Giovanni Foti1, Ronaldo Silva2, Niccolò Faccioli3, Alessandro Fighera3, Rossella Menghini3, Arianna Campagnola3, Giovanni Carbognin4. 1. Department of Radiology, IRCCS Sacro Cuore Don Calabria Hospital, Via Don A. Sempreboni 1, 37024, Negrar, VR, Italy. gfoti81@yahoo.it. 2. Clinical Research Unit, IRCCS Sacro Cuore Don Calabria Hospital, Negrar, Italy. 3. Department of Radiology, Verona University Hospital, Verona, Italy. 4. Department of Radiology, IRCCS Sacro Cuore Don Calabria Hospital, Via Don A. Sempreboni 1, 37024, Negrar, VR, Italy.
Abstract
OBJECTIVES: To evaluate the diagnostic accuracy of venous-phase dual-energy computed tomography (VP-DECT) in the identification of PE compared with standard CT pulmonary angiography (CTPA). METHODS: This prospective IRB-approved study included 61 consecutive oncology patients (35 females, 26 males, mean age 66.91 years) examined by CTPA and VP-DECT. DECT data were post-processed on a SyngoVia workstation to obtain monoenergetic images (MEI+). The diagnosis of PE was based on the presence of any vascular perfusion defects. DECT images were evaluated independently by two radiologists (8 and 16 years of experience). A consensus reading of CTPA images (two senior radiologists, 18 and 24 years of experience) represented the reference for diagnosis. The diagnostic accuracy values of VP-DECT on a per-patient and per-lobe basis were assessed. Interobserver agreement was calculated using k-statistics. A value of p < 0.05 was considered statistically significant. RESULTS: Thirty of 61 patients (49.18%) were diagnosed with PE by CTPA, with 57/366 lobes being involved (15.57%). The sensitivity and specificity of the per-patient analysis of VP-DECT images were 90.0% (27/30) and 100% (31/31) respectively, for both readers. As concerns the per-lobe analysis, the sensitivity ranged from 100% for the right lower lobe to 50% for the left upper lobe for reader 1, and from 100% for the left upper lobe to 69.23% for the lingula for reader 2. The interobserver agreement ranged from 0.8671 (patients' analysis) to 0.6419 (lobes' analysis). CONCLUSION: VP-DECT could be considered an accurate imaging tool for diagnosing PE in a selected, high-prevalence population, compared with CTPA. KEY POINTS: • With regard to the patients' analysis, venous-phase DECT sensitivity and specificity in diagnosing pulmonary embolism were 90% and 100%, respectively, for both readers. • With regard to the lobes' analysis, the sensitivity ranged from 100 to 50%, for reader 1, and from 100 to 69.23%, for reader 2, respectively. • The sensitivity and specificity of lung perfusion maps obtained from venous DECT were 73.33% and 67.74% as concerns the patients' analysis and 71.92% and 75.72% as regards the lobes' analysis, respectively.
OBJECTIVES: To evaluate the diagnostic accuracy of venous-phase dual-energy computed tomography (VP-DECT) in the identification of PE compared with standard CT pulmonary angiography (CTPA). METHODS: This prospective IRB-approved study included 61 consecutive oncology patients (35 females, 26 males, mean age 66.91 years) examined by CTPA and VP-DECT. DECT data were post-processed on a SyngoVia workstation to obtain monoenergetic images (MEI+). The diagnosis of PE was based on the presence of any vascular perfusion defects. DECT images were evaluated independently by two radiologists (8 and 16 years of experience). A consensus reading of CTPA images (two senior radiologists, 18 and 24 years of experience) represented the reference for diagnosis. The diagnostic accuracy values of VP-DECT on a per-patient and per-lobe basis were assessed. Interobserver agreement was calculated using k-statistics. A value of p < 0.05 was considered statistically significant. RESULTS: Thirty of 61 patients (49.18%) were diagnosed with PE by CTPA, with 57/366 lobes being involved (15.57%). The sensitivity and specificity of the per-patient analysis of VP-DECT images were 90.0% (27/30) and 100% (31/31) respectively, for both readers. As concerns the per-lobe analysis, the sensitivity ranged from 100% for the right lower lobe to 50% for the left upper lobe for reader 1, and from 100% for the left upper lobe to 69.23% for the lingula for reader 2. The interobserver agreement ranged from 0.8671 (patients' analysis) to 0.6419 (lobes' analysis). CONCLUSION: VP-DECT could be considered an accurate imaging tool for diagnosing PE in a selected, high-prevalence population, compared with CTPA. KEY POINTS: • With regard to the patients' analysis, venous-phase DECT sensitivity and specificity in diagnosing pulmonary embolism were 90% and 100%, respectively, for both readers. • With regard to the lobes' analysis, the sensitivity ranged from 100 to 50%, for reader 1, and from 100 to 69.23%, for reader 2, respectively. • The sensitivity and specificity of lung perfusion maps obtained from venous DECT were 73.33% and 67.74% as concerns the patients' analysis and 71.92% and 75.72% as regards the lobes' analysis, respectively.
Authors: Giovanni Foti; William Mantovani; Niccolò Faccioli; Giacomo Crivellari; Luigi Romano; Claudio Zorzi; Giovanni Carbognin Journal: Radiol Med Date: 2020-08-25 Impact factor: 3.469