- Email: [email protected]
Alexithymia and anxiety sensitivity in populations at high risk for panic disorder
Available online at www.sciencedirect.com
Comprehensive Psychiatry 53 (2012) 868 – 874 www.elsevier.com/locate/comppsych
Alexithymia and anxiety sensitivity in populations at high risk for panic disorder Michele Cucchi⁎, Daniele Cavadini, Vittoria Bottelli, Liana Riccia, Vera Conca, Paolo Ronchi, Enrico Smeraldi Department of Clinical Neurosciences, Scientific Institute and University, Vita-Salute San Raffaele, Via Stamira D'Ancona 20, 20127 Milan, Italy
Abstract Objective: Populations at high risk for panic disorder (PD) probably share with subjects with PD an underlying vulnerability involving features like anxiety sensitivity (AS) and alexithymia. The present study would verify if PD relatives (R) and subjects who have experienced 1 or more panic attacks (PAs) show different levels of AS and alexithymia with respect to healthy controls (HC). Methods: One hundred fifty-seven HCs, 30 subjects with PA, 64 R subjects, and 139 outpatients with PD were evaluated and compared on AS, alexithymia, and control variables. Results: Subjects with PD show higher alexithymia and AS levels compared with HCs; R subjects do not differ on ASI total score; and R females show more alexithymic features. Subjects with PA are comparable with HCs both on AS and alexithymia. Conclusions: Results confirm an impairment in emotional and bodily sensations information processing in subjects with PD but partially disconfirm the expectation of a difference between R subjects and subjects with PA with respect to HCs on AS and alexithymia. Emotional and bodily sensation competencies could be protective factors for PD in high-risk populations. © 2012 Elsevier Inc. All rights reserved.
1. Introduction Several authors have pointed out that in anxiety disorders, in particular panic disorder (PD), amplification of somatic sensations is often associated to subsequent dysfunctional cognitive appraisal of these sensations with a significant bias toward a danger-related and catastrophizing interpretational style [1,2]. The construct of anxiety sensitivity (AS) [3] refers to the tendency to fear anxiety-related sensations. According to Reiss [4], individuals with elevated AS experience amplified fear in response to stimuli that elicit anxiety and find their own anxiety symptoms to be particularly aversive. Many studies have found large associations between AS and PD features [5-8] (for a meta-analysis, see Naragon-Gainey [9]). Some authors evidenced that the way that individuals appraise anxiety- and panic-related symptoms may increase the risk of developing agoraphobia, over and above the risk of developing PD [10]. Prospective studies have shown that AS predicts the onset of anxiety and panic in both nonclinical and clinical populations [11]. ⁎ Corresponding author. Tel.: +39 02 26433229; fax: +39 02 26433265. E-mail address: [email protected] (M. Cucchi). 0010-440X/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.comppsych.2012.01.005
A second interesting psychological construct pertaining to bodily sensations (related to emotion-related activation) information processing is alexithymia. The alexithymia construct, as defined by Nemiah et al [12], has 3 salient features: (1) difficulty identifying and describing subjective feelings, (2) difficulty distinguishing between feelings and the bodily sensations of emotional arousal, and (3) constricted imaginal capacities, as evidenced by paucity of fantasies and externally oriented cognitive style. The alexithymia construct reflect deficits in the cognitive processing and regulation of emotions [13]. As reported by Galderisi and colleagues [14], empirical evidence of poor emotion processing in subjects with PD has been provided, including a high prevalence of alexithymia. Regarding PD, the psychological construct of alexithymia has been extensively studied from different points of view [15-18]: overall results proved an alexithymic deficit in subjects with PD. In particular, the first feature of alexithymia seems to distinguish healthy control (HC) subjects from subjects with PD and to predict symptom severity. However, as evidenced in the study by Marchesi et al [18], the possibility of a trait or state interpretation of this construct remains controversial.
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
Converging lines of evidence suggest that PD in adulthood may represent manifestations of an underlying constitutional vulnerability or diathesis for anxiety that is partly genetic and variably expressed during a life cycle [19]. It is well known that PD runs in families [20,21]. Individual differences in AS and alexithymia are hypothesized to emerge from both genetic and maladaptive learning experiences that lead to the acquisition of beliefs about the aversive consequences of arousal and anxietyrelated states [22]. A recent study by Stein and colleagues, examining the heritability of AS in a group of 179 monozygotic and 158 dizygotic twin pairs, suggests that variation in AS (particularly the physical aspect of AS) is explained by unique environmental and additive genetic influences [23]. Emotional maltreatment in childhood, particularly when it includes parental threatening, hostile, and rejecting behaviors, is associated with higher levels of AS in young adults [24]. In a familial study, van Beek and Griez [25] found that the first-degree relatives of subjects with PD were significantly more anxiety sensitive than HC subjects, but less anxiety sensitive than patients with PD, suggesting that AS runs in families. Moreover, Stein et al [26] observed a statistically significant interaction between levels of childhood emotional (or physical) maltreatment and 5-HTTLPR genotype. Specifically, S/S individuals with higher levels of maltreatment had significantly higher levels of AS than subjects in other groups, evidencing a gene-by-environment (serotonin transporter and childhood maltreatment) interaction process influencing AS. In the same way, several authors studied genetic and environmental factors associated with alexithymia. For example, Ham et al [27] showed an association between the Val/Val COMT genotype and higher Toronto Alexithymia Scale (20-item version; TAS-20) scores in a sample of 109 students. Valera and Berenbaum [28], in a study involving 45 monozygotic and 32 dizygotic twin pairs, concluded that familial influences contribute to alexithymia, with a specific effect of environmental factors on difficulty in identifying feelings and communicating emotions and a specific effect of shared genetic factors on externally oriented thinking. In a larger sample composed of 8785 twin pairs, Jørgensen et al [29], using a structural equation modeling, selected an ACE model including additive genetic, shared environmental, and nonshared environmental effects providing the best fit for all 3 facets of alexithymia as well as total alexithymia scores, with heritabilities of 30% to 33% and the remaining variance being explained by shared (12%-20%) and nonshared environmental effects (50%-56%). Two categories of subjects have been proposed to be populations at high risk for the development of PD: relatives of subjects affected by PD [20] and subjects who have had almost a panic attack without developing the psychiatric syndrome [30].
869
Genetic and environmental variables are thought to be involved both in the etiology of PD and influencing the development and regulation of metacognitive abilities such as the access to one's own emotions and feelings and the interpretation of bodily sensations, contributing to a description of a liability spectrum for developing the clinical features of the disorder. Actually, no studies have systematically investigated the differences in these variables between subjects with PD, high-risk populations (first-degree relatives and subjects who have had 1 or more panic attacks without developing PD), and healthy subjects. We hypothesize that the distribution of alexithymia and AS among these categories of subjects could provide important information about their specific role in characterizing low- and high-risk populations. In particular, we would verify if first-degree relatives of subjects with PD and subjects who have had almost a panic attack without developing the psychiatric syndrome show higher levels of AS and alexithymia with respect to HCs. 2. Methods 2.1. Sample After the approval of the study protocol by the local ethics committee, 4 distinct samples were recruited for the study purposes. The clinical sample, composed of 139 outpatients affected by PD with agoraphobia, was consecutively recruited from the Unit of General Psychiatry of the Department of Clinical Neurosciences of the University VitaSalute, San Raffaele Hospital of Milan, using a subsequent admission method of recruitment. All subjects undertook an average of 1-hour assessment, including a diagnostic and anamnestic interview, and, afterward, another 1-hour session with the administration of the Beck Depression Inventory II (BDI-II) [31], the TAS-20 [32] (Italian version; [33]), and the Anxiety Sensitivity Index (ASI) [34]. Axis I and II diagnoses were defined according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition's (DSM-IV) criteria, using the Italian version of the Structured Clinical Interview for DSM-IV Axis I and II Disorders. All the interviews are administered by trained clinicians who are expert in anxiety disorders' diagnosis and treatment, whereas the BDI-II, the TAS-20, and the ASI are self-administered and followed by trainee psychologists. Patients with comorbid diagnoses were not excluded, provided that PD was the main problem for which treatment was sought. Exclusion criteria were brain injury or trauma, any neurologic condition, psychosis, and substance abuse. Patients with PD were recruited before starting therapy (which could have been pharmacologic therapy, CBT, CBT plus pharmacotherapy, and hospitalization). Sixty-three patients were under medication at the time of the assessment, 32 of whom were taking selective serotonin reuptake inhibitors, 15 benzodiazepine, and 16 selective
870
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
serotonin reuptake inhibitors plus benzodiazepine. Twentyfive Patients with PD have 1 or more comorbid Axis I or II disorders: 13 subjects had major depressive disorder; 5, generalized anxiety disorder; 4, social phobia; 8, narcissistic personality disorder; 1, histrionic personality; and 2, avoidant personality disorder. Thirty-seven patients reported a positive history of family psychiatric disorders (Table 1). In the same period, we asked an independent sample of 198 outpatients afferent to our clinical service and affected by PD with agoraphobia, selected on the basis of the same inclusion criteria described previously, about the availability of a first-degree relative for participating to the study. First-degree relatives who gave consent, in the same way of the clinical group, undertook an average of 1-hour assessment, including a diagnostic and anamnestic interview: Axis I and II diagnoses were defined according to DSM-IV 's criteria, using the Italian nonpatient version of the Structured Clinical Interview for DSM-IV Axis I and II Disorders. Exclusion criteria were brain injury or trauma, any neurologic condition, and any psychiatric condition both in present and lifelong use and present or past use of psychotropic drugs. After this preliminary selection, 36 subjects were excluded because they did not meet the inclusion criteria, whereas 64 of them were admitted to the study. These subjects undertook another 1-hour session with the administration of the BDI-II [31], the TAS-20 [32] (Italian version; [33]) and the ASI [34]. All the interviews are administered by trained clinicians, whereas the BDI-II, the TAS20, and the ASI are self-administered and followed by trainee psychologists. The rationale for this selection is based on the sequent reason: first-degree relatives share with patients with PD a genetic vulnerability to PD [20]; however, the presence of other psychiatric conditions or the use of psychotropic drugs could further differentiate this population and introduce confounding variables to take into account in the statistical analysis. One hundred eighty-seven healthy subjects constituted the third group: they were recruited among a larger sample from the general population, as volunteers. They were assessed with the same procedure of the previous groups: exclusion criteria were brain injury or trauma, any neurologic condition, and any psychiatric condition both in present and lifelong use and present or past use of psychotropic drugs.
Table 1 Demographic characteristics of the 4 samples Variables
HC (n = 157) PD (n = 139) R (n = 64)
PA (n = 30)
Sex (male) 73 (46.50%) 60 (43.80%) 27 (42.19%) 14 (46.67%) Age (y) 34.54 ± 14.30 37.58 ± 13.58 45.58 ± 15.97 30.90 ± 12.17 Years of 13.83 ± 3.29 13.29 ± 3.41 14.41 ± 3.85 14.87 ± 2.71 education
The fourth group was a subsample from the HC group: healthy subjects who have had almost 1 episode of panic attack but have never developed PD, according to DSM-IV criteria, were included in this group (PA; n = 30). The subjects in the PA group were selected on the basis of the screening questions for panic attack included in the nonpatient Structured Clinical Interview for DSM-IV Axis I and II Disorders: if any of the screening questions were endorsed, the participant was asked about the symptoms that were present during the worst panic attack that occurred “out of the blue.” If at least 4 panic attack symptoms were reported, the participant was asked if at least 4 of the symptoms began suddenly and became intense within 10 minutes. All participants who further met the criteria for PD or other psychiatric disorder encoded in DSM-IV were excluded from the PA group. All HC and subjects with PA undertook another 1-hour session with the administration of the BDI-II [31], the TAS-20 [32] (Italian version; [33]), and the ASI [34], using the same procedures described for the clinical sample. 2.2. Measurement 2.2.1. Beck Depression Inventory II The BDI-II [31] is a standard self-reported questionnaire containing 21 items that measure the presence and severity of cognitive and somatic symptoms of depression on a scale from 0 to 63, with higher scores indicating greater depression. Each item evaluates a category according to a scale of 4 possible responses of increasing severity. It has been validated in patients with diabetes and screens effectively for major depression in this population. Standard cutoffs were used in this study: 0 to 15 (no depression), 16 to 30 (mild), 31 to 46 (moderate), and 47 or more (severe). 2.2.2. Toronto Alexithymia Scale The 20-item Italian version of the TAS-20 is used to assess alexithymia [33]. It is a 20-item self-report instrument, rated on a 5-point Likert scale, providing a total score and scores for each of the 3 subscales, which are as follows: difficulty in identifying feelings, difficulty in describing feelings, and externally oriented thinking. Subjects with a total score of 51 or less are nonalexithymic, whereas subject scoring higher than 61 on the scale are alexithymic; the scores ranging from 52 and 60 indicate the presence of a borderline area [13]. 2.2.3. Anxiety Sensitivity Index The ASI [34] is a questionnaire comprising 16 items that tap concerns about anxiety-related sensations. Respondents indicate their strength of endorsement for each item on a 5-point scale that ranges from 0 (very little) to 4 (very much). A body of evidence supports the reliability and validity of the ASI as a measure of the fear of anxiety symptoms [35]. The National Institute of Mental Health's Committee on Standardized Assessment for Panic Disorder recommended the ASI as a clinical outcome [36].
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
871
η = 0.31) are evidenced between the 4 groups: in all cases, the strength of the association between predictors and dependent variables is large, as evidenced by an effect size higher than 0.14. The strongest association is observed between predictors and ASI total score (η 2 = 0.31), as expected on the basis of its specificity for PD vulnerability. After post hoc multiple comparisons, scores of TAS-20 factor 1 are lower in HC subjects compared with the PD group (P b .0001), as well as scores of TAS-20 factor 2 compared with female subjects in the R group (P b .01), scores of TAS-20 total score compared with female subjects in the R group (P b .01) and in the PD group (P b .001), and ASI scores compared with the PD group (P b .0001). Scores of TAS-20 factor 1 are lower in the R group compared with the PD group (P b .01), as well as ASI scores compared with the PD group (P = .001). Scores of TAS-20 factor 1 are higher in the PD group compared with all groups (P between b.0001 and .009), as well as TAS-20 total scores compared with HC subjects (P b .001) and female subjects in the PA group (P b .01) and ASI scores compared with all groups (P between b .0001 and .05). Hartley F-max and Cochran C tests confirmed homogeneity of variances; TAS-20 and ASI scores are all normally distributed. See descriptive statistics in Table 3. 2
2.3. Statistical analysis All statistical analyses are carried out with Statistica 7.0 software package for Windows (STATSOFT Inc, Tulsa, OK). Descriptive statistics of the demographic variables and the psychometric tests scores have been computed for the 4 samples. Differences in age and education between groups have been tested using t tests, whereas differences in gender distribution are explored using the χ 2 test. χ 2 Tests are also used to detect differences in the distribution of alexithymia categories in the 4 groups. Differences in TAS-20 scales and ASI total score between groups are explored in an analysis of covariance design, controlling for depression (BDI), sex, age, and education effects. Post hoc multiple comparisons have been computed with Fisher least significant difference test. 3. Results 3.1. Descriptive statistics There are no significant differences between groups in terms of sex (χ 23 = 0.46, P = .93). The relatives (R) group shows higher age and less years of education compared with all the other groups (P b .0001), whereas subjects with PD show higher age (t167 = 2.48, P = .014) and less years of education (t167 = −2.36, P = .019) compared with subjects in the PA group.
4. Discussion
3.2. Alexithymia distributions Distribution of subjects with PD among the 3 TAS-20 categories—alexithymic, nonalexithymic, and borderline— differs significantly from all the other group distributions (P b .01): the percentage of alexithymic patients in the PD group is 25.90%, whereas in the other groups, it is largely smaller. Moreover, there are more alexithymic patients in first-degree relatives of subjects with PD than in HC subjects (χ 22 = 7.41, P = .02), whereas the remaining subjects in the PA group are comparable with both R and HC groups. For all frequencies and percentages of TAS-20 categories among the 4 samples, see Table 2. 3.3. Group comparisons Controlling for depression (BDI), sex, age, and education, significant differences in TAS-20 factor 1 (F3, 352 = 18.51, P b .0001, η 2 = 0.21) and total (F3, 352 = 7.77, P b .0001, η 2 = 0.17) and ASI total score (F3, 352 = 33.70, P b .0001, Table 2 Alexithymia categories distribution among the 4 samples
Nonalexithymic Border Alexithymic
HC (n = 157)
PD (n = 139)
R (n = 64)
PA (n = 30)
%
n
%
n
%
n
%
n
70.70 23.57 5.73
111 37 9
43.17 30.94 25.90
60 43 36
67.19 15.63 17.19
43 10 11
70.00 23.33 6.67
21 7 2
Cumulative and specific life events such as threat, interpersonal and health-related events in adulthood, and abuse or loss/separation experiences in childhood could be hypothesized to be influencing the pathogenesis and the maintenance of PD with some overlapping effect across the anxiety disorder spectra [26]. Besides genetic vulnerability factors, personality and behavioral characteristics such as AS, alexithymia, and cognitive appraisal might moderate the influence of these life events on the development of PD [29,37]. To identify the role of cognitive and emotional processes in characterizing populations at high risk for PD, we conduced a study in which a control group and a PD group were compared with first-degree relatives of subjects affected by PD and to subjects who have a history of sporadic panic attacks, without having developed the specific psychiatric disorder. On the basis of literature data, we expected an alexithymic deficit in subjects with PD [17] and a higher AS level [38] as compared with HC subjects. We verified the alexithymic deficit in our sample both in a categorical and a dimensional manner: as reported in the “Results” section, the distribution of subjects with PD among the 3 TAS-20 categories differs significantly from HC subject distribution, with the percentage of alexithymic patients being 25.90% in the PD group 5.73% in the HC group. Moreover, the 2 groups show significant differences in TAS-20 identifying feelings factor and total score, with
872
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
Table 3 Means and standard deviations of all the considered psychometric measures among the 4 groups
TAS-20
ASI
Variable
HC (n = 157)
PD (n = 139)
R (n = 64)
PA (n = 30)
F1 F2 F3 Total Total
13.72 ± 5.52 12.28 ± 4.39 17.21 ± 4.49 43.21 ± 10.72 16.83 ± 9.39
20.30 ± 6.25 14.09 ± 4.82 17.96 ± 4.23 52.36 ± 11.15 30.22 ± 10.39
15.02 ± 5.69 14.47 ± 4.83 18.84 ± 4.23 48.32 ± 11.31 18.10 ± 9.51
14.17 ± 6.73 12.30 ± 4.49 17.10 ± 4.19 43.57 ± 11.75 18.97 ± 9.14
Significant differences between the 4 groups are set in italics (P b .001). F1 indicates difficulty in identifying feelings; F2, difficulty in describing feelings; F3, externally oriented thinking.
subjects with PD having higher levels of alexithymic characteristics than the control subjects. As expected, AS was significantly higher in the PD group, replicating previous literature data. For example, Cox [39] found that patients with PD (n = 216) had a higher ASI score (mean ± SD, 37.44 ± 12.24) than a sample of college students (n = 365; mean ± SD, 21.85 ± 10.14), with t579 = 16.55, P b .001. As hypothesized, in our sample, PD relatives are more alexithymic than control subjects both in a categorical (percentage of alexithymic patients) and a dimensional manner (on the communicating emotions factor and the total score). These data are in accordance with previous studies, in which this construct has demonstrated heritabilities of 30% to 33% [29] and to be influenced by familial and genetic factors [27,28]. However, in our sample, this difference regards only female relatives of subjects with PD. In population studies, no significant differences between male and female subjects have been reported [40], and in the present work, these data are replicated. Nevertheless, epidemiologic data concerning PD [41] reveal an association between this diagnosis and sex, with a higher risk for women (female-tomale ratio ≈2:1). It is possible that being a female constitutes a predisposition not only to PD but also to PD information processing features, including alexithymia and AS: for this reason, female relatives of subjects with PD could manifest more evident differences from HC subjects on these relevant variables, as revealed in our sample. Regarding AS, HC and R groups do not differ in this variable between each other: they show a lower level of AS than that in subjects with PD. Compared with the literature, this result appears controversial. van Beek and Griez [25], for example, found that first-degree relatives of subjects with PD were significantly more anxiety sensitive than HC subjects, but less anxiety sensitive than patients with PD. In another study, van Beek et al [42] demonstrated that children of PD parents are not more anxiety sensitive than children of healthy parents and did not find correlations between AS scores of PD parents and their offspring, supposing that AS could be considered a developing vulnerability factor that only becomes important toward late adolescence or early adulthood. However, in the first cited study, the samples were relatively small (23 healthy first-degree relatives of patients with PD, 30 HCs, and 38 patients with PD), control variables could not be taken into account because of the statistical
analysis conduced, and the scores of their first-degree relatives were only moderately increased, although significantly higher than those of HCs. Although the present study was conduced on larger samples, using more complete analyses, further research is needed to confirm these results, for example, using adequate behavioral genetic methods on very large samples. Probably, as reported elsewhere [43], AS has more complex etiology than previously recognized, which could be comprehended only through a mix of dimension-specific and nonspecific etiologic factors. In the present study, the lack of statistical difference in alexithymia between the HC and the PA groups, which is contrary to our expectations, could be explained in different ways. For example, the PA sample size (n = 30) could be insufficient to detect significant differences: in this case, it could be reasonable to replicate this study by simply increasing the sample size. Moreover, we have not taken into account a potentially important variable: the spontaneous vs situational nature of the panic attacks experienced by the PA group. It is possible that the processes involved in these 2 different kinds of panic attacks are specific and partially independent from each other and that different biological, genetic, cognitive, and emotional mechanisms play a role in their manifestation: if subjects belonging to 1 of these 2 categories share a different amount of genetic or environmental vulnerability to PD, they could also share a different way of emotional information processing with subjects with PD. One more explanation has been considered: if some people experience sporadic panic attacks but do not develop PD, we could suppose that these subjects have some protective factors that prevent them from triggering anticipatory anxiety and dysfunctional cognitive appraisal of bodily sensations; the ability to correctly read internal states and bodily sensations is a good candidate to hold this role. A first limit of the present study is represented by the small sample size of the PA group compared with the other groups. This limitation has been, nonetheless, considered only partially influential because homogeneity of variances and normality of variables' distributions have been verified for all analyses. The second limit concerns the scale for measuring alexithymia (TAS-20): it is a self-report instrument that could be affected by nonidentifiable bias depending on variables such as personality traits or social desirability. Although this instrument has demonstrated good psychometric properties
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
and has been validated in different populations (clinical and nonclinical) and languages, it could be useful, as reported by other authors [44], to use, in future studies, multiple methods that eliminate the potential influence of measurement method-based response biases associated with monomethod assessment. A third limit, already partially discussed, could be identified in the characteristics of the PA group: it is possible that differences in the nature of panic attacks (situational vs spontaneous) have not been sufficiently considered in the analysis and that the exclusive or mixed presence of these 2 kinds of panic attacks not only has reference to a phenomenologic description of the disorder but also is related to biological, cognitive, and emotional mechanisms involved in its etiopathogenesis and clinical course. Fourth, the R group has an elevated average age (mean ± SD, 45.58 ± 15.97 years): it is well above the peak age of onset for PD (ie, 25 years [45]). Consequently, it may be argued that only those with an ASI score in the high range are at risk for PD and share with them, differently from HC subjects, a different amount of genetic or environmental vulnerability to PD. Future research could test this consideration, selecting a group of relatives of subjects with PD with different age distributions. A further limit to consider concerns the assessment methodology. In the 4 samples, we measured AS and alexithymia using a single self-administered instrument, the ASI (16-item version) and the TAS-20, respectively. Although these 2 psychological scales show good reliability and validity characteristics, as reported in the “Methods” section, further studies should comprehend multimethod assessment procedures. Last, the clinical group used in the present study consisted exclusively of patients diagnosed as having PD with agoraphobia. Consequently, the results cannot be generalized to patients affected by PD without agoraphobia. Despite the limitations considered above, the results of the present study are encouraging to explore, with more accurate techniques of assessment, larger sample sizes, and genetic studies, how these variables play a role not only in differentiating groups with a different level of risk for developing PD but also in influencing the clinical course of the disorder, its onset, and general population liability to this clinical condition. Acknowledgment We thank all the patients who participated in this study. References [1] Barlow DH, Craske MG. The phenomenology of panic. In: Rachman S, & Maser JD, editors. Panic: psychological perspectives. Hillsdale, NJ: Erlbaum; 1988. p. 11-35. [2] Clark DM. A cognitive approach to panic. Behav Res Ther 1986;24(4): 461-70.
873
[3] Reiss S, McNally R. In: & Reiss SBR, editor. Theoretical issues in behavior therapy. San Diego: Academic Press; 1985. [4] Reiss S. Expectancy model of fear, anxiety, and panic. Clin Psychol Rev 1991;11(2):141-53. [5] White KS, Brown TA, Somers TJ, Barlow DH. Avoidance behavior in panic disorder: the moderating influence of perceived control. Behav Res Ther 2006;44(1):147-57. [6] Wilson KA, Hayward C. Unique contributions of anxiety sensitivity to avoidance: a prospective study in adolescents. Behav Res Ther 2006;44(4):601-9. [7] Zvolensky MJ, Kotov R, Antipova AV, Schmidt NB. Cross cultural evaluation of smokers risk for panic and anxiety pathology: a test in a Russian epidemiological sample. Behav Res Ther 2003;41(10): 1199-215. [8] Schmidt NB, Koselka M. Gender differences in patients with panic disorder: evaluating cognitive mediation of phobic avoidance. Cognitive Ther Res 2000;24(5):533. [9] Naragon-Gainey K. Meta-analysis of the relations of anxiety sensitivity to the depressive and anxiety disorders. Psychol Bull 2010; 136(1):128-50. [10] Zvolensky M, Bernstein A, Vujanovic A. The interactive effects of anxiety sensitivity and emotion dysregulation in predicting anxietyrelated cognitive and affective symptoms. Cognitive Ther Res 2008; 32(6):803-17. [11] Benítez CI, Shea MT, Raffa S, Rende R, Dyck IR, Ramsawh HJ, et al. Anxiety sensitivity as a predictor of the clinical course of panic disorder: a 1-year follow-up study. Depress Anxiety 2009;26(4): 335-42. [12] Nemiah JC, Freyberger H, Sifneos PE. Alexithymia: a view of the psychosomatic process. In: & Hill OW, editor. Modem trends in psychosomatic medicine, Vol. 3. London: Butterworths; 1976. p. 430-9. [13] Taylor GJ, Bagby RM, Parker JDA. Disorders of affect regulation: alexithymia in medical and psychiatric illness. Cambridge: Cambridge University Press; 1997. [14] Galderisi S, Mancuso F, Mucci A, Garramone S, Zamboli R, Maj M. Alexithymia and cognitive dysfunctions in patients with panic disorder. Psychother Psychosom 2008;77(3):182-8. [15] Fukunishi I, Kikuchi M, Wogan J, Takubo M. Secondary alexithymia as a state reaction in panic disorder and social phobia. Compr Psychiatry 1997;38(3):166-70. [16] Bankier B, Aigner M, Bach M. Alexithymia in DSM-IV disorder: comparative evaluation of somatoform disorder, panic disorder, obsessive-compulsive disorder, and depression. Psychosomatics 2001;42(3):235-40. [17] De Berardis D, Campanella D, Gambi F, La Rovere R, Sepede G, Core L, et al. Alexithymia, fear of bodily sensations, and somatosensory amplification in young outpatients with panic disorder. Psychosomatics: Journal of Consultation Liaison Psychiatry 2007;48(3):239-46. [18] Marchesi C, Fonto S, Balista C, Cimmino C, Maggini C. Relationship between alexithymia and panic disorder: a longitudinal study to answer an open question. Psychother Psychosom 2005;74(1):56-60. [19] Klauke B, Deckert J, Reif A, Pauli P, Domschke K. Life events in panic disorder—an update on “candidate stressors.”. Depress Anxiety 2010;27(8):716-30. [20] Hettema JM, Neale MC, Kendler KS. A review and meta-analysis of the genetic epidemiology of anxiety disorders. Am J Psychiatry 2001; 158(10):1568-78. [21] Goldstein RB, Wickramaratne PJ, Horwath E, Weissman MM. Familial aggregation and phenomenology of 'early'-onset (at or before age 20 years) panic disorder. Arch Gen Psychiatry 1997;54(3): 271-8. [22] Stewart S. Heartbeat awareness and heart rate reactivity in anxiety sensitivity: a further investigation. J Anxiety Disord 2001;15(6): 535-53. [23] Stein MB, Jang KL, Livesley WJ. Heritability of anxiety sensitivity: a twin study. Am J Psychiatry 1999;156(2):246-51.
874
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
[24] Scher CD, Stein MB. Developmental antecedents of anxiety sensitivity. J Anxiety Disord 2003;17(3):253-69. [25] van Beek N, Griez E. Anxiety sensitivity in first-degree relatives of patients with panic disorder. Behav Res Ther 2003;41(8):949-57. [26] Stein MB, Schork NJ, Gelernter J. Gene-by-environment (serotonin transporter and childhood maltreatment) interaction for anxiety sensitivity, an intermediate phenotype for anxiety disorders. Neuropsychopharmacology: official publication of the American College of Neuropsychopharmacology 2008;33(2):312-9. [27] Ham B, Lee M, Lee Y, Kim M, Choi M, Oh K, et al. Association between the catechol O-methyltransferase Val108/158Met polymorphism and alexithymia. Neuropsychobiology 2005;52(3):151-4. [28] Valera EM, Berenbaum H. A twin study of alexithymia. Psychother Psychosom 2001;70(5):239-46. [29] Jørgensen MM, Zachariae R, Skytthe A, Kyvik K. Genetic and environmental factors in alexithymia: a population-based study of 8,785 Danish twin pairs. Psychother Psychosom 2007;76(6):369-75. [30] Goracci A, Fagiolini A, Salviulo C, Forgione RN, Martinucci M, Castroglovanni P. Panic agoraphobic spectrum in psychiatrically healthy subjects: impact on quality of life. Epidemiol Psichiatr Soc 2009;18(1):65-8. [31] Beck AT, Steer RA, Brown GK. Manual for the Beck Depression Inventory-II. San Antonio (Tex): Psychological Corporation; 1996. [32] Bagby RM, Parker JDA, Taylor GJ. The twenty-item Toronto Alexithymia Scale: I. Item selection and cross-validation of the factor structure. J Psychosom Res 1994;38(1):23-32. [33] Bressi C, Taylor G, Parker J, Bressi S, Brambilla V, Aguglia E, et al. Cross validation of the factor structure of the 20-item Toronto Alexithymia Scale: an Italian multicenter study. J Psychosom Res 1996;41(6):551-9. [34] Reiss S, Peterson RA, Gursky DM, McNally RJ. Anxiety sensitivity, anxiety frequency and the predictions of fearfulness. Behav Res Ther 1986;24(1):1-8.
[35] Zinbarg RE, Mohlman J, Hong NN. Dimensions of anxiety sensitivity. In: & Taylor S, editor. Anxiety sensitivity: theory, research, and treatment of the fear of anxiety. Mahwah (NJ): Lawrence Erlbaum; 1999. p. 83-114. [36] McNally RJ. Anxiety sensitivity and panic disorder. Biol Psychiatry 2002;52(10):938-46. [37] Taylor S, Jang KL, Stewart SH, Stein MB. Etiology of the dimensions of anxiety sensitivity: a behavioral-genetic analysis. J Anxiety Disord 2008;22(5):899-914. [38] Schmidt NB, Mitchell MA, Richey JA. Anxiety sensitivity as an incremental predictor of later anxiety symptoms and syndromes. Compr Psychiatry 2008;49(4):407-12. [39] Cox BJ. The nature and assessment of catastrophic thoughts in panic disorder. Behav Res Ther 1996;34(4):363-74. [40] Franz M, Popp K, Schaefer R, Sitte W, Schneider C, Hardt J, et al. Alexithymia in the german general population. Soc Psychiatry Psychiatr Epidemiol 2008;43(1):54-62. [41] Grant BF, Hasin DS, Stinson FS, Dawson DA, Goldstein RB, Smith S, et al. The epidemiology of DSM-IV panic disorder and agoraphobia in the united states: results from the national epidemiologic survey on alcohol and related conditions. J Clin Psychiatry 2006; 67(3):363-74. [42] van Beek N, Perna G, Schruers K, Muris P, Griez E. Anxiety sensitivity in children of panic disorder patients. Child Psychiatry Hum Dev 2005;35(4):315-24. [43] Taylor S, Jang KL, Stewart SH, Stein MB. Etiology of the dimensions of anxiety sensitivity: a behavioral-genetic analysis. J Anxiety Disord 2008;22(5):899-914. [44] Taylor GJ, Bagby RM. New trends in alexithymia research. Psychother Psychosom 2004;73(2):68-77. [45] American Psychiatric Association. Diagnostic and statistical manual of mental disorders, fourth edition, text revision (DSM-IV-TR). Washington (DC): American Psychiatric Association; 2000.
Comprehensive Psychiatry 53 (2012) 868 – 874 www.elsevier.com/locate/comppsych
Alexithymia and anxiety sensitivity in populations at high risk for panic disorder Michele Cucchi⁎, Daniele Cavadini, Vittoria Bottelli, Liana Riccia, Vera Conca, Paolo Ronchi, Enrico Smeraldi Department of Clinical Neurosciences, Scientific Institute and University, Vita-Salute San Raffaele, Via Stamira D'Ancona 20, 20127 Milan, Italy
Abstract Objective: Populations at high risk for panic disorder (PD) probably share with subjects with PD an underlying vulnerability involving features like anxiety sensitivity (AS) and alexithymia. The present study would verify if PD relatives (R) and subjects who have experienced 1 or more panic attacks (PAs) show different levels of AS and alexithymia with respect to healthy controls (HC). Methods: One hundred fifty-seven HCs, 30 subjects with PA, 64 R subjects, and 139 outpatients with PD were evaluated and compared on AS, alexithymia, and control variables. Results: Subjects with PD show higher alexithymia and AS levels compared with HCs; R subjects do not differ on ASI total score; and R females show more alexithymic features. Subjects with PA are comparable with HCs both on AS and alexithymia. Conclusions: Results confirm an impairment in emotional and bodily sensations information processing in subjects with PD but partially disconfirm the expectation of a difference between R subjects and subjects with PA with respect to HCs on AS and alexithymia. Emotional and bodily sensation competencies could be protective factors for PD in high-risk populations. © 2012 Elsevier Inc. All rights reserved.
1. Introduction Several authors have pointed out that in anxiety disorders, in particular panic disorder (PD), amplification of somatic sensations is often associated to subsequent dysfunctional cognitive appraisal of these sensations with a significant bias toward a danger-related and catastrophizing interpretational style [1,2]. The construct of anxiety sensitivity (AS) [3] refers to the tendency to fear anxiety-related sensations. According to Reiss [4], individuals with elevated AS experience amplified fear in response to stimuli that elicit anxiety and find their own anxiety symptoms to be particularly aversive. Many studies have found large associations between AS and PD features [5-8] (for a meta-analysis, see Naragon-Gainey [9]). Some authors evidenced that the way that individuals appraise anxiety- and panic-related symptoms may increase the risk of developing agoraphobia, over and above the risk of developing PD [10]. Prospective studies have shown that AS predicts the onset of anxiety and panic in both nonclinical and clinical populations [11]. ⁎ Corresponding author. Tel.: +39 02 26433229; fax: +39 02 26433265. E-mail address: [email protected] (M. Cucchi). 0010-440X/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.comppsych.2012.01.005
A second interesting psychological construct pertaining to bodily sensations (related to emotion-related activation) information processing is alexithymia. The alexithymia construct, as defined by Nemiah et al [12], has 3 salient features: (1) difficulty identifying and describing subjective feelings, (2) difficulty distinguishing between feelings and the bodily sensations of emotional arousal, and (3) constricted imaginal capacities, as evidenced by paucity of fantasies and externally oriented cognitive style. The alexithymia construct reflect deficits in the cognitive processing and regulation of emotions [13]. As reported by Galderisi and colleagues [14], empirical evidence of poor emotion processing in subjects with PD has been provided, including a high prevalence of alexithymia. Regarding PD, the psychological construct of alexithymia has been extensively studied from different points of view [15-18]: overall results proved an alexithymic deficit in subjects with PD. In particular, the first feature of alexithymia seems to distinguish healthy control (HC) subjects from subjects with PD and to predict symptom severity. However, as evidenced in the study by Marchesi et al [18], the possibility of a trait or state interpretation of this construct remains controversial.
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
Converging lines of evidence suggest that PD in adulthood may represent manifestations of an underlying constitutional vulnerability or diathesis for anxiety that is partly genetic and variably expressed during a life cycle [19]. It is well known that PD runs in families [20,21]. Individual differences in AS and alexithymia are hypothesized to emerge from both genetic and maladaptive learning experiences that lead to the acquisition of beliefs about the aversive consequences of arousal and anxietyrelated states [22]. A recent study by Stein and colleagues, examining the heritability of AS in a group of 179 monozygotic and 158 dizygotic twin pairs, suggests that variation in AS (particularly the physical aspect of AS) is explained by unique environmental and additive genetic influences [23]. Emotional maltreatment in childhood, particularly when it includes parental threatening, hostile, and rejecting behaviors, is associated with higher levels of AS in young adults [24]. In a familial study, van Beek and Griez [25] found that the first-degree relatives of subjects with PD were significantly more anxiety sensitive than HC subjects, but less anxiety sensitive than patients with PD, suggesting that AS runs in families. Moreover, Stein et al [26] observed a statistically significant interaction between levels of childhood emotional (or physical) maltreatment and 5-HTTLPR genotype. Specifically, S/S individuals with higher levels of maltreatment had significantly higher levels of AS than subjects in other groups, evidencing a gene-by-environment (serotonin transporter and childhood maltreatment) interaction process influencing AS. In the same way, several authors studied genetic and environmental factors associated with alexithymia. For example, Ham et al [27] showed an association between the Val/Val COMT genotype and higher Toronto Alexithymia Scale (20-item version; TAS-20) scores in a sample of 109 students. Valera and Berenbaum [28], in a study involving 45 monozygotic and 32 dizygotic twin pairs, concluded that familial influences contribute to alexithymia, with a specific effect of environmental factors on difficulty in identifying feelings and communicating emotions and a specific effect of shared genetic factors on externally oriented thinking. In a larger sample composed of 8785 twin pairs, Jørgensen et al [29], using a structural equation modeling, selected an ACE model including additive genetic, shared environmental, and nonshared environmental effects providing the best fit for all 3 facets of alexithymia as well as total alexithymia scores, with heritabilities of 30% to 33% and the remaining variance being explained by shared (12%-20%) and nonshared environmental effects (50%-56%). Two categories of subjects have been proposed to be populations at high risk for the development of PD: relatives of subjects affected by PD [20] and subjects who have had almost a panic attack without developing the psychiatric syndrome [30].
869
Genetic and environmental variables are thought to be involved both in the etiology of PD and influencing the development and regulation of metacognitive abilities such as the access to one's own emotions and feelings and the interpretation of bodily sensations, contributing to a description of a liability spectrum for developing the clinical features of the disorder. Actually, no studies have systematically investigated the differences in these variables between subjects with PD, high-risk populations (first-degree relatives and subjects who have had 1 or more panic attacks without developing PD), and healthy subjects. We hypothesize that the distribution of alexithymia and AS among these categories of subjects could provide important information about their specific role in characterizing low- and high-risk populations. In particular, we would verify if first-degree relatives of subjects with PD and subjects who have had almost a panic attack without developing the psychiatric syndrome show higher levels of AS and alexithymia with respect to HCs. 2. Methods 2.1. Sample After the approval of the study protocol by the local ethics committee, 4 distinct samples were recruited for the study purposes. The clinical sample, composed of 139 outpatients affected by PD with agoraphobia, was consecutively recruited from the Unit of General Psychiatry of the Department of Clinical Neurosciences of the University VitaSalute, San Raffaele Hospital of Milan, using a subsequent admission method of recruitment. All subjects undertook an average of 1-hour assessment, including a diagnostic and anamnestic interview, and, afterward, another 1-hour session with the administration of the Beck Depression Inventory II (BDI-II) [31], the TAS-20 [32] (Italian version; [33]), and the Anxiety Sensitivity Index (ASI) [34]. Axis I and II diagnoses were defined according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition's (DSM-IV) criteria, using the Italian version of the Structured Clinical Interview for DSM-IV Axis I and II Disorders. All the interviews are administered by trained clinicians who are expert in anxiety disorders' diagnosis and treatment, whereas the BDI-II, the TAS-20, and the ASI are self-administered and followed by trainee psychologists. Patients with comorbid diagnoses were not excluded, provided that PD was the main problem for which treatment was sought. Exclusion criteria were brain injury or trauma, any neurologic condition, psychosis, and substance abuse. Patients with PD were recruited before starting therapy (which could have been pharmacologic therapy, CBT, CBT plus pharmacotherapy, and hospitalization). Sixty-three patients were under medication at the time of the assessment, 32 of whom were taking selective serotonin reuptake inhibitors, 15 benzodiazepine, and 16 selective
870
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
serotonin reuptake inhibitors plus benzodiazepine. Twentyfive Patients with PD have 1 or more comorbid Axis I or II disorders: 13 subjects had major depressive disorder; 5, generalized anxiety disorder; 4, social phobia; 8, narcissistic personality disorder; 1, histrionic personality; and 2, avoidant personality disorder. Thirty-seven patients reported a positive history of family psychiatric disorders (Table 1). In the same period, we asked an independent sample of 198 outpatients afferent to our clinical service and affected by PD with agoraphobia, selected on the basis of the same inclusion criteria described previously, about the availability of a first-degree relative for participating to the study. First-degree relatives who gave consent, in the same way of the clinical group, undertook an average of 1-hour assessment, including a diagnostic and anamnestic interview: Axis I and II diagnoses were defined according to DSM-IV 's criteria, using the Italian nonpatient version of the Structured Clinical Interview for DSM-IV Axis I and II Disorders. Exclusion criteria were brain injury or trauma, any neurologic condition, and any psychiatric condition both in present and lifelong use and present or past use of psychotropic drugs. After this preliminary selection, 36 subjects were excluded because they did not meet the inclusion criteria, whereas 64 of them were admitted to the study. These subjects undertook another 1-hour session with the administration of the BDI-II [31], the TAS-20 [32] (Italian version; [33]) and the ASI [34]. All the interviews are administered by trained clinicians, whereas the BDI-II, the TAS20, and the ASI are self-administered and followed by trainee psychologists. The rationale for this selection is based on the sequent reason: first-degree relatives share with patients with PD a genetic vulnerability to PD [20]; however, the presence of other psychiatric conditions or the use of psychotropic drugs could further differentiate this population and introduce confounding variables to take into account in the statistical analysis. One hundred eighty-seven healthy subjects constituted the third group: they were recruited among a larger sample from the general population, as volunteers. They were assessed with the same procedure of the previous groups: exclusion criteria were brain injury or trauma, any neurologic condition, and any psychiatric condition both in present and lifelong use and present or past use of psychotropic drugs.
Table 1 Demographic characteristics of the 4 samples Variables
HC (n = 157) PD (n = 139) R (n = 64)
PA (n = 30)
Sex (male) 73 (46.50%) 60 (43.80%) 27 (42.19%) 14 (46.67%) Age (y) 34.54 ± 14.30 37.58 ± 13.58 45.58 ± 15.97 30.90 ± 12.17 Years of 13.83 ± 3.29 13.29 ± 3.41 14.41 ± 3.85 14.87 ± 2.71 education
The fourth group was a subsample from the HC group: healthy subjects who have had almost 1 episode of panic attack but have never developed PD, according to DSM-IV criteria, were included in this group (PA; n = 30). The subjects in the PA group were selected on the basis of the screening questions for panic attack included in the nonpatient Structured Clinical Interview for DSM-IV Axis I and II Disorders: if any of the screening questions were endorsed, the participant was asked about the symptoms that were present during the worst panic attack that occurred “out of the blue.” If at least 4 panic attack symptoms were reported, the participant was asked if at least 4 of the symptoms began suddenly and became intense within 10 minutes. All participants who further met the criteria for PD or other psychiatric disorder encoded in DSM-IV were excluded from the PA group. All HC and subjects with PA undertook another 1-hour session with the administration of the BDI-II [31], the TAS-20 [32] (Italian version; [33]), and the ASI [34], using the same procedures described for the clinical sample. 2.2. Measurement 2.2.1. Beck Depression Inventory II The BDI-II [31] is a standard self-reported questionnaire containing 21 items that measure the presence and severity of cognitive and somatic symptoms of depression on a scale from 0 to 63, with higher scores indicating greater depression. Each item evaluates a category according to a scale of 4 possible responses of increasing severity. It has been validated in patients with diabetes and screens effectively for major depression in this population. Standard cutoffs were used in this study: 0 to 15 (no depression), 16 to 30 (mild), 31 to 46 (moderate), and 47 or more (severe). 2.2.2. Toronto Alexithymia Scale The 20-item Italian version of the TAS-20 is used to assess alexithymia [33]. It is a 20-item self-report instrument, rated on a 5-point Likert scale, providing a total score and scores for each of the 3 subscales, which are as follows: difficulty in identifying feelings, difficulty in describing feelings, and externally oriented thinking. Subjects with a total score of 51 or less are nonalexithymic, whereas subject scoring higher than 61 on the scale are alexithymic; the scores ranging from 52 and 60 indicate the presence of a borderline area [13]. 2.2.3. Anxiety Sensitivity Index The ASI [34] is a questionnaire comprising 16 items that tap concerns about anxiety-related sensations. Respondents indicate their strength of endorsement for each item on a 5-point scale that ranges from 0 (very little) to 4 (very much). A body of evidence supports the reliability and validity of the ASI as a measure of the fear of anxiety symptoms [35]. The National Institute of Mental Health's Committee on Standardized Assessment for Panic Disorder recommended the ASI as a clinical outcome [36].
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
871
η = 0.31) are evidenced between the 4 groups: in all cases, the strength of the association between predictors and dependent variables is large, as evidenced by an effect size higher than 0.14. The strongest association is observed between predictors and ASI total score (η 2 = 0.31), as expected on the basis of its specificity for PD vulnerability. After post hoc multiple comparisons, scores of TAS-20 factor 1 are lower in HC subjects compared with the PD group (P b .0001), as well as scores of TAS-20 factor 2 compared with female subjects in the R group (P b .01), scores of TAS-20 total score compared with female subjects in the R group (P b .01) and in the PD group (P b .001), and ASI scores compared with the PD group (P b .0001). Scores of TAS-20 factor 1 are lower in the R group compared with the PD group (P b .01), as well as ASI scores compared with the PD group (P = .001). Scores of TAS-20 factor 1 are higher in the PD group compared with all groups (P between b.0001 and .009), as well as TAS-20 total scores compared with HC subjects (P b .001) and female subjects in the PA group (P b .01) and ASI scores compared with all groups (P between b .0001 and .05). Hartley F-max and Cochran C tests confirmed homogeneity of variances; TAS-20 and ASI scores are all normally distributed. See descriptive statistics in Table 3. 2
2.3. Statistical analysis All statistical analyses are carried out with Statistica 7.0 software package for Windows (STATSOFT Inc, Tulsa, OK). Descriptive statistics of the demographic variables and the psychometric tests scores have been computed for the 4 samples. Differences in age and education between groups have been tested using t tests, whereas differences in gender distribution are explored using the χ 2 test. χ 2 Tests are also used to detect differences in the distribution of alexithymia categories in the 4 groups. Differences in TAS-20 scales and ASI total score between groups are explored in an analysis of covariance design, controlling for depression (BDI), sex, age, and education effects. Post hoc multiple comparisons have been computed with Fisher least significant difference test. 3. Results 3.1. Descriptive statistics There are no significant differences between groups in terms of sex (χ 23 = 0.46, P = .93). The relatives (R) group shows higher age and less years of education compared with all the other groups (P b .0001), whereas subjects with PD show higher age (t167 = 2.48, P = .014) and less years of education (t167 = −2.36, P = .019) compared with subjects in the PA group.
4. Discussion
3.2. Alexithymia distributions Distribution of subjects with PD among the 3 TAS-20 categories—alexithymic, nonalexithymic, and borderline— differs significantly from all the other group distributions (P b .01): the percentage of alexithymic patients in the PD group is 25.90%, whereas in the other groups, it is largely smaller. Moreover, there are more alexithymic patients in first-degree relatives of subjects with PD than in HC subjects (χ 22 = 7.41, P = .02), whereas the remaining subjects in the PA group are comparable with both R and HC groups. For all frequencies and percentages of TAS-20 categories among the 4 samples, see Table 2. 3.3. Group comparisons Controlling for depression (BDI), sex, age, and education, significant differences in TAS-20 factor 1 (F3, 352 = 18.51, P b .0001, η 2 = 0.21) and total (F3, 352 = 7.77, P b .0001, η 2 = 0.17) and ASI total score (F3, 352 = 33.70, P b .0001, Table 2 Alexithymia categories distribution among the 4 samples
Nonalexithymic Border Alexithymic
HC (n = 157)
PD (n = 139)
R (n = 64)
PA (n = 30)
%
n
%
n
%
n
%
n
70.70 23.57 5.73
111 37 9
43.17 30.94 25.90
60 43 36
67.19 15.63 17.19
43 10 11
70.00 23.33 6.67
21 7 2
Cumulative and specific life events such as threat, interpersonal and health-related events in adulthood, and abuse or loss/separation experiences in childhood could be hypothesized to be influencing the pathogenesis and the maintenance of PD with some overlapping effect across the anxiety disorder spectra [26]. Besides genetic vulnerability factors, personality and behavioral characteristics such as AS, alexithymia, and cognitive appraisal might moderate the influence of these life events on the development of PD [29,37]. To identify the role of cognitive and emotional processes in characterizing populations at high risk for PD, we conduced a study in which a control group and a PD group were compared with first-degree relatives of subjects affected by PD and to subjects who have a history of sporadic panic attacks, without having developed the specific psychiatric disorder. On the basis of literature data, we expected an alexithymic deficit in subjects with PD [17] and a higher AS level [38] as compared with HC subjects. We verified the alexithymic deficit in our sample both in a categorical and a dimensional manner: as reported in the “Results” section, the distribution of subjects with PD among the 3 TAS-20 categories differs significantly from HC subject distribution, with the percentage of alexithymic patients being 25.90% in the PD group 5.73% in the HC group. Moreover, the 2 groups show significant differences in TAS-20 identifying feelings factor and total score, with
872
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
Table 3 Means and standard deviations of all the considered psychometric measures among the 4 groups
TAS-20
ASI
Variable
HC (n = 157)
PD (n = 139)
R (n = 64)
PA (n = 30)
F1 F2 F3 Total Total
13.72 ± 5.52 12.28 ± 4.39 17.21 ± 4.49 43.21 ± 10.72 16.83 ± 9.39
20.30 ± 6.25 14.09 ± 4.82 17.96 ± 4.23 52.36 ± 11.15 30.22 ± 10.39
15.02 ± 5.69 14.47 ± 4.83 18.84 ± 4.23 48.32 ± 11.31 18.10 ± 9.51
14.17 ± 6.73 12.30 ± 4.49 17.10 ± 4.19 43.57 ± 11.75 18.97 ± 9.14
Significant differences between the 4 groups are set in italics (P b .001). F1 indicates difficulty in identifying feelings; F2, difficulty in describing feelings; F3, externally oriented thinking.
subjects with PD having higher levels of alexithymic characteristics than the control subjects. As expected, AS was significantly higher in the PD group, replicating previous literature data. For example, Cox [39] found that patients with PD (n = 216) had a higher ASI score (mean ± SD, 37.44 ± 12.24) than a sample of college students (n = 365; mean ± SD, 21.85 ± 10.14), with t579 = 16.55, P b .001. As hypothesized, in our sample, PD relatives are more alexithymic than control subjects both in a categorical (percentage of alexithymic patients) and a dimensional manner (on the communicating emotions factor and the total score). These data are in accordance with previous studies, in which this construct has demonstrated heritabilities of 30% to 33% [29] and to be influenced by familial and genetic factors [27,28]. However, in our sample, this difference regards only female relatives of subjects with PD. In population studies, no significant differences between male and female subjects have been reported [40], and in the present work, these data are replicated. Nevertheless, epidemiologic data concerning PD [41] reveal an association between this diagnosis and sex, with a higher risk for women (female-tomale ratio ≈2:1). It is possible that being a female constitutes a predisposition not only to PD but also to PD information processing features, including alexithymia and AS: for this reason, female relatives of subjects with PD could manifest more evident differences from HC subjects on these relevant variables, as revealed in our sample. Regarding AS, HC and R groups do not differ in this variable between each other: they show a lower level of AS than that in subjects with PD. Compared with the literature, this result appears controversial. van Beek and Griez [25], for example, found that first-degree relatives of subjects with PD were significantly more anxiety sensitive than HC subjects, but less anxiety sensitive than patients with PD. In another study, van Beek et al [42] demonstrated that children of PD parents are not more anxiety sensitive than children of healthy parents and did not find correlations between AS scores of PD parents and their offspring, supposing that AS could be considered a developing vulnerability factor that only becomes important toward late adolescence or early adulthood. However, in the first cited study, the samples were relatively small (23 healthy first-degree relatives of patients with PD, 30 HCs, and 38 patients with PD), control variables could not be taken into account because of the statistical
analysis conduced, and the scores of their first-degree relatives were only moderately increased, although significantly higher than those of HCs. Although the present study was conduced on larger samples, using more complete analyses, further research is needed to confirm these results, for example, using adequate behavioral genetic methods on very large samples. Probably, as reported elsewhere [43], AS has more complex etiology than previously recognized, which could be comprehended only through a mix of dimension-specific and nonspecific etiologic factors. In the present study, the lack of statistical difference in alexithymia between the HC and the PA groups, which is contrary to our expectations, could be explained in different ways. For example, the PA sample size (n = 30) could be insufficient to detect significant differences: in this case, it could be reasonable to replicate this study by simply increasing the sample size. Moreover, we have not taken into account a potentially important variable: the spontaneous vs situational nature of the panic attacks experienced by the PA group. It is possible that the processes involved in these 2 different kinds of panic attacks are specific and partially independent from each other and that different biological, genetic, cognitive, and emotional mechanisms play a role in their manifestation: if subjects belonging to 1 of these 2 categories share a different amount of genetic or environmental vulnerability to PD, they could also share a different way of emotional information processing with subjects with PD. One more explanation has been considered: if some people experience sporadic panic attacks but do not develop PD, we could suppose that these subjects have some protective factors that prevent them from triggering anticipatory anxiety and dysfunctional cognitive appraisal of bodily sensations; the ability to correctly read internal states and bodily sensations is a good candidate to hold this role. A first limit of the present study is represented by the small sample size of the PA group compared with the other groups. This limitation has been, nonetheless, considered only partially influential because homogeneity of variances and normality of variables' distributions have been verified for all analyses. The second limit concerns the scale for measuring alexithymia (TAS-20): it is a self-report instrument that could be affected by nonidentifiable bias depending on variables such as personality traits or social desirability. Although this instrument has demonstrated good psychometric properties
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
and has been validated in different populations (clinical and nonclinical) and languages, it could be useful, as reported by other authors [44], to use, in future studies, multiple methods that eliminate the potential influence of measurement method-based response biases associated with monomethod assessment. A third limit, already partially discussed, could be identified in the characteristics of the PA group: it is possible that differences in the nature of panic attacks (situational vs spontaneous) have not been sufficiently considered in the analysis and that the exclusive or mixed presence of these 2 kinds of panic attacks not only has reference to a phenomenologic description of the disorder but also is related to biological, cognitive, and emotional mechanisms involved in its etiopathogenesis and clinical course. Fourth, the R group has an elevated average age (mean ± SD, 45.58 ± 15.97 years): it is well above the peak age of onset for PD (ie, 25 years [45]). Consequently, it may be argued that only those with an ASI score in the high range are at risk for PD and share with them, differently from HC subjects, a different amount of genetic or environmental vulnerability to PD. Future research could test this consideration, selecting a group of relatives of subjects with PD with different age distributions. A further limit to consider concerns the assessment methodology. In the 4 samples, we measured AS and alexithymia using a single self-administered instrument, the ASI (16-item version) and the TAS-20, respectively. Although these 2 psychological scales show good reliability and validity characteristics, as reported in the “Methods” section, further studies should comprehend multimethod assessment procedures. Last, the clinical group used in the present study consisted exclusively of patients diagnosed as having PD with agoraphobia. Consequently, the results cannot be generalized to patients affected by PD without agoraphobia. Despite the limitations considered above, the results of the present study are encouraging to explore, with more accurate techniques of assessment, larger sample sizes, and genetic studies, how these variables play a role not only in differentiating groups with a different level of risk for developing PD but also in influencing the clinical course of the disorder, its onset, and general population liability to this clinical condition. Acknowledgment We thank all the patients who participated in this study. References [1] Barlow DH, Craske MG. The phenomenology of panic. In: Rachman S, & Maser JD, editors. Panic: psychological perspectives. Hillsdale, NJ: Erlbaum; 1988. p. 11-35. [2] Clark DM. A cognitive approach to panic. Behav Res Ther 1986;24(4): 461-70.
873
[3] Reiss S, McNally R. In: & Reiss SBR, editor. Theoretical issues in behavior therapy. San Diego: Academic Press; 1985. [4] Reiss S. Expectancy model of fear, anxiety, and panic. Clin Psychol Rev 1991;11(2):141-53. [5] White KS, Brown TA, Somers TJ, Barlow DH. Avoidance behavior in panic disorder: the moderating influence of perceived control. Behav Res Ther 2006;44(1):147-57. [6] Wilson KA, Hayward C. Unique contributions of anxiety sensitivity to avoidance: a prospective study in adolescents. Behav Res Ther 2006;44(4):601-9. [7] Zvolensky MJ, Kotov R, Antipova AV, Schmidt NB. Cross cultural evaluation of smokers risk for panic and anxiety pathology: a test in a Russian epidemiological sample. Behav Res Ther 2003;41(10): 1199-215. [8] Schmidt NB, Koselka M. Gender differences in patients with panic disorder: evaluating cognitive mediation of phobic avoidance. Cognitive Ther Res 2000;24(5):533. [9] Naragon-Gainey K. Meta-analysis of the relations of anxiety sensitivity to the depressive and anxiety disorders. Psychol Bull 2010; 136(1):128-50. [10] Zvolensky M, Bernstein A, Vujanovic A. The interactive effects of anxiety sensitivity and emotion dysregulation in predicting anxietyrelated cognitive and affective symptoms. Cognitive Ther Res 2008; 32(6):803-17. [11] Benítez CI, Shea MT, Raffa S, Rende R, Dyck IR, Ramsawh HJ, et al. Anxiety sensitivity as a predictor of the clinical course of panic disorder: a 1-year follow-up study. Depress Anxiety 2009;26(4): 335-42. [12] Nemiah JC, Freyberger H, Sifneos PE. Alexithymia: a view of the psychosomatic process. In: & Hill OW, editor. Modem trends in psychosomatic medicine, Vol. 3. London: Butterworths; 1976. p. 430-9. [13] Taylor GJ, Bagby RM, Parker JDA. Disorders of affect regulation: alexithymia in medical and psychiatric illness. Cambridge: Cambridge University Press; 1997. [14] Galderisi S, Mancuso F, Mucci A, Garramone S, Zamboli R, Maj M. Alexithymia and cognitive dysfunctions in patients with panic disorder. Psychother Psychosom 2008;77(3):182-8. [15] Fukunishi I, Kikuchi M, Wogan J, Takubo M. Secondary alexithymia as a state reaction in panic disorder and social phobia. Compr Psychiatry 1997;38(3):166-70. [16] Bankier B, Aigner M, Bach M. Alexithymia in DSM-IV disorder: comparative evaluation of somatoform disorder, panic disorder, obsessive-compulsive disorder, and depression. Psychosomatics 2001;42(3):235-40. [17] De Berardis D, Campanella D, Gambi F, La Rovere R, Sepede G, Core L, et al. Alexithymia, fear of bodily sensations, and somatosensory amplification in young outpatients with panic disorder. Psychosomatics: Journal of Consultation Liaison Psychiatry 2007;48(3):239-46. [18] Marchesi C, Fonto S, Balista C, Cimmino C, Maggini C. Relationship between alexithymia and panic disorder: a longitudinal study to answer an open question. Psychother Psychosom 2005;74(1):56-60. [19] Klauke B, Deckert J, Reif A, Pauli P, Domschke K. Life events in panic disorder—an update on “candidate stressors.”. Depress Anxiety 2010;27(8):716-30. [20] Hettema JM, Neale MC, Kendler KS. A review and meta-analysis of the genetic epidemiology of anxiety disorders. Am J Psychiatry 2001; 158(10):1568-78. [21] Goldstein RB, Wickramaratne PJ, Horwath E, Weissman MM. Familial aggregation and phenomenology of 'early'-onset (at or before age 20 years) panic disorder. Arch Gen Psychiatry 1997;54(3): 271-8. [22] Stewart S. Heartbeat awareness and heart rate reactivity in anxiety sensitivity: a further investigation. J Anxiety Disord 2001;15(6): 535-53. [23] Stein MB, Jang KL, Livesley WJ. Heritability of anxiety sensitivity: a twin study. Am J Psychiatry 1999;156(2):246-51.
874
M. Cucchi et al. / Comprehensive Psychiatry 53 (2012) 868–874
[24] Scher CD, Stein MB. Developmental antecedents of anxiety sensitivity. J Anxiety Disord 2003;17(3):253-69. [25] van Beek N, Griez E. Anxiety sensitivity in first-degree relatives of patients with panic disorder. Behav Res Ther 2003;41(8):949-57. [26] Stein MB, Schork NJ, Gelernter J. Gene-by-environment (serotonin transporter and childhood maltreatment) interaction for anxiety sensitivity, an intermediate phenotype for anxiety disorders. Neuropsychopharmacology: official publication of the American College of Neuropsychopharmacology 2008;33(2):312-9. [27] Ham B, Lee M, Lee Y, Kim M, Choi M, Oh K, et al. Association between the catechol O-methyltransferase Val108/158Met polymorphism and alexithymia. Neuropsychobiology 2005;52(3):151-4. [28] Valera EM, Berenbaum H. A twin study of alexithymia. Psychother Psychosom 2001;70(5):239-46. [29] Jørgensen MM, Zachariae R, Skytthe A, Kyvik K. Genetic and environmental factors in alexithymia: a population-based study of 8,785 Danish twin pairs. Psychother Psychosom 2007;76(6):369-75. [30] Goracci A, Fagiolini A, Salviulo C, Forgione RN, Martinucci M, Castroglovanni P. Panic agoraphobic spectrum in psychiatrically healthy subjects: impact on quality of life. Epidemiol Psichiatr Soc 2009;18(1):65-8. [31] Beck AT, Steer RA, Brown GK. Manual for the Beck Depression Inventory-II. San Antonio (Tex): Psychological Corporation; 1996. [32] Bagby RM, Parker JDA, Taylor GJ. The twenty-item Toronto Alexithymia Scale: I. Item selection and cross-validation of the factor structure. J Psychosom Res 1994;38(1):23-32. [33] Bressi C, Taylor G, Parker J, Bressi S, Brambilla V, Aguglia E, et al. Cross validation of the factor structure of the 20-item Toronto Alexithymia Scale: an Italian multicenter study. J Psychosom Res 1996;41(6):551-9. [34] Reiss S, Peterson RA, Gursky DM, McNally RJ. Anxiety sensitivity, anxiety frequency and the predictions of fearfulness. Behav Res Ther 1986;24(1):1-8.
[35] Zinbarg RE, Mohlman J, Hong NN. Dimensions of anxiety sensitivity. In: & Taylor S, editor. Anxiety sensitivity: theory, research, and treatment of the fear of anxiety. Mahwah (NJ): Lawrence Erlbaum; 1999. p. 83-114. [36] McNally RJ. Anxiety sensitivity and panic disorder. Biol Psychiatry 2002;52(10):938-46. [37] Taylor S, Jang KL, Stewart SH, Stein MB. Etiology of the dimensions of anxiety sensitivity: a behavioral-genetic analysis. J Anxiety Disord 2008;22(5):899-914. [38] Schmidt NB, Mitchell MA, Richey JA. Anxiety sensitivity as an incremental predictor of later anxiety symptoms and syndromes. Compr Psychiatry 2008;49(4):407-12. [39] Cox BJ. The nature and assessment of catastrophic thoughts in panic disorder. Behav Res Ther 1996;34(4):363-74. [40] Franz M, Popp K, Schaefer R, Sitte W, Schneider C, Hardt J, et al. Alexithymia in the german general population. Soc Psychiatry Psychiatr Epidemiol 2008;43(1):54-62. [41] Grant BF, Hasin DS, Stinson FS, Dawson DA, Goldstein RB, Smith S, et al. The epidemiology of DSM-IV panic disorder and agoraphobia in the united states: results from the national epidemiologic survey on alcohol and related conditions. J Clin Psychiatry 2006; 67(3):363-74. [42] van Beek N, Perna G, Schruers K, Muris P, Griez E. Anxiety sensitivity in children of panic disorder patients. Child Psychiatry Hum Dev 2005;35(4):315-24. [43] Taylor S, Jang KL, Stewart SH, Stein MB. Etiology of the dimensions of anxiety sensitivity: a behavioral-genetic analysis. J Anxiety Disord 2008;22(5):899-914. [44] Taylor GJ, Bagby RM. New trends in alexithymia research. Psychother Psychosom 2004;73(2):68-77. [45] American Psychiatric Association. Diagnostic and statistical manual of mental disorders, fourth edition, text revision (DSM-IV-TR). Washington (DC): American Psychiatric Association; 2000.