Continuous RenalReplacementTherapyEdited byJohn A. KellumProfessor and Vice ChairDepartment of Critical Care MedicineUniversity of PittsburghPittsburgh, PennsylvaniaRinaldo BellomoProfessor of MedicineDirector of Intensive Care ResearchMelbourne UniversityMelbourne, AustraliaClaudio RoncoProfessor of Clinical Nephrology and MedicineDirector, Department of NephrologySt. Bortolo HospitalVicenza, Italy12010
We dedicate this volume to the nursing professionals thatdeliver CRRT—for without their hard work and dedication this therapywould not exist—and to the patients and their families in the hope thatwe can make some difference in their lives.
viiSigniﬁcant advances have occurred in the care of patients with acute kidneyinjury. Continuous renal replacement therapy (CRRT) has become the stan-dard of care for many critically ill patients with severe acute kidney injury, andmost major medical centers have developed the capability of providing CRRT.However, many hospitals lack the capacity, and many that have it underutilize it.Our goal with the CRRT handbook is to provide a concise but authoritativeguide to the use of CRRT. In a single, slim volume, we have covered the basicsto management of acute kidney injury both with and in addition to CRRT. Theintent of this book is to provide a quick reference for both novice and expe-rienced CRRT providers, to enrich existing expertise, and to achieve a betterunderstanding of this powerful treatment. Our ultimate goal is to improve out-comes for patients with acute kidney injury through teamwork and education.John A. KellumRinaldo BellomoClaudio Ronco2009Preface
Abbreviations xiiiContributors xviiPart 1 Theory1 The critically ill patient with acute kidney injuryAditya Uppalapati and John A. Kellum 32 History and rationale for continuous renalreplacement therapyIlona Bobek and Claudio Ronco 113 Terminology and nomenclatureIan Baldwin and Rinaldo Bellomo 194 Basic principles of solute transportZhongping Huang, Jeffrey J. Letteri, Claudio Ronco,and William R. Clark 255 Principles of ﬂuid managementRinaldo Bellomo and Sean M. Bagshaw 336 Indications, timing, and patient selectionJohn A. Kellum 397 Extended indications for continuous renalreplacement therapyRinaldo Bellomo and Ian Baldwin 478 Dose adequacy and assessmentZaccaria Ricci and Claudio Ronco 539 Acid-base and electrolyte disordersJohn A. Kellum 61Part 2 Practice10 Choosing a renal replacement therapyin acute kidney injuryJorge Cerdá and Claudio Ronco 7911 Vascular access for continuous renal replacement therapyAlexander Zarbock and Kai Singbartl 9312 The circuit and the prescriptionRinaldo Bellomo and Ian Baldwin 99Contents
xCONTENTS13 The membrane: size and materialZhongping Huang, Jeffrey J. Letteri, Claudio Ronco,and William R. Clark 10714 Fluids for continuous renal replacement therapyPaul M. Palevsky and John A. Kellum 11515 Alarms and troubleshootingZaccaria Ricci, Ian Baldwin, and Claudio Ronco 12116 Nonanticoagulation strategies to optimize circuitfunction in renal replacement therapyIan Baldwin 12917 AnticoagulationRinaldo Bellomo and Ian Baldwin 13518 Regional citrate anticoagulationNigel Fealy 14119 Drug dosing in continuous renal replacement therapyKimberly A. Maslonek, Kelly A. Killius, and John A. Kellum 147Part 3 Special Situations20 Renal replacement therapy in childrenMichael L. Moritz 15921 Therapeutic plasma exchange in critical care medicineJoseph E. Kiss 16722 MARS: molecular adsorbent recirculating systemNigel Fealy and Rinaldo Bellomo 17523 SorbentsDehua Gong and Claudio Ronco 18124 Hybrid therapiesDinna N. Cruz and Claudio Ronco 189Part 4 Organizational issues25 The ICU environmentYounghoon Kwon 19926 Patient care quality and teamworkKimberly Whiteman and Frederick J. Tasota 20527 Organizational aspects: developing policiesand procedures for continuous renalreplacement therapiesJorge Cerdá 213
xiCONTENTS28 Documentation, billing, and reimbursement forcontinuous renal replacement therapyKevin W. Finkel 22329 Machines for continuous renal replacement therapyClaudio Ronco 22930 Quality assurance for continuous renalreplacement therapiesIan Baldwin and Rinaldo Bellomo 24731 Educational resourcesIan Baldwin and Kimberly Whiteman 253Index 263
xviiContributorsSean M. Bagshaw, MDAssistant ProfessorDivision of Critical Care MedicineUniversity of Alberta HospitalUniversity of AlbertaEdmonton, CanadaIan Baldwin, RNClinical EducatorDepartment of Intensive CareAustin HospitalDepartment of Nursingand Health SciencesRMIT UniversityMelbourne, AustraliaRinaldo Bellomo, MDProfessor of MedicineDirector of Intensive Care ResearchDepartment of Intensive CareMelbourne UniversityMelbourne, AustraliaIlona Bobek, MDNephrologistDepartment of Nephrology, Dialysisand TransplantationSan Bartolo HospitalVicenza, ItalyJorge Cerdá, MD, FACP, FASNClinical Associate Professor ofMedicineAlbany Medical College and CapitalDistrict Renal PhysiciansAlbany, New YorkWilliam R. Clark, MDVice President, Medical Strategy andTherapy DevelopmentGambro Renal ProductsLakewood, ColoradoAssistant Clinical Professorof MedicineNephrology DivisionIndiana University School of MedicineIndianapolis, IndianaDinna N. Cruz, MD, MPHNephrologistDepartment of Nephrology, Dialysisand TransplantationSan Bortolo HospitalInternational Renal Research InstituteVicenzaVicenza, ItalyNigel Fealy, RNClinical Nurse EducatorDepartment of Intensive CareAustin HospitalHeidelberg, AustraliaKevin W. Finkel, MD, FACP,FASNProfessor and DirectorDivision of Renal Diseases andHypertensionUniversity of Texas Medical Schoolat HoustonHouston, Texas
xviiiCONTRIBUTORSDehua Gong, MDAssociate ProfessorResearch Institute of NephrologyJinling HospitalNanjing University School of MedicineNanjing, P.R.ChinaZhongping Huang, PhDAssistant ProfessorDepartment of MechanicalEngineeringWidener UniversityChester, PennsylvaniaJohn A. Kellum, MDProfessor of Critical Care Medicine,Medicine, Bioengineering, and Clinicaland Translational SciencesVice Chair of ResearchDepartment of Critical Care MedicineUniversity of Pittsburgh School ofMedicinePittsburgh, PennsylvaniaKelly A. Killius, PharmD, BCPSClinical Pharmacy Specialist,Emergency MedicineBoston Medical CenterBoston, MassachusettsJoseph E. Kiss, MDAssociate Professor of MedicineDepartment of MedicineDivision of Hematology/OncologyPittsburgh, PennsylvaniaYounghoon Kwon, MDStaff IntensivistHealth East Care SystemSaint Paul, MinnesotaJeffrey J. Letteri, BS, CHTDirector DevelopmentGambro, Inc.Lakewood, ColoradoKimberly A. Maslonek,PharmDClinical PharmacistCardiothoracic and Surgical ICUUniversity of Pittsburgh MedicalCenter PresbyterianPittsburgh, PennsylvaniaMichael L. Moritz, MDAssociate Professor of PediatricsDivision of NephrologyChildrens Hospital of Pittsburghof UPMCPittsburgh, PennsylvaniaPaul M. Palevsky, MDProfessor of MedicineUniversity of Pittsburgh School ofMedicinePittsburgh, PennsylvaniaZaccaria Ricci, MDConsultantIntensive Care UnitDepartment of Pediatric CardiacSurgeryBambino Gesù Children’s HospitalRome, ItalyClaudio Ronco, MDDirectorDepartment of Nephrology, Dialysisand TransplantationSan Bortolo HospitalVicenza, ItalyKai Singbartl, MD, EDICAssistant Professor of Critical CareMedicine and AnesthesiologyDepartment of CriticalCare MedicineUniversity of PittsburghPittsburgh, Pennsylvania
xixCONTRIBUTORSFrederick J. Tasota, RN, MSN,CCRNCritical Care Advanced Practice NurseUniversity of Pittsburgh MedicalCenter PresbyterianPittsburgh, PennsylvaniaAditya Uppalapati, MDResidentInternal MedicineUniversity of Pittsburgh MedicalCenterMcKeesport, PennsylvaniaKimberly Whiteman, RN,MSN, CCRNAdvanced Practice Nurse EducatorUniversity of Pittsburgh MedicalCenter PresbyterianPittsburgh, PennsylvaniaAlexander Zarbock, MDResidentDepartment of Anesthesiology andCritical Care MedicineUniversity of MuensterMuenster, Germany
3Chapter 1The critically ill patient withacute kidney injuryAditya Uppalapati and John A. KellumThe terms acute kidney injury (AKI) and acute renal failure (ARF) are not syn-onymous. While the term renal failure is best reserved for patients who havelost renal function to the point that life can no longer be sustained without in-tervention, AKI is used to describe the milder as well as severe forms of acuterenal dysfunction in patients. Although the analogy is imperfect, the AKI–ARFrelationship can be thought of as being similar to the relationship between acutecoronary syndrome and ischemic heart failure. AKI is intended to describe theentire spectrum of disease from being relatively mild to severe.In contrast, renal failure is deﬁned as renal function inadequate to clear thewaste products of metabolism despite the absence of or correction of hemody-namic or mechanical causes. Clinical manifestations of renal failure (either acuteor chronic) include the following:Uremic symptoms (drowsiness, nausea, hiccough, twitching)•Hyperkalemia•Hyponatremia•Metabolic acidosis•OliguriaPersistent oliguria may be a feature of ARF but nonoliguric renal failure is notuncommon. Patients may continue to make urine despite an inadequate glomer-ular ﬁltration. Although prognosis is often better if urine output is maintained,use of diuretics to promote urine output does not seem to improve outcome(and some studies even suggest harm).ClassiﬁcationInternational consensus criteria for AKI have been purposed. The acronymRIFLE is used to describe three levels of renal impairment (Risk, Injury, Failure)and two clinical outcomes (Loss and End-stage kidney disease), as shown inFigure 1.1.OliguriaClassiﬁcation
CHAPTER1ThecriticallyillpatientwithAKI4Incidence and etiology of acute kidney injuryThe classiﬁcation system includes separate criteria for serum creatinine andurine output. The criteria, which lead to the worst classiﬁcation, deﬁne the stageof AKI. Note that RIFLE-F is present even if the increase in serum creatinine(SCrt) is less than threefold, so long as the new SCrt is <4.0 mg/dL in the settingof an acute increase of at least 0.5 mg/dL. The ﬁgure shows that more patients(high sensitivity) will be included in the mild category, including some withoutactually having renal failure (less speciﬁcity). In contrast, at the bottom, the cri-teria are strict and therefore speciﬁc, but some patients will be missed.Incidence and progressionAcute kidney injury occurs in 35%–65% of ICU admissions and 5%–20% of gen-eral hospital admissions. Mortality rates increase signiﬁcantly with AKI, andmost studies show a—threefold to ﬁvefold increase in the risk of death amongpatients with AKI compared to patients without AKI. Furthermore, increases inseverity of AKI are associated with a stepwise increase in risk of death such thatpatients reaching RIFLE-F are far more likely to die before hospital dischargeIncidence and etiology of acute kidney injuryFigure 1.1 The RIFLE Criteria for diagnosis and staging of AKI—used to describe threelevels of renal impairment (Risk, Injury, Failure) and two clinical outcomes (Loss and End-stagekidney disease). Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal failure—deﬁnition, outcome measures, animal models, ﬂuid therapy and information technologyneeds: the Second International Consensus Conference of the Acute Dialysis Quality Initiative(ADQI) Group. Crit Care. 2004;8:R204-R212. Used with permission.*An alternative proposal is to deﬁne “Risk” to include any increase in serum creatinine of at least 0.3 mg/dL, over 48hours or less even if less than 50% increase.Creatinine criteriaRiskHighsensitivityHighspecificityInjuryFailureLossESRDUrine output criteriaIncreased creatininex 1.5*Persistent ARF = complete loss ofrenal function > 4 weeksEnd stage renal diseaseOliguriaIncreased creatinine x2Increased creatinine x3or creatinine ≥4mg/dL(Acute rise of≥0.5 mg/dL)UO < 0.5mL/kg/hx 6 hUO < 0.5mL/kg/hx 12 hUO < 0.3mL/kg/hx 24 h orAnuria x12 h
CHAPTER1ThecriticallyillpatientwithAKI5compared to patients who do not progress from RIFLE-R or RIFLE-I. Hospitalmortality rates for ICU patients with AKI are approximately as follows: R—9%,I—11%, F—26% compared to 6% for ICU patients without AKI. Unfortunately,more than 50% of patients with RIFLE-R progress to class I (in approximately1–2 days) or F (in approximately 3–4 days), and almost 30% of RIFLE-I progressto F.Risk factors for AKIRisk factors for developing AKI as deﬁned by RIFLE criteria are as follows:Sepsis•Increasing age, especially age > 62 years•Race—Black patients for developing RIFLE-F•Greater severity of illness as per Acute Physiology and Chronic Health•Evaluation (APACHE) III or Sepsis-related Organ Failure Assessment (SOFA)scoresPreexisting chronic kidney disease•Presiding admission to a non-ICU ward in the hospital•Surgical admissions more likely than medical admissions•Cardiovascular disease•Emergent surgeries•Being on mechanical ventilation•Etiology of AKIClinical features may suggest the cause of AKI and dictate further investigation.AKI is common in the critically ill, especially in patients with sepsis and otherforms of systemic inﬂammation (e.g., major surgery, trauma, burns), but othercauses must be considered. In sepsis, the kidney often has a normal histologicalappearance.Volume-responsive AKIIt is estimated that as many as 50% of cases of AKI are “ﬂuid responsive,” and theﬁrst step in managing any case of AKI is to ensure appropriate ﬂuid resuscitation.However, volume overload is a key factor contributing to the mortality attrib-utable to AKI, so ongoing ﬂuid administration to nonﬂuid responsive patientsshould be discouraged. In general, ﬂuid resuscitation should be guided by hemo-dynamic monitoring.Sepsis-induced AKISepsis is a primary cause or contributing factor in more than 50% of cases of AKI,which includes cases severe enough to require renal replacement therapy (RRT).Patients with sepsis, including those outside the ICU, develop AKI at rates ashigh as 40%. Septic shock appears to be an important factor in the developmentof sepsis-induced AKI; however, patients without overt shock do not appear tobe any less likely to develop AKI.
CHAPTER1ThecriticallyillpatientwithAKI6HypotensionHypotension is an important risk factor for AKI, and many patients with AKIhave sustained at least one episode of hypotension. Treating ﬂuid-responsiveAKI with ﬂuid resuscitation is clearly an important step, but many patients willalso require vasoactive therapy (e.g., dopamine, norepinephrine) to maintainarterial blood pressure. Despite a common belief among many practitioners,norepinephrine does not increase the risk of AKI compared to dopamine andrenal blood ﬂow actually increases with norepinephrine in animals with sepsis.Postoperative AKIRisk factors include hypovolemia, hypotension, major abdominal surgery, andsepsis. Surgical procedures (particularly gynecological) may be complicated bydamage to the lower urinary tract with an obstructive nephropathy. Abdominalaortic aneurysm surgery may be associated with renal arterial disruption.Cardiac surgery may be associated with atheroembolism and sustained periodsof reduced arterial pressure as well as systemic inﬂammation.Other causesNephrotoxins—may cause renal failure via direct tubular injury, interstitial•nephritis, or renal tubular obstruction. In patients with AKI, all potential neph-rotoxins should be withdrawn.Rhabdomyolysis—suggested by myoglobinuria and raised creatine kinase in•patients who have suffered a crush injury, coma, or seizures.Glomerular disease—red cell casts, hematuria, proteinuria, and systemic fea-•tures (e.g., hypertension, purpura, arthralgia, vasculitis) are all suggestive ofglomerular disease. Renal biopsy or speciﬁc blood tests (e.g., Goodpasture’ssyndrome, vasculitis) are required to conﬁrm diagnosis and guide appropriatetreatment.Hemolytic uremic syndrome—suggested by hemolysis, uremia, thrombocyto-•penia, and neurological abnormalities.Crystal nephropathy—suggested by the presence of crystals in the urinary•sediment. Microscopic examination of the crystals conﬁrms the diagnosis (e.g.,urate, oxalate). Release of purines and urate are responsible for acute renalfailure in the tumor lysis syndrome.Renovascular disorders—loss of vascular supply may be diagnosed by renog-•raphy. Complete loss of arterial supply may occur in abdominal trauma oraortic disease (particularly dissection). More commonly, the arterial supplyis partially compromised (e.g., renal artery stenosis) and blood ﬂow is fur-ther reduced by hemodynamic instability or locally via drug therapy [e.g., non-steroidal antiinﬂammatory drugs (NSAIDs), angiotensin-converting enzyme(ACE) inhibitors]. Renal vein obstruction may be due to thrombosis or ex-ternal compression (e.g., raised intra-abdominal pressure).Abdominal compartment syndrome—suggested by oliguria, a ﬁrm abdomen•on physical examination, and increased airway pressures (secondary to up-ward pressure on the diaphragms). Diagnosis is likely when sustained increased
CHAPTER1ThecriticallyillpatientwithAKI7intra-abdominal pressures (bladder pressure measured at end-expiration inthe supine position) exceed 25 mmHg. However, abdominal compartmentsyndrome may occur with intra-abdominal pressures as low as 10 mmHg.NephrotoxinsTable 1.1 lists some common nephrotoxins.Key referencesBellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal failure—deﬁnition,outcome measures, animal models, ﬂuid therapy and information technology needs:the Second International Consensus Conference of the Acute Dialysis Quality Initiative(ADQI) Group. Crit Care. 2004;8:R204-R212.Kellum JA. Acute Kidney Injury. Crit Care Med. 2008;36:S141-S145.Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients: a multina-tional, multicenter study. JAMA. 2005;294:813-818.Management of AKIIdentiﬁcation and correction of reversible causes of AKI is critical. All cases re-quire careful attention to ﬂuid management and nutritional support.Urinary tract obstructionLower tract obstruction requires the insertion of a catheter (suprapubic if thereis urethral disruption) to allow decompression. Ureteric obstruction requiresurinary tract decompression by nephrostomy or stent. A massive diuresis iscommon after decompression, so it is important to ensure adequate circulatingvolume to prevent secondary AKI.Hemodynamic managementFluid-responsive AKI may be reversible in its early stage. Careful ﬂuid manage-ment to ensure adequate circulating volume and any necessary inotrope orvasopressor support to ensure renal perfusion will help improve chances forrenal recovery. Admission to intensive care and use of hemodynamic monitoringshould be considered for all patients with AKI and is mandatory for patients notresponding to conservative therapy.Management of AKITable 1.1 Common nephrotoxinsAllopurinol Organic solventsAminoglycosides ParaquatAmphotericin PentamidineFurosemide Radiographic contrastHerbal medicines SulfonamidesHeavy metals ThiazidesNSAIDs
CHAPTER1ThecriticallyillpatientwithAKI8Glomerular diseaseSpeciﬁc therapy in the form of immunosuppressive drugs may be useful afterdiagnosis has been conﬁrmed.Interstitial nephritisAcute interstitial nephritis most often results from drug therapy. However, othercauses include autoimmune disease, and infection (e.g., Legionella, leptospirosis,Streptococcus, cytomegalovirus). Numerous drugs have been implicated, but themost common ones are as follows:Antibiotics (penicillins, cephalosporins, sulfa, rifampin, quinolones)•Diuretics (furosemide, bumetanide, thiazides)•NSAIDs (including selective COX-2 inhibitors)•Allopurinol•Cimetidine (rarely other H-2 blockers)•Proton pump inhibitors (omeprazole, lansoprazole)•Indinavir•5-Aminosalicylates•Urine sediment usually reveals white cells, red cells, and white cell casts.Eosinophiluria is present in about two-thirds of cases and speciﬁcity for interstitialnephritis is only about 80%. Other causes of AKI in which eosinophiluria is relativelycommon are rapidly progressive glomerulonephritis and renal atheroemboli.Discontinuation of the potential causative agent is a mainstay of therapy.Abdominal compartment syndromeAbdominal compartment syndrome is a clinical diagnosis in the setting of in-creased intra-abdominal pressure—pressures below 10 mmHg generally rule itout, while pressures above 25 mmHg make it likely. Baseline blood pressure andabdominal wall compliance inﬂuence the amount of intra-abdominal pressurethat can be tolerated. Surgical decompression is the only deﬁnitive therapy andshould be undertaken before irreversible end-organ damage occurs.Renal replacement therapyCRRT forms the mainstay of replacement therapy in critically ill patients whooften cannot tolerate standard hemodialysis due to hemodynamic instability.Hybrid techniques (discussed in Chapter 24) may be reasonable alternatives insettings where CRRT cannot be accomplished but outcome data are limited.Peritoneal dialysis is not usually sufﬁcient. Mortality in the setting of acute renalfailure in the critically ill is high (50%–60%). Renal recovery in survivors may beas high as 90% but recent studies suggest that sustained renal failure or incom-plete renal recovery is more common than previously thought (as many as 50%of survivors do not return to baseline renal function following an episode ofacute renal failure).
CHAPTER1ThecriticallyillpatientwithAKI9Clinical consequences of AKIUntil recently it was assumed that patients with AKI died not because of AKI it-self but because of their underlying disease. Several studies, however, have doc-umented a substantial mortality attributable to AKI after controlling for othervariables, including chronic illness and severity of underlying acute illness. Table1.2 lists some of the more important clinical consequences of AKI.Clinical consequences of AKITable 1.2 Clinical consequences of AKISystem Mechanisms ComplicationsElectrolytedisturbances1. Hyponatremia2. Hyperkalemia1. CNS (see below)2. Malignant arrhythmiasAcid-base(decreased chlorideexcretion, accu-mulation of organicanions like PO4,decreased albumin ldecreased buffering)1. Downregulation of Betareceptors, increased iNOS2. Hyperchloremia3. Impairing the insulin resistance4. Innate immunity1. Decreased cardiac output,blood pressure2. Lung, Intestinal injury,decreases gut barrier function3. Hyperglycemia, increasedprotein break down4. See belowCardiovascular 1. Volume overload 1. Congestive heart failure2. Secondary hypertensionPulmonary 1. Volume overload, decreasedoncotic pressure2. Inﬁltration and activation oflung neutrophils by cytokines3. Uremia1. Pulmonary edema, pleuraleffusions2. Acute lung injury3. Pulmonary hemorrhageGastrointestinal 1. Volume overload2. Gut ischemia and reperfusioninjury1. Abdominal compartmentsyndrome2. Acute gastric and duodenalulcerlbleeding; impairednutrient absorptionImmune 1. Decreased clearance of oxidantstress2. Tissue edema3. White cell dysfunction1. Increased risk of infection2. Delayed wound healingHematological 1. Decreased synthesis of RBCincreased destruction of RBC,blood loss2. Decreased production of erythro-poietin, von willebrand’s factor1. Anemia2. BleedingNervous system 1. Secondary hepatic failure,malnutrition, altered drugmetabolism2. Hyponatremia, acidosis3. Uremia1. Altered mental status2. Seizures, impairedconsciousness, coma3. Myopathy, neuropathylprolonged length onmechanical ventilationPharmacokineticsand dynamics1. Increased volume of distribution2. Decreased availability, albuminbinding, elimination1. Drug toxicity or under dosing
11Chapter 2History and rationalefor continuous renalreplacement therapyIlona Bobek and Claudio RoncoNew therapeutic advances have coped with an increasing clinical demand foradequate and effective renal replacement therapies in the critically ill patient.The history of continuous renal replacement therapy (CRRT) is one of thebest examples of multidisciplinary progress and collaboration between medicalknowledge and industrial technology toward therapy improvement.Medical demand/necessity for CRRTChange in the clinical picture of acute renal failure in the 1980sSevere sepsis was considered to be the underlying disease leading to ARF, andearlier ARF occurred frequently after abortions; however, ARFs epidemio-logical pattern and the involvement of other organs became more and moreclear after the 1990s:The cases of isolated (purely nephrological) ARF decreased due to early•diagnosis and better prophylaxis.More patients received increasingly extensive operations and survived serious•accidents.Number of intensive care unit (ICU) patients signiﬁcantly increased.•There was evidence of longer stay with possibility of better outcomes in ICU.•Change in the pathogenesis of ARFMain factors that are currently considered to be responsible for ARF are asfollows:Shock•Perfusion disturbances•Hypoxia•Medical demand/necessity for CRRT
CHAPTER2HistoryandrationaleforCRRT12Chronology/cornerstones of CRRT1960sThe idea of CRRT was born, but resources and technology were not available.Most ARF cases were treated with peritoneal dialysis (PD) because hemodi-alysis (HD) was difﬁcult to perform and it was not tolerated by intensive carepatients.1970sHenderson played an important role in developing the technical groundwork forhemoﬁltration (HF). Isolated ultraﬁltration (UF) and the use of convection forsolute removal was experimentally established.1977First description of an arterio-venous hemoﬁltration technique was given byKramer et al. in Göttingen, Germany.A vascular catheter that was accidentally placed into the femoral arterygave rise to the idea of using the systemic arterio-venous pressure differencein an extracorporeal circuit to generate the ultraﬁltrate, providing an effectivemethod for elimination of both ﬂuid and solutes.Heparin could be added before and fluid could be reinfused after theﬁltration. Continuous arterio-venous hemoﬁltration (CAVH) was soon acceptedworldwide in ICUs (Figure 2.1).Advantages of CAVH:Hemodynamic stability over conventional HD at that time•Simplicity•No necessity of blood pump•Continuous physiological ﬂuid removal•Limitations of CAVH:Low efﬁciency compared to HD•Reduced clearance capacity in the presence of high catabolic states•Necessity of additional, intermittent HD or HF•Complications associated with arterial access (indwelling catheters,•thrombosis)Reliance on arterial pressure to pump blood through the circuit•Danger of balancing errors•Necessity of continuous supervision by the staff•1979Continuous veno-venous hemoﬁltration (CVVH) was ﬁrst employed in ARFafter a cardiac surgery in Cologne, Germany. Any desired ﬁltrate volume couldbe arranged and uraemia could be controlled. A pump and control and balancingsystem became necessary (see Figure 2.2).Chronology/cornerstones of CRRT
CHAPTER2HistoryandrationaleforCRRT13Figure 2.1 The concept from Peter Kramer and Lee Henderson of continuous ﬁltration andthe ﬁrst patient treated with continuous arterio-venous hemoﬁltration in Vicenza, 1978.SubstitutionslösungHeparinBubbletrapVenous return clampGraduatedcylinderBlood pumpUltrafilterTo dialyzerVenous return linePo ABPoPrPi(1)(2)(2) (3)UltrafiltrateclampFigure 2.2 A typical system for continuous veno-venous hemoﬁltration (Hospal BM32).Minifilters
CHAPTER2HistoryandrationaleforCRRT141980sNumerous technical and methodical improvements in CRRT have contributedto the following:Changes in arterio-venous technique:Different types of catheters to obtain adequate blood ﬂow•Shorter blood line with no gadgets to reduce resistance•Positioning the collecting bag to apply a negative pressure•Optimization of treatment parameters and the concept of ﬁltration fraction•Changes in ﬁlter geometry and in the structure of ﬁber; an entire family of•diaﬁlters was created to fulﬁl the hemodynamic requirementsImplementation of CAVHD; to increase the efﬁcacy, dialyses ﬂuid was ﬁltered•through the external port of the ﬁlterCombination of hemofiltration and hemodialysis, that is, CAVHDF was•performedReplacement of the arterio-venous techniques by the pump-driven veno-venous techniques:Introduction of CVVH, employing blood pump to further increase efﬁciency•Use of double lumen catheters through jugular vein•Development of highly permeable polysulfone, polyacrylonitrile, and polyamid•membranes with a cut off between 15,000 and 50,000 daltonAvailability of bicarbonate-buffer solutions•Establishment of new anticoagulation methods, even for patients at high risk•of bleeding1982The use of CAVH in intensive care patients was approved by the Food and DrugAdministration (FDA) in the United States.1984For the ﬁrst time in the world history a neonate was treated with CAVH inVicenza, Italy (Figure 2.3).1990–2000Establishment of new technologies, modalities, and adequate dose of CRRT.Adoptive technology•Machines speciﬁcally created for CRRT (Figure 2.4)•Different modalities chosen for the need of the patient•The progression of dose delivery and prescription•CRRT is achievable in most of the ICUs worldwide•2000 to presentMultiorgan support therapyPatients do not die of ARF, but of multiorgan failure. The probability of deathis directly correlated to the number of failing organs other than the kidney and
CHAPTER2HistoryandrationaleforCRRT15Figure 2.3 The ﬁrst neonate treated in the world with continuous arterio-venous hemoﬁltra-tion (CAVH) and a special miniﬁlter (Vicenza, 1984).BloodpumpPressure monitoringBubble trapHeparinReinfusionpumpFigure 2.4 Evolution of continuous renal replacement therapy (CRRT) technology over theyears. The case of a single company.1985 1989YearsCRRT evolutionEvolutionProgress of CRRT2004 ?1977the severity of physiological disorders. The proper goal of extracorporeal bloodpuriﬁcation in ICU should be multiorgan support therapy (MOST). Treatmentshould not be directed at various organs as separate entities (Figure 2.5). Itshould be integrated and directed at patients. Therefore a wide range of sup-portive therapy in sepsis and liver failure were established, such as high-volume
CHAPTER2HistoryandrationaleforCRRT16Figure 2.5 The concept of MOST: Multiorgan supportive therapy. Blood can be circulatedby a platform through different ﬁltration/adsorption systems, leading to removal of speciﬁccompounds and support to different failing organs.SCUFCRRTLiverSupport - HVHFCPFA - CASTECLSM.O.S.Themoﬁltration (HVHF), coupled plasma ﬁltration and adsorption (CPFA),bioartiﬁcial liver, and endotoxin removal strategies.Acute dialysis quality initiativeAcute dialysis quality initiative (ADQI) is an ongoing process that seeks to pro-duce evidence-based recommendations for the prevention and management ofacute kidney injury (AKI) and on different issues concerning acute dialysis. Thefollowing goals have been achieved:Definition and classification of ARF [RIFLE criteria, acute kidney injury•network (AKIN)]Practice guidelines adopted in clinical practice (cardiac surgery-associated AKI)•Recent interests focus on timing of treatment initiation on patient survival andthe effect of dialysis modality on recovery of renal function in ARF.Future processes involve the online preparation of reinfusion ﬂuids duringhigh-volume hemoﬁltration, intracorporeal microﬂuidics and technology forplasma separation, intracorporeal ultraﬁltration plasma water extraction, bio-artiﬁcial tubulus, new sorbent techniques, nanotechnology, and wearable/trans-portable devices.Renal replacement therapy (RRT) has evolved from the concept that weneed to treat the dysfunction of a single organ, the kidney. However, CRRThas opened the door also to the concept of MOST. The future should requirea single multifunctional machine with a very user-friendly interface and ﬂex-ibility in parameters and prescription such that it can be used to respond to
CHAPTER2HistoryandrationaleforCRRT17different medical needs using different disposable layouts. The new generationof machines should be usable by different operators in different hospitals andsettings.Key referencesHenderson LW, Besarab A, Michaels A, Bluemle LW. Blood puriﬁcation by ultraﬁltrationand ﬂuid replacement (Diaﬁltration). Trans ASAIO. 1967;13:216-222.Henderson LW. Peritoneal ultraﬁltration dialysis: enhanced urea transfer using hypertonicperitoneal ﬂuid. JCI 1966;45:950-961.Kellum JA, Mehta R, Angus DC, Palevsky P, Ronco C; ADQI Workgroup. The ﬁrst inter-national consensus conference on continuous renal replacement therapy. Kidney Int.2002;62:1855-1863.Kramer P, Wigger W, Rieger J, Matthaei D, Scheler F. Arterio-venous hemoﬁltration: anew simple method for treatment of overhydrated patients resistant to diuretic. KlinWschr. 1977;55:1121-1122.Ronco C, Bellomo R, Brendolan A, et al. Effect of different doses in continuous veno-venous haemoﬁltration on outcomes of acute renal failure: a prospective randomizedtrial. The Lancet. 2000;355:26-30.Ronco C, Bellomo R. Acute renal failure and multiple organ dysfunction in the ICU: fromrenal replacement therapy (RRT) to multiple organ support therapy (MOST). Int J ArtifOrgans. 2002;25:733-747.Sieberth HG, History and development of continuous renal replacement (CRRT). CriticalCare Nephology. 1161-1167, Dordrecht: Kluwer Academic Publishers, 1998.
19Chapter 3Terminology andnomenclatureIan Baldwin and Rinaldo BellomoIntroductionAs with any specialized ﬁeld of therapy, speciﬁc terms and languages are used todescribe the use of renal replacement therapy (RRT).Key acronyms are also used to describe different extracorporeal circuits (EC)for the various techniques used for RRT. These terms generally differentiatesolute and solvent removal methods, treatment schedule or timing, and the in-tensity or “dose” of treatment.In addition, there are also speciﬁc terms used for the circuit itself and“system” components for the clinical device or machine used for RRT and forthe prescription of a treatment. For clinical care, clarity of prescription orders,research, audits, reporting, and publications it is necessary and useful to have acommon language.Deﬁnitions and relevant key termsContinuous renal replacement therapy (CRRT)Continuous renal replacement therapy is a general term referring to any extra-corporeal blood puriﬁcation therapy intended to substitute for impaired renalfunction over an extended period of time and applied for or aimed at beingapplied for 24 hours per day.Continuous veno-venous hemoﬁltration (CVVH)Continuous veno-venous hemoﬁltration is a technique of CRRT whereby bloodis driven through a highly permeable membrane by a peristaltic pump and via anEC originating in a central vein and also terminating in a central vein (Figure 3.1).The pressure generated by the pumped blood induces the passage of plasmawater (the solvent) across the membrane. This process is called ultraﬁltration.The ultraﬁltrate produced during the transit of blood through the membranecontains all molecules to which the membrane is permeable. As solvent movesacross the membrane it takes with it (solvent drag) many toxins that requireremoval. This process of blood puriﬁcation is called convection. The ﬂuid loss isIntroductionDeﬁnitions and relevant key terms
CHAPTER3Terminologyandnomenclature20replaced in part or completely with appropriate replacement solutions to achieveblood puriﬁcation while maintaining volume and electrolyte homeostasis.Continuous veno-venous hemodialysis (CVVHD)Continuous veno-venous hemodialysis is a technique of CRRT whereby bloodis driven through a highly permeable membrane by a peristaltic pump and viaan EC originating in a central vein and also terminating in a central vein butwhere solute removal is achieved by diffusion (exchange of solutes dependenton a concentration gradient) of molecules across a membrane. Such diffusionis achieved by pumping a toxin-free ﬂuid, which contains appropriate plasmaelectrolytes, into the nonblood side of the membrane and in a direction coun-tercurrent to that of blood (Figure 3.2). As this ﬂuid (dialysate) passes throughthe blood, molecules to which the membrane is permeable move from plasmawater to dialysate. The dialysate is then discarded.Continuous veno-venous hemodiaﬁltration (CVVHDF)Continuous veno-venous hemodiaﬁltration is a technique of CRRT that com-bines CVVH and CVVHD. During CVVHDF, solute removal is achieved by acombination of convection and diffusion (Figure 3.3). Blood is pumped into theEC from a central vein and returned into a central vein.Continuous arterio-venous techniquesContinuous arterio-venous techniques include all techniques of CRRT (hemo-ﬁltration, hemodialysis, and hemodiaﬁltration) whereby the patient’s bloodpressure (instead of pump) drives blood through a ﬁlter, which contains thehighly permeable membrane. This process is achieved via an EC originating in anartery and terminating in a vein. The method of blood puriﬁcation is otherwiseFigure 3.1 Continuous veno-venous hemoﬁltration (CVVH) circuit. In this circuit, the re-placement ﬂuid is being delivered before the ﬁlter in predilution mode.CVVHMembraneBlood pumpReplacementfluidHeaterWaste collection PatientPumpPump
CHAPTER3Terminologyandnomenclature21Figure 3.2 Continuous veno-venous hemodialysis (CVVHD) circuit.CVVHDBlood pumpPatientPumpDialysateHeaterSpentdialysateFigure 3.3 Continuous veno-venous hemodiaﬁltration (CVVHDF) circuit in predilution mode.CVVHDFBlood pumpPumpReplacementfluidPatientPumpDialysateHeaterHeaterDiafiltrate/wasteidentical to veno-venous techniques. They can be abbreviated in the same wayas veno-venous techniques except for the use of AV instead of VV. Thus, forexample, continuous arterio-venous hemoﬁltration would be abbreviated asCAVH. These techniques have all been abandoned in developed countries infavor of veno-venous techniques.For all techniques, ﬂuid balance is maintained by the difference between ﬂuidinput (dialysate and/or replacement ﬂuid or both) and output (spent dialysateand/or ultraﬁltrate or both). Both input and output can be regulated by pumps.
CHAPTER3Terminologyandnomenclature22If output is greater than input, there is a negative ﬂuid balance, which can be reg-ulated in intensity as deemed necessary.PredilutionPredilution is the administration of replacement ﬂuid into the patient’s bloodbefore its entry into the hemoﬁlter (preﬁlter delivery).PostdilutionPostdilution is the administration of replacement ﬂuid into the patient’s bloodafter its exit from the hemoﬁlter (postﬁlter delivery).Intermittent hemodialysis (IHD)Intermittent hemodialysis is a term that describes a diffusive blood puriﬁca-tion treatment during which blood and dialysate are circulated on the oppositesides of a semipermeable membrane in a countercurrent direction in order toachieve diffusive solute removal. IHD is performed using a machine, which ispurpose-built for this technique and which can generate dialysate ﬂow ratesthat are much higher than those used during CRRT. IHD machines can generatedialysate from tap water through a process of bacteria and endotoxin removaland reverse osmosis with subsequent electrolyte and buffer additives to pro-vide a dialysate ﬂuid for high ﬂow use (Figure 3.4). Prescription is commonly for3–4 hours per session with the frequency and intensity of such sessions regu-lated in response to perceived clinical need. Ultraﬁltration can also be achievedto remove ﬂuid by applying a negative pressure on the dialysate side of themembrane. This allows the removal of excess ﬂuid as clinically estimated.Figure 3.4 Intermittent hemodialysis circuit. Tap water is further puriﬁed by reverse osmosisand mixed with concentrates containing potassium, sodium, chloride, and bicarbonate toachieve physiological concentrations of each. The dialysate is then run countercurrent to bloodand discarded as waste.WasteHeaterR.O. K+HCO3-Electrolytes IHDTap waterBloodpumpPatient
CHAPTER3Terminologyandnomenclature23Slow low efﬁciency dialysis (SLED)Slow low efﬁciency dialysis is a dialysis treatment equivalent in nature to IHDbut where dialysate and blood ﬂow rates are reduced to provide a less efﬁcientclearance rate but with an extended time of treatment (e.g., 8–12 hours in-stead of 3–4 hours). The acronym SLEDD is used when this technique is applieddaily (D = daily). SLEDf is used when the technique includes some convectiveclearance in addition to diffusion (Figure 3.5). Extended daily dialysis (EDD) andextended daily dialysis with ﬁltration (EDDf) are also used to describe theserespective techniques.Extracorporeal circuit (EC)Extracorporeal circuit is the path for blood ﬂow outside the body. The ECincludes the plastic tubing carrying the blood to the ﬁlter (or hemoﬁlter or dia-lyzer) from the vascular access catheter and from the ﬁlter back to the body viathe access catheter again.Vascular access catheterVascular access catheter is a device inserted into a central vein to allow bloodto be pumped in and out of a ﬁlter. This device is typically in the form of a largecentral vein catheter (French gauge from 11.5 to 14.0) and has two lumens,one for outﬂow of blood from the body (typically referred to as “arterial”lumen) and one for the return of blood to the body (typically referred to as the“venous” lumen). Blood ﬂows between 150 and 300 mL/min can be typicallyachieved through such catheters.Figure 3.5 The circuit used to provide diffusive and convective clearance for SLED(f) usuallyfor a daily treatment of 6–12 hours. The dialysate and blood ﬂow rates are lower than that forintermittent hemodialysis.WasteHeaterR.O. K+HCO3-Electrolytes SLED(f)Tap waterFluidpumpBloodpumpPatient
CHAPTER3Terminologyandnomenclature24DiffusionDiffusion is a term that describes a type of solute transport across a semiper-meable membrane. During diffusion a solute has a statistical tendency to reachthe same concentration in the available distribution space on both sides of asemipermeable membrane. Thus, molecules move from the compartment withhigher concentration to the compartment with lower concentration.ConvectionConvection is a term that describes a type of solute transfer across a semi-permeable membrane in which solute is transported together with a solventby means of a process (ﬁltration) that occurs in response to a transmembranepressure gradient.Filter or dialyzerFilter or dialyzer is a tubular-shaped device that is made up of the plastic casingand the capillary ﬁbers of the semipermeable membrane within it.SummaryThere are many techniques of CRRT, and it is useful to understand theirnomenclature for ease of communication and understanding. Comparison ofsuch circuits with those of intermittent therapies such as IHD and SLED can fur-ther help one understand the mechanisms and principles involved in achievingblood puriﬁcation during CRRT. Familiarity with the various abbreviations usedin the clinical setting can help with rapid communication with other medical andnursing personnel.Key referencesBellomo R, Baldwin I, Ronco C. High-volume hemofiltration. Curr Opin Crit Care.2000;6:442-445.Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P; ADQI workgroup. Acute renalfailure-deﬁnition, outcome measures, animal models, ﬂuid therapy an dinformationtechnology needs: the second international consensus conference of the ADQI Group.Crit Care. 2004;8:R204-R212.Bellomo R, Ronco C. Nomenclature for CRRT. In: Bellomo R, Baldwin I, Ronco C, Golper T,eds. Atlas of Hemoﬁltration. London: WB Saunders; 2001:11-14.Bellomo R. Choosing a therapeutic modality: hemoﬁltration vs. hemodialysis vs. hemodia-ﬁltration. Semin Dial. 1996;9:88-92.Ronco C, Bellomo R. A nomenclature of continuous renal replacement therapies. ContribNephrol. 1995;116:28-33.Summary
25IntroductionRenal replacement therapy (RRT) is required in a signiﬁcant percentage ofpatients developing acute kidney injury (AKI) in an intensive care unit (ICU) set-ting. One of the foremost objectives of continuous renal replacement therapy(CRRT) is the removal of blood solutes retained as a consequence of decreasedor absent glomerular ﬁltration. Because prescription of CRRT requires goals tobe set with regard both to the rate and extent of solute removal, a thoroughunderstanding of the mechanisms by which solute removal occurs during CRRTis necessary. This chapter provides an overview of solute transfer during CRRT.Characterization of ﬁlter performance in CRRTClearanceQuantiﬁcation of dialytic solute removal is complicated by the confusion relatingto the relationship between clearance and mass removal for different therapies.By deﬁnition, solute clearance (K) is the ratio of mass removal rate (N) to bloodsolute concentration (CB):K = N/CB From a kinetic perspective, Figure 4.1 depicts the relevant ﬂows for determin-ing CRRT clearances while Figure 4.2 provides the solute clearance expressions,which differ from those used in conventional hemodialysis. In the latter therapy,the mass removal rate (i.e., the rate at which the dialyzer extracts solute fromblood into the dialysate) is estimated by measuring the difference in solute con-centration between the arterial and venous lines. In other words, a “blood-side”clearance approach is used. On the other hand, in CRRT, the mass removal rateIntroductionCharacterization of ﬁlter performance in CRRTChapter 4Basic principles of solutetransportZhongping Huang, Jeffrey J. Letteri, Claudio Ronco, andWilliam R. Clark
CHAPTER4Basicprinciplesofsolutetransport26Figure 4.1 Relevant ﬂow considerations for the determination of solute clearance in CRRT.The modality represented is CVVHDF.QACA QCCVQECECAQECE QDQRMass removal rateBlood concentrationClearanceFilter==ddiFigure 4.2 Solute clearance in CRRT.CVVHD/CVVHDFK = E·QDPostdilution CWHPredilution CWHK = S·QUFK = S·QUF·QBWQBW + QRConcentration in effluent dialysate/diafiltrateConcentration in bloodConcentration in bloodConcentration in filtrateS =E =((()))is estimated by measuring the actual amount of solute appearing in the efﬂuent.The mass removal rate is the product of the efﬂuent ﬂow rate (QE) and theefﬂuent concentration of the solute (CE).In continuous veno-venous hemodialysis (CVVHD) and continuous veno-venous hemodiaﬁltration (CVVHDF), the efﬂuent is dialysate and diaﬁltrate,respectively. For these therapies, the extent of solute extraction from the bloodis estimated by the equilibration ratio (E), also known as the degree of efﬂuentsaturation. The benchmark for efﬁciency in these therapies is the volume of ﬂuid(dialysate and/or replacement ﬂuid) required to achieve a certain solute clear-ance target (see below).Clearance in postdilution CVVH is the product of the SC (see below) andthe ultraﬁltration rate (QUF). For small solutes like urea and creatinine, the SC is
CHAPTER4Basicprinciplesofsolutetransport27essentially 1 (under normal ﬁlter operation). Therefore, small solute clearancein postdilution CVVH essentially is equal to the QUF. On the other hand, estima-tion of clearance in predilution CVVH has to account for the fact that the bloodsolute concentrations are reduced by dilution of the blood before it enters theﬁlter. Thus, the clearance has a “dilution factor” that is represented by the thirdterm on the right hand side of the second equation above. This term essentiallyis the ratio of the blood ﬂow rate (QB) to the sum of QB and the replacementﬂuid rate (QRF). (The actual blood ﬂow parameter, QBW, is blood water ﬂowrate.) In essence, the dilution factor can be viewed as a measure of the extentto which predilution differs from postdilution for a speciﬁc combination of QBand QUF.Sieving coefﬁcientWhen a dialyzer is operated as an ultraﬁlter (i.e., ultraﬁltration with no dialysateﬂow, e.g., CVVH), solute mass transfer occurs almost exclusively by convection.Convective solute removal is primarily determined by membrane pore size andtreatment ultraﬁltration rate. Mean pore size is the major determinant of a dia-lyzer’s ability to prevent or allow the transport of a speciﬁc solute. The sievingcoefﬁcient (SC) represents the degree to which a particular membrane permitsthe passage of a speciﬁc solute:SC = CUF/CP In this equation, CUF and CP are the solute concentrations in the ultraﬁltrate andthe plasma (water), respectively.Irrespective of membrane type, all ﬁlters in the “virgin” state have small sol-ute SC values of 1, and these values are typically not reported by dialyzer manu-facturers. SC values for solutes of larger molecular weight are more applicableand manufacturers frequently provide data for one or more middle moleculesurrogates, such as vitamin B12, inulin, cytochrome C, and myoglobin. As is thecase for solute clearance, the relationship between SC and solute molecularweight is highly dependent on membrane mean pore size.Sieving coefﬁcient (SC) data provided by manufacturers are usually derivedfrom in vitro experimental systems in which (nonprotein containing) aqueoussolutions are used as the blood compartment ﬂuid. In actual clinical practice,nonspeciﬁc adsorption of plasma proteins to a dialyzer membrane effectivelyreduces the permeability of the membrane. Consequently, in vivo SC valuesare typically less than those derived from aqueous experiments, sometimes bya considerable amount.Transmembrane solute removal mechanismsDiffusionDiffusion is the process of transport in which molecules that are present in asolvent and can freely move across a semipermeable membrane tend to moveTransmembrane solute removal mechanisms
CHAPTER4Basicprinciplesofsolutetransport28from the region of higher concentration into the region of lower concentra-tion (Figure 4.3). In reality, molecules present a random movement. However,since they tend to reach the same concentration in the available space occu-pied by the solvent, the number of particles crossing the membrane toward theregion of lower concentration is statistically higher. Therefore, this transportmechanism occurs in the presence of a concentration gradient for solutes thatare not restricted in diffusion by the porosity of the membrane. In addition totransmembrane concentration gradient, Fick’s Law states that diffusive solute isinﬂuenced by the following:Membrane characteristics: surface area, thickness, porosity•Solute diffusion coefﬁcient (primarily a function of molecular weight)•Solution temperature•Based on the previous discussion, the clearance of a given solute can bepredicted with reasonable certainty under a given set of operating conditions.However, several factors may lead to a divergence between theoretical andempirically derived values. As an example, protein binding or electrical chargesin the solute may negatively impact the ﬁnal clearance value. Conversely, con-vection may result in a measured clearance value that is signiﬁcantly greaterthan the value based on a “pure” diffusion assumption. Diffusion is an efﬁcienttransport mechanism for the removal of relatively small solutes, but as solutemolecular weight increases, diffusion becomes limited and the relative importanceof convection increases.ConvectionConvection is the mass transfer mechanism associated with ultraﬁltration ofplasma water. If a solute is small enough to pass through the pore structureof the membrane, it is driven (“dragged”) across the membrane in associationFigure 4.3 Mechanisms of diffusion and convection.Diffusion is solute transport across a semipermeable membrane—molecules move froman area of higher to an area of lower concentrationConvection is a process where solutes pass a cross the semipermeable membrane alongwith the solvent (‘‘solvent drag’’) in response to a positive transmembrane pressureEffectiveness less dependent onmolecular sizeBest for small molecule clearance
CHAPTER4Basicprinciplesofsolutetransport29with the ultraﬁltrated plasma water. This movement of plasma water is a conse-quence of a transmembrane pressure (TMP) gradient. Quantitatively, the ultra-ﬁltration ﬂux (JF), deﬁned as the ultraﬁltration rate normalized to membranesurface area, can be described byJF = KF TMP In this equation, KF is the membrane-speciﬁc hydraulic permeability (units: mL/h/mmHg/m2) and TMP is a function of both the hydrostatic and oncotic pressuregradients. Convective ﬂux of a given solute is a function mainly of the followingparameters:Ultraﬁltration rate•Blood solute concentration•Membrane sieving properties•In clinical practice, however, because plasma proteins and other factors modifythe “native” properties of the membrane, the ﬁnal observed SC is smaller thanthat expected from a simple theoretical calculation. As noted above, nonspe-ciﬁc adsorption of plasma proteins (i.e., secondary membrane formation) occursinstantaneously to an extracorporeal membrane after exposure to blood. Thischanges the effective permeability of the membrane and can be explained by theaction of proteins to essentially “plug” or block a certain percentage of mem-brane pores.Postdilution ﬂuid replacement tends to accentuate secondary membraneeffects because protein concentrations are increased within the membraneﬁbers (due to hemoconcentration). On the other hand, higher blood ﬂow rateswork to attenuate this process because the shear effect created by the blooddisrupts the binding of proteins to the membrane surface.Kinetic considerations for different CRRTtechniquesIn CVVH, high-ﬂux membranes are utilized and the prevalent mechanism ofsolute transport is convection. Ultraﬁltration rates in excess of the amountrequired for volume control are prescribed, requiring partial or total replace-ment of ultraﬁltrate losses with reinfusion (replacement) ﬂuid. As describedin greater detail elsewhere, replacement ﬂuid can either be infused before theﬁlter (predilution) or after the ﬁlter (postdilution). Postdilution hemoﬁltration isinherently limited by the attainable blood ﬂow rate and the associated ﬁltrationfraction constraint.On the other hand, from a mass transfer perspective, the use of predilutionhas several potential advantages over postdilution. First, both hematocrit andtotal blood protein concentration are reduced signiﬁcantly before the bloodenters into the hemoﬁlter. This effective reduction in the red cell and proteincontent of the blood attenuates the secondary membrane and concentrationKinetic considerations for different CRRTtechniques
CHAPTER4Basicprinciplesofsolutetransport30polarization phenomena described above, resulting in improved mass transfer.Predilution also favorably impacts mass transfer due to the augmented ﬂow inthe blood compartment because preﬁlter mixing of blood and replacement ﬂuidoccurs. This achieves a relatively high membrane shear rate, which also reducessolute-membrane interactions. Finally, predilution may also enhance mass trans-fer for some compounds by creating concentration gradients that induce solutemovement out of red blood cells.However, the major drawback of predilution hemoﬁltration is its relatively lowefﬁciency, resulting in relatively high replacement ﬂuid requirements to achievea given solute clearance. In a group of patients treated with a “traditional” bloodﬂow rate for CRRT, the efﬁciency loss associated with predilution has recentlybeen quantiﬁed. Troyanov et al. demonstrated the signiﬁcant negative effecton efﬁciency when a relatively low QB (less than 150 mL/min) is used with arelatively high QUF and QRF in predilution CVVH. This speciﬁc combination ofQB = 125–150 mL/min and QUF = 4.5 L/h (75 mL/min) is associated with a lossof efﬁciency of 30%–40% relative to postdilution for several different solutes. Inother words, to achieve the same solute clearance, 30%–40% more replacementﬂuid is required in predilution under these conditions, relative to postdilutionunder the same conditions. However, it should be noted that the likelihood ofachieving such an ultraﬁltration rate in postdilution is very remote at such a lowblood ﬂow rate, as this would require a ﬁltration fraction in excess of 50%. Thiscondition is likely to lead to very short-term ﬁlter patency.In CVVHDF, a high-ﬂux hemodiaﬁlter is used, and the operating principles ofhemodialysis and hemoﬁltration are combined. As such, this therapy may allowfor an optimal combination of diffusion and convection to provide clearancesover a very broad range of solutes. Dialysate is circulated in countercurrentmode to blood and, at the same time, ultraﬁltration is obtained in excess ofthe desired ﬂuid loss from the patient. The ultraﬁltrate is partially or totallyreplaced with reinfusion ﬂuid, either in predilution or postdilution mode. Later-generation CRRT machines allow a combination of predilution and postdilu-tion with the aim of combining the advantages of both reinfusion techniques.Information from the chronic hemodiaﬁltration literature suggests that a combi-nation of predilution and postdilution may be optimal in terms of clearance andoperational parameters. This may also be the case for CVVHDF in AKI, althoughthis possibility has not been assessed carefully. The optimal balance is most likelydictated by the speciﬁc set of CVVHDF operating conditions, namely blood ﬂowrate, dialysate ﬂow rate, ultraﬁltration rate, and ﬁlter type.Due to the markedly lower ﬂow rates used and clearances obtained inCVVHDF, the effect of simultaneous diffusion and convection on overall soluteremoval is quite different from the situation in chronic hemodiaﬁltration (HDF).In the latter application, diffusion and convection interact in such a manner thattotal solute removal is signiﬁcantly less than what is expected if the individ-ual components are simply added together. On the other hand, in CVVHDF
CHAPTER4Basicprinciplesofsolutetransport31the small solute concentration gradient along the axial length of the ﬁlter (i.e.,extraction) is minimal compared to the one that is seen in chronic HDF, in whichextraction ratios of 50% or more are the norm. Thus, the minimal diffusion-related change in small solute concentrations along the ﬁlter length allows anyadditional clearance related to convection to be simply additive to the diffusivecomponentTroyanov et al. have performed a direct clinical comparison of CVVHDF andpredilution CVVH with respect to urea and 2-microglobulin (B2M) clearanceat a “traditional” blood ﬂow rate of 125 mL/min. The study compared clear-ances at the same efﬂuent rate over an efﬂuent range of up to 4.5 L/h. As Figure4.4 indicates, urea clearance was higher in CVVHDF than in predilution CVVHand, in fact, the difference between the two therapies increased as efﬂuent rateincreased. These results are consistent with the “penalizing” effect of predilution,which is especially pronounced at low blood ﬂow rates. For B2M, the results arecontrary to the “conventional wisdom,” which would suggest that a purely con-vective therapy like CVVH should inherently be superior to a partly convectivetherapy like CVVHDF for clearance of a molecule of this size. However, onceagain, the penalty of predilution in CVVH is apparent, as the B2M clearancesfor the two modalities are equivalent except at very high efﬂuent rates (greaterthan 3.5 L/h). Until the impact of higher blood ﬂow rates on solute clearances inCRRT can be assessed, these and other data suggest that CVVHDF is a logicalmodality choice to achieve the broadest spectrum of solute molecular weightrange in the most efﬁcient way.Figure 4.4 Comparison of solute clearance in predilution CVVH and CVVHDF. FromTroyanov S, Cardinal J, Geadah D, et al. Solute clearances during continuous venvenoushaemoﬁltration at various ultraﬁltration ﬂow rates using Multiﬂow-100 and HF1000 ﬁlters.Nephrol Dial Transplant. 2003;18:961-966. Reprinted by permission of Oxford University Press.7060504030201000 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5Effluent (mL/h)Clearance(mL/min)Urea, CVVHM-100Urea, CVVHDFB2-Microglobulin, CVVHB2-Microglobulin, CVVHDF
CHAPTER4Basicprinciplesofsolutetransport32SummaryRational prescription of CRRT to critically ill patients with AKI is predicatedupon an understanding of the basic principles of solute and water removal. Inthis chapter, the major ways in which ﬁlter function is characterized have beenreviewed clinically. In addition, the fundamental mechanisms for solute and ﬂuidtransport have been discussed. Finally, these principles have been applied in atherapeutic context to the various CRRT modalities used by clinicians managingAKI patients.Suggested readingsBrunet S, Leblanc M, Geadah D, Parent D, Courteau S, Cardinal J. Diffusive and convectivesolute clearances during continuous renal replacement therapy at various dialysate andultraﬁltration ﬂow rates. Am J Kidney Dis. 1999;34:486-492.Clark WR, Turk JE, Kraus MA, Gao D. Dose determinants in continuous renal replace-ment therapy. Artif Organs. 2003;27:815-820.Henderson LW. Biophysics of ultrafiltration and hemofiltration. In: Jacobs C, ed.Replacement of Renal Function by Dialysis. 4th ed. Dordrecht: Kluwer AcademicPublishers; 1996:114-118.Huang Z, Letteri JJ, Clark WR, Ronco C. Operational characteristics of continuous renalreplacement therapy modalities used for critically ill patients with acute kidney injury.Int J Artif Organs. 2008;31:525-534.Huang Z, Letteri JJ, Clark WR, Zhang W, Gao D, Ronco C. Ultraﬁltration rate as dosesurrogate in pre-dilution hemoﬁltration. Int J Artif Organs. 2007;30:124-132.Troyanov S, Cardinal J, Geadah D, et al. Solute clearances during continuous venvenoushaemoﬁltration at various ultraﬁltration ﬂow rates using Multiﬂow-100 and HF1000ﬁlters. Nephrol Dial Transplant. 2003;18:961-966.Summary
33The control and optimization of ﬂuid balance is a clinically important compo-nent of continuous renal replacement therapy (CRRT). Inadequate ﬂuid removalis associated with peripheral edema and vital organ edema (i.e., pulmonaryedema). Such edema can retard weaning from mechanical ventilation or com-prise wound healing. Fluid overload has been identiﬁed as an independent pre-dictor of increased mortality in critically ill patients and is clearly undesirable.Similarly, excessive ﬂuid removal may contribute to hypovolemia with increaseddoses of vasopressor drug therapy, exposing the patients to the risks of unneces-sary beta and alpha receptor stimulation. Hypovolemia may induce hypotensionand, thereby, possibly aggravate organ injury and, speciﬁcally, retard or blockrenal recovery. Accordingly, careful clinical assessment of the patient’s ﬂuidstatus and careful prescription of CRRT to optimize ﬂuid balance, together withfrequent review of such assessment and prescription, represent a key aspect ofbest practice in the ﬁeld of CRRT.Patient ﬂuid balance: This term refers to the total balance over a 24-hour periodof ﬂuids administered (intermittent drugs, continuous infusion of drugs, blood,blood products, nutrient solutions, additional ﬂuids) and measurable ﬂuids re-moved (drainage from chest or abdomen, urine—if present, blood loss, andexcess ﬂuid removed by the CRRT machine).Machine (CRRT) ﬂuid balance: This term refers to the total balance over a 24-hourperiod of ﬂuids administered by the CRRT machine (dialysate or replacementﬂuid or both depending on the technique and any additional anticoagulant infu-sion) and ﬂuids removed by the CRRT machine (spent dialysate or ultraﬁltrateor both depending on the technique).Efﬂuent: It is the total amount of ﬂuid discarded by the machine. In continuousveno-venous hemoﬁltration (CVVH), this is the same as ultraﬁltrate. In contin-uous veno-venous hemodialysis (CVVHD), this is equal to the spent dialysate +any additional ultraﬁltrate generated by the machine. In continuous veno-venousIntroductionChapter 5Principles of ﬂuidmanagementRinaldo Bellomo and Sean M. Bagshaw
CHAPTER5Principlesofﬂuidmanagement34hemodiaﬁltration (CVVHDF), this is the same as the sum of spent dialysate andultraﬁltrate discarded by the machine (also called spent ultradiaﬁltrate).Dry weight: This is the patient’s normal/optimal weight before the onset of ill-ness. This weight is often available in detail in elective operative patients whereit is typically measured before the operation. In other cases, it might need to beestimated.Edema: This term refers to the accumulation of excess ﬂuid in the extracellularcompartment. In the subcutaneous tissue, it can be detected by the phenom-enon of pitting of the skin under pressure. In the lungs, if signiﬁcant, it can bedetected by radiography.Assessment of ﬂuid status: This term refers to the clinical process of estimatingthe patient’s intravascular and extravascular ﬂuid status. Such assessment is com-plex and imperfect. It requires consideration of vital signs, invasive and noninva-sive hemodynamic measurements, information of ﬂuid balance and body weight,and radiological information. Such assessment is necessary to guide ﬂuid balanceprescription during CRRT.Approach to ﬂuid balance during CRRTThe prescription of CRRT-related ﬂuid management and its integration intooverall patient ﬂuid management can be assisted by a speciﬁc order chart (Table5.1) for the machine ﬂuid balance.The above order chart will tell the nurse how to set the machine and howto achieve the planned hourly ﬂuid balance. However, in the intensive care unit(ICU), the ﬂuid needs of the patients are not static and require frequent review.For example, should the same patient require the administration of 600 mLof fresh frozen plasma over 2 hours prior to an invasive procedure, necessaryadjustments to the order should be made with speciﬁcation for the duration ofchange and the reasons (Table 5.2).The ﬂuid balance prescription related to the machine can be usefully relatedto the patient and a ﬂuid balance prescription describing the overall patient ﬂuidbalance goal for a 12-hour time period is useful for informing the nurse whatthe broad goals of ﬂuid therapy are in a given patient. This may be expressed inan additional prescription attached to the previous machine ﬂuid balance chart(Table 5.3).Approach to ﬂuid balance during CRRTTable 5.1 Example Order ChartTechnique Dialysateﬂow rateReplacementﬂuid ﬂowrateEfﬂuentﬂow rateAnticoagulantinfusion ﬂowrateMachineﬂuidbalanceCVVHDF 1000 mL/h 1000 mL/h 2300 mL/h 100 mL/h –200 mL/h
CHAPTER5Principlesofﬂuidmanagement35Practical considerationsThe above goals can be achieved by means of physician and nursing educationand by ensuring that no CRRT session can be started unless such orders areclearly and legibly written, signed, and accompanied by the physician’s printedname and contact number. They also require the regular recording of ﬂuid bal-ance on an hourly basis and its correct ﬁnal addition of ﬂuid losses and gains.This can be done in a computerized system or added by the nurse at the bedsideusing a pocket calculator and then charted. This process allows the creation ofa running hourly balance, which is useful in ensuring that progress is being madeat the appropriate speed, in the appropriate direction, and to the prescribedamount.Expected outcomes, potential problems,cautions, and beneﬁtsThe expected outcome of a systematic process for the prescription, delivery,and monitoring of ﬂuids during CRRT is the ability to ensure that the patientwill receive prescribed therapy in a safe and effective manner. This approach willminimize errors and their consequences (persistent ﬂuid overload or dangerousintravascular volume depletion).Despite this careful approach, problems can still arise. A relatively commonproblem is related to off-time (time during which CRRT is not operative dueto ﬁlter clotting or an out-of-ICU procedure or investigation). Under such cir-cumstances, the ﬂuid removal cannot proceed as planned. If the patient has 5hours of off-time, then the consequence may be that close to 1 L of plannedPractical considerationsExpected outcomes, potential problems,cautions, and beneﬁtsTable 5.3 Example Order Chart 3Patient MedicalrecordnumberOverall ﬂuidbalance frommidnight to12:00 (noon)Overall ﬂuidbalance from12:00 (noon)to midnightRight atrialpressurenotiﬁcation rangeName 00123 –1000 mL –1000 mL/h <6 or >15 mmHgTable 5.2 Example Order Chart 2Technique Dialysateﬂow rateReplacementﬂuid ﬂow rateEfﬂuentﬂow rateAnticoagulantinfusion ﬂowrateMachine ﬂuidbalanceCVVHDF 1000 mL/h 1000 mL/h 2600 mL/h 100 mL/h –500 mL/h(for 2 hoursonly duringFFP treatment)Note: FFP = Fresh frozen plasma
CHAPTER5Principlesofﬂuidmanagement36ﬂuid removal fails to occur (assuming ﬂuid balance of 200mL/h). Moreover, dur-ing this off-time, patients may be administered additional ﬂuid that will counterearlier ﬂuid balance goals. If this happens, the physician and the nurse need tobe alert to the consequences and respond appropriately. This my require anadjustment in ﬂuid removal during the ensuing 12 or 24 hours, which safely com-pensates for the off-time by increasing the hourly ﬂuid removal by, for example,an extra 100 mL/h. Due consideration needs to be paid to speciﬁc patientswhere such ﬂuid removal may be problematic. However, typically, machine ﬂuidremoval rates of 300–400 mL/h are well tolerated in ﬂuid-overloaded patients.Nonetheless, caution should be exerted and the patient’s condition should bereviewed frequently.Another relatively common problem is the frequent interruptions of therapydue to machine alarms. In some patients who are agitated or who have frequentleg ﬂexion in the presence of a femoral access catheter or who sit up and movein the bed in the presence of a subclavian access device, the machine pressurealarms may be frequently triggered. In addition, other alarms related to substi-tution ﬂuids bag or waste bag changes interrupt treatment. This may lead toperiods of 5–10 minutes over an hour and over a day create “lost treatmenttime” and failure to achieve ﬂuid balance goals. It is often prudent to prescribe agreater ﬂuid loss than desired to compensate for these factors. Most machinesallow the operator to check what the actual ﬂuid removal achieved was over agiven time period. Such checks should be done to ensure that the correct ﬂuidremoval is entered into the ﬂuid balance calculations; many nursing protocolsmandate ﬂuid balance check each hour particularly for inexperienced nurses.The beneﬁts of such continuous monitoring of ﬂuid delivery and removal aremany. They include frequent patient assessment, vigilance with regard to othersimultaneous therapies, attention to detail, avoidance of dangerous swings inﬂuid status, and competent and detailed machine operation.ConclusionAttention to ﬂuid balance during CRRT is of great clinical importance. Inadequateﬂuid removal leads to clinical complications, especially in relation to weaningfrom mechanical ventilation. Excessive ﬂuid removal can cause hypovolemia andhypotension, and retard renal recovery. Best practice in this ﬁeld can only beachieved by a systematic combination of frequent and thoughtful assessment,attention to detail, rigorous and vigilant monitoring of ﬂuid input and output,and clear and explicit description and prescription of the goals of therapy withregard to both machine settings and patient management.Conclusion
CHAPTER5Principlesofﬂuidmanagement37Key referencesBagshaw SM, Baldwin I, Fealy N, Bellomo R. Fluid balance error in continuous renal re-placement therapy: a technical note. Int J Artif Organs. 2007;30:435-440.Bagshaw SM, Bellomo R. Fluid resuscitation and the septic kidney. Curr Opin Crit Care.2006;12:527-530.Bagshaw SM, Bellomo R. The inﬂuence of volume management on outcome. Curr Opin CritCare. 2007;13:541-548.Bagshaw SM, Brophy PD, Cruz D, Ronco C. Fluid balance as a biomarker: impact ofﬂuid overload on outcome in critically ill patients with acute kidney injury. Crit Care.2008;12:169.
39Indications for renal replacement therapyIndications for renal replacement therapy (RRT) fall into two broad categories,so-called “renal” (i.e., to speciﬁcally address the consequences of renal failure)and “nonrenal” (without necessitating renal failure). Although the distinction isnot always precise, it is a reasonably easy way to categorize indications for RRT.Renal indicationsThe manifestations of acute kidney disease (as discussed in Chapter 1 and sum-marized in Table 6.1) include oliguria, (leading to volume overload), azotemia(leading to a host of clinical complications), hyperkalemia, and metabolic aci-dosis. While there is no consensus regarding the precise level of dysfunctionin any of these areas that should prompt initiation of RRT, general agreementexists on the following general indications for RRT:Volume overload (e.g., pulmonary edema)•Azotemia with uremic symptoms•Hyperkalemia (>6.0 mmol/L)•Metabolic acidosis (pH < 7.2) due to renal failure•Volume overloadVolume overload usually occurs in the setting of oliguria, but it may occur simplybecause urine output is insufﬁcient to maintain ﬂuid balance in the face of largeIndications for renal replacement therapyChapter 6Indications, timing, andpatient selectionJohn A. KellumTable 6.1 Diuretic dosingOral IV InfusionMetolazone 10–20 mg qdChlorothiazide 250–500 mg IVFurosemide 20–40 mg 6–24-hourly 5–80 mg 6–24-hourly 1–10 mg/hTorsemide 5–20 mg 6–24-hourly 5–20 mg 6–24-hourly 1–5 mg/hBumetanide 0.5–1 mg 6–24-hourly 0.5–2 mg 6–24-hourly 1–5 mg/h
CHAPTER6Indications,timing,andpatientselection40volume input—even if true oliguria is not present. Furthermore, most authoritiesrecommend therapy before volume overload becomes clinically manifest, andthus RRT may be used to “create space” for additional ﬂuids (e.g., nutritionalsupport, antibiotics) that are scheduled to be administered.There is a controversy regarding the role of diuretics in the setting of volumeoverload secondary to acute renal failure (ARF). While most clinicians willattempt diuretics prior to initiation of RRT, there is a wide variation as tohow long or intense such a trial will be or how the success will be deﬁned.Although it is obviously desirable to avoid RRT, there is little evidence to sug-gest that diuretics can be successful in achieving this goal and even the availableevidence suggests potential harm. Importantly, attempts to increase urine outputwith diuretics should only be directed toward treatment of volume overload orhyperkalemia, not oliguria per se. Large observational studies have failed to showbeneﬁt from diuretics in critically ill patients with oliguria, and some studies haveshown harm.Diuretic therapyA loop diuretic such as furosemide is given in a dose of 20–40 mg intravenously.If this dose is ineffective, a higher dose can be tried within 30–60 minutes. Higherdoses may be needed if the patient has previously received diuretic therapy (seeTable 6.2). If boluses doses of 80 mg every 6 hours are infective, an infusion maybe started (1–5 mg/h IV). A thiazide diuretic such as chlorothiazide (250–500mg IV) or metolazone (10–20 mg PO) can be used in conjunction with a loopdiuretic to improve diuresis. In general, there is no point in continuing diuretictherapy if it is not effective; loop diuretics in particular may be nephrotoxic.Discontinue all diuretics prior to initiating RRT.AzotemiaAzotemia, the retention of urea and other nitrogenous waste products, resultsfrom a reduction in glomerular ﬁltration rate (GFR) and is a cardinal feature ofkidney failure. However, like oliguria, azotemia represents not only disease butalso a normal response of the kidney to extracellular volume depletion or adecreased renal blood ﬂow. Conversely, a “normal” GFR in the face of volumedepletion could only be viewed as renal dysfunction. Thus, changes in urineoutput and GFR are neither necessary nor sufﬁcient for the diagnosis of renalpathology. Yet, no simple alternative for the diagnosis currently exists.Azotemia is also a biochemical marker of the uremic syndrome, a conditioncaused by a diverse group of toxins that are normally excreted but build up inthe circulation and in the tissues during renal failure. The clinical manifestationsof the uremic syndrome are shown in the Table 6.2.Although uremic symptoms correlate with the level of urea in the blood, therelationship between blood urea nitrogen (BUN) and uremic symptoms is notconsistent across individuals or even within a given individual at different times.Thus, there is no threshold level of BUN that deﬁnes uremia or provides a spe-ciﬁc indication for RRT. Instead, the provision of RRT and, indeed, decisions
CHAPTER6Indications,timing,andpatientselection41Table 6.2 Manifestations of renal failureSystem Complication(s) Mechanism(s) Clinical featuresCardiovascular Volume overload Salt/water retention Edema, heart failure,hypertensionElectrolyte andacid-baseHyponatremia,hyperkalemia,acidosis, azotemiaImpaired free waterexcretion, chlorideaccumulationHypotension, impairedglucose metabolism,decreased muscleprotein synthesis, cardiacdysrhythmiasGastrointestinal Impaired nutrientabsorption,GI bleeding,abdominalcompartmentsyndromeBowel edema,platelet dysfunction,volume overloadNausea, vomiting,decreased mucosal/intestinal absorption,increased intra-abdominalpressuresHematological Anemia, plateletdysfunctionDecreasederythropoietin,decreased vonWilibrand’s factorAnemia, bleedingImmune Infections, immunesuppressionImpaired neutrophilfunctionInfection, sepsisNervous Encephalopathy Uremic toxins,hyponatremiaAsterixis, delirium, comaRespiratory Pleural effusions,pulmonary edemaVolume overload,decreased oncoticpressure, ? directuremic toxicityPleural effusion, pulmonaryedema, respiratory failureregarding timing and intensity should be individualized to patients on the basis ofclinical factors and not solely on the basis of biochemical markers.HyperkalemiaHyperkalemia may be severe and can be life threatening. The risks of hyperkalemiaare greatest when it develops rapidly, where serum concentrations in excess of6 mmol/L may produce cardiac dysrhythmias. The earliest electocardiographicsign of hyperkalemia is peaking of the T waves. This ﬁnding is associated with car-diac irritability and should prompt emergent treatment. Temporary managementof severe hyperkalemia (while preparing for RRT) includes intravenous calciumchloride (10 mL of 10% solution) to reduce cardiac irritability and a combinationof insulin (10 units IV) and dextrose (50 mL D50) given together over 20 minutesto shift potassium intracellularly (blood glucose should be monitored).Metabolic acidosisRenal failure causes metabolic acidosis by retention of various acid anions(e.g., phosphate, sulfate) as well as from renal tubular dysfunction resulting inhyperchloremic acidosis. Clinical manifestations range from acute alterations
CHAPTER6Indications,timing,andpatientselection42in inﬂammatory cell function to chronic changes in bone mineralization. Mildalterations can be managed using oral sodium bicarbonate or calcium carbonate.RRT is effective in removing acids as well as correcting plasma sodium and chlo-ride balance and is generally targeted at maintaining an arterial pH > 7.30.“Nonrenal” indicationsSo-called nonrenal indications for RRT are to remove various dialyzablesubstances from the blood. These substances include drugs, poisons, contrastagents, and cytokines.Drug and toxin removalBlood puriﬁcation techniques have long been used for removal of variousdialyzable drugs and toxins. A list of common drugs and toxins that can bereadily removed using RRT is shown in Table 6.3. The majority of poisoningcases do not require treatment with RRT. Indeed, the drugs or toxins that aremost commonly responsible for poisoning-related fatalities are not amenable toRRT (e.g., acetaminophen, tricyclic antidepressants, short-acting barbiturates,Table 6.3 Common poisonings treated with RRTSubstance ExtracorporealmethodCommentsMethanol Hemodialysis RRT should be continued until the serum methanolconcentration is < 25 mg/dL and the anion-gapmetabolic acidosis and osmolal gap are normal.Rebound may occur up to 36 hours.Isopropanol Hemodialysis RRT effectively removes isopropanol and acetone,although it is usually unnecessary except in severecases (prolonged coma, myocardial depression,renal failure).Ethylene glycol Hemodialysis RRT should be continued until the ethylene glycollevel is <20 mg/dL and metabolic acidosis or othersigns of systemic toxicity have been resolved.Rebound may occur up to 24 hours.Lithium IHD/CRRT IHD removes lithium faster but rebound is asigniﬁcant problem and can be addressed effectivelywith CRRT.Salicylate IHD/CRRT Both IHD/CRRT have been reported in themanagement of salicylate poisoning.Theophylline IHD/CRRT/hemoperfusionRRT should be continued until clinical improvementand a plasma level < 20 mg/L is obtained. Reboundmay occur.Valproic acid IHD/CRRT/hemoperfusionAt supratherapeutic drug level plasma proteinsbecome saturated, and the fraction of unbound drugincreases substantially and becomes dialyzable.Note: Other treatments are also required for many of these substances.
CHAPTER6Indications,timing,andpatientselection43stimulants, and “street drugs”). In general, the size of the molecule and thedegree of protein binding determines the degree to which the substance canbe removed (smaller, nonprotein bound substances are easiest to remove).Continuous renal replacement therapy (CRRT) may be effective in removingsubstances with higher degrees of protein binding and is sometimes used toremove substances with very long plasma half-lives. Techniques such as sorbenthemoperfusion may also be used for this indication and are discussed further inChapter 23.The role of CRRT in the management of acute poisonings is not well estab-lished. There is relatively lower drug clearance per unit of time compared tointermittent hemodialysis (IHD) but CRRT has a distinct advantage in hemody-namically unstable patients who are unable to tolerate the rapid solute and ﬂuidlosses associated with IHD or even other techniques such as hemoperfusion.CRRT may also be effective for the slow, continuous removal of substanceswith large volumes of distribution, a high degree of tissue binding, or for sub-stances that are prone to “rebound phenomenon” (e.g., lithium, procainamide,and methotrexate). In such cases, CRRT may even be used as adjuvant therapywith IHD or hemoperfusion.Contrast agentsRRT has been used to remove radio-contrast agents for many years, but thepurpose of this treatment has changed over time. In the past, ionic, high-osmolarcontrast was used for imaging studies, and RRT was often used to remove thesesubstances and to remove ﬂuid in patients with renal failure who were at risk ofcongestive heart failure from the large osmotic load. These patients could notexcrete the contrast and would develop pulmonary edema after contrast ad-ministration. In more recent years, nonionic, low-osmolality, or even iso-osmo-lar agents have been developed, and the risk of pulmonary edema has decreasedsigniﬁcantly. However, all radio-contrast agents are nephrotoxic and CRRT isbeing advocated by some experts to help prevent so-called contrast nephrop-athy. Standard IHD has been shown to remove radio-contrast agents but doesnot appear to prevent contrast nephropathy. Despite less efﬁciency in removingcontrast, CRRT has been shown to result in less contrast nephropathy, partic-ularly when it has begun prior to or in conjunction with contrast administration(see Table 6.4). However, the effect is controversial and most centers do notcurrently offer RRT for prevention of contrast nephropathy.CytokinesMany endogenous mediators of sepsis can be removed using continuous veno-venous hemoﬁltration (CVVH) or continuous veno-venous hemodiaﬁltration(CVVHDF) (dialysis is not able to remove these mediators). This observationhas prompted many investigators to attempt to use CVVH as an adjunctivetherapy in sepsis. While it remains controversial as to whether CVVH offers ad-ditional beneﬁt in patients with renal failure and sepsis, available evidence doesnot support a role of CVVH for the removal of cytokines in patients without
CHAPTER6Indications,timing,andpatientselection44renal failure. If CVVH is capable of removing cytokines, the effect of standard“renal dose” CVVH appears to be small. However, some individuals appearto respond with improved hemodynamics, especially to higher doses of CVVH(also see Chapter 7).Key studiesLee PT, Chou KJ, Liu CP, et al. Renal protection for coronary angiography in advancedrenal failure patients by prophylactic hemodialysis. A randomized controlled trial. J AmColl Cardiol. 2007;50:1015-1020.Mehta RL, Pascual MT, Soroko S, et al; PICARD Study Group. Diuretics, mortality, andnonrecovery of renal function in acute renal failure. JAMA. 2002;288:2547-2553.Uchino S, Doig GS, Bellomo R, et al; B.E.S.T. Kidney Investigators. Diuretics and mortalityin acute renal failure. Crit Care Med. 2004;32:1669-1677.Timing of RRTWhen to initiate RRTThe simplest answer to the question “when should RRT be started?” wouldbe when the indications discussed above are met. Numerous attempts havebeen made to reach a consensus on timing of RRT. The Acute Dialysis QualityInitiative (ADQI) ﬁrst addressed this issue in 2000 but was unable to reach con-sensus beyond stating that a patient is considered to require RRT when he orshe has “an acute fall of GFR and has developed, or is at risk of, clinically signif-icant solute imbalance/toxicity or volume overload.” In essence this amountsto saying that RRT should begin when a patient has “symptomatic” ARF. Whatconstitutes symptomatic ARF is a matter of clinical judgment and how “at risk”Timing of RRTTable 6.4 Methods to reduce contrast nephropathyOral IV DosingaSaline 0.9% (154mEq/L)1 mL/kg/h begun 12 hours or3 mL/kg/h begun 1 hour prior toprocedure and 1 mL/kg/h continuing6 hours after procedureNaHCO3 in water 150 mEq/L 1 mL/kg/h begun 12 hours or3 mL/kg/h begun 1 hour priorto procedure and 1 mL/kg/hcontinuing 6 hours afterprocedureN-acetylcysteine 1200 mg every12 hours1200 mg every12 hoursbeginning 24 hours beforeand continuing 24 hours afterprocedure: aDosing ranges are provided as a general guide only—none of the above agents are approved for thisindication.
CHAPTER6Indications,timing,andpatientselection45is interpreted. Most, but not all, experts advise that RRT should begin beforeclinical complications occur, but it is often difﬁcult to know exactly when sucha point occurs. For example, subtle abnormalities in platelet function can beginearly in acute kidney injury (AKI) prior to when most clinicians would beginRRT.Observational studies of AKI using RIFLE criteria have provided two impor-tant pieces of information: ARF (stage F by RIFLE) is common among critically illpatients (10%–20% of ICU patients) and is associated with a 3- to10-fold increasein the risk of death prior to discharge. Given the profound increase in the riskof death, many investigators have asked why more patients do not receive RRT,yet many patients with ARF recover renal function without ever receiving RRT.Should these patients receive RRT? Current evidence is insufﬁcient to answerthis question, but given the low rates of complications associated with CRRT,and high risk of death associated with AKI, consideration should be given tostarting therapy early (e.g., when F criteria is present rather than waiting forcomplications to occur).When to stop RRTAn even more difﬁcult question to answer than when to start is when to stopRRT. Again the simplest answer would be “when renal function has recovered,”but two problems exist with this simple answer. First, it is not always easy to de-termine when renal function has recovered and it is also unclear what amount ofrecovery should be sought prior to cessation of therapy. In essence the questionis not dissimilar to so-called weaning from mechanical ventilation and very littleis actually known about how and when “weaning” from RRT should occur. Oneapproach that was used in the largest trial of dialysis intensity published to dateused the rule described in Table 6.5.Key studiesHoste EA, Clermont G, Kersten A, et al. RIFLE criteria for acute kidney injury is associatedwith hospital mortality in critical ill patients: A cohort analysis. Crit Care. 2006;10:R73.Kellum JA, Mehta RL, Levin A, et al; AKIN. Development of a clinical research agenda foracute kidney injury using an international, interdisciplinary, three-step modiﬁed Delphiprocess. Clin J Am Soc Nephrol. 2008;3:887-894.Table 6.5 Assessment for recovery of renal function if urinevolume > 30 mL/hCreatinine clearance Management of RRT<12 mL/min Continuation of RRT12–20 mL/min Clinician’s judgment>20 mL/min Discontinuation of RRTNote: 6 hours’ timed urine collections obtained for assessment of creatinine clearance.