Extracellular volume quantification by cardiac magnetic resonance imaging without hematocrit sampling : Ready for prime time?

BACKGROUND

Myocardial tissue characterization by cardiovascular magnetic resonance (CMR) T1 mapping currently receives increasing interest as a diagnostic tool in various disease settings. The T1-mapping technique allows non-invasive estimation of myocardial extracellular volume (ECV) using T1-times before and after gadolinium administration; however, for calculation of the myocardial ECV the hematocrit is needed, which limits its utility in routine application. Recently, the alternative use of the blood pool T1-time instead of the hematocrit has been described.

METHODS

The results of CMR T1 mapping data of 513 consecutive patients were analyzed for this study. Blood for hematocrit measurement was drawn when placing the i. v. line for contrast agent administration. Data from the first 200 consecutive patients (derivation cohort) were used to establish a regression formula allowing synthetic hematocrit calculation, which was then validated in the following 313 patients (validation cohort). Synthetic ECV was calculated using synthetic hematocrit, and was compared with conventionally derived ECV.

RESULTS

Among the entire cohort of 513 patients (mean age 57.4 ± 17.5 years old, 49.1% female) conventionally measured hematocrit was 39.9 ± 4.7% and native blood pool T1-time was 1570.6 ± 117.8 ms. Hematocrit and relaxivity of blood (R1 = 1/blood pool T1 time) were significantly correlated (r = 0.533, r = 0.284, p < 0.001). By linear regression analysis, the following formula was developed from the derivation cohort: synthetic hematocrit = 628.5 × R1 - 0.002. Synthetic and conventional hematocrit as well as ECV showed significant correlation in the validation (r = 0.533, r = 0.284, p < 0.001 and r = 0.943, r = 0.889, p < 0.001, respectively) as well as the overall cohort (r = 0.552, r = 0.305, p < 0.001 and r = 0.957, r = 0,916, p < 0.001). By Bland Altman analysis, good agreement between conventional and synthetic ECV was found in the validation cohort (mean difference: 0.007%, limits of agreement: -4.32 and 4.33%, respectively).

CONCLUSION

Synthetic ECV using native blood pool T1-times to calculate the hematocrit, is feasible and leads to almost identical results in comparison with the conventional method. It may allow fully automatic ECV-mapping and thus enable broader use of ECV by CMR T1 mapping in clinical practice.