabor A, Kazda A, Fencl V. Anion gap and hypoproteinaemia. Crit Care Med. 1998; 26: 1807–1810.

8. Fencl V, Jabor A, Kazda A, Figge J. Diagnosis of metabolic acid-base disturbances in critically illpatients. Am J Resp Crit Care Med. 2000; 162: 2246–2251.

9. Story DA, Poustie S, Bellomo R. Estimating unmea-sured anions in critically ill patients: anion gap,base deficit and strong ion gap.fi Anaesthesia. 2002; 57: 1102–1133.

10. Kaplan LJ, Kellum JA. Initial pH, base deficit, lactate, fianion gap, strong ion difference, and strong ion gappredict outcome from major vascular injury. Crit Care Med. 2004; 32: 1120–1124.

11. Smith I, Kumar P, Molloy S, et al. Base excess andlactate as prognostic indicators for patients ad-mitted to intensive care. Intensive Care Med. 2001; 27: 74–83.

12. De Backer D. Lactic acidosis. Intensive Care Med. 2003; 29: 699–702.

86 S. Blakeley

13. Bellomo R, Ronco C. The pathogenesis of lactic aci-dosis in sepsis. Curr Opin Crit Care. 1999; 5: 452–457.

14. Stacpoole PW, Wright EC, Baumgartner TG, et al. for the DCA-Lactic acidosis Study Group: Natural history and course of acquired lactic acidosis inadults. Am J Med. 1994; 97: 47–54.

15. Husain FA, Martin MJ, Mullenix PS, et al. Serum lactate and base defi cit as predictors of mortality fiand morbidity. Am J Surg. 2003; 185(5): 485–491.

16. Levraut J, Grimaud D. Treatment of metabolic acidosis. Curr Opin Crit Care. 2003; 9: 260–265.

17. Levraut J, Ciebiera JP, Jambou P, et al. Effect of con-tinuous venovenous hemofiltration with dialysis onfilactate clearance in critically ill patients. Crit Care Med. 1997; 25(1): 58–62.

18. Galla JH. Metabolic alkalosis. J Am Soc Nephrol. 2000; 11: 369–375.

87

AN-Acetylcysteine, 17, 28, 36NNAcid-base disturbances. See also

Acidosis; Alkalosistraditional versus

physiochemical approachto, 81–84

Acidosis, 22, 39lactic, 83–84metabolic, 60, 82, 83, 84

Acute kidney injury (AKI), 19–25causes of, 20–22

investigation of, 23, 24complications of, 22–23defi nition of, 19fiincidence and outcome of, 19–20

Acute tubular necrosis (ATN), 3, 20,22

casts in, 6, 23imaging of, 7–9as oliguria cause, 27as renal failure cause, 7, 21, 26,

27–28urea-to-creatinine ratio in, 2–3

Adenosine agonists, 28Alcohol, as rhabdomyolysis cause,

39Alkalosis, metabolic, 74, 75, 82,

84–85Allopurinol, 15Aminoglycosides, nephrotoxicity

of, 16, 33Ammoniagenesis, 75Anaphylaxis, dialysis-related, 67Anemia, acute kidney injury-

related, 22, 24Anesthesia, effect on renal

function, 33

Aneurysms, abdominal, 22, 35Angiotensin, 2Angiotensin-converting enzyme

inhibitors (ACEIs), 14, 16, 33

Angiotensin receptor blockers,nephrotoxicity of, 16

Anion gap, 81, 82, 83Anticoagulation, in renal

replacement therapy, 61–63Antidiuretic hormone (ADH), 4,

80Anuria, 4, 19, 27Aprotinin, as acute renal failure

risk factor, 34Aspartate transaminase, 38Azathioprine, 68Azotemia, 42

BBacteriuria, 5Bicarbonate, 40, 51, 60, 73, 83–84,

84, 85Biopsy, renal

for acute renal failure diagnosis, 27

“poor man’s,” 3for vasculitis diagnosis, 43

Blood fl ow, renalfleffect of prostaglandins on, 14regulation of, 33in rhabdomyolysis, 38, 39–40

CCalcifi cation, heterotopic, 39fiCalcium. See also Hypercalcemia;

Hypocalcemiaas hyperkalemia treatment, 73

Calcium channel antagonists, 17Calculi, ureteric, imaging studies

of, 9–10Cardiac arrest, hyperkalemia-

related, 71, 73Cardiac output, 29Cardiac surgery, as acute renal

failure risk factor, 34–35Casts, 6, 23, 26, 39Cation exchange resins, 73Central pontine myelinolysis, 79Chemical testing, of urine, 4–5Chloride depletion, 82, 83Coagulopathy, therapeutic plasma

exchange-related, 50Cockroft and Gault formula, for

creatinine clearance, 3Compartment syndromes, 20, 34,

38–39, 41Complement, 44, 46Compression injuries, as

rhabdomyolysis cause, 38Computed tomography (CT), renal,

7, 9–11Contrast agents, nephrotoxicity of,

10, 11, 13, 16–17, 27–28, 36Corticosteroids, 68Creatine kinase, 38Creatinine, 2–3

as acute kidney injury marker, 19

renal failure-related increase in, 24

Creatinine clearance, 3Crush injuries, as rhabdomyolysis

cause, 38Cyclooxygenase inhibitors, 15Cyclosporin, 68

Index

88 Index

DDaclizumab, 69Dextrose, 17, 73Diabetes insipidus, nephrogenic, 76Diabetes mellitus, 6, 33, 34Dialysis

optimal dose in, 65–67peritoneal, 64–67principles of, 64as renal vasculitis treatment, 44for urinary myoglobin

reduction, 39–40Dialysis equilibrium syndrome, 67Diethylene triamine pentaacetic

acid (DTPA) studies, 7, 10Digoxin, toxicity of, 731,25-Dihydroxyvitamin D3, 1Disseminated intravascular

coagulation, 46–47Diuretics

as acute renal failure prophylaxis, 34, 36

as acute renal failure treatment,36

loop, 15, 17, 27–28nephrotoxicity of, 16

Dopamine, as renal failure prophylaxis, 36

Dopamine agonists, 28Drugs. See also names of specific fi

drugsas hypokalemia cause, 74impaired renal clearance of, 23–24nephrotoxicity of, 14–18, 23,

33–34, 38, 39as rhabdomyolysis cause, 38, 39

EEdema

pulmonary, 43renal, 9

Electrocardiographyin hyperkalemia, 71–72in hypokalemia, 75

Electrolyte disturbances. See alsospecific electrolyte fidisturbances

acute kidney injury-related, 22Encephalopathy, uremic, 22Enteral support, in critically ill

patients, 29Eosinophilia, 24Erythropoietin, 1, 22Excretion, urinary, 1

FFamilial hypokalemic periodic

paralysis, 74Fasciotomy, 38Fluid composition, of the body, 1,

2Fluid resuscitation, in acute renal

failure patients, 28–29

GGlomerular filtration, cessation of, fi

33Glomerular filtration rate (GFR), 2fi

in acute kidney injury, 19assessment of, 3in metabolic alkalosis, 85

Glomerulonephritis, 6, 9, 26Glucose control, tight, 29, 36Glycosuria, renal, 5Goodpasture’s syndrome, 9

HHematuria, 4, 26, 42Hemodiafiltration, 64fi

continuous venovenous, 59Hemodialysis, 64

complications of, 67continuous, 57–63as hyperkalemia treatment, 74intermittent, 52, 54, 60

Hemofi lters, 50fiHemofi ltration, 52, 64fi

continuous venovenous, 52, 57,59

high-volume, 53prophylactic, 29for urinary myoglobin

reduction, 39–40Hemofi ltration machines, 58, fi

59–60Hemolytic uremic syndrome, 46,

47Hemorrhage, gastrointestinal, 23Henoch-Schönlein purpura, 43Heparin, 61–62Homeostasis, role of the kidney in,

1Hypercalcemia, 24Hyperglycemia, 22Hyperkalemia, 16, 22, 71–73Hypernatremia, 79–80Hyperphosphatemia, 22, 24, 39Hypertension, 13, 42

malignant, 47

Hypocalcemia, 39, 50Hypokalemia, 50, 74–76, 82, 85,

8575, 73076Hypomagnesemia, 75Hyponatremia, 22, 77–79, 79Hypoperfusion, renal, 9, 20.6Hypotension

acute renal failure-related, 26, 29anesthesia-related, 33dialysis-related, 67as postoperative renal failure

cause, 33prerenal failure-related, 20relative, 36in surgical patients, 33therapeutic plasma exchange-

related, 50Hypovolemia, 21, 49, 85

IImaging, of acute renal failure,

7–13. See also Computed tomography (CT); Magnetic resonance angiography (MRA); Magnetic resonance imaging (MRI); Scintigraphic imaging

diagnostic algorithm for, 12Immunosuppression, renal failure-

related, 22Immunosuppressive therapy, in

renal transplant recipients, 68–69

Infectionin renal transplant recipients, 69as rhabdomyolysis cause, 38

Inotrope therapy, for acute renalfailure, 29

Insulin therapy, 36, 73Intensive care unit (ICU) patients

acute kidney injury in, 19acute renal failure in, 26intrinsic renal failure in, 20, 21obstructive renal failure in, 22

Inulin clearance, 3Ischemia, renal, 33

KKidney

hydronephrotic, 9normal, ultrasound imaging of,

7, 8normal functions of, 1polycystic, 7

Index 89

LLactate, 60, 82, 83Lactate dehydrogenase, 38Left ventricular end diastolic

pressure (LVEDP), 29

MMagnetic resonance angiography

(MRA), renal, 10–11Magnetic resonance imaging

(MRI), renal, 7Mannitol, 28, 34Mean arterial pressure (MAP), 29,

33Mercaptoacetyltriglycerine (MAG3)

studies, 7, 8, 9, 10, 11Microangiopathy, thrombotic,

46–47Mineral disturbances, acute kidney

injury-related, 22Mycophenolate mofetil, 69Myoglobin, in urine, 39–40

NNatriuretic peptides, 28, 36Nephritic syndrome, 42–43Nephritis

interstitial, 6, 9, 15, 33–34systemic lupus erythromatosus-

related, 45Nephrotic syndrome, 5, 11, 16Neutrophilia, 24Nonsteroidal anti-infl ammatory fl

drugs, nephrotoxicity of,14–16, 33

Nutrition/nutritional support, 29,67

OOliguria, 4, 27, 34Oxygen consumption, renal, 33

PPeritoneal dialysis, 64–67Plasma exchange, therapeutic,

49–50, 59Plasma volume, calculation of, 49Polyclonal antibodies, 69Polyuria, 4Postrenal failure, 7Potassium. See also Hyperkalemia;

Hypokalemiarenal excretion of, 71, 73

Potassium supplementation, 76

Pregnancy, in systemic lupus erythromatosus patients, 46

Prerenal failure, 7, 21as acute kidney injury cause, 20defi nition of, 20fidifferentiated from intrinsic

renal failure, 23, 24Prostacyclin, 63Prostaglandins, renal, 14–15Proteinuria, 4–5Pseudohyperkalemia, 72–73Pseudohyponatremia, 77Purpura, thrombotic

thrombocytopenic, 46, 47

RRed blood cells, in urine, 6Renal artery, dissection or

occlusion of, 10–11, 13, 16Renal disease, end-stage, renal

replacement therapy for,64–70

Renal dysfunction. See also Acute kidney injury (AKI)

RIFLE classifi cation system for,fi19, 20

Renal failureacute

acute kidney injury-related, 19diagnosis of, 26, 35–36differentiated from chronic

renal failure, 7drug-related, 14–18imaging of, 7–13laboratory investigations of,

26–27medical management of,

26–32multisystem causes of, 42–48pathophysiology of, 33prevention of, 27–30, 36–37rhabdomyolysis-related, 38risk factors for, 26, 27, 33–34supportive strategies for, 28–29in surgical patients, 33–37treatment of, 35–36

chronicdifferentiated from acute renal

failure, 7imaging of, 8

intrinsic, 20–21differentiated from prerenal

failure, 23, 24postrenal or obstructive, 22

Renal functionabrupt and sustained decrease

in. See Acute kidney injury (AKI)

assessment of, 1–6, 26Renal index (RI), 9Renal injury. See also Acute kidney

injury (AKI)drug-induced, 14–18

Renal replacement therapy, 28, 51–56. See also Dialysis; Hemodialysis; Renaltransplantation

for acute kidney injury, 20anticoagulation in, 62basics of, 51–52continuous, 51, 52, 53, 54, 57–59,

61for hyperkalemia, 74

effect on lactic acidosis, 84for end-stage renal disease, 64–70hemofi lters in, 60fihybrid, 52indications for, 53–54intermittent, 51, 53, 54replacement and dialysis fluid fl

in, 59–60selection of dosage and mode in,

54–55technical aspects of, 57–63vascular access in, 52, 60–61for volume overload, 22

Renal transplantation, 67–69Renal vein, thrombosis of, 11, 13Renin, 1, 2Respiratory failure, 43Resuscitation, in surgical patients,

36Rhabdomyolysis, 28, 38–41

diagnosis of, 38–39drug-related, 14treatment of, 40

RIFLE system, for renal dysfunction classification,fi19, 20

SSalbutamol, 73Scintigraphic imaging

diethylene triamine pentaaceticacid (DTPA) studies, 7, 10

mercaptoacetyltriglycerine(MAG3) studies, 7, 8, 9, 10, 11

of ureteric obstruction, 10

90 Index

Sepsisdialysis-related, 67treatment of, 53

Sodium bicarbonate, 40, 73, 84Statins, as rhabdomyolysis cause,

39Strong ion difference (SID), 81, 83Surgical patients

acute kidney injury in, 20acute renal failure in, 33–37

Syndrome of inappropriate antidiuretic hormone secretion (SIADH), 77, 78, 79

Systemic lupus erythromatosus, 45–46

TTacrolimus, 68Tamm-Horsfall protein, 40Thrombocytopenia, 24Toxins

as rhabdomyolysis cause, 38uremic, retention of, 22

Transurethral resection of the prostate (TURP) syndrome, 77

Tubulointerstitial disease, 26Tumor lysis syndrome, 14

UUltrafi ltration, slow continuous, 53,fi

59Ultrasound, renal, 7, 23, 27

of normal kidneys, 7, 8

of renal vein thrombosis, 11of ureteric obstruction, 9

Ureaaccumulation of, 22, 24measurement of, 2

Urea-to-creatinien ratio, 2–3Uremia, 22–23Uremic patients, cardiovascular

risk factors in, 65Ureteric obstruction, imaging

studies of, 9–10Urinalysis, 3–6

for acute kidney injury evaluation, 23, 24

for acute renal failure evaluation,26–27

for glomerulonephritis/acute interstitial nephritisdifferentiation, 9

in surgical patients, 35for systemic lupus

erythromatosus evaluation, 45

Urinealkalinization of, 14, 40, 41appearance of, 3chemical testing of, 4–5microscopy of, 5–6

in rhabdomyolysis, 39, 40osmolality of, 23, 24, 80specifi c gravity of, 3, 24fivolume of, 3–4, 27

Urine output, in acute kidney injury, 19–20

Urolithiasis, 27

Urological surgery, as acute renal failure risk factor, 35

VVascular surgery, as acute renal

failure risk factor, 34Vasculitis

as acute renal failure cause, 26,42–45

clinical presentation of, 42–43etiology of, 42treatment of, 43–44Wegener’s renal, 44

Venography, of renal vein thrombosis, 11

Volume depletionas acute renal failure cause, 26assessment and correction of, 29as postoperative renal failure

cause, 33prerenal failure-related, 21

Volume expansion, intravascular, 40

Volume overload, acute kidney injury-related, 22

WWegener’s granulomatosis, 9Wegener’s renal vasculitis, 44White blood cells, in urine, 5–6

XX-rays

chest, 23, 44renal, 7