The aim of this study was to compare the effectiveness of subcutaneous and intravenous fluid therapy in hydrating, elderly acute stroke patients. Thirty-four such patients, needing parenteral fluids because of impaired consciousness or dysphagia, were randomly allocated to receive either subcutaneous or intravenous fluids (2 litres of dextrose-saline/24 hours). Serum osmolality was measured before starting fluid therapy (Day 1) and on Days 2 and 3. An analysis of covariance of the osmolalities showed no statistical difference between the two groups (P = 0.12). The total cost of cannulae used over the 3 days for the subcutaneous route was approximately a third of that for the intravenous route. Complication rates were similar for the two groups. The results suggest that subcutaneous fluid therapy is an effective alternative to the intravenous route.