RESUMO
BACKGROUND: The optimal heart-to-contralateral chest (H/CL) ratio threshold for non-invasive diagnosis of transthyretin cardiac amyloidosis (ATTR-CA) using Tc99m pyrophosphate (PYP) imaging in a population with low pretest probability is not known. METHODS: Using myocardial PYP retention by SPECT as the reference standard, we evaluated the diagnostic performance of different semi-quantitative and quantitative (H/CL chest ratio) planar parameters obtained from 3-hour PYP imaging in a prospectively recruited cohort of minority older adults with heart failure and increased LV wall thickness. RESULTS: Of 229 patients, 14 were found to have ATTR-CA (6.1%). No PYP uptake (grade 0) was observed in 77% of scans, all grade 3 scans were ATTR-CA, and only 4 of 11 (36%) grade 2 scans were ATTR-CA. An H/CL threshold of ≥ 1.4 maximized specificity (99%) and positive predictive value (93%) but resulted in decreased sensitivity (93%), compared to the ≥ 1.3 threshold which had 100% sensitivity. CONCLUSION: Among patients with a low pretest likelihood of ATTR-CA, planar interpretation, while useful to exclude disease, must be interpreted with caution. H/CL ratio threshold of ≥ 1.3 resulted in clinically important misclassifications. These data suggest that quantitative planar imaging thresholds may not be appropriate to apply in low pretest likelihood populations being evaluated for ATTR-CA.
