Randomized comparison of closed-loop feedback computer-controlled with manual-controlled infusion of phenylephrine for maintaining arterial pressure during spinal anaesthesia for Caesarean delivery

Abstract

BackgroundClosed-loop feedback computer-controlled infusion has not been described for administering phenylephrine to maintain arterial pressure (AP) during spinal anaesthesia for Caesarean delivery. We aimed to compare AP control using this automated system with a previously described manual infusion system.MethodsWe randomly allocated 222 healthy subjects having spinal anaesthesia for scheduled Caesarean delivery to have systolic AP maintained near baseline with phenylephrine (100 μg ml-1) by computer-controlled infusion utilizing a proportional algorithm or manual-controlled infusion utilizing an on-off algorithm. AP control was assessed by comparing the proportion of systolic AP measurements within ±20% of baseline and by performance error (PE) calculations.ResultsA total of 212 subjects finished the study. In the computer-control group, 97% of systolic AP recordings fell within ±20% of baseline compared with 95% in the manual-control group (P=0.0004). For computer-control compared with manual-control, wobble was smaller [median 3.5 (inter-quartile range 2.5-4.8)% vs 4.2 (3.3-5.9)%, P=0.003], but there was no difference in the median PE [2.9 (0.3-4.7)% vs 1.9 (0-4.2)%], median absolute PE [4.7 (3.5-5.6)% vs 4.7 (3.8-6.7)%], or divergence [-0.01 (-0.03-0)% vs -0.06 (-0.26-0.08)%]. Fewer interventions per subject for controlling AP were required in the computer-control group [2 (2-2) vs 10 (8-13), P<0.001]. There were no differences in measured clinical outcomes.ConclusionsWithin the constraints of the studied algorithms, closed-loop feedback computer-controlled phenylephrine infusion provided better AP control with fewer interventions required compared with manual-controlled infusion. © The Author [2012]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.