Direct chest area measurement: A potential anthropometric replacement for BMI to inform cardiac CT dose parameters?

• Published in: Journal of cardiovascular computed tomography 2011-07, Vol.5 (4), p.240-246
• Main Authors: Ghoshhajra, Brian B., MD, MBA, Engel, Leif-Christopher, MD, Major, Gyöngyi Petra, Verdini, Dan, MD, Sidhu, Manavjot, MD, Károlyi, Mihály, MD, Abbara, Suhny, MD, Hoffmann, Udo, MD, MPH, Kalra, Manudeep, MD, Brady, Thomas J., MD
• Format: Article
• Language: English
• Subjects: Adult; Aged; Analysis of Variance; Body Mass Index; Boston; Cardiac CT angiography; Cardiovascular; Contrast-to-noise ratio; Coronary Angiography - methods; Coronary Artery Disease - diagnostic imaging; Female; Humans; Image quality; Linear Models; Linear regression; Male; Middle Aged; Predictive Value of Tests; Radiation Dosage; Radiation dose; Radiographic Image Interpretation, Computer-Assisted; Retrospective Studies; Second-generation dual-source CT; Thorax - anatomy & histology; Tomography, X-Ray Computed; Tube voltage
• ISSN: 1934-5925; 1876-861X
• DOI: 10.1016/j.jcct.2011.06.003

Background Significant dose reductions for coronary CT angiography (CTA) can be achieved with reduced tube potential (kV); however, the potential effect on image quality is unknown. Objective We sought to evaluate anthropometric measures (chest area, chest circumference, and chest attenuation) as potentially better predictors of the appropriate tube potential versus body mass index (BMI). Methods Consecutive patients (n = 183) who underwent routine coronary CTA (with standard department protocols, whereby tube potential was selected according to BMI) between April 2010 and October 2010 were included. All anthropometric measures were obtained by tracing a region of interest encompassing an entire axial full field-of-view image at the mid left atrial level. Linear regression was used to stratify patients into 4 chest area classes (very small to large). Patients were also stratified by standard BMI classes (underweight to obese). Qualitative and quantitative image quality parameters were compared between concordant and discordant chest area and BMI classes. Results A strong correlation was observed between patients’ BMI and chest area ( r = 0.84; P < 0.001) and between BMI and chest circumference ( r = 0.82, P < 0.001). Concordance between chest area class and BMI class was found in 61%, whereas 17.6% of patients were “potentially underdosed” (chest area class > BMI class) and 21.4% were “potentially overdosed” (chest area class < BMI class). Signal and contrast of the proximal coronaries and left ventricle were significantly different between the groups. Conclusion Patients’ chest area and BMI classes were frequently discordant, potentially leading to overdosing or underdosing when using BMI to select tube potential.