Cardiovascular stress reactivity and recovery in bulimia nervosa and binge eating disorder

International Journal of Psychophysiology 78 (2010) 163–168

Contents lists available at ScienceDirect

International Journal of Psychophysiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j p s yc h o

Cardiovascular stress reactivity and recovery in bulimia nervosa and binge eating disorder Nadine Messerli-Bürgy a,b,⁎, Claudia Engesser c, Eva Lemmenmeier c, Andrew Steptoe a, Kurt Laederach-Hofmann c a b c

Psychobiology Group, Department of Epidemiology and Public Health, University College London, United Kingdom Institute of Psychology, Clinical Psychology and Psychotherapy Unit, University of Berne, Switzerland Department of Endocrinology, Diabetology, and Clinical Nutrition, Autonomic Lab, University of Berne, Switzerland

a r t i c l e

i n f o

Article history: Received 11 January 2010 Received in revised form 24 June 2010 Accepted 20 July 2010 Available online 25 July 2010 Keywords: Eating disorder Heart rate variability Stress response Bulimia Binge eating

a b s t r a c t Objective: Stress plays a role in the pathology of bulimia nervosa and binge eating disorders, but it is unclear whether they involve similar disturbances of biological stress responses. Patients and methods: We recruited 25 patients with binge eating behavior, 12 with bulimia nervosa (BN) and 13 with binge eating disorder (BED), and compared them with 13 obese non-binge eaters (NBED). We measured heart rate variability in response to mental stress tasks, and concentrations of leptin, glucose and insulin in the blood. Results: Heart rate stress reactivity was highest in BN patients. Heart rate variability did not change during mental stress in BN and BED patients, but reduced as expected in the NBED group. During post-stress recovery, heart rate variability decreased in BN, was maintained in BED and increased as expected only in the NBED group. Conclusions: BN and BED patients exhibit limitations in autonomic stress reactivity and recovery capacity. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Bulimia nervosa (BN) and binge eating disorder (BED) are both characterized by recurrent episodes of binge eating, but differ in compensatory behavior. Bulimic patients regularly use selfinduced vomiting or misuse laxatives to control their weight, whereas individuals with binge eating disorder usually suffer from uncontrolled weight gain and obesity (Munsch and Beglinger, 2005). It is thought that stress plays an important role in the pathology of both disorders (Pinaquy et al., 2003; Koo-Loeb et al., 1998). Heightened negative mood has frequently been described as a precursor to binge eating episodes in BN and BED (Levine and Marcus, 1997; Eldredge and Agras, 1996; Telch and Agras, 1996) as well as obesity. Individuals with BN and BED experience greater negative mood before binge eating episodes and perceive challenging situations as more stressful compared with healthy individuals (Pinaquy et al., 2003; Wolff et al., 2000; Hansel and Wittrock, 1997; Kjelsas et al., 2004). Additionally, several studies have shown an increase in hunger and desire to binge following experimental stress

⁎ Corresponding author. Dept. of Clinical Psychology and Psychotherapy, University of Berne, 3012 Berne, Switzerland. Tel.: + 41 31 631 54 28; fax: + 41 31 631 41 55. E-mail address: [email protected] (N. Messerli-Bürgy). 0167-8760/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ijpsycho.2010.07.005

tasks (Cattanach et al., 1988; Gluck et al., 2004; Tuschen-Caffier and Vogele, 1999). Therefore, understanding the physiological mechanisms underlying stress responses would be useful for the identification of physiological correlates of binge eating patterns in BN and BED disorders. There is evidence that people with BN and BED differ in autonomic functioning from healthy individuals. Patients with BN exhibit blunted sympathetic activation in response to mental stress (KooLoeb et al., 1998) and reduced 24 h blood pressure (Cong et al., 2004). Furthermore, results of a comparison of BED and controls suggested higher stress vulnerability in the BED population (Friederich et al., 2006). However, obesity might confound these results, as other studies on obese subjects have shown a change of stress response profiles due to obesity itself (Laederach-Hofmann et al., 2000; Valensi et al., 1995). Similarities in biological stress vulnerability could be a key to understanding binge eating behavior in BN and BED. But studies of autonomic stress responses in BN and BED individuals are rather scarce. The purpose of this study was to assess the stress responses of patients with similar binge eating behavior but different diagnoses. In order to investigate the capacity of the autonomic nervous system to adapt to stressful conditions, we measured heart rate variability (HRV) and assessed autonomic activity during mental stress testing. We hypothesized that BN and BED patients would show similar disturbances of autonomic stress responses in comparison with nonbinge eaters (NBED).

164

N. Messerli-Bürgy et al. / International Journal of Psychophysiology 78 (2010) 163–168

2. Materials and methods 2.1. Participants and study protocol Thirty-eight women agreed to participate. Bulimia nervosa (BN) was diagnosed in 12 participants, binge eating disorder (BED) in 13 and obesity without binge eating behavior (NBED) in 13 participants, according to DSM IV diagnostic criteria (Amercian Psychological Association, 1994). Individuals had to show stabilized electrolyte conditions, unchanged weight levels during the past 10 to 12 weeks and had to undergo a routine physical examination and diagnostic interview conducted by an experienced psychiatrist before being asked to participate in the study. Individuals with a medical history of cardiovascular disease, metabolic disease, other disease or medication that could influence autonomic functioning were excluded. All participants were recruited consecutively from the psychiatric outpatient clinic at the University Hospital of Berne and from a private clinic for eating disorders ‘Klinik Wysshoelzli’, both in Switzerland. All participants were referred to these specialized units by their family doctors, and were asked to participate only if the inclusion criteria for the study were fulfilled. Participants gave written informed consent to the study procedures which were approved by the local Research Ethics Committee. 2.2. Questionnaires The German version of Stunkard and Messick's Three-Factor Eating Questionnaire (Pudel and Westenhoefer, 1989; Stunkard and Messick, 1985) was used to measure the three scales ‘cognitive restraint’, a tendency to control food intake in order to prevent weight gain; ‘disinhibition’ a term describing a combination of emotional and external eating, and ‘hunger’. Additionally, symptoms of eating disorder such as fear of weight gain, dissatisfaction with one's own figure, binge behavior, nausea and vomiting after eating, feelings of external eating pressures and excessive demands, perfectionism, interpersonal reservations and fear of one's own feelings were assessed by the self-report Eating Disorder Inventory (Diehl and Staufenbiel, 1994). Finally, the German version of Hospital Anxiety and Depression Scale (Herrmann, 1997; Zigmond and Snaith, 1983) was administered to assess symptoms of depression and anxiety. 2.3. Cardiovascular assessment and stress testing Autonomic function testing was performed in 3 conditions: at rest, during mental stress, and recovery from mental stress. Each condition lasted 10 minutes according to the guidelines of the joint European/ North American task force (1996) on heart rate variability. All testing sessions were carried out between 08:00 am and 12:00 pm. Patients were required not to do any physical activity during the 24 h before the testing session, not to drink caffeine-containing liquids, and were tested in a fasting state (following overnight fast). All participants were tested in their first week after menstruation only (Ettinger et al., 1998; Matsumoto et al., 2007). The laboratory had no natural light and was heated to room temperature (18–20 °C). After fitting of measuring devices an equilibration period of 20 minutes took place. The mental stress test involved administration of the Stroop color/ word interference task using standardized methods (Hoshikawa and Yamamoto, 1997). The task difficulty was titrated on the basis of the individual's performance by testing system ‘Wiener Testsystem’ (Schuhfried GmbH, Austria). Heart rate, blood pressure, and respiration were recorded simultaneously with a computer-based system (Task Force Monitor ®, CN-Systems, Graz, Austria). The Einthoven lead II of the electrocardiogram was used to detect the R-peaks of the electrocardiogram (ECG). Blood pressure was measured non-invasively and continuously from the fingers (Habenbacher et al., 2002). The

continuous blood pressure signal was equilibrated in five minute intervals by oscillometric blood pressure determination in the contralateral arm. Respiration frequency was recorded using impedance pneumography and stroke volume was estimated using impedance cardiography. The ECG and beat-to-beat blood pressure signals were sampled with a frequency of 1000 Hz and changes of thoracic impedance due to respiration at 50 Hz. HRV was assessed using spectral analysis. Previous analyses have shown that the commercial algorithm used to prepare heart period data for spectral analysis (TFM Software V2.4, CN-Systems, Graz, Austria) fails to detect all artifacts. After applying this method, we additionally visually displayed heart period data to screen out any residual outliers and artifacts. Using this conservative procedure, fewer than 10% of the total data were problematic and had to be excluded. ECG sequences that showed premature beats or other artifacts were also excluded from the analysis. A commercial software package was then employed to determine HRV with an adapted autoregressive model as proposed by Bianchi et al. (1997), using a recursive least squares algorithm. Power spectral densities (in ms2) were quantified by the area within the frequency bands which were determined according to international guidelines (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). We calculated total power (HRV-TP) in the range 0.003–0.4 Hz; low-frequency (LF) power, range 0.04–0.15 Hz; and high-frequency (HF) power, range 0.15–0.4 Hz. Data were skewed, so all spectral values were normalized by natural logarithms (ln) (Malliani et al., 1991). HRV-TP is an index of total heart rate variability (HRV). LF HF ratio reflects sympathetic/parasympathetic balance, with higher values indicating greater sympathetic control (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). By contrast, the HF power component is a marker of parasympathetic influence on cardiac chronotropy. Baroreflex sensitivity was calculated by using hemodynamic data assessed with impedance cardiography (ICG) using the dZ/dt paradigm which is part of the TFM program. 2.4. Blood samples Blood samples for hormone assays and glucose were obtained fasting after the heart rate variability testing session between 8:00 and 12:00 h in all participants. Glucose was assessed by using hexokinase enzyme assay (Roche, Modular P800, Roche Diagnostics Switzerland) and insulin by microparticle-enzyme immunoassay (MEIA), Abbott Axsym. Plasma leptin concentrations were determined by a LEP-R40 radioimmunoassay kit purchased from Mediagnost (Reutlingen, Germany). Sensitivity of the method was 0.04 ng/ ml. The homeostasis model assessment index (HOMA-R) of insulin resistance was calculated (Gutt et al., 2000). 2.5. Statistics Basic descriptive statistics and repeated measures analysis of variance were used to analyze the data, with trial (rest, mental stress, and recovery) as the within-subject factor and patient group (BN, BED and NBED) as the between-subject factor. Analyses were performed with a PC-based SPSS 14.0 (SPSS Inc. Chicago, Illinois, USA). Data are reported as means ± standard deviations (SD). A p-value of less than 0.05 was considered significant. 3. Results 3.1. Participant characteristics Participants differed in age and BMI (Table 1). Post hoc tests indicated that the BN group was younger on average than the other

N. Messerli-Bürgy et al. / International Journal of Psychophysiology 78 (2010) 163–168 Table 1 Baseline results in women with bulimia nervosa (BN), binge eating (BED), and nonbinge obese (NBED). Baseline

BMI (kg/m2) Age (years) Glucose (mmol/l) Insulin (mmol/l) HOMA-R (mmol IE/l) Leptin (μg) HR (bpm) sBP (mm Hg) dBP (mm Hg) MAP (mm Hg) HRV-LF (ln ms2) HRV-HF (ln ms2) HRV-TP (ln ms2) LF HF ratio Baroreceptor sensitivity (ms/mm Hg)

BN (n = 12)

BED (n = 13)

NBED (n = 13)

M (SD)

M (SD)

M (SD)

23.15(5.61) 24.43(5.70) 4.51(0.61) 10.01(10.20) 2.33(2.44) 14.52(10.83) 74.63(11.48) 104.64(17.35) 66.23(10.54) 77.94(11.88) 5.78(0.45) 5.42(0.94) 6.63(0.49) 2.19(2.30) 16.60(6.74)

37.87(6.42) 33.92(7.04) 5.09(0.45) 17.83(11.31) 4.06(2.61) 46.42(25.51) 80.89(11.51) 114.16(15.18) 73.55(11.04) 86.47(12.28) 4.54(1.13) 3.86(1.71) 5.90(1.30) 3.37(3.90) 10.66(5.45)

35.99(5.00) 41.15(8.97) 5.48(0.89) 18.46(12.61) 4.73(3.31) 50.70(26.64) 77.36(9.60) 123.41(23.58) 76.72(16.17) 91.56(17.79) 4.23(0.90) 3.69(1.07) 5.68(1.07) 2.19(1.28) 9.64(3.66)

pvalue 0.001 0.001 0.008 0.191 0.172 0.003 0.363 0.062 0.131 0.070 0.001 0.004 0.083 0.456 0.075

165

3.2. Blood pressure and heart rate There were no differences in baseline heart rate, diastolic or mean arterial pressure between groups, after controlling for BMI (see Table 3). Systolic blood pressure (sBP) differed marginally (F(2,34) = 3.14, p = 0.056) between groups at baseline, being highest in the NBED and lowest in the BN group. Repeated measures analysis of variance confirmed there was a significant trial effect for heart rate (F(2,34)=23.52, p =0.001), systolic (F(2,34) =19.19, p =0.001) and diastolic blood pressure (F(2,34) =7.20, p = 0.002) and mean arterial pressure (MAP) (F(2,34) = 12.79, p = 0.001). Mental stress induced increased heart rate and blood pressure, with values returning towards resting levels during the recovery period. Blood pressure stress reactivity did not differ between groups, but the group by trial interaction was significant for heart rate (F (4,70)= 2.78, p =0.033). As can be seen in Table 3, the heart rate response to stress was greater in BN patients than in the other two groups.

3.3. Heart rate variability two groups. There were also significant BMI differences between the three groups (F(2,37) = 24.24, p = 0.001). The average BMI of patients in the BED and NBED groups were both in the obesity range, (means 37.87 ± 6.42 and 35.99 ± 5.00 kg/m2 respectively), whereas the BN group was in the normal BMI range (23.15 ± 5.61 kg/m2). Differences in the concentration of glucose and plasma leptin were found between the three groups, but not in insulin or HOMA-R (see Table 1). There was no difference in personality traits or anxiety and depression level between the three groups (Table 2). Regarding eating behavior, the groups differed significantly in restrained eating (F(2,30) = 8.45, p = 0.001), hunger (F(2,28) = 4.33, p = 0.023), binge eating behavior (F(2,17) = 4.06, p = 0.039) and feelings of external eating pressures (F(2,12) = 7.82, p = 0.009) as expected. Patients with BN showed more restrained eating than the BED and NBED groups, and more binge eating than the NBED group. Furthermore, the BN group had fewer feelings of hunger than did BED patients, but also experienced more feelings of external eating pressure (see Table 2).

At baseline, there were significant group differences in LF power (HRV-LF) (F(2,37) = 10.56, p = 0.001) and HF power (HRV-HF) (F (2,37) = 11.23, p = 0.012). As can be seen in Table 1, both LF and HF power values were greater in the BN than BED or NBED groups. There were no differences in baseline LF HF ratio, total power (HRV-TP) or baroreceptor sensitivity (BRS). Analysis of HRV data across trials indicated that there were significant changes in LF power (F(2,34) = 13.31, p = 0.001), total power (F(2,34) = 20.67, p = 0.001), LF HF ratio (F(2,34) = 4.48, p = 0.019), and in baroreceptor sensitivity (F(2,32) = 6.61, p = 0.004), but not in HF power (p = 0.095) (see Table 3). LF power and LF HF ratio increased during mental stress, reflecting the well established pattern of sympathetic activation, and this was accompanied by inhibition of baroreceptor reflex sensitivity. The three groups did not differ in total power, LF HF ratio or baroreceptor sensitivity responses over the session. However, there were significant interactions between trial and group in the analyses of LF (F(4,70) = 9.85, p = 0.001) and HF (F(4,70) = 5.96, p = 0.001) power, which are illustrated in Fig. 1. Only the NBED group showed the anticipated

Table 2 Self-reported psychopathology in women with bulimia nervosa (BN), binge eating (BED), and non-binge obese (NBED). BN (n = 12)

BED (n = 13)

NBED (n = 13)

BN vs BED

BN vs NBED

BED vs NBED

M ± SD

M ± SD

M ± SD

p-value

p-value

p-value

TFEQ Restraint eating Disinhibition Hunger

14.50 ± 3.78 9.90 ± 4.31 4.20 ± 3.65

7.72 ± 4.73 12.55 ± 3.84 9.77 ± 4.60

8.08 ± 4.42 11.41 ± 3.17 7.41 ± 4.20

** ns *

** ns ns

ns ns ns

EDI-2 Fear of weight gain Dissatisfaction with one's own figure Binge eating Nausea and vomiting after eating Feelings of external eating pressures Feelings of excessive demands and inferiority Perfectionism and drive for achievement Interpersonal reservations Fear of one's own feelings

15.50 ± 3.16 9.13 ± 3.64 12.38 ± 6.32 8.50 ± 4.84 3.38 ± 2.00 8.25 ± 3.62 9.25 ± 2.60 7.38 ± 3.46 7.50 ± 3.96

10.50 ± 5.06 11.75 ± 0.50 10.50 ± 3.41 3.66 ± 2.51 2.50 ± 0.70 7.00 ± 3.46 6.25 ± 4.99 7.50 ± 0.57 7.75 ± 4.99

10.71 ± 4.49 11.00 ± 2.97 8.71 ± 3.03 3.33 ± 3.38 0.50 ± 1.00 7.66 ± 2.44 9.22 ± 2.94 6.77 ± 2.72 5.37 ± 3.20

ns ns ns ns ns ns ns ns ns

** ns * ns ** ns ns ns ns

ns ns ns ns ns ns ns ns ns

8.50 ± 3.78 6.20 ± 3.33

7.66 ± 4.06 6.22 ± 2.81

8.75 ± 4.18 7.00 ± 3.43

ns ns

ns ns

ns ns

HADS Anxiety Depression *p b 0.05, **p b 0.01.

N. Messerli-Bürgy et al. / International Journal of Psychophysiology 78 (2010) 163–168

p-value

* ns ns ns ***

***

ns

ns ns

Group

p-value

ns ns ns ns *

*

ns

ns ns

Trial

p-value

reduction in HF power during mental stress. Differences also emerged during the post-stress recovery period in terms of change of LF power (F(2,37) = 13.89, p = 0.001) and change of HF power (F(2,37) = 6.38, p = 0.004). Following stress, LF and HF power diminished below the resting level in the BN group but not in the BED or NBED groups (see Fig. 1). Both the BED and NBED groups showed a partial failure of poststress recovery of LF power, suggestive of persistent sympathetic activation. Additionally, there were marked differences between groups in HF power during recovery, with the expected increase in HF only occurring in the NBED group.

* ** 3.00(2.44) 9.11(3.82) 3.13(0.92) 8.23(3.99) 2.19(1.28) 9.64(3.66) 3.80(2.90) 11.12(5.80) 4.72(3.92) 10.01(5.30) 3.37(3.90) 10.66(5.45)

*p b 0.05**p b 0.01. ***p b 0.001.

2.20(1.35) 10.71(3.61) 2.19(2.30) 16.60(6.74)

2.48(2.15) 13.80(4.85)

*** 5.92(1.15) 5.81(1.27) 5.68(1.07) 6.28(1.39) 6.25(1.39) 5.90(1.30) 6.73(0.78) 6.63(0.49)

7.16(0.62)

ns 3.75(1.21) 3.54(1.14) 3.69(1.07) 3.88(1.82) 3.88(1.90) 3.86(1.71) 5.42(0.86) 5.42(0.94)

4.25(1.44)

87.20(18.01) 114.46(21.03) 73.04(15.73) 86.18(17.06) 6.02(0.81) 74.63(11.48) 104.64(17.35) 66.23(10.54) 77.94(11.88) 5.78(0.45)

74.36(11.64) 106.48(20.53) 65.38(15.22) 78.67(16.67) 4.82(1.03)

80.89(11.51) 114.16(15.18) 73.55(11.04) 86.47(12.28) 4.54(1.13)

84.97(9.20) 122.19(14.19) 79.12(9.11) 93.63(10.36) 5.10(1.36)

79.94(11.02) 116.93(15.31) 74.99(10.70) 89.30(12.27) 4.99(1.49)

77.36(9.60) 123.41(23.58) 76.72(16.17) 91.56(17.79) 4.23(0.90)

85.71(16.05) 140.90(28.58) 86.93(23.18) 104.74(22.80) 4.49(0.97)

77.82(9.91) 129.92(24.61) 80.22(18.52) 96.67(19.59) 4.54(0.97)

*** *** ** *** ***

4. Discussion

HR (bpm) sBP (mm Hg) dBP (mm Hg) MAP (mm Hg) HRV-LF (ln ms2) HRV-HF (ln ms2) HRV-TP (ln ms2) LF HF ratio Baroreceptor sensitivity (ms/mm Hg)

M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD)

M (SD)

Rest

M (SD)

Rest

BED

Rest

Recovery Mental stress BN

Table 3 Cardiovascular and spectral values, and baroreceptor sensitivity (controlled for BMI).

Mental stress

Recovery

NBED

Mental stress

Recovery

Statistical effects

Inter action

166

We studied autonomic and cardiovascular responses to mental stress in different types of binge eating disorder. We hypothesized that patients with binge eating but different diagnoses (BN versus BED) might have impairments in autonomic adaptation during stress exposure, in comparison with non-binge obese patients. We observed clear-cut cardiovascular activation during stress exposure in all three groups, consistent with previous studies of healthy individuals. Similarly, baroreflex sensitivity declined during mental stress task as expected (Steptoe and Sawada, 1989). However, important differences in physiological response between groups were also observed. Heart rate reactivity was greatest in the BN group, and smallest in the BED group (see Fig. 1). Additionally, the BN group had greater LF power at baseline, suggesting chronic shifts in autonomic balance towards the sympathetic nervous system in this condition. There were also differences in post-stress HRV reactivity and recovery. We anticipated that LF power would increase and HF power would reduce during stress, returning to baseline levels in the post-stress recovery period. Instead, neither the BED nor NBED groups showed complete recovery in LF power following stress. Additionally, HF power decreased rather than increased during recovery in the BN group, and was insensitive to stress in the BED group. These phenomena may reflect an impaired ability of the cardiovascular system to adapt to stress exposure in BN and BED individuals. Previous studies have suggested differences between BED and NBED groups in hormonal concentrations and emotional factors such as depression (Pinaquy et al., 2003; Gluck et al., 2004). They have also shown a correlation between hormonal concentration and BMI. We did find differences in blood measurement such as leptin and glucose, but not insulin or HOMA-R in our study. Patients with NBED showed higher levels of glucose than BED. This might be due to the fact that the sustained stimulation of beta cells in NBED might lead to the early development of insulin resistance (Roehrig et al., 2009). We were not able to identify differences in terms of emotional upset such as depression or anxiety between BED and NBED. Furthermore, other studies have found differences in other dimensions of eating pathology between the groups BED and BN (Ardovini et al., 1999). Dimensions such as nausea, vomiting and disinhibition did not differ between the groups as expected. However, the categorization was made according to the diagnostic criteria of DSM IV after interviewing the patients, and we suspect that the questionnaires used did not have the sensitivity that is evident in the clinical diagnosis. We have also found that BED and NBED patients referred to these specialized units have a long history of eating problems and weight cycling, and this might be reflected in similarities in eating pathology across the three groups. Previous studies have also shown disturbances of autonomic function in BN. Patients with BN seem to exhibit a stress-induced sympathetic dysregulation which becomes apparent as a reduced sympathetic activation in response to mental stress (Koo-Loeb et al., 1998; Pirke, 1996; Pirke et al., 1992). Our results are more complex, in that heart rate reactivity was enhanced in BN patients, while the elevated LF power during baseline suggests that resting sympathetic

N. Messerli-Bürgy et al. / International Journal of Psychophysiology 78 (2010) 163–168

167

Fig. 1. LF power and HF power during trials for each group (BN = bulimia nervosa, BED = binge eating disorder, NBED = non-binge obese).

tone was increased. But BN patients did not differ from restrained eaters or controls in terms of psychophysiological reactivity in another study (Tuschen-Caffier and Vogele, 1999). Previous research has suggested higher psychophysiological arousal to binge cues in BED patients (Vogele and Florin, 1997) and an increased inhibition of HF power during stress in comparison with NBED individuals (Friederich et al., 2006). Our findings are inconsistent with this work, since we found that HF inhibition during stress was if anything greater in NBED than BED patients. This discrepancy in results may be because of the difference in sample size and depression. Our sample was rather small and showed fewer symptoms of depression. To our knowledge, no study has compared autonomic activity and the impact of mental stress in BN and BED individuals. Nor are we aware of previous studies comparing BN and BED individuals with NBED obese. Our findings therefore add to the evidence that patients who binge (BN and BED) have disturbed autonomic stress responses. These results need to be interpreted with caution because of the limitations of our study. The sample size was relatively small. It is possible that a larger sample size would have revealed differences in blood pressure or HRV responses between the two eating disordered groups. However, the differences between groups were very small; power calculations based on these results indicate that the sample size required to detect significant differences in blood pressure or HF HRV of this magnitude would be more than 1750 patients per group. We used an experimental approach and induced stress in a laboratory situation, so do not know whether similar factors operate in relation to binge eating in the real lives of BN and BED individuals (Gluck, 2006). We tested a NBED obese group to control for the effects of obesity, as obesity itself is known to influence stress responses. However, we did not include a normal weight NBED group. Additionally, we did not collect subjective ratings of emotional state or cognitive involvement, so we do not know if participants in the three groups were equally engaged in the tasks. It is possible, for example, that the BED group was not as engaged in the tasks as the other groups, and this is why they did not show a HRV response.

However, the tasks were administered in a standardized fashion, and the difficulty of the Stroop test was titrated to the participant's performance, reducing the likelihood of this explanation. Previous studies had emphasized the importance of abdominal adiposity index by waist–hip ratio (WHR) as an influence on biological responses to laboratory stressors (Epel et al., 2000; Moyer et al., 1994). Unfortunately, waist–hip ratio was not routinely measured at both recruitment sites for our sample, so could not be included in the analysis. Overall, patients with BN and BED exhibited a blunted capacity for autonomic stress responsivity and impaired post-stress recovery. These results suggest abnormal physiological stress reactivity in patients with binge eating that might affect mood and emotional eating behaviors such as increasing food intake or binge eating in stressful situations. Acknowledgments We gratefully acknowledge the postdoctoral Fellowship from the Swiss National Science Foundation to N. Messerli-Burgy (PBBE1117004). References Amercian Psychological Association, 1994. Diagnostic and statistical manual of mental disorders. American Psychiatric Association, Washington, DC. Ardovini, C., Caputo, G., Todisco, P., Dalle Grave, R., 1999. Binge eating and restraint model: psychometric analysis in binge eating disorder and normal weight bulimia. Eur. Eat. Disorders Rev. 7, 293–299. Bianchi, A.M., Mainardi, L.T., Meloni, C., Chierchia, S., Cerutti, S., 1997. Continuous monitoring of the sympatho-vagal balance through spectral analysis. IEEE Eng. Med. Biol. Mag. 16 (5), 64–73. Cattanach, L., Malley, R., Rodin, J., 1988. Psychologic and physiologic reactivity to stressors in eating disordered individuals. Psychosom. Med. 50, 591–599. Cong, N.D., Saikawa, T., Ogawa, R., Hara, M., Takahashi, N., Sakata, T., 2004. Reduced 24 hour ambulatory blood pressure and abnormal heart rate variability in patients with dysorexia nervosa. Heart 90, 563–564. Diehl, J.M., Staufenbiel, T., 1994. Inventar zum Essverhalten und Gewichtsproblemen (IEG). Ess-Stoerungs-Inventar (ESI). Supplement of IEG {Eating behavior and

168

N. Messerli-Bürgy et al. / International Journal of Psychophysiology 78 (2010) 163–168

weight problems inventory (EWI). Eating Disorder Inventory. Supplement to the EWI} Verlag D. Klotz, Eschborn, Germany. Eldredge, K., Agras, W., 1996. Weight and shape overconcerns and emotional eating in binge eating disorder. Int. J. Eat. Disord. 19, 73–82. Epel, R.D., McEwen, B., Seeman, T., Matthews, K., Castellazzo, G., Brownell, K.D., Bell, J., Ickovics, J.R., 2000. Stress and body shape: stress-induced cortisol secretion is consistently greater among women with central fat. Psychosom. Med. 62, 623–632. Ettinger, S.M., Silber, D.H., Gray, K.S., Smith, M.B., Yang, Q.X., Kunselman, A.R., Sinoway, L.I., 1998. Effects of the ovarian cycle on sympathetic neural outflow during static exercise. J. Appl. Physiol. 85 (6), 2075–2081. Friederich, H.C., Schild, S., Schellberg, D., Quenter, A., Bode, C., Herzog, W., Zipfel, S., 2006. Cardiac parasympathetic regulation in obese women with binge eating disorder. Int. J. Obes. 30, 534–542. Gluck, M.E., 2006. Stress response and binge eating disorder. Appetite 46, 26–30. Gluck, M.E., Geliebter, A., Hung, J., Yahav, E., 2004. Cortisol, hunger, and desire to binge eat following a cold stress test in obese women with binge eating disorder (BED). Psychosom. Med. 66, 876–881. Gutt, M., Davis, C.L., Spitzer, S.B., Llabre, M.M., Kumar, M., Czarnecki, E.M., Schneiderman, N., Skyler, J.S., Marks, J.B., 2000. Validation of the insulin sensitivity index (ISI (0, 120)): comparison with other measures. Diabetes Res. Clin. Pract. 47 (3), 177–184. Habenbacher, W., Marte, W., Skrabal, F., 2002. Cross validation between invasive and non-invasive measurement of continuous blood pressure: CN-System International Technical Report. Hansel, S., Wittrock, D.A., 1997. Appraisal and coping strategies in stress situations: a comparison of individual who binge eat to controls. Int. J. Eat. Dis. 21, 89–93. Herrmann, C., 1997. International experiences with the Hospital Anxiety and Depression Scale – a review of validation data and clinical results. J. Psychosom. Res. 42, 17–41. Hoshikawa, Y., Yamamoto, Y., 1997. Effects of Stroop color-word conflict test on the autonomic nervous system responses. Am. J. Physiol. 272 (3 Pt 2), H1113–H1121. Kjelsas, E., Borsting, I., Gudde, C.B., 2004. Antecedents and consequences of binge eating episodes in women with an eating disorder. Eat. Weight Disord. 9, 7–15. Koo-Loeb, J.H., Pedersen, C., Girdler, S.S., 1998. Blunted cardiovascular and catecholamine stress reactivity in women with bulimia nervosa. Psychiatry. Res. 80, 13–27. Laederach-Hofmann, K., Mussgay, L., Ruddel, H., 2000. Autonomic cardiovascular regulation in obesity. J. Endocrinol. 164, 59–66. Levine, M., Marcus, M., 1997. Eating behavior following stress in women with and without bulimic symptoms. Ann. Behav. Med. 19, 132–138. Malliani, A., Pagani, M., Lombardi, F., Cerutti, S., 1991. Cardiovascular neural regulation explored in the frequency domain. Circulation 84 (2), 482–492.

Matsumoto, T., Ushiroyama, T., Kimura, T., Hayashi, T., Moritani, T., 2007. Altered autonomic nervous system activity as a potential etiological factor of premenstrual syndrome and premenstrual dysphoric disorder. Biopsychosoc. Med. 1, 24. Moyer, A., Rodin, J., Grilo, C., Cummings, N., Larson, L., Rebuffe-Scrive, M., 1994. Stressinduced cortisol response and fat distribution in women. Obes. Res. 2, 255–262. Munsch, S., Beglinger, C., 2005. Obesity and Binge Eating Disorder: from Research to Clinical Practice. Bibliotheca Psychiatrica 171, Karger, Basel. Pinaquy, S., Chabrol, H., Simon, C., Louvet, J.P., Barbe, P., 2003. Emotional eating, alexithymia, and binge-eating disorder in obese women. Obes. Res. 11, 195–201. Pirke, K.M., 1996. Central and peripheral noradrenalin regulation in eating disorders. Psychiatry. Res. 62, 43–49. Pirke, K., Platte, P., Laessle, R., Seidl, M., Fichter, M., 1992. The effect of a mental challenge test of plasma norepinephrine and cortisol in bulimia nervosa and in controls. Biol. Psychiatry. 32, 202–206. Pudel, V., Westenhoefer, J., 1989. Fragebogen zum Essverhalten. Handanweisung. Hogrefe, Goettingen, Germany. Roehrig, M., Masheb, R.M., White, M.A., Grilo, C.M., 2009. The metabolic syndrome and behavioral correlates in obese patients with binge eating disorder. Obesity (Silver Spring) 17 (3), 481–486. Steptoe, A., Sawada, Y., 1989. Assessment of baroreceptor reflex function during mental stress and relaxation. Psychophysiology 26 (2), 140–147. Stunkard, A.J., Messick, S., 1985. The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J. Psychsom. Res. 29, 71–83. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996. Heart rate variability: standards of measurement, physiological interpretation, and clinical use. Eur. Heart J. 17 (3), 354–381. Telch, C.F., Agras, W.S., 1996. Do emotional states influence binge eating in the obese? Int. J. Eat. Disord. 20, 271–279. Tuschen-Caffier, B., Vogele, C., 1999. Psychological and physiological reactivity to stress: an experimental study on bulimic patients, restrained eaters, and controls. Psychother. Psychosom. 68, 333–340. Valensi, P., Thi, B.N., Lormeau, B., Paries, J., Attali, J.R., 1995. Cardiac autonomic function in obese patients. Int. J. Obes. Relat. Metab. Disord. 19, 113–118. Vogele, C., Florin, I., 1997. Psychophysiological responses to food exposure: an experimental study in binge eaters. Int. J. Eat. Dis. 21, 147–157. Wolff, G.E., Crosby, R.D., Roberts, J.A., Wittrock, D.A., 2000. Differences in daily stress, mood, coping, and eating behavior in binge eating and non binge eating college women. Addict. Behav. 25 (2), 205–216. Zigmond, A.S., Snaith, R.P., 1983. The hospital anxiety and depression scale. Acta. Psychiatr. Scand. 67, 361–370.