2.1.

Breath Ethylene in Humans

The relation among ethylene (${\mathrm{C}}_2{\mathrm{H}}_4$), free radicals, and SCZ disease can be explained by the oxidative stress. In a normal healthy human body, the generation of pro-oxidants in the form of reactive oxygen species (ROS) and reactive nitrogen species (RNS) is stored by the antioxidant defense.²⁵ When it gets exposed to psychiatric disorder, adverse physicochemical, environmental, or pathological agents, atmospheric pollutants, radiation, toxic chemicals, or overnutrition, the antioxidant defense is shifted and replaced with pro-oxidants with a role in the initiation of oxidative damage/stress.²⁶

Oxidative stress has been associated with the pathophysiology of SCZ. In contrast with other organs in the body, brain tissues exhibit high vulnerability to oxidative stress because of their high oxygen consumption, high content of polyunsaturated fatty acids (PUFA), and low level of antioxidant defenses in addition to a high metal content, which can catalyze the formation of ROS/RNS. Under physiological conditions, the potential for free radical mediated damage is counteracted by the antioxidant defense system, which is composed of a series of enzymatic and nonenzymatic components. The critical antioxidant enzymes include superoxide dismutase, catalase, and glutathione peroxidase. In SCZ, the antioxidant defense is considered to be weak and oxidative stress to be present. Superoxide dismutase converts free radicals into hydrogen peroxide, which is then decomposed into water and oxygen by catalase, thereby preventing the formation of hydroxy radicals that initiate lipid peroxidation (LP).²⁷

LP is a free radical mediated process and the initiation of a peroxidative sequence is due to the attack by any species, which can abstract a hydrogen atom from a methylene group (${\mathrm{CH}}_2$), together with an electron on the carbon atom (${}^{•}\mathrm{CH}$). The resultant carbon radical is stabilized by molecular rearrangement to produce a conjugated diene to give a lipid peroxyl radical (${\mathrm{LOO}}^{•}$). These radicals can further abstract hydrogen atoms from other lipid molecules to form lipid hydroperoxides (LOOH) and at the same time propagate LP further. The process of LP ends with many products including malondialdehyde, 4-hydroxynonenal, and a variety of hydrocarbons, including pentane, ethane, and ethylene.^25–29

Most previous studies in SCZ have been invasive,^30,31 requiring samples of blood or cerebrospinal fluid or indirect measures of antioxidant enzyme levels have been used. A new way to measure LP noninvasively in humans is to measure free radical damage by analyzing early products of oxidation like exhaled hydrocarbons. Breath analysis is an emerging methodology that, being noninvasive and rapid, is ideally suited to clinical monitoring.

A recent study³² has correlated the systemic oxidative stress with changes in brain metabolism defining ethane (${\mathrm{C}}_2{\mathrm{H}}_6$) as a terminal product of the oxidation of omega-3 PUFA. Ethylene is a product of the LP of linoleic acid and can assess free radical damage.^33,34

Given the correlation of breath ethylene with brain metabolism,^27,31,32 measuring the breath concentration of this compound may represent a useful means to examine oxidative stress in SCZ.

2.2.

Breath Ammonia in Humans

Ammonia is disposed primarily by the formation of urea in the liver but can also be produced by all tissues during the metabolism of different compounds.^35–37 Elevated blood (breath) ammonia causes pathophysiologic changes (hyperammonemia) in the central nervous system.

Hyperammonemia is not a true disease, but it is a sign that specific abnormalities may be present that cause blood ammonia to become elevated.^35,38,39

The kidneys generate ammonia from glutamine (which is then excreted into the urine as ${\mathrm{NH}}_4^{+}$), or from the hydrolysis of glutamine (by intestinal glutaminase).³⁷

Ammonia is also formed from urea and absorbed from the intestine then is removed by the liver (severe impairment of metabolic liver function will produce increased blood ammonia).³⁷

Formation of urea in the liver is quantitatively the most important disposal route for ammonia. Urea travels in the blood from the liver to the kidneys, where it passes into the glomerular filtrate. As small molecules, ammonia can penetrate the blood–lung barrier and appear in exhaled breath.

Higher concentrations of ammonia in the blood can cause ammonia intoxication and cell damage (for example, somnolence, tremors, slurring of speech, Helicobacter pylori infection, and so on).^35–38

Generally speaking, exhaled breath analysis (called breath test) can be represented as follows: production of the biomarker during a particular biochemical reaction or a complex metabolic process; diffusion of biomarker through tissues and input into hematic flow; possible intermediate accumulation (buffering); possible trapping of biomarker by utilization and assimilation systems or natural chemical transformation; transport to the lungs; transmembrane diffusion to the air space of lungs; diffusion of biomarkers and their mixing with inhaled air in the alveolar space of lungs; release of biomarkers in the breathing air; collection of a breath sample; and assessment of the biomarkers in the breath sample.

3.1.

Subjects

SCZ patients were recruited and informed consent was obtained from staff at the C.S.C.C.H.S. Center, Calarasi, Romania (all gave their informed consent to participate in this research, which was approved by the institutional review boards of both institutions). The trial protocol was reviewed by ward consultants at the C.S.C.C.H.S. Center, and patients matched for age, gender, and smoking status.

The diagnosis of SCZ was made based on the criteria of SCZ disorders as evaluated in the Complex Evaluation Service of the C.S.C.C.H.S. Center.

A total of 15 subjects (6 males and 9 females, age range from 20 to 23 years, $\text{mean}\pm \mathrm{SD}$: $21.46\pm 1.45$) who had been previously diagnosed as suffering from SCZ and 19 subjects without any history of psychiatric illness or other diseases and nonsmokers were selected as a control group (15 males and 4 females, age range from 25 to 33, $\mathrm{mean}\pm \mathrm{SD}$: $30.05\pm 1.96$) and included in the study.

The control subjects were non- or ex-smokers, nonalcoholic, nonrenal, nondiabetic, and free from psychiatry disorders, somatic diseases, or brain tumors, and had never been treated with antidepressant or antipsychotic medications.

The SCZ group comprising 15 patients were nonsmokers, nonalcoholic, nonrenal, nondiabetic, and on a range of drug therapies with an antipsychotic and anxiolitic treatment: Levomepromazine.

Some studies^12,39,40 indicate that Levomepromazine may be useful in a small number of patients with severe aggression; the drug appears to be efficacious not only in controlling aggression but also lethargy, stereotypy, irritability, and hyperactivity symptoms.

Prior to the analysis of breath, the subjects were asked to avoid for at least 6 h, before or at any time during the breath sample collection, alcohol and coffee, food or beverages, and to refrain from exercise in the morning. On the day prior to the test, products such as onions, leeks, eggs, and garlic should be avoided.

3.2.

Breath Collection

To collect a clean breath air sample, we used aluminized multipatient collection bags (750 mL aluminum-coated bags), designed to collect multiple samples from patients and hold a sample for a maximum of 6 h. The alveolar breath sampling procedure was performed in accordance with previous studies.^41–44 Briefly, after an approximately normal inspiration, the subject places the mouthpiece in his mouth, forming a tight seal around it with the lips. A normal expiration is then made through the mouth in order to empty the lungs of as much air as required to provide the breath sample. When an adequate sample is collected, the subject stops exhaling and the samples of exhaled gas from the schizophrenic subjects can be transferred into the PA cell. To remove any residual contaminants, all of these bags were thoroughly cleaned by flushing with nitrogen gas (purity 99.9999%) and subsequently evacuated for breath sample collection. All of the collected samples were analyzed within 3 h after sampling over a period of three months.

3.3.

CO₂ LPAS Analyses

The ${\mathrm{CO}}_2$ LPAS used for the gas content measurement and presented in this report is schematically shown in Fig. 1 and is also described in other works.^23–45 In brief, LPAS utilizes a line-tunable ${\mathrm{CO}}_2$ laser and a PA cell, where the gas is analyzed.

Fig. 1

Schematic of the ${\mathrm{CO}}_2$ laser photoacoustic spectroscopy instruments.

The experimental setup consists of a homebuilt, line-tunable and frequency stabilized ${\mathrm{CO}}_2$ laser. This laser, emitting radiation in the 9.2 to $10.8\mu \mathrm{m}$ region on 73 different vibrational-rotational lines, has a maximum power of 6.5 W on the 10P(20) line.^23,24,41

Our laser beam was modulated by a high-quality, low-vibration noise and variable-speed (4 to 4000 Hz) mechanical chopper model DigiRad C-980 or C-995 (30 aperture blade), operated at the appropriate resonant frequency of the cell (564 Hz).

We used a dual-phase, digital lock-in amplifier Stanford Research Systems model SR 830 with the following characteristics: full scale sensitivity, 2 nV to 1 V; input noise, $6\mathrm{nV}(\mathrm{rms})/\surd \mathrm{Hz}$ at 1 kHz; dynamic reserve, $>100\mathrm{dB}$; frequency range, 1 mHz to 102 kHz; time constants, $10\mu \mathrm{s}$ to 30 s, or up to 30,000 s.

The PA cell has a total volume of $\sim 1.0{\mathrm{dm}}^3$, and is made of stainless steel and Teflon to reduce the outgassing problems. The PA cell consists of an acoustic resonator tube, windows, gas inlets and outlets, microphones, and an acoustic filter to suppress the window noise. The PA cell windows are made of ZnSe and positioned at the Brewster angle to their mounts. The resonant conditions are obtained as longitudinal standing waves in an open tube (excited in its first longitudinal mode). To achieve an optimum signal, we chose a long absorption path length of 300 mm and an inner diameter of the pipe of 7 mm. The fundamental longitudinal wave, therefore, has a nominal wavelength of 600 mm and a resonance frequency of 564 Hz.

The two buffer volumes placed near the Brewster windows have a length of 75 mm and a diameter of 57 mm. The inner wall of the stainless steel resonator tube is highly polished. It is centered inside the outer stainless steel tube with Teflon spacers. A massive spacer is positioned at one end to prevent bypassing of gas in the flow system; the other is partially open to avoid the formation of closed volumes. Gas is admitted and exhausted through two ports located near the ends of the resonator tube. The perturbation of the acoustic resonator amplitude by the gas flow noise is thus minimized. The acoustic waves generated in the PA cell are detected by four Knowles electrets miniature microphones (sensitivity $20\mathrm{mV}/\mathrm{Pa}$ each) in series, mounted flush with the wall. They are situated at the loops of the standing wave pattern at an angle of 90 deg to one another. The electrical output from these microphones is summed and the signal is selectively amplified by the lock-in amplifier.^23,24

Comparing with other values reported in the literature [minimum detectable concentration of 3.8⁴⁶ parts per billion by volume (ppbV)], our PA system is one of the most sensitive instruments, having a responsivity of $405\mathrm{cmV}/\mathrm{W}$ and being able to measure a minimum detectable concentration of 0.9 ppbV.

We used a modular software architecture (Keithley TestPoint software) aimed at controlling the experiments, collecting data, and preprocessing information. It helps to automate the process of collecting and processing the experimental results. The software transfers powermeter readings, normalizes data, and automatically stores files. It allows the user to record parameters such as the PA cell responsivity (a constant used to normalize raw data), gas absorption coefficient, number of averaged samples at every measurement point, sample acquisition rate, and the total number of measurement points. This software interfaces the lock-in amplifier, the chopper, the laser powermeter, and the gas flowmeter. It allows the user to set or read input data and instantaneous values for the PA voltage, average laser power after chopper, and trace gas concentration.^23,24

Of great significance in these determinations is the gas handling system due to its role in ensuring gas purity in the PA cell. It can be used to pump out the cell, to introduce the sample gas in the PA cell at a controlled flow rate, and monitor the total and partial pressures of gas mixtures. Also, the gas handling system can perform several functions without necessitating any disconnections.²⁴

${\mathrm{CO}}_2$ LPAS performs well in terms of sensitive and selective detection of trace gas and it allows near on-line measurements.

The calibration measurements (concentration-dependent response) for both ammonia and ethylene (Fig. 2) were experimentally determined using commercially prepared, certified gas mixtures containing 0.96 ppmV ethylene diluted in pure nitrogen and 10 ppmV ammonia diluted in pure nitrogen.^23,24

Fig. 2

The concentration-dependent response for (a) ethylene and (b) ammonia.

For calibration, we examined this reference mixture at a total pressure of $\sim 1013\mathrm{mbar}$ and a temperature of 23°C, using the commonly accepted values: $30.4{\mathrm{cm}}^{-1}{\mathrm{atm}}^{-1}$ (for ethylene) and $57{\mathrm{cm}}^{-1}{\mathrm{atm}}^{-1}$ (for ammonia).

To analyze the gas from the bags, we evacuated the extra gas and then we flushed the system with pure nitrogen at atmospheric pressure for few minutes; then the exhaled air sample can be transferred to the cell using a controlled flow rate.

Because ammonia is a highly adsorbing compound and the results of successive measurements are often altered by the molecules previously adsorbed on the pathway and cell wall, an intensive cycle of ${\mathrm{N}}_2$ washing was performed between samples in order to have a maximum increase of 10% for the background PA signal (to ensure the quality of each measurement). It has to be underlined that the measured PA signal is due mainly to the absorption of ammonia and ethylene, but some traces of ${\mathrm{CO}}_2$, ${\mathrm{H}}_2\mathrm{O}$, ethanol, etc., influenced the measurements (overall contribution is $<10\%$).

The response to all absorbing species at a given laser wavelength (PA signal) decreased considerably when we inserted a KOH trap (with a volume $>100{\mathrm{cm}}^3$), proving that amounts of ${\mathrm{CO}}_2$ and ${\mathrm{H}}_2\mathrm{O}$ vapors in the breath can significantly alter the results, thus making their removal compulsory.⁴⁷

An important parameter in the measurements is the responsivity $R$ ($\mathrm{cmV}/\mathrm{W}$) of the PA cell, which depends on the pressure of the gas inside the cell. Taking into account the fact that the initial pressure in the sample bags filled by the healthy humans and by the subjects with different disorders differs from one case to other, it is necessary to know the pressure dependence of the PA cell responsivity (Fig. 3).

Fig. 3

The responsivity of the photoacoustic cell against the pressure.

The exhaled air sample was transferred to the PA cell at 600 standard cubic centimeters per minute, and the total pressure of the gas in the PA cell was measured, then applying the correction factor for the responsivity according to the calibration curve from Fig. 3.

The responsivity of the PA cell was determined by using a calibrated mixture (Linde Gas) of 0.96 ppmV ($\pm 2\%$) ${\mathrm{C}}_2{\mathrm{H}}_4$ diluted in nitrogen 6.0 (purity 99.9999%) and of 10 ppmV ($\pm 5\%$) ${\mathrm{NH}}_3$ diluted in nitrogen 5.0 (purity 99.999%).²⁴ The pressure dependence of the responsivity was always measured at the center of the ${\mathrm{CO}}_2$ laser line by using a frequency stabilized laser (instability $3\times {10}^{-8}$).

The absorption coefficients of ethylene and ammonia at different ${\mathrm{CO}}_2$ laser wavelengths were precisely measured previously^23,24,48,49 and the ${\mathrm{CO}}_2$ laser was kept tuned at the 10P (14) line ($10.53\mu \mathrm{m}$) where ethylene exhibit a strong peak, corresponding to an absorption coefficient of $30.4{\mathrm{cm}}^{-1}{\mathrm{atm}}^{-1}$ and at 9R(30) ${\mathrm{CO}}_2$ laser line ($9.22\mu \mathrm{m}$), where the ammonia absorption coefficient has the maximum value of $57{\mathrm{cm}}^{-1}{\mathrm{atm}}^{-1}$.

4.1.

Results

In this study, ethylene and ammonia concentrations from breath samples were measured before/after the treatment with Levomepromazine in SCZ patients, and the results were compared with healthy controls using ${\mathrm{CO}}_2$ LPAS.

Figure 4 shows the average concentrations of breath ethylene for SCZ patients, before and 30 min after ingestion of Levomepromazine treatment compared to the ethylene concentrations of a healthy group control.

Fig. 4

Breath ethylene biomarker in 15 schizophrenia (SCZ) patients and 19 age-matching control people.

As an observation of our primary result of interest, we see that the mean ethylene level of SCZ patients is higher (0.07 ppm) compared to the mean ethylene level of healthy subjects (0.008 ppm). In addition, at 30 min after the start of the treatment with Levomepromazine, the mean ethylene level of SCZ patients is smaller (0.066 ppm) than before the treatment (but still high compared with the control subjects).

Using gas chromatography and mass spectrometry, previous studies^30,50 reported an increase in exhaled ethane (like ethylene, ethane is also a hydrocarbon derived from n-3 PUFA) of patients with SCZ (e.g., 5.15 ppb or 8 ppbV) compared with those of the healthy controls (e.g., 2.63 ppb or 2.5 ppbV).

It is important to mention that the SCZ patients from the previous studies^30,50 had not been in receipt of psychotropic medication for three weeks prior to participating in the study but had received medication for the purpose of the study.

So our findings confirm previous determinations that oxidative stress is increased in SCZ and that this is unlikely to be a consequence of antipsychotic medications because the breath biomarkers after the treatment were not significantly increased.

As ethylene is produced as a byproduct of oxidative stress, ammonia is produced as a byproduct of amino acids and protein ingestion.

Figure 5 shows the average concentrations of breath ammonia for SCZ patients, before and 30 min after ingestion of Levomepromazine compared to the ammonia concentrations of a healthy group control.

Fig. 5

Breath ammonia biomarker in 15 SCZ patients and 19 age-matching control people.

It should be pointed out that the mean ammonia level of SCZ patients is higher (2.02 ppm) compared to the mean ammonia level of healthy subjects (0.29 ppm). At 30 min after the start of the treatment with Levomepromazine, the mean ammonia level is even higher (2.2 ppm).

Other possible confounding variables, such as age or sex, showed no statistically significant differences between the two groups.

4.2.

Discussion

Oxidative stress seems to be a key piece in the SCZ pathophysiology. When oxidants exceed the antioxidant defense, biological systems suffer oxidative stress with damage to biomolecules and functional impairment.

The possible responsible factor for the differences between the concentrations of breath ethylene before and after the treatment with Levomepromazine could be explained by the difference between untreated and treated SCZ patients. Most invasive measurements of oxidative stress in patients with SCZ have been made on peripheral tissues.^51–56 There is a lack of information on oxidative processes in cerebrospinal fluid and brain. It must be mentioned that traces of oxidative damage may originate from various sources in the body, and consequently, such a peripheral indicator may not necessarily reflect the conditions of the oxidative stress parameters in the brain.⁵⁵

Our measurements are based on the detection of biomarkers from breath and are in good agreement with those (based on oxidative stress analysis) reported in the literature.^57–66

While the majority of invasive studies have reported decreased antioxidant defense in patients with SCZ, there are also some studies where the opposite has been reported.^67–74

Several factors, such as the differences in measuring techniques, differences in material tested, exposure to antipsychotic treatment, sampling of patients at different stages of the disease, lifestyle, and dietary patterns, may be responsible for this discrepancy.

Our study also reviewed the efficacy of Levomepromazine in patients with SCZ, and the findings indicate that breath ethylene decreases after the treatment and breath ammonia increases after the treatment (but not significantly). So, while the oxidative stress is mildly reduced after the treatment, a mild impairment of metabolic liver function will produce increased blood (breath) ammonia.

Taking into consideration that the Levomepromazine is achieved in 2 to 3 h depending on the route of administration,⁷⁵ at 30 min after the administration, there is no significant chance in the chemical levels from breath of patients. The physiological basis of these findings is still speculative and future studies are needed that would clearly identify the etiologic relation between breath biomarkers and treatment with Levomepromazine.

The relation between level of ammonia in the exhaled breath and SCZ could be explained by the treatment with Levomepromazine that can lead to a deficiency of amino acids which are required to detoxify toxins in the liver.⁷⁶ Along with their useful effects, most medicines can cause unwanted side effects, although not everyone experiences them.

Levomepromazine at SCZ patients, seems that, mildly reduced kidney function resulting an insufficient detoxification pathways with a very small accumulation of ammonia in the breath.⁷⁷

The most important route for ammonia is the formation of urea in the liver; then the urea is transported to the blood from the liver to the kidneys and lastly appear in the exhaled breath of SCZ patients.

From the results of this study, the ammonia breath of SCZ patients were identified in higher concentrations (at treated patients) when compared to the healthy group.

Our data support a dysregulation of energy metabolism in SCZ and suggests new markers that may contribute to a better understanding of this disease. Both the feasibility and the importance of monitoring exhaled ammonia and exhaled ethylene from different subjects have been shown.