Cardiac Amyloidosis — Multimodality Imaging
Definition
Multimodality imaging in cardiac amyloidosis integrates echocardiography, cardiovascular magnetic resonance (CMR), and radionuclide bone-avid scintigraphy to diagnose, subtype, risk-stratify, and monitor AL and ATTR cardiac amyloidosis. No single modality can independently diagnose amyloidosis and confirm its subtype; the three modalities are complementary and serve different diagnostic roles. The gold standard non-invasive diagnostic algorithm combines bone-avid scintigraphy grading with exclusion of monoclonal protein.
Key Concepts
Formal Diagnostic Criteria (Table 1, Part 2)
Four diagnostic pathways exist; all non-biopsy pathways require "typical imaging features" as defined below. (sources/imaging-amyloidosis-2nd-aha-2021, rating: very high)
Pathway 1 — Endomyocardial biopsy (all subtypes): Congo red + apple-green birefringence; typing by immunohistochemistry and/or mass spectrometry
Pathway 2 — Extracardiac biopsy + typical imaging:
- ATTR: extracardiac biopsy-proven ATTR AND typical imaging features
- AL: extracardiac biopsy-proven AL AND (typical imaging features OR abnormal age-adjusted NT-proBNP/troponin with other causes excluded)
Pathway 3 — Non-biopsy ATTR diagnosis:
- 99mTc-PYP/DPD/HMDP Grade ≥2 AND no clonal plasma cell process (FLC + serum/urine immunofixation) AND typical imaging features
- When Grade ≥2 co-exists with any abnormal FLC/immunofixation or MGUS → NOT diagnostic; refer to specialist amyloid centre
Typical Imaging Features (any ONE, with other causes excluded):
| Modality | Parameter | Threshold |
|---|---|---|
| Echo | LV wall thickness | >12 mm |
| Echo | Apical sparing GLS ratio | Average apical LS / average (mid+basal) LS >1 |
| Echo | Diastolic dysfunction | Grade ≥2 |
| CMR | LV wall thickness | >ULN for sex on SSFP |
| CMR | Global ECV | >0.40 |
| CMR | LGE | Diffuse LGE pattern |
| CMR | Gadolinium kinetics | Myocardium nulls before blood pool |
| PET | Target-to-background (LV:blood pool) | >1.5 (18F-florbetapir or florbetaben) |
| PET | Retention index | >0.030 min⁻¹ |
Diagnostic Algorithm (Operational Flow)
- The consensus non-invasive diagnostic algorithm: (1) clinical/imaging suspicion; (2) serum/urine immunofixation + serum FLC assay to exclude monoclonal plasma cell dyscrasia; (3) 99mTc-PYP/DPD/HMDP scintigraphy if ATTR suspected. (sources/imaging-amyloidosis-aha-2021, rating: very high)
- Grade ≥2 scintigraphy + no monoclonal protein + typical imaging features = ATTR-CA without biopsy (PPV 100% [95%CI 98–100], n=1,498). (sources/imaging-amyloidosis-aha-2021, rating: very high)
- If any monoclonal protein detected → histological diagnosis required.
- If scintigraphy and monoclonal screen both negative → cardiac amyloidosis very unlikely (rare ATTRv mutations may be scintigraphy-negative).
- TTR gene sequencing in all confirmed ATTR-CA to distinguish ATTRwt from ATTRv.
Echocardiography — Red Flags and Key Parameters
- Morphological:
- LV wall thickness >1.2 cm without other explanation (most sensitive trigger)
- Relative wall thickness >0.42
- Biatrial enlargement and dysfunction
- Thickened interatrial septum (>0.5 cm) and valve leaflets
- Pericardial effusion
- Sparkling/granular myocardial texture (non-specific; also seen in ESRD)
- Small A wave in sinus rhythm (reflects LA failure; stroke risk marker)
- Functional:
- Tissue Doppler "5-5-5 sign": s', e', and a' all <5 cm/s at mitral annulus — highly suggestive in advanced disease; may not be sensitive early
- Reduced longitudinal function with preserved LVEF and radial/circumferential shortening
- Low stroke volume index despite preserved LVEF ("paradoxical low output")
- Grade 2–3 diastolic dysfunction with high E/A (>1.5) and short E deceleration time (<150 ms) in overt disease
- Strain Imaging (most specific echo feature):
- Reduced global LV longitudinal strain (GLS absolute value <–15%)
- "Cherry-on-top" sign on bullseye map: apical longitudinal strain preserved; basal and mid-LV segments severely impaired — most specific echo sign for cardiac amyloidosis; differentiates from HCM and aortic stenosis (which reduce strain at sites of maximum hypertrophy)
- Myocardial contraction fraction (MCF = stroke volume / myocardial volume): reduced; independent of chamber geometry
- Reporting categories: Not suggestive / Strongly suggestive / Equivocal for cardiac amyloidosis
- (sources/imaging-amyloidosis-aha-2021, rating: very high)
CMR — Tissue Characterisation
- LGE patterns:
- Subendocardial LGE: more prevalent in AL amyloidosis
- Transmural LGE: more prevalent in ATTR amyloidosis
- Both patterns basal-predominant initially; biventricular transmural in advanced disease
- LGE sensitivity 85–90%, specificity ~92% (meta-analysis, 7 studies)
- PSIR (phase-sensitive inversion recovery) technique is mandatory for amyloidosis LGE — standard mag-IR is unreliable because the optimal null-point cannot be selected; PSIR is operator-independent. Partially protein-bound contrast agents (gadolinium-BOPTA/MultiHance) must NOT be used — ECV technique and LGE pattern unreliable.
- TI scout gadolinium kinetics sign: myocardium nulls before or at the same time as blood pool (reversed from normal) — reflects ECV approaching plasma volume; rapid and confirmatory adjunct to PSIR LGE
- Native T1 mapping: elevated in both AL and ATTR; composite intra+extracellular signal; sensitivity 92%/specificity 91% (ShMOLLI); useful when contrast contraindicated
- ECV mapping (key quantitative marker):
- ECV >0.40 highly suggestive of cardiac amyloidosis
- Elevated even when LGE and conventional testing suggest no cardiac involvement → early disease marker
- Tracks amyloid burden and cardiomyocyte response longitudinally
- Prognostic in both AL and ATTR independently of known predictors
- T2 mapping: marker of myocardial oedema; independent prognostic factor in AL; not yet standard clinical protocol
- CMR cannot reliably distinguish AL from ATTR cardiac amyloidosis
- CMR reporting categories: Not suggestive / Strongly suggestive / Equivocal
- (sources/imaging-amyloidosis-aha-2021, rating: very high)
99mTc-PYP/DPD/HMDP Scintigraphy — Visual Grading System
-
The three tracers appear interchangeable (no direct head-to-head comparison); PYP preferred in USA; DPD/HMDP preferred in Europe
-
Visual grading (relative to rib uptake):
Grade Interpretation 0 No myocardial uptake; normal bone uptake 1 Myocardial uptake < rib uptake 2 Myocardial uptake = rib uptake 3 Myocardial uptake > rib uptake with mild/absent rib uptake -
Grade 2 or 3 = diagnostic of ATTR-CA if monoclonal protein excluded
-
Grade 1 = possible AL, AA, or ApoA1 amyloidosis; not diagnostic of ATTR; further work-up required
-
Grade ≥2 reported in >20% of AL patients → monoclonal exclusion is non-negotiable
-
H/CL (heart/contralateral lung) ratio: ≥1.5 at 1h or ≥1.3 at 3h — supports ATTR-CA only when SPECT confirms myocardial uptake; ratio alone is insufficient for diagnosis (removed as standalone criterion in addendum)
-
(sources/imaging-amyloidosis-aha-2021, rating: very high)
Addendum Protocol Updates (Critical)
Following reports of incorrect ATTR diagnoses in lower-prevalence populations, the protocol was updated:
- SPECT mandatory in all studies — confirms myocardial vs blood pool/bone uptake; SPECT/CT fusion helpful
- Preferred timing: 2–3 hours post-injection (1 hour is optional only; 1-hour planar-only NOT recommended)
- If excess blood pool activity at any timepoint → repeat SPECT at 3 hours
- H/CL ratio ≥1.5 removed as standalone "strongly positive" criterion — visual grade + SPECT is primary method
- Serum/urine immunofixation + FLC must be stated as recommended in every scintigraphy report
- (sources/imaging-amyloidosis-aha-2021, rating: very high)
Prognosis
- Echo: LV longitudinal strain, E deceleration time, pericardial effusion, RV wall thickness predict outcomes; no formal echocardiographic staging system validated
- CMR: LGE pattern independently predicts prognosis in AL and ATTR; ECV predicts in both; native T1 predicts in AL; T2 predicts in AL
- Scintigraphy: H/CL ratio >1.5 and H/WB ratio correlate with survival; combining cardiac uptake with septal thickness or NYHA class improves stratification; visual grade alone not independent predictor
- mIBG: Late HMR <1.6 predicts poor prognosis in ATTRv; cardiac denervation precedes diphosphonate scintigraphy uptake in TTR carriers → potentially earlier detection marker
- (sources/imaging-amyloidosis-aha-2021, rating: very high)
Appropriate Utilization — Key Rules (Part 2)
Evidence-based appropriateness ratings across 32 clinical scenarios (modified Delphi, 7 clinical experts; A = Appropriate, M = May Be Appropriate, R = Rarely Appropriate). (sources/imaging-amyloidosis-2nd-aha-2021, rating: very high)
Scintigraphy is RARELY APPROPRIATE when:
- AL amyloidosis is biopsy-proven or suspected (elevated FLC/monoclonal gammopathy/multiple myeloma) — cannot subtype amyloid in this context
- Monitoring disease progression or therapy response in biopsy-proven AL or ATTR
Scintigraphy is APPROPRIATE when:
- Asymptomatic TTR gene carrier (initial or recurrent)
- African-Americans >60 yr with unexplained HF or increased LV wall thickness
- Non-African-Americans >60 yr with HF + increased LV wall thickness
- Low-flow low-gradient aortic stenosis
- HF + unexplained peripheral sensorimotor neuropathy
- Known/suspected familial amyloidosis
- New or worsening symptoms in TTR carrier or confirmed ATTR
Echo is APPROPRIATE for nearly everything (27/30 indications); only high-frequency monitoring intervals (every 6 months) rated May Be Appropriate
CMR is the preferred modality for amyloid burden quantification (A(9) in both AL and ATTR biopsy-proven disease); Rarely Appropriate only for very frequent monitoring (every 6 months)
Emerging red-flag indications (bilateral carpal tunnel syndrome, biceps tendon rupture, unexplained neuropathy/arrhythmia in elderly men): Echo Appropriate; CMR and scintigraphy May Be Appropriate — insufficient evidence to rate higher
Monitoring interval consensus: Every 24 months Appropriate for echo and CMR in biopsy-proven disease; every 12 months Appropriate for echo, May Be Appropriate for CMR; every 6 months May Be Appropriate (echo) or Rarely Appropriate (CMR)
Management Monitoring
- Serial echocardiography: reasonable for LV dysfunction monitoring and anticoagulation guidance (LA function + TEE for LAA); GLS emerging as therapy-response marker
- CMR ECV: most promising quantitative response marker; retrospective data show ECV and LV mass decrease with deep hematologic response in AL
- Serial SPECT scintigraphy: NOT recommended for disease monitoring — not validated for this purpose; quantification not reliably correlated with treatment response
- No imaging technique has been prospectively validated for therapy response with survival endpoint
- (sources/imaging-amyloidosis-aha-2021, rating: very high)
AL vs ATTR Differentiation by Imaging
| Feature | AL Amyloidosis | ATTR Amyloidosis |
|---|---|---|
| LGE pattern | Subendocardial predominant | Transmural predominant |
| Scintigraphy | Grade 0–1 (usually); Grade ≥2 in >20% | Grade 2–3 (characteristic) |
| mIBG denervation | Less common | More common (esp. ATTRv) |
| CMR distinction | Neither AL nor ATTR reliably distinguished by CMR alone | |
| Echo distinction | Cherry-on-top pattern present in both; does not differentiate subtype |
Contradictions / Open Questions
- False-positive scintigraphy risk in lower-prevalence populations: The 100% PPV was derived from expert centres with high pre-test probability (NYHA >II, advanced disease). The addendum acknowledges increased false-positive diagnoses as scintigraphy expanded to lower-prevalence settings. The PPV in primary care or unselected populations has not been validated. (sources/imaging-amyloidosis-aha-2021)
- CMR vs scintigraphy for early disease detection: ECV by CMR is elevated before LGE or scintigraphy becomes abnormal in some patients. Whether ECV-based early detection improves outcomes has not been studied prospectively. (sources/imaging-amyloidosis-aha-2021)
- No validated imaging endpoint for therapy response: Neither scintigraphy, CMR, nor echocardiography has been prospectively validated against survival as a therapy-response measure. Clinical trials of tafamidis and RNA-interference agents relied on biomarkers rather than imaging endpoints. (sources/imaging-amyloidosis-aha-2021)
- Bone scintigraphy vs. CMR — complementary roles not standardised: Scintigraphy establishes ATTR subtype; CMR quantifies fibrosis burden and prognosis. The choice and sequence between the two modalities is not standardised. Also noted in entities/ATTR-Amyloidosis from (sources/esc-cmp-2023).
- AUC monitoring interval disagreement: There was lack of consensus (†) among the 7-expert panel for echo every 6 months (M(5)) and CMR/echo every 6 months in biopsy-proven ATTR/AL. This reflects genuine practice variation across expert amyloidosis centres. (sources/imaging-amyloidosis-2nd-aha-2021)
- PET amyloid-binding tracers — diagnostic thresholds are off-label and based on limited pilot data: The target-to-background >1.5 and retention index >0.030 min⁻¹ thresholds derive from small pilot studies; these agents are FDA-approved for brain amyloid in Alzheimer's disease but not for cardiac amyloidosis. The evidence base is substantially weaker than for scintigraphy. (sources/imaging-amyloidosis-2nd-aha-2021)
Connections
- Related to entities/ATTR-Amyloidosis
- Related to concepts/Late-Gadolinium-Enhancement
- Related to concepts/LV-Diastolic-Function
- Related to entities/Heart-Failure — ATTR-CA in HFpEF
- Related to entities/HCM — differential of increased LV wall thickness
- Related to concepts/Aortic-Stenosis — ATTR-CA co-existing with AS; AUC for low-flow low-gradient AS
- Related to entities/Atrial-Fibrillation — LAA thrombus risk in sinus rhythm; unexplained arrhythmia as red flag
- Related to concepts/Cancer-Therapy-Related-CV-Toxicity — 99mTc-PYP SPECT rated +++ in Table 2 for cardiac amyloidosis in cardio-oncology context (sources/imaging-cardio-oncology-aha-2024)