Performance assessment of an adaptive model-based feedback controller: comparison between atracurium, mivacurium, rocuronium and vecuronium.

The performance of an adaptive model-based controller for the administration of atracurium, mivacurium, rocuronium and vecuronium was compared in 159 adult surgical patients. The degree of neuromuscular block was set to 90% for atracurium, rocuronium and vecuronium and to 95% for mivacurium. Performance was assessed by calculating the median prediction error (bias), median absolute performance error (inaccuracy), divergence, wobble, the mean offset and the mean standard deviation from the setpoint. All indices of controller performance showed minimal deviation of the actual neuromuscular block from the setpoint. Although the controller appeared to be able to control rocuronium induced block at 90% and mivacurium induced block at 95% better than atracurium and vecuronium block at 90%, the differences in the controller performance between the four studied relaxants were small and have hardly any clinical significance. We conclude that a model-based adaptive controller is useful in the administration of atracurium, mivacurium, rocuronium or vecuronium.