Role of cardiovascular magnetic resonance in suspected cardiac amyloidosis: late gadolinium enhancement pattern as mortality predictor
Role of cardiovascular magnetic resonance in suspected cardiac amyloidosis: late gadolinium enhancement pattern as mortality predictor
M. Baroni 0 1 2 3
S. Nava 0 1 2 3
G. Quattrocchi 0 1 2 3
A. Milazzo 0 1 2 3
C. Giannattasio 0 1 2 3
A. Roghi 0 1 2 3
P. Pedrotti 0 1 2 3
0 Cardiologia 1, A. De' Gasperis Heart Center, ASST Grande Ospedale Metropolitano Niguarda , Milano , Italy
1 Cardiologia 3, A. De' Gasperis Heart Center, ASST Grande Ospedale Metropolitano Niguarda , Milano , Italy
2 Health Science Department, Bicocca University , Milano , Italy
3 Cardiologia 4, A. De' Gasperis Heart Center, ASST Grande Ospedale Metropolitano Niguarda , Milano , Italy
Background Cardiac magnetic resonance (CMR) has gained a central role in the diagnosis of cardiac amyloidosis (CA). While the diagnostic role of a typical late gadolinium enhancement (LGE) pattern (global subendocardial enhancement coupled with accelerated contrast washout) has been identified, evidence is still conflicting regarding the prognostic role of such examination. Methods and results We retrospectively analysed all patients referring for CMR at Niguarda Hospital (Milan, Italy) from January 2006 to January 2015 for suspected CA. Primary outcome was all-cause mortality. We identified 42 patients and divided them into 2 groups, according to the presence (Group A) or absence (Group B) of a typical amyloidosis LGE pattern. At the end of the followup (median 37 months, interquartile range 10-50 months), 31 patients (74%) had died. The hazard ratio for all-cause death was 3.2 (95% confidence interval [CI] 1.5-6.4, p < 0.01) for Group A versus Group B. Median survival time was 17 months (95% CI 7-42 months) for Group A and 70 months (95% CI 49-94 months) for Group B (p < 0.01). Multivariate analysis did not find any adjunctive predictive role for biventricular volumes and ejection fraction, indexed left ventricular mass, transmitral E/e' at echocardiography, age at diagnosis or serum creatinine. Conclusion In our population, a typical LGE pattern was significantly associated with higher mortality. Moreover, patients with a typical LGE pattern showed a globally worse prognosis. Our data suggest that the LGE pattern may play a central role in prognostic stratification of patients with suspected CA, thus prompting further diagnostic and therapeutic measures.
Amyloidosis; Cardiac magnetic resonance
Infiltrative cardiomyopathies are a wide spectrum of rare
cardiac diseases characterised by chronic deposition of
pathological substance into cardiac muscle, thus causing
its progressive stiffening, diastolic heart failure and an
overall poor outcome. Amyloidosis, caused by a misfolded,
insoluble aggregated protein featuring a characteristic
betasheet structure, is the most common aetiology [
different amyloid proteins have been described over the
years, only two forms of cardiac amyloidosis (CA) account
for the majority of cases. AL amyloidosis, caused by
immunoglobulin light chains produced by a mutated plasma
cell clone, is the most common form accounting for about
two thirds of cases of CA. Transthyretin-related (ATTR)
amyloidosis can reveal itself as an inherited or a sporadic
(senile systemic) form, with a variable range of severity
]. Over the past few years, cardiac magnetic resonance
(CMR) has gained a central role in the diagnostic workup
of infiltrative cardiomyopathies. Allowing tissue
characterisation, CMR has proven to be a key diagnostic tool
when routine cardiac echography and the patient’s history
are suggestive of CA. A typical pattern of late gadolinium
enhancement (LGE) has been described in patients with
CA, characterised by global subendocardial enhancement
with accelerated gadolinium washout [
While the diagnostic accuracy of CMR has been
highlighted by several studies, evidence is still conflicting about
the prognostic role of this type of examination for patients
with suspected CA. This study aimed to assess the
prognostic value of CMR in a group of consecutive patients
routinely referred for a CMR scan in the work-up of
suspected CA in a tertiary referral centre.
We searched the database of the CMR Laboratory of
Niguarda Hospital (Milan, Italy) from January 2006 to January
2012 for patients referred for suspected CA. We included
patients with known systemic amyloidosis (ATTR or AL),
as well as patients referred for a suspected infiltrative
disorder at clinical and echocardiographic assessment, with
evidence of congestive heart failure and diastolic
dysfunction or myocardial hypertrophy with a discordance between
electrocardiogram (ECG) QRS voltage and left ventricular
wall thickness. We accurately excluded patients with
evidence of increase of left ventricular mass due to valvular
diseases, hypertension or hypertrophic cardiomyopathy.
Sources of data
Baseline clinical, laboratory and echocardiographic data
were obtained from internal medical records. When a
parameter was repeatedly available, only the closest to the
time of CMR was considered.
Echocardiographic data included left ventricular
volumes, left ventricular mass (Devereux formula) and
ejection fraction, as well as Doppler markers for diastolic
dysfunction (mitral E/e’ ratio). Laboratory markers
included blood cell count, renal function and N-terminal
probrain natriuretic peptide (NT-proBNP).
Patients were scheduled for contrast-enhanced CMR after
the exclusion of contraindications to CMR and
gadoliniumbased contrast agents. Scans were performed on a 1.5 T
clinical scanner (Siemens Avanto, Erlangen, Germany) using
a four-element phased-array receiver coil. All images were
acquired with retrospective ECG gating and during repeated
single breath-holds. Balanced steady state free precession
cine images (echo time/repetition time, 1.6/3.2 ms; flip
angle, 60°; typical pixel size, 2.4 × 1.4 mm) were acquired in
three long-axis planes and in contiguous short-axis slices
(slice thickness, 8 mm; gap, 2 mm) from the
atrioventricular groove to the apex. LGE images were acquired 5 and
10 minutes after a 0.1 mmol/kg infusion of gadopentetate
dimeglumine (Bayer AG Magnevist, Leverkusen, Germany)
on planes matching cine images, with a segmented
inversion-recovery gradient echo sequence (echo time/repetition
time, 1.4/5.4 ms; flip angle, 10°; slice thickness, 8 mm; gap,
2 mm; typical spatial resolution, 2.2 × 1.5 mm; inversion
times adjusted to null normal myocardium), after inversion
time identification with a scout acquisition [
the technical challenge to correctly null the myocardium in
CA patients, also phase-sensitive inversion recovery (PSIR)
sequences were acquired.
Image analysis was performed by 2 experienced
observers (PP, AR) blinded to patient identity on a
commercially available cardiovascular magnetic resonance
workstation (Leonardo, Siemens, Erlangen, Germany). Ventricular
volumes and myocardial mass were quantified from
shortaxis cine images by standard methods [
] and indexed to
body surface area. The observers independently determined
the dichotomous presence or absence of myocardium LGE
by reviewing all post-contrast CMR images; discordant
interpretations were resolved by consensus. LGE distribution
was classified as: typical LGE pattern, with global
subendocardial enhancement with or without any transmural
enhancement and atypical LGE pattern, with any other
nonischaemic pattern (Fig. 1). Due to the lack of a
standardised method for quantitative description of washout kinesis,
this parameter was assessed by the ratio of blood and
myocardial T1 values, measured 5 minutes after gadolinium
infusion. Myocardial T1 was measured on the left
ventricular subendocardium of the interventricular septum while
blood T1 was measured in the middle of the left
ventricular chamber (Fig. 2) in a mid-ventricular short-axis slice.
A blood/myocardium value <1 (i. e. more gadolinium in the
myocardium than in the blood 5 minutes after injection) was
considered a marker of accelerated contrast washout.
Endomyocardial biopsy and genetic analysis
Patients were referred to endomyocardial biopsy or genetic
analysis at the discretion of the treating physician.
Endomyocardial biopsy was performed via the internal jugular vein
or the femoral vein with a standard bioptome. At least three
samples were taken from the right ventricular apex and
interventricular septum and processed at the pathologist’s
For genetic analysis, DNA was extracted from whole
blood and amplified by polymerase chain reaction assay,
and the coding region of the TTR gene was sequenced
looking for most common spot mutations.
No. at risk
Primary endpoint was all-cause mortality. Data was
obtained from patients’ records and from the national health
service’s electronic archives. Death status was determined
as of month and year. The most recent date of the clinical
follow-up was recorded for survived patients.
Statistical analysis was performed with Med Calc v.14
(Medcalc Software, Ostend, Belgium) and SPSS v.22
software (IBM corp, Armonk, USA). Continuous variables
were tested for normality with the Shapiro-Wilk test.
Normal variables are presented as mean ± standard deviation
(SD), otherwise as median and interquartile range (IQR).
Differences in the mean values were compared using a
twotailed t-test for continuous paired and unpaired variables
and a Chi-square test for nominal values. A p-value <0.05
was considered statistically significant. Survival was
evaluated with a Cox proportional hazards regression analysis,
providing hazard ratios with 95% confidence intervals and
Kaplan-Meier curves. Variables selected for clinical
relevance were first explored with the univariate Cox regression
model and then entered into the multivariate model.
Of 7,800 CMR scans performed in the aforementioned
period, 42 patients (31 males, mean age at CMR 58 ±
12 years) were included for analysis. Patients were divided
into Group A and Group B, based on LGE pattern
analysis. Patients in Group A (28; 67%) had a typical LGE
pattern, as previously defined, while patients in Group B
(14; 33%) had an atypical LGE pattern. A detailed
description of the diagnostic findings is summarised in Tab. 1.
Patients in Group A had more frequently LGE of atrial
walls and an accelerated contrast washout. There were no
significant differences in clinical features nor in other
imaging findings between the 2 groups (Tab. 2). After a median
follow up of 37 months (IQR 10–50 months), 31 patients
(74%) had died. Patients in Group A had an hazard ratio
for all-cause death of 3.2 compared to patients in Group B
(95% CI 1.5–6.4, p < 0.01) (Fig. 3). Median survival time
was 17 months (95% CI 7–42 months) for Group A and
70 months (95% CI 49–94 months) for Group B (p < 0.01).
Biventricular volumes and ejection fraction, indexed left
ventricular mass (I-LVM), E/e’ at echo, age at diagnosis
and serum creatinine didn’t show any adjunctive
predictive value at multivariate analysis (Tab. 3). Endomyocardial
biopsy or genetic analysis was performed in 15 patients
(54%) of Group A and in 13 patients (93%) of Group B.
Diagnosis of amyloidosis was confirmed in 13 patients in
Group A (4 mutated ATTR, 1 senile systemic ATTR, 8 AL)
and in 1 patient in Group B (1 senile systemic ATTR, p <
0.05) (Tab. 1), thus CMR sensitivity and specificity for the
diagnosis of CA were 93% and 87%, respectively.
LGE late gadolinium enhancement, EMB endomyocardial biopsy, GA genetic analysis, AL amyloid light chain, WT wild type,
ATTR transthyretin-related, NP not performed
CMR cardiac magnetic resonance, LV EDV left ventricular end diastolic volume, LV EF left ventricular ejection fraction, IVS interventricular
septum, I-LV EDV indexed left ventricular end diastolic volume, I-LVM indexed left ventricular mass, I-RV EDV indexed right ventricular end
diastolic volume, RV EF right ventricular ejection fraction
CMR is the gold standard for structural and functional
evaluation of patients with known or suspected cardiomyopathy.
The prognostic value of CMR has been documented in
], dilated [
] and hypertrophic cardiomyopathy
]. In CA, the evidence of global subendocardial LGE
with accelerated contrast washout has shown an excellent
diagnostic accuracy .
LVEDV left ventricular end diastolic volume, RVEDV right ventricular end diastolic volume, I-LVM indexed
left ventricular mass, LV EF left ventricular ejection fraction, RV EF right ventricular ejection fraction, MRI
magnetic resonance imaging
Data from our study suggest that the presence of typical
LGE patterns is a strong independent predictor of
mortality for patients with suspected CA. Currently, little
evidence is available about the role of LGE as predictor of
adverse events in CA. Maceira et al. didn’t find any
significant correlation between LGE patterns and survival at
4 years in 29 patients with proven AL CA, while a
correlation was found for gadolinium kinetics [
a trend towards a worse outcome was observed for patients
with positive LGE, the sample size was probably too small
to draw any definitive conclusion. The larger prospective
study by White et al. documented the prognostic value of
transmural LGE in a cohort of patients with proven
systemic amyloidosis and suspected cardiac involvement [
Considering that CA is a progressive disease and the
degree of LGE probably reflects the continuum of myocardial
involvement, the former study mainly identified patients
with advanced disease featuring an intrinsic worse
prognosis. This observation agrees with the findings by Fontana
et al. [
]. who demonstrated that transmural global LGE
was a stronger death predictor than subendocardial LGE.
In our population, such prognostic difference can’t be
observed because of the relatively small quota of
transmural LGE. Recently, a large meta-analysis of 425 patients
demonstrated an increased risk of death after a median
follow-up of 25 months in patients with positive LGE (odds
ratio 4.96, 95% CI 1.9–12.9) [
heterogeneity of inclusion criteria, imaging protocols and follow-up
length within the studies could have determined a loss of
events, especially for patients with early stage disease. To
the best of our knowledge, the present study has the longest
follow-up available for this category of patients. Our data
show that the median survival time is significantly shorter
for patients with a typical LGE pattern for CA. Conversely,
the absence of a typical LGE was associated with better
long-term prognosis, independently from typical predictors
of worse outcome for heart failure patients like left
ventricular ejection fraction or E/e’ (as a surrogate of diastolic
]. I-LVM, often considered a marker of
cardiac involvement in infiltrative diseases [
], did not show
any adjunctive predictive role compared to LGE. Our data,
derived from a “real world” scenario, suggest that the LGE
pattern can be considered a reliable tool for the diagnosis of
CA. Moreover, the degree of myocardial involvement may
play a central role in prognostic stratification of patients
with suspected CA. In our population, the mere presence of
LGE in the heart was associated with a significantly worse
outcome. This probably accounts for the possibility that
myocardial involvement could be one of the last steps of
a progressive systemic disease. However, over the course of
the past few years, new therapeutic approaches have been
developed for targeted therapy of most forms of CA [
Although there still is no solid data about the efficacy of
such interventions on patients with heavy myocardial
involvement, an adjunctive tool for prompt identification of
high-risk patients could accelerate the planning of further
invasive and therapeutic strategies. On the other hand, if
these new therapies do not prove to be effective in this
subgroup of patients, severe myocardial involvement at CMR
could reasonably be considered an exclusion criterion for
the administration of such treatments.
Some limitations of the present study should be
acknowledged. This is a retrospective study carried out in a tertiary
referral centre, thus with a high probability of including
patients with severe disease. In recent years, the introduction
of parametric mapping has been shown to be very promising
in patients affected by infiltrative disorders, allowing a
better characterisation of the disease and an early diagnosis. At
the time of the study, T1 mapping was not yet available at
our centre, so it could not be included in the scan protocol
of these patients. Besides, T1 mapping is not yet completely
standardised across centres, while LGE acquisition and
assessment are well established and are part of routine CMR
clinical examination, thus rendering the analysis of LGE
pattern an easily obtainable tool for the clinical evaluation
of patients with CA. T1 blood/myocardial ratio is not a
validated technique for the assessment of contrast washout and
inversion times were chosen at the operators’ discretion
during acquisition of LGE sequences. However, this method is
a physiologically logical surrogate for semi-quantitative
description of contrast washout and showed good correlation
with inspective evaluation of gadolinium kinetics. Another
significant limitation of the study is the exclusion of
patients with severe renal failure due to contraindication to
a magnetic resonance contrast agent. Considering that most
forms of systemic amyloidosis show a progressive renal
involvement, this limitation could result in the exclusion of
some patients with an advanced stage disease, potentially
leading to a selection bias.
In conclusion, the LGE pattern proved to be an
affordable diagnostic tool and an independent adverse prognostic
predictor for patients with suspected cardiac amyloidosis.
Early identification of high-risk subjects could guide
towards a more aggressive diagnostic workup, helping to
improve therapeutic choices.
Conflict of interest M. Baroni, S. Nava, G. Quattrocchi, A. Milazzo,
C. Giannattasio, A. Roghi and P. Pedrotti declare that they have no
Open Access This article is distributed under the terms of the
Creative Commons Attribution 4.0 International License (http://
creativecommons.org/licenses/by/4.0/), which permits unrestricted
use, distribution, and reproduction in any medium, provided you give
appropriate credit to the original author(s) and the source, provide a
link to the Creative Commons license, and indicate if changes were
1. Kyle RA. Amyloidosis . Circulation. 1995 ; 91 : 1269 - 71 .
2. Rapezzi C , Merlini G , Quarta CC , et al. Systemic cardiac amyloidoses: disease profiles and clinical courses of the 3 main types . Circulation . 2009 ; 120 : 1203 - 12 .
3. Fontana M , Chung R , Hawkins PN , Moon JC . Cardiovascular magnetic resonance for amyloidosis . Heart Fail Rev . 2015 ; 20 : 133 - 44 .
4. Vogelsberg H , Mahrholdt H , Deluigi CC , et al. Cardiovascular magnetic resonance in clinically suspected cardiac amyloidosis: noninvasive imaging compared to endomyocardial biopsy . J Am Coll Cardiol . 2008 ; 51 : 1022 - 30 .
5. Kramer CM , Barkhausen J , Flamm SD , Kim RJ , Nagel E , Society for Cardiovascular Magnetic Resonance Board of Trustees Task Force on Standardized Protocols. Standardized cardiovascular magnetic resonance (CMR) protocols 2013 update . J Cardiovasc Magn Reson . 2013 ; 15 : 91 .
6. Kelle S , Roes SD , Klein C , et al. Prognostic value of myocardial infarct size and contractile reserve using magnetic resonance imaging . J Am Coll Cardiol . 2009 ; 54 : 1770 - 7 .
7. Gulati A , Jabbour A , Ismail TF , et al. Association of fibrosis with mortality and sudden cardiac death in patients with nonischemic dilated cardiomyopathy . JAMA . 2013 ; 309 : 896 - 908 .
8. Chan RH , Maron BJ , Olivotto I , et al. Prognostic value of quantitative contrast-enhanced cardiovascular magnetic resonance for the evaluation of sudden death risk in patients with hypertrophic cardiomyopathy . Circulation . 2014 ; 130 : 484 - 95 .
9. Chan RH , Maron BJ , Olivotto I , et al. Significance of late gadolinium enhancement at right ventricular attachment to ventricular septum in patients with hypertrophic cardiomyopathy . Am J Cardiol . 2015 ; 116 : 436 - 41 .
10. Maceira AM , Joshi J , Prasad SK , et al. Cardiovascular magnetic resonance in cardiac amyloidosis . Circulation . 2005 ; 111 : 186 - 93 .
11. Maceira AM , Prasad SK , Hawkins PN , Roughton M , Pennell DJ . Cardiovascular magnetic resonance and prognosis in cardiac amyloidosis . J Cardiovasc Magn Reson . 2008 ; 10 : 54 .
12. White JA , Kim HW , Shah D , et al. CMR imaging with rapid visual T1 assessment predicts mortality in patients suspected of cardiac amyloidosis . JACC Cardiovasc Imaging . 2014 ; 7 : 143 - 56 .
13. Fontana M , Pica S , Reant P , et al. Prognostic value of late gadolinium enhancement cardiovascular magnetic resonance in cardiac amyloidosisCLINICAL PERSPECTIVE . Circulation. 2015 ; 132 : 1570 - 9 .
14. Raina S , Lensing SY , Nairooz RS , et al. Prognostic value of late gadolinium enhancement CMR in systemic amyloidosis . JACC Cardiovasc Imaging . 2016 ; 9 ( 11 ): 1267 - 77 .
15. Nagueh SF , Appleton CP , Gillebert TC , et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography . J Am Soc Echocardiogr . 2009 ; 22 : 107 - 33 .
16. Kristen AV , aus dem Siepen F , Scherer K , et al. Comparison of different types of cardiac amyloidosis by cardiac magnetic resonance imaging . Amyloid . 2015 ; 22 : 132 - 41 .
17. Palladini G , Milani P , Merlini G . Novel strategies for the diagnosis and treatment of cardiac amyloidosis . Expert Rev Cardiovasc Ther . 2015 ; 13 : 1195 - 211 .