Extracellular Volume Fraction Based on Cardiac Magnetic Resonance T1 Mapping: An Effective Way to Evaluate Cardiac Injury Caused by Cardiac Amyloidosis in Patients with Multiple Myeloma

Abstract

Multiple myeloma (MM) is a hematological malignancy of plasma cell origin. Cardiac amyloidosis (CA) is a common form of heart damage caused by MM and is associated with a poor prognosis. This study was a prospective cohort study and was aimed at evaluating the clinical predictive value of extracellular volume fraction (ECV) based on cardiovascular magnetic resonance (CMR) T1 mapping for cardiac amyloidosis and cardiac dysfunction in MM patients. Fifty-one newly diagnosed MM patients in Zhongnan Hospital of Wuhan University were enrolled in the study. A total of 19 patients (19/51; 37.25%) developed CA. The basal ECV of CA group was significantly higher than that of the non-CA group (p < 0.01). Multivariate logistic regression analysis showed that basal ECV (OR = 1.551, 95% CI 1.084-2.219, p < 0.05) and LDH1 level (OR = 1.150, 95% CI 1.010-1.310, p < 0.05) were two independent risk factors for CA. Further study demonstrated that basal ECV in the heart failure group was significantly higher than that of the nonheart failure group (p < 0.01). Notably, ROC curve showed that basal ECV had a good predictive value for CA and heart failure, with AUC of 0.911 and 0.893 (all p < 0.01), and the best cutoff values of 38.35 and 37.45, respectively. Taken together, basal ECV is a good predictor of CA and heart failure for MM patients.

Figure 1

CMR pre- and postcontrast T1 mapping and calculated ECV mapping of representative MM patients (a) without or (b) with CA.

Figure 2

The comparison of biochemical parameters between patients with or without CA. (a) β2-microglobulin (β2-MG) (CA, n = 19; non-CA, n = 32). (b) Lactate dehydrogenase 1 (LDH1) (CA, n = 19; non-CA, n = 32). (c) High-Sensitivity Troponin I (hs-TNI) (CA, n = 19; non-CA, n = 32). (d) Pro-brain natural peptide (pro-BNP) (CA, n = 19; non-CA, n = 32). (e) Left ventricular ejection fraction (LVEF) (CA, n = 19; non-CA, n = 32). (f) Myocardial mass (CF/BAS) (CA, n = 19; non-CA, n = 32). (g) Basal ECV (CA, n = 19; non-CA, n = 32). (h) Mid ECV (CA, n = 19; non-CA, n = 32). (i) Apical ECV (CA, n = 19; non-CA, n = 32).

Figure 3

The comparison of biochemical parameters between patients with or without heart failure. (a) ALB (heart failure, n = 30; nonheart failure, n = 21). (b) β2-MG (heart failure, n = 30; nonheart failure, n = 21). (c) Creatine Kinase-MB (CK-MB) (heart failure, n = 30; nonheart failure, n = 21). (d) hs-TNI (heart failure, n = 30; nonheart failure, n = 21). (e) pro-BNP (heart failure, n = 30; nonheart failure, n = 21). (f) End-diastolic volume (EDV)-CF/BAS (heart failure, n = 30; nonheart failure, n = 21). (g) End-systolic volume (ESV)-CF/BAS (heart failure, n = 30; nonheart failure, n = 21). (h) Myocardial mass (CF/BAS) (heart failure, n = 30; nonheart failure, n = 21). (i) Basal ECV (heart failure, n = 30; nonheart failure, n = 21). (j) Apical ECV (heart failure, n = 30; nonheart failure, n = 21). (k) Mid ECV (heart failure, n = 30; nonheart failure, n = 21). (l) Left ventricular ejection fraction (LVEF) (heart failure, n = 30; nonheart failure, n = 21).

Figure 4

The correlation between basal ECV and other laboratory parameters in 51 MM patients. (a) pro-BNP, (b) ALB, (c) β2-MG, (d) hs-TNI, and (e) LDH1. The correlated analysis between pro-BNP and LDH1 was shown (f). Spearman's correlation analysis and equation of residuals plots were shown.

Figure 5

ROC curve for predicting CA by (a) basal ECV, (b) hs-TNI, (c) LDH1, (d) pro-BNP, (e) LVEF, and (f) myocardial mass (CF/BSA).

Figure 6

ROC curve for predicting heart failure by (a) basal ECV, (b) ALB, (c) β2-MG, (d) CK-MB, (e) EDV (CF/BSA), (f) ESV (CF/BSA), (g) myocardial mass (CF/BSA), and (h) hs-TNI.