To the Editor:
—In recent years, the journal ANESTHESIOLOGY has published several reports on the analgesic effectiveness of the cholinesterase inhibitor neostigmine. Although there is good evidence for a spinal action of neostigmine, 1,2a rationale for a peripheral mechanism of action is lacking. Intrathecal injection of neostigmine produces analgesic effects in animals, 3including humans, 4accompanied by a high incidence of side effects. The inhibition of spinal cholinesterase results in an increase of endogenous acetylcholine, which is most likely released from intrinsic cholinergic neurons within the dorsal horn of the spinal cord. 5These cholinergic neurons terminate in the vicinity of primary afferents, 5which express muscarinic receptors. 6,7Consistently, analgesic effects of intrathecal neostigmine could be reversed by muscarinic receptor antagonists. 8The analgesic effect may be explained by a muscarinic presynaptic inhibition of glutamatergic afferents, similar to how it has been described in the neostriatum. 9An important prerequisite for the effectiveness of neostigmine is a tonic cholinergic activity. 10,11
Although these mechanisms are clearly described for the spinal cord, evidence is lacking for the periphery. Therefore, the negative result of adding neostigmine to a mepivacaine axillary plexus block in the study by Bouaziz et al. 12is not surprising. How would neostigmine exert an effect within the nerve sheath of the axillary plexus? An endogenous release of acetylcholine does not exist. Interestingly, the authors hypothesized a peripheral action of neostigmine based on the demonstration of peripheral muscarinic receptors only 6; however, this is not enough. In their discussion, they refer to studies that investigated the effects of muscarinic agonists in spinal cord slice preparations i n vitro 13or intrathecally in vivo . 14However, these effects are not described for peripheral nerve endings, and neostigmine should not be confused with muscarinic agonists. Therefore, a lack of neostigmine effectiveness within the nerve sheath of the axillary plexus should have been anticipated. For the same reasons, the results of experimental 15and clinical 16studies that showed analgesic effectiveness of intraarticular neostigmine are surprising. In rats, the intraarticular injection of neostigmine produces moderate analgesia to thermal stimuli, 15which is allegedly reversible by intraarticular atropine, although data were not shown in that study. In patients undergoing arthroscopic meniscus repair, intraarticular injection of 500 mg neostigmine resulted in a significant difference in pain intensity at 1 h postoperatively, in total consumption of intravenous rescue analgesics, and in time to first analgesic use. 16To support their hypothesis of a peripheral site of action, the authors refer to “preclinical data suggesting peripheral antinociceptive effects of acetylcholine.”17However, acetylcholine is an agonist at muscarinic receptors. Again, the cholinesterase inhibitor neostigmine should not be confused with muscarinic agonists. In both cases, the question arises: Which cells in the periphery are responsible for a tonic cholinergic activity? This question has not been addressed in either of these studies; therefore, the mechanism of action of peripherally applied neostigmine is still lacking, and the identification of the presumed source of peripheral acetylcholine is urgently needed.