Epidural ketamine in healthy children: What's the point? Authors' reply

To the Editor: We note with concern, as we have in the past with regard to other agents (1,2), the continuing policy of this journal to publish clinical studies of epidural or intrathecal injection of drugs for which preclinical toxicity studies are lacking or have demonstrated toxicity. The current study (3) involves the epidural injection of S(+)-ketamine in healthy children undergoing inguinal hernia repair. This study was conducted with a double-blind methodology, and the concept of equipoise states that for a double-blind trial to be performed, the investigators must genuinely be uncertain as to which treatment is better. Undertaking such a study presumes a favorable risk-benefit ratio. In the case of healthy children, we must assume a particularly favorable ratio. What is the risk of epidural injection of S(+)-ketamine? The article provides no citations or discussion regarding preclinical toxicity assessment or regulatory approval of the epidural administration of S(+)-ketamine. It should be noted that some but not all preclinical studies have indicated neurotoxicity as a result of neuraxial ly administered ketamine and other NMDA antagonists (4–9), and one case report (10) demonstrated neurotoxicity in postmortem spinal cord tissue, which is consistent with the neuropathology observed in animals studies. Although not all studies report neurotoxicity as a result of ketamine and some have speculated that the preservative present in some ketamine formulations may be responsible, there clearly is controversy in the preclinical literature as to the safety of this agent for either epidural or intrathecal injection. In general, repeated exposure for 28 days in two species of drug in concentrations or doses far exceeding the clinically anticipated dose is required for regulatory approval in the United States to proceed with clinical trials, and there is no such support in the published literature with regard to ketamine. To move one step further, what is the risk of epidural injection of S(+)-ketamine combined with clonidine? Again, no citations or discussion is provided in the article. Have the compatibility and stability of such mixtures, even if acutely prepared, been established? Although clonidine lacks neurotoxicity when administered intrathecally alone (11), it does reduce spinal cord blood flow (12), which could potentiate toxicity from other agents. Given the conflicting nature of the neurotoxicity literature with ketamine alone, it seems only prudent to examine the toxicity of this combination before clinical trials. Risk aside, what is the benefit of epidural injection of S(+)-ketamine, either alone or with clonidine, in these healthy children with minor surgery? Certainly, improved pain relief from lingering caudal analgesia compared with systemic therapy would not be expected to alter mortality in this healthy population, although it might reduce morbidity, such as sedation or nausea from systemic analgesics. The stated advantage is that accidental intravascular injection would not cause cardiovascular or central nervous system complications, as can occur with local anesthetics. Such events are a concern, particularly in anesthetized children. Yet, the incidence of this catastrophe is very small and known, compared with the unknown incidence (if any) of neurotoxicity from these drugs or of nightmares from central nervous system administration of ketamine. We do not imply that the authors performed an improper study. Their institutional ethics committee approved the study, and informed consent was doubtless obtained. Nevertheless, we find it difficult to imagine the language of such a consent form that would not set off alarm bells, particularly if it included clear language that, in contrast to other therapies (e.g., local anesthetics), the safety of these drugs is completely unknown and that the risks include unanticipated side effects, including permanent paralysis. Regardless of the nature of these issues, there is another party to this scenario, and that party is this journal. We think—and have previously stated so in letters to the editor (1,2) —that it is the ethical responsibility of the journal (i.e., its editors and referees) to insist on discussion of preclinical toxicity issues for epidural or intrathecal injection of novel drugs (even those with previous exposure in small numbers of humans) before acceptance and publication. This responsibility is not trivial, and its importance arises from the position of authority ascribed to a peer-reviewed paper in an authoritative journal. Reading the bottom line of this paper, one might conclude, as these authors stated, that “…(the combination of S(+)-ketamine with clonidine) is a safe and effective alternative to common local anesthetics for caudal blocks in pediatric anesthesia.” Yet, in fact, this report establishes nothing of the kind, and such a statement is not supported by this report or any published report of which we are aware. The presence of such an assertion in the present case should be a source of concern to the journal and its editors. James C. Eisenach, MD Tony L. Yaksh, PhD