Xenon in the treatment of panic disorder: an open label study

Background

One of the most common anxiety disorders is panic disorder (PD), with a 12 month prevalence in the US and in Europe estimated at 1.8 and 2.7% of the population, respectively [1, 2]; the main clinical feature of which is an unexpected panic attack (PA) that arises in the absence of any situational or emotional triggers, reaching its peak intensity within minutes, that is manifested by intense physical and cognitive symptoms, such as fear of recurrence, general health concerns, and behavioral changes [3, 4]. In addition to spontaneous PA, its other forms include situationally predisposed PA, “symptomatically mild” (“minor”) PA [5], in which less than 4 out of 13 symptoms listed in the DSM-IV are present, and “nocturnal” panic attacks that occur during phase 2 of the sleep cycle [6].

Panic disorder in its “pure” form is found only in 24.6% of cases, in 36.7% of cases it is accompanied by a comorbid disorder, in 13.3% —by 2, and in 23.5%—by 3 or more mental disorders, mainly anxiety and diseases of depressive spectrum [7]. Lecrubier et al. had shown that individuals with isolated panic attacks are more prone to the development of depression (45.6%) than to development of a panic disorder [8].

To date, the greatest clinical evidence of efficacy in the treatment of PD has been demonstrated with the use of selective serotonin reuptake inhibitors (SSRIs), serotonin–norepinephrine reuptake inhibitors (SNRIs) (drugs of first choice) and benzodiazepine tranquilizers (drugs of second choice) [9]. Disadvantages of SSRIs/SNRIs therapy include delayed onset of therapeutic effect (2–6 weeks), and side effects at the start of therapy, which can limit its use in the treatment of PD, especially given the importance of achievement of rapid anxiolytic effect [10, 11]. While benzodiazepines have immediate onset of action, their side effect profile is significant and includes excessive sedation, slow reaction time, dizziness, and possible paradoxical reactions such as anxiety [12]. In addition, knowledge of benzodiazepines’ high risk of dependence [13, 14], often forces patients with PD leading an active lifestyle, to seek other, alternative methods of treatment. These side effects in themselves can exacerbate PD and become the triggers of panic attacks. Historically, 18% of patients receiving SSRIs and 15% of patients receiving benzodiazepines drop-out from clinical studies [8]. To our knowledge, the contemporary scientific literature contains practically no data on the treatment of refractory PD, where both first- and second-line treatment options are ineffective. Additionally, there are frequent clinical scenarios where PD exists concomitantly with other psychiatric comorbidities or where panic attacks don’t reach the diagnostic threshold of a panic disorder, but nevertheless have a significant impact on the course of the underlying disease and impair social functioning. Thus, the search for an alternative treatment modality with non-habit forming anxiolytic effect and minimal side effects aimed at rapid relief of panic attacks represents an urgent unmet need in the treatment of PD.

Anxiety symptoms can occur in a variety of mental and substance abuse disorders. In particular, “neurovegetative” (poor sleep, sweating, loss of appetite, tremors, high blood pressure), anxiety and depressive symptoms are well established components of opioid and alcohol dependence and withdrawal, which are currently being treated with psychotropic drugs, including benzodiazepines, valproate and antiadrenergic agents [8, 15].

Xenon is a monatomic inert gas with very low chemical reactivity. It is colorless, odorless and heavy. Xenon has a very low blood-gas partition coefficient, rapidly penetrates the blood–brain barrier, which makes it an ideal general anesthetic. Xenon is a competitive N-methyl-d-aspartate (NMDA) receptor antagonist, which it exhibits through binding to glycine site of glutamatergic NMDA receptor [16]. In addition, xenon reduces excitatory neurotransmission through downregulation of 5-HT3 [17], nicotinic acetylcholine [18], potassium channel [19], HCN channel [20], and AMPA [21]. It also increases inhibitory neurotransmission by upregulating TREK1 [22]. Of relevance to fear associated conditions such as PD, the role of NMDA receptors in modulation of fear memories has previously been suggested [23]. Accordingly, Meloni et al. demonstrated that administration of xenon gas in a rat model of fear memory reconsolidation—a state in which recalled memories become susceptible to modification, reduced conditioned fear induced freezing [24].

Other psychiatric uses of xenon have been explored, for example, promising results on the use of inhaled xenon in opioid and alcohol withdrawal states, based on its pharmacokinetic effects have been reported [25–28]; in particular, its anti-stress properties, decreased sensitivity to pain and improved adaptation [29, 30]. However, there is paucity of research on the use of xenon outside of anesthesiology and addiction. According to some authors, its therapeutic properties are likely based on its effect on the glutamatergic system neuromodulation [31] via inhibition of NMDA receptors and reduction in binding of glutamate [31–33]. It was also shown that xenon at a dose of 30–50–70% in the gas mixture does not alter plasma concentrations of dopamine and norepinephrine, but causes a significant reduction of the level of adrenaline and cortisol [34].

It should also be noted that due to the biochemical inertness xenon, it exhibits no acute or chronic toxicity [35], embryotoxic or teratogenic effects, it is non-allergenic [18, 33], and does not alter the integrity of brain structures [36].

In recent studies it was demonstrated that the glutamatergic system plays a significant role in the regulation of anxiety. In particular, blockade of glutamatergic transmission in the periaqueductal gray matter lead to restoration of normal behavior in animals, and glutamate antagonists exhibited anxiolytic properties in experimental conditions [37].

In addition, preclinical studies have shown that blocking the glycine site NMDA-glutamate receptors results in anxiolytic effects [38]. Some supporting evidence that implicates glutamate in the pathogenesis of anxiety disorders stems from the efficacy of pregabalin, in which the mechanism of action is associated with inhibition of glutamate release [39]. Thus, on the basis of clinical data previously obtained from the use of xenon in anesthesiology and addiction medicine, as well as based on its receptor activity profile (reduction of glutamatergic neurotransmission) it can be expected that xenon possesses an independent anti-anxiety effect. The preliminary experience of using xenon in the outpatient treatment of various psychiatric and addictive diseases has been marked by its clear anxiolytic effect, which triggered our desire to study xenon’s effect on specific anxiety disorders. PD was selected because of its paroxysmal, easily quantifiable nature and a high degree of recurrence, and also because the “panic attack” phenomenon occurs widely in other anxiety states.

The clinical study presented aimed to: (a) study efficacy and adverse effect profile of xenon in acute treatment as a monotherapy for “pure” PD; (b) assess efficacy and adverse effect profile of xenon in treatment of PD in the presence of other mental illness comorbidities; and (c) quantify the duration of xenon’s therapeutic effect.

Methods

Results

The two groups of patients, with “pure” PD (group 1, n = 42) and with “comorbid” PD (group 2, n = 39) were well matched (Table 1). For patients in group 2 the following comorbid disorders were most commonly observed: mixed anxiety-depressive disorder (43.6%), bipolar affective disorder (10.3%), recurrent depressive disorder (10.3%), obsessive–compulsive disorder (5.1%), and other nonpsychotic mental disorders (12.8%, heading F48).

Changes in the subscale of “anxiety” in Hospital Anxiety and Depression Scale (HADS_T) are presented in Fig. 1. The total score on this scale in both groups corresponded to the level of “clinically severe anxiety” (17.7 and 19.0, respectively), and showed a decrease (−4.6 and 5.7 points, respectively) after 3 sessions (V3) of xenon administration (13.3 and 13.3, respectively). By the end of treatment (V6), the overall scores in both groups corresponded to the category of the “norm” for HADS_T Scale. Statistical analysis of the changes in SAS Scale using paired test samples is presented in Table 2.

Analysis of Clinical Global Impression Scale Improvement Subscale (CGI-I) changes after the third treatment shows a more significant improvement with xenon treatment (“marked improvement” on the CGI-I) in group 1 than in group 2 (40.5 and 10.3%, respectively, when compared to baseline). This trend persisted after 6 treatments: the indicator “very much improved” in patients with “pure” PD was 52.4%, while for those with “comorbid” PD it was only 12.8% (Table 3). According to the Clinical Global Impression Scale Severity of the Disease Subscale (CGI-S) (Table 4), before the start of treatment, both groups of patients demonstrated a pronounced degree of impairment: the indicator “significantly pronounced disease” was at 90.5 and 87.2%, respectively. After the third procedure, reduction in the severity of disorders was more pronounced in group 1: the indicator “moderately severe disease” was 48.7 and 11.9%, respectively. At the same time, upon completion of xenon treatments the differences between the two groups disappeared and most patients in both groups reached the “borderline” level (82.1 and 88.1%, respectively).

Thus, by analyzing the changes in the indices of psychometric scales (HADS_T, CGI-I, CGI-S) it can be concluded that the use of xenon treatment in PD produced rapid onset of action, statistically significant clinical improvement, and complete cessation of panic attacks after the 6th treatment.

Symptom changes on the SAS scale are presented in Fig. 2. The initial presentation in both groups corresponded to “high level of anxiety”(72.7 and 64.1, respectively). One month after treatment, all patients showed a decrease in the SAS total score, although it was more pronounced in group 1: 36.5 points (which corresponds to “no anxiety”) against 46.8 points in group 2 (“minimum degree of anxiety”). Furthermore, these parameters remained approximately at the same level throughout the study follow up (34.5 and 47.9, respectively). Statistical analysis of changes in SAS scale using paired test samples are presented in Table 5.

As noted above, in the modern classifications, in addition to “major” episodes that meet the criteria of a panic attack based on the number of symptoms, “limited symptom” (“minor”) panic attacks have been described, which, nevertheless, have an impact on social functioning and quality of life. As seen in Fig. 3, the mean number of “major” panic attacks per month in group 2 was even greater than that of the group 1 (7.7 and 11.7, respectively), while the number of “minor” attacks were slightly higher in group 1, or patients with “pure” PD (44.8 and 41.7, respectively). 6 months after treatment “major” panic attacks were absent in both groups, while “minor” panic attacks occurred in a very few cases (.3–1, respectively).

The results of SAS scales and lack of panic attacks after 6 months of treatment indicate the sustained anxiolytic effect of xenon administration.

As has already been noted, most often encountered comorbid mental conditions in group 2 included depressive disorders. The high degree of severity of depressive symptoms indicate that the traditional in these cases antidepressant therapy was ineffective. While the effect of xenon directly on depression is beyond the scope of this study, the analysis of HADS_T subscale “Depression” of HADS_D scale (Table 6) warrants some observations on this topic. According to HADS_D, “clinically severe depression” was absent in 66.7% of the patients in group 1 before the start of treatment, while it was present in 92.3% of group 2. After 3 xenon treatments it was absent in 90.5% of patients in group 1, but was still present in the majority of patients in group 2 (82.1%). By the end of the active phase of treatment “clinically severe depression” was negligible in patients of group 1 (2.4%), and it decreased to 46.2% in group 2.

Xenon therapy was generally well tolerated, side effects, mainly headache and dizziness, were rare and lead to only 5 patients dropping out from the study (5.8%). After carefully reviewing the data from these patients, it should be noted that four of them were found to have clinical symptoms of mild organic brain disease of vascular origin (F06.71 heading ICD-10), which is indirectly confirmed by the results of head and neck Doppler ultrasound. It was previously demonstrated that inhalation of xenon can increase cerebral blood flow [6].

Discussion

Despite the fact that SSRIs, SNRIs and benzodiazepines have proven efficacy in the treatment of PD, the delayed onset of action for the former and the side effects and the risk of dependence for the latter limit its use in the most active cohort of patients with PD. In addition, there are currently insufficient data on the treatment of refractory PD and on effective augmentation strategies, as well as on the treatment PD with comorbid mental illnesses.

The inert gas xenon was first shown to possess anesthetic properties over 50 years ago [43]. Over the last 10 years the interest in xenon as an inhalational anesthetic has increased due to several characteristics associated with its use: cardiovascular stability, rapid induction and emergence from anesthesia, and its analgesic effects—all of which make it an ideal anesthetic [44]. As a result, xenon has become more routinely used as an anesthetic agent in Europe and Japan and has garnered increasing interest in the United States although, primarily due to the higher cost of xenon as compared to other inhalational anesthetics, it has not yet received FDA approval. Evidence suggests that xenon’s biological effects may be mediated through its ability to potently block the NMDA receptors [45].

Furthermore, xenon has distinct advantages over other NMDA antagonists, such as ketamine, for future translation to the clinical setting. First, subsedative concentrations of xenon that would sufficiently block the NMDA receptor without producing anesthesia could potentially be administered briefly in a safe and effective manner in the outpatient setting with minimal medical monitoring. Second, in contrast to existing NMDA receptor blockers like ketamine, xenon has been shown to inhibit NMDA receptor activity through competitive inhibition of the co-agonist glycine at the glycine site of the NMDA receptor [45]—a mechanism devoid of psychotomimetic effects.

On the basis of the result of this study’s clinical and psychometric data with the use of scales assessing both the severity of anxiety (SAS, HADS_T), and evaluation of treatment effect in general (CGI-I, CGI-S), several preliminary conclusions can be made.

Firstly, when using xenon as an acute treatment of PD, reduction in both frequency and severity of panic attacks and anxiety level was observed during the first three treatment sessions, and by the end of treatment the vast number of patients experienced complete resolution of panic attacks while anxiety symptoms decreased to a subclinical level. This conclusion is also true for patients with PD with comorbidities. Secondly, after treatment of both “pure” PD and PD with psychiatric comorbidities, anxiolytic effect of xenon was maintained for at least 6 months, which clinically manifested by cessation of panic attacks. Thirdly, treatment with xenon was well tolerated, although it is possible that patients with symptoms of cerebral insufficiency may experience such adverse events as headache and dizziness. Fourthly, although the effect of xenon on depressive disorders was beyond the scope of this study, the analysis of the changes in HADS_D scale allows us to make a preliminary conclusion: in patients with PD and comorbid depressive symptoms xenon treatment can not only lead to improvement in anxiety symptoms, but also to reduction of depression severity. In this sense, we see as highly relevant a conduct of the study on the use of xenon as an adjuvant therapy for those patients with depressive disorders and concomitant anxiety (with anxious distress according to DSM-5), which have not sufficiently responded to treatment with conventional pharmacotherapy. Fifthly, the current first line of treatment of PD, in addition to pharmacotherapy, includes cognitive-behavioral therapy (CBT) [7]. At the same time, in clinical practice, the use of CBT in patients with severe PD in the early stages is often hampered by the high intensity of the somatic symptoms of anxiety and hypochondriacal fears, for relief of which providers often resort to the use of benzodiazepines. This approach can affect cognitive function and impede effective personal involvement in psychotherapy. In this sense, xenon treatment may be a good alternative to benzodiazepine tranquilizers because of its good tolerability, rapidity of onset and its lack of addictive potential.

The main methodological limitation of this study is its open design. In order to determine the place xenon in the treatment of PD additional randomized, placebo-controlled clinical trials of xenon and psychotropic substances used for the treatment of PD (SSRIs, SNRIs, benzodiazepines) are needed. The design of such studies should distinguish between direct anxiolytic effect of xenon and a potential placebo effect.

Conclusions

The present study is the first work on the use of xenon in panic disorder that can give impetus to a more intensive research into xenon’s place in the treatment of anxiety and depressive disorders as both the adjunct and a potential alternative to the currently used psychotropic pharmacotherapy. Given the accepted use of subanesthetic concentrations of xenon in imaging, and established safety profile of concentrations similar to the ones utilized in the current study, future investigation of xenon based therapeutics in prospective double blind placebo controlled trials is warranted.