Background and Goal of Study
The variability of response to drugs used in anaesthesia, including neuromuscular blocking drugs (NMB) such as rocuronium(1), is often not appreciated by providers. Fixed-rate infusion regimes imply individual response is relatively constant(2). Commercial drug interaction displays such as GE-Navigator allow visualisation of the kinetics of several drugs including rocuronium in routine practice(3). When combined with use of qualitative NMB monitoring we can evaluate drug kinetics in real time. 

The aim of this study was to explore the inter- and intra-patient variability in response to rocuronium in a wide range of routine clinical cases.

Materials and Methods
New Zealand National Ethics Committee approval. 
Patients over 18yr undergoing surgery with planed use of >1 dose of rocuronium were eligible. NMB monitored using GE NMT module (mechnoSensor) with train of four (TOF) monitored every 20 sec. Rocuronium doses entered into Navigator which calculates rocuronium effect-site concentration using models of Cooper and Laud. Rocuronium concentration at return of the second twitch (T2) of the TOF was used as an endpoint . 
We calculated the average rocuronium concentration at first return of T2 and used interquartile range (IQR) and Median Absolute Deviation (MAD) as measures of dispersion. For subjects with >1 rocuronium doses we calculated the IQR & MAD of the concentrations at each return of T2. 

Results and Discussion
We have results from 42 subjects: 24F; 18M. Mean age 56 (SD17)yr; weight 83(19)kg; BMI 30.0(8.1)kg/m2. Initial rocuronium range 0.30mg/kg - 0.85 mg/kg (mean 0.53mg/kg). 36 received >1 rocuronium dose.
First recovery to return of T2 occurred at a mean rocuronium concentration of 1.08 (sd 0.54)ng/ml; range 0.21 to 2.7ng/ml. IQR 0.74ng/ml and MAD 0.90ng/ml.
Mean IQR of individual responses was 0.14(0.10)ng/ml & MAD 0.10 (0.06)ng/ml.
There was no correlation between rocuronium concentration at T2 and age or weight.

Conclusions
Response to rocuronium varied between subjects reinforcing the need for NMB monitoring. The response of an individual is relatively constant. 
Use of predictive drug models in combination with measures of effect may facilitate preemptive rather than reactive drug dosing and simplify the use of relaxant infusions.

References
1. Cooper RA, Maddineni VR, Mirakhur RK et al Br J Anaesth 1993 71:222-6
2. Harrison MJ Anaesthesia 1997 52:37-40
3. Gin T Anesth Analg 2010 111:256-8

Exploring the variability in response to rocuronium in routine clinical practice using GE-Navigator