Introduction:
CMR is considered the ‘gold standard’ for noninvasive LV and RV mass quantitation. This concept is predominantly based on GRE sequences yet never validated in humans undermining its credibility. SSFP offers superior image contrast potentially redefining the CMR gold standard for mass.
Objective:
To establish correlations between SSFP-CMR LV and RV mass vs. pathology mass using Htx pts.
Methods:
Over 3 years, 68 pts were evaluated including 48 HTx explants obtained immediately from the OR upon HTx. They were prepared (removed PM/AICD and/or LV/RVAD) and scanned
ex vivo
via SSFP for CMR LV/RV mass (1.5T GE, WI). Via an
A
utomatic
T
hresholding program for segmentation with
M
anual
T
rimming (ATMT) of extraneous cardiac tissue a 3D LV model was generated. The weight of total heart
including
papillary and trabeculae for LV and RV was measured via high-fidelity scale. Correlations were between 3D CMR vs. pathology with B-A plots. A separate cohort of 20 non-Htx pts underwent
in vivo
validation.
Results:
Of 48 Htx pts, 3 (6%) were excluded due to poor images (peri-surgical destruction) leaving 45 (94%) for analysis. Correlations: 1) Total CMR 3D mass (448±116g) vs. total pathology mass (444±118g; r=0.99, p<.001); 2) 3D CMR-LV mass (297±95g) vs. pathology-LV mass (308±96g; r =0.95, p<.001) 3) 3D CMR-RV mass (152±44g) vs pathology-RV mass (132±41g; r = 0.76, p<.001). The equation y = 1.01x - 6.6b regressed the LV (r = 0.95). Avg bias between CMR and pathology for total, LV and RV mass was +4.4g. For the validation cohort LVEDV r=0.99; LVESV r= 0.99.
Conclusions:
SSFP-CMR accurately defines total cardiac, LV and RV mass as compared to explanted hearts (despite variable surgical removal of instrumentation). Thus, while GRE was the original ‘gold standard’ for LV mass, SSFP, despite its universal acceptance as the ‘
de facto
gold standard is now formally validated in a first-ever human autopsy study. Further, the regression equation now
includes
, not
excludes
papillary and trabecular mass.