Deep Clinical Phenotyping of Perugini Grade 1 on Bone Scintigraphy: An Imaging-Phenotype Association Study
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Purpose
A planar Perugini grade 1 scan is equivocal for transthyretin amyloid cardiomyopathy (ATTR-CM), has no dedicated management pathway and is biologically heterogeneous, spanning early or low-burden ATTR, AL amyloidosis and non-amyloid blood-pool activity. Rather than investigating grade 1 as a step toward amyloid confirmation, we characterized the clinical phenotype it marks and tested whether that phenotype, rather than amyloid, drives its prognosis.
Methods
We studied 9,170 consecutive patients who underwent [ 99m Tc]Tc-DPD bone scintigraphy between 2010 and 2020, graded on planar imaging by blinded expert consensus. Using an imaging-phenotype association (IPA) framework analogous to genome-wide association analysis, we related Perugini grade to 1,243 clinical, laboratory, echocardiographic, cardiac magnetic resonance and ICD-10 comorbidity variables from electronic health records, adjusting for age, sex and cancer history. The outcome was a composite of heart failure hospitalization or death.
Results
Grade 1 occurred in 175 (1.9%) and grade ≥2 in 142 (1.6%) patients. Grade ≥2 reproduced the canonical infiltrative ATTR phenotype, with greater septal thickness, higher extracellular volume, neuropathy and atrial fibrillation, validating the framework. Grade 1 instead marked a non-infiltrative cardiometabolic phenotype dominated by hypertension, chronic ischemic heart disease, high BMI, anemia, reduced eGFR and atrial enlargement, with low extracellular volume and septal thickness arguing against meaningful infiltration. Grade 1 carried worse outcomes than grade 0 (adjusted HR 1.30, 95% CI 1.07 to 1.59). Within grade 1, a cardiometabolic phenotype defined by hypertension, high BMI and low hematocrit, with or without chronic ischemic heart disease, identified patients whose risk matched or exceeded that of grade ≥2 (median HR 2.46 versus HR 1.76), indicating that grade 1 prognosis is driven substantially by comorbid cardiometabolic disease rather than amyloid burden.
Conclusions
Perugini grade 1 is neither uniform early amyloid nor a benign artifact but a mixed-etiology, predominantly non-infiltrative cardiometabolic phenotype that carries clinically meaningful risk. A simple cardiometabolic phenotype stratifies this risk, supporting reinterpretation of grade 1 and a management focus on cardiorenal comorbidity alongside selective amyloid workup. The IPA framework provides a scalable approach to characterizing cardiovascular imaging biomarkers.
