Abstract
Background
Nitrous oxide (N₂O)-related neurotoxicity is a significant public health concern among young people in the UK. Recognition necessitates timely diagnosis, abstinence from N₂O consumption and replacement of vitamin B12, usually via intramuscular (IM) hydroxocobalamin. This service development project evaluated a self-injection programme (SIP) compared with a nurse-led approach, within an established ambulatory care pathway, with the aim of improving treatment adherence and completion.
Methods
Between June and December 2024, a total of 46 patients presenting with N₂O-induced neurological symptoms were included in the evaluation. Patients were given the choice of self-injecting (SIP, n=25) or attending the hospital for nurse-led administration (non-SIP, n=21). Clinical outcomes, adherence and functional improvement (including the 10 metre walk test (10MWT)) were assessed.
Results
Most patients were young (median age of 23), male (n=29, 63%), of Asian or Asian British ethnicity (n=29, 63%), and regularly using N2O (n=32, 70%). SIP patients had higher adherence, missing fewer IM B12 doses compared with non-SIP (79.7% vs 20.3%, p<0.001). Clinical recovery rates were comparable, with 74% achieving full or substantial improvement across both groups. Walking speed (10MWT) improved across both groups following treatment and did not differ between groups.
Conclusion
Self-injection of IM hydroxocobalamin is a feasible and likely cost-effective alternative to nurse-led administration while maintaining clinical efficacy.
Keywords: NEUROTOXICOLOGY, NEUROPATHY, B12 DEFICIENCY
WHAT IS ALREADY KNOWN ON THIS TOPIC
Recreational N₂ON misuse is an increasingly recognised cause of functional vitamin B₁₂ deficiency and neurological complications among young people in the UK. Standard treatment involves intramuscular hydroxocobalamin injections administered by healthcare professionals; however, adherence is often poor in ambulatory care settings, resulting in missed doses and incomplete recovery.
WHAT THIS STUDY ADDS
This service evaluation demonstrates that a structured self-injection programme for hydroxocobalamin is a feasible and effective alternative to nurse-led administration. Patients who self-injected showed significantly better adherence, with comparable neurological recovery and functional improvement.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Self-injection pathways may enhance accessibility, reduce the burden on urgent care services and support wider implementation of patient-administered therapies for conditions requiring frequent parenteral treatment.
Introduction
Recreational nitrous oxide (N₂O) is a significant public health concern for young people in the UK.1 Although use may have decreased in recent years as a result of increased awareness and legislative changes, still 4.2% of 16–24-year olds reported using N₂O in 2022–2023, making it the third most commonly used drug in this age group in the UK.2
N₂O misuse has been associated with neurological complications resulting from vitamin B12 inactivation, leading to demyelination of the spinal cord and peripheral nerves.3,6 The characteristic clinical presentation of N₂O-induced myeloneuropathy includes paraesthesia, muscle weakness and balance problems.7 8 N₂O misuse has also been linked to thrombotic complications, including unprovoked arterial and venous thrombosis.9 10
In response to a rise in N₂O-related harms, the Royal London Hospital (RLH) created a dedicated N₂O Same Day Emergency Care (SDEC) pathway in July 2021 to streamline the assessment, treatment and follow-up of patients with N₂O-induced neurotoxicity.3 Treatment of N₂O-induced neurological symptoms involves administering intramuscular (IM) hydroxocobalamin (vitamin B12) injections on alternate days until no further neurological improvement is observed, along with discontinuation of N₂O use.3 11 Through the pathway, patients are required to attend the SDEC three times a week for IM injections. A high number of did not attend (DNA) rates were observed in a previous evaluation of the pathway obscuring the clinical outcomes in a significant proportion of patients.12
We initiated a service development project to provide patients with the option to self-administer IM hydroxocobalamin while on the pathway. In this model, nurses trained patients or their caregivers to perform the injections at home. The objective was to evaluate whether this approach had comparable clinical outcomes, improved treatment adherence, reduced DNA rates and increased capacity in the SDEC unit to accommodate other urgent cases, while still avoiding unnecessary hospital admissions.
Methods
Context
This work was conducted within the context of clinical service development where a new self-injection protocol (SIP) was introduced and compared with in-person, nurse-administered hydroxocobalamin injections.
Participants
Patients in East London presenting with neurological symptoms due to N2O were referred to the RLHNitrous oxide Same Day Emergency Care (N2O-SDEC) pathway for evaluation, treatment and follow-up care. Referrals were received from accident and emergency (A&E), the urgent treatment centre or local general practitioners, with the majority presenting through A&E. The evaluation period was between 3 June 2024 and 17 December 2024.
During this period, patients were given the choice between self-administration of hydroxocobalamin following appropriate training or receiving nurse-led injections at the hospital. To evaluate the new service, patients were categorised into two groups: SIP and non-SIP.
Data collection
Care Records Service (CRS) Millennium, an electronic patient record system at RLH, was used to collect clinical and demographic information. Vitamin B12 and methylmalonic acid and/or homocysteine levels were collected for all patients as part of routine diagnostic workup. Information collected included the presenting complaint and quantification of N2O use (including pattern, frequency and quantity), the delay between symptom onset and initial presentation, follow-up adherence, treatment compliance and outcomes at discharge. Data were extracted from CRS Millennium and pseudonymised for analysis.
Clinical outcomes were based on symptoms, objective clinical signs and assessments. These included the 10 metre walk test (MWT), sensory and motor examination, Romberg’s test and overall gait stability. Based on patients’ recovery, outcomes were categorised into four groups: full recovery, substantial improvement, partial improvement and no improvement. Full recovery was defined as the resolution of all neurological deficits, with patients returning to their baseline functional status. Substantial improvement was characterised by marked progress in these clinical outcomes. Partial improvement was defined by persistent but improved symptoms, while no improvement indicated minimal or no clinical improvement despite treatment.
Descriptive and summary statistics were used to report demographic characteristics and N2O usage. Independent samples’ t-tests were conducted on continuous data (eg, age) and χ2 test of independence or Fisher’s exact test were conducted on categorical data (eg, clinical features). For the 10MWT analysis, the Wilcoxon Signed-Rank test was used to compare paired baseline and follow-up data, and the Mann-Whitney U test was used to compare unpaired 10MWT outcomes between SIP and non-SIP groups.
Missing data were minimal, primarily due to missed appointments or follow-up assessments. These were reported descriptively, and no imputation methods were used.
Potential sources of bias were considered, including reliance on self-reported adherence data among SIP participants, which may be subjected to reporting bias. Demographic variables were compared between groups to assess baseline imbalances that could confound clinical outcomes.
Results
Overview
A total of 52 patients were referred to the N2O-SDEC pathway at RLH between 3 June 2024 and 17 December 2024. Six patients (11.5%) were excluded from subsequent analyses due to non-attendance at two consecutive initial SDEC appointments. A total of 25 patients (54.3%) opted to self-inject hydroxocobalamin during this period (SIP), while 21 patients (45.7%) attended the medical SDEC to receive their injections from a nurse (non-SIP).
Demographics
Demographics were similar across groups (see table 1). Of the 46 patients included in this service development project, most were young with a median age of 23 years (IQR 21–27) in both groups. The majority of SIP were male (72%, n=18), while the non-SIP group had a closer gender balance (female 47.6%, n=10). No differences in ethnicity were present across groups, with most patients identifying as Asian or Asian British (n=29, 63%) or black (n=13, 28%). Only half of all patients (50%) were employed or in education.
Table 1. Characteristics of self-injecting and non-self-injecting patients.
| SIP (n=25) | Non-SIP (n=21) | Total (n=46) | Test statistic (df) | P value | |
|---|---|---|---|---|---|
| Age (median) | 23 (17–36) | 23 (17–45) | 23 (17–45) | t(44)= −0.53 | 0.611 |
| Gender | χ²(1)= 0.013 | 0.91 | |||
| Male | 18 (72%) | 11 (52.4%) | 29 (63%) | ||
| Female | 7 (28%) | 10 (47.6%) | 17 (37%) | ||
| Ethnicity | χ²(2)= 3.67 | 0.159 | |||
| South Asian | 19 (76%) | 10 (47.6%) | 29 (63%) | ||
| Black | 5 (20%) | 8 (38.1%) | 13 (28.3%) | ||
| Mixed/White/other | 1 (4%) | 3 (14.3%) | 4 (8.7%) | ||
| Employment | χ²(3)= 3.45 | 0.323 | |||
| Employed | 10 (40%) | 8 (38.1%) | 18 (39.1%) | ||
| Unemployed | 12 (48%) | 9 (42.9%) | 21 (45.7%) | ||
| Studying | 1 (4%) | 4 (19%) | 5 (10.9%) | ||
| Unknown | 2 (8%) | 0 | 2 (4.3%) | ||
| N2O use | χ²(1)= 2.07 | 0.027 | |||
| Regular | 21 (84%) | 11 (52.4%) | 32 (69.6%) | ||
| Sporadic | 4 (16%) | 6 (28.6%) | 10 (21.7%) | ||
| Amount per week (cylinders) | 1–42 (median 5) | 0.5–40 (median 6) | 5.5 (IQR 2–7.5) | ||
| Not specified | 0 | 4 (19%) | 4 (8.7%) | ||
| Other drugs | χ²(2)= 0.04 | 0.842 | |||
| Yes | 5 (20.8%) | 5 (23.8%) | 10 (22%) | ||
| No | 18 (75%) | 16 (76.2%)) | 34 (74%) | ||
| Unknown | 1 (4.2%) | 1 (4.8%) | 2 (4%) |
“Regular use” was defined as nitrous oxide use on a weekly or more frequent basis, whereas “sporadic use ” referred to less frequent patterns of use.
N₂O, nitrous oxide; Non-SIP, non-self-injecting programme; SIP, self-injection programme; t, two-sample t-test.
N2O use
Regular N2O use was reported by 32/42 patients (70%). There was some evidence for a difference in N₂O usage patterns between the two groups (see table 1). Regular N₂O use was defined as self-reported weekly or more frequent ongoing use; a higher proportion of SIP (84%) reported regular use compared with non-SIP (52.4%). Ten individuals (22%) reported sporadic use of N2O; four in the SIP group and six in the non-SIP group. Weekly N2O consumption was similar between the two groups, ranging from 1 to 42 large cylinders in SIP (median of 5 cylinders), and 0.5 to 40 large cylinders in non-SIP (median of 6 cylinders). Each large N2O cylinder contains approximately 600 g of N2O.7
Four patients in the non-SIP group reported using N₂O but did not specify the amount or frequency of use. However, their clinical examinations and investigations were consistent with those of other patients in the study.
In addition to N2O use, 22% of all patients reported the use of other recreational drugs.
Clinical presentation
There were no significant group differences (p>0.16 to 1.00) in symptoms at baseline (see table 2). All patients presented with sensory disturbance in the limbs. Paraesthesia in both upper and lower limbs was present in approximately two-thirds of all patients (65%).
Table 2. Clinical characteristics.
| Clinical features | SIP (n=25) | Non-SIP (n=21) | Total (n=46) | Test statistic (df) | P value | |
|---|---|---|---|---|---|---|
| Clinical presentation | Paraesthesia/numbness—UL and/or LL | 25 (100%) | 21 (100%) | 46 (100%) | NA | NA |
| Weakness | 11 (44%) | 6 (28.6%) | 17 (37%) | χ²(1)= 1.45 | 0.23 | |
| Unsteadiness | 9 (36%) | 10 (47.6%) | 19 (41%) | χ²(1)= 0.67 | 0.41 | |
| Pain | 6 (24%) | 5 (23.8%) | 11 (24%) | χ²(1)= 0.01 | 0.92 | |
| Cognition* | 6 (24%) | 1 (4.8%) | 7 (15%) | FET | 0.06 | |
| Falls | 3 (12%) | 3 (14.3%) | 6 (13%) | FET | 0.99 | |
| Vision changes | 1 (4%) | 2 (9.5%) | 3 (6.5%) | FET | 0.59 | |
| Sleep disturbances | 3 (12%) | 0 | 3 (6.5%) | FET | 0.24 | |
| Urinary problems | 4 (16%) | 2 (9.5%) | 6 (13%) | FET | 0.68 | |
| Bowel problems | 1 (4%) | 1 (4.8%) | 2 (4.3%) | FET | 0.99 | |
| Stiffness | 2 (8%) | 1 (4.8%) | 3 (6.5%) | FET | 0.99 | |
| Genital numbness | 0 | 1 (4.8%) | 1 (2.2%) | FET | 0.47 | |
| Lhermitte’s sign | 4 (16%) | 2 (9.5%) | 8 (17.4%) | FET | 0.68 | |
| Speech disturbance | 2 (8%) | 1 (4.8%) | 3 (6.5%) | FET | 0.99 | |
| Examination findings | Reduced power | 8 (32%) | 5 (23.8%) | 13 (28.3%) | χ²(1)= 0.45 | 0.50 |
| Abnormal sensory | 23 (92%) | 18 (85.7%) | 41 (89%) | χ²(1)= 0.67 | 0.41 | |
| Ataxic gait | 10 (40%) | 8 (38.1% | 18 (39%) | χ²(1)= 0.02 | 0.89 | |
| Romberg sign | 13 (52%) | 9 (42.9%) | 22 (48%) | χ²(1)= 0.45 | 0.50 | |
| Treatment | Vitamin B12 injections (median) | 12 (6–30) | 12 (6–42) | MW=216 | 0.15 | |
| Treatment delay (days) | 11.5 (1-90) | 14 (0–180) | MW=238 | 0.61 | ||
| Number of missed B12 injections (n=64) | 13 (20.3%) | 51 (79.7%) | 64 | χ²(1)= 15.67 | <0.001 |
Cognition=self-reported memory issues, confusion, drowsiness, hallucinations.
MW=Mann-Whitney U test.FET, Fisher’s Exact test; LL, lower limbs; MW, Mann-Whitney U test; SIP, self-injection programme; UL, upper limbs.
Muscle weakness was noted in 17/46 patients (37%), with a higher prevalence among SIP (44%) compared with non-SIP (29%). Nineteen patients (41%) reported unsteadiness or difficulty maintaining balance, with a similar frequency across the two groups.
Physical examination
Baseline clinical examination did not significantly differ between groups (see table 2). Abnormal sensory findings were the most common clinical sign, observed in 89% of all patients. Reduced muscle strength, primarily in the lower limbs, was observed in 13/46 patients (28%), with a slightly higher proportion affected in the SIP. Difficulty with balance was evident in 18/46 individuals (39%). Romberg’s sign was positive in around half of all patients (48%, n=22).
Treatment
On average, patients in both groups received a median of 12 injections of hydroxocobalamin (ie, 4 weeks of treatment). The number of injections ranged from 6 to 30 in SIP and 6 to 42 in non-SIP.
Missed or delays to treatment
A total of 64 injections were missed during this 6-month period. Missed injections were significantly more common among non-SIP 79.7% (n=51), compared with 20.3% (n=13) in SIP (p<0.001, see table 2).
Delays from symptom onset to hospital presentation ranged from 0 to 180 days, with an average delay of 11.5 days in SIP, compared with 14 days in the non-SIP group. There was no significant difference in treatment delay between the SIP and non-SIP groups.
Missed review appointments
A total of 39 face-to-face appointments were missed over the 6-month period across the 46 patients, with no significant difference between SIP (54%) and non-SIP (46%) groups, respectively (see table 3).
Table 3. Missed appointments.
| DNA appointment | SIP (n) | Non-SIP (n) | Total DNAs (%) | Test statistic (df) | P value |
|---|---|---|---|---|---|
| First week | 0 | 2 | 2 (5%) | χ²(1)= 0.50 | 0.48 |
| Second week | 6 | 4 | 10 (26%) | χ²(1)= 0.40 | 1.00 |
| Third week | 10 | 6 | 16 (41%) | χ²(1)= 0.09 | 0.77 |
| Fourth week | 2 | 3 | 5 (13%) | χ²(1)= 0.22 | 0.64 |
| Fifth week | 2 | 2 | 4 (10%) | χ²(1)= 0.00 | 1.00 |
| Sixth week | 1 | 1 | 2 (5%) | χ²(1)= 0.00 | 1.00 |
| Total | 21 (54%) | 18 (46%) | 39 (100%) | χ²(1)= 0.06 | 0.81 |
DNA, did not attend; SIP, self-injection programme.
A pattern in patient attendance was evident in both groups, with more than two-thirds of missed appointments (67%, n=26) occurring during the second and third weeks of treatment (figure 1). The third week had the highest rate of failed attendance, accounting for 41% (n=16) of all missed face-to-face appointments. The fourth and fifth weeks combined accounted for a total of nine missed appointments (23%).
Figure 1. Distribution of DNA appointments across weekly intervals.
Outcomes
Clinical outcomes were categorised into full recovery, substantial improvement, partial improvement and no improvement. No differences were observed between the two groups (see table 4).
Table 4. Clinical outcomes.
| Group | Full recovery | Substantial improvement | Partial improvement | No improvement | Test statistic (df) | P value (χ² test) |
|---|---|---|---|---|---|---|
| SIP (n=25) | 3 (12%) | 15 (60%) | 5 (20%) | 2 (8%) | χ²(3) = 0.63 | 0.889 |
| Non-SIP (n=21) | 4 (19%) | 12 (57%) | 3 (14.3%) | 2 (9.5%) | ||
| Total (n=46) | 7 (15%) | 27 (59%) | 8 (17%) | 4 (9%) |
SIP, self-injection programme.
Substantial improvement was the predominant outcome across both groups, accounting for more than half of the evaluated patients (n=27, 59%); 15/46 patients (33%) were SIP, while 12/46 (26%) were non-SIP. Patients in this category showed significant recovery with some residual symptoms that did not limit their daily life activities.
Seven individuals (15%) showed full recovery to baseline, with the majority (five out of the seven patients) presenting with mild residual symptoms. A comparable proportion of patients (17%, n=8) demonstrated a partial recovery but with persistent symptoms. Approximately 10% showed no improvement (n=4, 9%).
Ten-metre walk test
The 10MWT was used as quantitative outcome measure. Median 10MWT times were compared between SIP and non-SIP at baseline and follow-up to assess functional recovery (table 5).
Table 5. Comparison of 10-MWT between SIP and Non-SIP groups at baseline and follow-up.
| Group | Time point | Median (IQR, s) | SE | Test statistic | P value |
|---|---|---|---|---|---|
| SIP | Baseline (n=25) | 16 (12–20) | 2.6 | WSR=16.5 | 0.002 |
| Follow-up (n=18) | 12 (10–15) | 0.84 | |||
| Non-SIP | Baseline (n=21) | 13 (11–20) | 2.5 | WSR=11 | 0.04 |
| Follow-up (n=15) | 11 (10–13) | 0.81 | |||
| SIP vs non-SIP (baseline, follow-up) | MW=101 | 0.22 |
MW, Mann-Whitney U test; 10MWT, Ten-metre walk test; SIP, self-injection programme; WSR, Wilcoxon Signed-Rank test.
At baseline, both groups showed comparable walking speeds, with the SIP group demonstrating a slightly slower median time (16s, IQR: 12–20) compared with non-SIP (13s, IQR: 11–20), though this difference was not statistically significant. Both groups demonstrated improved walking speed at follow-up, with the SIP group improving to 11s and the non-SIP to 12s.
Discussion
N₂O-related neurotoxicity is a public health concern in East London and other parts of the UK.7 Treatment of neurological symptoms involves administering IM hydroxocobalamin on alternate days until no further neurological improvement is observed, along with discontinuation of N₂O use.3 This service development project compared adherence and clinical outcomes between patients who self-administered hydroxocobalamin injections (SIP) and those who had hydroxocobalamin injections administered by a medical SDEC nurse (non-SIP). Our findings indicate that self-administration of vitamin B12 may be a feasible and effective alternative to nurse-led administration, with the potential to improve treatment adherence.
Treatment adherence emerged as a key distinguishing factor between the groups. The SIP group had fewer missed injections (20.3%) than non-SIP (79.7%). However, self-reporting bias cannot be excluded, since SIP patients may have overestimated their adherence. Self-administration of medication has been shown to improve compliance in other settings.13,15
Missed review appointments followed a notable pattern, with the highest rates of non-attendance occurring in the third week of follow-up (n=16, 41%). This trend was consistent across both groups and may suggest that early symptom resolution contributed to decreased attendance as observed in other outpatient settings.16
Clinical outcomes were broadly comparable between groups with no significant differences observed. Substantial improvements were reported in over half of all patients (59%). The proportion of patients with partial improvement or no improvement was also similar between the two groups, suggesting that self-injection does not negatively impact recovery trajectories.
These findings have important implications for the management of N₂O-induced neurotoxicity. The feasibility of self-injection may reduce the burden on healthcare services, particularly in high-demand settings such as emergency care. Additionally, improved treatment adherence in SIP suggests that empowering patients with self-care strategies could enhance long-term management.
N₂O usage patterns differed between the two groups, with a higher proportion of SIP reporting regular use compared with non-SIP. This difference was not statistically significant, which might reflect the modest sample size. Individuals with heavier N₂O consumption may have been more motivated to adopt self-injection, potentially to maintain treatment continuity and improve outcomes. However, it is not clear whether higher treatment adherence among SIP was a result of greater motivation for recovery or simply a consequence of logistical convenience and accessibility.
Clinical presentations were largely consistent between the two groups, with paraesthesia, muscle weakness and unsteadiness being the most common symptoms. All patients presented with paraesthesia or numbness, a finding that aligns closely with previous studies.7 17 18 The higher proportion of SIP patients with muscle weakness (44% vs 28.6%) might suggest a greater degree of neurological impairment at presentation due to their more frequent N₂O use. The demographic characteristics of the cohort also align with previously reported data on N₂O-associated neurological toxicity, indicating a young patient population with a predominance of male and Asian ethnicity.719,21 Only half of the participants were either employed or in education. SIP may therefore improve accessibility for individuals who have daytime commitments, offering greater flexibility and reducing the indirect costs associated with travel.
This project offers valuable insights into the feasibility and effectiveness of self-injection of hydroxocobalamin in a high-demand ambulatory care pathway, which can be applied to other healthcare settings. The project also highlights the potential of self-injection to improve treatment adherence, while also optimising healthcare resources by reducing missed appointments and reducing workload from emergency care pathways.
Several limitations should be considered: (1) adherence rates among SIP may be subjected to self-reporting bias, potentially overestimating compliance; (2) the project primarily focuses on short-term outcomes, leaving uncertainty regarding long-term neurological recovery and relapse rates; (3) as a single-centre study in a metropolitan cohort of young N₂O users, predominantly Asian or Asian British; the findings may have limited generalisability to settings with different demographic or healthcare profiles.
Conclusion
This project demonstrated that self-injection of IM hydroxocobalamin is a feasible and effective alternative to standard nurse-led administration for patients with N₂O-induced neurotoxicity. It may improve adherence while maintaining comparable clinical outcomes. The integration of self-administration into acute care pathways may enhance accessibility to treatment, reduce healthcare burden and improve patient autonomy.
Footnotes
Funding: MH was funded by grants from QMUL Impact Fund and Tower Hamlets Council
Data availability free text: The data used in this study were collected as part of routine clinical care and are stored securely within the Barts Health NHS Trust electronic patient record system. Deidentified data may be made available upon reasonable request, subject to appropriate approvals and data sharing agreements with Barts Health NHS Trust.
Patient consent for publication: Not applicable.
Provenance and peer review: Not commissioned; externally peer-reviewed.
Ethics approval: This study was conducted as a clinical service development project at Barts Health NHS Trust and was exempt from formal NHS Research Ethics Committee approval, in accordance with Health Research Authority (HRA) guidance. No identifiable patient data were used, and all data were collected as part of routine clinical care.
Data availability statement
Data are available in a public, open access repository.
References
- 1.Zaloum SA, Mair D, Paris A, et al. Tackling the growing burden of nitrous oxide-induced public health harms. Lancet Public Health. 2025;10:e257–63. doi: 10.1016/S2468-2667(24)00298-6. [DOI] [PubMed] [Google Scholar]
- 2.Office for National Statistics Drug misuse in England and wales - appendix table. 2020. [19-Feb-2025]. https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/datasets/drugmisuseinenglandandwalesappendixtable Available. Accessed.
- 3.Paris A, Lake L, Joseph A, et al. Nitrous oxide-induced subacute combined degeneration of the cord: diagnosis and treatment. Pract Neurol. 2023;23:222–8. doi: 10.1136/pn-2022-003631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Keddie S, Adams A, Kelso ARC, et al. No laughing matter: subacute degeneration of the spinal cord due to nitrous oxide inhalation. J Neurol. 2018;265:1089–95. doi: 10.1007/s00415-018-8801-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Evans EB, Evans MR. Nangs, balloons and crackers: Recreational nitrous oxide neurotoxicity. Aust J Gen Pract . 2021;50:834–8. doi: 10.31128/AJGP-10-20-5668. [DOI] [PubMed] [Google Scholar]
- 6.Wong J, Viyasar T, Layton B, et al. The dangers of recreational inhalation of nitrous oxide. Br J Hosp Med. 2021;82:1–8. doi: 10.12968/hmed.2021.0322. [DOI] [PubMed] [Google Scholar]
- 7.Mair D, Paris A, Zaloum SA, et al. Nitrous oxide-induced myeloneuropathy: a case series. J Neurol Neurosurg Psychiatry . 2023;94:681–8. doi: 10.1136/jnnp-2023-331131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Asmussen Frank V, MacLean S, Herold MD. Nitrous oxide use among young people – new trends, policy challenges, and knowledge gaps. DAT . 2020;20:383–92. doi: 10.1108/DAT-09-2020-0062. [DOI] [Google Scholar]
- 9.Joerres C, Patyjewicz M, Cetin M, et al. Assessing the prevalence of nitrous oxide usage in patients aged 35 years or under presenting with unprovoked VTE between 2021-2023. Thrombosis Update . 2024;16:100184. doi: 10.1016/j.tru.2024.100184. [DOI] [Google Scholar]
- 10.Patyjewicz M, Mair D, Zaloum SA, et al. Recreational nitrous oxide and thrombotic events: a case series. BMJ Neurol Open . 2024;6:e000619. doi: 10.1136/bmjno-2023-000619. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Noh T, Osman G, Chedid M, et al. Nitrous oxide-induced demyelination: Clinical presentation, diagnosis and treatment recommendations. J Neurol Sci. 2020;414:116817. doi: 10.1016/j.jns.2020.116817. [DOI] [PubMed] [Google Scholar]
- 12.Zaloum SA, Paris A, Mair D, et al. Evaluation of an ambulatory care pathway for patients with nitrous oxide-induced myeloneuropathy. BMJ Neurol Open . 2024;6:e000737. doi: 10.1136/bmjno-2024-000737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ito M, Kamata M, Uchida H, et al. Introduction of self-injection increases compliance with dupilumab administration in adult patients with atopic dermatitis: A single-center retrospective study. JAAD Int . 2020;1:91–2. doi: 10.1016/j.jdin.2020.06.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hajialibeigloo R, Mazlum SR, Mohajer S, et al. Effect of self-administration of medication programme on cardiovascular inpatients’ medication adherence and nurses’ satisfaction: A randomized clinical trial. Nurs Open. 2021;8:1947–57. doi: 10.1002/nop2.870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.van Herpen-Meeuwissen LJM, van Onzenoort HAW, van den Bemt PMLA, et al. The Effect of Self-Administration of Medication During Hospitalization on Patient’s Self-Efficacy and Medication Adherence After Discharge. Patient Prefer Adherence. 2022;16:2683–93. doi: 10.2147/PPA.S375295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Alturbag M. Cureus; 2024. Factors and reasons associated with appointment non-attendance in hospitals: a narrative review.https://www.cureus.com/articles/245772-factors-and-reasons-associated-with-appointment-non-attendance-in-hospitals-a-narrative-review Available. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zheng D, Ba F, Bi G, et al. The sharp rise of neurological disorders associated with recreational nitrous oxide use in China: a single-center experience and a brief review of Chinese literature. J Neurol. 2020;267:422–9. doi: 10.1007/s00415-019-09600-w. [DOI] [PubMed] [Google Scholar]
- 18.Kulkantrakorn K, Chunhachatrachai P, Kulkantrakorn W. Nitrous oxide abuse and associated neurological diseases. BMC Neurol. 2024;24:461. doi: 10.1186/s12883-024-03966-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Malek N, Nihat A, Mattoscio M, et al. A recent surge of nitrous oxide misuse around London which merits a public health warning. Postgrad Med J. 2024;100:174–8. doi: 10.1093/postmj/qgad121. [DOI] [PubMed] [Google Scholar]
- 20.Zhang J, Xie D, Zou Y, et al. Key Characteristics of Nitrous Oxide-Induced Neurological Disorders and Differences Between Populations. Front Neurol. 2021;12:627183. doi: 10.3389/fneur.2021.627183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zheng R, Wang Q, Li M, et al. Reversible Neuropsychiatric Disturbances Caused by Nitrous Oxide Toxicity: Clinical, Imaging and Electrophysiological Profiles of 21 Patients with 6–12 Months Follow-up. NDT. 2020;16:2817–25. doi: 10.2147/NDT.S270179. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data are available in a public, open access repository.