Concentration and Detection Time of Nitrous Oxide in Blood Following Controlled Inhalation

ABSTRACT

Recreational use of nitrous oxide (laughing gas, N₂O) is becoming increasingly common and abuse is often seen in relation to driving, posing significant traffic‐safety concerns. Only a few studies exist on blood concentrations of N₂O and its detectability over time after use. In this study, 11 volunteers received controlled administration of 50% N₂O for 10 min, after which blood samples were drawn and analyzed for N₂O by headspace‐gas chromatography–mass spectrometry (HS‐GC–MS). Pharmacokinetic modelling indicated that elimination of N₂O can best be described by a two‐compartment model with half‐lives of 2.4 and 31 min. Although both the pharmacological effect and intoxication typically disappear within minutes after intake, N₂O remained detectable in blood for an average of 62 min at a cutoff of 0.2 mL/L and 132 min at a cutoff of 0.05 mL/L.

In a study with controlled inhalation of nitrous oxide (N₂O), elimination followed a two‐compartment model with half‐lives of 2.4 and 31 min, and N₂O remained detectable in blood samples for 1–3 h after inhalation.

1. Introduction

Nitrous oxide (N₂O) is a colorless and inert gas used for medical purposes as an inhalation anesthetic for mild sedation [1]. When inhaled, it quickly induces analgesia and euphoria for a few minutes before the effect declines. Reported side effects include mood changes, altered body awareness, impaired coordination, dizziness, disorientation, and fainting [2, 3].

Owing to the established use as an inhalation anesthetic, the uptake phase (wash‐in) and circulation phase (maintenance) during inhalation are well described [4, 5, 6, 7, 8]. In contrast, the elimination phase (wash‐out) has only been investigated through exhaled breath sampling, as this approach enables real‐time monitoring in clinical settings. O'Reilly et al. showed that the elimination profiles derived from breath analysis followed a two‐phased exponential decay (two‐compartment model), with reported half‐lives of approximately 2 and 20 min [9]. This elimination profile was recently corroborated in a study employing a lung simulator by Jiménez et al. [10]. Although the presence of a third, extremely rapid elimination phase or an additional model compartment has been suggested, this has never been conclusively confirmed [9].

Recreational use of N₂O while driving is an increasing concern across Europe [11, 12, 13]. A recent Danish study reported an increase in the number of blood samples submitted for forensic analysis of N₂O in traffic‐related cases between 2020 and 2022 [14]. For driving under the influence of drugs in general, legislation is most often based on either proving impairment or on fixed limits (per se), where limits are set for drug concentrations in biological samples such as blood, oral fluid or exhaled breath [15]. Previously N₂O has been measured in both exhaled breath and blood samples [10, 14, 16]. Since intoxication due to N₂O typically only lasts few minutes proving impairment can be impractical or impossible, and quick collection of biological samples is needed. However, little is known about the detection time of N₂O in blood after intake. Therefore, establishing reference concentrations and detection times of N₂O in blood is crucial for accurate interpretation in cases with suspected driving under the influence of N₂O.

The aim of this study was to determine concentrations and detection times of N₂O in blood after controlled inhalation using headspace‐gas chromatography–mass spectrometry (HS‐GC–MS). The elimination profile of N₂O was investigated to determine the half‐lives.

2. Materials and Methods

2.1. Study Design

The study was conducted at the Department of Anesthesiology, Surgery and Trauma Centre, Copenhagen University Hospital—Rigshospitalet, Denmark. Healthy volunteers were recruited among local undergraduate and graduate medical students. All participants provided written informed consent and received monetary compensation for their participation. Both male and female participants were included. Each participant underwent a screening process to verify eligibility for inclusion in the study. The screening procedure included a questionnaire addressing relevant medical history and a blood test for vitamin B₁₂ levels. Female participants were required to provide a negative pregnancy test (hCG urine test) prior to N₂O administration. Detailed inclusion and exclusion criteria are provided in the Supporting Information. In total, 20 participants were screened for eligibility, of whom 12 met the inclusion criteria, and 11 completed the full experimental visit (Figure S1).

2.2. Nitrous Oxide Administration

Kalinox, a pre‐mixed medicinal‐grade gas containing 50% N₂O and 50% O₂, was supplied in 5 L (170 bar) canisters by Air Liquide Danmark A/S (Horsens, Denmark). Two canisters were used throughout the study. The active canister was connected to a demand valve and single use anesthesia masks, secured using elastic head harnesses. Participants received the 50% N₂O mixture for 10 min.

2.3. Blood Samples

Blood samples were collected from a cannula inserted into the left radial artery of the participant by trained medical staff. Blood samples were collected before inhalation of N₂O and at 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 40, 50, 60, 70, 80, 90, 120, 150, 180, 240, and 300 min after cessation of N₂O inhalation. The exact sampling time for each blood sample was recorded. All samples were collected in 5 mL grey‐top Vacutainer tubes and stored at −20°C until analysis.

2.4. Nitrous Oxide Analysis

Quantitative analysis of N₂O in whole blood was performed by HS‐GC–MS. Aliquots of 250 μL headspace gas phase above the blood in the sample tubes were transferred with a gastight syringe to sealed, empty 20 mL headspace vials. Xenon gas (3 μL) was added as an internal standard using a separate gastight syringe. The headspace vials were analyzed on an Agilent 7890A GC with a 7697A headspace autosampler and 5975C MS system fitted with a HP‐PLOT/Q column (15 m × 320 μm × 20 μm). Duplicate measurements were performed for all samples. Correction factors were applied to correct for the thermodynamic equilibrium in the blood tube prior to extraction. Correction factors were calculated for each sample based on the specific volume of blood in the collection tube using a gas to blood partition coefficient of 0.67. The method, including the applied correction factors, was validated within a concentration range of 0.20–100 mL/L with imprecision and bias below 23% and 17%. Limit of detection (LOD) was 0.05 mL/L.

2.5. Data Analysis

Nonlinear regression analysis of individual time‐concentration profiles was performed in GraphPad Prism (Version 10.6.0) [17]. Both two‐ and three‐compartment first‐order elimination models were assessed according to Equations (1) and (2), respectively:

$$c=A·e^{-\mathit{\alpha t}}+B·e^{-\mathit{\beta t}}$$ (1)

$$c=A·e^{-\mathit{\alpha t}}+B·e^{-\mathit{\beta t}}+C·e^{-\mathit{\gamma t}}$$ (2)

where c is the N₂O concentration measured in the blood, α, β, and γ are rate‐constants for the first, second, and third compartments, respectively, and A, B, and C are the concentrations at zero accounted for by the respective compartments. Least squares regression with a maximum of 10,000 iterations with default convergence criteria, 1/Y² weighting, and no special handling of outliers was used for all models. All measured concentrations above 0.05 mL/L (LOD, 1/4 of lower limit of quantification, LLOQ) were included as continuous data for model analysis to minimize model bias [18].

The two‐ and three‐compartment models for each participant were then compared using both the extra sum‐of‐squares F‐test and the Akaike information criterion (AIC) to assess which model provided the best fit. Alpha was set to 0.05, with the two‐compartmental model representing the null hypothesis.

2.6. Ethical Approval

Ethical approval was obtained through scientific and ethical review by national competent authorities and ethics committees under the European Clinical Trial Regulation, EU No 536/2014 (EU CT number: 2023‐509993‐49‐00). Additionally, the study was approved by the Danish Data Protection Agency through the University of Copenhagen [514‐0971/24‐3000].

3. Results

A total of 286 blood samples were collected from 11 participants, with up to 27 samples collected from each individual. For one participant, blood sampling was discontinued 80 min after cessation of N₂O administration due to technical issues with the cannula. For two other participants, sampling was terminated after 240 min due to mild discomfort at the cannula insertion site.

The mean maximum concentration was 17 mL/L, with individual concentrations ranging from 6.4 to 50 mL/L in the first sample taken after end of administration (Table 1). N₂O remained detectable above the LLOQ, 0.20 mL/L, for an average of 62 min (range 31–90 min) and above the LOD, 0.05 mL/L, for 132 min (range 70–180 min). At the 60‐min sampling point, the mean concentration was 0.26 mL/L.

TABLE 1.

Maximum concentrations and detection times of N₂O in blood and calculated half‐lives from two and three phase pharmacokinetic modelling for each participant.

Participant	Maximum concentration (mL/L)	Detection time at LLOQ (0.2 mL/L) (min)	Detection time at LOD (0.05 mL/L) (min)	Preferred model	Two phase decay half‐lives (min)		Three phase decay half‐lives (min)
					First	Second	First	Second	Third
A	15.1	40	90	Three phase	1.85	24.7	0.67	3.70	31.1
B	12.7	60	150	Two phase	2.61	31.3	0.04	3.04	32.1
C	19.5	51	120	Two phase	1.82	26.1	0.07	1.94	26.5
D	8.5	61	120	Two phase	2.65	35.6	1.47	4.07	38.6
E	20.2	91	180	Three phase	2.86	34.4	1.15	7.83	41.9
F	20.6	90	120	Two phase	2.22	25.4	0.06	2.51	25.9
G	49.7	81	150	Two phase	1.57	31.6	a	1.57	31.6
H	20.7	60	120	Two phase	1.97	25.5	a	1.97	25.5
I	6.4	41	150	Three phase	3.69	39.2	2.19	14.0	94.9
J	7.4	31	70	Three phase	1.79	15.7	0.23	2.62	21.3
K	8.2	71	180	Two phase	3.40	46.9	a	3.40	46.9
Mean	17.2	62	132		2.40	30.6	0.73	4.24	37.8

^a No stable value was identified.

The individual time‐concentration profiles showed an initial decline in N₂O concentration immediately after cessation of N₂O inhalation, followed by a markedly slower decline occurring 10–20 min later (Figure 1). The deviation from a straight line in the semi‐logarithmic plot indicates that the elimination follows a multi‐compartmental model, with elimination from the central compartment being faster than elimination from one or more peripheral compartments.

FIGURE 1.

Semi‐logarithmic time‐concentration profiles of N₂O in blood following a controlled 10‐min inhalation in 11 participants (A–K). Individual measurements are shown as green dots, and the best fitted curves are represented by grey lines. Lower limit of quantification (LLOQ) at 0.20 mL/L is indicated by the dashed line.

Time‐concentration profiles were modeled using nonlinear least‐square regression and fitted to two‐ and three‐compartment models. Comparison between models was performed using F‐test. Both models provided adequate fits; for four participants the elimination profiles were best described by the three‐compartmental model, while for the remaining seven participants elimination profiles were better described by the two‐compartmental model. For six participants, convergence of pharmacokinetic parameters related to the initial concentration and slope was not achieved with the three‐compartment model. Calculated half‐lives for both models are given in Table 1, and additional model parameters are provided in Table S1.

For the two‐compartment models, a mean half‐life of 2.4 min (range 1.6–3.7 min) for the initial phase and 30 min (range 16–47 min) for the second phase were calculated. For the three‐compartment models, a mean half‐life of 0.73 min (range 0.043–2.2 min) for the initial phase and 4.25 min (range 1.6–14 min), and 38 min (range 21–95 min) for the two slower phases were identified.

4. Discussion

The results obtained in this study provide useful insight into the expected timeframe and blood concentrations of N₂O following a controlled 10‐min inhalation of a 50% N₂O mixture. Furthermore, the elimination profile was found to be more complex than the previously proposed single 5‐min half‐life [1]. N₂O was detectable in blood for 1–3 h depending on individual and cutoff.

The average maximum measured concentration of N₂O was 17 mL/L (range 6.4–50 mL). This concentration range is comparable to levels detected in whole blood samples from traffic‐related cases in Denmark [14], but lower than those measured in previous clinical studies involving N₂O. An average venous blood concentration of 89 mL/L has been reported in adult patients after inhalation of a 39% N₂O mixture for 10 min, whereas arterial blood concentrations in the range 170–220 mL/L have been measured during surgical anesthesia [19, 20].

The two‐compartment elimination model identified in this study is consistent with models reported in exhaled breath studies [9, 10]. Our findings align with those of Jimenéz et al., who observed a slow redistribution phase with a half‐life of approximately 30 min, commencing 20 min after administration [10], and O'Reilly et al., who reported exhalation‐phase half‐lives of approximately 1.5–2 min [9]. Additionally, O'Reilly et al. reported a steep decline in alveolar N₂O concentration within the first 30 s after inhalation cessation, consistent with the sub‐1‐min half‐life estimated in our three‐compartment model [9]. However, more frequent sampling is required to obtain reliable pharmacokinetic parameters for a possible very rapid, initial elimination. The simpler two‐compartment model is appropriate and adequate to explain the measured N₂O concentrations in the blood samples.

The study protocol involved continuous inhalation of a 50% N₂O mixture, which limits generalizability to recreational use, where bolus dosing with intermittent breathing from balloons is more common [21]. Bolus dosing might alter maximum concentrations and half‐lives, but due to safety concerns, experiments with 100% N₂O were not possible. Repeated or prolonged inhalation of pure nitrous oxide is likely to give higher maximum concentrations. The use of a demand‐valve hinders accurate calculation of the inhaled dose of N₂O, as respiratory parameters can vary between participants, likely contributing to intra‐individual differences. The study population consisted of young, healthy individuals, reflecting the demographic most commonly associated with recreational use [13]. Although a previous study has not identified age‐related differences in N₂O concentrations or elimination half‐life [9], further investigations should include participants with a broader age range and varying health status.

5. Conclusion

After 10 min of controlled inhalation of 50% N₂O by 11 participants, N₂O remained detectable in blood for an average of 62 and 132 min at cutoff concentrations of 0.2 and 0.05 mL/L, respectively. Immediately after the end of exposure, blood concentrations ranged from 6.4 to 50 mL/L and then declined rapidly. The elimination profile was best described by a two‐compartment model with mean half‐lives of 2.4 and 31 min. These findings provide important insights into blood concentrations of inhaled N₂O, demonstrating that it can remain detectable in blood well after intoxication has subsided, and supports its interpretation in forensic toxicology, including investigations of impaired driving.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Flow diagram for inclusion and exclusion of study participants, with the number of participants excluded for the different reasons.

Table S1: Pharmacokinetic model comparison and parameters, with calculated half‐lives in minutes.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.