To the Editor:—
A recent paper1reported experiments, using laboratory rats, on the effect of intraamygdala infusion of a γ-aminobutyric acid type A antagonist on propofol-induced amnesia for inhibitory avoidance training, as well as on expression of activity-regulated cytoskeleton-associated protein in the hippocampus. This work may elucidate the neural mechanisms of the amnestic effects of propofol, as well as the neurobiological mechanisms of general anesthesia and memory more generally.
The experiments reported in this paper required stereotaxic neurosurgery to implant cannulae aimed at the basolateral amygdala, and rats were subsequently euthanized for determination of activity-related cytoskeleton-associated protein levels or histologic verification of cannula placement. The authors used chloral hydrate in both procedures; for surgical anesthesia in the first and euthanasia in the second. Chloral hydrate is not a suitable drug in either case. Chloral hydrate is regarded by many to produce hypnosis and not anesthesia.2It does not provide analgesia and causes marked respiratory depression at doses required for surgical anesthesia.3Apart from its inadequate anesthetic properties, 20% chloral hydrate is extremely irritating and therefore unsuitable for intraperitoneal use. It is associated with ileus in rats,4as well as peritonitis and gastric ulcers.5Its use by intraperitoneal injection for survival surgery is not recommended.5Thus, it is not the most refined choice of agent for the surgical procedure in which cannulae are chronically implanted to make drug infusions into the amygdala. The authors also used a higher dose of chloral hydrate for euthanasia. However, chloral hydrate is not an acceptable agent for euthanasia according to the guidelines of the American Veterinary Medical Association†; its use for this purpose has been proscribed for some time.6
There are no scientific justifications for using chloral hydrate for these experiments, as many other agents would be more suitable for both surgical anesthesia and euthanasia without interfering with the experimental endpoints. Indeed the chloral hydrate-induced hypoxemia which must occur during euthanasia as respiration becomes depressed†, may compromise the experimental aims in terms of measuring protein and messenger ribonucleic acid levels of an activity-related protein. The noxious stimulus of an intraperitoneal irritant is not only inhumane, but if it leads to peritonitis the rats will be abnormal at the time of testing.
It seems that chloral hydrate has traditionally been used to provide anesthesia where the avoidance of agents with known receptor interactions is desirable. But it is likely that chloral hydrate has unknown receptor interactions. Therefore choosing a different agent whose receptor interactions are better characterized could be beneficial, not only in terms of animal welfare but also in terms of data interpretation.
The publication of this paper in Anesthesiology concerns us, because the standard of laboratory animal anesthesia used in this research is not acceptable.
*Oxford University, Oxford, United Kingdom. general@psy.ox.ac.uk
