Mayur Virarkar1, Dhakshinamoorthy Ganeshan2, Catherine Devine2, Roland Bassett3, Vishnupriya Kuchana2, Priya Bhosale2. 1. Departmment of Diagnostic Radiology, The University of Texas MD Anderson Cancer Center, Houston, TX, United States of America. Electronic address: mkvirarkar@mdanderson.org. 2. Departmment of Diagnostic Radiology, The University of Texas MD Anderson Cancer Center, Houston, TX, United States of America. 3. Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX, United States of America.
Abstract
PURPOSE: To perform a meta-analysis of the literature to compare the diagnostic performance of 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) versus 18F-FDG positron emission tomography/magnetic resonance imaging (PET/MRI) for gynecological malignancies of the pelvis. MATERIALS AND METHODS: We searched for English-language studies, published through May 2019, on the diagnostic accuracy of PET/CT and PET/MRI for gynecological malignancies. To reduce inter-study heterogeneity, we focused primarily on studies in which both PET/CT and PET/MRI were performed on the entire study cohort. We pooled the sensitivity, specificity, diagnostic odds ratio, and area-under-the-receiver-operating-characteristic curve values for PET/CT and PET/MRI and determined the 95% confidence intervals (CIs). RESULTS: Out of 30 studies, nine met the inclusion criteria. On patient-based analysis, PET/CT had a pooled sensitivity and specificity of 62.6% (95% CI: 47.1%-76%) and 91.6% (95% CI: 81.9%-96.3%), respectively, compared with 73.3% (95% CI: 63.1%-81.6%) (P = 0.22) and 91.2% (95% CI: 81.8%-96%) (P = 94) for PET/MRI. On lesion-based analysis, PET/CT had a pooled sensitivity and specificity of 81.5% (95% CI: 59.3%-93.1%) and 86.6% (95% CI: 77.3%-92.5%), respectively, compared with 84.7% (95% CI: 66.8%-93.8%) (P = 0.77) and 89.3% (95% CI: 85.2%-92.3%) (P = 0.51) for PET/MRI. The diagnostic odds ratios for PET/CT compared with PET/MRI were not significantly different in the patient-based (P = 0.48) and lesion-based analyses (P = 0.4). CONCLUSION: Compared with PET/CT, PET/MRI had slightly better diagnostic performance to that of 18F-FDG PET/CT in the gynecological malignancies on lesion level (44 vs 26) and patient level analysis (28 vs 17). However, the differences between results showed no statistical significance (P = 0.4 and 0.48, respectively).
PURPOSE: To perform a meta-analysis of the literature to compare the diagnostic performance of 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) versus 18F-FDG positron emission tomography/magnetic resonance imaging (PET/MRI) for gynecological malignancies of the pelvis. MATERIALS AND METHODS: We searched for English-language studies, published through May 2019, on the diagnostic accuracy of PET/CT and PET/MRI for gynecological malignancies. To reduce inter-study heterogeneity, we focused primarily on studies in which both PET/CT and PET/MRI were performed on the entire study cohort. We pooled the sensitivity, specificity, diagnostic odds ratio, and area-under-the-receiver-operating-characteristic curve values for PET/CT and PET/MRI and determined the 95% confidence intervals (CIs). RESULTS: Out of 30 studies, nine met the inclusion criteria. On patient-based analysis, PET/CT had a pooled sensitivity and specificity of 62.6% (95% CI: 47.1%-76%) and 91.6% (95% CI: 81.9%-96.3%), respectively, compared with 73.3% (95% CI: 63.1%-81.6%) (P = 0.22) and 91.2% (95% CI: 81.8%-96%) (P = 94) for PET/MRI. On lesion-based analysis, PET/CT had a pooled sensitivity and specificity of 81.5% (95% CI: 59.3%-93.1%) and 86.6% (95% CI: 77.3%-92.5%), respectively, compared with 84.7% (95% CI: 66.8%-93.8%) (P = 0.77) and 89.3% (95% CI: 85.2%-92.3%) (P = 0.51) for PET/MRI. The diagnostic odds ratios for PET/CT compared with PET/MRI were not significantly different in the patient-based (P = 0.48) and lesion-based analyses (P = 0.4). CONCLUSION: Compared with PET/CT, PET/MRI had slightly better diagnostic performance to that of 18F-FDG PET/CT in the gynecological malignancies on lesion level (44 vs 26) and patient level analysis (28 vs 17). However, the differences between results showed no statistical significance (P = 0.4 and 0.48, respectively).