Traditionally, nitrous oxide (N₂O) was widely used in anesthetic practices (1). With the propensity to induce euphoria and simple accessibility, the recreational use of N₂O has expanded and become popular (2).

However, prolonged exposure to N₂O could have consequences. In 1978, neuropathy caused by N₂O abuse was first reported in three cases, followed by descriptions of 12 additional cases of abuse or occupational exposure (3, 4). It is widely believed that N₂O could oxidize the cobalt ion of Vitamin B12 (VitB12), disrupt the function of cobalamin, and cause a disturbance in myelin sheath synthesis and maintenance (5). Demyelination of white matter in the brain, the dorsal and lateral spinal cord, and the peripheral nerves can later lead to gait disturbances, paresthesia, autonomic dysfunction, and cognitive impairment.

Due to a lack of supervision measures and a large population, the number of cases of N₂O-related neurological disorders in China has increased rapidly since 2016 (6–12). Published studies have typically focused on symptom presentations, imaging characteristics, laboratory findings, and distinctions from other subacute combined degenerations (SCDs). The relationship between symptoms and the presentation of clinical severity has been understudied and N₂O abuse is still an overlooked etiology for SCD or polyneuropathy among young patients.

Thus, we conducted this retrospective observational study with detailed data collected from multiple cases in a single center to investigate the characteristics of N₂O consumption history, clinical features, as well as medical examination results, and their effects on the overall neurological impairments of these individuals.

We conducted this retrospective observational study on 20 individuals with N₂O-induced neuropathy from a tertiary medical center. We found that clinical characteristics, imaging features, and serum VitB12 levels were not related to the overall clinical severity; however, a decrease in Hb and RBC counts and an increase in MCV were associated with a worse clinical presentation upon admission.

The FDRS score was to quantify the severity of clinical manifestations and the effect on physical activity in SCDs, including N₂O-related neuropathies (11, 13, 14). There may be objective factors related to the FDRS sum score, such as demographic characteristics, N₂O consumption, hospitalization characteristics, imaging, and laboratory results. Therefore, these may indicate a worse clinical picture or outcome and provide insight into the mechanism of N₂O consumption-related neurological damage. We found that decreased Hb and RBC counts were associated with a lower FDRS score on admission in individuals who had or had not previously taken VitB12 supplements. Similarly, a higher MCV was associated with a lower FDRS score on admission in individuals who had previously taken VitB12 supplements. Although these relationships have not been reported in other recreational N₂O-induced neuropathies, postsurgical high MCV has been shown to be associated with a more severe clinical picture and poorer recovery (tendency) in anesthetic N₂O-induced neuropathies (16, 17). In other SCDs unrelated to N₂O, a correlation between a lower Hb level and a more severe clinical presentation has also been identified (18). Thus, changes in the RBC index on admission may be a sensitive indicator of more severe clinical presentation.

The majority of studies, including our own, failed to find a correlation between N₂O cumulative consumption and the severity of initial clinical presentation or outcome, regardless of different evaluation ways (17, 19–21). There were contradictory findings regarding VitB12 and clinical severity, possibly because different studies included varying proportions of VitB12-supplemented patients before admission (11, 17, 19, 20). Thus, we categorized our patients according to whether they had taken VitB12 supplements before admission. Even after categorizing the data, we were unable to identify any correlations between serum VitB12 level and the FDRS sum score. Meanwhile, studies have found total VitB12 (19, 21) and active VitB12 levels (19) did not correlate with cumulative nitrous oxide use (19). These data suggested that N₂O-induced neurotoxicity might not be dose-dependent. Among our participants, the mean N₂O exposure time was 7 months, while the mean symptom onset-to-admission days was only 8.5 days. In addition to this finding, 75% of the patients reported increases in the use of N₂O prior to admission, suggesting that symptoms could be caused by predisposing subclinical functional declines of VitB12 and a rapid process of cobalamin inactivation. Similar conditions were observed in N₂O-anesthetized patients with postoperative neuropathy, and their N₂O exposure was relatively rapid. Studies have found that some of them had presurgical recreational N₂O use caused MCV increase but were asymptomatic before surgery (16, 22). It also provides evidence that a subclinical abnormality associated with N₂O, combined with rapid consumption of N₂O, may have contributed to the onset of symptoms.

In addition to the factors associated with a worse clinical presentation, the demographics of these patients also warranted consideration. Although these characteristics were not related to clinical severity in our study, they may indicate which group of people is more likely to consume N₂O. Therefore, clinicians would be alerted to inquire about N₂O consumption if they encounter neuropathy symptoms in a patient. According to our patients, most were in their early twenties and unemployed. Past neuropsychological history was common, for example, depression, bipolar disorder, and sleep disorders. More than half of them regularly and heavily consumed cigarettes, alcohol, or narcotics, almost doubling or tripling the rates in the general same-aged population in China (23, 24). Similar to our cohort, high psychiatric comorbidity, substance use, and unemployment rates have been reported in other N₂O cohorts (21, 25–27). Social factors could also have an impact on mental health. For example, the recent COVID-19 pandemic has also been shown to lead to an increase in drug and alcohol abuse, including the consumption of N₂O (28). In addition, 30% of our patients initially concealed their use of N₂O due to the aforementioned psychosocial factors and the cultural perception in China that the consumption of psychoactive substances is unacceptable. The concealment of substance abuse could be complicated if patients present with atypical symptoms upon admission, such as neuropsychiatric symptoms. There was an unemployed 19-year-old patient who initially presented with delusions, hallucinations, and depression in our cohort. During hospitalization in the psychiatric ward, he developed an unsteady gait and frequent falling. The consultant neurologist diagnosed him with Guillain–Barré syndrome because he was unable to obtain a positive N₂O consumption history. After discharge, the patient re-used N₂O and was readmitted to the hospital; it was only then that a positive history of consumption was obtained, and the cause of the initial psychiatric symptoms was identified. Therefore, when encountering a patient with neuropathy with the aforementioned demographic traits, physicians should also focus on collecting a detailed substance use history, including consumption of N₂O.

Neuroimaging and other medical examinations helped make the diagnosis, but they lacked sensitivity and specificity. Abnormal spinal MRI was found in 37.5% of our patients, lower than other N₂O cohorts (47–100%), but the cervical spinal cord was consistent to be found as the most frequently involved segment among published literature (6, 10–12, 19, 21, 29). Results from other large cohorts have shown the lesions on MRI often extend 4–6 spinal segments and are presented as inverted “V,” “Triangle,” or “oval” shapes in the posterior column of the spinal cord (6, 10). Compared with other SCDs, Gao et al. have found that N₂O-related SCDs had higher rates of spinal MRI abnormality, wider spinal lesions on sagittal MRI, fewer involved spinal segments, and higher incidence of inverted “V” sign (10). Since there are substantial differences in percentages of abnormalities between cohorts and there are no specific MRI findings in N₂O cases, the findings have limited significance. On the EMGs, lower limb motor axonal dysfunction was predominant among our cases, similar to findings from the initial report in 1978 (3, 4) and a detailed EMG study of two N₂O cases (30). However, the EMG findings reported from published studies were heterogeneous, with most patients reported to have both axonal and demyelinating sensorimotor neuropathies (11, 19, 31). N₂O may interfere with the VitB12-dependent enzymes (methionine synthase and MMCoA mutase) and cause a lack of myeline methylation, axonal regeneration disruption, and myelin integrity destruction, according to published studies on the mechanisms of N₂O neural injury (21). Thus, both axonal and demyelinating neuropathies were detectable on EMG. The inconsistent clinical findings among studies might be due to the size of included patients and the different ways of EMG analysis. The motor neuropathy observed in N₂O cases may be attributed to factors other than VitB12 deficiency, as VitB12 is known to affect the sensory nerves predominantly (32, 33). For example, N₂O-induced repeated hypoxia and destruction of cytokine balance could also participate in the pathogenesis processes (6, 34).

First, because the participants were hospitalized cases from a tertiary medical center, the clinical presentation was likely to be more severe than in non-hospitalized cases. Second, the cohort had a low natural follow-up rate, multiple psychosocial factors, and a tendency to conceal N₂O use. These factors made it challenging to conduct thorough follow-ups. As a result, we were unable to provide detailed recovery status information and analyze follow-up data for these individuals.

Recreational N₂O use could cause severe neuropathies. Changes in the RBC index could be an early indicator of a more severe clinical presentation. High rates of psychiatric comorbidity and complex psychosocial factors make it more difficult to obtain a patient's medical history and to manage the patient. Clinicians should collect N₂O consumption histories from patients with neuropathy with specific psychosocial characteristics. An emphasis should also be placed on population education and control measures regarding access to N₂O.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Ethics Committee of West China Hospital of Sichuan University. The patients/participants or their legal guardians provided their written informed consent to participate in this study.

HG, QC, and DZ were involved in conceptualization. HG, RW, YZ, LQ, and HC helped in data collection and analysis. HG and RW wrote the original draft of the manuscript. QC reviewed and edited the draft. All authors contributed to the article and approved the submitted version.