Multi-modality imaging discrepancies in key prognostic parameters in hypertrophic cardiomyopathy

Background Multimodality imaging plays a crucial role in evaluating prognosis in patients with hypertrophic cardiomyopathy (HCM). Intermodal discrepancies may have significant implications for risk stratification. Purpose Evaluate inter-modal reporting discrepancies of key parameters between transthoracic echocardiography (TTE) and cardiac magnetic resonance (CMR) in HCM patients. Methods Retrospective assessment of consecutive HCM patients with TTE and CMR performed within 6 months for reported maximum wall thickness (MWT), left atrial diameter (LAd), and presence of left ventricular apical aneurysm (LVAA). CMR reports were considered gold standard. Estimated 5-year risk of sudden cardiac death (SCD) was calculated using the European Society of Cardiology algorithm at the time of the CMR. Results Between Sept’16 and May’23, 1000 consecutive HCM patients (71% male, median age 60) with TTE and CMR within 6 months (median difference 2 months, IQR 1.0-3.5 months) were assessed. Of 793 patients with paired MWT data, median difference was 0.5mm (IQR –1.0 to 2.75mm, p<0.0001), (Figure 1). There was agreement to within 2mm in 61% of cases. However, TTE over and underestimated MWT by >2mm in 15% and 26% cases respectively, with a difference of ≥5mm reported in 19% cases. Furthermore, at the diagnostic threshold of 15mm, there were 110 (14%) false negatives for HCM diagnosis using TTE alone. There were 9 cases of massive hypertrophy (≥30mm) that were not identified by TTE compared to CMR. Of the 323 patients with paired LAd data, median difference was -1.0mm (IQR, -4 to 3mm, p=0.218). LVAA was identified by TTE in only 29/62 (46%) of cases (p<0.0001). Upon second read, spectral Doppler evidence of LVAA without clear reporting (n=29), burn out (n=4), microaneurysm (n=2), and suboptimal assessment of the apical cavity (n=3) contributed to false negative LVAA in TTE (Figure 2). In the 279 patients with paired MWT and LAd data, CMR parameter use produced a small but statistically significant increase in SCD risk score (1.6 vs 1.7%, p=0.002). A total of 7 patients (2.5%) were reclassified by one risk category depending on modality, with 66% upgraded by CMR. Conclusion Clinically significant intermodal discrepancies exist in key imaging parameters in a sizeable proportion of HCM patients with potentially important ramifications for diagnosis and prognostication. Accurate reporting of pathognomonic spectral Dopplers can increase confidence of LVAA identification on TTE.