AmyloidoticCardiomyopathy:Multidiscipl inaryApproach to Diagnosisand Treatment

David C. Seldin, MD, PhDa,b,*, John L. Berk, MDa,c,Flora Sam, MDa,d, Vaishali Sanchorawala, MDa,b

KEYWORDS

� Amyloidosis � Cardiomyopathy � Transthyretin� Immunoglobulin light chain � Matrix metalloproteinase� Autologous stem cell transplantation � Biomarkers

AMYLOIDOTIC CARDIOMYOPATHY: NOTA RARE DISEASE

Amyloidosis is the generic term for a family offibrillar protein deposition diseases. In the mid-19th century, pathologist Rudolf Virchow foundamorphous-appearing deposits in sections ofkidney, spleen, heart, and other tissues in autop-sies performed on patients succumbing to untreat-able infections. The investigator called themstarchlike or amyloid. It is now known that amyloiddeposits are composed of regular 10-nm poly-meric protein fibrils; the authors surmise thatVirchow was observing amyloid composed offibrils of serum amyloid A protein, an acute phasereactant and the cause of amyloid A amyloidosis(AA) or secondary amyloidosis. AA is still found inpatients with chronic mycobacterial or bacterialinfections; those with rare hereditary periodic feversyndromes, such as familial Mediterranean fever;and those with refractory autoimmune diseases.

Funding support: NIH R01 NS051306 (J.L.B.); HL079099, HThe authors have no conflicts of interest.a Amyloidosis Treatment and Research Program, DepaMedicine, Boston Medical Center, K5, 72 East Concord Sb Hematology-Oncology Section, Department of MedicBoston, MA 02118, USAc Pulmonary Section, Department of Medicine, FGH1, 72d Division of Cardiology, Whitaker Cardiovascular Institu* Corresponding author. Amyloidosis Treatment and RUniversity School of Medicine, Boston Medical Center, KE-mail address: dseldin@bu.edu

Heart Failure Clin 7 (2011) 385–393doi:10.1016/j.hfc.2011.03.0091551-7136/11/$ – see front matter � 2011 Elsevier Inc. All

Although AA most often involves the kidney, about10% of patients develop cardiac involvement.1

Other forms of amyloidosis more frequentlyaffect the heart and are more likely to be diag-nosed by cardiologists. Familial amyloidosis (AF)is caused by inheritance of mutations in DNAencoding abundant serum proteins that becomeprone to misfolding and aggregation. Inheritedmutations of transthyretin (TTR) are the mostcommon cause of AF, producing syndromes offamilial amyloidotic cardiomyopathy (FAC) andfamilial amyloidotic polyneuropathy (FAP), de-pending on the tissue tropism of the particularmutant. TTR mutations are common in regions ofPortugal, Japan, and Sweden but can be foundworldwide. More than 100 polymorphisms of TTRhave been reported, of which more than 80 areamyloidogenic.2 Other serum proteins that cancause inherited amyloidoses include fibrinogen,lysozyme, gelsolin, and apolipoproteins.

L095891, and HL102631 (F.S.)

rtment of Medicine, Boston University School oftreet, Boston, MA 02118, USAine, Boston Medical Center, 72 East Concord Street,

East Concord Street, Boston, MA 02118, USAte, 715 Albany Street, W507, Boston, MA 02118, USAesearch Program, Department of Medicine, Boston5, 72 East Concord Street, Boston, MA 02118.

rights reserved. heartfailure.th

eclinics.com

Box 1Symptoms and signs of cardiac amyloidosis

Common symptoms

� Fatigue� Exertional dyspnea� Edema

Uncommon symptoms

� Jaw or buttock claudication� Atypical chest pain

Physical findings

� Rales� Pitting edema� Elevated jugular venous pressure� Hepatojugular reflux� Adventitious third heart sound

Electrocardiographic findings

� Low-voltage electrocardiograph� Atrial fibrillation� Ventricular arrhythmias

Echocardiographic findings

� Concentric hypertrophy� Thickened interventricular septum� Diastolic dysfunction� Systolic dysfunction (late)

Seldin et al386

Immunoglobulin light chain amyloidosis (AL),formerly termed primary amyloidosis, is causedby fibrils composed of immunoglobulin light chains(LCs). Immunoglobulin AL can occur not only inclonal B lymphoproliferative diseases, usuallyin low-grade plasma cell dyscrasias, but also inmultiple myeloma, B cell lymphoma, chroniclymphocytic leukemia, or Waldenstrom macro-globulinemia. As many as 50% of patients withAL have cardiac involvement, which is oftenrapidly progressive and fatal if untreated.The hereditary and acquired forms of amyloid-

osis are relatively rare. However, just as normalaging is accompanied by the development of Abplaques in the brain, it is also accompanied bythe formation of amyloid deposits in other tissues,of which the heart is the organ in which this depo-sition is most clinically apparent. This syndrome istermed senile systemic amyloidosis (SSA), some-times termed senile cardiac amyloidosis. In SSA,the fibrils are formed of wild-type TTR protein.Histopathologic evidence of SSA can be found in10% to 25% of people older than 80 years and inalmost all centenarians,3 but how frequently SSAcauses amyloidotic cardiomyopathy (ACMP) isuncertain. Nonetheless, in an aging population,SSA and the associated ACMP are increasinglyrecognized by cardiologists investigating diastolicor systolic hypertrophic cardiomyopathy (CMP)and heart failure (HF).Thus, ACMP can occur as a consequence of

a blood disorder, an inherited genetic disease, ora part of normal aging. As more and more effectivetherapies are developed for these disorders, accu-rate and timely diagnosis has become increasinglyimportant for patients.

DIAGNOSIS OF ACMP: CLINICAL SUSPICIONPLUS APPROPRIATE LABORATORYINVESTIGATION

First and foremost, the key to the diagnosis ofamyloidosis is to consider the diagnosis. Amyloid-osis, not syphilis, is the great mimic of the twenty-first century because it can masquerade as morecommon disorders capable of causing nephroticsyndrome and renal insufficiency, cholestatic liverfailure, malabsorption and gastrointestinal bleed-ing, or peripheral or autonomic neuropathy. ForACMP, the typical presentation is with symptomsof HF with preserved left ventricular function, man-ifesting first as nonspecific symptoms of fatigue,exertional dyspnea, and hypotension (Box 1).These symptoms progress over time, eventuallyto systolic HF. As discussed later, HF due toamyloidosis is refractory to many of the usual inter-ventions. The amyloidotic heart is also highly

susceptible to heart block and arrhythmia, likelyas a result of deposition of fibrils in the conductingsystem. Progressive HF and sudden death are themost common causes of death in patients withACMP because of ventricular arrhythmias orpulseless electrical activity (PEA). Atrial arrhyth-mias are also common.Amyloidosis should be suspected in patients

who present with unexplained nephrotic syn-drome, cholestatic liver disease, autonomicneuropathy, peripheral neuropathy, CMP withconcentric hypertrophy, or combinations of thesesyndromes. Pathognomonic symptoms and signsof amyloidosis include macroglossia and periorbi-tal purpura, and these findings should be promptlypursued.With respect to the heart, patients usually come

to the attention of cardiologists for evaluation ofexertional dyspnea, fatigue, or hypotension. Chestpain due to amyloid in epicardial coronary vesselsis rare, although patients with amyloidosis in smallarterioles and capillaries may develop atypicalchest pain, jaw claudication with chewing, orbuttock claudication with ambulation. Hypoten-sion can occur because of a combination ofcardiac dysfunction and autonomic insufficiency.Such symptoms are evaluated by echocardiog-

raphy to assess cardiac structure and function.

Amyloidotic Cardiomyopathy 387

Textbooks describe a classic “sparkly” pattern onechocardiography that was actually an artifactproduced by low-resolution imaging and is notseen as commonly with modern equipment. Thecommon echocardiographic features of ACMPare concentric hypertrophy of the ventricular walls,biatrial enlargement, and abnormal diastolic fillingparameters and, commonly, mild to moderatevalvular regurgitation. HF symptoms can occurwith minimal ventricular wall thickening. The stan-dard electrocardiogram can provide a tip-off to thepresence of an infiltrative CMP because the limblead voltages are often reduced rather thanincreased as in other hypertrophic CMPs. Cardiacmagnetic resonance imaging can identify struc-tural and functional abnormalities consistent withamyloidosis, and a phenomenon of delayed sub-endocardial enhancement with gadolinium hasbeen reported.4 The only specific scan for amyloiddeposits is done using labeled serum amyloid P(SAP) component,5 an accessory protein thatbinds to amyloid fibrils and has been postulatedto protect amyloid fibrils from degradation.However, this scan is not useful for diagnosingcardiac amyloidosis because the tracer pools inthe heart; furthermore, SAP scanning is not avail-able in the United States.

The gold standard for diagnosing amyloidosis isa tissue biopsy demonstrating the presence offibrils that bind the dye Congo red, producingapple-green birefringence under polarized micros-copy (Table 1). Fibrils can also be identified byelectron microscopy as rodlike bundles that are10 nm in diameter. The most accessible site fordemonstration of fibrils is in fat readily obtainedby aspiration from the abdominal wall using an18-gauge syringe needle after administering a localanesthetic. The result of the fat test is positive in65% to 95% of cases, depending on the type of

Table 1Workup of cardiac amyloidosis

Assay A

Congo Red Fibrils 1

Serum and Urine Immunofixation Electrophoresisfor Monoclonal Immunoglobulin

1

Serum Free LCs 1

Abnormal TTR Isoelectric Focusing �Abnormal TTR Gene Sequence �Kappa or Lambda Immunohistochemistry 1

TTR Immunohistochemistry �Amyloid A Immunohistochemistry �Mass Spectrometry Protein Identification 1

systemic amyloidosis.6 If the result is negative ina patient suspected of having amyloidosis, thenext step is usually to proceed with biopsy of anaffected organ, although biopsies of salivaryglands and the rectum are other sources of tissuethat frequently show positive results. In patientswith cardiac symptoms and signs indicatingamyloidosis without clinically detectable involve-ment of other organs, an endomyocardial biopsywould be performed next. Biopsy material shouldbe fixed in formalin and also in paraformaldehydefor immunoelectron microscopy, in case immuno-histochemical studies fail to identify the subunitprotein by light microscopy.

Once amyloid fibrils are identified, the next stepis to establish the type of amyloid protein, a criticalstep in determining treatment. Multiple modalitiesare used to do this step. Patients with AL almostalways have evidence of a clonal plasma cellprocess by one or more of the following tests:bone marrow immunohistochemistry or flow cy-tometry for k and l LCs, serum or urine immunofix-ation electrophoresis (the standard serum andurine immunofixation electrophoreses are usuallynormal because there is rarely enough monoclonalimmunoglobulin to be detected with these tech-niques), or serum free LC testing using the FreeliteAssay, a relatively new test marketed internation-ally by the Binding Site Co. Normally, almost allLCs are associated with immunoglobulin heavychains to form tetramers. In plasma cell diseases,the levels of free LC are elevated and can be de-tected by nephelometry using specific antibodies.Unlike whole immunoglobulins, LCs are rapidlyexcreted by the kidney and have a half-life ofonly 6 hours; so, measurement of serum-freeLCs provides an early measure of diseaseresponse in patients undergoing treatment, aidingdiagnosis.

L ATTR AA SSA Source

1 1 1 Pathology laboratory

� � � Clinical laboratory

� � � Binding Site, Quest

1 � � Specialty laboratory

1 � � Specialty laboratory, Athena

� � � Pathology laboratory

1 � 1 Pathology laboratory

� 1 � Pathology laboratory

1 1 1 Specialty laboratory

Seldin et al388

Patients with TTR amyloidosis (ATTR) or otherhereditary forms of amyloidosis have DNA muta-tions that can be detected by gene sequencing.However, the presence of a mutation or a plasmacell disorder is not conclusive for identifying theamyloid type, and in most cases, immunochem-ical or biochemical identification is essential.However, amyloid fibrils are notoriously stickyand bind many antibodies nonspecifically, socareful controls, immunoelectron microscopywith gold-labeled antibodies, and extraction ormicrodissection of the fibrils and identification bymass spectrometry are important confirmatorytests, which are available in specialized centers.Misdiagnosis and inappropriate application ofchemotherapy should be avoided.

Table 2Cardiac biomarker staging system

StagingNT-proBNP, 332 ng/L;CTnT, 0.035 mg/L

MedianSurvival (mo)

I Neither elevated 26

MECHANISMS OF CARDIAC DYSFUNCTIONIN ACMP

Cardiac dysfunction in amyloidosis has been attrib-uted to thedepositionof amyloid fibrils andphysicaldisruption of the integrity of the myocardial syncy-tium. However, clinical observation has shownthat the degree of cardiac dysfunction is not neces-sarily proportional to the thickness of the walls;thus, it has been hypothesized that amyloid mayhave other effects on the heart. For example,CMP in AL has the worst prognosis and most rapidprogression, although wall thickness in patientswith this condition may be much less than in thosewith ATTR or SSA7,8 in whom wall thickness canexceed 2 cm. A likely explanation is that prefibrillarLC oligomers have direct toxic effects on cardio-myocyte function, impairing excitation-contractioncoupling via increased oxidant stress9,10 mediatedby p38 MAPK signaling.11

Extracellular amyloid fibrils also seem to disruptcardiac matrix homeostasis and alter extracellularmatrix (ECM) turnover.12 Regulated ECM turnoveris critical for the maintenance of myocyte-myocyteforce coupling and proper myocardial function.Disruption of the ECM alters matrix metalloprotei-nases (MMPs) and their tissue inhibitors in ALCMP. As a result, matrix degradation is impaired.Reactive oxygen species alter myocardial MMPactivity by translational and posttranslationalmechanisms, activating MMPs and decreasingfibrillar collagen synthesis in cardiomyocytes,contributing to the accelerated clinical manifesta-tions of the disease.

II One elevated 11

III Both elevated 4

Data from Dispenzieri A, Gertz MA, Kyle RA, et al. Serumcardiac troponins and N-terminal pro-brain natriureticpeptide: a staging system for primary systemic amyloid-osis. J Clin Oncol 2004;22:3751.

CARDIAC BIOMARKERS AND RISKSTRATIFICATION IN AL

The extent of ACMP is the single most importantdeterminant of outcome in AL.13 The concentration

of cardiac biomarkers, B-type natriuretic peptide(BNP) and troponins, has been shown to beelevated with cardiac stress and cardiomyocytedamage caused by amyloidogenic LC. Thesebiomarkers have been useful for both diagnosisand prognosis.Criteria for the assessment of cardiac involve-

ment at baseline and of cardiac response aftertreatment have been established by an interna-tional consensus panel.14 Short of endomyocardialbiopsy, electrocardiography and echocardiog-raphy were accepted as the gold standards forthe diagnosis of heart involvement by amyloidosis.Recently, it has been shown that serum cardiactroponin T (CTnT) and serum cardiac troponin I(CTnI) as well as BNPs (either BNP or N-terminal-proBNP [NT-proBNP]) are highly sensitive markersof cardiac involvement, and normal values excludeclinically significant cardiac amyloidosis.15 Further-more, an analysis of outcomes for 242 patientswithAL demonstrated that patients could be dividedinto 3 prognostic groups based on elevation ofNT-proBNP and troponin levels (NT-proBNP>332ng/L, CTnT>0.035 mg/L, and CTnI>0.1 mg/L).16

Patients with normal proBNP and troponin levelsin this study had a median survival of about 2 yearsandwere designated as stage I. Patientswith eitherbiomarker elevated were categorized into stage IIand had a median survival of slightly less thana year. Patients with both cardiac biomarkerselevated were categorized into stage III, corre-sponding to amedian survival of only 3 to 4months(Table 2).In addition, among patients with cardiac involve-

ment, cardiac troponins provide quantitative infor-mation about myocardial damage. Mediansurvivals of patients with elevated CTnT andCTnI levels were 6 and 8 months, respectively,and worse than that for those with undetectablevalues (22 and 21 months, respectively). Aftermultivariate analysis, CTnT provided a betterpredictor of survival than CTnI.17 Nonetheless,

Amyloidotic Cardiomyopathy 389

the baseline CTnI has also been shown to be anexcellent predictor of relative survival.18

Based on these studies, biomarker criteria forcardiac involvement and improvement andprogression after therapy are now being incorpo-rated into the consensus for organ involvementand response.19 Furthermore, reductions in levelsof cardiac biomarkers from baseline also correlatewith improvement in survival after treatment inpatients with cardiac involvement; a 30% or 300ng/L decrease in the NT-proBNP level from base-line correlates with improved survival, whereas anincrease of that magnitude correlates withprogression and worse survival posttreatment.

AL TREATMENT

Treatment of AL is aimed at eradicating the plasmacell clone that produces the deleterious fibril-formingLC. The first attempt todo thiswasbyusingoral dosing of the alkylating chemotherapymelphalan, accompanied by prednisone. In AL, 2outcomes are assessed: hematologic responses(reduction in the plasma cell clone and LC produc-tion) and clinical responses (improvement in organfunction). Studies have demonstrated that theseoutcomes are linked, and treatments that signifi-cantly reduce production of the LC fibril precursor,for example, by 90%ormore, are those that can beassociated with clinical improvement. Melphalanand prednisone are relatively ineffective becausepartial (50%) responses occur in only 20% to25% of patients and deeper responses andimprovements in organ function were rare. Thus,there was little impetus to make the diagnosis andinitiate therapy, and most patients died of theirdisease within the first 1 to 2 years of diagnosis.

About 15 years ago, the authors and othercenters took advantage of accumulating datafrom the myeloma field indicating that high-doseintravenous melphalan (HDM) supported by trans-plant of autologous bone marrow stem cells (SCT)could produce complete or near-complete hema-tologic responses and transferred this approach,with some trepidation, to patients with AL. Overthe next few years, it was learned that a high rateof complete hematologic responses can beachieved, and organ function can then improve.However, inexperienced centers have hadtreatment-related mortality (TRM) rates as highas 30% to 40%. In a multicenter randomized trialfrom France that compared outcomes of HDM/SCT to oral melphalan chemotherapy, 25% ofenrolled patients in the HDM/SCT group did notreceive their transplant and another 25% diedbecause of TRM.20 At Boston Medical Center,the authors have observed that the TRM in the

early years was 14%21 and more recently hasbeen reduced to less than 5%. Thus, with carefulselection of patients and expert multidisciplinarycare, morbidity and mortality during HDM/SCTcan be minimized.

The first step in this treatment involves harvest-ing hematopoietic stem cells. At present, this har-vesting is almost always accomplished byleukapheresis of peripheral blood after administra-tion of high-dose myeloid growth factors, usuallygranulocyte colony-stimulating factor (G-CSF),that promote hematopoietic stem cell divisionand egress from the bone marrow. It is rare tosubject patients to bone marrow harvest. Forthose patients who fail to mobilize enough cellswith G-CSF, plerixafor, an antagonist of theCXCR4 chemokine receptor, can be administered,which promotes release of stem cells from thebone marrow stroma. Hematopoietic stem cellsare then cryopreserved while patients undergotreatment with high doses of antiplasma cellchemotherapy that is myeloablative. Melphalan isthe most useful alkylating chemotherapy agentfor this purpose. With the reinfusion of stem cells1 to 2 days after chemotherapy, the hematopoieticsystem is soon reconstituted, with neutrophilengraftment typically occurring 10 days later andplatelet and erythrocyte engraftments a few daysafter. It is this period of pancytopenia during whichpatients are at highest risk of infection and alsomucositis and enteritis because of the melphalan.During this period, shifts in fluid and electrolytelevels, stress, fever, fatigue, cytokines, and infec-tion provide a significant stress on the heart, andit is not infrequent for patients to have their firstatrial or ventricular arrhythmia during this period.Exacerbation of HF symptoms frequently occurs.Even the administration of high doses of G-CSFduring the prechemotherapy period can precipi-tate such events. Guidelines for the managementof congestive HF (CHF) and arrhythmias in patientswith amyloidosis are provided below.22

However, if this treatment can be administeredsafely, the outcomes are excellent. Patients whohave cessation of LC production after HDM/SCTare able to recover organ function. The authorshave demonstrated significant improvement innephrotic syndrome and recently in wall thick-ness by echocardiography.23 Hepatomegaly alsoregresses in many patients. More subjectivesymptoms of fatigue, lightheadedness, anorexia,and gut dysmotility also improve, as does qualityof life. However, this improvement is a slowprocess, and patients often require extensivesupportive care for 6 to 12 months as their perfor-mance status and immunologic function slowlyimproves.

Seldin et al390

Patients with advanced ACMP are at high risk ofcomplications, and such patients should be identi-fied using clinical criteria or biomarkers and gener-ally excluded from undergoing HDM/SCT. Thestandard alternative chemotherapy regimen isconsidered to be pulse oral melphalan combinedwith dexamethasone, which is fairly well toleratedand can produce hematologic responses andorgan improvement.24 Nonetheless, patients withACMP are poorly tolerant of multiday pulses ofhigh-dose corticosteroids, and the authors gener-ally use a less-intensive regimen of dexametha-sone 1 day a week, instead of 4 consecutivedays once or twice a month. It is also commonfor patients with significant ACMP or edema dueto nephrotic syndrome to require dose reductionfrom the typical starting dose of 40 mg each dayto 20 mg or even 10 mg.Multiple myeloma treatment has undergone

a transformation in the last 5 or so years with theadvent of novel antiplasma cell agents. Theseagents fall into 2 major classes of proven efficacy:the immunomodulator drugs (so-called IMiDs), ofwhich thalidomide was the first in class, andthe proteasome inhibitors, of which bortezomib(Velcade) was the first and only one so far achievingFood and Drug Administration approval. TheIMiDs are thought to act on the bone marrowmicroenvironment, affecting stromal-plasma cellinteractions, cytokines, and angiogenesis. Theproteasome inhibitors seem to be more directlyplasma cytotoxic, perhaps, because plasma cellsare antibody factories particularly susceptible todisruption of intracellular proteolytic pathways.Thalidomide, its newer analogs lenalidomide andpomalidomide, and bortezomib have all beenstudied in patients with AL and have shown toprovide effective antiplasma cell activity. Thesedrugs clearly have a role in the treatment of AL,alone or in combination. It is too soon to arguethat these agents can replace HDM/SCT in good-risk patients, but further clinical trials will undoubt-edly demonstrate that these agents have a role inpatients who are ineligible for HDM/SCT or in thosein whom the treatment fails or the conditionrelapses or perhaps in induction or asmaintenancetherapy.However, these agents are not benign, particu-

larly in patientswith AL. Thalidomidewasmarketedin Europe as a sedative, until a high rate of birthdefects was observed; thus, IMiDs are highly regu-lated, and pregnancy must be avoided during theiruse. In addition, IMiDs are prothrombotic, andpatients must take aspirin or full warfarin anticoa-gulation when they are on IMiDs. IMiDs also havecardiac effects; thalidomide has been reported tocause bradycardia, and lenalidomide has been

noted recently to raise the BNP level. IMiDs canalso produce worsening azotemia in patients withamyloid renal disease.25 These agents are used atlower doses in AL than in multiple myeloma, andpatients must be monitored closely for the earlier-mentioned and other side effects.Bortezomibhasadifferent spectrumof toxicities,

with neuropathy being a common dose-relatedside effect. The cardiac side effects of bortezomibare less well understood, although exacerbationsof CHF and arrhythmia have been seen and cardio-toxicity has been replicated in animal models.26

These agents should still be considered to beinvestigational in patients with AL.

HEREDITARY TTR AMYLOIDOSIS

Variant ATTR arises from point mutations in exons1 to 4 of the TTR exons on chromosome 18, result-ing in more than 100 identified single amino acidsubstitutions. Disease prevalence is estimated at1 in 100,000 people in the United States, givingATTR orphan disease status (<200,000 affectedin the United States). One ATTR mutation, V122I,is found in 3.9% of the African American popula-tion; however, the rate of clinical expression isundefined, and African Americans are still morelikely to present with symptomatic AL CMP thanFAC.27 In contrast, the ATTR carrier frequency innorthern Portugal is 100 times more than the prev-alence in the United States, and in certain northernSwedish communities, ATTR affects 3% to 5% ofthe population.The clinical spectrum of ATTR is variable and

dependent on the nature of the mutation. MostTTR mutations induce peripheral and/or auto-nomic neuropathy (FAP), followed by amyloidCMP (FAC) and, less frequently, renal disease.V30 M typically induces neuropathy and rarelyaffects the heart, whereas T60A and V122I almostexclusively produce CMP and infrequently affectperipheral nerves. FAC is predominant for approx-imately 40 ATTR variants. The course of TTR FACdiffers from AL-related heart dysfunction. Incontrast to the rapidly deteriorating course ofAL CMP, FAC develops slowly and patients areoften asymptomatic until the amyloid involvementof the myocardium is advanced in late stages ofdisease. Although oligomeric LCs themselvesseem to be deleterious on myocardial contractility,variant TTR does not seem to affect heart function(L. W. Connors, PhD, personal communication,2010). However, ATTR CMP more often presentswith conduction delays or complete heart blockthan AL CMP.TTR is almost exclusively produced by the liver,

with minor quantities made by the choroid plexus

Amyloidotic Cardiomyopathy 391

and retinal epithelium. In an effort to eliminatevariant TTR production and prevent continuedamyloid fibril formation, liver transplant was firstperformed in Sweden in 1990. To date, morethan 1700 liver transplants have been undertakenin patients with ATTR (see the Familial AmyloidoticPolyneuropathy World Transplant Registry athttp://www.fapwtr.org/). Although well tolerated,liver transplants have not proven the panaceathey were predicted to be. Once fibril formationhas begun, it seems that in some cases wild-type TTR made by the transplanted liver can beincorporated into the amyloid deposits and thedisease can progress.28,29 These findings led torecommendations to limit liver transplant in ATTRto patients with early neurologic disease andminimal signs of cardiac amyloid deposition. Forthose with ATTR CMP, 2 therapeutic optionsremain: heart and liver transplant or experimentalmedical therapies.

Over the past decade, experiments examiningthe thermodynamic landscape of ATTR fibrilformation determined that disaggregation of nativeTTR tetramer represented the critical and rate-limiting steps. Small molecules occupying thethyroid-binding sites raised the activation barrierfor TTR tetramer dissociation. Using x-ray crystal-lography to characterize the thyroid-binding siteand high throughput modeling, Dr Jeffery Kelly(The Scripps Research Institute) identifieda nonsteroidal antiinflammatory drug (NSAID), di-flunisal, as a candidate small-molecule inhibitorof ATTR fibril formation. At the same time, FoldRxPharmaceutical (Cambridge, MA, USA) developeda proprietary small-molecule inhibitor, tafamidis,designed to maximize ATTR tetramer binding andeliminate potential NSAID toxicities. Both agentsinhibit ATTR fibril formation in vitro. These indepen-dent investigational programs led to separateinternational multicenter randomized placebo-controlled studies; results are pending. Small-molecule inhibitors, RNA interference, or proteinstabilizers appear to be the future of ATTRmanagement, particularly in patients with amyloidCMP at disease presentation in whom liver trans-plant may not halt progression.

MANAGEMENT OF HF AND ARRHYTHMIASIN PATIENTS WITH ACMP

Regardless of the specific treatment directedagainst the plasma cell dyscrasia, supportivecare to decrease symptoms and support organfunction plays an important role in the manage-ment of disease and requires coordinated careby specialists in multiple disciplines.

The mainstay of the treatment of HF in ACMP issodium restriction and the use of diuretics; higherdoses may be required if the serum albumin levelis low as a result of concomitant nephroticsyndrome. Furthermore, achieving a balancebetween HF and intravascular volume depletion isparticularly challenging, especially in patientswith autonomic nervous system involvement ornephrotic syndrome.Diuretic resistance is commonin patients with severe nephrotic syndrome, andmetolazone or spironolactone may be required inconjunction with loop diuretics. In a patient withanasarca, intravenous diuresis is often neededbecause absorption of diuretics may be impaired.Diuretic-resistant large pleural effusions may indi-cate the presence of pleural amyloid.30 Sucheffusionscanbemanagedwith repeated thoracent-esesorplacementof indwellingpleural catheters forcontinuous drainage; pleurodesis is often ineffec-tiveand is frequentlycomplicatedbypneumothoraxbecause of friable pleural tissue.

In contrast to other types of HF, in ACMP, thereis no evidence that drugs such as b-blockers orangiotensin-converting enzyme (ACE) inhibitorsare beneficial; in fact, their use can be quitedangerous. Any of these drugs should be usedcautiously in ACMP, starting with a low doseadministered in a monitored setting becauseeven small doses may result in profound hypoten-sion. b-Blockers and calcium channel blockersmay produce hypotension because of their nega-tive inotropic effect.31–33 Patients with ACMPmay be hypersensitive to ACE inhibitors becausein the setting of amyloid-induced autonomicdysfunction, there may be increased reliance onthe renin-angiotensin system for maintenance ofadequate blood pressure. There are no publisheddata on the use of intravenous inotropic or vasodi-lator drugs in patients with severe HF resultingfrom amyloidosis. Renal-dose dopamine (1–3 mg/kg/min) can be helpful for the treatment ofanasarca, provided that renal function is unim-paired. Patients may be particularly prone todysrhythmias at higher doses of dopamine orwith dobutamine therapy.

Digoxin is also not generally useful in amyloid-osis, and patients may be at an increased risk ofdigoxin toxicity, despite therapeutic serum digoxinlevels. Digoxin has been shown to bind avidly toamyloid fibrils,34 leading to high local levels ofthe drug in the myocardium.

Thus, options for medical management ofpatients with severe ACMP are limited. In highlyselected cases, orthotopic heart transplant maybe considered. Early experience with cardiactransplant in AL suggested that mortality did notdiffer from that in other disorders,35 but with longer

Seldin et al392

follow-up, greater mortality than expected wasobserved, usually because of disease progressionin the heart or other organs.36 As a result of theseobservations, many solid organ transplantationcenters consider AL a contraindication to hearttransplant. However, with the advent of high-dose chemotherapy and stem cell transplant, it ispossible to transplant the heart and performchemotherapy 6 to 12 months later to eliminatethe underlying plasma cell dyscrasia, preventingamyloid deposition in the transplanted heart andother organs. Several patients have been treatedsuccessfully with this combined approach andhave an actuarial survival that is similar to that ofpatients undergoing heart transplant for otherindications.37,38

Arrhythmias occur frequently in patients withACMP,22 yet the optimal management of arrhyth-mias in these patients with ACMP remains contro-versial. Atrial fibrillation can often be suppressedwith amiodarone. Patients who fail amiodaroneor do not tolerate it may be candidates for ablationprocedures. Patients with atrial fibrillation shouldbe anticoagulated if possible, although patientswho have amyloidosis are often prone to bleedingbecause of capillary fragility or deficiency of circu-lating clotting factors. Furthermore, in severeACMP, the atria as well as the ventricles are infil-trated, and atrial contractile dysfunction may bepresent even during sinus rhythm, predisposingto atrial thrombus formation.39 It is thereforeprudent to anticoagulate patients with amyloidosisif there is defective left atrial mechanical activityon echocardiography, even in the absence offibrillation.40

Management of ventricular arrhythmias ischallenging, and evidence-based guidelines arelacking. Patients with recurrent syncope or symp-tomatic ventricular arrhythmias may also betreated with amiodarone, which can suppress ec-topy but has not been proved to reduce mortalityfrom sudden death because of ventricular fibrilla-tion or PEA. Implantable defibrillators have beenused, again without clear proof of efficacy. The in-filtrated myocardium may be difficult to cardiovert.Nonetheless, anecdotal evidence supports theuse of implantable defibrillators in selected cases.

SUMMARY

ACMP occurs in the setting of genetic diseases,blood dyscrasias, chronic infections and inflamma-tion, and advanced age. Cardiologists are on thefront lines of diagnosis of ACMP when evaluatingpatients with unexplained dyspnea, CHF, orarrhythmias. Noninvasive detection of diastoliccardiac dysfunction andunexplained left ventricular

hypertrophy should be followed by biopsy todemonstrate the presence of amyloid depositsand appropriate genetic, biochemical, and immu-nologic testing to accurately define the type ofamyloid. Growing treatment options exist for thesediseases, and timely diagnosis and institution oftherapy is essential for preservation of cardiacfunction.

REFERENCES

1. Girnius S, Dember L, Doros G, et al. The changing

face of AA amyloidosis: a single centre experience.

Amyloid 2010;17(Suppl 1): 72.

2. Connors LH, Richardson AM, Theberge R, et al.

Tabulation of transthyretin (TTR) variants as of 1/1/

2000. Amyloid 2000;7:54.

3. Gustavsson A, Jahr H, Tobiassen R, et al. Amyloid

fibril composition and transthyretin gene structure

in senile systemic amyloidosis. Lab Invest 1995;

73:703.

4. Ruberg FL, Appelbaum E, Davidoff R, et al. Diag-

nostic and prognostic utility of cardiovascular

magnetic resonance imaging in light-chain cardiac

amyloidosis. Am J Cardiol 2009;103:544.

5. Hawkins PN, Aprile C, Capri G, et al. Scintigraphic

imaging and turnover studies with iodine-131

labeled serum amyloid P component in systemic

amyloidosis. Eur J Nucl Med 1998;25:701.

6. Libbey CA, Skinner M, Cohen AS. Use of abdominal

fat tissue aspirate in the diagnosis of systemic

amyloidosis. Arch Intern Med 1983;143:1549.

7. Dubrey SW, Cha K, Skinner M, et al. Familial and

primary (AL) cardiac amyloidosis: echocardiograph-

ically similar diseases with distinctly different clinical

outcomes [see comments]. Heart 1997;78:74.

8. Ng B, Connors LH, Davidoff R, et al. Senile systemic

amyloidosis presenting with heart failure: a compar-

ison with light chain-associated amyloidosis. Arch

Intern Med 2005;165:1425.

9. Brenner DA, Jain M, Pimentel DR, et al. Human amy-

loidogenic light chains directly impair cardiomyo-

cyte function through an increase in cellular

oxidant stress. Circ Res 2004;94:1008.

10. Liao R, Jain M, Teller P, et al. Infusion of light chains

from patients with cardiac amyloidosis causes dia-

stolic dysfunction in isolated mouse hearts. Circula-

tion 2001;104:1594.

11. Shi J, Guan J, Jiang B, et al. Amyloidogenic light

chains induce cardiomyocyte contractile dysfunc-

tion and apoptosis via a non-canonical p38alpha

MAPK pathway. Proc Natl Acad Sci U S A 2010;

107:4188.

12. Biolo A, Ramamurthy S, Connors LH, et al. Matrix

metalloproteinases and their tissue inhibitors in

cardiac amyloidosis: relationship to structural,

Amyloidotic Cardiomyopathy 393

functional myocardial changes and to light chain

amyloid deposition. Circ Heart Fail 2008;1:249.

13. Obici L, Perfetti V, Palladini G, et al. Clinical aspects

of systemic amyloid diseases. Biochim Biophys Acta

2005;1753:11.

14. Gertz MA, Comenzo R, Falk RH, et al. Definition of

organ involvement and treatment response in immu-

noglobulin light chain amyloidosis (AL): a consensus

opinion from the 10th International Symposium on

Amyloid and Amyloidosis, Tours, France, 18-22 April

2004. Am J Hematol 2005;79:319.

15. Palladini G, Campana C, Klersy C, et al. Serum

N-terminal pro-brain natriuretic peptide is a sensitive

marker of myocardial dysfunction in AL amyloidosis.

Circulation 2003;107:2440.

16. Dispenzieri A, Gertz MA, Kyle RA, et al. Serum

cardiac troponins and N-terminal pro-brain natri-

uretic peptide: a staging system for primary

systemic amyloidosis. J Clin Oncol 2004;22:3751.

17. Dispenzieri A, Kyle RA, Gertz MA, et al. Survival in

patients with primary systemic amyloidosis and

raised serum cardiac troponins. Lancet 2003;361:

1787.

18. Apridonidze T, Comenzo R, Hoffman J, et al.

Troponin as a prognostic marker in cardiac amyloid-

osis. Amyloid 2010;17:167a.

19. Gertz M, Merlini G. Definition of organ involvement

and response to treatment in AL amyloidosis: an up-

dated consensus opinion. Amyloid 2010;17:48a.

20. Jaccard A, Moreau P, Leblond V, et al. High-dose

melphalan versus melphalan plus dexamethasone

for AL amyloidosis. N Engl J Med 2007;357:1083.

21. Skinner M, Sanchorawala V, Seldin DC, et al. High-

dose melphalan and autologous stem-cell trans-

plantation in patients with AL amyloidosis: an

8-year study. Ann Intern Med 2004;140:85.

22. Goldsmith YB, Liu J, Chou J, et al. Frequencies

and types of arrhythmias in patients with systemic

light-chain amyloidosis with cardiac involvement

undergoing stem cell transplantation on telemetry

monitoring. Am J Cardiol 2009;104:990.

23. Meier-Ewert HK, Sanchorawala V, Berk J, et al.

Regression of cardiac wall thickness following

chemotherapy and stem-cell transplantation for AL

amyloidosis. Amyloid 2010;17(Suppl 1):150.

24. Palladini G, Perfetti V, Obici L, et al. Association of

melphalan and high-dose dexamethasone is effec-

tive and well tolerated in patients with AL (primary)

amyloidosis who are ineligible for stem cell trans-

plantation. Blood 2004;103:2936.

25. Specter R, Sanchorawala V, Seldin DC, et al. Kidney

dysfunction during lenalidomide treatment for AL

amyloidosis. Nephrol Dial Transplant 2011;26:881.

26. Nowis D, Maczewski M, Mackiewicz U, et al. Cardi-

otoxicity of the anticancer therapeutic agent borte-

zomib. Am J Pathol 2010;176:2658.

27. Connors LH, Prokaeva T, Lim A, et al. Cardiac

amyloidosis in African Americans: comparison of

clinical and laboratory features of transthyretin

V122I amyloidosis and immunoglobulin light chain

amyloidosis. Am Heart J 2009;158:607.

28. Dubrey SW, Davidoff R, Skinner M, et al. Progression

of ventricular wall thickening after liver transplanta-

tion for familial amyloidosis. Transplantation 1997;

64:74.

29. Liepnieks JJ, Benson MD. Progression of cardiac

amyloid deposition in hereditary transthyretin

amyloidosis patients after liver transplantation.

Amyloid 2007;14:277.

30. Berk JL, Keane J, Seldin DC, et al. Persistent pleural

effusions in primary systemic amyloidosis: etiology

and prognosis. Chest 2003;124:969.

31. Gertz MA, Falk RH, Skinner M, et al. Worsening of

congestive heart failure in amyloid heart disease

treated by calcium channel-blocking agents. Am J

Cardiol 1985;55:1645.

32. Gertz MA, Skinner M, Connors LH, et al. Selective

binding of nifedipine to amyloid fibrils. Am J Cardiol

1985;55:1646.

33. Pollak A, Falk RH. Left ventricular systolic dysfunc-

tion precipitated by verapamil in cardiac amyloid-

osis. Chest 1993;104:618.

34. Rubinow A, Skinner M, Cohen AS. Digoxin sensitivity

in amyloid cardiomyopathy. Circulation 1981;63:

1285.

35. Hosenpud JD, Uretsky BF, Griffith BP, et al.

Successful intermediate-term outcome for patients

with cardiac amyloidosis undergoing heart trans-

plantation: results of a multicenter survey. J Heart

Transplant 1990;9:346.

36. Hosenpud JD, DeMarco T, Frazier OH, et al.

Progression of systemic disease and reduced

long-term survival in patients with cardiac amyloid-

osis undergoing heart transplantation. Follow-up

results of a multicenter survey. Circulation 1991;84:

III338.

37. Dey BR, Chung SS, Spitzer TR, et al. Cardiac trans-

plantation followed by dose-intensive melphalan

and autologous stem-cell transplantation for light

chain amyloidosis and heart failure. Transplantation

2010;90:905.

38. Lacy MQ, Dispenzieri A, Hayman SR, et al. Autolo-

gous stem cell transplant after heart transplant for

light chain (AL) amyloid cardiomyopathy. J Heart

Lung Transplant 2008;27:823.

39. Modesto KM, Dispenzieri A, Cauduro SA, et al. Left

atrial myopathy in cardiac amyloidosis: implications

of novel echocardiographic techniques. Eur Heart J

2005;26:173.

40. Feng D, Syed IS, Martinez M, et al. Intracardiac

thrombosis and anticoagulation therapy in cardiac

amyloidosis. Circulation 2009;119:2490.