Reproducibility and reliability of flow quantification using CMR 2D-phase contrast and 4D-Flow in secondary mitral valve regurgitation

Abstract

Accurate quantification of mitral valve regurgitation (MVR) is crucial for patient management. While different MVR quantification methods based on cardiac magnetic resonance imaging (CMR) exist, their reproducibility and reliability remain uncertain. This study aims to evaluate the reproducibility of different CMR 2D-phase contrast (PC) and 4D-flow MVR quantification methods. The inter-reader and intra-reader reproducibility were assessed using intraclass correlation coefficients (ICC). Seven methods were evaluated: 2D-PC standard (LVSV minus aortic flow), 2D-PC mitral-aortic (mitral inflow minus aortic flow), 2D-PC direct (quantifying mitral backflow), 4D-flow standard, 4D-flow mitral-aortic, 4D-flow direct, and volumetric method (LVSV minus RVSV) in 32 patients (74.8 ± 9.8 years, 28% females) with secondary MVR, analyzed independently by two experienced readers. A total of 26 patients were included in the analysis for 2D-PC and 15 for 4D-flow methods. Among all techniques, 2D-PC standard was the most reliable method with both good inter-reader (ICC = 0.85, p < 0.001) and intra-reader agreement (ICC = 0.87, p < 0.001). The 4D-flow standard (ICC = 0.97, p < 0.001) and the volumetric method (ICC = 0.81, p < 0.001) showed excellent and good intra-reader agreements, respectively, but only moderate inter-reader reproducibility (ICC = 0.52, p = 0.027 and ICC = 0.71, p < 0.001). In patients with secondary MVR, 2D-PC standard method demonstrated the highest reproducibility, while 4D-flow methods showed excellent intra-reader reliability but more variable inter-reader agreement. Standardized post-processing protocols and training would likely enhance the clinical application of these techniques. Future studies should investigate these methods in larger, diverse cohorts and correlate findings with clinical outcomes.

Keywords: 2D-PC; 4D-flow; Cardiovascular; Challenges; Data analysis; Flow; Heart; Validation; Valves; Velocity & flow.

Fig. 1

Illustration of MVR quantification methods. 2D-PC_standard, CMR flow gold standard (Left Ventricle Stroke Volume [LV SV]– Aortic Forward Flow Volume derived from 2D-PC plane [AoPC]); 2D-PC_MVAAo, Forward Flow through Mitral Valve (MVPC) - AoPC; 2D-PC_MVdirect, directly quantifying flow through Mitral Valve (Backward flow through Mitral Valve [RegurgVol_MV]); Volumetric (LV SV– Right Ventricle Stroke Volume [RV SV]); 4D-flow_standard, similar to 2D-PC_standard using 4D-flow sequence instead of 2D-PC (LV SV– AoPC); 4D-flow_MVAAo, similar to 2D-PC_MVAAo using 4D-flow sequence instead of 2D-PC (MVPC– AoPC); 4D-flow_MVdirect, directly quantifying flow through Mitral Valve (Backward flow through Mitral Valve [RegurgVol_MV]); AoPC, Aortic Forward Flow (Outflow_AAo); MVPC, Mitral Valve Forward Flow (Inflow_MV); EDV, End Diastolic Volume; ESV, End Systolic Volume

Fig. 2

Box plots of Mitral Valve Regurgitation (MVR) volumes measured using various CMR flow quantification methods. Measurements were taken with seven distinct methods, based on a cohort comprising 26 patients with 2D-PC data, of whom 15 patients also underwent 4D-flow acquisition. Each method was analyzed by two readers: once by reader A and twice by the same reader (B1 and B2). Outliers (Z-score > 2) were removed from the data to enhance clarity. Each method is illustrated with its corresponding box plot, with individual measurement points overlayed in red

Fig. 3

Heatmap of Intra-Reader and Inter-Reader Intraclass Correlation Coefficient (ICC) values for MVR volume measurements across various CMR flow quantification methods. The analysis includes the same 15 patients who underwent both 2D-PC and 4D-flow acquisitions. Each cell displays the ICC value, with p-values annotated below. Methods include both 2D and 4D-flow measurements, as well as Volumetric analysis, comparing reliability between repeated measurements by the same reader (Intra-Reader) and measurements by different readers (Inter-Reader). The color scale represents ICC values, with darker red colors indicating stronger reliability and darker blue colors indicating weaker reliability. (ICC > = 0.9, excellent reliability; ICC = 0.75–0.9, good reliability; ICC = 0.5–0.75, moderate reliability; ICC < 0.5, poor reliability.)

Fig. 4

Bland-Altman plots comparing inter- and intra-reader measurements of MVR volumes across various CMR flow quantification methods. The analysis includes 26 patients for 2D-PC methods and 15 patients for 4D-flow methods. Each plot displays the mean of two measurements on the x-axis and the difference between them on the y-axis, with limits set from 0 to 80 on the x-axis and − 80 to + 80 on the y-axis. Horizontal dashed lines indicate the mean difference (gray) and the limits of agreement (± 1.96 SD, red). Inter-reader plots (comparing two different readers) are shown first for each method, followed by intra-reader plots (comparing repeated measurements by the same reader), providing insights into the consistency of measurements across and within readers for each quantification method

Fig. 5

MVR severity classification of mitral valve regurgitation (MVR) across various CMR flow quantification methods. The stacked bar chart shows the distribution of a subset of cases, the patients with both 2D-PC and 4D-flow acquisitions (n = 15), classified as Mild, Moderate, or Severe for each method. Percentages within the bars indicate the relative contribution of each severity category, providing a detailed comparison of the classification outcomes for different measurement techniques