Sixty patients (mean age 80 years) with cognitive impairment who requi
red parenteral fluids for at least 48 h were randomized to receive eit
her intravenous (i.v.) or subcutaneous (s.c.) fluids. There was no sig
nificant difference in the mean volume of fluid prescribed over 48 h i
n the two groups (s.c. 3.3 litres vs. i.v, 3.6 litres) or in the propo
rtion of prescribed fluids actually administered (s.c. 0.82 vs. i.v. 0
.76). After adjusting for baseline differences, there was no differenc
e between serum urea or creatinine levels in the two groups at 48 h. A
gitation related to the infusion was reported in 11 (37%) patients rec
eiving s.c. fluids and 24 (80%) patients receiving i.v. fluids Ip < 0.
005). The cost of the cannulae used during the study was pound 6.80 fo
r the s.c. group and pound 28.70 for the i.v. group. Local oedema was
noted in 2 patients in the s.c. group and led to re-siting of the infu
sion in 1 patient. No other complication was noted. These results sugg
est that s.c. fluid therapy is the treatment of choice in nonurgent si
tuations for confused patients who require parenteral fluids.