211 The diuretic effect of steroids in pateints with heart failure: a rapid review, narrative synthesis and meta-analysis

Introduction Steroids may have a diuretic effect. It is unknown whether they are a useful treatment for venous congestion due to heart failure (HF). Methods A rapid systematic review was conducted following PRISMA guidelines (PROSPERO CRD420200620029). Medline and CENTRAL databases were searched for studies involving adult HF patients receiving steroid. Randomized and observational studies were included. For those data on which meta-analysis (MA) was possible, a random effects model was used to calculate Cohen’s D and unstandardised mean difference (UMD) for continuous variables, and odds and log odds ratios (OR) for categorical variables. Risk of bias was evaluated using ROB-2 and ROBINS-I tools. Results From an initial 4349 records, 17 observational studies, and 6 randomised controlled trials (RCT) involving 22,815 patients were included (table 1). Oral steroid was associated with greater weight loss (Cohen’s D 0.69 (0.11 – 1.27); P=0.02), although the absolute differences were small (UMD 0.94 kg (standard error = 0.51; P=0.06) (figure 1). Urine volume and natriuresis also tended to be greater with steroid treatment but no MA was possible. Treatment with steroid had no effect on renal function or electrolyte concentrations. Hyperglycaemia was the most common side effect (4-35%) with an OR of 11.19 ((3.22 – 38.90); P