Author + information
- Received March 4, 2005
- Revision received May 18, 2005
- Accepted May 24, 2005
- Published online September 20, 2005.
- Enrica Perugini, MD⁎,
- Pier Luigi Guidalotti, MD†,
- Fabrizio Salvi, MD‡,
- Robin M.T. Cooke, MA⁎,
- Cinzia Pettinato, MD†,
- Letizia Riva, MD⁎,
- Ornella Leone, MD§,
- Mohsen Farsad, MD†,
- Paolo Ciliberti, MD⁎,
- Letizia Bacchi-Reggiani, MSc, MBiostat⁎,
- Francesco Fallani, MD⁎,
- Angelo Branzi, MD⁎ and
- Claudio Rapezzi, MD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Prof. Claudio Rapezzi, Istituto di Cardiologia, Policlinico S. Orsola-Malpighi, via Massarenti 9, 40138 Bologna, Italy.
Objectives We investigated the diagnostic accuracy of 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (99mTc-DPD) scintigraphy for differentiation of monoclonal immunoglobulin light-chain (AL) and transthyretin (TTR)-related cardiac amyloidosis.
Background Differential diagnosis between TTR-related and AL amyloidosis is often complex and time-consuming.
Methods Patients under routine observation with TTR-related/AL systemic amyloidosis and echocardiographic evidence of cardiac involvement were studied with 99mTc-DPD scintigraphy.
Results Patients with cardiac involvement of TTR-related (group A; n = 15) and AL (group B; n = 10) etiology were comparable for left ventricular mass and renal function. Heart and heart/whole-body tracer retention were significantly higher (p < 0.05) in group A as compared with group B and with 10 unaffected controls. At visual scoring, cardiac 99mTc-DPD uptake was present in all group A patients and absent in all group B patients; thus, using genotyping/immunohistochemistry as the reference technique, the accuracy of 99mTc-DPD scintigraphy for distinction of TTR-related and AL etiology was 100%. Cardiac 99mTc-DPD uptake was also absent among unaffected controls. Using echocardiography as the reference standard for recognition of cardiac involvement, sensitivity and specificity of scintigraphy were both 100% for group A patients; in group B, sensitivity was 0% and specificity was 100% (accuracy, 50%). Eleven patients with myocardial 99mTc-DPD uptake underwent 99mTc-methylene diphosphonate (99mTc-MDP) scintigraphy; all patients showed a 99mTc-MDP myocardial visual score of 0.
Conclusions Etiology is a third major cause—in addition to type of organ-involved (soft-tissue/heart) and tracer type—of scintigraphic variability in cardiac amyloidosis. This is a highly relevant consideration for future studies. We conclude that 99mTc-DPD scintigraphy is a useful step in the workup of the differential diagnosis of TTR versus AL etiology in patients with documented cardiac amyloidosis.
Cardiac involvement has major clinical and prognostic implications in amyloidosis. Acquired monoclonal immunoglobulin light-chain (AL) and transthyretin (TTR)-related (familial and wild-type/senile) disease are the most frequent causes of cardiac amyloidosis (1). In hereditary amyloidosis, cardiac involvement can limit short-term and long-term results of orthotopic liver transplantation (OLT) (2,3) and influence decisions to perform combined heart-liver transplantation (4). Severe cardiac involvement in AL amyloidosis can limit optimal hematologic treatments, including bone marrow transplantation (5).
Differential diagnosis between TTR-related and AL amyloidosis is often complex and time-consuming. Echocardiography is considered the gold standard for noninvasive diagnosis of amyloidotic cardiomyopathy (1,6), but it is not etiologically informative. A recent study showed a 10% misdiagnosis rate between familial and AL amyloidosis (7). Although genotyping and/or immunohistochemical analysis are mandatory for a definitive diagnosis, they are not routine examinations and thus require accurate patient selection (7,8).
Various tracers (mainly phosphonates) have been used in the past for scintigraphic assessment of cardiac amyloidosis, and they have produced heterogeneous results (9–22). These discrepancies might partially be explained by the variable composition of the amyloid substance (14,17). We hypothesized that this apparent limit might translate into a noninvasive tool for diagnostic evaluation of etiology. In this pilot study, we investigated the uptake of 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (99mTc-DPD) in a population of patients with TTR-related or AL cardiac amyloidosis. The main goal of the study was to assess the diagnostic accuracy and predictive value of 99mTc-DPD scintigraphy for differentiation of TTR-related and AL cardiac amyloidosis using genotyping/immunohistochemistry as a reference standard. Subsidiary goals were to explore patterns of myocardial and soft-tissue DPD uptake and relationships between scintigraphic and echocardiographic/histologic findings.
Study design and selection criteria
All patients under routine observation in our cardiology and neurology units between January 2004 and January 2005 with a diagnosis of TTR-related or AL amyloidosis with echocardiographically demonstrated cardiac involvement were enrolled in this study (no exclusion criteria were applied). External controls were 10 unaffected patients (6 men, 4 women; median age, 68 years; age range, 55 to 74 years) undergoing scintigraphy for oncologic reasons. All participants provided informed consent for participation. Ethical approval was not required for this noninvasive study.
Diagnosis of amyloidosis was defined by histologic documentation of Congo-red staining and apple-green birefringence under cross-polarized light in at least one involved organ (1). Cardiac involvement was echocardiographically defined as end-diastolic thickness of the interventricular septum >1.2 cm (in the absence of any other cause of ventricular hypertrophy) plus two or more of the following features: 1) homogeneous atrioventricular valve thickening; 2) atrial septal thickening; and 3) sparkling appearance of the ventricular septum (6). A clear-cut distinction between TTR-related and AL amyloidosis was based on genotyping and/or immunohistochemistry (7). Diagnosis of familial TTR-related amyloidosis was defined by a documented TTR mutation at DNA analysis; TTR-related wild-type (senile) amyloidosis by positive immunohistochemistry for TTR in the absence of any TTR mutation at DNA analysis (23). Monoclonal immunoglobulin light-chain amyloidosis was diagnosed by the presence of monoclonal plasma cells in the bone marrow, plus: 1) negative immunohistochemistry for TTR, and 2) absence of any TTR mutation at DNA analysis (7).
Genotyping/immunohistochemistry and histology
For genotyping, DNA was isolated from peripheral blood and exons 2, 3, and 4 of the TTRgene (accession number m11844) were amplified by polymerase chain reaction (Takara ExTaq polymerase, TaKaRa Shuzo, Otsu, Japan) by using primers previously described (24). Amplified DNA fragments were directly sequenced by using an ABI Prism 3100 automated sequencer (Applied Biosystems, Foster City, California).
Endomyocardial biopsies were microwave fixed/processed, and multiple 2-μm sections were tested for the presence of amyloid by Congo-red staining and apple-green birefringence under cross-polarized light microscopy. Amounts of amyloid deposits were semiquantitatively classified as mild (<30% involvement of the tissue fragments), moderate (30% to 60% involvement), or severe (>60% involvement) (25). Immunohistochemical analysis was performed by the Labeled StreptAvidin Biotin (LSAB) method using an antibody against TTR (R. P. Linke, Max Plank Institute of Biochemistry, Martinsried, Germany).
Scintigraphy protocol and image analysis
Scintigraphy was always performed ≤2 months after echocardiography and genotyping/immunohistochemistry. All patients received 740 MBq of 99mTc-DPD intravenously; a dual-head camera (ECAM, Siemens Medical Systems, Hoffman Estates, Illinois) equipped with low-energy, high-resolution collimators was used. Whole body scans were obtained five minutes (early) and three hours (late) after injection. In patients showing cardiac uptake, a myocardial single-photon emission computed tomography (SPECT) study was acquired after the late whole-body scan. Image analysis was independently performed by three experienced analysts (P.G., C.P., and M.F.) blinded to all patient data (any disagreement was settled by group discussion). Visual scoring of cardiac retention was routine (i.e., score 0, absent cardiac uptake and normal bone uptake; score 1, mild cardiac uptake, inferior to bone uptake; score 2, moderate cardiac uptake accompanied by attenuated bone uptake; score 3, strong cardiac uptake with mild/absent bone uptake). Semiquantitative analysis of heart retention, whole-body retention, and heart/whole-body ratio were evaluated from region of interest (ROI) drawings in the standard manner. In brief, on anterior images rectangular ROIs were drawn over the heart, and irregular ROIs over the kidneys and bladder. These ROIs were copied and mirrored on posterior images, and for each ROI geometric means of the two projections were calculated. All ROIs were corrected for background counts. Total counts in the images were considered as whole-body counts. Early whole-body counts were used to represent the injected activity. Whole-body retention was evaluated by comparing counts in the late images (corrected for decay and scan speed, and subtracting the activity in the urinary tract and bladder) with the counts in the early whole-body images. Heart retention was evaluated by comparing decay-corrected counts of the heart in late images with counts in early whole-body images. Heart/whole-body ratios were calculated by dividing counts in the heart by counts in late whole-body images. Soft tissue uptake (particularly skeletal muscle and splanchnic organs) was also visually assessed using an analogous scoring system to that used for the heart.
Patients showing cardiac uptake of 99mTc-DPD were invited to receive 99mTc-methylene diphosphonate (99mTc-MDP) scintigraphy using the same protocol and heart and soft-tissue visual scoring systems.
Statistical analysis was performed using SPSS 11.0 statistical software (SPSS Inc., Chicago, Illinois). Continuous variables are expressed as median (interquartile range), categorical variables as percentages. When comparing subgroups, the Fisher exact test or contingency tables were used (the former when the expected frequency of counts in at least one cell of 2 × 2 tables was <5), alongside the Mann-Whitney U or the Kruskal-Wallis test. For multiple comparisons, the Dunn test was also applied. Any p values <0.05 were considered statistically significant.
Twenty-five cardiac amyloidosis patients were enrolled, all of whom completed the study protocol. Table 1summarizes the principal clinical, electrocardiographic, and echocardiographic findings in the patient population, subdivided into two groups according to etiology. Of the 15 patients with TTR-related amyloidosis with cardiac involvement (group A), 10 had familial TTR amyloidosis (carrying Glu89Gln [n = 1], Val30Met [n = 2], Ala36Pro [n = 2], Thr49Ala [n = 2], Phe33Val [n = 1], or Ile68Leu [n = 2] TTR mutations), and 5 had senile TTR. No difference was apparent regarding age, renal function, and left ventricular (LV) mass (all p > 0.40). None of the patients had any associated cardiovascular disease, including systemic hypertension and coronary artery disease (the latter was excluded on clinical grounds, as well as by coronary angiography in three patients with suspected angina).
Figure 1shows representative examples of the spectrum of 99mTc-DPD uptake among patients and control patients. Semiquantitative and visual scintigraphic findings are summarized in Figure 2and Table 2,according to the two subgroups of the patient population and the unaffected controls. At semiquantitative analysis, heart tracer retention and heart/whole-body retention ratio were both about two-fold higher in group A as compared with group B and also with controls (Fig. 2). Smaller differences were also observed between group B and controls, plausibly because of intense mediastinal/sternal uptake in a single patient in group B (see asterisk in Fig. 2A). Regarding the visual score, myocardial uptake (score >0) was detected in 15 of 15 (100%) group A patients, as compared with 0 of 10 (0%) in group B and 0 of 10 (0%) in the control group.
We were able to compare the macroscopic, histologic, and 99mTc-DPD scintigraphic appearance of myocardium from one group A patient who underwent combined heart–liver transplantation (one month after scintigraphy) and one group B patient who died (after three months). Myocardial uptake of 99mTc-DPD occurred only in the patient with TTR-related amyloidosis, despite similar LV mass and myocardial infiltration (Fig. 3).
Eleven patients with myocardial uptake of 99mTc-DPD agreed to undergo 99mTc-MDP scintigraphy. The visual score for 99mTc-MDP myocardial uptake was 0 in all of these patients. A representative example of differences in myocardial uptake of the two tracers can be seen in Figure 3. The visual score for 99mTc-MDP soft tissue uptake was 0 in 8 patients and 1 in the remaining 3.
Diagnostic accuracy of 99mTc-DPD scintigraphy for differentiation of TTR-related and al cardiac amyloidosis
We assessed the diagnostic accuracy of 99mTc-DPD scintigraphy (visual score >0) for differentiation of TTR-related and AL cardiac amyloidosis using genotyping/immunohistochemistry as the reference technique. Within our subpopulation of amyloidosis patients with echocardiographic evidence of cardiac involvement (groups A and B), sensitivity and specificity of 99mTc-DPD scintigraphy were both 100%.
Diagnostic accuracy of 99mTc-DPD scintigraphy for detection of amyloidotic cardiomyopathy within different etiologic settings
We also considered the diagnostic accuracy of a scintigraphy visual score >0 between each of the two groups of patients and the control patients for detection of cardiac amyloidosis using echocardiography as the reference standard. Sensitivity and specificity of scintigraphy were both 100% for group A (TTR-related patients). By contrast, for group B (AL patients), sensitivity was 0% and specificity was 100%. Thus, the overall accuracy for detection of amyloid of both types was only 50%. These findings further illustrate the concept that etiology can be a major cause of variability in the diagnostic performance of 99mTc-DPD scintigraphy.
Patterns of myocardial uptake of 99mTc-DPD
All but one of the group A patients had widespread involvement of all myocardial segments (Fig. 4A),whereas the remaining patient showed localized septal accumulation (Fig. 4B). These observations paralleled the patterns of increased wall thickness visible at echocardiography (Fig. 4).
99mTc-DPD scintigraphy findings in soft tissue
Regarding extracardiac uptake of 99mTc-DPD, at semiquantitative analysis whole-body retention was significantly higher in both groups A and B with respect to controls (Table 2, Fig. 2C). The increased uptake of 99mTc-DPD was visually detectable at the muscular and/or splanchnic level in 13 of 15 group A patients and 5 of 5 group B patients. No difference was apparent between groups A and B. The visual score for 99mTc-MDP soft tissue uptake was 0 in eight patients and 1 in the remaining three.
Relationships between scintigraphic and echocardiographic/histologic findings
We investigated whether the degree of myocardial 99mTc-DPD uptake as assessed by visual score could be related to LV mass and/or interventricular septal thickness. No evidence of relationships with LV mass or interventricular septal thickness emerged (p = 0.54 and p = 0.43, respectively, using Kruskal-Wallis test). As can be seen from Figure 5,the vast majority of patients had a similar spectrum of LV masses irrespective of visual score at scintigraphy.
In myocardial biopsies (available from 8 of 15 group A patients and all 10 group B patients), the degree of amyloidotic infiltration was similar in patients with or without myocardial 99mTc-DPD uptake (severe infiltration in 5 of 8 vs. 7 of 10 patients, respectively; p = 0.87).
To our knowledge, this is the first assessment of 99mTc-DPD scintigraphy as a potential diagnostic tool for assessment of etiology in cardiac amyloidosis. Our findings could have potential clinical implications (especially in the light of recent therapeutic advances), and they also could help clarify the conflicting data in published reports on the use of bone tracers in the scintigraphic evaluation of amyloidosis.
Noninvasive detection of amyloid has been investigated using three main groups of radiolabeled tracers: 123I-labeled serum amyloid P protein (26,27); aprotinin (28,29); and bone tracers, mainly 99mTc-pyrophosphate (9–15) and 99mTc-HDP (20,21). Labeled serum amyloid P protein localizes rapidly and specifically to amyloid deposits (in proportion to the amount of amyloid) and persists there, allowing quantitative monitoring of amyloid deposition and regression during therapy (low quality of heart visualization and limited availability are current limitations of this tracer [12,13]). 99mTc-aprotinin (a protease inhibitor) has been mainly studied in AL patients, in whom it accumulates in various organs, including myocardium: available experience is limited and quantitative assessment is not feasible (14,15). Data on bone tracers are numerous but heterogeneous. 99mTc-pyrophosphate is the most widely studied: because it binds avidly to many types of amyloid, it was a popular imaging agent for cardiac amyloidosis during the 1980s. However, amyloidotic uptake of this tracer has not been quantitated, and it has lower diagnostic sensitivity than echocardiography (2–8,10).
Relatively few studies are available on the diagnostic performance of 99mTc-DPD (16,17), 99mTc-methylene diphosphonate, or 99mTc-hydroxymethylene-diphosphonate in amyloidosis (18–20). The numbers of patients included in these studies were often low, and the type of amyloidosis was frequently unspecified. None of these studies used immunohistochemistry/genotyping to exclude misdiagnosis between TTR and AL forms. Both positive and negative myocardial uptake of bone tracers has been reported in patients with definite cardiac amyloidosis, generating a confusing scenario.
In the present study, abnormal cardiac uptake of 99mTc-DPD (whether at visual scoring or semiquantitative analysis) occurred exclusively in patients with familial or senile TTR-related cardiac amyloidosis, despite quite similar signs of amyloidotic myocardial infiltration in the two groups, as expressed by comparable values of interventricular septal thickness and LV mass. The strength of these findings was reinforced by the use of immunohistochemistry/genotyping for differentiation between AL and TTR-related amyloidosis. The data are broadly in line with the single published study of 99mTc-DPD scintigraphy in a small series of patients in familial TTR-related cardiac amyloidosis (17). In the only patient with AL amyloidosis reported in the literature to have myocardial uptake of 99mTc-DPD, the etiologic diagnosis was based only on the presence of abnormal protein at serum and urine electrophoresis (a potentially misleading finding) (16).
In our study sample, selective myocardial uptake of 99mTc-DPD provided 100% accuracy for differentiation of TTR-related and AL etiology of cardiac amyloidosis. This potential diagnostic application is confined to amyloidotic patients with myocardial involvement and cannot be extended to patients with involvement limited to soft tissues and/or splanchnic organs. Differential diagnosis of TTR-related and AL cardiac amyloidosis is often problematic, especially from the clinical standpoint. Electrocardiographic and echocardiographic findings are superimposable, irrespective of the composition of the deposits (30,31). Clinically, AL cardiac amyloidosis more often affects older patients and is associated with a more rapid course with prominent symptoms of heart failure. However, exceptions are frequent. For example, patients carrying the TTR Glu89Gln mutation (4) can experience an accelerated course strongly influenced by cardiologic problems. Furthermore, the first clinical manifestations of amyloidosis can be cardiologic with a variety of other mutations. A relatively frequent diagnostic pitfall occurs in elderly patients with monoclonal gammopathy and echocardiographic signs of cardiac amyloidosis. Because of the high prevalence of subtle monoclonal gammopathies in the general population, it is not possible automatically to assume a diagnosis of AL cardiac amyloidosis in such cases (7) (as in one of our patients with TTR-related cardiac amyloidosis casually associated with a monoclonal gammopathy). Despite these obstacles, a correct differential diagnosis of TTR-related and AL cardiac amyloidosis is now mandatory in the light of the availability of highly specific therapeutic options: OLT or combined heart-liver transplantation for familial TTR (2–4); high-dose chemotherapy and bone marrow transplantation for AL amyloidosis (5). Moreover, definite diagnosis of a hereditary TTR form enables appropriate genetic counseling.
Regarding our wider understanding of the conflicting scintigraphic data available on amyloidosis, the present study offers a new interpretative key. Our analysis reveals that the etiology of the amyloidosis is a third major variable, in addition to the type of organ involved (soft tissues versus heart) and bone tracer type. Etiologic variability must be taken into account in future studies of scintigraphy in cardiac amyloidosis.
What possible explanations are there for selective myocardial uptake of 99mTc-DPD in TTR-related cardiac amyloidosis? Analysis of the echocardiographic and histologic characteristics of the two groups largely excludes the possibility of different levels of myocardial infiltration (Figs. 3 and 5). The affinity of all bone tracers for calcium determines their uptake by bone (and especially by bone metastases and post-traumatic lesions), tumors, infected/inflammatory tissues, and some neoplastic lesions (18). It has been hypothesized that uptake of phosphonates in amyloidotic tissues might be explained by high calcium concentration in amyloid deposits (10,32,33). A model capable of fully explaining our findings would have to account for all three forms of variability (etiology, type of organ involved, tracer type), and in particular why 99mTc-DPD uptake is massive in myocardium of patients with TTR-related amyloidosis, absent in myocardium of AL patients, and present in soft tissues of both groups. No such model currently exists. Onset of amyloidosis is a very complex process, eventually leading to tissue deposition of amyloid fibrils (34). Although amyloid fibrils share a common chemical structure (i.e., serum amyloid protein) irrespective of etiology, some characteristics vary between specific etiologies (allowing differential immunostaining). We can only speculate that 99mTc-DPD might bind to specific fragments of the mutated TTR protein or of the wild-type protein (present in the senile form), and that this phenomenon occurs predominantly in the myocardium either because of the large amount of amyloid or some unknown features of its spatial arrangement at this level.
It should be noted that in our study sample, 99mTc-MDP, unlike 99mTc-DPD, did not show selective myocardial uptake in 11 patients with TTR-related cardiac amyloidosis. This finding agrees with one reported case (9), but contrasts with a further two (19,20) with an apparently definite diagnosis of TTR-related cardiac amyloidosis.
Our findings have potential clinical implications in a clinical setting, where misdiagnosis rates between hereditary TTR and AL amyloidosis can be as high as 7% (7) and where the repercussions can be disastrous (chemotherapy with/without bone marrow transplantation instead of OLT). In patients with diagnosis of cardiac amyloidosis associated with monoclonal gammopathy, negative 99mTc-DPD scintigraphy would reinforce a diagnostic hypothesis of AL etiology, whereas a positive test result would direct the diagnostic workup toward DNA analysis and immunohistochemistry. Also, 99mTc-DPD scintigraphy could be used as an orientative diagnostic step in cases of suspected senile (wild-type TTR) cardiac amyloidosis.
Considering the restricted number of patients (especially in the AL cardiac amyloidosis subgroup), independent confirmation is required on a larger study sample. Our tertiary center included cardiology/neurology units particularly active in the field of amyloidosis, and especially familial amyloidotic polyneuropathy. Therefore, the results cannot automatically be extended to populations with a lower prevalence of the disease. It should be noted that our study does not provide information on the rarer hereditary forms of amyloidosis related to mutations of apolipoprotein A, fibrinogen, lisozyme, or the acquired AA type.
Etiology is an important determinant of 99mTc-DPD scintigraphic variability in cardiac amyloidosis, a highly relevant consideration for future studies. Our analysis indicates that 99mTc-DPD scintigraphy could offer a useful first step in the workup of differential diagnosis of TTR versus AL etiology in patients with documented cardiac amyloidosis.
- Abbreviations and Acronyms
- monoclonal immunoglobulin light-chain
- cardiac amyloidosis
- left ventricle/ventricular
- orthotopic liver transplantation
- region of interest
- single-photon emission computed tomography
- 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid
- 99mTc-methylene diphosphonate
- Received March 4, 2005.
- Revision received May 18, 2005.
- Accepted May 24, 2005.
- American College of Cardiology Foundation
- Eriksson P.,
- Backman C.,
- Bjerle P.,
- Eriksson A.,
- Holm S.,
- Olofsson B.
- Worsley D.F.,
- Lentle B.C.
- Singer R.,
- Schnabel A.,
- Strasser R.H.
- Ferlini A.,
- Fini S.,
- Salvi F.,
- Patrosso M.C.,
- Vezzoni P.,
- Forabosco A.
- Schaadt B.K.,
- Hendel H.W.,
- Gimsing P.,
- Jonsson V.,
- Pedersen H.,
- Hesse B.
- Dubrey S.W.,
- Cha K.,
- Skinner M.,
- LaValley M.,
- Falk R.H.