Defining Biventricular Abnormalities by Cardiac Magnetic Resonance in Pre-Dialysis Patients with Chronic Kidney Disease

INTRODUCTION: The aim of the study was to investigate biventricular structural and functional abnormalities in pre-dialysis patients across stages of chronic kidney disease (CKD) by cardiac magnetic resonance (CMR). METHODS: Fifty-one CKD patients with CMR exams were retrospectively analyzed. Patients were divided into three groups according to estimated glomerular filtration rate (eGFR): CKD 1 group (patients with normal eGFR≥90 mL/min/1.73 m(2), n = 20), CKD 2-3 group (patients with eGFR< 90 to ≥30 mL/min/1.73 m(2), n = 14), and CKD 4-5 group (patients with eGFR<30 mL/min/1.73 m(2), n = 17). Twenty-one age- and sex-matched healthy controls (HC) were recruited. CMR-derived left ventricular (LV) and right ventricular (RV) structural and functional measures were compared. Association between CMR parameters and clinical measures was assessed. RESULTS: There was an increasing trend in RV mass index (RVMi) and LV mass index (LVMi) with the occurrence and development of CKD from HC group to CKD 4-5 group although no significant difference was observed between CKD 1 group and HC group. LV global radial strain and LV global circumferential strain dropped and native T1 value elevated significantly in CKD 4-5 group compared with the other three groups (all p < 0.05), while RV strain measures, RV ejection fraction, and LV ejection fraction showed no significant difference among 4 groups (all p > 0.05). Elevated LV end-diastolic volume index (β = 0.356, p = 0.016) and RV end-systolic volume index (β = 0.488, p = 0.001) were independently associated with RVMi. Increased systolic blood pressure (β = 0.309, p = 0.004), LV end-systolic volume index (β = 0.633, p < 0.001), and uric acid (β = 0.261, p = 0.013) were independently associated with LVMi. Meanwhile, serum phosphorus (β = 0.519, p = 0.001) was independently associated with native T1 value. CONCLUSION: In pre-dialysis CKD patients, left and right ventricular remolding has occurred. RVMi and LVMi were the first changed CMR indexes in the development of CKD when eGFR began to drop. Because fluid volume overload was the independent risk factor for RVMi and LVMi increase, reasonable controlling fluid volume overload may slow down the progression of biventricular remolding and may reduce related cardiovascular disease risk.