Neville Gai PhD, Presenter: Nothing to Disclose

Veit Sandfort MD, Abstract Co-Author: Nothing to Disclose

Songtao Liu MD, Abstract Co-Author: Nothing to Disclose

Colin Yi BA, Abstract Co-Author: Nothing to Disclose

Joao A. C. Lima MD, Abstract Co-Author: Research Grant, Toshiba Corporation

David A. Bluemke MD, PhD, Abstract Co-Author: Research support, Siemens AG

To show that standard gadolinium dose administration (in mmol/kg) results in biased post-contrast T1 values which can result in erroneous interpretation of results even after normalizing for dose, time and GFR. After application of the corrected dose values based on accurate estimates of blood volume, the results show a dramatic change.

605 subjects who had undergone  T1 mapping using the MOLLI sequence were retroactively selected from the MESA study based on availability of a complete set of physiological parameters. Precontrast blood and myocardial  T1 values were corrected for heart rate bias. Post-contrast blood and myocardial T1 values obtained at time=12min and dose≈0.15 mmol/kg were normalized for slight differences in dose and for GFR using an analytical model. The given dose in mmol/kg was then adjusted based on an accurate estimate for blood volume proposed by Lemmens (Obesity Surgery 2006;16:773-776). The adjusted values of dose were again used to normalize post-contrast blood and myocardial T1 values  for dose and GFR. Post contrast blood T1 values were compared with subject BMI using Pearson correlation. As an example of how this bias can affect interpretation, we analyzed post-contrast myocardial T1 values before and after dose adjustment for subjects with metabolic syndrome (METS, n=177) and with no METS (n=428) using Student’s t-test.   

Post-contrast blood T1 showed a strong correlation with BMI (R = –0.257, P<0.001) prior to adjustment for dose (based on blood volume) and no significant correlation after correction based on Lemmens estimate for blood volume (R= –0.011, P=0.79). Post-contrast myocardial T1 showed significant differences between METS and non-METS groups (P<0.0001) prior to dose adjustment and no significance (P=0.96) after correction.      

Contrast based on mmol/kg overestimates dose at higher BMIs. A correction for dose based on accurate blood volume estimate is proposed. Multiple linear regression analysis models can reveal significant relationships and correct for covariates. However, there can be confusion whether both BMI (risk factor for METS) and T1 (measure of fibrosis) could be used as independent measures of METS. Our analysis helps circumvent an erroneous and confusing deduction.  

Standard contrast dosing (in mmol/kg) needs to be corrected for bias due to BMI which can lead to erroneous interpretation based on post contrast T1 values.

Gai, N, Sandfort, V, Liu, S, Yi, C, Lima, J, Bluemke, D, Incorrect Dose Administration in Cardiac T1 Studies: Correction and Application to a MESA Case Study Based on BMI.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14004421.html