Fick versus flow: a real-time invasive cardiovascular magnetic resonance (iCMR) reproducibility study

Abstract

Background: Cardiac catheterization and cardiovascular magnetic resonance (CMR) imaging have distinct diagnostic roles in the congenital heart disease (CHD) population. Invasive CMR (iCMR) allows for a more thorough assessment of cardiac hemodynamics at the same time under the same conditions. It is assumed but not proven that iCMR gives an incremental value by providing more accurate flow quantification.

Methods: Subjects with CHD underwent real-time 1.5 T iCMR using a passive catheter tracking technique with partial saturation pulse of 40° to visualize the gadolinium-filled balloon, CMR-conditional guidewire, and cardiac structures simultaneously to aid in completion of right (RHC) and left heart catheterization (LHC). Repeat iCMR and catheterization measurements were performed to compare reliability by the Pearson (PCC) and concordance correlation coefficients (CCC).

Results: Thirty CHD (20 single ventricle and 10 bi-ventricular) subjects with a median age and weight of 8.3 years (2-33) and 27.7 kg (9.2-80), respectively, successfully underwent iCMR RHC and LHC. No catheter related complications were encountered. Time taken for first pass RHC and LHC/aortic pull back was 5.1, and 2.9 min, respectively. Total success rate to obtain required data points to complete Fick principle calculations for all patients was 321/328 (98%). One patient with multiple shunts was an outlier and excluded from further analysis. The PCC for catheter-derived pulmonary blood flow (Qp) (0.89, p < 0.001) is slightly lower than iCMR-derived Qp (0.96, p < 0.001), whereas catheter-derived systemic blood flow (Qs) (0.62, p = < 0.001) was considerably lower than iCMR-derived Qs (0.94, p < 0.001). CCC agreement for Qp at baseline (C1-CCC = 0.65, 95% CI 0.41-0.81) and retested conditions (C2-CCC = 0.78, 95% CI 0.58-0.89) were better than for Qs at baseline (C1-CCC = 0.22, 95% CI - 0.15-0.53) and retested conditions (C2-CCC = 0.52, 95% CI 0.17-0.76).

Conclusion: This study further validates hemodynamic measurements obtained via iCMR. iCMR-derived flows have considerably higher test-retest reliability for Qs. iCMR evaluations allow for more reproducible hemodynamic assessments in the CHD population.

Keywords: Cardiac catheterization; Congenital heart disease; Device tracking; Interventional CMR; Magnetic resonance imaging; Reproducibility.

Fig. 1

Invasive cardiovascular magnetic resonance (iCMR) environment and equipment. A CMR team (zone 3) adjusting images with direct visualization of the interventionalist performing the iCMR procedure. B Depiction of sterile draping within the iCMR environment. C Interventionalist equipment includes an MR-conditional catheter and guidewire. The FDA cleared and CE marked guidewire has three passive markers, coated with nanoparticles, that produces a distinct susceptibility artifact (0 mm, 20 mm, and 40 mm from the tip). D Interventionalist performing an iCMR procedure (zone 4) with real-time CMR guidance on adjacent projector screen (E)

Fig. 2

Test–retest catheter-based Fick and CMR-based flow reliability. A series of two conditions were performed to evaluate intra- and inter-rater reliability between catheter-based Fick and CMR-derived flow hemodynamics. The first condition was baseline catheter-based Fick (right heart catheterization (RHC)/left heart catherization (LHC)) and CMR-based flow (pulmonic flow (Qp)/systemic flow (Qs)) measurements. The second condition was repeat measurement under the same conditions (Cath: RHC/LHC + CMR: Qp/Qs flows). Dashed white arrow – Gadolinium-filled balloon; Solid white arrow – MR-conditional guidewire; Blue line –CMR flow vector

Fig. 3

Basic subject demographics. (CoA = Coarctation of the aorta; PH = Pulmonary hypertension; iNO = Inhaled Nitric Oxide; TOF = tetralogy of Fallot; PA = Pulmonary artery; OHT = Orthotopic heart transplantation)

Fig. 4

Summary of basic iCMR hemodynamics. Comparison of mean and standard deviation (std dev) measurements for catheterization and CMR hemodynamics for condition 1 (C1) and condition 2 (C2). Qp = Pulmonary blood flow; Qs = Systemic blood flow

Fig. 5

iCMR intra-rater reliability testing. Bland–Altman plots depicting Pearson correlation coefficients (PCC) to measure test–retest reliability testing between conditions 1 (C1) and 2 (C2) for catheter-based Fick hemodynamics for (A) Qp (B) Qs and CMR-derived flow hemodynamics for (C) Qp and (D) Qs. Qp = Pulmonary blood flow; Qs = Systemic blood flow

Fig. 6

iCMR inter-rater reliability testing. Bland–Altman plots depicting concordance correlation coefficient (CCC) to measure agreement between catheter-based Fick and CMR-derived flow hemodynamics in condition 1 (C1) and condition 2 (C2) for (A) Qp and (B) Qs. Qp = Pulmonary blood flow; Qs = Systemic blood flow

Fig. 7

iCMR flow scatter plots. Comparison of Cath vs CMR blood flow measurements for (A) Qp and (B) Qs. Subjects are grouped based on their underlying anatomy (single ventricle and biventricular). C1 and C2 for each subject is connected by a black line. The dashed line represents an ideal linear relationship. The subjects where VO2 was assumed are outlined in red. Qp = Pulmonary blood flow; Qs = Systemic blood flow; SV = Single ventricle; BV = Biventricular; C1 = Condition 1; C2 = Condition 2; VO2 = Oxygen consumption