Bioimpedance Spectroscopy Reveals Important Association of Fluid Status and T(1) ‐Mapping by Cardiovascular Magnetic Resonance

BACKGROUND: Extracellular matrix expansion is a key pathophysiologic feature in heart failure and can be quantified noninvasively by cardiac magnetic resonance T(1)‐mapping. Free water within the interstitial space of the myocardium, however, may also alter T(1)‐mapping results. PURPOSE: To investigate the association between systemic fluid status and T(1)‐mapping by cardiac magnetic resonance. STUDY TYPE: Prospective, observational single‐center study. POPULATION: Two‐hundred eighty‐five consecutive patients (44.4% female, 70.0 ± 14.9 years old) scheduled for cardiac MR due to various cardiac diseases. SEQUENCE AND FIELD STRENGTH: 1.5‐T scanner (Avanto Fit, Siemens Healthineers, Erlangen, Germany). For T(1) ‐mapping, electrocardiographically triggered modified‐Look‐Locker inversion (MOLLI) recovery sequence using a 5(3)3 prototype on a short‐axis mid‐cavity slice and with a four‐chamber view was performed. ASSESSMENTS: MR parameters including native myocardial T(1)‐times using MOLLI and extracellular volume (MR‐ECV) were assessed, and additionally, we performed bioimpedance analysis (BIA). Furthermore, demographic data and comorbidities were assessed. STATISTICS: Wilcoxon's rank‐sum test, chi‐square tests, and for correlation analysis, Pearson's correlation coefficients were used. Regression analyses were performed to investigate the association between patients' fluid status and T(1)‐mapping results. A P‐value <0.05 was considered statistically significant. RESULTS: The mixed cohort presented with a mean overhydration (OH) of +0.2 ± 2.4 liters, as determined by BIA. By MR, native T(1)‐times were 1038 ± 51 msec and MR‐ECV was 31 ± 9%. In the multivariable regression analysis, only OH was significantly associated with MR‐ECV (adj. beta: 0.711; 95% CI: 0.28 to 1.14) along with male sex (adj. beta: 2.529; 95% CI: 0.51 to 4.55). In linear as well as multivariable analysis, only OH was significantly associated with native T(1) times (adj. beta: 3.750; 95% CI: 1.27 to 6.23). CONCLUSION: T(1)‐times and MR‐ECV were significantly associated with the degree of OH on BIA measurement. These effects were independent from age, sex, body mass index, and hematocrit. Patients' volume status may thus be an important factor when T(1)‐time and MR‐ECV values are interpreted. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 3