Systemic amyloidosis is a multiorgan disease resulting from extracellular deposition of insoluble amyloid fibrils, formed from aggregated, misfolded proteins (1, 2). Cardiac amyloidosis is predominantly caused by light chain amyloid (AL) and transthyretin amyloidosis [ATTR] (2); other amyloid subtypes are less prevalent and rarely demonstrate cardiac involvement (2).

ATTR subtypes include wild-type and hereditary, both displaying heterogeneity in clinical symptoms and organ involvement (1, 2). Cardiac involvement in ATTR is common, with median survival without treatment of 2–4 years, presenting commonly with heart failure with preserved ejection fraction, conduction disorders and atrial fibrillation (1, 3–9).

Cardiac amyloidosis diagnosis can be difficult, with definitive diagnosis usually based on tissue immunohistochemistry and/or mass spectrometry proteomic analysis of cardiac biopsy specimens (10, 11). Cardiac amyloidosis subtype identification is crucial, particularly as novel treatment options emerge for ATTR (12, 13). Several non-invasive screening tests are now available including transthoracic echocardiography (TTE) and cardiac magnetic resonance imaging (12). Recently, technetium-99m 3,3-diphosphono-1,2-propanodicarboxylic acid (^99mTc-DPD) scintigraphy emerged as an important diagnostic modality with strong tracer avidity for the transthyretin protein, with high sensitivity and specificity (10, 11, 14, 15). Most cardiac amyloidosis studies using ^99mTc-DPD scintigraphy employed qualitative visual grading or semi-quantitative grading with heart-to-body ratios (15, 16). Few studies employed quantitative ^99mTc-DPD tracer uptake analysis, and consequently, little is known about tracer uptake correlation with specific structural and functional cardiac changes (16, 17).

This study evaluated echocardiographic biventricular structural and functional involvement and determined their correlation with ^99mTc-DPD tracer uptake quantitative analysis. We hypothesized that the extent of structural and functional abnormalities on TTE should be commensurate with ^99mTc-DPD tracer uptake amount.

We retrospectively reviewed 42 consecutive patients with wild-type, hereditary ATTR and asymptomatic transthyretin (TTR) variant carriers who attended the Westmead Hospital Amyloidosis Clinic, a statewide referral service for amyloidosis patient evaluation in New South Wales, Australia. Ethics was submitted and approved by the Western Sydney Local Health District (WSLHD) Research Ethics and Governance Committee in keeping with the Declaration of Helsinki. Two patients were excluded as ^99mTc-DPD reference imaging was not available for quantitative analysis.

In total, 40 patients (wild-type n = 17; hereditary ATTR and asymptomatic TTR variant carriers n = 23) were included for analysis after screening to ensure absence of a monoclonal gammopathy. All had undergone TTE and ^99mTc-DPD scintigraphy within a 12-month period. Figure 1 demonstrates the algorithm to identify cardiac involvement. Thirty-four patients were determined to have cardiac ATTR on ^99mTc-DPD scintigraphy; the remaining six patients had hereditary ATTR or were asymptomatic TTR variant carriers confirmed by ATTR genotyping, without cardiac involvement. Both TTR variant carriers and hereditary ATTR patients had positive family histories and were identified to have an amyloidogenic mutation, with hereditary ATTR patients additionally having clinical symptoms or proven amyloid deposition (Val122Ile n = 5, Val71Ala n = 1, Ala97Ser n = 3, Val30Met n = 6, Ser52Pro n = 1, Thr60Ala n = 5, Ser77Tyr n = 2). One patient identified to have cardiac involvement had mild aortic stenosis. All patients had serum creatinine (allowing derivation of estimated glomerular filtration rate), troponin I, N-terminal-pro hormone brain natriuretic peptide (NT-proBNP) and electrocardiography. Demographic and clinical information were collected for all patients. Sixteen patients had mild hypertension (14/34 in those with cardiac amyloidosis and 2/6 in those without).

Forty patients comprised the study cohort (30 males; median age 68.8 ± 22 years) with 34 patients with cardiac involvement. The remaining 6 patients were genotyped as carrying an amyloidogenic TTR variant with one having peripheral neuropathy. TTE and ^99mTc-DPD scintigraphy were performed with a median time of 7 days between scans (inter-quartile range 99.8 days); 88% of patients had scans within 4 months of each other. Six patients with pacemakers were excluded from the electrocardiogram analysis. Demographic, biochemical, electrocardiogram and TTE data for the group are presented in Tables 1A,B. Fourteen of the 34 patients with cardiac amyloidosis had mild hypertension and 6 had atrial flutter/fibrillation.

Echocardiographic LV parameters were obtained for all patients. Due to limited image quality, RV fractional area change, RV FWS and RV GLS could not be obtained in two patients and RV wall thickness and RV s' velocity could not be performed in one patient.

The ^99mTc-DPD scintigraphy signal obtained for each ventricle demonstrated a median LV value of 272% (interquartile range 462%) and median RV value of 203% (interquartile range 265%). There was a significant correlation of NT-proBNP levels with LV and RV ^99mTc-DPD tracer uptake (Table 2).

We examined the relationship between structural and functional echocardiographic parameters for LV, RV and regional myocardial ^99mTc-DPD uptake (Figure 2, Table 2).

Echocardiographic LV structural parameters of average wall thickness and LVMI demonstrated significant correlation with ^99mTc-DPD tracer uptake (Figures 3A,B). LV diastolic functional parameters of average e'and E/e' demonstrated good correlation with ^99mTc-DPD uptake (Figures 3C,D); however, peak E velocity demonstrated weak correlation. Of the LV systolic functional parameters evaluated, LVEF demonstrated a weak inverse correlation while LV GLS demonstrated a strong correlation (Figures 3E,F).

RV wall thickness demonstrated positive correlation with RV ^99mTc-DPD uptake (Figure 4A). Of the RV functional parameters evaluated (fractional area change, tricuspid annular plane systolic excursion, RV s' velocity, RV FWS and RV GLS), RV GLS was the only parameter demonstrating a modest positive correlation (r = 0.632, p-value < 0.001) (Figure 4B), while other RV functional parameters demonstrated weak correlations (Table 2).

Of the 6 patients without ^99mTc-DPD scintigraphic evidence of cardiac involvement, all demonstrated normal biventricular wall thickness and function (Table 3). Patients with ^99mTc-DPD scintigraphic evidence of cardiac ATTR had reduced LV GLS, increased average wall thickness and LVMI compared to those without. Patients with ^99mTc-DPD scintigraphic evidence of cardiac ATTR were found similarly to have RV wall thickness increased and functional RV parameters reduced (Table 3).

Receiver operating characteristic analysis was performed on echocardiographic LV and RV parameters to determine diagnostic accuracy in identifying cardiac ATTR based on ^99mTc-DPD scintigraphy. Average LV wall thickness and LVMI demonstrated the highest diagnostic accuracy (area under the curve = 0.96 for both); there was moderate sensitivity (77%) and high specificity (100%) for a cutoff of 12.2 mm for wall thickness, whereas LVMI had lower sensitivity (71%), but similar specificity (100%) for a cutoff value of 110 g/m² (Figure 5A). LV GLS had an area under the curve of 0.86, with moderate sensitivity (79%) and high specificity (100%) using a cutoff value of GLS of −17% (Figure 5A).

On assessment of RV parameters, RV wall thickness demonstrated an area under the curve of 0.96 with high sensitivity (87.1%) and specificity (100%) using a cutoff value of 5.5 mm (Figure 5B). RV GLS had an area under the curve of 0.88 with sensitivity and specificity of 83% using a cutoff value of −18.6% (Figure 5B). There was no significant difference in LV or RV parameters using DeLong's method [Figures 5A,B] (26).

The salient findings from this study are the demonstration of robust correlation of LV and RV structural and functional echocardiographic parameters with ^99mTc-DPD tracer uptake analysis, using ^99mTc-DPD SPECT-CT as a quantitative measure of ATTR deposition in myocardial tissue. Our results demonstrate that increasing ^99mTc-DPD tracer uptake is associated with greater structural change with increased biventricular wall thickness. Decreased LV diastolic and systolic function, as well as RV systolic function, were also associated with ^99mTc-DPD tracer uptake. In addition, we observed hereditary ATTR patients and asymptomatic TTR variant carriers without myocardial ^99mTc-DPD uptake demonstrating normal echocardiographic structure and function.

Structural echocardiographic parameters (LV and RV wall thickness and LVMI) demonstrated stronger correlation with ^99mTc-DPD uptake compared to functional echocardiographic parameters. This stronger correlation is probably because ^99mTc-DPD uptake reflects the amount of myocardial amyloid content and not the functional consequences thereof. Moreover, coexistent cardiac pathologies not reflected by ^99mTc-DPD tracer uptake could impact on functional myocardial properties.The presence of hypertension can result in increased LV wall thickness (27). Our group has previously demonstrated a reduction in LV GLS in hypertensive patients as well as those with AL amyloidosis vs. controls (28). Sixteen patients (14/34 in the cardiac amyloidosis group and 2/6 without cardiac involvement) had hypertension. The coexistent hypertension was only mild (mean systolic BP was 126 mmHg), but could have altered LV GLS. Such a confounding effect may only reduce correlations with tracer uptake on ^99mTc-DPD scintigraphy. Despite this, correlations demonstrated were robust between LV wall thickness and GLS and LV tracer uptake on ^99mTc-DPD scintigraphy.

It has previously been demonstrated that patients with atrial fibrillation may have impaired LV GLS compared to those in sinus rhythm independent of sex, age, heart rate, LVEF and LV mass (29). A variation in LV GLS should reduce correlation with tracer uptake compared to those in sinus rhythm, and although 15% of the study cohort was in atrial fibrillation, the overall correlation with tracer uptake was strong. Aortic stenosis has been previously demonstrated to be associated with cardiac ATTR, which can also impact some echocardiographic parameters (30, 31). However, in the study group only one patient had mild aortic stenosis in the cardiac ATTR group and unlikely to make a significant impact on the outcomes.

^99mTc-DPD scintigraphy, initially utilized for bone scintigraphy, was found incidentally to demonstrate avidity for cardiac amyloid (16). A large multicentre trial demonstrated both high sensitivity and specificity for cardiac ATTR detection (32). The mechanism of ^99mTc-DPD binding to ATTR fibrils is yet to be identified (33). Although, some patients with AL amyloidosis demonstrate ^99mTc-DPD tracer uptake, the absence of monoclonal protein in serum and urine affords diagnostic sensitivity >99% and specificity 100% (32).

^99mTc-DPD scintigraphy image acquisitions vary, with methods using either thorax targeted or whole body planar images; more recently, SPECT-CT imaging has been employed (34). ^99mTc-DPD image assessment has been performed by qualitative visual grading using the Perugini method or semi-quantitative assessment with heart-to-body region ratios (15, 34). Incorporating ^99mTc-DPD SPECT-CT imaging has allowed exploration of quantitative methods to evaluate cardiac amyloid burden (16, 17). The Perugini method was evaluated previously as a prognostic tool, however did not demonstrate any survival difference between different grades (15). NT-proBNP is a significant predictor of mortality in ATTR patients (15). The correlation between ^99mTc-DPD tracer uptake and NT-proBNP levels suggest that quantitative ^99mTc-DPD scintigraphy could potentially provide prognostic information; however, this requires validation.

Echocardiography plays an important role in screening and longitudinal follow-up of cardiac ATTR patients (10, 35). Characteristic features of cardiac ATTR include biventricular increase in wall thickness and increased myocardial echogenicity (referred as a “speckled” appearance) (1, 7, 36). Diastolic dysfunction is present; however, LVEF is usually preserved until advanced disease (1). Cardiac ATTR strain analysis demonstrates subclinical systolic impairment despite preserved LVEF, with reduced basal and relatively preserved apical segmental longitudinal strain, producing the characteristic “apical sparing” strain pattern (1, 36). Increased LV wall thickness results in worsening GLS and eventually LVEF (7, 36). Previous studies have shown LVEF demonstrates lower sensitivity than LV GLS for identification of early or subclinical LV dysfunction, and this finding has been recapitulated in the present analysis where LVEF was preserved, with no correlation with tracer uptake (37, 38).

Only few studies evaluating echocardiographic changes in ATTR patients with ^99mTc-DPD tracer uptake employed a combination of qualitative visual grading, semi-quantitative analysis and planar imaging to quantify ^99mTc-DPD tracer uptake (39–41). In congruence with our findings, these earlier studies demonstrated ^99mTc-DPD tracer uptake correlates with LV wall thickness, mitral annular systolic velocity, tricuspid annular plane systolic excursion, E/e' and LV GLS (39–41). To our knowledge, our study is the first to quantitate ^99mTc-DPD tracer uptake in ventricular chambers and correlate this with biventricular structural and functional echocardiographic parameters. Additionally, prior studies focused primarily on LV parameters, with limited RV assessment, and only one previous study evaluated LV GLS (39–41).

Given the good correlation, we demonstrate the potential utility of TTE for longitudinal patient follow-up alongside quantitative ^99mTc-DPD scintigraphy for monitoring both disease progression and response to emerging therapeutic agents. Echocardiography could be particularly useful in regional centers, where availability of quantitative ^99mTc-DPD scintigraphy is limited.

We have demonstrated an association between echocardiographic structural and functional biventricular parameters with quantitative ^99mTc-DPD scintigraphy uptake. Additionally, we demonstrated that absence of ^99mTc-DPD tracer uptake is associated with normal structural and functional echocardiographic measurements, suggesting absence of cardiac ATTR deposition. Thus, in addition to their diagnostic utility for cardiac ATTR ^99mTc-DPD scintigraphy and echocardiography may both play significant roles in longitudinal follow-up of ATTR patients for monitoring disease progression and evaluating response to therapy.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by WSLHD HREC. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

All authors have contributed to the design, data collection, analysis, and manuscript preparation.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.