Decreased food pleasure and disrupted satiety signals in chronic low back pain

Ò

PAIN 155 (2014) 712–722

www.elsevier.com/locate/pain

Decreased food pleasure and disrupted satiety signals in chronic low back pain Paul Geha a,b,⇑, Ivan deAraujo a,b, Barry Green b, Dana M. Small a,b a b

Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA The John B. Pierce Laboratory, New Haven, CT, USA

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

a r t i c l e

i n f o

Article history: Received 6 September 2013 Received in revised form 10 December 2013 Accepted 20 December 2013

Keywords: Chronic back pain Obesity Pleasure Satiety

a b s t r a c t Chronic low back pain (CLBP) and obesity are interrelated, but the physiological mechanisms linking the 2 conditions remain to be determined. Functional brain imaging data from CLBP patients show functional and structural alterations in areas mediating the attribution of hedonic value to food. Accordingly, we hypothesized that CLBP patients would exhibit alteration in the hedonic perception of highly palatable, calorie-containing foods. CLBP patients and matched healthy controls initially rated their perception of highly palatable puddings of varying fat content and sugary drinks of varying sucrose content without ingesting significant amounts of either stimulus. In a subsequent intake test, hungry participants ingested their preferred pudding ad libitum. Compared to healthy controls, CLBP patients exhibited significantly lower ratings of food pleasure when sampling the fat puddings but not when sampling the sugary drinks. In contrast, the patients’ sensory evaluation of these stimuli was not different from those of healthy controls. In addition, whereas in healthy controls caloric intake from pudding closely matched hedonic ratings and decreased hunger after ad libitum pudding intake, such effect was totally abolished in CLBP patients. Our data thus reveal a decoupling between hedonic perception and fat calorie intake in CLBP patents, suggesting altered hedonic perception of fat as a potential mechanism linking CLBP to overeating and obesity. Published by Elsevier B.V. on behalf of International Association for the Study of Pain.

1. Introduction Chronic pain and obesity both constitute a huge burden to affected individuals and to society [28,43,66], and prevalence of both is increasing [32,77]. Evidence suggests that these conditions are interrelated. For example, the prevalence of obesity is higher in those with chronic pain [53,74,98,107], and the prevalence of chronic pain is higher in those who are obese [43,83,88]. This interaction is believed to negatively affect treatment response in chronic pain [87] and success rates of weight loss programs [104]. Despite its clinical relevance, little is known about the neurobiological mechanisms underlying the epidemiological association between chronic pain and obesity [50]. Mounting evidence indicates that the alarming increase in the prevalence of obesity results from an interaction between the abundance of palatable energy-dense foods that act to stimulate brain reward systems and individual variations in the responsivity of these systems [54,103]. Among the brain reward systems, the ⇑ Corresponding author. Address: Department of Psychiatry, Yale University School of Medicine, 290 Congress Ave, New Haven, CT 06519, USA. Tel.: +1 203 752 8256. E-mail address: [email protected] (P. Geha).

ventral striatum (VS) and medial prefrontal cortex (mPFC) have been consistently implicated as critical to the expression of appetitive and consummatory feeding behaviors [4,48,49,68,80,93]. Consistent with an association between obesity and chronic pain, chronic low back pain (CLBP) patients exhibit disrupted rewardrelated behaviors concomitantly to altered activity in VS and mPFC [2,6,9–11]. Moreover, in CLBP patients, back pain intensity ratings correlate with activity levels in VS and mPFC [6] and with VS–mPFC functional connectivity [9]. In particular, mPFC shows increased activity in CLBP patients compared to healthy subjects [5,8], an effect reversed by successful CLBP treatment [10,46,47]. More importantly, the strength of functional connectivity between VS and mPFC predicts the likelihood that a subacute back pain patient with backache for 6 to 12 weeks will seek care for back pain 1 year later [11]. Finally, VS volume shrinks only in subacute back pain patients whose pain persists after 1 year but not in those who recover [11]. Collectively, these results strongly support a role for VS–mPFC circuits in chronic pain. VS–mPFC circuits are also well known to mediate the attribution of hedonic value such as disliking and liking to negative and positive reinforcers, respectively [36,39,81,85], including palatable food [4,15,25,40,49,57–60,75,90,91,96]. Animal and human studies also show that in VS, both hedonic responses to food and pain relief

0304-3959/$36.00 Published by Elsevier B.V. on behalf of International Association for the Study of Pain. http://dx.doi.org/10.1016/j.pain.2013.12.027

Ò

P. Geha et al. / PAIN 155 (2014) 712–722

are mediated by l-opioid receptor signaling [4,86,93,97,110,111], which is in turn altered in different chronic pain conditions [29,45,64]. Interestingly, l-opioid receptor inverse agonist administration in humans decreases hedonic perception and caloric intake of palatable fat– or carbohydrate-rich foods but not of foods low in sugar and fat [72]. These findings support a critical role for VS and mPFC opioid signaling in the hedonic perception of food. Given the strong evidence for overlapping neural circuits for chronic pain and hedonic perception of palatable food, we set out to test whether patients with CLBP have altered hedonic perception of highly palatable foods. 2. Methods 2.1. Subjects All subjects provided written informed consent to participate in the study, which was approved by the Yale University institutional review board. Nineteen healthy subjects (4 men) and 18 CLBP patients (4 men) participated in this study. Subjects were recruited through flyers in the New Haven area and advertisements posted on the Internet. We received approximately 50 responses to our ads from patients with back pain. Thirty-three CLBP patients were screened in detail, 18 of whom participated in the study. Subjects were briefly screened at first to check (1) the location of pain, (2) whether they were otherwise healthy, (3) whether they were nonsmokers, and (4) whether they had pain duration of more than 2 years. If they passed this initial brief screening, a more detailed screening was conducted where we assessed demographics; location, possible cause, duration, and radiation of the pain; nonopiate analgesic medication use; medical assessments of the back pain; substance misuse; recent or past history of opioid medication use; complete medical and psychiatric history; recent or past fluctuations in body weight; history of olfactory or taste impairments; or nasal sinuses surgery. Healthy control subjects were likewise screened. In addition, the presence of any current back pain and any history of back pain of more than 6 weeks’ duration were exclusion criteria. Because we have a large database of healthy subjects, we used it to target recruitment of healthy participants whose gender, age, and body mass index (BMI) would be within the range of recruited CLBP patients. Participants had no history of psychiatric disorders, other medical conditions, and loss of consciousness, chemosensory impairment, or food allergies. To be included in the study, CLBP patients had to (1) fulfill the International Association for the Study of Pain criteria of chronic back pain [70], (2) not be currently, or during the month before the study, receiving any opioid analgesics, and (3) have a pain duration of at least 2 years. We chose to include patients with at least 2 years of back pain to make sure that they were in the time window when VS and mPFC alterations would have set in [11]. CLBP diagnosis was confirmed on the basis of history collected by experienced clinician (PG). Briefly, all patients had pain more days than not for more than 2 years, primarily localized to the lumbosacral region, including buttocks and thighs, with and without radiation. All participants were financially compensated $60 for taking part in both sessions of the study. 2.2. Stimuli A set of 4 pudding samples were prepared with 0%, 1.6%, 3.1%, and 6.9% fat weight by weight (w/w) [69]. The samples were prepared by mixing instant pudding (Jell-O, Kraft Foods) in milk (Guida’s Dairy) with varying fat content. The sugar content was held constant between the 4 stimuli at 4.6% (w/w). In order to maximize liking ratings, subjects were asked to pick a preferred flavor out of a choice of vanilla and chocolate during the prestudy screening

713

interview. A set of 4 Kool-Aid-based orange-flavored juices with 0, 0.018, 0.1, and 0.56 M sucrose concentration were also prepared. 2.3. Procedures Subjects were asked to participate in 2 sessions on 2 different days. During session 1, they sampled and rated food stimuli as described below. During session 2 they were offered a pudding to consume ad libitum. The maximum interval between the 2 sessions was 7 days. Session 1. Subjects presented to the laboratory between 9 AM and 3 PM. They were asked to arrive neither hungry nor full and to rate their hunger level upon arrival using a visual analog scale (VAS; 0 = ‘‘I am not hungry at all’’ and 100 = ‘‘I have never been more hungry’’). Testing continued only if hunger ratings were less than 30. If they rated hunger at greater than 30, they were given a small snack and were asked to wait 30 min, after which the hunger ratings were repeated. We thought it was important to test subjects in the absence of hunger or satiety in order to minimize homeostatic effects on food liking. Before testing, each subject was trained to use the General Labeled Magnitude Scale (gLMS) to rate overall intensity and sweetness [41], the Labeled Hedonic Scale (LHS) to rate liking or disliking [62], and the VAS to rate hunger, fullness, thirst, oiliness, fattiness, creaminess, and wanting of the stimuli. The gLMS is a computerized psychophysical tool that requires subjects to rate the perceived intensity of a stimulus along a vertical axis lined with adjectives that are spaced semilogarithmically on the basis of experimentally determined intervals to yield ratioquality data. The LHS was derived using similar methods as the gLMS but asks subjects to rate hedonic liking or disliking [62]. Subjects were then asked if they preferred the chocolate or vanilla pudding. The preferred pudding was used to conduct the remainder of the experiment. The different pudding or juice stimuli were presented in 3 blocks with the order of presentation randomized. Subjects sampled 5 mL of the juice and expectorated without swallowing; for the pudding, they sampled approximately 3 to 5 mL at the tip of a spoon without swallowing. After tasting each sample, subjects used the scales to rate their perceptions. They rinsed in between samples and paused for 30 s before taking the next sample. Subjects who chose chocolate pudding were asked to wear a blindfold during testing because we could not equate the color of the different concentrations. At the end of the session, another set of hunger, fullness, and thirst ratings was obtained. Session 2. Subjects presented hungry around lunchtime between noon and 2 pm. They were asked to eat breakfast and then refrain from eating anything until the time of testing. Subjects were tested only if their hunger level at arrival was rated >30 on the VAS. Otherwise they were rescheduled for a different day. First, percentage body fat was assessed using air displacement plethysmography (Bod-Pod; Cosmed). Because the percentage of body fat that is considered healthy differs in men and women (21%–25% range in men and 30%–35% range in women) [27,94], we divided the absolute output values from plethysmography by 31% and 21% for women and men, respectively. Two CLBP patients refused to undergo body fat mass assessment. Immediately after, subjects rated hunger, fullness, and thirst; they were then offered pudding and instructed to eat as much as they liked. For each subject, the pudding given a maximum liking rating during session 1 was used. Hunger, fullness, and thirst were rated after ad libitum pudding consumption. Subjects also provided ratings for intensity, liking, fullness, thirst, oiliness, fattiness, creaminess, and wanting after consumption. Questionnaires. Subjects were also asked to fill out feeding behavior questionnaires, and CLBP patients also completed painrelated questionnaires. CLBP was assessed using the following pain questionnaires and scales: VAS for pain intensity, the short form of

714

Ò

P. Geha et al. / PAIN 155 (2014) 712–722

the McGill Pain Questionnaire (MPQ) [67], the Washington Neuropathic Pain Scale (NPS) [34], and the pain DETECT questionnaire [33]. All subjects filled out the Beck Anxiety Index (BAI) [14], the Beck Depression Index (BDI) [13], and the Edinburgh Inventory Handedness Scale (EIHS). Feeding behavior was assessed in all subjects using the following questionnaires: impulsivity was measured with the Barratt Impulsiveness Scale Version 11 (BIS-11) [78]; eating style was measured with the Dutch Eating Behavior Questionnaire (DEBQ) [99], the Three Factor Eating Questionnaire (TFEQ) [95], the Power of Food Scale (PFS) [63], and the Binge Eating Scale (BES) [37]; reward sensitivity was measured with Behavioral Inhibition System/Behavioral Activation System [21]; and hours of physical exercise in the past 7 days was reported using the International Physical Activity Questionnaire [23]. Continuous back pain ratings. We collected continuous back pain ratings to verify self-reports of pain in the CLBP patients. Foss et al. [31] demonstrated that subjective rating of CLBP intensity without any outside stimulation over a span of 6 to 12 min has a fractal dimension D different from 1.5, indicating that it is not a random Gaussian process [82]. ‘‘Imagined pain’’ rating, on the other hand, reported by healthy subjects was a Gaussian random process, also called a ‘‘random walk,’’ with D = 1.5. Furthermore, CLBP intensity ratings also correlate with activity in the mPFC [5,6,10], suggesting that continuous subjective rating of intensity is reliable in capturing the CLBP patients’ subjective experience of pain. We therefore used the method to investigate whether our sample of CLBP patients pain ratings have D different from a random walk. Such a finding would support the diagnosis of CLBP and help rule out patients who falsely invent their symptoms. Time-varying signals have a Euclidian dimension (E) of 1. When they fluctuate nonperiodically, they can have fractal dimensions spanning between 1 and 2. The rougher their fluctuations, the higher their fractal dimension [79]. A pure random walk formed by up–down Gaussian independent steps has a fractal dimension of D = 1.5. Fractal time series are ‘‘scale free’’ over a certain range of time, meaning that the variance of the time series does not converge but continues to increase by a power law as sample length increases. We followed the same procedures for collecting continuous ratings of CLBP intensity described by Foss et al. [31] and used previously by the authors [10,36]. Patients indicate continuously their level of pain through a linear potentiometer device that is attached to the thumb and index finger of the dominant hand. Voltage output from the finger device is collected and calibrated by a computer running the software LabView (National Instruments). Patients are seated in front of a computer monitor, which displays the extent of their finger span by a colored bar (the y-axis has an intensity scale of 0 to 100), providing visual feedback of their rating. Ratings are sampled at 20 Hz. First, patients are trained to use the finger span device. To this end, they are presented with a moving bar on the computer screen that varies in time and are instructed to rate the length with the finger span device over a 1 min trial. Only subjects able to follow the bar at a consistency level that results in a correlation coefficient r > 0.75 between rating and bar fluctuations are included in the study, within 2 attempts. All our patients were able to achieve this criterion. Patients are then instructed to rate the fluctuations of their own ongoing back pain for a period of 6 min. They are instructed that the maximum thumb–finger span should be used to indicate maximum imaginable intensity of pain (level 100), while thumb and finger touching should indicate absence of pain (level 0). Calculation of fractal dimension. Fractal dimension was calculated by rescaled range analysis [30,65] and following per Foss et al. [31] Rescaled range analysis measures the extent to which the range R (ie, maximum minus minimum value of pain) spanned during a fluctuation trajectory and depends on the number of steps or time in the trajectory s. A characteristic of scale-free trajectories

is that R and s obey a power law, ie, R a sa. To empirically compute a, the scaling exponent for a given time series of length N samples, we determine the average R for different length scale s subsets of the original sample. s is varied from the largest possible, when s = N, to s = 8 samples. We find the length scale, R, by taking the distance between the maximum and minimum values of the subseries after detrending. We then scale R by the standard deviation S of the step size of this subseries because series with larger steps will naturally have a larger length scale, regardless of their scaling properties. We find the average value of R/S of all subseries at each time scale (for s = N, there is only 1 subseries; for s = N/2, there are 2 subseries, and so on) and use them to create the ‘‘scaling plot,’’ log10(R/S) as a function of log10(s). If the time series exhibit scale-free fluctuations, this relationship is linear with a slope a. The fractal dimension D is related to the scaling exponent by the relation D = 2 a [30]. In order to generate a set of control time series, we generated surrogate controls from pain ratings by reversing in time the recorded time series. This procedure preserves all the statistical properties of the original ratings but scrambles the relationship between the ratings and the actual pain fluctuations, thus controlling for nonspecific fluctuations. Statistical analysis. Statistical analysis was performed by Statistica 10 software (StatSoft). Between-group analysis was performed by unpaired t test and mixed 2-way ANOVAs with stimulus concentration as a repeated measure after averaging across the 3 presentations of the same stimulus concentration within each subject. Within-group analyses were performed by paired t tests. We used regression analyses to investigate the relationships between pudding rating, caloric consumption, and internal state. 3. Results 3.1. Subjects’ characteristics Demographic and clinical data for both CLBP patients and healthy participants are presented in Table 1. Individual patients’ clinical back pain data are presented in Table 2. We were not able to obtain continuous CLBP intensity ratings in 2 patients as a result of technical difficulties. Patient’s continuous CLBP ratings exhibited fractal properties (Fig. 1), with the average fractal dimension D = 1.54 ± 0.02 (n = 16), very close to the value (D = 1.55) published by Foss et al. [31] for a similar patient group. Patient’s rating D was significantly different (P = .002, paired t test) from their surrogate time series obtained by inverting their pain ratings in time that had an average D = 1.49 ± 0.001, which is very close to 1.5 the D of random Gaussian time series. The properties of the ratings therefore suggest that our patients were accurately classified as having CLBP. The average age, BMI, percentage body fat, and years of education did not differ between CLBP patients and healthy subjects (Table 1). However, we observed an accelerated positive increase in percentage body fat with age in CLBP patients compared to healthy subjects (Fig. 2). On average, both CLBP patients and healthy subjects had less than minimal depression (BDI score <9) and less than minimal anxiety (BAI score <7) with no significant difference between the groups (BDI, P = .69; BAI, P = .2; unpaired t test). The 2 groups did not differ in their eating style or in the average number of hours spent in physical activity during the past 7 days before testing (P = .53) [23] (Table 1). 3.2. CLBP alters the perception of affective, but not sensory, properties of palatable stimuli Internal state. During tasting session 1, CLBP patients and healthy subjects did not differ on hunger (F1,36 = 0.01, P = .94), fullness (F1,36 = 0.27, P = .61), or thirst (F1,36 = 2.00, P = .17) ratings (repeated measures ANOVA, group effects) associated with the

Ò

715

P. Geha et al. / PAIN 155 (2014) 712–722 Table 1 Demographic and clinical characteristics.a Parameter

CLBP

Healthy

P Value

Age, y Male sex n, (%) EHIS BMI, kg/m2 % Body fat Years of education BDI BAI Calories consumed BIS-11 BIS BAS TFEQ—Restraint TFEQ—Disinhibition TFEQ—Hunger PFS—Food availability PFS—Food presence PFS—Food tasted BES DEBQ IPAQ, h VAS pain Pain duration, y Pain DETECT questionnaire MPQ—Sensory, session 1 MPQ—Affective, session 1 MPQ—Sensory, session 2 MPQ—Affective, session 2 NPS—Total, session 1 NPS—Total, session 2

36.5 ± 2.5 4 (22.2%) 77.8 26.3 ± 1.7 1.0 ± 0.1 17.4 ± 1.1 5.9 ± 1.5 7.4 ± 1.9 205.7 ± 34.6 60.2 ± 2.3 3 ± 0.1 3.2 ± 0.1 9.5 ± 1.3 4.8 ± 0.9 4.7 ± 0.9 10.9 ± 1.2 12.7 ± 1.2 12.3 ± 0.8 5.5 ± 1.5 2.0 ± 0.9 174.8 ± 57.6 3.6 ± 0.5 8.2 ± 1.6 7.9 ± 1.3 8.7 ± 1.3 1.9 ± 0.5 4.1 ± 0.7 1.2 ± 0.5 24.8 ± 3.7 18.4 ± 3.3

34.5 ± 2.7 4 (22.2%) 86.9 27.1 ± 1.8 1.1 ± 0.1 15.4 ± 3.7 5.2 ± 1.4 4.4 ± 1.1 222.1 ± 45.8 56.7 ± 1.9 3.1 ± 0.2 3.2 ± 0.1 10.5 ± 1.2 5.4 ± 0.7 4.4 ± 0.9 12.3 ± 6.7 14.3 ± 6.3 12.1 ± 4.3 6.1 ± 1.6 2.1 ± 1.1 222.4 ± 48.9 ... ... ... ... ... ... ... ... ...

.72 .5 .5 .68 .45 .15 .69 .2 .77 .25 .9 .68 .59 .58 .8 .49 .41 .85 .79 .8 .53

b

CLBP, chronic low back pain; EHIS, Edinburgh Handedness Inventory Score; BMI, body mass index; BDI, Beck Depression Index; BAI, Beck Anxiety Index; BIS-11, Barratt Impulsiveness Scale; BIS, Behavioral Inhibition System; BAS, Behavioral Activation System; TFEQ, Three Factor Eating Questionnaire; PFS, Power of Food Scale; BES, Binge Eating Scale; DEBQ, Dutch Eating Behavioral Questionnaire; IPAQ, International Physical Activity Questionnaire; VAS, visual analog scale; MPQ, McGill Pain Questionnaire; NPS, Neuropathic Pain Scale. a Values are expressed as mean ± standard deviation or n (%). b Calculated by t test or v2 test.

pudding and juice stimuli. In addition, their ratings did not change between the beginning and the end of the sessions (time effect): hunger, F1,36 = 2.1, P = .16; fullness, F1,36 = 1.1, P = .30; F1,36 = 1.70 thirst, P = .20, including no detectable group by time interaction

effects: hunger, F1,35 = 1.36, P = .25; fullness, F1,35 = 0.70, P = .41; thirst, F1,35 = 0.44, P = .51 (Fig. 3, Table 3). Pudding ratings. CLBP patients and healthy subjects sampled puddings with 4 different concentrations of fat (0%, 1.6%, 3.1%, and 6.9% weight by weight) without ingestion during session 1. Given that we asked subjects to evaluate the stimuli over 7 psychophysical measures, we corrected for multiple comparisons by the Bonferroni method. Therefore, groups were considered significantly different if the P value was <.05/7 = 0.007. CLBP patients rated puddings as significantly less liked (mixed-measures ANOVA group effect: F1,35 = 8.75, P = .006) than healthy subjects (Fig. 4, Table 4). Group differences in wanting did not survive multiple comparisons’ correction (F1,35 = 5.46, P = .025) (Fig. 4, Table 4). In contrast, the 2 groups did not differ on their sensory ratings of the puddings’ intensity (F1,35 = 0.58, P = .45), sweetness (F1,35 = 1.73, P = .2), fattiness (F1,35 = 0.57, P = .46), creaminess (F1,35 = 0.07, P = .79), or oiliness (F1,35 = 0.07, P = .79) (Fig. 4, Table 4). These results indicate that CLBP patients differed mostly from healthy subjects when rating the stimuli’s affective property (ie, liking) but not when rating their sensory properties. CLBP patients and healthy subjects did not show Group  Concentration interactions for pudding ratings liking (F3,105 = 0.11, P = .95), wanting (F3,105 = 1.22, P = .31), sweetness (F3,105 = 1.17, P = .32), intensity (F3,105 = 0.20, P = .90), or fattiness (F3,105 = 0.36, P = .79) but showed a trend toward an interaction when rating creaminess (F3,105 = 2.70, P = .051) (Fig. 4, Table 4). Similar numbers of CLBP patients and controls chose chocolate pudding and wore the blindfold (9 CLBP and 12 controls) with no significant difference between the 2 proportions (P = .21). Thus, between-group choice patterns between chocolate and vanilla cannot account for the observed results. Kool-Aid ratings. CLBP patients and healthy subjects tasted orange-flavored Kool-Aid with 4 different concentrations of sucrose (0, 0.018, 0.1, and 0.56 M) without swallowing during session 1. Unlike puddings, CLBP and healthy subjects’ ratings of Kool-Aid did not differ on liking (F1,35 = 0.93, P = .34), wanting (F1,35 = 2.97, P = .094), sweetness (F1,35 = 4.15, P = .049), or intensity (F1,35 = 2.98, P = .093). There was no significant Group  Concentration interaction for liking (F3,105 = 1.44, P = .23), sweetness (F3,105 = 1.01, P = .39), or intensity (F3,105 = 0.42, P = .74) of the juices. Wanting ratings, on the other hand, showed a significant Group  Concentration interaction (F3,105 = 18.34, P < 10 5). This effect

Table 2 CLBP patients’ clinical characteristics. Patient code

Sex

Age, y

Duration, y

Clinical history

Pain radiation

VAS

MPQ—Affective score

MPQ—Sensory score

Neuropathic Pain Scale (total)

727 1086 1104 1124 1125 1143 1144 1148 1149 1150 1151 1152 1153 1163

F F F F F F F F F F M M F F

40 25 34 27 48 38 49 21 45 35 40 32 44 21

8 8 17 2.2 3 2 4 9.5 24 10 3 4 7 2.7

None None Above knee None None None None None None None None None None Above knee

5.2 3.5 3 2.5 6.6 3 7 3 5.8 0.7 1.8 0.7 0.7 3.5

0 2 0 0 4 1 1 1 4 4 1 1 1 1

3 8 5 8 18 4 5 7 9 15 3 8 6 9

40 19 14 5 24 32 23.5 29.5 23.5 17.5 10 8.5 4.5 24.5

1168 1169 1172 1174

F M M F

53 22 22 49

12 3.4 3.5 23.4

Nonspecific Nonspecific Nonspecific Nonspecific Nonspecific Nonspecific Nonspecific Scoliosis Nonspecific Scoliosis; corrective surgery Nonspecific Nonspecific Nonspecific Sports injury, L4–L5 disc degeneration Nonspecific Nonspecific Nonspecific Nonspecific

Below knee Above knee None Below knee

7 2.9 2.7 7

7 0 2 9

14 9 2 23

43 14 14 42.5

CLBP, chronic low back pain; VAS, visual analog scale; MPQ, McGill Pain Questionnaire.

716

Ò

P. Geha et al. / PAIN 155 (2014) 712–722

Fig. 1. Example of continuous CLBP ratings from 4 different patients. Pain intensity ratings were collected continuously over a period of 6 min. CLBP pain intensity fluctuation exhibited fractal properties, with a fractal dimension D different from 1.5, a value characteristic of random Gaussian time series. CLBP, chronic low back pain.

3.3. CLBP disrupts the association between subjective liking and caloric consumption

Fig. 2. CLBP patients exhibited accelerated increase in percentage of body fat with increasing age compared to healthy subjects. Regression analysis correcting for the effect of gender showed that the slope for the effect of increasing age on percentage of body fat was significant in CLBP patients (bCLBP = 0.027 ± 0.007, F1,13 = 12.6, P = .004) but not in healthy subjects (bhealthy = 0.004 ± 0.007, F1,16 = 0.4, P = .53), with the difference between the 2 slopes being significant (P = .02). CLBP, chronic low back pain.

arose because concentration influenced wanting ratings for controls (F3,54 = 35.69, P < 10 5) but not patients (F3,51 = 0.98, P = .41) (Fig. 4, Table 4).

On the day of the ad libitum feeding assessment (session 2), CLBP patients and healthy subjects once again did not differ on their hunger (CLBP, 66.0 ± 3.3; healthy, 69.5 ± 2.9; P = .42, unpaired t test), fullness (CLBP, 5.7 ± 2.8; healthy, 9.7 ± 3.1; P = .35) or thirst ratings (CLBP, 39.8 ± 6.0; healthy, 50.1 ± 5.7; P = .22) at presentation to the laboratory (Fig. 5A). CLBP patients’ and healthy subjects’ total caloric consumption from preferred pudding also did not significantly differ (calories for CLBP patients = 205.7 ± 34.6; calories for healthy subjects = 200.7 ± 43.1; P = .93) (Fig. 5C). Both groups showed a significant drop in hunger after consumption (CLBP, 41.4 ± 5.3; P < 10 5; healthy, 37.7 ± 5.6; P < 10 5, paired t test) (Fig. 5A, left), and increase in fullness (CLBP, 26.6 ± 5.7; P < .01; healthy, 36.4 ± 6.8; P < .01) (Fig. 5A, middle) but no significant change in thirst ratings (CLBP; P = .55; healthy; P = .47) after consumption (Fig. 5C, right). Change in hunger was similar in both groups because the Hunger  Group interaction was not significant (F1,34 = 0.95; P = .34, mixed 2-way ANOVA) (Fig. 5A, left). In addition, CLBP patients and healthy subjects did not show any significant difference in their ratings of the preferred pudding after

Ò

P. Geha et al. / PAIN 155 (2014) 712–722

717

Fig. 3. Bar plots of internal state ratings before and after session 1. There were no significant differences between CLBP patients and healthy subjects. CLBP, chronic low back pain.

Table 3 Internal state ratings for session 1.

4. Discussion

Internal state

Healthy

CLBP

Group effect

Hunger before Hunger after Fullness before Fullness after Thirst before Thirst after

26.5 ± 4.4 27.3 ± 5.2 40.4 ± 6.4 40.9 ± 6.9 38 ± 5.6 36.5 ± 5.3

22.1 ± 5.6 29.5 ± 5.5 33.8 ± 5.8 40.2 ± 5.9 25.9 ± 6.1 21.5 ± 5.5

NS NS NS

CLBP, chronic low back pain; NS, not significant at P < .05 for group effect repeatedmeasures ANOVA.

consumption (liking, P = .30; wanting, P = .34; sweetness, P = .73; intensity, P = .45; fattiness, P = .52; creaminess, P = .60; and oiliness, P = .57; unpaired t test) (Fig. 6). Given that pudding liking was significantly less in CLBP patients compared to healthy subjects (in session 1), we wanted to investigate whether liking of preferred pudding collected during session 1 determined caloric intake when participants came back for pudding consumption by regression analysis. Liking of preferred pudding was significantly correlated with caloric intake in healthy subjects (r2 = 0.36, P = .009) but not in CLBP patients (r2 = 0.06, P = .35), indicating that the relationship between liking of food and caloric ingestion is disrupted in the patient group (Fig. 5B). Caloric intake did not, however, depend on liking ratings collected during session 2 after ingestion for either healthy subjects (P = .31) or CLBP patients (P = .75). To address whether the choice of flavor influences the relationship between liking ratings collected during session 1 and caloric intake, we used a general linear model where caloric intake was the dependent variable, flavor a categorical predictor, and liking ratings from session 1 a continuous predictor. Liking ratings significantly predicted caloric intake in healthy subjects (F1,16 = 8.2; r2 = 0.35; P = .012) but not in CLBP patients (F1,15 = 0.08; r2 = 0.00; P = .79). There was no effect of flavor on caloric intake in either healthy (F1,16 = 0.01; P = .91) or CLBP participants (F1,15 = 0.43; P = .52). Therefore, flavor identity (and hence the blindfold) did not account for the relationship between liking and caloric intake. 3.4. CLBP disrupts the association between changes in hunger ratings and caloric consumption Next, we investigated whether caloric intake depended on the change in hunger ratings after consumption of the preferred pudding in session (drops in hunger ratings = hunger before consumption hunger after consumption). Whereas in healthy subjects drops in hunger ratings was significantly associated with caloric intake (r2 = 0.57; P < 10 3), this effect was completely absent in CLBP patients (r2 = 0.01; P = .69) (Fig. 5D). This result points to a decoupling between subjective hunger and caloric intake in CLBP patients.

CLBP and obesity are interrelated, but the physiological mechanisms linking the 2 conditions remain to be determined. Neuroimaging data from CLBP patients reveal functional and structural alterations in areas mediating the attribution of hedonic value to food [5,6,8,9,11]. We therefore speculated that CLBP may be associated with altered hedonic responses to food, which could in turn affect feeding behavior to promote weight gain. Accordingly, the aim of this work was to test whether hedonic responses to food are disrupted in CLBP patients. In keeping with our hypothesis, we observed that CLBP patients displayed blunted hedonic reactions to palatable puddings. This finding was specific because, first, ratings of the sensory properties of the puddings were unaffected by CLBP, and second, the hedonic blunting did not generalize to the sugar-containing drinks. In addition, whereas pudding liking and drops in hunger ratings after pudding consumption correlated with caloric intake in healthy subjects, no such relationship was found in CLBP patients, suggesting a disruption in hedonic control of palatable food intake and deficient satiety mechanisms in this population. 4.1. Blunted hedonic responses to fat/sugar puddings in CLBP We found that CLBP patients report significantly lower liking ratings for palatable puddings compared to individuals in a matched control group. Whereas control subjects rated the puddings as between ‘‘like moderately’’ and ‘‘like very much,’’ CLBP patients rated the puddings between ‘‘like slightly’’ and ‘‘like moderately.’’ Thus, the statistically significant differences are also perceptually meaningful. In contrast, no differences were observed between CLBP patients and healthy controls on sensory ratings of intensity, sweetness, creaminess, fattiness, and oiliness, indicating that the difference observed in hedonic ratings is not related to a deficit in oral detection of food. In a previous work, Small and Apkarian [89] demonstrated that CLBP patients exhibit increased oral sensitivity to pure gustatory stimuli. We did not observe this increase in the current study, possibly because, unlike the previous work, we used complex flavored food stimuli. This behavioral finding is in accordance with the well-documented alterations in the VS–mPFC circuitry in CLBP patients [5,6,8–11] and the known role of this circuit in the attribution of hedonic value to food. It is believed that the mPFC encodes the subjective value of all aversive and rewarding stimuli on a common scale [38,39,61,71,85]; notably, the VS receives dense afferent inputs from this region [44]. Ratings of experienced pleasantness correlate strongly with activity in both regions [60] and in particular with pleasantness ratings of fat taste [40,84]. Altered activity and connectivity in the VS–mPFC circuitry observed in CLBP patients [5,6,8,9,11] could therefore disrupt hedonic perception of highly palatable food and result in anhedonia.

718

Ò

P. Geha et al. / PAIN 155 (2014) 712–722

Fig. 4. Psychophysical ratings for CLBP patients (n = 18) and healthy control subjects (n = 19) during session 1. (A–D) (Left) Participants’ ratings for each of the 4 fat concentrations of puddings. (Middle) Ratings for each of the 4 sucrose concentrations of orange-flavored Kool-Aid. (Far right) Bar plot depicting mean ± standard error of the mean (SEM) across all the concentrations for pudding (left) and juice (right). Puddings’ fattiness, creaminess, and oiliness ratings are depicted in E, F, and G, respectively. ⁄ P < .007. CLBP, chronic low back pain.

Notably, group differences in affective perception emerged only for the more palatable high fat/high sugar pudding and not for the Kool-Aid drinks, which were rated overall less liked and did not contain fat. Mu-opioid agonist binding in the

VS plays an important role in coding food palatability [4,15,96,97,108,109], especially in the case of fat [105,108,109] and blockade of l-opioid receptors leads to a specific decrease in the ingestion of highly palatable food [72,97,108]. Critically,

Ò

719

P. Geha et al. / PAIN 155 (2014) 712–722 Table 4 Comparison of CLBP patients’ and healthy subjects’ ratings of puddings and Kool-Aid during session 1.a Food

Group, F1,35

Pudding Liking Wanting Sweetness Intensity Fattiness Creaminess Oiliness

8.75* 5.46 1.73 0.58 0.57 0.07 0.07

Kool-Aid Liking Wanting Sweetness Intensity

0.93 2.97 4.15 2.98

Stimulus concentration, F3,105 2.38 2.69 1.64 0.43 4.3 7.28* 1.59 62.66* 27.07* 135.18* 12.61*

Group  Concentration, F3,105 0.11 1.22 1.17 0.2 0.36 2.68 1.22 1.44 18.34* 1.01 0.42

CLBP, chronic low back pain. a F Values are results of a mixed 2-way ANOVA where group (CLBP vs healthy) was a factor and stimulus concentration the repeated measure. P < .007.

*

Fig. 5. Results for session 2. (A) Hunger of healthy participants and CLBP patients (left) dropped, and fullness (middle) increased significantly after consumption of the preferred pudding ad libitum. Thirst (right) ratings did not change significantly after consumption or differ between the groups (C) Healthy participants and CLBP patients consumed the same amount of calories. Relationship between caloric intakes from the most preferred pudding during the ad libitum consumption and pudding liking (B) and drop in hunger after consumption (D). ⁄⁄P < 10 3,  P < .01. CLBP, chronic low back pain.

decreased l-opioid receptor availability in the VS is a consistent finding in chronic pain patients [29,45,64], raising the possibility that alterations in opioid signaling associated with chronic pain may also give rise to the reduced pleasure derived from the highly palatable food that we observed in the current study. However, l-opioid binding in the parabrachial nucleus [22,56], periaqueductal gray matter [1,20,76,112], hypothalamus [3,111], amygdala [52,73,106,111], and insula [7,18,25,26,35,51] also play a role in both food reward and pain. Therefore, it is possible that these circuits might also contribute to the blunted hedonic responses observed in CLBP patients.

4.2. Hedonic control of feeding is disrupted in CLBP Anhedonia, defined as reduced sensitivity to reward, has been linked to obesity [24] and failure of weight loss interventions [55]. It is therefore possible that the hedonic blunting we observed underlies the association between obesity and CLBP. In keeping with this possibility, we found that liking ratings correlated with caloric intake in the control group but not in the CLBP group. This suggests the existence of a disruption in the hedonic control of palatable food intake in CLBP. In healthy individuals, food palatability is an important determinant of food choice [19]. We also observed

Ò

720

P. Geha et al. / PAIN 155 (2014) 712–722

Acknowledgments Supported in part by pilot funds from the John B. Pierce Laboratory. We thank Dr Julie Mennella for sharing her stimulus preparation protocol with us. We also thank Drs A. Vania Apkarian and M.N. Baliki for sharing with us their functional brain imaging results of CLBP patients. Finally, we thank all the members of the Neuropsychology and Physiology of Flavor and Feeding Laboratory for helping with subject recruitment and preparation of food stimuli. References

Fig. 6. Bar plots for psychophysical ratings of the consumed pudding in session 2. None of the measures differed between CLBP patients and healthy controls. CLBP, chronic low back pain; VAS, visual analog scale.

evidence for a decoupling of hunger perceptions and intake. Specifically, decreases in hunger ratings after pudding consumption correlated with caloric intake in healthy subjects, but no such relationship was found in CLBP patients. Thus, our findings suggest that hedonic and homeostatic control of feeding is disrupted in CLBP. Exactly how these changes are related to the association between obesity and CLBP remains unanswered. One possibility is that decreasing the salience of a reward value, coupled with decreasing the ability of signals that are critical for goal-directed behavior, allow habitual response to dominate behavior [12]. Another possibility is that the blunted liking results in CLBP patients requiring more palatable food to reach the same level of reward [16]. It is important to note that the disruption we observed between drop in hunger and caloric intake in patients occurs across subjects. Therefore, we cannot rule out the possibility that the drop in hunger and caloric intake could be significantly correlated within an individual CLBP patient tested across multiple different sessions. Importantly, BMI and adiposity were not different between patients and healthy subjects, so these factors are unlikely to account for the present findings. Nevertheless, we observed increased age-dependent body fat gain in the chronic pain group, supporting the possibility that CLBP leads to overeating. Finally, CLBP patients and healthy subjects did not differ on any of the questionnaires assessing their eating style, demonstrating that our samples were well matched for self-reported eating habits. 4.3. Summary We found blunted hedonic responses to palatable food and a disruption of the ability of hedonic and internal state signals to guide intake in CLBP patients. Given the lack of evidence for the fear avoidance model [17,42,92,100–102], which posits that higher rates of obesity are associated with reduced physical activity to avoid pain, we suggest that future research exploring the role of altered hedonic signals is warranted. In particular, one important question that remains to be determined is whether the alterations in food reward observed here account for the association between obesity and CLBP. Conflict of interest The authors report no conflict of interest.

[1] Anseloni VC, Ren K, Dubner R, Ennis M. A brainstem substrate for analgesia elicited by intraoral sucrose. Neuroscience 2005;133:231–43. [2] Apkarian AV, Sosa Y, Krauss BR, Thomas PS, Fredrickson BE, Levy RE, Harden R, Chialvo DR. Chronic pain patients are impaired on an emotional decisionmaking task. PAINÒ 2004;108:129–36. [3] Atweh SF, Kuhar MJ. Autoradiographic localization of opiate receptors in rat brain. II. The brain stem. Brain Res 1977;129:1–12. [4] Baldo BA, Pratt WE, Kelley AE. Control of fat intake by striatal opioids. In: Montmayeur JP, le Coutre J, editors. Fat detection: taste, texture, and post ingestive effects. Boca Raton, FL: CRC Press; 2010 [chapter 13]. [5] Baliki MN, Baria AT, Apkarian AV. The cortical rhythms of chronic back pain. J Neurosci 2011;31:13981–90. [6] Baliki MN, Chialvo DR, Geha PY, Levy RM, Harden RN, Parrish TB, Apkarian AV. Chronic pain and the emotional brain: specific brain activity associated with spontaneous fluctuations of intensity of chronic back pain. J Neurosci 2006;26:12165–73. [7] Baliki MN, Geha PY, Apkarian AV. Parsing pain perception between nociceptive representation and magnitude estimation. J Neurophysiol 2009;101:875–87. [8] Baliki MN, Geha PY, Apkarian AV, Chialvo DR. Beyond feeling: chronic pain hurts the brain, disrupting the default-mode network dynamics. J Neurosci 2008;28:1398–403. [9] Baliki MN, Geha PY, Fields HL, Apkarian AV. Predicting value of pain and analgesia: nucleus accumbens response to noxious stimuli changes in the presence of chronic pain. Neuron 2010;66:149–60. [10] Baliki MN, Geha PY, Jabakhanji R, Harden N, Schnitzer TJ, Apkarian AV. A preliminary fMRI study of analgesic treatment in chronic back pain and knee osteoarthritis. Mol Pain 2008;4:47. [11] Baliki MN, Petre B, Torbey S, Herrmann KM, Huang L, Schnitzer TJ, Fields HL, Apkarian AV. Corticostriatal functional connectivity predicts transition to chronic back pain. Nat Neurosci 2012;14:1117–9. [12] Balleine BW, O’Doherty JP. Human and rodent homologies in action control: corticostriatal determinants of goal-directed and habitual action. Neuropsychopharmacology 2010;35:48–69. [13] Beck A, Steer R. Beck depression inventory. San Antonio: Psychological Corporation; 1993. [14] Beck A, Steer R. Manual for the beck anxiety inventory. San Antonio: Psychological Corporation; 1993. [15] Berridge KC. ‘‘Liking’’ and ‘‘wanting’’ food rewards: brain substrates and roles in eating disorders. Physiol Behav 2009;97:537–50. [16] Blum K, Gardner E, Oscar-Berman M, Gold M. ‘‘Liking’’ and ‘‘wanting’’ linked to reward deficiency syndrome (RDS): hypothesizing differential responsivity in brain reward circuitry. Curr Pharm Des 2012;18:113–8. [17] Bousema EJ, Verbunt JA, Seelen HA, Vlaeyen JW, Knottnerus JA. Disuse and physical deconditioning in the first year after the onset of back pain. PAINÒ 2007;130:279–86. [18] Burkey AR, Carstens E, Jasmin L. Dopamine reuptake inhibition in the rostral agranular insular cortex produces antinociception. J Neurosci 1999;19:4169–79. [19] Caltabiano ML, Shellshear J. Palatability versus healthiness as determinants of food preferences in young adults: a comparison of nomothetic and idiographic analytic approaches. Aust N Z J Public Health 1998;22:547–51. [20] Carstens E, Yokota T, Zimmermann M. Inhibition of spinal neuronal responses to noxious skin heating by stimulation of mesencephalic periaqueductal gray in the cat. J Neurophysiol 1979;42:558–68. [21] Carver CS, White TL. Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: the BIS/BAS scales. J Pers Soc Psychol 1994;67:319–33. [22] Coizet V, Dommett EJ, Klop EM, Redgrave P, Overton PG. The parabrachial nucleus is a critical link in the transmission of short latency nociceptive information to midbrain dopaminergic neurons. Neuroscience 2010;168:263–72. [23] Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, Pratt M, Ekelund U, Yngve A, Sallis JF, Oja P. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003;35:1381–95. [24] Davis C, Strachan S, Berkson M. Sensitivity to reward: implications for overeating and overweight. Appetite 2004;42:131–8. [25] De Araujo IE, Rolls ET. Representation in the human brain of food texture and oral fat. J Neurosci 2004;24:3086–93.

Ò

P. Geha et al. / PAIN 155 (2014) 712–722 [26] De Araujo IE, Geha PY, Small DM. Orosensory and homeostatic functions of the insular taste cortex. Chem Percept 2012;5:64–79. [27] De Lorenzo A, Bianchi A, Maroni P, Iannarelli A, Di Daniele N, Iacopino L, Di Renzo L. Adiposity rather than BMI determines metabolic risk. Int J Cardiol 2013;166:111–7. [28] Deyo RA, Mirza SK, Turner JA, Martin BI. Overtreating chronic back pain: time to back off? J Am Board Fam Med 2009;22:62–8. [29] DosSantos MF, Martikainen IK, Nascimento TD, Love TM, Deboer MD, Maslowski EC, Monteiro AA, Vincent MB, Zubieta JK, DaSilva AF. Reduced basal ganglia mu-opioid receptor availability in trigeminal neuropathic pain: a pilot study. Mol Pain 2012;8:74. [30] Feder J. Fractals: physics of solids and liquids. New York: Plenum Press; 1988. [31] Foss JM, Apkarian AV, Chialvo DR. Dynamics of pain: fractal dimension of temporal variability of spontaneous pain differentiates between pain States. J Neurophysiol 2006;95:730–6. [32] Freburger JK, Holmes GM, Agans RP, Jackman AM, Darter JD, Wallace AS, Castel LD, Kalsbeek WD, Carey TS. The rising prevalence of chronic low back pain. Arch Intern Med 2009;169:251–8. [33] Freynhagen R, Baron R, Gockel U, Tolle TR. PainDETECT: a new screening questionnaire to identify neuropathic components in patients with back pain. Curr Med Res Opin 2006;22:1911–20. [34] Galer BS, Jensen MP. Development and preliminary validation of a pain measure specific to neuropathic pain: the Neuropathic Pain Scale. Neurology 1997;48:332–8. [35] Geha PY, Baliki MN, Chialvo DR, Harden RN, Paice JA, Apkarian AV. Brain activity for spontaneous pain of postherpetic neuralgia and its modulation by lidocaine patch therapy. PAINÒ 2007;128:88–100. [36] Glimcher PW. Understanding dopamine and reinforcement learning: the dopamine reward prediction error hypothesis. Proc Natl Acad Sci U S A 2011;108:15647–54. [37] Gormally J, Black S, Daston S, Rardin D. The assessment of binge eating severity among obese persons. Addic Behav 1982;7:47–55. [38] Grabenhorst F, D’Souza AA, Parris BA, Rolls ET, Passingham RE. A common neural scale for the subjective pleasantness of different primary rewards. Neuroimage 2010;51:1265–74. [39] Grabenhorst F, Rolls ET. Value, pleasure and choice in the ventral prefrontal cortex. Trends Cogn Sci 2011;15:56–67. [40] Grabenhorst F, Rolls ET, Parris BA, d’Souza AA. How the brain represents the reward value of fat in the mouth. Cereb Cortex 2010;20:1082–91. [41] Green BG, Dalton P, Cowart B, Shaffer G, Rankin K, Higgins J. Evaluating the ‘‘Labeled Magnitude Scale’’ for measuring sensations of taste and smell. Chem Senses 1996;21:323–34. [42] Griffin DW, Harmon DC, Kennedy NM. Do patients with chronic low back pain have an altered level and/or pattern of physical activity compared to healthy individuals? A systematic review of the literature. Physiotherapy 2012;98:13–23. [43] Guh DP, Zhang W, Bansback N, Amarsi Z, Birmingham CL, Anis AH. The incidence of co-morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC Public Health 2009;9:88. [44] Haber SN, Knutson B. The reward circuit: linking primate anatomy and human imaging. Neuropsychopharmacology 2010;35:4–26. [45] Harris RE, Clauw DJ, Scott DJ, McLean SA, Gracely RH, Zubieta JK. Decreased central mu-opioid receptor availability in fibromyalgia. J Neurosci 2007;27:10000–6. [46] Hashmi JA, Baliki MN, Huang L, Parks EL, Chanda ML, Schnitzer T, Apkarian AV. Lidocaine patch (5%) is no more potent than placebo in treating chronic back pain when tested in a randomised double blind placebo controlled brain imaging study. Mol Pain 2012;8:29. [47] Hashmi JA, Baria AT, Baliki MN, Huang L, Schnitzer TJ, Apkarian AV. Brain networks predicting placebo analgesia in a clinical trial for chronic back pain. PAINÒ 2012;153:2393–402. [48] Humphries MD, Prescott TJ. The ventral basal ganglia, a selection mechanism at the crossroads of space, strategy, and reward. Prog Neurobiol 2010;90:385–417. [49] Ikemoto S. Dopamine reward circuitry: two projection systems from the ventral midbrain to the nucleus accumbens–olfactory tubercle complex. Brain Res Rev 2007;56:27–78. [50] Janke EA, Collins A, Kozak AT. Overview of the relationship between pain and obesity: what do we know? Where do we go next? J Rehabil Res Dev 2007;44:245–62. [51] Jasmin L, Burkey AR, Granato A, Ohara PT. Rostral agranular insular cortex and pain areas of the central nervous system: a tract-tracing study in the rat. J Comp Neurol 2004;468:425–40. [52] Johansen JP, Cain CK, Ostroff LE, LeDoux JE. Molecular mechanisms of fear learning and memory. Cell 2011;147:509–24. [53] Kadam UT, Jordan K, Croft PR. Clinical comorbidity in patients with osteoarthritis: a case-control study of general practice consulters in England and Wales. Ann Rheum Dis 2004;63:408–14. [54] Kenny PJ. Reward mechanisms in obesity: new insights and future directions. Neuron 2011;69:664–79. [55] Keranen AM, Rasinaho E, Hakko H, Savolainen M, Lindeman S. Eating behavior in obese and overweight persons with and without anhedonia. Appetite 2010;55:726–9. [56] Klop EM, Mouton LJ, Hulsebosch R, Boers J, Holstege G. In cat four times as many lamina I neurons project to the parabrachial nuclei and twice as many

[57] [58]

[59] [60] [61] [62] [63]

[64]

[65] [66]

[67] [68]

[69]

[70] [71]

[72]

[73] [74]

[75]

[76]

[77]

[78] [79] [80]

[81]

[82]

[83]

[84] [85]

[86]

721

to the periaqueductal gray as to the thalamus. Neuroscience 2005;134:189–97. Kringelbach ML. The human orbitofrontal cortex: linking reward to hedonic experience. Nat Rev Neurosci 2005;6:691–702. Kringelbach ML, O’Doherty J, Rolls ET, Andrews C. Activation of the human orbitofrontal cortex to a liquid food stimulus is correlated with its subjective pleasantness. Cereb Cortex 2003;13:1064–71. Kringelbach ML, Stein A, van Hartevelt TJ. The functional human neuroanatomy of food pleasure cycles. Physiol Behav 2012;106:307–16. Kuhn S, Gallinat J. The neural correlates of subjective pleasantness. Neuroimage 2012;61:289–94. Levy DJ, Glimcher PW. The root of all value: a neural common currency for choice. Curr Opin Neurobiol 2012;22:1027–38. Lim J, Wood A, Green BG. Derivation and evaluation of a labeled hedonic scale. Chem Senses 2009;34:739–51. Lowe MR, Butryn ML, Didie ER, Annunziato RA, Thomas JG, Crerand CE, Ochner CN, Coletta MC, Bellace D, Wallaert M, Halford J. The Power of Food Scale. A new measure of the psychological influence of the food environment. Appetite 2009;53:114–8. Maarrawi J, Peyron R, Mertens P, Costes N, Magnin M, Sindou M, Laurent B, Garcia-Larrea L. Differential brain opioid receptor availability in central and peripheral neuropathic pain. PAINÒ 2007;127:183–94. Mandelbrot BB. The fractal geometry of nature. New York: W. H. Freeman; 1982. Mehra M, Hill K, Nicholl D, Schadrack J. The burden of chronic low back pain with and without a neuropathic component: a healthcare resource use and cost analysis. J Med Econ 2012;15:245–52. Melzack R. The short-form McGill Pain Questionnaire. PAINÒ 1987;30: 191–7. Mena JD, Sadeghian K, Baldo BA. Induction of hyperphagia and carbohydrate intake by mu-opioid receptor stimulation in circumscribed regions of frontal cortex. J Neurosci 2011;31:3249–60. Mennella JA, Finkbeiner S, Reed DR. The proof is in the pudding: children prefer lower fat but higher sugar than do mothers. Int J Obesity 2012;36:1285–91. Merskey H, Bogduk N. Classification of chronic pain. 2nd ed. Seattle: IASP; 1994. Monosov IE, Hikosaka O. Regionally distinct processing of rewards and punishments by the primate ventromedial prefrontal cortex. J Neurosci 2012;32:10318–30. Nathan PJ, O’Neill BV, Bush MA, Koch A, Tao WX, Maltby K, Napolitano A, Brooke AC, Skeggs AL, Herman CS, Larkin AL, Ignar DM, Richards DB, Williams PM, Bullmore ET. Opioid receptor modulation of hedonic taste preference and food intake: a single-dose safety, pharmacokinetic, and pharmacodynamic investigation with GSK1521498, a novel mu-opioid receptor inverse agonist. J Clin Pharmacol 2012;52:464–74. Neugebauer V, Li W, Bird GC, Han JS. The amygdala and persistent pain. Neuroscientist 2004;10:221–34. Neumann L, Lerner E, Glazer Y, Bolotin A, Shefer A, Buskila D. A crosssectional study of the relationship between body mass index and clinical characteristics, tenderness measures, quality of life, and physical functioning in fibromyalgia patients. Clin Rheumatol 2008;27:1543–7. Niznikiewicz MA, Delgado MR. Two sides of the same coin: learning via positive and negative reinforcers in the human striatum. Dev Cogn Neurosci 2011;1:494–505. Nolan-Poupart S, Veldhuizen MG, Geha P, Small DM. Midbrain response to milkshake correlates with ad libitum milkshake intake in the absence of hunger. Appetite 2013;60:168–74. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 2006;295:1549–55. Patton JH, Stanford MS, Barratt ES. Factor structure of the Barratt impulsiveness scale. J Clin Psychol 1995;51:768–74. Peitgen HO, Jurgens H, Saupe D. Chaos and fractals: new frontiers of science. New York: Springer; 1992. Petrovich GD, Ross CA, Holland PC, Gallagher M. Medial prefrontal cortex is necessary for an appetitive contextual conditioned stimulus to promote eating in sated rats. J Neurosci 2007;27:6436–41. Plassmann H, O’Doherty JP, Rangel A. Appetitive and aversive goal values are encoded in the medial orbitofrontal cortex at the time of decision making. J Neurosci 2010;30:10799–808. Rangarajan G, Ding M. Integrated approach to the assessment of long range correlation in time series data. Phys Rev E Stat Phys Plasmas Fluids Relat Interdiscip Topics 2000;61:4991–5001. Ray L, Lipton RB, Zimmerman ME, Katz MJ, Derby CA. Mechanisms of association between obesity and chronic pain in the elderly. PAINÒ 2011;152:53–9. Rolls ET, McCabe C. Enhanced affective brain representations of chocolate in cravers vs non-cravers. Eur J Neurosci 2007;26:1067–76. Rushworth MF, Noonan MP, Boorman ED, Walton ME, Behrens TE. Frontal cortex and reward-guided learning and decision-making. Neuron 2011;70:1054–69. Scott DJ, Heitzeg MM, Koeppe RA, Stohler CS, Zubieta JK. Variations in the human pain stress experience mediated by ventral and dorsal basal ganglia dopamine activity. J Neurosci 2006;26:10789–95.

722

Ò

P. Geha et al. / PAIN 155 (2014) 712–722

[87] Sellinger JJ, Clark EA, Shulman M, Rosenberger PH, Heapy AA, Kerns RD. The moderating effect of obesity on cognitive-behavioral pain treatment outcomes. Pain Med 2010;11:1381–90. [88] Shiri R, Karppinen J, Leino-Arjas P, Solovieva S, Viikari-Juntura E. The association between obesity and low back pain: a meta-analysis. Am J Epidemiol 2010;171:135–54. [89] Small DM, Apkarian AV. Increased taste intensity perception exhibited by patients with chronic back pain. PAINÒ 2006;120:124–30. [90] Small DM, Gregory MD, Mak YE, Gitelman D, Mesulam MM, Parrish T. Dissociation of neural representation of intensity and affective valuation in human gustation. Neuron 2003;39:701–11. [91] Small DM, Zatorre RJ, Dagher A, Evans AC, Jones-Gotman M. Changes in brain activity related to eating chocolate: from pleasure to aversion. Brain 2001;124:1720–33. [92] Smeets RJ, van Geel KD, Verbunt JA. Is the fear avoidance model associated with the reduced level of aerobic fitness in patients with chronic low back pain? Arch Phys Med Rehabil 2009;90:109–17. [93] Smith KS, Berridge KC. Opioid limbic circuit for reward: interaction between hedonic hotspots of nucleus accumbens and ventral pallidum. J Neurosci 2007;27:1594–605. [94] Snitker S. Use of body fatness cutoff points. Mayo Clin Proc 2010;85:1057. author reply 1057–8. [95] Stunkard AJ, Messick S. The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res 1985;29:71–83. [96] Taha SA, Katsuura Y, Noorvash D, Seroussi A, Fields HL. Convergent, not serial, striatal and pallidal circuits regulate opioid-induced food intake. Neuroscience 2009;161:718–33. [97] Taha SA, Norsted E, Lee LS, Lang PD, Lee BS, Woolley JD, Fields HL. Endogenous opioids encode relative taste preference. Eur J Neurosci 2006;24:1220–6. [98] Ursini F, Naty S, Grembiale RD. Fibromyalgia and obesity: the hidden link. Rheumatol Int 2011;31:1403–8. [99] Van ST, Rookus MA, Bergers GP, Frijters JE, Defares PB. Life events, emotional eating and change in body mass index. Int J Obes 1986;10:29–35. [100] van Weering MG, Vollenbroek-Hutten MM, Tonis TM, Hermens HJ. Daily physical activities in chronic lower back pain patients assessed with accelerometry. Eur J Pain 2009;13:649–54.

[101] Verbunt JA, Seelen HA, Vlaeyen JW, van de Heijden GJ, Heuts PH, Pons K, Knottnerus JA. Disuse and deconditioning in chronic low back pain: concepts and hypotheses on contributing mechanisms. Eur J Pain 2003;7:9–21. [102] Verbunt JA, Smeets RJ, Wittink HM. Cause or effect? Deconditioning and chronic low back pain. PAINÒ 2010;149:428–30. [103] Volkow ND, Wise RA. How can drug addiction help us understand obesity? Nat Neurosci 2005;8:555–60. [104] Wachholtz A, Binks M, Eisenson H, Kolotkin R, Suzuki A. Does pain predict interference with daily functioning and weight loss in an obese residential treatment-seeking population? Int J Behav Med 2010;17:118–24. [105] Will MJ, Franzblau EB, Kelley AE. Nucleus accumbens mu-opioids regulate intake of a high-fat diet via activation of a distributed brain network. J Neurosci 2003;23:2882–8. [106] Will MJ, Franzblau EB, Kelley AE. The amygdala is critical for opioid-mediated binge eating of fat. Neuroreport 2004;15:1857–60. [107] Wilson AC, Samuelson B, Palermo TM. Obesity in children and adolescents with chronic pain: associations with pain and activity limitations. Clin J Pain 2010;26:705–11. [108] Zhang M, Gosnell BA, Kelley AE. Intake of high-fat food is selectively enhanced by mu opioid receptor stimulation within the nucleus accumbens. J Pharmacol Exp Ther 1998;285:908–14. [109] Zhang M, Kelley AE. Opiate agonists microinjected into the nucleus accumbens enhance sucrose drinking in rats. Psychopharmacology (Berl) 1997;132:350–60. [110] Zubieta JK, Bueller JA, Jackson LR, Scott DJ, Xu Y, Koeppe RA, Nichols TE, Stohler CS. Placebo effects mediated by endogenous opioid activity on muopioid receptors. J Neurosci 2005;25:7754–62. [111] Zubieta JK, Smith YR, Bueller JA, Xu Y, Kilbourn MR, Jewett DM, Meyer CR, Koeppe RA, Stohler CS. Regional mu opioid receptor regulation of sensory and affective dimensions of pain. Science 2001;293:311–5. [112] Zubieta JK, Stohler CS. Neurobiological mechanisms of placebo responses. Ann N Y Acad Sci 2009;1156:198–210.