Nitrous oxide (N2O) is an inhaled anaesthetic gas and a popular intoxicant. Excessive recreational use can cause spinal cord myelopathy. Previous studies have discussed the medical management. However, none have specified the sensorimotor rehabilitation management. This case report documents the investigations, physical rehabilitation and functional outcomes in two cases of N2O-associated myelopathy. Both presented with lower limb strength and sensorimotor integration impairments resulting in ataxic ambulation. Dorsal column signal abnormality was observed on T2-weighted MRI in one case. Myelopathy was diagnosed based on clinical presentation and both were treated with vitamin B12. Rehabilitation was conceived and specified using the Rehabilitation Treatment Specification System (RTSS). Both cases achieved independent indoor gait on hospital discharge, and full function at 9 months in one case. Appropriate and timely medical management and reasoned rehabilitation provided excellent functional outcomes for N2O-related myelopathy. By using the RTSS, reasoned rehabilitation efficacy can be tested in the future.
- Neurology (drugs and medicines)
- Healthcare improvement and patient safety
- Medical management
- Spinal cord
- Physiotherapy (rehabilitation)
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
- Neurology (drugs and medicines)
- Healthcare improvement and patient safety
- Medical management
- Spinal cord
- Physiotherapy (rehabilitation)
The analgesic and intoxicating effects of the colourless inorganic gas nitrous oxide (N2O) have been known for over 200 years.1 2 As an inhaled analgesic, N2O acts as an N-methyl-D-aspartate antagonist which decreases excitatory neurotransmission throughout the central nervous system via non-competitive glutamate inhibition.3 When prescribed as a mixture with 50% oxygen, it can be safely administered by trained professionals (eg, nurses, midwives, dentists, paramedics) during painful or distressing procedures.4 As an inhaled intoxicant without oxygen, N2O (colloquially sweet-air, nos, noz, nox, hippie-crack5) elicits short-lived feelings of euphoria with pleasurable psychedelic and empathogenic effects.6
Recreational N2O intoxication is popular. The UK lifetime-user prevalence rate from a non-probabilistic sample is 38.6%; typical users appear to be male clubbers in their 20s.7 The catering industry’s development of small 8 g steel recharging canisters for whipped cream dispensers containing pressurised N2O (bulbs, whippits, nangs) has made them ubiquitous to buy cheaply (50p–£2 per canister8) usually in packs of 100. Use from large cylinders (2–10 kg) has become more prevalent recently.9 N2O remains legal to buy in the UK although this is now under review10 not least due to recent media concern about the increase in chronic abuse,11 and the alarming increase in use reported in the Netherlands, which has led it to plan for formal controls in 2023.12 Consuming N2O directly from the canister is dangerous—the pressure (~200 kPa) can cause barotrauma,13 and the temperature of the gas near the outlet (−55°C14) can cause frostbite injuries.15 Instead, the gas is typically transferred into balloons via either the whipped cream dispenser or a secondary device (creamers or crackers) easily available that can dispense the canister gas into balloons inserted over the device spout. The preferred recreational use is to inhale or rebreathe the N2O via the balloon at parties and festivals.16
Although recreational N2O use is relatively safe compared with other drugs, acute harm includes falls, accidental injury, confusion7 or vomit aspiration.17 Acute death is rare. A 2016 systematic review of case reports found evidence of 29 case deaths mostly due to acute asphyxiation due to hypoxia with users consuming N2O with a mask or plastic bag.18 In the UK, 17 deaths were reported between 2006 and 201219 mainly due to accidental asphyxia when N2O displaces oxygen if used in enclosed spaces.20 Most users inhale low doses per session (1–2 canisters), but overuse exists (~100 canisters), which can lead to persistent neurological sequelae. Chronic clinical toxicity is due to N2O’s interaction with vitamin B12 converting it from an active to an inactive form by irreversible oxidation.21 By impairing methylation reactions and DNA synthesis, oxidation results in an accumulation of homocysteine and impaired maintenance of the myelin sheath.22 Thus, B12 deficiency from excessive N2O use can cause demyelination of the central and peripheral nervous system and a diagnosis of subacute combined degeneration (SACD). If demyelination presents in the spinal cord, the dorsal columns are classically affected, with sometimes the lateral but rarely the anterior columns. The associated diagnostic procedures, neural impairments (progressive vibratory and proprioceptive sensory abnormalities, ascending paraesthesia, ataxic gait, hyporeflexia or hyper-reflexia, and rarely muscle weakness and loss of sphincter control23–25), and medical treatment options have been described in medical case reports and observational studies of SACD of the spinal cord previously.22 25–36
In contrast, rehabilitative treatment options for N2O abuse SACD are not as pervasive in the literature. It is as if patients are expected to recover naturally with cursory reassurance and good luck. One published case study advocates physicians to be cognisant that tailored rehabilitation for a patient’s functional and neuropsychiatric impairments can limit the extent of neurological damage,37 and another clinical review advises medical clinicians to refer to physiotherapy if a patient’s mobility is affected.38 Yet, rehabilitation treatment is not well specified even in these welcome cases.
This lack of rehabilitation specificity is not unusual in research into rehabilitation treatment efficacy where objective definitions of treatments of interest, or even ‘usual care’,39 are required. For surgical or pharmacological treatments, it is usual for active treatment ingredients to be obvious and be specified as chemical structures or anatomical results. In contrast, rehabilitation specification is challenging for several reasons. A binary definition of rehabilitation is tenuous because there are non-binary categories of rehabilitation,40 rehabilitation ingredients are not obvious, rehabilitation treatments attempt to change multiple interacting patient functions (eg, muscle strength, movement fluidity, participation in occupation) and specific treatment ingredient definitions do not exist, let alone their downstream effect on patient function.41 To date, the active ingredients of treatment at the macro level of acute inpatient rehabilitation are rarely specified, which equates with the ‘black box’ approach where little insight into the active treatment ingredients is given which is a typical criticism in the rehabilitation science and clinical communities.42 The recently developed coherent framework (the Rehabilitation Treatment Specification System (RTSS)) addresses the black box problem by developing methods that specify, and ultimately measure, the rehabilitative content (ingredients), that via theoretical or observed processes (mechanisms of action), cause a change in desired systems (targets).43 Because of relative rarity of both the clinical condition and the specification of its rehabilitation using the RTSS, the following report of two cases with spinal cord injury post-N2O overuse is intended to provide an exemplar of holistic medical and rehabilitation management. It was prepared following the Case Report guidelines.44
The patients were both male. Case 1 was in his early teens (mass 71.7 kg, height 183.0 cm, body mass index (BMI) 21.4 kg/m2) and case 2 was in his 20s (54.1 kg, 173.5 cm, BMI 18.0 kg/m2) on presentation to our large, urban hospital emergency department. Case 1 presented with a 2-day history of gradually worsening lower leg weakness, sensory disturbance in a stocking distribution, reduced balance and mobility. Case 2 presented with a 3-month history of altered leg sensation/weakness commencing with bilateral distal sensory disturbance progressing to pins and needles in a stocking distribution with bilateral leg weakness and reduced mobility.
Neither disclosed any medical history, nor recalled using other recreational drugs except for case 1 who chose to smoke cannabis occasionally. Both confirmed consuming five to six units of alcohol socially per week. History of N2O consumption was similar for both cases and varied from 50 to 100 canisters per month (usually with large quantities in a single sitting). However, both cases describe an acute escalation in consumption to 100–200 canisters in the week prior to hospital admission. Case 1 was in full-time education, and case 2 worked part time as a football coach. Both cases lived in rented flats; case 1 with his mother, and case 2 with his parents and siblings.
Both patients presented with normal cognition, orientation and fluent speech. The cranial nerves and upper limb examination were normal. There were no cerebellar signs. The sensorimotor lower limb impairments (table 1) included distal lower limb weakness measured using the Medical Research Council ordinal rating scale45 and translated into diminished ambulatory function with both cases requiring aids to transfer or walk. Neither case presented with any urinary or bowel dysfunction.
Methylmalonic acid (MMA) serum levels were elevated above the normal threshold level of 243 nmol/L,46 more so in case 1 (table 1) indicative of vitamin B12 deficiency. Serum folate levels in both cases were unremarkable (3.8 μg/L and 4.9 μg/L for cases 1 and 2, respectively; both within the normal range of 3.1–20.5 µg/L). Other investigations are summarised in table 1.
Thorough history taking identified a gradual onset of symptoms over several days, ruling out a vascular cause of neurological dysfunction such as spinal or cerebral stroke. Further neurological assessment and clinical investigations (table 1), along with a history of excessive N2O intake, led to a diagnosis of SACD of the spinal cord causing sensorimotor myelopathy.
Intramuscular hydroxocobalamin injections (1 mg) were provided daily for 1 week and then on alternate days for a further week. The patient was advised to cease all recreational drug use including N2O inhalation. Physiotherapy treatment (total with therapist 445 min) was provided in 11 sessions over a 12-day inpatient stay (mean 40.5 min/session). Clinical assessment revealed bilateral lower limb sensorimotor weakness, and a total score of 7 out of 56 on the Berg Balance Scale (BBS—a 14-item ordinal measure that assesses static balance, functional mobility and falls risk47) indicated difficulties in functional mobility and balance particularly a reliance on visual sensory cues, and poor planning of whole body transitional movement and reactions to centre-of-mass excursions. Seeing as the mean (±SD) BBS for a sample of ambulatory spinal cord injured-paraplegic participants (American Spinal Injury Association (ASIA) Impairment Scale D (AIS-D))48 is 44.8 (±13.0),49 case 1’s functional mobility and balance deficits at initial assessment were profound, a finding supported by their slow observed average (10 m) self-selected gait velocity (0.29 m/s) indicative of a home-only ambulator.50 Overall activities of daily living function was moderate (modified Barthel Index51 12 out of 20). The aims of physical rehabilitation treatment were safe community ambulation with or without walking aids. Treatment specification was undertaken based on the RTSS43 (table 2) and includes interventions designed to affect volitional rehabilitation behaviour using a behaviour change model that considers three essential conditions: capability, opportunity and motivation (COM-B system).52
Intramuscular hydroxocobalamin injections (1 mg) for 1 week were provided which was reduced to alternating days for a further week. He was also prescribed oral paracetamol 1 g four times a day, codeine phosphate 30 mg four times a day and gabapentin 300 mg three times a day for symptom control. Similarly to case 1, the patient was also advised to cease recreational N2O inhalation. Physiotherapy treatment (370 min) was provided in 12 sessions over a 14-day inpatient stay (30.9 min/session). Assessment revealed similar impairments to case 1: bilateral lower limb sensorimotor weakness (table 1). A combined time of 0/120 s on the modified Clinical Test of Sensory Interaction on Balance (CTSIB-M—a performance measure quantifying postural standing control under four sensory conditions each tolerated for up to 30 s for a maximum score of 120 s53 54) indicated their sensorimotor balance control was profoundly impaired. In fact, case 2’s impairments were so profound compared with case 1; they were unable to ambulate at all and required maximal assistance to transfer safely. The aims of treatment were nonetheless the same as in case 1 and remained safe community ambulation with or without walking aids. Treatment specification was undertaken based on the RTSS (table 2).
While folic acid treatment was not indicated to correct abnormal serum folate levels in either of our cases, it should be noted that correcting any diagnosed folate insufficiency should be undertaken only after initial phase B12 replacement to prevent worsening of the neurological spinal cord degeneration.55
Outcome and follow-up
Cases 1 and 2 improved with treatment and their MMA concentration levels were within the normal range prior to hospital discharge. There were however residual sensorimotor deficits. In case 1, plantarflexor strength was graded at 5/5 on reassessment prior to hospital discharge, but dorsiflexion strength had only recovered to 3/5 bilaterally. Proprioception was no longer impaired at the ankle, but distal impairments remained bilaterally. In addition, there were residual sensory impairments to light touch on the left in an L1–S1 and on the right an L5 dermatomal distribution. He was independently mobile with foot-up splints (due to dorsiflexor fatigue over longer distances) and two elbow crutches up to 45 m but could not ambulate outdoors. Average gait speed increased from 0.29 to 0.41 m/s, an increase of 0.12 m/s, which is a clinically meaningful difference56 but still less than the comfortable average gait velocity threshold of 0.49 m/s that delineates community from household-only ambulation.50 His sensorimotor recovery translated to a meaningful improvement in his BBS (47 of 56), and a Barthel score of 17 out of 20 on discharge. The patient was duly referred to local community physiotherapy services available at their discharge residence for treatment as specified.
On discharge, case 2 had residual impaired sensation to light touch bilaterally in an L4–S1 dermatomal distribution. Proprioception at the first metatarsals remained absent bilaterally. He was independently mobile indoors with two elbow crutches up to 60 m and was able to control the knee avoiding recurvatum during stance phases in ≥90% of gait cycles. Indoor independent gait velocity at discharge was 0.38 m/s. However, he could not ambulate outdoors independently similarly to case 1. His recovery translated to an improved CTSIB-M (71/120 s) revealing residual reliance on visual sensory information, and a Barthel score of 19 out of 20. He too was duly referred for outpatient therapy for treatment as specified; no community therapy options were available at his discharge residence.
Case 1 chose to not participate in further therapy or medical follow-up. In contrast, case 2 attended further physiotherapy and was reviewed in an outpatient medical clinic 3 weeks post-hospital discharge. He was ambulating independently but self-reported mildly impaired balance. While proprioception was intact on clinical examination, he presented with symmetrical, lower limb impaired light touch sensation in a stocking distribution, and could only tolerate the Romberg position (eyes closed)57 for 22 s without excessive sway or losing balance.
At a subsequent outpatient medical clinic review at 9 months post-hospital discharge, case 2 reported a full recovery including being able to sprint on a treadmill in his physiotherapy sessions. Repeat neurophysiology tests showed improved conduction velocities and normal sensory studies; however, compound muscle action potentials in the lower limbs remained below the normal range.
We observed profound sensorimotor impairment caused by excessive N2O abuse which led to SACD of the spinal cord in two cases. The cases presented with typical features that collectively confirmed the diagnosis. First, a clinical myelopathy anatomically localised to the spinal cord (confirmed by MRI studies in case 1); second, evidence of vitamin B12 deficiency; and lastly, an absence of identifiable central or peripheral nervous system pathology in keeping with the clinical findings.23 The lack of confirmatory imaging studies in case 2 is not unusual—a published series of 54 patients with dorsal column injury reported low sensitivity with imaging confirming the diagnosis only in 8 patients (14.8%).58 Previous literature has reported excessive N2O use and vitamin B12 deficiency cases whose main presenting symptoms were weakness, paraesthesia and ataxic gait,22 28 32 falls,31 33 absent or impaired deep tendon reflexes and diminished vibration sense,26 29 hyporeflexia,34 limb proprioceptive abnormalities27 30 and wide-based steppage gait,35 not dissimilar to the present cases.
While resolution of symptoms following vitamin B12 therapy is sometimes incomplete in older patients (mean (±SD) age 55.9 (±15.5) years), it tends to be complete in younger patients (39.8 (±18.8) years).23 So it was not surprising that the two young cases presented in this paper progressed in their recovery and made a near full sensorimotor recovery at 9 months, at least in the second case. Because their sensorimotor function returned to near normal, we can presume that the two cases did not re-abuse N2O while we followed them up. If they had re-abused N2O inhalation during the follow-up period, then B12 would have been deactivated very rapidly with a commensurate deterioration in neurology and we would have expected to have observed a relapse in balance and ambulation dysfunction.59 Re-abuse could be a sign of addiction to N2O although N2O inhalation addiction is contentious within the literature.18 60 61 While clinical discourse to persuade cessation of use appears to have been successful in these two cases, clinicians should be aware that patients presenting with acute recreational N2O toxicity should be provided with necessary psychosocial support if difficulties in self-selected cessation of abuse are suspected.38
The positive outcomes overall in the two cases presented here are almost certainly due to the medical diagnostic procedures and subsequent clinical decision to commence vitamin B12 treatment in a timely way with regular monitoring. What is less clear is whether the equally timely and novel specification of rehabilitation interventions using the RTSS led to a more reasoned and tailored physical therapy approach. To our knowledge, this is the first specified rehabilitation report using the RTSS in cases of this type. By acting as a coherent framework for specifying rehabilitation interventions based on the clinician’s treatment theory, the RTSS fractionates multidimensional and progressive rehabilitation interventions into entities (treatment components) that are amenable to clinical adoption (including among clinical colleagues) and empirical research.41 Thus, the specification described herein acts first as a reasoned approach for fellow rehabilitation clinicians to use with these patients downstream of the acute hospital; second, as an example for other clinicians treating similar cases; and lastly, as a record of the specified treatment to inform researchers. It is worth noting though that the treatment specifications provided are examples peculiar to the two cases presented. They are not meant as a prescriptive regime for all patients presenting with sensorimotor system impairments due to recreational N2O toxicity and functional B12 deficiency. Nonetheless, we contend that in successfully specifying the treatment in this way, the patient’s treatment was reasoned and tailored and thus verifies the RTSS’s utility.
The impairments (bilateral lower limb sensorimotor) and aims of treatment (restitution of normal ambulatory function) were specified identically between cases. Treatments were also identical based on organ function targets (muscle strengthening and stretching), and skills and habits targets (function balance and ambulation practise tasks). Representation targets were identical too. They were designed to influence cognitive processes when managing the musculoskeletal integrity and injurious falls risks inherently associated with distal lower limb paralysis, and ensure the consequences of failing to voluntary manage those risks were understood.
However, there were treatment specification differences between cases because case 2 initially presented with impairments so profound that no ambulatory function was possible. Case 2 presented with proximal lower limb joint symmetrical weakness and more widespread proprioceptive loss. Therefore, therapy time was prioritised towards specifying a more widespread and explicit strengthening programme and more rudimentary static balance tasks compared with case 1.
What's more, when case 2 progressed to ambulating, knee recurvatum (KR) was observed during some, but not all, stance phases. KR is a relatively common impairment observed in other neurological pathologies (eg, 30%–50% of ambulatory hemiplegic stroke survivors62 63). In normal continuous gait, coordination between pretibial muscle contraction (which restrains ankle plantarflexion and allows the tibia to move forward over the foot during weight acceptance) and triceps surae/quadriceps contraction (crucial to restraining tibial forward motion during mid-stance while the femoral segment progresses forward distal to it with extension at the knee) controls knee flexion and forward tibial movement over the stationary foot in stance phases.64
While the main cause for gait KR is premature overactivity or spasticity of the plantarflexors which prevents the knee from flexing during loading response in early stance63 65 or vasti hypertonia,66 67 we reasoned that KR in case 2 was due to dorsiflexion/plantarflexion muscle weakness, poor proprioception and reliance on forward placed walking aids. This is because no muscle hyper-reflexia was observed on assessment at the ankle or knee. The variability of KR observed in case 2 was probably due to instances when knee hyperextension was simply deployed to stabilise the knee in stance by adopting a quadriceps-avoidance strategy sometimes seen in anterior cruciate ligament deficient gait68 in combination with plantarflexion weakness during mid-stance.69 Our theory therefore was that treatment ingredients targeting muscle weakness and sensory integration (including proprioception) in static balance would ameliorate the KR in case 2. Passive treatment ingredients designed to target KR more directly to support the approach would include issuing of rigid ankle/foot orthotics which were avoided here. Instead, gait re-education was deployed via instructions based on implicit motor learning to avoid KR with delayed knowledge-of-performance feedback.70
Specified treatments were not then all identical between cases and support the argument that treatment was tailored for the two cases. This is one of the advantages of the RTSS. Its coherent framework is intended to administer the complexity of rehabilitation practice in the interactions of clinical impairments, dysfunctions and participation in society in order to specify individuals’ rehabilitation treatment.43 Yet, the two cases’ rehabilitation treatment aims were similar and were focused on ambulation function. Acute adult inpatient physical rehabilitation often includes stressing motor systems toward safe, independent ambulation because ambulatory function is a key milestone for discharge planning. This is important within the goal of reducing acute hospital length of stay (LOS) which is itself dependent on coherent, equitable and resourced rehabilitation services downstream of an acute setting. But the problem with yoking rehabilitation with minimising LOS is that the rehabilitation aim is often achievement of minimal adaptive ambulatory function necessary for a safe discharge. The aim instead should be the restitution 71 of ambulatory function, especially since stroke72 and spinal cord-injured73 neurorehabilitation patients express it as their primary goal. Restitution of ambulatory performance therefore requires rehabilitative progression towards premorbid performance, not merely an assessment of current ambulatory performance.
The RTSS includes accurate specification of rehabilitation progression and was included throughout the two cases described. For ambulatory function, a separate treatment ingredient was included so the two cases were proficient in self-monitoring their exertional fatigue—a common symptom burden for patients with acute spinal cord injury.74 That allowed the ambulatory treatment’s progression to be specified in both cases by attributing the therapeutic dose of ambulatory practise (distance or time walked) with the patient’s increase in self-rated exertional fatigue (using a dual Numerical and Face Rating Scale75) during ambulatory treatment ingredients.
Similar separate treatment ingredients were included in our specification designed for the cases to attain skills and habits in managing the risk to their musculoskeletal integrity by applying ankle resting splints and foot-up devices when ambulating. We contend that without the RTSS’s coherent framework, there would be a risk that therapists might not have clinically reasoned tailored treatments as thoroughly, nor been as accurate in specifying it.
In conclusion, the combination of timely medical treatment and specific and clinically reasoned rehabilitation in a structured format using the RTSS provided excellent functional outcomes in both the acute and subacute phases post-injury in two individuals who sustained N2O-induced spinal cord injury. Stronger conclusions can be drawn when future studies determine the impact of rehabilitation on clinical outcomes at the impairment, activity and participation levels in this patient cohort by virtue of using the RTSS.
Case 2 and his next of kin were able to provide their perspective of their assessment and treatment.
Before the hospital admission we had no idea about the possible side effects of the inhaled canisters. Lots of the young people where we live took them at parties and we believed that they were risk free, we had never heard of others experiencing medical complications from taking them.
The profound loss of ambulation was terrifying particularly because there were no initial explanation why these symptoms had occurred.
When we initially went to hospital the doctors weren’t sure what was the cause of the weak legs and the inability to walk. After the scans the doctors explained that there was a problem with the nerves in the back and they weren’t totally sure if my son’s walking would fully recover. They thought that the gas canisters were the cause because of some of the blood test results. We were all shocked that this had happened and it was scary to hear that someone so young might have long term effects that they would have to live with. It was hard for us not to know what the future may hold for him, we felt very worried.
However, the medical assessment and treatment and the inpatient physical therapy rehabilitation delivered enabled meaningful recovery – a recovery that was complete after receiving longer term community physiotherapy. We feel lucky that my son has made a full recovery, it was slow to begin with and he had to work hard in physio sessions in the hospital. Unfortunately it took quite a long time for him to get seen in the physio clinic after leaving hospital, but he’s now managing to do everything he could do before the injury. We are thankful for all of the care that my son received.
Rates of nitrous oxide recreational use are rising, meaning that incidents of abuse and associated subacute combined degeneration pathology may also rise leading to profound sensorimotor dysfunction.
Timely diagnosis and medical treatment to reverse vitamin B12 deficiency in combination with skilled physical rehabilitation led to favourable outcomes.
Specified treatment from rehabilitation clinicians in the acute phase after neurological pathology in these two cases shows that timely, tailored and theoretically reasoned treatments can be put into practice if the goal is to restitute normal movement function.
Patient consent for publication
Contributors CB—conceptualisation, methodology, validation, investigation, resources, data curation, writing (review and editing). JB—conceptualisation, methodology, validation, investigation, resources, writing (review and editing). JG—conceptualisation, methodology, validation, resources, writing (original draft), writing (review and editing), visualisation, investigation, writing (review and editing). GDJ—conceptualisation, validation, resources, data curation, writing (original draft), writing (review and editing), visualisation, project administration.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.