Diagnostic accuracy of an integrated multimodality imaging approach in patients with cardiac masses

Background The diagnostic evaluation of cardiac masses (CMs) remains challenging. A multimodality imaging approach is essential for accurate CMs characterization. However, a direct comparison of combined imaging strategies, including cardiac magnetic resonance (CMR), cardiac computed tomography (CCT), and 18F-Fluorodeoxyglucose positron emission tomography (18F-FDG PET), has never been provided for CMs. Objectives To compare the diagnostic accuracy of multiparametric scores at CMR, CCT, and 18F-FDG PET, individually and in combination, in determining the presence and the malignant nature of suspected CMs compared to histology. Methods This prospective study included patients with suspected CMs who underwent echocardiography, CCT, CMR, and 18F-FDG PET within two months from first medical contact. Definite diagnosis was confirmed by histology or radiological resolution after anticoagulation for thrombi. Diagnostic performance was assessed using receiver operating characteristic (ROC) curves and compared with the DeLong test. A Bonferroni correction was applied to account for multiple comparisons. Classification Tree Analysis evaluated the incremental diagnostic value of PET after CMR or CCT. Results Among 142 patients, 77 (54%) had malignant CMs. Echocardiography had the lowest diagnostic accuracy (Area Under the Curve, AUC, 0.77 [95% Confidence Interval, CI,: 0.71–0.84]), missing the detection of 12.6% of CMs. CMR was the most accurate modality (AUC 0.94 [95% CI: 0.90–0.98]), significantly outperforming CCT (AUC 0.85 [95% CI: 0.80–0.91], Bonferroni-adjusted p=0.024) and 18F-FDG PET (AUC 0.85 [95% CI: 0.79–0.91], Bonferroni-adjusted p=0.048). The combination of CMR and 18F-FDG PET achieved the highest diagnostic accuracy (AUC 0.97 [95% CI: 0.94–1.00]; Positive Predictive Value, PPV, 95.1%; Negative Predictive Value, NPV, 100%). CCT combined with 18F-FDG PET also showed incremental diagnostic accuracy (AUC 0.92 [95% CI: 0.87–0.97]; PPV 91.2%; NPV 93.5%). Classification Tree Analysis confirmed that the integration of 18F-FDG PET with CMR or CCT increased malignancy detection (97.2% and 92.3% correctly classified, respectively). Similarly, integrating CMR and CCT significantly improved accuracy (AUC 0.95 [95% CI: 0.91–0.99]; PPV 93.8%; NPV 96.8%) compared to CCT alone (Bonferroni- adjusted p=0.006). Conclusion Although CMR is the most accurate imaging modality, its combination with 18F-FDG PET in inconclusive cases offers the best diagnostic performance for malignancy detection. When CMR is not feasible, combining CCT with PET offers a reliable alternative.