Impact of Overhydration on Left Ventricular Hypertrophy in Patients With Chronic Kidney Disease

OBJECTIVE: Volume overload is a frequent feature related to left ventricular hypertrophy (LVH) in dialysis patients, but its influence on patients with chronic kidney disease (CKD) not on dialysis has not been accurately uncovered. This article was to examine the relationship between overhydration (OH) and LVH in patients with CKD not yet on dialysis. METHODS: A total of 302 patients with CKD stages 1–4 were included. Participants were divided into different subgroups according to occurring LVH or not, and OH tertiles. Clinical and laboratory parameters were compared among groups. Spearman correlation analyses were adopted to explore the relationships of echocardiographic findings with the clinical and laboratory characteristics. Binary logistic regression models were performed to estimate the odds ratios (ORs) for the associations between OH and LVH. Restricted cubic splines were implemented to assess the possible non-linear relationship between OH and LVH. LVH was defined as left ventricular mass index (LVMI) >115 g/m(2) in men and >95 g/m(2) in women. RESULTS: Of the enrolled patients with CKD, the mean age was 45.03 ± 15.14 years old, 165 (54.6%) cases were men, and 65 (21.5%) cases had LVH. Spearman correlation analyses revealed that OH was positively correlated with LVMI (r = 0.263, P < 0.001). After adjustment for age, gender, diabetes, body mass index (BMI), systolic blood pressure (SBP), hemoglobin, serum albumin, estimated glomerular filtration rate (eGFR), and logarithmic transformation of urinary sodium and urinary protein, multivariate logistic regression analyses demonstrated that both the middle and highest tertile of OH was associated with increased odds of LVH [OR: 3.082 (1.170–8.114), P = 0.023; OR: 4.481 (1.332–15.078), P = 0.015, respectively], in comparison to the lowest tierce. Restricted cubic spline analyses were employed to investigate the relationship between OH and LVH, which unfolded a significant non-linear association (P for non-linear = 0.0363). Furthermore, patients were divided into two groups according to CKD stages. The multivariate logistic regression analyses uncovered that increased odds of LVH were observed in the middle and the highest tertile of OH [OR: 3.908 (0.975–15.670), P = 0.054; OR: 6.347 (1.257–32.054), P = 0.025, respectively] in patients with stages 1–2. CONCLUSION: These findings suggest that a higher level of OH was associated with a higher occurrence of LVH in patients with CKD not on dialysis, especially in patients with CKD stages 1–2.