David J Murphy1, Sinead H McEvoy2, Sri Iyengar3, Gudrun Feuchtner4, Ricardo C Cury5, Carl Roobottom6, Stephan Baumueller7, Hatem Alkadhi8, Jonathan D Dodd9. 1. Department of Radiology, St Vincent's University Hospital, Elm Park, Dublin 4, Ireland. Electronic address: david.murphy@st-vincents.ie. 2. Department of Radiology, St Vincent's University Hospital, Elm Park, Dublin 4, Ireland. Electronic address: s.mcevoy@st-vincents.ie. 3. Department of Radiology, Plymouth Hospitals NHS Trust, Plymouth Devon PL6 8DH, United Kingdom. Electronic address: sri.iyengar@nhs.net. 4. Department of Radiology, Innsbruck Medical University, Anichstr. 35, A-6020 Innsbruck, Austria. Electronic address: Gudrun.Feuchtner@i-med.ac.at. 5. Department of Radiology, Baptist Cardiac and Vascular Institute, 8900 North Kendall Drive, Miami, FL 33176, United States. Electronic address: r.cury@baptisthealth.net. 6. Department of Radiology, Plymouth Hospitals NHS Trust, Plymouth Devon PL6 8DH, United Kingdom; Plymouth University Peninsula Schools of Medicine and Dentistry. Electronic address: carl.roobottom@nhs.net. 7. Institute for Diagnostic and Interventional Radiology, University Hospital Zurich, Raemistrasse 100, CH-8091 Zurich, Switzerland. Electronic address: Hatem.Alkadhi@usz.ch. 8. Institute for Diagnostic and Interventional Radiology, University Hospital Zurich, Raemistrasse 100, CH-8091 Zurich, Switzerland. Electronic address: stephan.baumueller@usz.ch. 9. Department of Radiology, St Vincent's University Hospital, Elm Park, Dublin 4, Ireland. Electronic address: jonniedodd@gmail.com.
Abstract
OBJECTIVES: To assess the diagnostic accuracy of standard axial 64-slice chest CT compared to aortic valve image plane ECG-gated cardiac CT for bicuspid aortic valves. MATERIALS AND METHODS: The standard axial chest CT scans of 20 patients with known bicuspid aortic valves were blindly, randomly analyzed for (i) the appearance of the valve cusps, (ii) the largest aortic sinus area, (iii) the longest aortic cusp length, (iv) the thickest aortic valve cusp and (v) valve calcification. A second blinded reader independently analyzed the appearance of the valve cusps. Forty-two age- and sex-matched patients with known tricuspid aortic valves were used as controls. Retrospectively ECG-gated cardiac CT multiphase reconstructions of the aortic valve were used as the gold-standard. RESULTS: Fourteen (21%) scans were scored as unevaluable (7 bicuspid, 7 tricuspid). Of the remainder, there were 13 evaluable bicuspid valves, ten of which showed an aortic valve line sign, while the remaining three showed a normal Mercedes-Benz appearance owing to fused valve cusps. The 35 evaluable tricuspid aortic valves all showed a normal Mercedes-Benz appearance (P=0.001). Kappa analysis=0.62 indicating good interobserver agreement for the aortic valve cusp appearance. Aortic sinus areas, aortic cusp lengths and aortic cusp thicknesses of ≥ 3.8 cm(2), 3.2 cm and 1.6mm respectively on standard axial chest CT best distinguished bicuspid from tricuspid aortic valves (P<0.0001 for all). Of evaluable scans, the sensitivity, specificity, positive and negative predictive values of standard axial chest CT in diagnosing bicuspid aortic valves was 77% (CI 0.54-1.0), 100%, 100% and 70% respectively. CONCLUSION: The aortic valve is evaluable in approximately 80% of standard chest 64-slice CT scans. Bicuspid aortic valves may be diagnosed on evaluable scans with good diagnostic accuracy. An aortic valve line sign, enlarged aortic sinuses and elongated, thickened valve cusps are specific CT features.
OBJECTIVES: To assess the diagnostic accuracy of standard axial 64-slice chest CT compared to aortic valve image plane ECG-gated cardiac CT for bicuspid aortic valves. MATERIALS AND METHODS: The standard axial chest CT scans of 20 patients with known bicuspid aortic valves were blindly, randomly analyzed for (i) the appearance of the valve cusps, (ii) the largest aortic sinus area, (iii) the longest aortic cusp length, (iv) the thickest aortic valve cusp and (v) valve calcification. A second blinded reader independently analyzed the appearance of the valve cusps. Forty-two age- and sex-matched patients with known tricuspid aortic valves were used as controls. Retrospectively ECG-gated cardiac CT multiphase reconstructions of the aortic valve were used as the gold-standard. RESULTS: Fourteen (21%) scans were scored as unevaluable (7 bicuspid, 7 tricuspid). Of the remainder, there were 13 evaluable bicuspid valves, ten of which showed an aortic valve line sign, while the remaining three showed a normal Mercedes-Benz appearance owing to fused valve cusps. The 35 evaluable tricuspid aortic valves all showed a normal Mercedes-Benz appearance (P=0.001). Kappa analysis=0.62 indicating good interobserver agreement for the aortic valve cusp appearance. Aortic sinus areas, aortic cusp lengths and aortic cusp thicknesses of ≥ 3.8 cm(2), 3.2 cm and 1.6mm respectively on standard axial chest CT best distinguished bicuspid from tricuspid aortic valves (P<0.0001 for all). Of evaluable scans, the sensitivity, specificity, positive and negative predictive values of standard axial chest CT in diagnosing bicuspid aortic valves was 77% (CI 0.54-1.0), 100%, 100% and 70% respectively. CONCLUSION: The aortic valve is evaluable in approximately 80% of standard chest 64-slice CT scans. Bicuspid aortic valves may be diagnosed on evaluable scans with good diagnostic accuracy. An aortic valve line sign, enlarged aortic sinuses and elongated, thickened valve cusps are specific CT features.
Authors: Mathias Hillebrand; Dietmar Koschyk; Pia Ter Hark; Helke Schüler; Meike Rybczynski; Jürgen Berger; Amit Gulati; Alexander M Bernhardt; Christian Detter; Evaldas Girdauskas; Stefan Blankenberg; Yskert von Kodolitsch Journal: Cardiovasc Diagn Ther Date: 2017-08