Liberal Versus Restricted Fluid Administration in Heart Failure Patients

Restrictive fluid intake is recommended, in addition to standard pharmacologic treatment, in the treatment of patients with chronic heart failure (CHF). However, this recommendation lacks firm scientific evidence. We conducted a systematic review and meta-analysis of published randomized controlled trials to estimate the effect of fluid restriction in patients with heart failure.Randomized controlled trials were identified in the MEDLINE, EMBASE, and Cochrane databases using the search-keywords "fluid" and "heart failure". Outcomes were compared in heart failure patients with liberal and restricted fluid intake. Pooled risk ratios (RR) and weighted mean differences (WMD) were calculated using random effects models. Studies focusing on decompensated heart failure were analyzed separately.Six small randomized trials comparing liberal and restricted fluid intake met the inclusion criteria. Significant heterogeneity was noted in the reported studies for several outcomes. There were no differences in readmission rate (5 studies, pooled RR = 1.32; 95% CI: 0.86 to 2.01; P = 0.2), mortality rate (5 studies, pooled RR = 1.50; 95% CI: 0.87 to 2.57; P = 0.14), perceived thirst (4 studies, WMD = -0.7; 95% CI: -2.58 to 1.17; P = 0.46), duration of intravenous diuretics (2 studies, WMD = 0.17; 95% CI: -1.26 to 1.6; P = 0.81) or serum sodium levels (WMD = -1.61; 95% CI: -3.28 to 0.07; P = 0.06) between the liberal fluid intake group and the restrictive fluid intake group. Mean serum creatinine and BNP levels were significantly higher in the liberal fluid group: WMD 0.20 (95% CI: 0.15 to 0.25; P < 0.00001) and 172.59 (95% CI: 67.38 to 277.8; P = 0.001), respectively. There was no difference in any of the outcomes after correcting for heterogeneity.While studies to date are limited by heterogeneity and small sample sizes, the combined data suggest similar clinical outcomes in patients with CHF managed with liberal and restrictive fluid intake. Larger studies are needed to confirm our findings.