Echocardiographic predictors of cardiac amyloidosis: validation of the IWT score

Abstract Background Cardiac amyloidosis (CA) is a progressive infiltrative cardiomyopathy. Accurate diagnosis of CA is critical, given that treatment improves mortality and morbidity. Echocardiography is the first line test for suspected amyloidosis. Multiparametric scores have been shown to increase the diagnostic accuracy of echocardiography for cardiac amyloidosis. Specifically, the increased wall thickness (IWT) score was shown be a sensitive and specific tool to rule in/out CA. We sought to improve the specificity of our echo lab for CA and reduce unnecessary downstream testing. We planned to validate the IWT score within our local echocardiography lab. We also hypothesized that patients with an IWT score 2 (low probability of CA) could safely forego further testing for CA. Methods This study was a retrospective cohort design. A prospectively kept database was used to identify all patients who underwent an echocardiogram at our lab followed by a technetium pyrophosphate scan during calendar years of 2023 and 2024. Echocardiographic images were reviewed by an experienced sonographer. All parameters incorporated in the IWT score were re-measured for accuracy. A chart review was performed by an amyloid cardiologist. The results of technetium pyrophosphate scans, cardiac MRI, cardiac biopsies, clinical biochemistry were used to identify patients as amyloid positive or negative. Statistical Analysis The performance of the IWT score as a linear variable was analyzed using logistic regression. The performance of the IWT score as categorical variable using a 3x2 contingency table (IWT score ≤2 , IWT score 3-5, IWT score 5. Results A total of 79 patients were identified and reviewed. Six patients were excluded from analyses due to poor echo image quality. The IWT score was calculated in the remaining 73 patients. A total of 17 patients were diagnosed with CA. The IWT score showed moderate predictive value. In univariate logistic regression, the odds ratio was 1.66 with a p value of 0.0033. The results of IWT score reported as a categorical variable are reported in figure 1. The IWT score demonstrated moderate preditive power as a categorical variable. The chi squared statistic was 14.95(p: 0.0006). The negative predictive value of an IWT score ≤2 was 86%. the PPV of an IWT score of 5 was 83.3%. ROC curve analysis of LA strain parameters showed that both reservoir strain and contractile strain offered moderate predictive value with an AUC of 0.71. (Figure 2). Conclusion The IWT score, and left atrial strain, both showed moderate diagnostic accuracy for CA. A low IWT score has high negative predictive value. However, 5/17 patients with cardiac amyloidosis had low IWT score. Our results suggest that the IWT score is useful, but a meaningful proportion of patients with cardiac amyloidosis will have subtle echo findings with low IWT scores.IWT score