Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder Andrea Woznica, Kristin Vickers n, Naomi Koerner, Katie Fracalanza Department of Psychology, Ryerson University, Toronto, Ontario, Canada
ar t ic l e i nf o
a b s t r a c t
Article history: Received 19 July 2014 Received in revised form 11 March 2015 Accepted 2 May 2015
The inhalation of 35% carbon dioxide (CO2) induces panic and anxiety in people with panic disorder (PD) and in people with various other psychiatric disorders. The anxiogenic effect of CO2 in people with eating disorders has received sparse attention despite the fact that PD and bulimia nervosa (BN) have several common psychological and neurobiological features. This study compared CO2-reactivity across three groups of participants: females with BN, females with PD, and female controls without known risk factors for enhanced CO2-reactivity (e.g., social anxiety disorder, first degree relatives with PD). Reactivity was measured by self-reported ratings of panic symptomatology and subjective anxiety, analyzed as both continuous variables (change from room-air to CO2) and dichotomous variables (positive versus negative responses to CO2). Analyses of each outcome measure demonstrated that CO2reactivity was similar across the BN and PD groups, and reactivity within each of these two groups was significantly stronger than that in the control group. This is the first study to demonstrate CO2hyperreactivity in individuals with BN, supporting the hypothesis that reactivity to this biological paradigm is not specific to PD. Further research would benefit from examining transdiagnostic mechanisms in CO2-hyperreactivity, such as anxiety sensitivity, which may account for this study's results. & 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Carbon dioxide Biological challenge Panic disorder Bulimia nervosa
1. Introduction A single-breath inhalation of 35% carbon dioxide (CO2) enriched air balanced with 65% oxygen, known as the 35% CO2 challenge, is a valid and reliable experimental procedure to provoke anxiety and panic-like symptoms (Griez et al., 1987; Vickers et al., 2012). Over two decades of research have established that people with panic disorder (PD) experience hyperreactivity to this concentration of CO2, such that they react with significantly greater panic symptomatology and subjective anxiety compared to people without psychiatric disorders (Vickers et al., 2012). CO2-hyperreactivity has also been observed in people with other disorders including situational and natural environment phobias (Verburg et al., 1994), social anxiety disorder (SAD; Schmidt and Richey, 2008), premenstrual dysphoric disorder (PMDD; Harrison et al., 1989), and possibly post-traumatic stress disorder (PTSD), although findings conflict (Muhtz et al., 2011; Talesnik et al., 2007). Hyperreactivity has similarly been demonstrated in those without a PD diagnosis but with sporadic unexpected panic attacks (Perna et al., 1995a) and in those with a family history of PD (van Beek and Griez, 2000). In contrast, CO2-hyperreactivity has not been observed in other diagnostic groups, including generalized anxiety disorder (GAD; Verburg et al., 1995), obsessive–compulsive disorder (OCD; Griez
n
Corresponding author. Tel.: þ 1 416 979 5000x7727; fax: þ 1 416 979 5273. E-mail address: [email protected] (K. Vickers).
et al., 1990), animal phobias (Verburg et al., 1994), and mood disorders (Perna et al., 1995b). Only one CO2 study has tested people with eating disorders (EDs), finding that an ED diagnosis did not predict hyperreactivity (Perna et al., 2004). At face value, this suggests that EDs are different from the aforementioned disorders in which hyperreactivity has been observed. In Perna et al.'s study, the ED group was composed of people with bulimia nervosa (BN) and people with anorexia nervosa (AN), likely on the assumption that BN and AN have more in common with one another than with other disorders. However, evidence converges to suggest that questions about CO2hyperreactivity should be pursued in BN specifically, as BN has much in common with PD, including a shared genetic loading (Kendler et al., 1995), serotonin deficiencies (Kaye et al., 1998), and significant anxiety symptoms produced by sodium lactate infusion (e.g., Lindy et al., 1988). Additionally, comorbidity rates between PD and EDs characterized by cycles of bingeing and purging are high relative to comorbidity between PD and restricting anorexia. For example, Godart et al. (2003) reported the prevalence of PD in an AN sample (AN-Restrictive subtype 5.4%; AN-Binge/Purge subtype 14.5%) and in a BN sample (BN-Purging subtype 20.9%). Most other studies have demonstrated similar findings (e.g., see Godart et al., 2002, Swinbourne and Touyz, 2007 for reviews; see also Kaye et al., 2004). Accordingly, the main objective of the present study was to examine whether hyperreactivity to 35% CO2 characterizes females
http://dx.doi.org/10.1016/j.psychres.2015.05.050 0165-1781/& 2015 Elsevier Ireland Ltd. All rights reserved.
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
A. Woznica et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
with BN compared to females with PD and female controls who do not have known risk factors for enhanced CO2-reactivity. One would expect high degrees of reactivity to 35% CO2 in those with a diagnosis of PD, as the CO2 provocation is known to aggravate panic in those with panic histories (e.g., Perna et al., 1995a). Additionally, evidence suggests that the serotonergic abnormalities that occur in PD (Coplan et al., 1992; Maron and Shlik, 2006) are associated with enhanced reactivity to CO2. However, a history of unexpected panic attacks is an exclusionary criterion for participants with BN in the current study; thus we cannot attribute hyperreactivity in BN to panic history as we potentially could in previous studies, including those using sodium lactate (e. g., Lindy et al., 1988). Therefore, in the current study, although individuals with BN are characterized by a serotonin deficiency similar to other psychodiagnostic groups that hyper-react to CO2 (e.g., PD, PMDD, SAD), they are without panic histories and thus cannot be expected to react to CO2 to the same extent as individuals with PD. It was therefore hypothesized that reactivity in the BN group would be intermediate between that of the PD and control groups, such that those with PD would be most reactive and control participants least reactive to CO2.
2. Method
2.2. Self-report measures of CO2-reactivity CO2-reactivity was assessed by subjective anxiety on the Subjective Units of Distress Scale (SUDS; Wolpe, 1973) and panic symptomatology on the Acute Panic Inventory (API; Liebowitz et al., 1984). Participants completed these two measures at baseline and immediately after each inhalation (room-air, CO2-enriched air). The SUDS is a visual analog scale for anxiety (VAS-A) measuring subjective anxiety from 0 (no anxiety at all) to 100 (the worst anxiety imaginable). The API is a 17-item questionnaire assessing the symptoms of physical and cognitive arousal associated with panic attacks. Participants rate the severity of each symptom from 0 (absent) to 3 (severe) and a total symptom score (TSS) is derived. In this study, internal consistency of the API according to Cronbach's alpha (calculated at three time points) ranged from 0.81 to 0.88. VAS-A and TSS were examined across groups in two ways: (1) as continuous variables – the change in scores from post-room-air inhalation to post-CO2 inhalation; and (2) as dichotomous variables – the presence or absence of a positive response to CO2. VAS-A change scores were calculated according to the method described by Perna et al. (2004), and a positive VAS-A response was defined as an increase of Z 26 points in anxiety from room-air to CO2 (Battaglia and Perna, 1995). A positive TSS response was defined as an increase of Z 4 items on the API, regardless of the intensity of the increase on any single item. 2.3. Apparatus Two gas mixtures were used: room-air and 35% CO2-enriched air. A breathing circuit contained a disposable mouthpiece and bacterial/viral filter connected via tubing to a two-way non-rebreathing valve. The inspiratory valve port attached to a manual stopcock on which one port fed room-air and the other connected to a nondiffusing gas collection bag filled with CO2-enriched air. A clinical vital signs monitor provided safety measures of BP, heart rate and oxygen saturation, and a pneumotach yielded tidal volume.
2.1. Participants 2.4. Procedure Three groups of female adults were included in this study: 14 with a principal diagnosis of BN-Purging Type, 15 with a principal diagnosis of PD, and 30 controls. All participants were recruited from the community via advertisements. A total of 80 participants met inclusion criteria during a telephone screen (over 100 additional participants were screened and found ineligible). Eligible participants were invited to the laboratory. Of these, 16 were subsequently excluded for not meeting eligibility requirements (medical and/or psychological) during more rigorous in-person screening procedures. The CO2-challenge was terminated early for two participants due to abnormal blood pressure, and for one participant with intense self-reported anxiety during the procedure. Data from two additional participants were excluded from analyses because their tidal volume during the CO2-inhalation was less than 80% of their vital capacity. Thus a final sample size of 59 remained. Psychiatric diagnoses according to DSM-IV-TR criteria were established using the Mini-International Neuropsychiatric Interview (MINI; Sheehan et al., 1998), a widely used semi-structured diagnostic interview with psychometric properties comparable to the Structured Clinical Interview for DSM-IV-TR Axis-I Disorders (SCID-I; First et al., 2001). The MINI-Screen was first administered over the telephone along with a screening question for specific phobias from the SCID-I. The complete MINI and specific phobia section from the SCID-I were subsequently administered in the laboratory. Diagnostic criteria for each study group were confirmed: BN-P, without comorbid PD; PD, without any comorbid ED; and neither PD nor an ED. Exclusionary for all participants was the endorsement of situational or natural environmental phobias, SAD, or PMDD since individuals with these diagnoses have displayed CO2-hyperreactivity in previous studies (Harrison et al., 1989; Schmidt and Richey, 2008; Verburg et al., 1994). Additional exclusions for the control and BN groups only were a history of sporadic unexpected panic attacks and PD in first or second-degree relatives. A senior-level doctoral student in clinical psychology was trained and supervised by PhD-level psychologists to administer this diagnostic interview. Medical exclusion criteria were consistent with previous studies (e.g., Perna et al., 2004); they were assessed over the telephone and again in the laboratory on a self-report questionnaire. Participants were required to have been deemed healthy by a medical doctor via a physical examination within the past year, and have none of the following: cardiovascular, respiratory, renal, or neurological disorders; complicated migraines; head injury; current pregnancy; family history of cerebral aneurysm or hemorrhage, or hemiplegic migraine. Also exclusionary was the use of psychotropic medications except for benzodiazepines occasionally (less than twice/ week and not within five half lives of the laboratory visit) or use of a medication that can affect heart rate (e.g., beta-blockers). Participants were asked to refrain from nicotine and caffeine for 4-hours prior to the laboratory visit, and the experiment was terminated if significant changes in blood pressure (BP) occurred during the CO2-challenge (e.g., systolic BP reaches Z 170 or r 90).
The study consisted of one two-hour laboratory visit. After completing eligibility screening and demographic measures, participants were informed that they would be inhaling two harmless gas mixtures containing different percentages of CO2 and O2, which might cause transitory discomfort or panic-like symptoms. Participants were connected to the clinical monitor, which measured vitals every 60-seconds. They were then connected to the breathing circuit and vital capacity was measured. A single blind, placebo-controlled method was used for the inhalations; participants took one breath of room-air (placebo) and subsequently one breath of 35% CO2-enriched air (e.g., Harrison et al., 1989). Only inhalations that were at least 80% of vital capacity were considered valid. Immediately following each inhalation, participants completed reactivity measures. Study procedures were approved by Ryerson University's Research Ethics Board. 2.5. Data analytic strategy To examine group differences in CO2-reactivity, analyses of variance (ANOVA) and Tukey's HSD pairwise comparisons were used to compare the degree of VAS-A and TSS change from room-air to CO2. Chi-square tests were used to compare the proportions of VAS-A and TSS positive responses to CO2.
3. Results 3.1. Preliminary analyses Missing values (less than 10% of the total data matrix and consistent with missing-at-random data) were replaced via mean substitution. All continuous variables approximated normality in their distributions. 3.2. Demographic and characteristics of the sample Age did not differ significantly across the BN (mean age: 25.147 S.D. 7.67 years), PD (mean age: 26.00 77.92 years), and control (mean age: 24.13 75.95 years) groups, F(2, 56) ¼ 0.38, p¼ 0.68, η2p ¼ 0:01. Ethnicity also did not differ significantly across the BN (n¼ 6; 42.9%), PD (n ¼10; 66.7%) and control (n¼ 9; 30.0%) groups, χ2(18) ¼25.87, p ¼0.10, ϕ ¼0.66. In total, 18.6% (n ¼11) of the sample reported symptoms consistent with at least one other
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
A. Woznica et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Table 1 Means and standard deviations of VAS-A and TSS scores after each inhalation (raw scores), and proportions, means, and standard deviations of reactivity after CO2 inhalation.
VAS-A after room-air VAS-A after CO2 TSS after room-air TSS after CO2 VAS-A % change* TSS change* Positive VAS-A responses to CO2* Positive TSS responses to CO2*
Bulimia nervosa (n¼ 14)
Panic disorder (n¼ 15)
Controls (n¼ 30)
13.29 7 16.25 21.647 22.87 2.647 2.17 5.007 3.57 11.047 14.61 2.36 71.98 3 (21%) 5 (36%)
21.13 717.00 30.60 7 20.33 7.477 3.87 9.73 74.85 7.187 28.24 2.27 73.95 5 (33%) 5 (33.3%)
8.23 7 14.79 8.17 713.69 2.4 7 4.66 2.337 3.25 10.32 725.23a 0.0773.00a 0 (0%) 1 (3%)
VAS-A ¼ visual analog scale for anxiety; TSS ¼total symptom score. a Negative mean change scores imply that some control participants reported less anxiety and fewer panic symptoms after CO2 compared to room-air. Given that we would not expect a significant increase in anxiety or panic post-CO2 in this group, a reasonable explanation for this finding is that the anticipatory anxiety experienced by some control participants steadily decreased from the first to the second inhalation after learning that the first inhalation (room-air) did not produce changes in bodily sensations. n po 0.05.
psychiatric diagnosis. The proportions of participants who endorsed a mood disorder (i.e., major depressive disorder; MDD) were consistent across groups, χ2(2)¼ 3.38, p ¼0.18, ϕ ¼0.24. The proportions of participants who endorsed an anxiety disorder (apart from PD in the PD group) were significantly different across groups, χ2(2) ¼12.12, p¼ 0.002, ϕ ¼0.45, with a greater proportion in the PD group compared to both the BN group, χ2(1)¼4.27, p ¼0.04, ϕ ¼0.38, and the control group, χ2(1)¼10.24, p ¼0.001, ϕ ¼ 0.48. Within the PD group, a total of seven participants (46.67%) endorsed comorbid disorders. Specifically, one participant endorsed OCD and MDD; one participant endorsed GAD and MDD; two participants endorsed PTSD; one participant endorsed GAD; and one participant endorsed MDD. Within the BN group, one participant (7.1%) endorsed PTSD and two participants (14.3%) endorsed MDD. Within the control group, one participant (3.3%) endorsed OCD and MDD. In addition, participants in the BN group reported a mean total of 4.93 74.65 binges and 6.36 77.54 purges per week over the past three months. Participants in the PD group reported a mean total of 2.87 72.77 panic attacks over the past month. 3.3. Comparison of CO2-reactivity across groups Table 1 displays means and standard deviations of VAS-A and TSS raw scores after each inhalation, means and standard deviations of VAS-A and TSS change scores, and proportions of VAS-A and TSS positive responses. The four measures of reactivity were not perfectly correlated (see Table 2); consequently, results for each measure are presented below. ANOVAs revealed a significant effect of group on VAS-A and TSS change scores, F(2, 56) ¼4.87, p ¼0.01, η2p ¼ 0:15, and F(2, 56) ¼ 5.01, p¼ 0.01, η2p ¼ 0:15, respectively. Planned comparisons established that the increase in VAS-A scores from room-air to CO2 was significantly greater in the BN group, t(56) ¼2.74, p ¼0.01, d ¼0.73, and the PD group, t(56) ¼ 2.30, p ¼0.03, d ¼0.62, compared to the control group. VAS-A change scores did not differ between the BN and PD groups, t(56) ¼0.43, p¼ 0.67, d¼ 0.12. Similarly, the increase in TSS scores was significantly greater in the BN group, t (56) ¼2.60, p ¼0.01, d ¼0.69, and the PD group, t(56) ¼2.56, p ¼0.01, d ¼0.68, compared to the control group. TSS change scores did not differ between the BN and PD groups, t(56) ¼ 0.08, p ¼0.93, d ¼0.02. Chi-square tests revealed a significant difference across groups in the proportion of participants with positive VAS-A responses to CO2, χ2(2)¼ 10.45, p ¼0.05, ϕ ¼0.42; the proportions of BN and PD participants with positive VAS-A responses were similar, χ2(2)¼ 0.51, p ¼0.47, ϕ ¼0.13, and significantly greater than controls, χ2(1)¼11.25, p ¼0.001, ϕ ¼0.50 and χ2(1) ¼6.90, p¼ 0.01,
Table 2 Intercorrelations among measures of CO2-reactivity. VAS-A positive responses VAS-A positive responses TSS positive responses VAS-A change TSS change
TSS positive responses
VAS-A change
TSS change
1 0.59*
1
1 0.70*
1
0.56* 0.52*
0.48* 0.61*
VAS-A ¼visual analog scale for anxiety; TSS¼ total symptom score. n
p o 0.01.
ϕ ¼0.40, respectively. A significant difference across groups was also found in the proportions of participants with positive TSS responses, χ2(2) ¼9.46, p ¼0.01, ϕ ¼ 0.40; the proportions of BN and PD participants who exhibited positive TSS responses were similar, χ2(1) ¼0.02, p ¼0.89, ϕ ¼ 0.03, and significantly greater than controls, χ2(1)¼ 8.50, p ¼0.004, ϕ ¼0.44, and χ2(2) ¼7.79, p¼ 0.01, ϕ ¼0.42, respectively.
4. Discussion The assertion that CO2-hyperreactivity is a specific marker for PD pathology (Verburg et al., 1998) has been questioned by more recent studies linking CO2-hyperreactivity to other psychiatric diagnoses. Consequently, the purported specificity of this response warrants continued investigation. The current study furthered this line of inquiry by examining CO2-reactivity amongst individuals with BN, a clinical group that has received little attention in this research area yet is known to possess commonalities with PD. Both panic symptomatology and subjective anxiety were used to measure reactivity, and each was analyzed as a continuous variable (change from room-air to CO2) and as a dichotomous variable (positive versus negative responses to CO2). It was hypothesized that participants with PD would be most reactive and control participants would be least reactive; and participants with BN would display intermediate levels of reactivity. Contrary to our hypothesis, analyses of each outcome measure demonstrated that CO2-reactivity was similar across the PD and BN groups. In addition, reactivity within the control group was significantly lower than that in both the BN and PD groups. The finding that control and BN participants react differently to CO2 is novel and worthy of further investigation.
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
4
A. Woznica et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
That noted, heterogeneity was present within each clinical group in the degree of CO2-reactivity. Thus, the statement that those with BN and PD were more reactive than control participants should be understood to mean that enhanced reactivity occurred in these clinical groups on average but only certain individuals within each group demonstrated a positive response to CO2. In fact, the proportion of PD participants demonstrating a positive VAS-A response in the current study (33%) was lower than that in the previous investigation of CO2-reactivity in ED (79%; Perna et al., 2004). Our participants were recruited from the community compared to an outpatient hospital setting in the previous study and thus likely had substantially less severe PD symptoms. Given that the present findings demonstrate similar reactivity between BN and PD, there is a question of whether a BN diagnosis itself predicts CO2-hyperreactivity, or if another factor drives the relationship. Along those lines, previous studies have examined potential cognitive correlates of CO2-hyperreactivity, such as high levels of anxiety sensitivity (AS; Reiss and McNally, 1985) and low levels of distress tolerance (DT; Simons and Gaher, 2005), given their associations with panic and anxiety psychopathology (e.g., Eifert et al., 1999; Keough et al., 2010; Kutz et al., 2010; Leyro et al., 2010; Perna et al., 2003). These constructs are also associated with BN (Anestis et al., 2007, 2008) and thus may help to explain the shared reactivity between BN and PD in this study. Additional factors that relate to both BN and PD have been shown to enhance reactivity in studies using different CO2 percentages. These factors include: maladaptive coping strategies (Spira et al., 2004), emotional avoidance (Karekla et al., 2004; Levitt et al., 2004), and the experience of early separation-related adverse events (Battaglia et al., 2009; Ogliari et al., 2010). Consideration of these factors in future studies may elucidate a shared cognitive factor between BN and PD that can help to explain the enhanced reactivity displayed by a proportion of BN participants in the current study. More generally, studies of CO2-reactivity might be picking up on underlying vulnerabilities to anxiety and panic that exist across diagnostic groups. In the present study specifically, the possible susceptibility of individuals with BN to panic and anxiety psychopathology is illustrated, and further research may identify more specific mechanisms that account for this relationship. This line of research relates to the growing interest in identifying transdiagnostic mechanisms of dysfunction (e.g., Cuthbert and Insel, 2013) that cut across the current psychiatric diagnostic categories (e.g., American Psychiatric Association, 2013). CO2 can be used in future research to help identify transdiagnostic mechanisms that underlie the expression of distinct disorders. Undoubtedly, this research can have important implications for the treatment of BN. For example, if AS and DT are found in future studies to consistently mediate CO2-hyperreactivity, then individuals with BN may benefit from explicit affective-regulation training in treatment, which would be expected to directly impact levels of AS and DT. Such a component is not currently included in cognitive behavioral therapy (CBT), a first-line treatment for BN (NICE, 2004; Wilson, 2005). However, studies have shown that dialectical behavior therapy is helpful for individuals with BN who have affect regulatory problems (Chen et al., 2008). Building on the present study in these ways has the potential to inform the conceptualization and treatment of BN. On the other hand, an additional possibility to consider is that subthreshold symptoms of specific psychiatric disorders are contributing to this study's findings. For example, in participants without a diagnosis of SAD, social anxiety symptoms may still affect CO2 response. A final possibility is that there is no underlying commonality between BN and PD that can explain the observed shared CO2-hyperreactivity. Perhaps those with BN and those with PD show enhanced reactivity for different reasons.
The current study has several strengths. First, this is the only known study using a homogenous ED sample of females with BN only to test the specificity of the CO2-challenge. Perna et al. (2004) utilized a more diverse ED sample; psychological and physiological differences exist between ED types, complicating interpretation of the non-significant findings of that previous study. Second, in the current study, persons who endorsed certain risk factors known to enhance CO2-reactivity (e.g., SAD, first degree relatives with PD) were excluded from participating. Third, given the lack of a single, consensual definition of reactivity in CO2-challenges (Vickers, 2012), this study utilized both continuous and dichotomous measures that are commonly used by other researchers. Limitations of this study are also noteworthy. Some participants within each group had comorbid mood and anxiety disorders. Excluding such individuals from study participation would have prevented these variables from potentially confounding the results; yet doing so may have also resulted in a non-ecologically valid sample. We should note that the greater proportion of comorbid anxiety disorders in the PD group could not have accounted for the shared hyperreactivity between BN and PD; the BN and control groups did not differ in their proportions of comorbid anxiety disorders, nonetheless the BN group experienced greater reactivity compared to the control group. In addition, relatively small samples sizes in the BN and PD groups were present and perhaps lacked power to detect differences between them; that noted, examination of raw change scores (as shown in Table 1) demonstrates that the BN group, on average, was more reactive than the PD group according to three of the four reactivity measures. In sum, the current study is the first to our knowledge to demonstrate CO2-hyperreactivity in individuals with BN. Replicating results with a larger sample size will enhance confidence in the finding that a BN diagnosis is an important factor in CO2hyperreactivity. Further research may also illuminate vulnerability factors for CO2-hyperreactivity that are shared among individuals with BN and PD. This line of research ultimately has important implications for both the specificity of CO2-hyperreactivity and the conceptualization and treatment of BN.
Acknowledgments This research was supported in part by a Canadian Institutes of Health Research Canada Graduate Scholarship and an Ontario Graduate Scholarship awarded to Andrea Woznica for her masters-level research, and by a Ryerson University Faculty of Arts Grant to Kristin Vickers.
References American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders, 5th ed. American Psychiatric Association, Washington, DC. Anestis, M.D., Holm-Denoma, J.M., Gordon, K.H., Schmidt, N.B., Joiner, T.E., 2008. The role of anxiety sensitivity in eating pathology. Cognitive Therapy and Research 32, 370–385. Anestis, M.D., Selby, E.A., Fink, E.L., Joiner, T.E., 2007. The multifaceted role of distress tolerance in dysregulated eating behaviors. International Journal of Eating Disorders 40, 718–726. Battaglia, M., Perna, G., 1995. The 35% CO2 challenge in panic disorder: optimization by receiver operating characteristic (ROC) analysis. Journal of Psychiatric Research 29, 111–119. Battaglia, M., Pesenti-Gritti, P., Medland, S.E., Ogliari, A., Tambs, K., Spatola, C.A.M., 2009. A genetically informed study of the association between childhood separation anxiety, sensitivity to CO2, panic disorder, and the effect of childhood parental loss. Archives of General Psychiatry 66, 64–71. Chen, E.Y., Matthews, L., Allen, C., Kuo, J.R., Linehan, M.M., 2008. Dialectical behavior therapy for clients with binge eating disorder or bulimia nervosa and borderline personality disorder. International Journal of Eating Disorders 41, 505–512.
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
A. Woznica et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎ Coplan, J.D., Gorman, J.M., Klein, D.F., 1992. Serotonin related functions in panicanxiety: a critical overview. Neuropsychopharmacology 6, 189–200. Cuthbert, B.N., Insel, T.R., 2013. Toward the future of psychiatric diagnosis: the seven pillars of RDoC. BMC Medicine 11, 1–8. Eifert, G.H., Zvolensky, M.J., Sorrell, J.T., Hopko, D.R., Lejuez, C.W., 1999. Predictors of self-reported anxiety and panic symptoms: an evaluation of anxiety sensitivity, suffocation fear, heart-focused anxiety, and breath-holding duration. Journal of Psychopathology and Behavioral Assessment 21, 293–305. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 2001. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition. (SCID-I/P) Biometrics Research. New York State Psychiatric Institute, New York. Godart, N.T., Flament, M.F., Perdereau, F., Jeammet, P., 2002. Comorbidity between eating disorders and anxiety disorders: a review. International Journal of Eating Disorders 32, 253–270. Godart, N.T., Flament, M.F., Curt, F., Perdereau, F., Lang, F., Venisse, J.L., Halfon, O., Bizouard, P., Loas, G., Corcos, M., Jeammet, P., Fermanian, J., 2003. Anxiety disorders in subjects seeking treatment for eating disorders: a DSM-IV controlled study. Psychiatry Research 117, 245–258. Griez, E., de Loof, C., Pols, H., Zandbergen, J., Lousberg, H., 1990. Specific sensitivity of patients with panic attacks to carbon dioxide inhalation. Psychiatry Research 31, 193–199. Griez, E.J., Lousberg, H., van den Hout, M.A., van der Molen, G.M., 1987. CO2 vulnerability in panic disorder. Psychiatry Research 20, 87–95. Harrison, W.M., Sandberg, D., Gorman, J.M., Fyer, M., Nee, J., Uy, J., Endicott, J., 1989. Provocation of panic with carbon dioxide inhalation in patients with premenstrual dysphoria. Psychiatry Research 27, 183–192. Karekla, M., Forsyth, J., Kelly, M.M., 2004. Emotional avoidance and panicogenic responding to a biological challenge procedure. Behavior Therapy 35, 725–746. Kaye, W., Bulik, C., Thornton, L., Barbarich, N., Masters, K., 2004. Comorbidity of anxiety disorders with anorexia and bulimia nervosa. American Journal of Psychiatry 161, 2215–2221. Kaye, W., Gendall, K., Strober, M., 1998. Serotonin neuronal function and selective serotonin reuptake inhibitor treatment in anorexia and bulimia nervosa. Biological Psychiatry 44, 825–838. Kendler, K.S., Walters, E.E., Neale, M.C., Kessler, R.C., Heath, A.C., Eaves, L.J., 1995. The structure of the genetic and environmental risk factors for six major psychiatric disorders in women: phobia, generalized anxiety disorder, panic disorder, bulimia, major depression, and alcoholism. Archives of General Psychiatry 52, 374–383. Keough, M.G., Riccardi, C., Timpano, K.R., Mitchell, M.A., Schmidt, N.B., 2010. Anxiety symptomatology: the association with distress tolerance and anxiety sensitivity. Behavior Therapy 41, 567–574. Kutz, A., Marshall, E., Bernstein, A., Zvolensky, M.J., 2010. Evaluating emotional sensitivity and tolerance factors in the prediction of panic-relevant responding to a biological challenge. Journal of Anxiety Disorders 24, 16–22. Levitt, J.T., Brown, T.A., Orsillo, S.M., Barlow, D.H., 2004. The effects of acceptance versus suppression of emotion on subjective and psychophysiological response to carbon dioxide challenge in patients with panic disorder. Behavior Therapy 35, 747–766. Leyro, T.M., Zvolensky, M.J., Bernstein, A., 2010. Distress tolerance and psychopathological symptoms and disorders: a review of the empirical literature among adults. Psychological Bulletin 136, 576–600. Liebowitz, M.R., Gorman, J.M., Fyer, A.J., Dillon, D.J., Klein, D.F., 1984. Effects of naloxone on patients with panic attacks. American Journal of Psychiatry 141, 995–997. Lindy, D.C., Walsh, B.T., Gorman, J.M., Roose, S.P., Gladis, M., Devlin, M.J., Glassman, A.H., 1988. Lactate infusions in patients with bulimia. Psychiatry Research 26, 287–292. Maron, E., Shlik, J., 2006. Serotonin function in panic disorder: important, but why? Neuropsychopharmacology 31, 1–11.
5
Muhtz, C., Yassouridis, A., Daneshi, J., Braun, M., Kellner, M., 2011. Acute panicogenic, anxiogenic and dissociative effects of carbon dioxide inhalation in patients with post-traumatic stress disorder (PTSD). Journal of Psychiatric Research 45, 989–993. NICE, 2004. Eating Disorders: Core Interventions in the Treatment and Management of Anorexia Nervosa, Bulimia Nervosa, and Related Eating Disorders NICE Clinical Guideline No. 9. National Institute for Clinical Excellence, London. Ogliari, A., Tambs, K., Harris, J., Scaini, D., Maffei, C., Reichborn-Kjennerud, T., Battaglia, M., 2010. The relationships between adverse events, early antecedents, and carbon dioxide reactivity as an intermediate phenotype of panic disorder. Psychotherapy and Psychosomatics 79, 48–55. Perna, G., Barbini, B., Cocchi, S., Bertani, A., Gasperini, M., 1995b. 35% CO2 challenge in panic and mood disorders. Journal of Affective Disorders 33, 189–194. Perna, G., Casolari, A., Bussi, R., Cucchi, M., Arancio, C., Bellodi, L., 2004. Comparison of 35% carbon dioxide reactivity between panic disorder and eating disorder. Psychiatry Research 125, 277–283. Perna, G., Gabriele, A., Caldirola, D., Bellodi, L., 1995a. Hypersensitivity to inhalation of carbon dioxide & panic attacks. Psychiatry Research 57, 267–273. Perna, G., Romano, P., Caldirola, D., Cocchi, M., Bellodi, L., 2003. Anxiety sensitivity and 35% CO2 reactivity in patients with panic disorder. Journal of Psychosomatic Research 54, 573–577. Reiss, S., McNally, R.J., 1985. Expectancy model of fear. In: Reiss, S., Bootzin, R.R. (Eds.), Theoretical Issues in Behavior Therapy. Academic Press, New York, pp. 107–121. Schmidt, N.B., Richey, J.A., 2008. Social anxiety symptoms uniquely predict fear responding to 35% CO2 challenge. Journal of Psychiatric Research 42, 851–857. Sheehan, D., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry 59, 22–33. Simons, J.S., Gaher, R.M., 2005. The distress tolerance scale: development and validation of a self-report measure. Motivation and Emotion 29, 83–102. Spira, A.P., Zvolensky, M.J., Eifert, G.H., Feldner, M.T., 2004. Avoidance-oriented coping as a predictor of anxiety-based physical stress: a test using biological challenge. Journal of Anxiety Disorders 18, 309–323. Swinbourne, J.M., Touyz, S.W., 2007. The co-morbidity of eating disorders and anxiety disorders: a review. European Eating Disorders Review 15, 253–274. Talesnik, B., Berzak, E., Ben-Zion, I., Kaplan, Z., Benjamin, J., 2007. Sensitivity to carbon dioxide in drug-naïve subjects with post-traumatic stress disorder. Journal of Psychiatric Research 41, 451–454. van Beek, N., Griez, E., 2000. Reactivity to a 35% CO2 challenge in healthy firstdegree relatives of patients with panic disorder. Biological Psychiatry 47, 830–835. Verburg, C., Griez, E., Meijer, J., 1994. A 35% carbon dioxide challenge in simple phobias. Acta Psychiatrica Scandinavica 90, 420–423. Verburg, K., Griez, E., Meijer, J., Pols, H., 1995. Discrimination between panic disorder and generalized anxiety disorder by 35% carbon dioxide challenge. American Journal of Psychiatry 152, 1081–1083. Verburg, K., Pols, H., de Leeuw, M., Griez, E., 1998. Reliability of the 35% carbon dioxide provocation challenge. Psychiatry Research 78, 207–214. Vickers, K., 2012. Hypersensitivity to hypercapnia: definition/(s). Psychiatry Research 197, 7–12. Vickers, K., Jafarpour, S., Mofidi, A., Rafat, B., Woznica, A., 2012. The 35% carbon dioxide test in stress and panic research: overview of effects and integration of findings. Clinical Psychology Review 32, 153–164. Wilson, G.T., 2005. Psychological treatment of eating disorders. Annual Review of Psychology 1, 439–465. Wolpe, J., 1973. The Practice of Behavior Therapy. Pergamon Press, New York.
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder Andrea Woznica, Kristin Vickers n, Naomi Koerner, Katie Fracalanza Department of Psychology, Ryerson University, Toronto, Ontario, Canada
ar t ic l e i nf o
a b s t r a c t
Article history: Received 19 July 2014 Received in revised form 11 March 2015 Accepted 2 May 2015
The inhalation of 35% carbon dioxide (CO2) induces panic and anxiety in people with panic disorder (PD) and in people with various other psychiatric disorders. The anxiogenic effect of CO2 in people with eating disorders has received sparse attention despite the fact that PD and bulimia nervosa (BN) have several common psychological and neurobiological features. This study compared CO2-reactivity across three groups of participants: females with BN, females with PD, and female controls without known risk factors for enhanced CO2-reactivity (e.g., social anxiety disorder, first degree relatives with PD). Reactivity was measured by self-reported ratings of panic symptomatology and subjective anxiety, analyzed as both continuous variables (change from room-air to CO2) and dichotomous variables (positive versus negative responses to CO2). Analyses of each outcome measure demonstrated that CO2reactivity was similar across the BN and PD groups, and reactivity within each of these two groups was significantly stronger than that in the control group. This is the first study to demonstrate CO2hyperreactivity in individuals with BN, supporting the hypothesis that reactivity to this biological paradigm is not specific to PD. Further research would benefit from examining transdiagnostic mechanisms in CO2-hyperreactivity, such as anxiety sensitivity, which may account for this study's results. & 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Carbon dioxide Biological challenge Panic disorder Bulimia nervosa
1. Introduction A single-breath inhalation of 35% carbon dioxide (CO2) enriched air balanced with 65% oxygen, known as the 35% CO2 challenge, is a valid and reliable experimental procedure to provoke anxiety and panic-like symptoms (Griez et al., 1987; Vickers et al., 2012). Over two decades of research have established that people with panic disorder (PD) experience hyperreactivity to this concentration of CO2, such that they react with significantly greater panic symptomatology and subjective anxiety compared to people without psychiatric disorders (Vickers et al., 2012). CO2-hyperreactivity has also been observed in people with other disorders including situational and natural environment phobias (Verburg et al., 1994), social anxiety disorder (SAD; Schmidt and Richey, 2008), premenstrual dysphoric disorder (PMDD; Harrison et al., 1989), and possibly post-traumatic stress disorder (PTSD), although findings conflict (Muhtz et al., 2011; Talesnik et al., 2007). Hyperreactivity has similarly been demonstrated in those without a PD diagnosis but with sporadic unexpected panic attacks (Perna et al., 1995a) and in those with a family history of PD (van Beek and Griez, 2000). In contrast, CO2-hyperreactivity has not been observed in other diagnostic groups, including generalized anxiety disorder (GAD; Verburg et al., 1995), obsessive–compulsive disorder (OCD; Griez
n
Corresponding author. Tel.: þ 1 416 979 5000x7727; fax: þ 1 416 979 5273. E-mail address: [email protected] (K. Vickers).
et al., 1990), animal phobias (Verburg et al., 1994), and mood disorders (Perna et al., 1995b). Only one CO2 study has tested people with eating disorders (EDs), finding that an ED diagnosis did not predict hyperreactivity (Perna et al., 2004). At face value, this suggests that EDs are different from the aforementioned disorders in which hyperreactivity has been observed. In Perna et al.'s study, the ED group was composed of people with bulimia nervosa (BN) and people with anorexia nervosa (AN), likely on the assumption that BN and AN have more in common with one another than with other disorders. However, evidence converges to suggest that questions about CO2hyperreactivity should be pursued in BN specifically, as BN has much in common with PD, including a shared genetic loading (Kendler et al., 1995), serotonin deficiencies (Kaye et al., 1998), and significant anxiety symptoms produced by sodium lactate infusion (e.g., Lindy et al., 1988). Additionally, comorbidity rates between PD and EDs characterized by cycles of bingeing and purging are high relative to comorbidity between PD and restricting anorexia. For example, Godart et al. (2003) reported the prevalence of PD in an AN sample (AN-Restrictive subtype 5.4%; AN-Binge/Purge subtype 14.5%) and in a BN sample (BN-Purging subtype 20.9%). Most other studies have demonstrated similar findings (e.g., see Godart et al., 2002, Swinbourne and Touyz, 2007 for reviews; see also Kaye et al., 2004). Accordingly, the main objective of the present study was to examine whether hyperreactivity to 35% CO2 characterizes females
http://dx.doi.org/10.1016/j.psychres.2015.05.050 0165-1781/& 2015 Elsevier Ireland Ltd. All rights reserved.
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
A. Woznica et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
with BN compared to females with PD and female controls who do not have known risk factors for enhanced CO2-reactivity. One would expect high degrees of reactivity to 35% CO2 in those with a diagnosis of PD, as the CO2 provocation is known to aggravate panic in those with panic histories (e.g., Perna et al., 1995a). Additionally, evidence suggests that the serotonergic abnormalities that occur in PD (Coplan et al., 1992; Maron and Shlik, 2006) are associated with enhanced reactivity to CO2. However, a history of unexpected panic attacks is an exclusionary criterion for participants with BN in the current study; thus we cannot attribute hyperreactivity in BN to panic history as we potentially could in previous studies, including those using sodium lactate (e. g., Lindy et al., 1988). Therefore, in the current study, although individuals with BN are characterized by a serotonin deficiency similar to other psychodiagnostic groups that hyper-react to CO2 (e.g., PD, PMDD, SAD), they are without panic histories and thus cannot be expected to react to CO2 to the same extent as individuals with PD. It was therefore hypothesized that reactivity in the BN group would be intermediate between that of the PD and control groups, such that those with PD would be most reactive and control participants least reactive to CO2.
2. Method
2.2. Self-report measures of CO2-reactivity CO2-reactivity was assessed by subjective anxiety on the Subjective Units of Distress Scale (SUDS; Wolpe, 1973) and panic symptomatology on the Acute Panic Inventory (API; Liebowitz et al., 1984). Participants completed these two measures at baseline and immediately after each inhalation (room-air, CO2-enriched air). The SUDS is a visual analog scale for anxiety (VAS-A) measuring subjective anxiety from 0 (no anxiety at all) to 100 (the worst anxiety imaginable). The API is a 17-item questionnaire assessing the symptoms of physical and cognitive arousal associated with panic attacks. Participants rate the severity of each symptom from 0 (absent) to 3 (severe) and a total symptom score (TSS) is derived. In this study, internal consistency of the API according to Cronbach's alpha (calculated at three time points) ranged from 0.81 to 0.88. VAS-A and TSS were examined across groups in two ways: (1) as continuous variables – the change in scores from post-room-air inhalation to post-CO2 inhalation; and (2) as dichotomous variables – the presence or absence of a positive response to CO2. VAS-A change scores were calculated according to the method described by Perna et al. (2004), and a positive VAS-A response was defined as an increase of Z 26 points in anxiety from room-air to CO2 (Battaglia and Perna, 1995). A positive TSS response was defined as an increase of Z 4 items on the API, regardless of the intensity of the increase on any single item. 2.3. Apparatus Two gas mixtures were used: room-air and 35% CO2-enriched air. A breathing circuit contained a disposable mouthpiece and bacterial/viral filter connected via tubing to a two-way non-rebreathing valve. The inspiratory valve port attached to a manual stopcock on which one port fed room-air and the other connected to a nondiffusing gas collection bag filled with CO2-enriched air. A clinical vital signs monitor provided safety measures of BP, heart rate and oxygen saturation, and a pneumotach yielded tidal volume.
2.1. Participants 2.4. Procedure Three groups of female adults were included in this study: 14 with a principal diagnosis of BN-Purging Type, 15 with a principal diagnosis of PD, and 30 controls. All participants were recruited from the community via advertisements. A total of 80 participants met inclusion criteria during a telephone screen (over 100 additional participants were screened and found ineligible). Eligible participants were invited to the laboratory. Of these, 16 were subsequently excluded for not meeting eligibility requirements (medical and/or psychological) during more rigorous in-person screening procedures. The CO2-challenge was terminated early for two participants due to abnormal blood pressure, and for one participant with intense self-reported anxiety during the procedure. Data from two additional participants were excluded from analyses because their tidal volume during the CO2-inhalation was less than 80% of their vital capacity. Thus a final sample size of 59 remained. Psychiatric diagnoses according to DSM-IV-TR criteria were established using the Mini-International Neuropsychiatric Interview (MINI; Sheehan et al., 1998), a widely used semi-structured diagnostic interview with psychometric properties comparable to the Structured Clinical Interview for DSM-IV-TR Axis-I Disorders (SCID-I; First et al., 2001). The MINI-Screen was first administered over the telephone along with a screening question for specific phobias from the SCID-I. The complete MINI and specific phobia section from the SCID-I were subsequently administered in the laboratory. Diagnostic criteria for each study group were confirmed: BN-P, without comorbid PD; PD, without any comorbid ED; and neither PD nor an ED. Exclusionary for all participants was the endorsement of situational or natural environmental phobias, SAD, or PMDD since individuals with these diagnoses have displayed CO2-hyperreactivity in previous studies (Harrison et al., 1989; Schmidt and Richey, 2008; Verburg et al., 1994). Additional exclusions for the control and BN groups only were a history of sporadic unexpected panic attacks and PD in first or second-degree relatives. A senior-level doctoral student in clinical psychology was trained and supervised by PhD-level psychologists to administer this diagnostic interview. Medical exclusion criteria were consistent with previous studies (e.g., Perna et al., 2004); they were assessed over the telephone and again in the laboratory on a self-report questionnaire. Participants were required to have been deemed healthy by a medical doctor via a physical examination within the past year, and have none of the following: cardiovascular, respiratory, renal, or neurological disorders; complicated migraines; head injury; current pregnancy; family history of cerebral aneurysm or hemorrhage, or hemiplegic migraine. Also exclusionary was the use of psychotropic medications except for benzodiazepines occasionally (less than twice/ week and not within five half lives of the laboratory visit) or use of a medication that can affect heart rate (e.g., beta-blockers). Participants were asked to refrain from nicotine and caffeine for 4-hours prior to the laboratory visit, and the experiment was terminated if significant changes in blood pressure (BP) occurred during the CO2-challenge (e.g., systolic BP reaches Z 170 or r 90).
The study consisted of one two-hour laboratory visit. After completing eligibility screening and demographic measures, participants were informed that they would be inhaling two harmless gas mixtures containing different percentages of CO2 and O2, which might cause transitory discomfort or panic-like symptoms. Participants were connected to the clinical monitor, which measured vitals every 60-seconds. They were then connected to the breathing circuit and vital capacity was measured. A single blind, placebo-controlled method was used for the inhalations; participants took one breath of room-air (placebo) and subsequently one breath of 35% CO2-enriched air (e.g., Harrison et al., 1989). Only inhalations that were at least 80% of vital capacity were considered valid. Immediately following each inhalation, participants completed reactivity measures. Study procedures were approved by Ryerson University's Research Ethics Board. 2.5. Data analytic strategy To examine group differences in CO2-reactivity, analyses of variance (ANOVA) and Tukey's HSD pairwise comparisons were used to compare the degree of VAS-A and TSS change from room-air to CO2. Chi-square tests were used to compare the proportions of VAS-A and TSS positive responses to CO2.
3. Results 3.1. Preliminary analyses Missing values (less than 10% of the total data matrix and consistent with missing-at-random data) were replaced via mean substitution. All continuous variables approximated normality in their distributions. 3.2. Demographic and characteristics of the sample Age did not differ significantly across the BN (mean age: 25.147 S.D. 7.67 years), PD (mean age: 26.00 77.92 years), and control (mean age: 24.13 75.95 years) groups, F(2, 56) ¼ 0.38, p¼ 0.68, η2p ¼ 0:01. Ethnicity also did not differ significantly across the BN (n¼ 6; 42.9%), PD (n ¼10; 66.7%) and control (n¼ 9; 30.0%) groups, χ2(18) ¼25.87, p ¼0.10, ϕ ¼0.66. In total, 18.6% (n ¼11) of the sample reported symptoms consistent with at least one other
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
A. Woznica et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Table 1 Means and standard deviations of VAS-A and TSS scores after each inhalation (raw scores), and proportions, means, and standard deviations of reactivity after CO2 inhalation.
VAS-A after room-air VAS-A after CO2 TSS after room-air TSS after CO2 VAS-A % change* TSS change* Positive VAS-A responses to CO2* Positive TSS responses to CO2*
Bulimia nervosa (n¼ 14)
Panic disorder (n¼ 15)
Controls (n¼ 30)
13.29 7 16.25 21.647 22.87 2.647 2.17 5.007 3.57 11.047 14.61 2.36 71.98 3 (21%) 5 (36%)
21.13 717.00 30.60 7 20.33 7.477 3.87 9.73 74.85 7.187 28.24 2.27 73.95 5 (33%) 5 (33.3%)
8.23 7 14.79 8.17 713.69 2.4 7 4.66 2.337 3.25 10.32 725.23a 0.0773.00a 0 (0%) 1 (3%)
VAS-A ¼ visual analog scale for anxiety; TSS ¼total symptom score. a Negative mean change scores imply that some control participants reported less anxiety and fewer panic symptoms after CO2 compared to room-air. Given that we would not expect a significant increase in anxiety or panic post-CO2 in this group, a reasonable explanation for this finding is that the anticipatory anxiety experienced by some control participants steadily decreased from the first to the second inhalation after learning that the first inhalation (room-air) did not produce changes in bodily sensations. n po 0.05.
psychiatric diagnosis. The proportions of participants who endorsed a mood disorder (i.e., major depressive disorder; MDD) were consistent across groups, χ2(2)¼ 3.38, p ¼0.18, ϕ ¼0.24. The proportions of participants who endorsed an anxiety disorder (apart from PD in the PD group) were significantly different across groups, χ2(2) ¼12.12, p¼ 0.002, ϕ ¼0.45, with a greater proportion in the PD group compared to both the BN group, χ2(1)¼4.27, p ¼0.04, ϕ ¼0.38, and the control group, χ2(1)¼10.24, p ¼0.001, ϕ ¼ 0.48. Within the PD group, a total of seven participants (46.67%) endorsed comorbid disorders. Specifically, one participant endorsed OCD and MDD; one participant endorsed GAD and MDD; two participants endorsed PTSD; one participant endorsed GAD; and one participant endorsed MDD. Within the BN group, one participant (7.1%) endorsed PTSD and two participants (14.3%) endorsed MDD. Within the control group, one participant (3.3%) endorsed OCD and MDD. In addition, participants in the BN group reported a mean total of 4.93 74.65 binges and 6.36 77.54 purges per week over the past three months. Participants in the PD group reported a mean total of 2.87 72.77 panic attacks over the past month. 3.3. Comparison of CO2-reactivity across groups Table 1 displays means and standard deviations of VAS-A and TSS raw scores after each inhalation, means and standard deviations of VAS-A and TSS change scores, and proportions of VAS-A and TSS positive responses. The four measures of reactivity were not perfectly correlated (see Table 2); consequently, results for each measure are presented below. ANOVAs revealed a significant effect of group on VAS-A and TSS change scores, F(2, 56) ¼4.87, p ¼0.01, η2p ¼ 0:15, and F(2, 56) ¼ 5.01, p¼ 0.01, η2p ¼ 0:15, respectively. Planned comparisons established that the increase in VAS-A scores from room-air to CO2 was significantly greater in the BN group, t(56) ¼2.74, p ¼0.01, d ¼0.73, and the PD group, t(56) ¼ 2.30, p ¼0.03, d ¼0.62, compared to the control group. VAS-A change scores did not differ between the BN and PD groups, t(56) ¼0.43, p¼ 0.67, d¼ 0.12. Similarly, the increase in TSS scores was significantly greater in the BN group, t (56) ¼2.60, p ¼0.01, d ¼0.69, and the PD group, t(56) ¼2.56, p ¼0.01, d ¼0.68, compared to the control group. TSS change scores did not differ between the BN and PD groups, t(56) ¼ 0.08, p ¼0.93, d ¼0.02. Chi-square tests revealed a significant difference across groups in the proportion of participants with positive VAS-A responses to CO2, χ2(2)¼ 10.45, p ¼0.05, ϕ ¼0.42; the proportions of BN and PD participants with positive VAS-A responses were similar, χ2(2)¼ 0.51, p ¼0.47, ϕ ¼0.13, and significantly greater than controls, χ2(1)¼11.25, p ¼0.001, ϕ ¼0.50 and χ2(1) ¼6.90, p¼ 0.01,
Table 2 Intercorrelations among measures of CO2-reactivity. VAS-A positive responses VAS-A positive responses TSS positive responses VAS-A change TSS change
TSS positive responses
VAS-A change
TSS change
1 0.59*
1
1 0.70*
1
0.56* 0.52*
0.48* 0.61*
VAS-A ¼visual analog scale for anxiety; TSS¼ total symptom score. n
p o 0.01.
ϕ ¼0.40, respectively. A significant difference across groups was also found in the proportions of participants with positive TSS responses, χ2(2) ¼9.46, p ¼0.01, ϕ ¼ 0.40; the proportions of BN and PD participants who exhibited positive TSS responses were similar, χ2(1) ¼0.02, p ¼0.89, ϕ ¼ 0.03, and significantly greater than controls, χ2(1)¼ 8.50, p ¼0.004, ϕ ¼0.44, and χ2(2) ¼7.79, p¼ 0.01, ϕ ¼0.42, respectively.
4. Discussion The assertion that CO2-hyperreactivity is a specific marker for PD pathology (Verburg et al., 1998) has been questioned by more recent studies linking CO2-hyperreactivity to other psychiatric diagnoses. Consequently, the purported specificity of this response warrants continued investigation. The current study furthered this line of inquiry by examining CO2-reactivity amongst individuals with BN, a clinical group that has received little attention in this research area yet is known to possess commonalities with PD. Both panic symptomatology and subjective anxiety were used to measure reactivity, and each was analyzed as a continuous variable (change from room-air to CO2) and as a dichotomous variable (positive versus negative responses to CO2). It was hypothesized that participants with PD would be most reactive and control participants would be least reactive; and participants with BN would display intermediate levels of reactivity. Contrary to our hypothesis, analyses of each outcome measure demonstrated that CO2-reactivity was similar across the PD and BN groups. In addition, reactivity within the control group was significantly lower than that in both the BN and PD groups. The finding that control and BN participants react differently to CO2 is novel and worthy of further investigation.
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
4
A. Woznica et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎
That noted, heterogeneity was present within each clinical group in the degree of CO2-reactivity. Thus, the statement that those with BN and PD were more reactive than control participants should be understood to mean that enhanced reactivity occurred in these clinical groups on average but only certain individuals within each group demonstrated a positive response to CO2. In fact, the proportion of PD participants demonstrating a positive VAS-A response in the current study (33%) was lower than that in the previous investigation of CO2-reactivity in ED (79%; Perna et al., 2004). Our participants were recruited from the community compared to an outpatient hospital setting in the previous study and thus likely had substantially less severe PD symptoms. Given that the present findings demonstrate similar reactivity between BN and PD, there is a question of whether a BN diagnosis itself predicts CO2-hyperreactivity, or if another factor drives the relationship. Along those lines, previous studies have examined potential cognitive correlates of CO2-hyperreactivity, such as high levels of anxiety sensitivity (AS; Reiss and McNally, 1985) and low levels of distress tolerance (DT; Simons and Gaher, 2005), given their associations with panic and anxiety psychopathology (e.g., Eifert et al., 1999; Keough et al., 2010; Kutz et al., 2010; Leyro et al., 2010; Perna et al., 2003). These constructs are also associated with BN (Anestis et al., 2007, 2008) and thus may help to explain the shared reactivity between BN and PD in this study. Additional factors that relate to both BN and PD have been shown to enhance reactivity in studies using different CO2 percentages. These factors include: maladaptive coping strategies (Spira et al., 2004), emotional avoidance (Karekla et al., 2004; Levitt et al., 2004), and the experience of early separation-related adverse events (Battaglia et al., 2009; Ogliari et al., 2010). Consideration of these factors in future studies may elucidate a shared cognitive factor between BN and PD that can help to explain the enhanced reactivity displayed by a proportion of BN participants in the current study. More generally, studies of CO2-reactivity might be picking up on underlying vulnerabilities to anxiety and panic that exist across diagnostic groups. In the present study specifically, the possible susceptibility of individuals with BN to panic and anxiety psychopathology is illustrated, and further research may identify more specific mechanisms that account for this relationship. This line of research relates to the growing interest in identifying transdiagnostic mechanisms of dysfunction (e.g., Cuthbert and Insel, 2013) that cut across the current psychiatric diagnostic categories (e.g., American Psychiatric Association, 2013). CO2 can be used in future research to help identify transdiagnostic mechanisms that underlie the expression of distinct disorders. Undoubtedly, this research can have important implications for the treatment of BN. For example, if AS and DT are found in future studies to consistently mediate CO2-hyperreactivity, then individuals with BN may benefit from explicit affective-regulation training in treatment, which would be expected to directly impact levels of AS and DT. Such a component is not currently included in cognitive behavioral therapy (CBT), a first-line treatment for BN (NICE, 2004; Wilson, 2005). However, studies have shown that dialectical behavior therapy is helpful for individuals with BN who have affect regulatory problems (Chen et al., 2008). Building on the present study in these ways has the potential to inform the conceptualization and treatment of BN. On the other hand, an additional possibility to consider is that subthreshold symptoms of specific psychiatric disorders are contributing to this study's findings. For example, in participants without a diagnosis of SAD, social anxiety symptoms may still affect CO2 response. A final possibility is that there is no underlying commonality between BN and PD that can explain the observed shared CO2-hyperreactivity. Perhaps those with BN and those with PD show enhanced reactivity for different reasons.
The current study has several strengths. First, this is the only known study using a homogenous ED sample of females with BN only to test the specificity of the CO2-challenge. Perna et al. (2004) utilized a more diverse ED sample; psychological and physiological differences exist between ED types, complicating interpretation of the non-significant findings of that previous study. Second, in the current study, persons who endorsed certain risk factors known to enhance CO2-reactivity (e.g., SAD, first degree relatives with PD) were excluded from participating. Third, given the lack of a single, consensual definition of reactivity in CO2-challenges (Vickers, 2012), this study utilized both continuous and dichotomous measures that are commonly used by other researchers. Limitations of this study are also noteworthy. Some participants within each group had comorbid mood and anxiety disorders. Excluding such individuals from study participation would have prevented these variables from potentially confounding the results; yet doing so may have also resulted in a non-ecologically valid sample. We should note that the greater proportion of comorbid anxiety disorders in the PD group could not have accounted for the shared hyperreactivity between BN and PD; the BN and control groups did not differ in their proportions of comorbid anxiety disorders, nonetheless the BN group experienced greater reactivity compared to the control group. In addition, relatively small samples sizes in the BN and PD groups were present and perhaps lacked power to detect differences between them; that noted, examination of raw change scores (as shown in Table 1) demonstrates that the BN group, on average, was more reactive than the PD group according to three of the four reactivity measures. In sum, the current study is the first to our knowledge to demonstrate CO2-hyperreactivity in individuals with BN. Replicating results with a larger sample size will enhance confidence in the finding that a BN diagnosis is an important factor in CO2hyperreactivity. Further research may also illuminate vulnerability factors for CO2-hyperreactivity that are shared among individuals with BN and PD. This line of research ultimately has important implications for both the specificity of CO2-hyperreactivity and the conceptualization and treatment of BN.
Acknowledgments This research was supported in part by a Canadian Institutes of Health Research Canada Graduate Scholarship and an Ontario Graduate Scholarship awarded to Andrea Woznica for her masters-level research, and by a Ryerson University Faculty of Arts Grant to Kristin Vickers.
References American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders, 5th ed. American Psychiatric Association, Washington, DC. Anestis, M.D., Holm-Denoma, J.M., Gordon, K.H., Schmidt, N.B., Joiner, T.E., 2008. The role of anxiety sensitivity in eating pathology. Cognitive Therapy and Research 32, 370–385. Anestis, M.D., Selby, E.A., Fink, E.L., Joiner, T.E., 2007. The multifaceted role of distress tolerance in dysregulated eating behaviors. International Journal of Eating Disorders 40, 718–726. Battaglia, M., Perna, G., 1995. The 35% CO2 challenge in panic disorder: optimization by receiver operating characteristic (ROC) analysis. Journal of Psychiatric Research 29, 111–119. Battaglia, M., Pesenti-Gritti, P., Medland, S.E., Ogliari, A., Tambs, K., Spatola, C.A.M., 2009. A genetically informed study of the association between childhood separation anxiety, sensitivity to CO2, panic disorder, and the effect of childhood parental loss. Archives of General Psychiatry 66, 64–71. Chen, E.Y., Matthews, L., Allen, C., Kuo, J.R., Linehan, M.M., 2008. Dialectical behavior therapy for clients with binge eating disorder or bulimia nervosa and borderline personality disorder. International Journal of Eating Disorders 41, 505–512.
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
A. Woznica et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎ Coplan, J.D., Gorman, J.M., Klein, D.F., 1992. Serotonin related functions in panicanxiety: a critical overview. Neuropsychopharmacology 6, 189–200. Cuthbert, B.N., Insel, T.R., 2013. Toward the future of psychiatric diagnosis: the seven pillars of RDoC. BMC Medicine 11, 1–8. Eifert, G.H., Zvolensky, M.J., Sorrell, J.T., Hopko, D.R., Lejuez, C.W., 1999. Predictors of self-reported anxiety and panic symptoms: an evaluation of anxiety sensitivity, suffocation fear, heart-focused anxiety, and breath-holding duration. Journal of Psychopathology and Behavioral Assessment 21, 293–305. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 2001. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition. (SCID-I/P) Biometrics Research. New York State Psychiatric Institute, New York. Godart, N.T., Flament, M.F., Perdereau, F., Jeammet, P., 2002. Comorbidity between eating disorders and anxiety disorders: a review. International Journal of Eating Disorders 32, 253–270. Godart, N.T., Flament, M.F., Curt, F., Perdereau, F., Lang, F., Venisse, J.L., Halfon, O., Bizouard, P., Loas, G., Corcos, M., Jeammet, P., Fermanian, J., 2003. Anxiety disorders in subjects seeking treatment for eating disorders: a DSM-IV controlled study. Psychiatry Research 117, 245–258. Griez, E., de Loof, C., Pols, H., Zandbergen, J., Lousberg, H., 1990. Specific sensitivity of patients with panic attacks to carbon dioxide inhalation. Psychiatry Research 31, 193–199. Griez, E.J., Lousberg, H., van den Hout, M.A., van der Molen, G.M., 1987. CO2 vulnerability in panic disorder. Psychiatry Research 20, 87–95. Harrison, W.M., Sandberg, D., Gorman, J.M., Fyer, M., Nee, J., Uy, J., Endicott, J., 1989. Provocation of panic with carbon dioxide inhalation in patients with premenstrual dysphoria. Psychiatry Research 27, 183–192. Karekla, M., Forsyth, J., Kelly, M.M., 2004. Emotional avoidance and panicogenic responding to a biological challenge procedure. Behavior Therapy 35, 725–746. Kaye, W., Bulik, C., Thornton, L., Barbarich, N., Masters, K., 2004. Comorbidity of anxiety disorders with anorexia and bulimia nervosa. American Journal of Psychiatry 161, 2215–2221. Kaye, W., Gendall, K., Strober, M., 1998. Serotonin neuronal function and selective serotonin reuptake inhibitor treatment in anorexia and bulimia nervosa. Biological Psychiatry 44, 825–838. Kendler, K.S., Walters, E.E., Neale, M.C., Kessler, R.C., Heath, A.C., Eaves, L.J., 1995. The structure of the genetic and environmental risk factors for six major psychiatric disorders in women: phobia, generalized anxiety disorder, panic disorder, bulimia, major depression, and alcoholism. Archives of General Psychiatry 52, 374–383. Keough, M.G., Riccardi, C., Timpano, K.R., Mitchell, M.A., Schmidt, N.B., 2010. Anxiety symptomatology: the association with distress tolerance and anxiety sensitivity. Behavior Therapy 41, 567–574. Kutz, A., Marshall, E., Bernstein, A., Zvolensky, M.J., 2010. Evaluating emotional sensitivity and tolerance factors in the prediction of panic-relevant responding to a biological challenge. Journal of Anxiety Disorders 24, 16–22. Levitt, J.T., Brown, T.A., Orsillo, S.M., Barlow, D.H., 2004. The effects of acceptance versus suppression of emotion on subjective and psychophysiological response to carbon dioxide challenge in patients with panic disorder. Behavior Therapy 35, 747–766. Leyro, T.M., Zvolensky, M.J., Bernstein, A., 2010. Distress tolerance and psychopathological symptoms and disorders: a review of the empirical literature among adults. Psychological Bulletin 136, 576–600. Liebowitz, M.R., Gorman, J.M., Fyer, A.J., Dillon, D.J., Klein, D.F., 1984. Effects of naloxone on patients with panic attacks. American Journal of Psychiatry 141, 995–997. Lindy, D.C., Walsh, B.T., Gorman, J.M., Roose, S.P., Gladis, M., Devlin, M.J., Glassman, A.H., 1988. Lactate infusions in patients with bulimia. Psychiatry Research 26, 287–292. Maron, E., Shlik, J., 2006. Serotonin function in panic disorder: important, but why? Neuropsychopharmacology 31, 1–11.
5
Muhtz, C., Yassouridis, A., Daneshi, J., Braun, M., Kellner, M., 2011. Acute panicogenic, anxiogenic and dissociative effects of carbon dioxide inhalation in patients with post-traumatic stress disorder (PTSD). Journal of Psychiatric Research 45, 989–993. NICE, 2004. Eating Disorders: Core Interventions in the Treatment and Management of Anorexia Nervosa, Bulimia Nervosa, and Related Eating Disorders NICE Clinical Guideline No. 9. National Institute for Clinical Excellence, London. Ogliari, A., Tambs, K., Harris, J., Scaini, D., Maffei, C., Reichborn-Kjennerud, T., Battaglia, M., 2010. The relationships between adverse events, early antecedents, and carbon dioxide reactivity as an intermediate phenotype of panic disorder. Psychotherapy and Psychosomatics 79, 48–55. Perna, G., Barbini, B., Cocchi, S., Bertani, A., Gasperini, M., 1995b. 35% CO2 challenge in panic and mood disorders. Journal of Affective Disorders 33, 189–194. Perna, G., Casolari, A., Bussi, R., Cucchi, M., Arancio, C., Bellodi, L., 2004. Comparison of 35% carbon dioxide reactivity between panic disorder and eating disorder. Psychiatry Research 125, 277–283. Perna, G., Gabriele, A., Caldirola, D., Bellodi, L., 1995a. Hypersensitivity to inhalation of carbon dioxide & panic attacks. Psychiatry Research 57, 267–273. Perna, G., Romano, P., Caldirola, D., Cocchi, M., Bellodi, L., 2003. Anxiety sensitivity and 35% CO2 reactivity in patients with panic disorder. Journal of Psychosomatic Research 54, 573–577. Reiss, S., McNally, R.J., 1985. Expectancy model of fear. In: Reiss, S., Bootzin, R.R. (Eds.), Theoretical Issues in Behavior Therapy. Academic Press, New York, pp. 107–121. Schmidt, N.B., Richey, J.A., 2008. Social anxiety symptoms uniquely predict fear responding to 35% CO2 challenge. Journal of Psychiatric Research 42, 851–857. Sheehan, D., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry 59, 22–33. Simons, J.S., Gaher, R.M., 2005. The distress tolerance scale: development and validation of a self-report measure. Motivation and Emotion 29, 83–102. Spira, A.P., Zvolensky, M.J., Eifert, G.H., Feldner, M.T., 2004. Avoidance-oriented coping as a predictor of anxiety-based physical stress: a test using biological challenge. Journal of Anxiety Disorders 18, 309–323. Swinbourne, J.M., Touyz, S.W., 2007. The co-morbidity of eating disorders and anxiety disorders: a review. European Eating Disorders Review 15, 253–274. Talesnik, B., Berzak, E., Ben-Zion, I., Kaplan, Z., Benjamin, J., 2007. Sensitivity to carbon dioxide in drug-naïve subjects with post-traumatic stress disorder. Journal of Psychiatric Research 41, 451–454. van Beek, N., Griez, E., 2000. Reactivity to a 35% CO2 challenge in healthy firstdegree relatives of patients with panic disorder. Biological Psychiatry 47, 830–835. Verburg, C., Griez, E., Meijer, J., 1994. A 35% carbon dioxide challenge in simple phobias. Acta Psychiatrica Scandinavica 90, 420–423. Verburg, K., Griez, E., Meijer, J., Pols, H., 1995. Discrimination between panic disorder and generalized anxiety disorder by 35% carbon dioxide challenge. American Journal of Psychiatry 152, 1081–1083. Verburg, K., Pols, H., de Leeuw, M., Griez, E., 1998. Reliability of the 35% carbon dioxide provocation challenge. Psychiatry Research 78, 207–214. Vickers, K., 2012. Hypersensitivity to hypercapnia: definition/(s). Psychiatry Research 197, 7–12. Vickers, K., Jafarpour, S., Mofidi, A., Rafat, B., Woznica, A., 2012. The 35% carbon dioxide test in stress and panic research: overview of effects and integration of findings. Clinical Psychology Review 32, 153–164. Wilson, G.T., 2005. Psychological treatment of eating disorders. Annual Review of Psychology 1, 439–465. Wolpe, J., 1973. The Practice of Behavior Therapy. Pergamon Press, New York.
Please cite this article as: Woznica, A., et al., Reactivity to 35% carbon dioxide in bulimia nervosa and panic disorder. Psychiatry Research (2015), http://dx.doi.org/10.1016/j.psychres.2015.05.050i
