Author + information
- Received November 23, 2015
- Revision received April 3, 2016
- Accepted April 12, 2016
- Published online July 5, 2016.
- Arnt V. Kristen, MDa,∗ (, )
- Eva Brokbals, MSa,
- Fabian aus dem Siepen, MDa,
- Ralf Bauer, MDa,
- Selina Hein, MDa,
- Matthias Aurich, MDa,
- Johannes Riffel, MDa,
- Hans-Michael Behrens, MScb,
- Sandra Krüger, BScb,
- Peter Schirmacher, MDc,
- Hugo A. Katus, MDa,d and
- Christoph Röcken, PhDb
- aDepartment of Cardiology, Angiology, and Respiratory Medicine, University of Heidelberg, Heidelberg, Germany
- bInstitute of Pathology, Christian-Albrechts-University, Kiel, Germany
- cInstitute of Pathology, University of Heidelberg, Heidelberg, Germany
- dDZHK (German Center for Cardiovascular Research), Site Heidelberg/Mannheim, Heidelberg, Germany
- ↵∗Reprint requests and correspondence:
Dr. Arnt V. Kristen, Department of Cardiology, Angiology, and Respiratory Medicine, University Hospital Heidelberg, Im Neuenheimer Feld 410, D-69120 Heidelberg, Germany.
Background Cardiac amyloid load has not been analyzed for its effect on mortality in patients with amyloid light-chain (AL) cardiac amyloidosis.
Objectives This study retrospectively compared histological amyloid load with common clinical predictors of mortality.
Methods This study assessed 216 patients with histologically confirmed cardiac amyloidosis at a single center with electrocardiography, echocardiography, and laboratory testing.
Results AL amyloid deposits were usually distributed in a reticular/pericellular pattern, whereas transthyretin amyloid (ATTR) more commonly showed patchy deposits. Median amyloid load was 30.5%; no amyloid load was above 70%. During follow-up (median 19.1 months), 112 patients died. Chemotherapy had a significant effect on overall survival in AL amyloidosis (16.2 months vs. 1.4 months; p = 0.003). Patients with <20% AL amyloid load who responded to chemotherapy showed significantly better survival than nonresponders. According to univariate analysis, predictors of survival in AL amyloidosis included sex, Karnofsky index, New York Heart Association (NYHA) functional class, diastolic blood pressure, estimated glomerular filtration rate, N-terminal pro–B-type natriuretic peptide, mineralocorticoid receptor antagonists, low voltage, ineligibility for chemotherapy, response to chemotherapy, and amyloid load. Independent predictors of mortality by multivariate analysis included NYHA functional class (III vs. II), estimated glomerular filtration rate, responders to chemotherapy, and amyloid load. In ATTR amyloidosis, survival correlated with NYHA functional class, diastolic blood pressure, and use of diuretic agents. Following Cox regression analysis, NYHA functional class (III vs. II; p < 0.05) remained the only independent predictor of patient survival in ATTR amyloidosis.
Conclusions Early identification of subjects with AL amyloid is essential given that in late-stage disease with extensive amyloid load, our data suggested that outcomes are not affected by administration of chemotherapy.
Amyloidosis constitutes a group of diseases with different etiologies, characterized by extracellular deposition of proteins oriented in a β-sheet structure. A total of 31 autologous physiological proteins that can form amyloid, each linked to a unique etiology, have been identified (1). Depending on the anatomical distribution and extent of amyloid formation, it can disrupt normal tissue architecture and function, leading to progressive organ failure and eventually death. The most clinically relevant forms are immunoglobulin-derived light-chain (AL) and transthyretin-derived (ATTR) amyloidosis (2). Clinical presentation of AL amyloidosis is multifactorial, involving almost every organ with predominance of kidney and heart involvement (3). ATTR amyloidosis can occur in a hereditary form due to a point mutation in the TTR gene or as a wild-type variant; the latter almost exclusively affects elderly men. In general, ATTR amyloidosis is characterized by senso-motoric polyneuropathy and/or cardiomyopathy (4–6). Median survival is about 4 months in patients with AL amyloidosis and manifest systolic left ventricular (LV) heart failure and 3 to 6 years in ATTR amyloidosis (7).
In view of the clinical and prognostic significance of cardiac amyloidosis, several noninvasive diagnostic tools have been reported for predicting AL amyloidosis mortality, including electrocardiography (ECG) and echocardiography (8–10). In recent years, cardiac magnetic resonance (CMR) imaging has increasingly been used for tissue characterization using late gadolinium enhancement (LGE) or T1 mapping (11). Skeletal scintigraphy also has demonstrated high sensitivity and specificity for diagnosing ATTR amyloidosis (12,13). Cardiac biomarkers such as troponin T and N-terminal pro–B-type natriuretic peptide (NT-proBNP) provide potent prognostic information in AL amyloidosis (3,14,15).
However, prior to clinical and serological tests, a sound diagnosis and classification of cardiac amyloidosis is mandatory. Clinical management is subtype-dependent and may require, for instance, chemotherapy and autologous bone marrow transplantation in AL amyloidosis or liver transplantation in ATTR amyloidosis. Endomyocardial biopsy (EMB) is often crucial for the differential diagnosis of hypertrophic cardiomyopathies. Despite the importance of EMB and thorough clinical assessment of patients experiencing cardiac amyloidosis, few studies have compared cardiac histology with clinical data (16). In this study, we carried out a comprehensive retrospective study testing the following hypotheses: 1) AL and ATTR amyloid show unique distribution patterns in the heart; 2) AL and ATTR amyloid show distinct clinical presentations; 3) amyloid load in tissue specimens correlates with clinico-pathological patient characteristics and (cardiac) disease severity; and 4) amyloid load is a prognostic and/or predictive biomarker.
This project was approved by the local ethics committee of the University Hospital in Heidelberg conforming to the Declaration of Helsinki.
Patient cohort and clinical assessment
From the Heidelberg Amyloidosis Center, we identified all patients with cardiac AL and ATTR amyloidosis who had undergone EMB and a detailed clinical assessment and were studied histologically by the Amyloid Registry Kiel. From 2004 to 2015, we identified a total of 216 patients.
Screening for monoclonal gammopathy by serum/urine immunofixation electrophoresis and the serum free light-chain test was carried out routinely (Binding Site, Schwetzingen, Germany). Patient records were analyzed for time of diagnosis and disease-specific parameters, including organ involvement, standard blood tests, 12-lead ECG, and echocardiography. Cardiac troponin T was measured by either a fourth-generation assay or a high-sensitivity assay (Roche Diagnostics, Mannheim, Germany). Due to lack of comparability of the different troponin assays, abnormal values were defined as >0.03 μg/l (fourth-generation assay) or >14 pg/ml (high-sensitivity assay). NT-proBNP was measured using Elecsys proBNP (Roche Diagnostics). In patients with AL amyloidosis, response to treatment was defined according to international consensus criteria (17,18).
ECGs were analyzed for a low-voltage pattern, defined as QRS complex deflection <0.5 mV in any limb leads or the sum of the S-wave deflection in V1–2 and R-wave deflection in V5–6 <1.5 mV. Transthoracic echocardiograms were analyzed for surrogate markers of cardiac amyloidosis, including left atrial diameter, diastolic interventricular septum thickness, diastolic posterior wall thickness, LV end-diastolic diameter, or LV end-systolic diameter. LV end-diastolic volume, LV end-systolic volume, stroke volume, ejection fraction (EF), and LV mass were calculated as reported previously (19). LVEF <45% was considered markedly impaired. Mitral annular systolic velocity was measured by pulsed-wave Doppler tissue imaging with sample volume on the lateral mitral annulus in the apical 4-chamber view. LV myocardial volume was defined as LV mass divided by the mean density of myocardium (1.05 g/ml). Myocardial contraction fraction was calculated as stroke volume divided by myocardial volume (20).
Formalin-fixed and paraffin-embedded tissue samples were used throughout this study. Amyloid was detected by Congo red staining, viewed under polarized light, showing green-yellow-orange birefringence. Immunohistochemical classification of amyloid was carried out as described elsewhere (21–23). Additional information on description of deoxyribonucleic acid extraction is in the Online Appendix.
Quantification of amyloid load
Immunohistochemically stained slides were scanned using a Leica SCN400 whole slide scanner (Leica Biosystems, Nussloch, Germany) with 40× magnification. The slide images were exported as overview and with 9× magnification (pixel width of about 1.2 μm). The percentage of the amyloid area was evaluated using ImageJ version 1.47v (U.S. National Institutes of Health, Bethesda, Maryland) by counting immunohistochemically stained and nonstained pixels. Artifacts surrounding the specimen within the original image were removed manually using photo editing software. Processing in Image J was done using the “color threshold” function to filter pixels on the basis of ranges of hue, saturation, and brightness values in the HSB color model. Background was detected by filtering pixels of high brightness and low saturation. The specimen’s pixel count was calculated by subtracting the background pixel count from the total pixel count. Stained areas (red pixels) were detected by filtering the corresponding range of hue values in combination with a lower threshold of saturation. Threshold values were adjusted individually under supervision of a pathologist (C.R.) to compensate for variations of the staining process.
Statistical analyses were carried out with SPSS version 20 (IBM Corporation, Armonk, New York) and R Version 3.2.0 (R Foundation for Statistical Computing, Vienna, Austria). Continuous data were expressed as median (interquartile range [IQR]). Categorical variables were expressed as absolute numbers (percentages). Amyloid load was divided into 3 groups (<20%, 20% to 40%, and >40%). To test for significant differences of continuous variables between amyloid types and between amyloid load groups, we applied the Kruskal-Wallis test and Mann-Whitney U test. For categorical clinical variables, we applied Fisher exact test, and for pairwise comparisons, the 2-sample test for equality of proportions with continuity correction using R. Overall survival, defined as the time between EMB and death, was analyzed using the Kaplan-Meier method. Differences of median overall survival were assessed using the log-rank test. We carried out univariate Cox regression separately for all continuous and categorical clinical variables. All variables having p ≤ 0.1 in univariate Cox regression were reported and included in a multivariate Cox regression to identify independent prognostic variables. Multivariate results are shown after backward-LR method using p = 0.05 as the exclusion limit. All p values were taken from 2-tailed tests. We assumed a significance level of p < 0.05.
To account for false discoveries due to multiple testing, we applied the Simes (Benjamini-Hochberg) procedure per test group (i.e., separately for comparisons with amyloid type, amyloid load, and survival tests) (24). All p values were given unadjusted, but were marked where they lost significance after Simes' multiple testing procedure.
The characteristics of the EMB cohort are summarized in Table 1. In total, 81 biopsies were obtained from the left and 73 from the right ventricular myocardium (unknown in 62 patients). AL amyloidosis was diagnosed in 107 patients (ALλ n = 91 [85%] and ALκ n = 16 [15%]) and ATTR in 109 patients. Molecular-genetic testing showed TTR-wild type (wt-ATTR) in 76 patients and mutant TTR (mt-ATTR) in 33 patients, including p.Val50Met (n = 9); p.Val40Ile (n = 6); p.Cys30Arg (n = 2); p.Glu74Lys (n = 2); p.Val142Ile (n = 2); p.Thr79Lys (n = 2); and p.Asp38Glu, p.Ala39Asp, p.His51Asn, p.Gly67Glu, p.Ser70Arg, p.Thr80Ala, p.Val114Ala, p.Thr126Asn, p.Ile127Val, and unknown (each n = 1), respectively. No correlation was found between biopsy site and amyloid type. No sex-based differences were observed, except for higher prevalence of males with wt-ATTR.
Clinical presentation of cardiac amyloidosis
The clinical, ECG, and echocardiographic findings showed several significant differences between patients experiencing AL, wt-ATTR, and mt-ATTR amyloidosis (Table 1). These differences can be categorized into amyloid-type specific differences (AL vs. ATTR), putative age-dependent differences (mt-ATTR vs. wt-ATTR), and differences probably related to both amyloid type and patient age (AL vs. wt-ATTR).
Amyloid-type specific differences (AL vs. ATTR) were found with regard to NT-proBNP serum levels (significantly higher in AL) and LV mass (significantly lower in AL) (Table 1). These 2 characteristics showed no differences between wt- and mt-ATTR amyloid.
Differences probably related to patient age (mt-ATTR vs. wt-ATTR) included estimated glomerular filtration rate (eGFR) and medication use, with patients with wt-ATTR showing a significantly lower eGFR and more frequently taking beta-blocker or angiotensin-converting enzyme inhibitors/angiotensin-1 receptor antagonist therapy. Additionally, disease-specific characteristics of patients with wt-ATTR included a higher age at diagnosis, more commonly being male, and an insignificantly higher prevalence of atrial fibrillation in wt-ATTR when compared with mt-ATTR and AL amyloidosis (Table 1).
Finally, many differences could not be related to either age (no significant difference between mt- and wt-ATTR) or amyloid type (no significant difference between AL vs. mt-ATTR) and may be related to a mixture of both and included Karnofsky index, diastolic blood pressure, heart rate, PQ interval, QRS duration, low voltage pattern, thickness of the septal wall, LV end-diastolic volume, and pericardial effusion (Table 1).
In total, 94 (88%) patients with AL amyloidosis received chemotherapy. The clinico-pathological patient characteristics did not significantly differ between responders (complete response, n = 8; very good partial response, n = 14; partial response, n = 28) and nonresponders (n = 14). Clinical variables did not differ between responders and nonresponders when divided according to amyloid load (<20% vs. 20% to 40% vs. >40%). In 30 patients, early death prevented assessment of hematological response status.
Amyloid was found in the interstitium and vessel walls with distinct distribution patterns in AL and ATTR amyloidosis (25). AL amyloid deposits were usually more evenly distributed in a reticular/pericellular pattern around myocytes, and the width of the deposits increased with progressive amounts of AL amyloid. Cardiac ATTR amyloid was patchy, with the number and size of individual deposits varying depending on the overall amount of ATTR amyloid (Figure 1). Progressive amyloid deposition was accompanied by an increasing loss of cardiomyocytes.
The median amyloid load was 30.5% (IQR: 18.3% to 42.1%) without any significant difference between biopsy sites. Most patients had an amyloid load of 30% to 40%, with none >60% (AL amyloid) or >70% (wt-ATTR amyloid) (Online Figures 1 and 2). The amyloid load was higher in wt- and mt-ATTR amyloidosis compared with AL amyloidosis (Table 1). However, this difference was not statistically significant. No difference was found in the median amyloid load between mt- and wt-ATTR (Online Figure 3).
Correlating amyloid load with clinico-pathological parameters showed an increase of NT-proBNP in ATTR amyloidosis. Patients with <20% amyloid load harbored a median serum NT-proBNP of 2,579 ng/ml (IQR: 1,274 to 7,249 ng/ml); a 20% to 40% load, 2,742 ng/ml (IQR: 1,201 to 5,280 ng/ml); and a >40% load, 4,832 ng/ml (IQR: 2,701 to 10,387 ng/ml) (<20% vs. 20% to 40%: p = 0.494; <20% vs. >40%: p = 0.043; 20 to 40% vs. >40%: p = 0.002). However, this difference lost statistical significance after multiple testing adjustments. Detailed correlations of amyloid load with clinico-pathological parameters according to amyloid type are shown in Online Table 1.
Survival data were available for 213 patients. During follow-up (median 19.1 months; range 2.5 to 145.7 months), 112 (52.6%) patients died (AL n = 73 [68.2%]; mt-ATTR n = 15 [48.4%]; wt-ATTR n = 24 [32.0%]). Mortality was significantly lower in wt-ATTR amyloidosis compared with AL amyloidosis. In total, 94 patients received chemotherapy (autologous stem cell transplantation n = 16). Additional first-line treatment regimens were melphalan/dexamethasone (n = 27), bortezomib/dexamethasone (n = 42), and melphalan/prednisolone (n = 9). Patients deemed ineligible for any type of chemotherapy were excluded from survival analyses to avoid bias. Six patients with mt-ATTR received a liver transplant. Chemotherapy had a significant effect on overall survival in AL amyloidosis: 16.2 months (95% confidence interval [CI]: 11.3 to 21.1 months) versus 1.4 months (95% confidence interval: 0 to 31 months) (p = 0.003). No difference was found in mt-ATTR amyloidosis after liver transplantation (Online Figure 4).
Patients with AL amyloidosis who had <20% (median 12.5 months [95% CI: 3.5 to 21.4 months]) and >40% amyloid load (median 6.7 months [95% CI: 0 to 15.1 months]) had shorter survival than patients harboring a 20% to 40% load (median 24.5 months [95% CI: 0.6 to 48.4 months]) (p = 0.024) (Figure 2). Survival depended significantly on chemotherapy (data not shown). Responders (complete response, very good partial response, and partial response post-chemotherapy) and nonresponders (no response, progressive disease) with <20% amyloid showed a significant difference in overall survival (Central Illustration). No difference was found between responders and nonresponders if amyloid load was 20% to 40% or >40% (Central Illustration).
Univariate and multivariate analysis
Finally, we correlated patient survival with all clinico-pathological patient characteristics. In univariate analyses, survival of patients with AL amyloidosis correlated with sex, Karnofsky index, New York Heart Association (NYHA) functional class, diastolic blood pressure, eGFR, NT-proBNP, medication with mineralocorticoid receptor antagonists, low voltage pattern, ineligibility for chemotherapy, and amyloid load. All variables having p < 0.1 in univariate analyses (Table 2) were included in a multivariate Cox regression, which showed NYHA functional class, eGFR, amyloid load, and response to chemotherapy to be independent predictors of patient survival (Table 2). A separate analysis excluding 13 patients ineligible for any chemotherapy revealed similar results (Table 2).
In ATTR amyloidosis (wt-ATTR, mt-ATTR), patient survival correlated with NYHA functional class, diastolic blood pressure, and medication with diuretic agents. Karnofsky index, troponin T, QRS duration, LVEF, and impaired right ventricular function showed p values <0.1 and were included in the subsequent multivariate Cox regression analysis, which rendered NYHA functional class as the only independent predictor of patient survival (Table 2).
The present data demonstrate that EMBs revealed important information beyond histological diagnosis in AL amyloidosis. Quantification of amyloid load at diagnosis indicated survival benefit of patients with an amyloid load <20%. In contrast, no survival benefit for future chemotherapy was observed if amyloid load was ≥20% even with hematological response. In general, cardiac manifestation is common in AL and ATTR amyloidosis (26). More than one-half of the patients with AL amyloidosis present with cardiac symptoms at diagnosis. Our study confirmed some distinct differences between AL and ATTR amyloidosis, including lower LV mass but higher NT-proBNP values in AL amyloidosis (7), most likely related to toxic effects of amyloidogenic light chains causing apoptosis of cardiomyocytes (27). Differences observed between patients with mt- and wt-ATTR, including eGFR and medication use, were probably related to patient age. Additionally, disease-specific characteristics of wt-ATTR included higher age at diagnosis, more commonly male sex, and higher prevalence of atrial fibrillation.
Histological diagnosis and differentiation of cardiac amyloidosis
Despite diagnostic improvements, including CMR imaging, echocardiography with speckle tracking, and skeletal scintigraphy (10,28,29), EMB remains a crucial step in the diagnostic work-up of unexplained cardiac hypertrophy and the gold standard for diagnosing cardiac amyloidosis (30). Moreover, distinct patterns of amyloid aid in the differentiation of AL (reticular, pericellular) and ATTR (patchy) amyloid, to some extent even without immunohistochemical staining. However, at least in the setting of ATTR subjects who have a concomitant plasma cell dyscrasia, a misdiagnosis rate of 10% was reported despite application of immunohistochemistry (31). In our center, combined use of clinical examination and specialized pathology yielded a high accuracy for correct amyloid typing (22).
Due to potential lack of therapeutic implications as well as the latent risk of pericardial effusion, physicians often avoid performing cardiac biopsies despite strong suspicion of amyloidosis. However, due to improvement of biopsy forceps, LV EMBs can be performed easily and safely (32). In our center, no fatal complication occurred after restriction to LV biopsies.
All primary antibodies used in the present study were highly specific to detect a single amyloid precursor protein (21,33). Digital conversion of immunohistochemically stained areas appears to represent the extent of amyloid deposited in this particular visual field. However, whether the amyloid burden obtained from a visual field of an EMB represents whole cardiac amyloid load precisely remains debatable. A comparable software-based method for quantification of amyloid deposits has been reported previously in whole hearts obtained from heart transplantation or autopsies (34). Unfortunately, we were unable to compare biopsy and autopsy results due to lack of autopsy cases. Generally, a diffuse, predominantly subendocardial LGE was indicative of cardiac amyloid (28,35,36). LGE was associated with amyloid deposition in explanted hearts (37), and diverse patterns of amyloid were reported in explanted and autopsy hearts, including diffuse, segmental, and subendocardial infiltration (34). In this series, the extent of amyloid infiltration of the right ventricle was higher in AL amyloidosis than in ATTR amyloidosis. No differences between AL and ATTR amyloidosis were reported regarding LV amyloid deposition (34).
Amyloid load and severity of cardiac manifestation
In the present study, amyloid load was equally distributed in the different amyloid subtypes, but in only 1 specimen (a patient with wt-ATTR amyloidosis) was amyloid load higher than 60%, potentially indicating the maximal tolerable cardiac amyloid load. Amyloid load is claimed to be higher in ATTR rather than AL amyloidosis despite minor symptoms. In the present study, higher amyloid load was observed in ATTR amyloidosis, although the difference was not statistically significant. This pathological observation aligned with previous imaging data (12).
In our series, amyloid load correlated with some indicators of severity of cardiac amyloidosis (NT-proBNP and LV wall thickness) in patients with ATTR amyloidosis but not in AL amyloidosis (15). However, in AL amyloidosis, amyloid load correlated with patient survival. This might be explained by selection bias, as our patients had exclusively histologically proven cardiac involvement in contrast to other studies that analyzed patients independent of organ manifestation. Correlation of troponin T would be of remarkable interest, as it has been reported to be a strong predictor of cardiac amyloidosis severity (14). Unfortunately, results were not available due to a change of troponin assays during the sample collection.
In general, EMBs carry the risk of a sampling error, which might explain the missing correlation with noninvasive indicators of severity of cardiac amyloidosis. Skeletal scintigraphy and CMR imaging with T1 mapping were alternative tools for noninvasive quantification of cardiac amyloid (11). Bone tracers are capable of differentiating and quantifying cardiac ATTR amyloid. T1 mapping allows detection of even minor changes of the myocardial texture. Unfortunately, data were not available for the present cohort due to the long observation period.
Amyloid load and survival
In the present study, patients with AL amyloidosis had a significantly worse prognosis over ATTR amyloidosis (median overall survival: 15.7 months vs. ≥38.9 months) (Table 1). In general, AL exhibits poorer survival than ATTR amyloidosis (38). However, according to more recent reports on survival of patients with ATTR amyloidosis, distinct differences between individual mutations were reported. Shorter survival was observed in patients with mainly cardiac manifestation due to p.Val142Ile transthyretin gene variant, and survival of patients with wt-ATTR amyloidosis appears to be worse than expected (39,40). However, in our cohort, NYHA functional class was the only independent prognosticator of patient survival. Presence of cardiac amyloid itself might be associated with poorer survival in ATTR; however, this was beyond the scope of the current analysis.
Besides diagnostic purposes (30), our study demonstrated that quantification of AL amyloid load in EMBs may have prognostic and predictive value for future treatment. Following multivariate analysis, amyloid load remained an independent predictor of patient survival: patients with cardiac amyloid >40% had a worse prognosis. With regard to the predictive value, Kaplan-Meier plots illustrated a significant difference between responders and nonresponders when the amyloid load was <20%. No difference was found in patients with ≥20% amyloid load. Thus, early identification of subjects with AL amyloid is essential, as our preliminary data suggested that in late stages of disease with extensive amyloid load, chemotherapy administration does not affect outcomes. This finding may illustrate different mechanisms of cardiac damage. It has long been recognized that cardiac impairment in AL amyloidosis is related to the toxic effects light chains exert on cardiomyocytes. This effect is probably most effective at the early stage of the disease (i.e., when amyloid load is <20%). However, in more advanced cardiac AL amyloidosis, the deposits increasingly exert biomechanical effects on cardiac function, resulting in impaired diastolic blood pressure.
On the basis of our data, we hypothesized that 20% to 40% cardiac amyloid load marks a threshold at which the primarily toxic and primarily biomechanic effects of cardiac AL amyloid are temporarily balanced out. This conjecture was supported by the lower median NT-proBNP serum levels observed in patients with AL who had 20% to 40% (6,989 ng/ml) amyloid load compared with patients with <20% (8,613 nm/ml) or >40% (8,746 ng/ml) (Online Table 1). Moreover, quantitative amyloid load in EMBs appears to identify patients who are at high risk for fatal outcome and, hence, might be an additional parameter for selecting patients for heart transplantation if solely cardiac manifestation is present (41).
According to the present study, in ATTR amyloidosis, EMBs did not reveal any benefit. Noninvasive methods using skeletal scintigraphy show high sensitivity and specificity for identifying cardiac ATTR amyloid, and no additional benefit of amyloid quantification was reported. Thus, EMBs appear not to be needed in these patients, when the diagnosis was already reached by other measures.
This was a retrospective analysis of a large cohort of patients with cardiac amyloidosis confirmed by EMB. It was limited by the lack of skeletal scintigraphy, CMR imaging with T1 mapping, echocardiography with speckle tracking, and high-sensitivity troponin T as potential diagnostic tools for early manifestation of cardiac amyloidosis. Moreover, evaluation of response status was not achievable for 30 subjects due to limited survival of patients with cardiac amyloidosis. Thus, the power of survival analyses needs to be confirmed prospectively in an independent, larger patient cohort.
Early identification of subjects with AL amyloid is essential, as in late stages of disease with extensive amyloid load, preliminary data suggest that chemotherapy does not affect outcomes. These results need to be confirmed in an independent cohort of AL amyloidosis and a larger group of patients with different types of ATTR amyloidosis.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: In patients with AL amyloidosis, evaluation of the cardiac amyloid load by EMB has both diagnostic and prognostic value, as those with high loads gain less benefit from available chemotherapy.
TRANSLATIONAL OUTLOOK: Further studies are needed to determine whether selection of patients with AL amyloidosis for chemotherapy or transplantation can be guided effectively by quantification of the endomyocardial amyloid burden.
For an expanded Methods section as well as supplemental figures and a table, please see the online version of this article.
Dr. Röcken was supported by grants from the German Research Foundation (Grant-No. Ro 1173/11) and the Federal Ministry of Education and Research (GERAMY). All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- light-chain amyloid
- transthyretin amyloid
- estimated glomerular filtration rate
- endomyocardial biopsy
- mutant-type transthyretin amyloid
- N-terminal pro-B-type natriuretic peptide
- New York Heart Association
- wild-type transthyretin amyloid
- Received November 23, 2015.
- Revision received April 3, 2016.
- Accepted April 12, 2016.
- American College of Cardiology Foundation
- Kristen A.V.,
- Giannitsis E.,
- Lehrke S.,
- et al.
- Dubrey S.,
- Ackermann E.,
- Gillmore J.
- Rapezzi C.,
- Merlini G.,
- Quarta C.C.,
- et al.
- Buss S.J.,
- Emami M.,
- Mereles D.,
- et al.
- Fontana M.,
- Banypersad S.M.,
- Treibel T.A.,
- et al.
- Perugini E.,
- Guidalotti P.L.,
- Salvi F.,
- et al.
- Castaño A.,
- DeLuca A.,
- Weinberg R.,
- et al.
- Palladini G.
- Rahman J.E.,
- Helou E.F.,
- Gelzer-Bell R.,
- et al.
- Palladini G.,
- Dispenzieri A.,
- Gertz M.A.,
- et al.
- Schonland S.O.,
- Hegenbart U.,
- Bochtler T.,
- et al.
- Benjamini Y.,
- Hochberg Y.
- Liao R.,
- Jain M.,
- Teller P.,
- et al.
- Chimenti C.,
- Frustaci A.
- Röcken C.,
- Peters B.,
- Juenemann G.,
- et al.
- Rapezzi C.,
- Quarta C.C.,
- Obici L.,
- et al.