Neurological disorders caused by recreational use of nitrous oxide—a retrospective study from a German metropolitan area and review of the literature

Abstract

Background

The recreational use of nitrous oxide (N₂O) has seen a worldwide rise in the recent years, resulting in an increased incidence of neurological complications due to N₂O-induced functional vitamin B₁₂ deficiency. Here, we report on a cohort of patients admitted to a tertiary care center with neurological symptoms in the context of recreational N₂O use between 2020 and 2024.

Methods

We screened the database of the University Hospital Frankfurt for patients ≥ 18 years of age who presented with neurological deficits and a history of N₂O consumption between January 2020 and December 2024. We analyzed the spectrum of neurological deficits as well as radiological and laboratory findings.

Results

We identified a total of 20 patients, 16 males and 4 females, with a median age of 21 years. We found a steady increase in the number of cases, with no cases in 2020 and 2021 and a definite peak in 2024. The mean daily N₂O consumption was 2500 g. All patients reported sensory deficits; 85% had gait disturbances and 70% had motor deficits. Less frequent symptoms included pain, bladder or bowel dysfunction, fatigue and spasticity. The median score on the modified Rankin scale (mRS) was 2, with some patients being wheelchair-bound. The most frequently observed lesion pattern was combined myelo-polyneuropathy. T2-hyperintense myelon lesions were observed in 11 of 15 patients (73.3%). Surprisingly, laboratory work-up revealed normal vitamin B₁₂ levels in nearly all patients (95%), whereas homocysteine and methylmalonic acid levels were prominently elevated in all patients (100%). In addition, 13 patients (65%) presented with hematological abnormalities. All of the patients who presented for follow-up (20%) reported continued use of N₂O. There was no neurological improvement in any of these cases.

Conclusions

Our study confirms that the increasing incidence of N₂O-induced neurotoxicity reported in other countries can also be observed in Germany. Therefore, it underlines the relevance of the current debate on health policies. In addition, our study highlights the pitfalls of vitamin B12 laboratory testing and emphasizes the need to address substance addiction in treatment.

Background

Nitrous oxide (N₂O) has been used as an anesthetic with strong analgesic and anxiolytic effects for over 150 years [12]. This substance was already used for recreational purposes in the eighteenth century, e.g. at laughing gas parties. Although the use of N₂O has generally been considered safe, it has long been known to damage neurological function through inactivation of vitamin B_12, which, in turn, causes neurotoxicity by impairing myelin synthesis [15, 17, 25]. This leads to neurological symptoms, most frequently presenting as paresthesia, peripheral polyneuropathy or subacute combined degeneration of the spinal cord [10, 17]. Myelopathy primarily affects the dorsal columns and pyramidal tracts in the spinal cord. In advanced stages of myelopathy, destruction of the myelin sheaths leads to demyelination with subsequent axonal degeneration [29].

In recent years, there has been a sharp increase in the recreational use of N₂O in the EU as well as in the USA and Canada, particularly among people aged 18–24 years [6, 8, 16, 33]. This development is accompanied by a rising incidence of neurological complications in the context of N₂O consumption, as shown by recent studies from France and the Netherlands [4, 33].

For Germany, larger case studies are scarce. However, in the past two years, an unprecedented increase in relatively young patients with neurological complications due to vitamin B₁₂ deficiency has been anecdotally reported from several large German cities [19]. Here, we performed a systematic analysis of all patients who were admitted to Frankfurt University Hospital due to N₂O-associated neurotoxicity.

Methods

Patient selection

We screened the database of the University Hospital Frankfurt for patients ≥ 18 years who presented with neurological deficits and a history of recreational N₂O consumption between January 2020 and December 2024. More specifically, we conducted a keyword search for the terms “nitrous oxide”, “laughing gas”, “vitamin B12”, “hypovitaminosis” or “subacute degeneration of the spinal cord” in both the electronic medical records as well as the radiological database of our department. Basic epidemiological data, neurological symptoms and laboratory results on vitamin B₁₂ metabolism were extracted from clinical records and deidentified. MR imaging of the central nervous system was viewed whenever available.

Data visualization and statistical analysis

For data visualization, we used GraphPad Prism 10.1.2. (Boston, MA, USA) and CorelDRAW Graphics Suite 2019 (Alludo, Ontario, Canada). Statistical analysis was performed with GraphPad Prism 10.1.2 and Microsoft Excel (Microsoft, WA, USA). Unless otherwise indicated, values are given as the means.

Review of the literature

We searched the MEDLINE®-indexed literature using the PubMed search engine from the National Centre for Biotechnology Information (www.pubmed.gov) from January 2010 until February 2025. We used a combined search approach including the following keywords: “nitrous oxide” [MeSH terms]” OR “laughing gas” (MeSH terms) AND “myelopathy” [all fields] OR “neuropathy” [all fields]. We excluded studies with ≤ 5 patients, studies not focusing on neurological disorders and articles that were not publicly available. In addition, we examined the reference list of all identified studies and review articles for studies that might be of relevance.

Results

Patient cohort and nitrous oxide consumption

We identified a total of 20 patients with a history of recreational N₂O consumption and neurological symptoms who were treated at the Department of Neurology/University Frankfurt (Table 1). The majority of patients (80%) were male. The median age of the patient cohort was 21 years (range 19–30). Patient interviews did not indicate consumption of other drugs or alcohol abuse. Most of the patients were in an apprenticeship or in a low to middle income job, none had a university degree. The reported amount of consumption of N₂O ranged from 40 g to 8000 g per day. The mean daily consumption was 2500 g. Assuming an average quantity of 8 g per container, this corresponds to 312 balloons per day. Notably, two patients reported frequent exposure to N₂O but consistently denied active consumption. The median duration of consumption before the first evaluation at our hospital was 11 months, with a range of 1 month to two years.

Table 1 Basic demographic data and consumption patterns of nitrous oxide

Increasing incidence of nitrous oxide-associated neurotoxicity between 2020 and 2024

During the investigated period from 2020–2024, we observed a significant increase in the number of patients treated with N₂O-associated neurotoxicity at our department (Fig. 1). No cases occurred before 2022 according to the database of the hospital. The number of cases then rose noticeably in 2023, and reached a maximum thus far in 2024 with 13 of the 20 cases (65%).

Fig. 1

Development of the number of cases of nitrous oxide-related neurological symptoms between 2020 and 2024

Clinical findings

All patients complained of sensory deficits (Fig. 2a). Most had additional neurological symptoms, presenting primarily as gait disturbance (85%, n = 17) and/or motor deficits (70%, n = 14). Pain (n = 5), bladder/bowel dysfunction (n = 2), fatigue (n = 2) and spasticity (n = 1) were reported less often. The most frequently observed clinical pattern of N₂O-associated neurotoxicity was combined myeloneuropathy (45%), while 30% of the patients had neuropathy, and 25% had myelopathy exclusively (Fig. 2b). We observed no correlation between the quantity or duration of N₂O consumption and the damage pattern (Table 1). The clinical severity of the symptoms at the time of the first clinical examination was 2 on the modified Rankin scale (mRS) (Fig. 2c).

Fig. 2

Clinical findings. a Frequency of neurological symptoms given as a percentage of the entire cohort. The bars in light blue are depicted as percentages of all patients reporting sensory or motor deficits. b Distribution of the damage pattern given as a percentage of the cohort. c Clinical severity quantified by the modified Rankin scale

Imaging and laboratory testing

MR imaging was performed in 15 of the 20 patients (75%). Among these patients, 11 (73.3%) had pathological hyperintensity in the T2-weighted sequences in the posterior funiculi of the spinal cord (Fig. 3a), indicating subacute combined degeneration of the spinal cord (formerly referred to as funicular myelosis). The level of vitamin B₁₂ was within the normal or lower-normal range in all but one patient (Fig. 3b). Three patients even had significantly elevated vitamin B₁₂ levels, which was due to self-initiated substitution prior to admission to our clinic. These patients had no particular dietary habits or other substance dependencies. All patients had normal values for transcobalamine I. In contrast, homocysteine and methylmalonic acid were well above the upper limit of normal in all 20 patients. Thirteen out of 20 patients (65%) had at least one hematological abnormality. These included alterations in the mean corpuscular hemoglobin (MCH), volume (MCV) or red cell distribution width (RCDW) (Fig. 3c). No patient had low hemoglobin levels.

Fig. 3

Diagnostic findings. a Representative T2-weighted MR image of the upper spine showing hyperintense lesions of the posterior funiculi (inverted “v” sign). b Results of the laboratory work-up of vitamin B₁₂ metabolism. The reference ranges are marked in blue. c Results of the laboratory hemoglobin, mean corpuscular volume and hemoglobin and red cell distribution width. The reference ranges are marked in blue

Treatment and follow-up

All patients were treated with 1000 µg of vitamin B₁₂ per day i.m. for 10 days followed by 1000 μg of vitamin B₁₂ weekly or monthly. We recommend determining the dosage and duration of treatment on the basis of the level of methylmalonic acid at follow-up. Of note, only 4 of the 20 patients (20%) presented for a neurological follow-up examination in our department within three months after discharge. None of these four patients experienced any improvement in their clinical condition. Notably, all four patients continued to use N₂O.

Discussion

The rising popularity of recreational nitrous oxide use

Between 2020 and 2024, we observed an unprecedented increase in relatively young adults who presented with N₂O-induced neurological symptoms. This coincides with an increased consumption of N₂O among adolescents in the catchment area of our hospital as evidenced by a recent urban drug report [21, 22]. In this report, the lifetime prevalence of N₂O increased from 7% in 2020 to 17% in 2022 [22] and 14% in 2023 [23]. More than 60% of the survey participants reported high popularity of this substance among their peers. The increase in the recreational use of N₂O has been reported by other European countries and worldwide [5, 6].

In our cohort, 80% were male and had a median age of 21 years. This finding is in line with the majority of studies published on recreational N₂O use between 2010 and 2025 (Table 2). Specifically, 17 out of 24 studies had a majority of male patients and the median age across all studies was 22 years. Studies from France and the Netherlands have identified a low level of education, employment in the low-wage sector or unemployment as factors that predict a high prevalence of laughing gas consumption [4, 32]. In addition, surveys from the Netherlands have found that consumers often had a non-Dutch background. It must be emphasized that socioeconomic data was rarely collected in the studies we identified. Hence, representative surveys at national level are needed to better understand the socioeconomic and cultural background of the consumers.

Table 2 Summary of studies on N₂O-induced neurotoxicity between 2010 and 2025

The rising popularity of N₂O worldwide has been attributed in part to its perception as a harmless substance [21, 23] as well as to its easy-access supply, e.g., through online shops and kiosks [6, 21]. A recent development is the sale of large cylinders of up to 2 kg, allowing for heavy use at reduced costs. In a recent study from France involving 181 patients between 2019 and 2021, the average daily consumption was 1200 g [4]. Notably, the authors reported that this number was 27,000 times greater than that reported in 2017 for the same urban area. In our cohort, 85% of the patients with active use consumed ≥ 400 g/day, and the mean daily consumption was 2500 g. Hence, the growing number of patients presenting with neurological sequelae from N₂O use could also result from a subgroup of users with excessive daily intake. Nevertheless, cases of neurological damage resulting from infrequent or passive consumption have also been reported [28, 30], which was also observed in our study. Consequently, no safe level of use can be defined at this point.

In our literature search, it was striking that nitrous oxide consumption was either not reported at all or that consumption was quantified in very heterogeneous units, which limited comparability. Some of the authors reported that patients did not want to provide any information on their consumption or that the information given was very vague. Clinicians should take this into account when taking the patient’s medical history.

Diagnostic findings

A crucial aspect for clinical management is the diagnosis at the earliest timepoint. From a clinical point of view, sensory deficits (hyp-/paresthesia) are the most common, often accompanied by gait disturbance, muscle weakness and reduced muscle reflexes (Table 2, Fig. 2). It is therefore not surprising that subacute polyneuropathy is often assumed as the initial differential diagnosis [9]. Myelopathy was present in about every second patient in our literature search. However, symptoms that suggest damage of the CNS (e.g. spasticity, increased tendon reflexes or pathologic reflexes) were often not detectable.

It has been reported that the neurotoxicity of N₂O is dose dependent [34]. Furthermore, the patterns of use could impact the clinical symptoms of patients in that the effect on the spinal cord is dependent on the quantity of N₂O, whereas the presence of polyneuropathy may depend on the duration of exposure to N₂O [3]. We did not observe such a correlation in our cohort, although this could be due to the limited number of cases.

In the case of suspected or confirmed nitrous oxide consumption, the diagnostic work-up primarily includes laboratory evaluation for vitamin B₁₂ metabolism. However, the use of vitamin B₁₂ levels for diagnosis remains problematic because of its low specificity. Although studies have shown reduced vitamin B₁₂ levels in up to 70% of consumers [26, 35], our literature search underlines that total circulating vitamin B₁₂ is not a reliable indicator of nitrous oxide-induced neurological disorders. Low vitamin B₁₂ levels varied between 17% and 75%. In our cohort, only one patient (5%) had low vitamin B₁₂ levels. In addition, it must be emphasized that in some studies, including ours, patients had already started self-treatment with vitamin B₁₂ before seeking medical help. Thus, relying on vitamin B₁₂ levels can mask the functional impairment of vitamin B₁₂ metabolism.

The inactivation of vitamin B₁₂ by N₂O incapacitates the enzymatic activity of methylmalonyl-CoA-mutase and methionine synthase, resulting in the accumulation of both methylmalonic acid and homocysteine [2]. In all studies in which both vitamin B₁₂ and homocysteine were examined, homocysteine was significantly more frequently elevated. This is in line with our own findings, as well as a recent review [20]. Hence, homocysteine and methylmalonic acid are much more reliable markers for N₂O-induced disruption of vitamin B₁₂ metabolism and for monitoring treatment success [11, 20]. Since the analysis of methylmalonic acid is costly, homocysteine is usually preferred. This is reflected in the results of our literature search: methylmalonic acid was only examined in 5 of 24 studies. Whether homocysteine or methylmalonic acid are indicative of the mechanisms of N₂O-induced neurotoxicity is still under investigation [13].

Treatment and outcome

The cornerstone of treatment is vitamin B₁₂ supplementation. There is a consensus that supplementation should initially be carried out at high intensity, although there are different schemes for implementation. It is recommended that the substitution is initially carried out daily [14] or every other day [27] via i.m. injections of 1000 µg of hydroxycobalamin over the course of one to two weeks or until there is no further improvement of symptoms. Supplementation can then be reduced to a weekly or 3-monthly dose, depending on whether the patient had a pre-existing vitamin B₁₂ deficiency.

There are little data on long-term follow-up, which is reflected in our literature search. In almost all studies that reported on the clinical outcome, including ours, a high rate of loss to follow-up was observed. Furthermore, information on N₂O abstinence was frequently missing.

Overall, a high percentage of patients show some neurological improvement (Table 2), either at discharge or at follow-up. In the 13 studies with follow-up, the percentage of complete recovery ranged from 0 to 95%. In a Chinese review with 51 patients at follow-up, 67% recovered completely whereas 33% had residual symptoms [35]. Notably, this study recorded only one relapse during the observation period. In two systematic case reviews, partial recovery was most frequently observed [10, 18], and progression of the disease was only seen if patients continued consumption. This could explain the poor outcome of our follow-up patients and emphasizes that cessation of N₂O consumption is crucial for recovery [18].

Reports of excessive recreational use have foregrounded the debate of whether N₂O has addictive potential. Previous studies have shown that the majority of heavy users show symptoms of substance use disorders, with up to 90% reporting the use of N₂O in larger quantities and for longer periods than intended [1, 7, 24]. In addition to psychological addiction, reports have shown that frequent users tend to inhale larger quantities over time to experience similar physical effects [31]. In our cohort, all patients who presented for follow-up continued to use N₂O despite their considerable neurological deficits. In a Dutch study, 80% of the participants unsuccessfully tried to cease N₂O use [24]. This highlights the need to address substance addiction in treatment.

Conclusions

In summary, the increasing number of patients presenting with neurological sequelae from the use of N₂O could be a result of several factors, including its rising popularity, low costs, excessive intake fueled by widespread availability, and the addictive potential of the gas. These aspects explain why some countries, e.g. France and the Netherlands, have introduced health policy changes that strongly restrict open access to N₂O [5]. In Germany, changes in legislation have not yet been implemented. Our study from a tertiary care center in Germany confirms the increasing incidence of N₂O-induced neurotoxicity especially in young adults that has been observed in other countries. It therefore underlines the relevance of the current health policy debate. Target group-specific measures to raise awareness of the health risks appear necessary, for example via social media or schools. The high median daily consumption in our cohort and the continued use observed at follow-up stress the addictive potential of N₂O. Addiction counseling should therefore be an integral part of treatment.