Synthetic hematocrit derived from the longitudinal relaxation of blood can lead to clinically significant errors in measurement of extracellular volume fraction in pediatric and young adult patients

Frank J. Raucci, David A. Parra, Jason T. Christensen, Lazaro E. Hernandez, Larry W. Markham, Meng Xu, James C. Slaughter, Jonathan H. Soslow

Extracellular volume fraction (ECV) is altered in pathological cardiac remodeling and predicts death and arrhythmia. ECV can be quantified using cardiovascular magnetic resonance (CMR) T1 mapping but calculation requires a measured hematocrit (Hct). The longitudinal relaxation of blood has been used in adults to generate a synthetic Hct (estimate of true Hct) but has not been validated in pediatric populations. One hundred fourteen children and young adults underwent a total of 163 CMRs with T1 mapping. The majority of subjects had a measured Hct the same day (N = 146). Native and post-contrast T1 were determined in blood pool, septum, and free wall of mid-LV, avoiding areas of late gadolinium enhancement. Synthetic Hct and ECV were calculated and intraclass correlation coefficient (ICC) and linear regression were used to compare measured and synthetic values. The mean age was 16.4 ± 6.4 years and mean left ventricular ejection fraction was 59% ± 9%. The mean measured Hct was 41.8 ± 3.0% compared to the mean synthetic Hct of 43.2% ± 2.9% (p < 0.001, ICC 0.46 [0.27, 0.52]) with the previously published model and 41.8% ± 1.4% (p < 0.001, ICC 0.28 [0.13, 0. 42]) with the locally-derived model. Mean measured mid-free wall ECV was 30.5% ± 4.8% and mean synthetic mid-free wall ECV of local model was 29.7% ± 4.6% (p < 0.001, ICC 0.93 [0.91, 0.95]). Correlations were not affected by heart rate and did not significantly differ in subpopulation analysis. While the ICC was strong, differences between measured and synthetic ECV ranged from -8.4% to 4.3% in the septum and -12.6% to 15.8% in the free wall. Using our laboratory's normal cut-off of 28.5%, 59 patients (37%) were miscategorized (53 false negatives, 6 false positives) with published model ECV. The local model had 37 miscategorizations (20 false negatives, 17 false positives), significantly fewer but still a substantial number (23%). Our data suggest that use of synthetic Hct for the calculation of ECV results in miscategorization of individual patients. This difference may be less significant once synthetic ECV is calculated and averaged over a large research cohort, making it potentially useful as a research tool. However, we recommend formal measurement of Hct in children and young adults for clinical CMRs.