Nitrous oxide: A difficult case for classification as an addictive substance

Evidence for a claim about nitrous oxide as an addictive substance is far from clear-cut. Back et al.’s article invites a thorough revisit of the DSM-5 criteria to disentangle neuropharmacological evidence and to look at concrete epidemiological data to base an informed opinion on this matter. Substances of dependence activate a common neurobiological route of the mesocorticolimbic dopamine system that classically translates into craving, tolerance and withdrawal symptoms upon discontinuation of the drug, as a recent excellent review has reiterated 1. Although nitrous oxide has been used and studied since the 19th century, the existence of these symptoms among nitrous oxide users have scarcely, if ever, been reported 2. Reading Back et al.’s article 3 invites an exploration of the very aspects that comprise addiction and a critical look at the DSM-5 criteria to understand whether or not nitrous oxide is an addictive substance. Unlike other addictive substances, nitrous oxide failed to induce reinforcement in animal studies 3, 4, whereas human studies yielded mixed results at best 5, 6. As a neuroscientist with more than two decades of experience in the field of addiction, the lack of solid neuropharmacological evidence leaves me sceptical. In some of my own research and based upon older studies, I found evidence of mechanisms of action through which nitrous oxide is able to increase dopaminergic firing 7, either through blockade of the N-methyl-D-aspartate (NMDA)-receptor system or through direct release of endogenous opioids 8, 9. However, given the extremely short half-life of nitrous oxide, it is doubtful whether these pathways invoke the typical long-lasting mesocorticolimbic changes, as is well known from pioneering work by Volkow and others 1. In fact, the controversy regarding some of the core addiction traits, such as craving, withdrawal or tolerance, have led to nitrous oxide and other volatiles or fumes to be placed under a separate category of ‘inhalant use disorder’ in the DSM-5 10. Nevertheless, the exercise by Back et al. into the possible addictive properties of nitrous oxide on the basis of DSM-5 criteria is not without merit. Nitrous oxide has an image of being relatively innocent and ‘fun’ and ideal for social use, especially among younger users 11. It could also function as an alternative for substances which are deemed too dangerous because of their illicit status 11, 12. However, binge use has led to decreased vitamin B12 and subsequent physical damage 13, 14. Therefore, the existence of harmful use is obvious, but this is not the same as addiction. In their exercise, Back et al. found strongest evidence for harmful use (criterion 8), negative social consequences (mainly criterion 6), spending a substantial amount of time using (criterion 3) and using more than intended (criterion 1). However, most evidence is qualitative or anecdotal, something that Back et al. also acknowledge. In light of the fact that mainly young users seem to be attracted to nitrous oxide use, one might regard it as part of risk-taking behaviour, as young people tend to take more uninformed risks anyway. For instance, binge alcohol use is typically prevalent among adolescents and young adults 15, and binge-use is definitely not addiction per se 16. Hence, the criterion of using more nitrous oxide than intended can also be seen in this light. In addition, it is not always clear from these studies whether social problems were already present or not, so disentangling cause and effect is difficult. Finally, physical harm caused by misuse could very probably be due to uninformed decisions, as many users were not aware of the potentially dangerous consequences of vitamin B12 deficiency 13. Therefore, my main argument here is that the complex neuropharmacological phenomenon of addiction is not always covered by the mere presence of separate DSM-5 criteria. Finally, my foremost concern about qualifying nitrous oxide as an addictive substance is the absence of epidemiological data. It would stand to reason that if nitrous oxide has addictive properties, following its popularity we would see a clear rise of people in addiction treatment. However, this is not the case; Fildago et al. failed to find any epidemiological data 17, aside from reported case studies, and concluded that it has merely a weak addiction potential, thereby justifying nitrous oxide’s separate status in the DSM-5. Taking all evidence together, it would be wise to treat nitrous oxide (and other volatiles) as non-typical substance(s) of misuse and to refrain from the (burdened) term addiction, at least until more convincing evidence emerges. Notwithstanding, it is a worthwhile endeavour to revisit the basics of addiction and to critically reassess substances from time to time, for which I owe gratitude to Back et al. None. Data sharing is not applicable to this article as no new data were created or analyzed in this study.