Arterial pressure is still one of the most important measures in estim
ating the required dose of inhaled anaesthetics. It is measured easily
and reacts rapidly which makes it suitable as a variable for feedback
control of depth of anaesthesia. Fuzzy logic, a novel approach to fee
dback control, was used to control arterial pressure in 10 patients du
ring intraabdominal surgery by automatic adjustment of the concentrati
on of isoflurane in fresh gas. During anaesthesia, fuzzy control perio
ds of 45-min duration were alternated randomly with human control peri
ods of equal duration. During the skin incision period (-3 to +12 min)
48.2% of all fuzzy control pressure values were within +/-10% of the
desired mean arterial pressure compared with 40.4% of the human contro
l values (P < 0.05). The corresponding values for the remainder of the
operation were 78.3% and 83.2%, respectively. Thus fuzzy outperformed
human control at skin incision, but was slightly inferior during the
rest of the operation. We conclude that fuzzy logic is a promising new
technique for control of isoflurane delivery during routine anaesthes
ia.