Acute panicogenic, anxiogenic and dissociative effects of carbon dioxide inhalation in patients with post-traumatic stress disorder (PTSD)

Abstract

Background

Increased anxiety and panic to inhalation of carbon dioxide (CO₂) has been described in patients with anxiety disorders, especially panic disorder, compared to healthy subjects. Post-traumatic stress disorder (PTSD) has been hypothesised to resemble panic disorder and is currently classified as an anxiety disorder in DSM-IV. However, there are only very few data available about the sensitivity of patients with PTSD to CO₂.

Methods

In 10 patients with PTSD, 10 sex- and age-matched healthy subjects and 8 patients with panic disorder we assessed anxiety, panic, dissociative and PTSD symptoms before and after a single vital capacity inhalation of 35% CO₂.

Results

Patients with PTSD showed an increased anxiety, panic and dissociative reaction to the inhalation of 35% CO₂ compared to healthy participants. PTSD subjects’ responses were indistinguishable from those of panic patients. Additionally, PTSD-typical symptoms like post-traumatic flashbacks were provoked in patients with PTSD after the inhalation of CO₂.

Conclusions

In our sample, PTSD was associated with an increased CO₂ reactivity, pointing to an increased susceptibility of PTSD patients to CO₂ challenge.

Introduction

Autonomic arousal and arousability, clinically characterised by symptoms such as shortness of breath, rapid heartbeat, and trembling, are shared features of the DSM-IV anxiety disorder constructs panic disorder (PD) and post-traumatic stress disorder (PTSD) (Brown and McNiff, 2009). In addition to salient similarities in the phenomenology of panic attacks and re-experiencing (intrusive) symptoms of PTSD (Mellman and Davis, 1985), some panic provoking agents, such as intravenous sodium lactate, yohimbine, meta-chlorophenylpiperazine (mCPP) and cholecystokinin-tetrapeptide (CCK-4) have been reported to elicit not only panic attacks, but also post-traumatic flashbacks or other PTSD symptoms in patients with PTSD (Jensen et al., 1997, Kellner et al., 1998, Kellner et al., 2000 Kellner et al., 2000, Southwick et al., 1993, Southwick et al., 1997; Rainey et al., 1987; Southwick et al., 1993, 1997). Only the putative panicogen flumazenil did not induce panic anxiety or PTSD symptoms in these patients (Coupland et al., 1997, Randall et al., 1995).

Another useful paradigm to study panic anxiety is inhalation of carbon dioxide (CO₂) (Rassovsky and Kushner, 2003). Inhalation of a single deep breath of 35% CO₂ has proved to be a reliable panicogenic challenge model in patients with PD (Verburg et al., 1998a). Healthy subjects who underwent a double inhalation of different dosages of CO₂ reported a negative affect with somatic and cognitive symptoms similar to a panic attack correlating to the amount of CO₂ administered (Griez et al., 2007). Individual response to CO₂ inhalation has been investigated as a putative intermediate phenotype of susceptibility to panic disorder in the general population (Ogliari et al., 2010).

One primary theory of the mechanisms of the CO₂ responsitivity is that of Klein (1993), who suggested that panic disorder patients have increased CO₂ sensitivity of their central chemoreceptors, and thus are more likely to experience a ‘suffocation false alarm’ that in turn leads to panic. In preclinical investigations an acid-sensing ion channel in the amygdala has been identified as an important chemosensor for hypercapnia eliciting fear responses (Ziemann et al., 2009). Further research suggests that the administration of CO₂ leads to non-specific activation of a complex brain fear-network including the amygdala, the hippocampus and the medial prefrontal cortex (Gorman et al., 2000, Gorman et al., 2001). Potentially the noradrenergic system, particularly the locus coeruleus, plays an important role in this network-activation (Bailey et al., 2003). In a recent review Esquivel et al. (2010) postulated that inhalation of CO₂ leads to disturbances in cerebral pH homeostasis and provokes negative emotions ranging from breathlessness to panic. These non-specific interoceptive disturbances might be subsequently misinterpreted as catastrophic (“danger of impending suffocation”) by patients with anxiety disorders and lead to immediate urgency to breathe (Rassovsky et al., 2006; Colasanti et al., 2008).

For patients suffering from PTSD, only one investigation has so far characterised the effect of CO₂ inhalation: Talesnik et al. (2007) reported a lack of sensitivity to a single vital capacity inhalation of 35% CO₂ measuring panic anxiety and post-traumatic symptoms in patients with PTSD. Findings about possible alterations of CO₂ partial pressure have been contradictory in PTSD patients: End-tidal CO₂ partial pressure at rest was found to be significantly lower compared to healthy controls by Rosen et al. (1994), while Blechert et al. (2007) did not detect respective differences between patients with PTSD compared to healthy controls and patients with PD.

We studied the effects of CO₂ (35%) on panic anxiety and PTSD symptoms in subjects with PTSD, healthy matched controls and, to be sure of the assay sensitivity, panic patients as an additional positive control group. We hypothesised increased susceptibility concerning provocation of panic, anxiety and dissociative symptoms in PTSD patients versus healthy control subjects in this paradigm and expected provocation of post-traumatic flashbacks and other PTSD symptoms.