The aim of the study was to investigate the offset time of atracurium when
given by continuous infusion on a paediatric intensive care unit and to loo
k for evidence of tolerance. Over a period of 8 months, 20 mechanically ven
tilated children had a steady-state infusion of atracurium discontinued to
enable the assessment of their level of sedation. The offset time of atracu
rium was assessed by train-of-four (TOF) stimulation of the ulnar nerve. Th
e initial TOF reading was documented as was the time taken to reach a TOF r
atio of 0.9. Thirty-five assessments were carried out. The mean offset time
of atracurium was 28.7 min (SEM 1.76 min, range 8-56 min). There was no co
rrelation between the dose of atracurium. at discontinuation and the offset
time of the infusion. The duration of infusion was negatively correlated w
ith the offset time of atracurium, and this effect was most prominent in ch
ildren who had received infusions for longer than 48 h. When given by conti
nuous infusion, the offset time of atracurium is very variable between indi
vidual patients. Infusions administered for longer than 48 h are associated
with a significant reduction in the offset time as a result of increasing
tolerance.