Original ArticleThick Maximum Intensity Projections for the Assessment of Left Ventricular Function With 64-Slice Computed TomographyHusmann, Lars MD*; Leschka, Sebastian MD*; Desbiolles, Lotus MD*; Schepis, Tiziano MD†; Koepfli, Pascal MD†; Gaemperli, Oliver MD†; Seifert, Burkhardt PhD‡; Flohr, Thomas G. PhD§; Frauenfelder, Thomas MD*; Marincek, Borut MD*; Kaufmann, Philipp A. MD†∥; Alkadhi, Hatem MD*Author Information From the *Institute of Diagnostic Radiology, the †Cardiovascular Center, the ‡Department of Biostatistics, and the ∥Center for Integrative Human Physiology, University of Zurich, Zurich, Switzerland; and §Siemens Medical Solutions, Computed Tomography CTE PA, Forchheim, Germany. Received May 5, 2006, and accepted for publication, after revision, June 30, 2006. This research has been supported by the National Center of Competence in Research, Computer Aided and Image Guided Medical Interventions (NCCR CO-ME) of the Swiss National Science Foundation. Reprints: Hatem Alkadhi, MD, Department of Medical Radiology, Institute of Diagnostic Radiology, University Hospital Zurich, Raemistrasse 100, CH-8091 Zurich, Switzerland. E-mail: [email protected]. Investigative Radiology: October 2006 - Volume 41 - Issue 10 - p 746-752 doi: 10.1097/01.rli.0000236906.05265.4f Buy Metrics Abstract Objective: The objective of this study was to assess the accuracy of thick maximum intensity projections (MIP) from computed tomography (CT) data sets mimicking projection images from biplane ventriculography for evaluation of left ventricular (LV) parameters. Materials and Methods: Fifty-eight patients underwent 64-slice CT. Multiphase images were reconstructed in 10% steps of the RR interval. MIP images (70-mm thickness) of the contrast-enhanced LV in fixed 30° right anterior oblique (RAO)/60° left anterior oblique (LAO) and in adapted short-/long-axis planes were reconstructed. LV parameters were calculated using the area–length method formula. Three-dimensional assessment with semiautomated software served as reference standard. Results: Use of thick MIP reconstructions had a high intermethod reliability (86–94%) compared with the 3-dimensional approach. Smaller measurement errors were found for thick MIP reconstructions in adapted short-/long-axis planes. A significant projection error (3.0%, P < 0.001) of thick MIP reconstructions was found using fixed 30° RAO/60° LAO compared with adapted short-/long-axis reconstructions. Conclusion: Thick MIP reconstructions with adapted short-/long-axis planes allow an accurate assessment of LV parameters compared with the established 3-dimensional method. © 2006 Lippincott Williams & Wilkins, Inc.