Nitrous oxide abuse direct measurement for diagnosis and follow-up: update on kinetics and impact on metabolic pathways

Gas chromatography-mass spectrometry

Most N₂O analytical methods rely on gas chromatographic techniques. Environmental studies have shown that gas chromatography-isotope ratio mass spectrometry [26] and gas chromatography-mass spectrometry (GC-MS) [27], [28], [29] are used. These mass spectrometry-based approaches have also been used in other fields of study, such as investigation of biological pathways [30, 31]. GC coupled with a thermal conductivity detector (TC) and a micro-ionization cross-section detector (MICS) are alternative methods for measuring N₂O [32]. Methodological shortcomings include limited sensitivity and the use of internal standards that are not optimally selective at the molecular level for N₂O measurement. Due to the lack of acceptable internal standards, external calibration is required, which is not the best strategy for gas analysis given the potential risk of leaks during sampling, extraction, and analysis [33, 34]. Even though previous published GC-MS techniques have improved sensitivity, the use of pre-loaded hermetic gas syringes should alleviate any quantification difficulties caused by probable leakage.

A variant of GC-MS called Headspace-GC-MS (HS-GC-MS) is a method also used to measure N₂O. The sample is placed in a hermetically sealed container for HS-GCMS analysis. To create a headspace, the sample is heated to boiling point. The GC injection needle draws the gaseous substance into the headspace and injects it into the chromatographic column. In 2009, Poli et al. used HS-GC-MS to measure N₂O in numerous tissues on patients who had been deceased for several days [35]. During autopsy, blood, liver, bile, kidney, fat, and brain samples were collected; a urine sample was also taken from the eighth deceased patient [35]. All of the tissue samples were immediately weighed and then transferred to 10-mL headspace (HS) vials with airtight plugs. The gaseous samples from the sample bags and syringe were also transferred to HS vials, which were completely filled [35]. In, 2015, Giuliani et al. suggest a new HS-GC-MS method for the quantification of N₂O by using hydrogen sulfide as internal standard [36]. Although the authors validated this method, no independent study incorporated HS-GC-MS as their method, and its implementation in medical laboratories appears challenging due to its preparation difficulty and lack of reproducibility. Hence, use of HS-GC-MS for N₂O measurement could be promising but requires further technical development and clinical validation studies for use in laboratory medicine. A few medical laboratories already employ HS-GC-MS for measurement of alcohols and glycols, so these laboratories that have the equipment and expertise could certainly investigate measuring and optimizing N₂O by HS-GC-MS without significant cost [37].

Infrared spectroscopy

Infrared spectroscopy (IR), based on the molecule’s ability to absorb light at a specific wavelength, is another technique for measuring N₂O. As described by Heusler et al., IR techniques are ideal to measure in air, but not in biological fluids such as blood or urine, where GC-MS is the preferred choice [38]. Closed path methods are preferred with greater sensitivity. The advantage of open path systems is that the concentration can be measured continuously, which can be useful when using N₂O for anesthesia to control flow rate or to prevent occupational exposure [39], [40], [41]. However, interest in using IR technology for measuring direct unique or recreational N₂O consumption is more limited [42].

Fourier transform infrared spectrometers (FTIR) are well described. This technique uses broadband IR radiation covering the entire IR spectrum, enabling many gases to be analysed at the same time with high precision, which explains why this technique is so popular for atmospheric measurements [25, 43, 44]. Another IR technique that can be used to measure N₂O is infrared laser-absorption spectroscopy [45]. In contrast to FTIR spectroscopy, which measures the complete spectrum, laser spectroscopy uses a single wavelength to boost sensitivity. Specificity can be enhanced by working at low pressure, which results in narrower lines [42].

We have not found data on the use of IR methods for the determination of N₂O in biological fluids other than exhaled air.

Consumption-dependent quantity

There are two types of N₂O exposure: therapeutic and recreational. In cases of analgesia without loss of consciousness, an equimolar mixture of oxygen and N₂O (EMONO) is utilized or diluted in oxygen at concentrations of 50–70 % [46]. The anesthesiologist adapts the flow rate to the patient ventilatory rate, but this practice is waning.

The average flow rate recommended according to an SFAR study (French Society of Anesthesiologists and Reanimators) was 1 L/min or less. Thus, patient receives N₂O at a rate of about 500 mL/min [47], or at 0.9 g/min of N₂O since the gaseous form has a density of 0.0018 g/mL (Figure 2). EMONO is still frequently used in dental surgery, particularly on young patients. The majority of treatments run no longer than 40 min, and the N₂O concentration used ranges from 20 to 50 %, depending on the complexity of the intervention [48], [49], [50].

Figure 2:

Summary of global toxicokinetic of nitrous oxide. Nitrous oxide is absorbed by the respiratory tract alone. It is then distributed to all tissues. Due to its low solubility, the body is rapidly saturated. It is not metabolized, and is rapidly eliminated in unchanged form, mainly in exhaled air. A fraction is eliminated via the urine, and a small proportion via the transdermal route.

The levels of N₂O used for recreational purposes are difficult to quantify since they differ depending on the consumer and the mode of consumption (Figure 1). A canister holds 8 g of N₂O, while a cylinder holds 640 g. Several sources claim that casual N₂O users only consume a few dozen canisters per occasion (equal to about 100 g of N₂O). However, recreational use of N₂O is on the rise. Many people now use N₂O in cylinders rather than canisters, with some people consuming numerous cylinders per day (Figure 1). Some claim to have consumed up to 25 cylinders per day (equal to 16 kg of N₂O) (Figure 2). As a result, recreational exposures can be substantially higher than medical doses.

Uptake

The absorption of a gas is the result of the balance between the concentration inhaled and concentration in the alveoli, blood and tissues. Because N₂O has relatively low solubility in blood (and other tissues), it equilibrates very quickly through tissue saturation. Due to its poor solubility in blood and tissues, the increase in alveolar concentration upon exposure is very quick, resulting in rapid saturation of the body. After 5 min of treatment, the alveolar fraction of N₂O equals 90 % of the fraction inhaled [51]. The rate of increase in alveolar concentration is influenced by inhaled concentration. The greater the inhaled concentration, the faster the alveolar concentration rises [52]. Uptake is altered in cases of pulmonary disease such as chronic obstructive pulmonary disease (COPD) [53]. This has been demonstrated in patients undergoing surgery and receiving N₂O. They were classified into three groups according to the severity of their respiratory impairment, on the basis of the Tiffeneau ratio. When comparing patients with COPD to healthy patients, concentration measurements in inspired and exhaled air revealed significative differences, especially with regard to equilibration time [53].

Partition coefficients

Because studying human tissue is highly challenging, extrapolations are frequently made from findings made during pre-clinical studies on animals or in vitro models. Tissue/gas distribution coefficients of N₂O and cyclopropane were determined in rabbits and compared to the distribution coefficients of these agents in homogenates of identical in vitro tissues [54]. Rabbits anesthetized by parenteral injection of sodium pentobarbitone were used and received N₂O mixed with oxygen (70/30). In order to maintain a constant arterial partial pressure, exhaled gas samples were evaluated by GC during the experiment, and the inhaled concentration was adjusted accordingly. Blood, brain, heart, kidney, liver and skeletal muscle were extracted. Equilibrating homogenates of the same tissues (unused in vivo samples where any remained) with a known partial pressure of N₂O yielded the in vitro coefficients. A discrepancy of 6 % was detected between than in vitro, higher, and in vivo distribution coefficients. For the six animals used, a three-way cross classification analysis of variance supported the fact that this difference is statistically significant, but the authors pointed out that this difference, although real, still allows us to have confidence in the in vitro distribution coefficients in order to extrapolate to humans.

Kety et al. conducted in vitro and in vivo experiments in dogs to evaluate the solubility of N₂O in whole blood and brain as well as the partition coefficient between the two. Human and dogs appear to have equivalent values [55]. In human whole blood, they found a solubility of 0.412. This solubility explains how quickly the alveolar concentration rises to the inspired concentration. The solubility in human entire brain is 0.437, giving a brain/blood partition coefficient of 1.06 (Figure 2).

As a result of lower tissue/blood partition coefficients, the equilibration of N₂0 in most tissues occurs rapidly.

Blood distribution

N₂O does not bind to any protein and circulates freely in the blood. Anesthetists, surgeons and nurses were subjected to an occupational exposure study to determine their blood levels of N₂O. For all processes, the N₂O flow rate was standardized at 6 L/min. Except for surgeons, from whom collections were taken at the end of the operation, blood samples (blood gas syringes) were collected every hour. Simultaneously, an air sample was taken from the subject’s breathing zone for 60 s. A correlation was showed between blood and breathing zone atmosphere concentrations (n=116) but for some samples (32 %, n=37), blood concentration could not be accurately matched to atmospheric concentration [56]. Nonetheless, this study holds additional significance because they checked the N₂O blood levels of subjects before exposure, which was undetectable in any of the groups. It leads to the conclusion that the N₂O blood level following exposure and its correlation to the atmospheric concentration are solely due to current exposure. A quantitative study comparing variance in blood and atmospheric concentrations revealed more stable blood concentrations. This shows that this is the result of blood N₂O saturation rather than a single exposure [56].

In subjects receiving N₂O directly for dental analgesia, N₂O was detected in the blood after 10 min of inhalation at concentrations 2 times lower than those measured in the nasal mask and up to 5 times lower than the concentration set on the machine [57]. After 5 min of 100 % oxygen inhalation, venous concentration divided by 2.

In 2010, after several patients’ death from hospital poisoning with N₂O, a team of researchers showed that N₂O was still present in their blood one month later [35]. These patients had mistakenly received N₂O instead of oxygen for periods ranging from 25 to 125 min. During the autopsy (one month after death), various samples were taken, including blood, and N₂O concentration measured by HS-GC-MS. Large quantities of N₂O were still measurable 1 month after death (11.29–2,152.04 mg/L). However, measurements taken so long after death are questionable since the tissues could have released N₂O stored in the blood due to post-mortem autolysis and falsely elevated the measurements.

Involvement for laboratory medicine

Data on N₂O blood distribution are lacking in direct consumers of recreational dose(s). Recreationally absorbed doses are not equivalent to those absorbed by professionals exposed in hospitals. Chronic consumption of large doses has not been studied. The absence of detection in professionals just prior to experimental exposure suggests that, with regular exposure to low environmental doses, N₂O is rapidly distributed and eliminated. Measuring it in the context of a diagnosis would therefore appear to be uninformative, due to the possibility of false negatives related to rapid disappearance from the blood. Again, these data are related to occupational exposure and the findings could be different for recreational consumption (Figure 2).

Metabolic transformations

The CNESST (Commission on standards, equity, health and safety at work) sheet on N₂O states that it is poorly metabolized in the liver, as the liver extracts only a tiny fraction of quantity received (Figure 2).

Because of its redox characteristics, N₂O can be reduced by a variety of substances found in the body. This was demonstrated in 1952 when bacteria were exposed to an anaerobic, N₂O-containing environment [58]. Pressure changes have been reported as a result of the transition of N₂O into N₂. The enzymes involved are adaptable and bacteria that have evolved to nitrates and nitrites have also adapted to N₂O. As a result, N₂O is a common intermediate in denitrification. In 1985, another experiment studied the denitrification by bacteria. The catalyzed equation is: N₂O + 2e⁻ + 2H⁺ → N₂ + H₂O [59].

Because bacteria may change N₂O, a small amount of N₂O can be digested by the microbiota. Rats (male and female) were sacrificed to recover a portion of gut and large intestine. The intestinal contents and wall were separated, and two of them received antibiotics (tetracycline, neomycin and bacitracin) for three days prior to sacrifice. Human intestinal contents were also studied at the same time by collecting faeces samples from one of the researchers. To avoid contamination from the intestinal bolus, the wall was cleaned. After diluting, samples were incubated with radiolabeled pure N₂O or a mixture of N₂O/O₂ at various doses. The intestinal wall was incubated for 3 h and the contents for 16 h at 37 °C. The radiolabeled N₂ generated by the reaction was analysed by mass spectrometry and quantified using an internal N₂ standard. These findings demonstrated that human and rat intestinal contents metabolize N₂O, as labelled N₂ concentrations were measured in incubated samples [60]. At concentration of 10 and 20 %, and to a lesser extent at 5 %, oxygen inhibits this process. The intestinal wall samples had a very low quantity of N₂, supporting the fact that metabolism is carried out by the intestinal microbiota [60].

Pulmonary elimination

As described in the absorption/distribution section, elimination of gases depends on solubility and partition coefficient. The blood/gas and tissue/gas coefficients reflect how readily some tissues can store xenobiotics. On the other hand, N₂O is eliminated extremely rapidly. When N₂O administration is stopped, the significant difference in concentration between the alveoli and the blood results in a rapid decrease in N₂O concentration. Salanitre et al. demonstrated in 1962 that the uptake and elimination curves are comparable [61]. The aim of their study was to validate or refute the equivalence between N₂O uptake and excretion. N₂O was administered to five subjects at subanesthetic doses (10 % N₂O/90 % O₂), and the amount absorbed and eliminated was measured. A good correlation between N₂O uptake and excretion was found, although for some patient curves not (but it could be attributable to the method) [61]. The excretion curves data were compared to a previous study with anesthetized patients. The only difference was an upward displacement of the curve in the anesthetized group, which might be explained by anatomic and physiological differences between the two groups, as well as a higher saturation in N₂O in the anesthetized group.

The concept of diffusion hypoxia or diffusion anoxia should not be underestimated. The diffusion hypoxia is caused by the fact that N₂O returns to the alveoli faster than nitrogen leaves them leading to a dilution of the oxygen concentration in the alveoli. To simulate gas exchange in the pulmonary alveolus, an in vitro experiment was performed. Because the solubility of N₂O in blood is similar to that of water, water could represent blood. Air represents the pulmonary alveolus. A flask containing 50 % water and 50 % air is saturated with N₂O. N₂O is 35 times more soluble in water than nitrogen, and 20 times more soluble than oxygen. At equilibrium, N₂O partial pressures in water and air are equivalent. When the flask is brought into contact with air, the equilibrium of partial pressures is disturbed. The partial pressure in air is lower than that in water. As a result, N₂O diffuses from the water into the air, diluting the oxygen and nitrogen in the air [62]. This explains the hypoxia that may result from N₂O inhalation during the reoxygenation period.

In 1990, an experimental investigation was conducted to evaluate biological exposure to N₂O in occupational exposed subjects (surgeons, anesthetists and nurses). Exhaled air and urine were obtained from 20 participants, working in five operating rooms, as well as samples of the ambient air in the operating room. A correlation (r=0.911) was found between concentrations in air and exhaled air (r=0.911) [63]. These subjects were not directly exposed to N₂O. Einarsson et al., investigated the lung clearance rate of N₂O in surgical patients in 1993 [64]. Thirty-six subjects were divided into six groups based on the duration of their N₂O exposure (30, 60 and 120 min) and the type of ventilation after this exposure (normo- or hypoventilation). End-tidal N₂O concentrations were evaluated using infrared spectroscopy and inhaled concentrations of 30 % oxygen and 70 % N₂O. During normal ventilation, the concentration of N₂O dropped rapidly after the end of the exposure, from 66–70 % to 6–9 % at 5 min and 2–4 % at 30 min [64] (Table 1) (Figure 2). It decreased a bit more slowly during hypoventilation (16–23 % at 5 min and 5–9 % at 30 min).

Furthermore, there was no effect of exposure duration on the rate of removal, indicating that the organism had reached maximum saturation after 30 min. There is no evidence of difference in the decline in N₂O rate according to age [65]. After 30 min of N₂O inhalation, the levels and rates of exhalation are not significantly different between elderly males (63–69 yo) and young adult males (22–30 yo). The rate of elimination therefore appears to be similar in all subjects, independent of age.

Urinary excretion

Numerous studies have evaluated urinary N₂O content following exposure, particularly in occupationally-exposed patients, because urine is an easy-to-recover matrix in medical staff. In 1995, urine and blood samples of patients working in nine hospital operating rooms were collected at the end of a professional exposure session. Personal passive samplers were used to monitor atmospheric N₂O in the operating room (n=80) and then compared to urine and blood levels in relevant participants (Table 1). A control group of urine (n=25) and blood (n=27) morning samples of blood donors were used to assess exposure due to atmospheric N₂O levels. This data revealed a significant correlation between both urinary and blood levels with environmental N₂O concentrations [66]. Finally, when compared to blood donors, the occupationally exposed group had significantly higher values. Particularly high and unexpected values were found in some patients (n=10). In this study, those extreme values were associated to participants who had a bacterial urine infection, even though they were asymptomatic. Another study found that urinary tract infections had an effect on N₂O levels in urine [67]. To study the kinetics of N₂O production, a healthy donor’s urine was inoculated with Enterobacter aerogenes. Six vials were infected and incubated at 37 °C. The level of N₂O was measured at 0, 2, 5, 8, 24 and 48 h after inoculation. Gram negative bacilli produce high amounts of N₂O (78.5–464.7 μg/L) compared to others bacteria (0.1–4.2 μg/L) [67]. The production is time dependent and follows exponential kinetics (it is low for the first 5 h, increases exponentially up to 24–36 h, and the equilibration is reaching at 48 h) [67]. N₂O concentrations measured in urine infected with Gram-positive cocci and yeasts were not significantly different from the control sample. To study the correlation between N₂O concentration and the different bacterial strain, 16 urine samples from patients were collected. After acidification and storage of urine at 4 °C, N₂O concentration was measured within 24 h by headspace GC. There was a strong correlation between the bacteria involved and the level of N₂O in urine (high for Gram negative bacteria, lower for Gram positive cocci and almost non-existent for sterile samples). Because N₂O is most likely produced in the bladder, four urine samples were analysed immediately after collection and varying quantities of N₂O were detected, which could be linked to the time of infection onset [67]. This means that the amount of N₂O measured in the urine may be greater the longer the infection has been present. The levels produced by bacteria in cases of urinary tract infection are well above those found in cases of occupational exposure to N₂O. The average urinary N₂O concentration in subjects working in the operating theatre was 26 μg/L [66]. As a comparison, in 2018, N₂O level was measured in urine of a cohort of dentists and dental assistants exposed to N₂O. Concentrations ranged from 2.54 to 25.4 μg/L [68]. It is therefore difficult to differentiate between exposure and urinary tract infection when N₂O is detected in urine (Figure 2).

In 1983, volunteers were exposed to N₂O (850 ± 25 ppm) in a 20 m³ chamber [69]. The exposure was intermittent, lasting 5–15 min and totaling 20–60 min. The last exposure was followed by a 30 min in N₂O-free air. Participants were equipped with in a gastight container. At the end of the experiment, urine was discharged into a syringe and was equilibrated with an equivalent volume of N₂O-free air. The syringe was shaken in a water bath at 38 °C for 20 min to obtain equilibration. N₂O concentration was measured by infrared spectrophotometer and GC. A strong correlation was discovered between the bag and urine concentrations, suggesting that 90 % of urinary N₂O is caused by environmental exposure [69]. As a result, the urine matrix may be a useful indicator of N₂O environmental exposure. Another cohort confirmed these results in 1985. The subjects were outfitted with pump-bag sampling to monitor the N₂O content, which they wore all day long (98 days of work in total). N₂O was analysed by IR or GC. They collected urine before starting anesthetic work and at the end (Table 1). They also investigated the effects of ventilation, close scavenging and types of anesthetics (mask, intubation, epidural) in several departments (pediatric, gynecological, general, orthopaedical and thoracic surgery). There was a positive correlation between N₂O in bags and urine headspace N₂O, this correlation being influenced by intra-individual factors and handling of urine specimens [70]. Duplicate samples were stored in a refrigerator or at room temperature and analysed up 24 h after (correlation coefficient between pump bag concentration and urine headspace gas measurement=0.94) [70]. The investigators concluded that urine specimens can be used in a practical way.

Similar studies on occupational exposure have been carried out, in several countries [68, 71]. These studies showed that a quantifiable concentration of N₂O was found in human urine after exposure, but the kinetics were not studied. There is no data on the time lapse after which it can be recovered. However, a correlation was found by Trevisan et al. between the concentration in exhaled air just after a 6-h exposure and the urinary concentration 2–3 min after the end of the exposure [63]. As previously mentioned, the concentration drops in exhaled air 5 min after exposure [64]. It may be hypothesized that urinary concentration falls in the same way after exposure ceases. This is consistent with an occupational exposure study which showed that urinary N₂O concentrations decrease highly after exposure ends. Urine and blood samples were collected the next morning from 25 exposed participants, approximately 18 h after the exposure ended. The mean N₂O level found was 1.52 μg/L compared to 40 μg/L at the end of operating session (Table 1). Moreover, this concentration so long after exposure is no different from the concentration of 1 μg/L found in control subjects (blood donors) [66].

However, these data are not derived from patients who were directly exposed, so they should be interpreted with caution. Furthermore, urine matrix is frequently difficult to obtain, and exposed patients may not want to provide. Moreover, N₂O doses differ from occupational to abuse. Hence, urine may be a complicated matrix for laboratory medicine and it may be more interesting to focus on blood or exhaled air.

Transdermal elimination

Cullen et al. used an in vitro test to demonstrate how N₂O diffused through the skin [72]. Healthy skin samples were recovered from amputated limbs. Each specimen (n=7) was immersed in a water bath in a diffusion chamber (37 °C or/and 24°). A steady flow of humidified N₂O (85.1 %) was applied at a constant rate of 1 L per minute. The gas spread through the skin to the other side of the nitrogen-filled glass box. Gas samples were analysed by GC. The N₂O rate diffusion through skin at 37 °C was 10.38 mL/min/m² with a mean difference of 2.47 mL/min/m² between 37 and 24 °C.

In 1968, the percutaneous loss in surgical patients anesthetized with N₂O was measured. Patients were subjected to a flow of 80 % N₂O, with 20 % O₂. A gas sample was collected at the start and then every 10 min and analysed by GC. The N₂O percutaneous loss was measured at 3.6 mL/m²/min or 214 mL/m²/h in an average sized adult after 100 min of anesthesia (plateau between 60 and 100 min) [73]. This is much lower than the rates found in vitro [72]. This may be due to the fact that the amount of N₂O reaching the skin in patients does not represent the total intake, and that it diffuses and is distributed elsewhere in the body. To describe the influence of temperature on this process, ice and heating patches were applied and the desired temperature was maintained for 30 min. There was a linear relationship between temperature (20–40 °C) and skin losses. The increase of skin temperature opening capillaries nearer the skin surface which decrease the path length for diffusion who increasing percutaneous loss. The conclusion was drawn that diffusion is the most important limiting factor for percutaneous loss [73].

Maternal milk

There is no information available about excretion in breast milk. According to Hale et al., N₂O is unlikely to be expelled in breast milk or ingested by newborns (albeit no experiments have been performed) [74].

Involvement for laboratory medicine

In the context of driving under the influence of N₂O and because N₂O elimination is primarily pulmonary, it may be useful to detect it with a technique based on a breathalyzer test kit. Elimination is however too rapid, considering that 5 min after the last inhalation, it has nearly entirely disappeared. Regarding the use of urine samples, in addition to the fact that patients are not always compliant for this type of sampling, data is lacking from patients who are direct consumers. Urine measurement seems possible even a few days after inhalation, but in addition to the above limitations, predicting pathological cut-off is complicated and there is a lack of studies in subjects chronically consuming large quantities of N₂O. Also, patients with urinary tract infections may have false positive results.

As direct detection of N₂O is complicated to implement routinely due to the specific methods and difficult to interpret due to the lack of kinetics data, it is necessary to look at the effects of N₂O on metabolism in order to find indirect markers.

Involvement for laboratory medicine

Consuming N₂O has metabolic side effects that are starting to be well understood, at least for the cytoplasmic pathway. It is still not well understood how N₂O enters the mitochondria, or at least how it affects this metabolic pathway. Vitamin B12, homocysteine and MMA assays are already available in medical laboratories. The techniques used are those traditionally employed, namely immunochemistry, liquid chromatography and mass spectrometry [106], [107], [108], [109], [110]. The combined use of these markers would appear to be relevant in assessing the metabolic damage attributed to N₂O consumption [104]. Mass spectrometry panels for homocysteine, methionine and MMA may be used to provide a complete interpretation of the patient’s profile.