Influence of drugs on the response characteristics of the LiDCO sensor: an in vitro study

Abstract

In a previous study, the authors found a large bias (50%) for lithium (LiDCO) compared with thermodilution cardiac output measurement methods in ponies receiving i.v. infusions of xylazine, ketamine, and midazolam. This prompted the authors to examine the effect of drugs on the LiDCO sensor. Drugs and lithium were dissolved in 0.9% saline to produce the following solutions: saline, saline-lithium, saline-drug, and saline-drug-lithium. The drug concentrations were overlapping the range of clinical interest as estimated from the published literature. These 38°C solutions were pumped through the LiDCO sensor in predetermined order. Sensor voltages were measured. Differences between lithium-induced voltage changes in the absence and presence of drugs indicated erroneous lithium detections that, if they occurred in vivo, may cause biases in LiDCO measurements. Clonidine, detomidine, dexmedetomidine, medetomidine, romifidine, xylazine, ketamine, S-ketamine, lidocaine, and rocuronium caused concentration-dependent increases in sensor voltages and negative biases in lithium detection that were mathematically equivalent to greater than +10% biases in LiDCO. The drug-induced voltage changes correlated with calculated biases in LiDCO (r(2)=0.91). Atipamezole, acepromazine, butorphanol, diazepam, midazolam, and guaifenesin caused minimal or no interaction in this study. A number of drugs influenced the accuracy of the LiDCO sensor in vitro but, based on published pharmacokinetic data, only xylazine, ketamine, lidocaine, and rocuronium may cause biases at clinically relevant concentrations. These findings need to be confirmed in vivo. Relevant (>3 mV) changes in sensor voltages due to the presence of drugs may indicate possible interactions with the LiDCO sensor.