Effects of Opioid Blockade With Naltrexone and Distraction On Cold and Ischemic Pain in Hypertension

 

C  2006)  Journal of Behavioral Medicine, Vol. 30, No. 1, February 2007 (  DOI: 10.1007/s10865-006-9084-1

Effects of Opioid Blockade with Naltrexone and Distraction on Cold and Ischemic Pain in Hypertension Christopher Ring,1,5 Christopher R. France,2 Mustafa al’Absi,3 Louise Beesley,4 Louisa Edwards,1 David McIntyre,1 Douglas Carroll,1 and Una Martin4  Accepted for publication: October 16, 2006 Published online: January 5, 2007 

Essential hypertension is characterised by reduced pain sensitivity. Hypertensive hypoalgesia has been attributed to elevated endogenous opioids and/or increased activation of descending pain modulation systems. A double-blind placebo-controlled design compared the effects of naltrexone and placebo on cold and ischemic pain in unmedicated newly-diagnosed patients with essential hypertension. Patients performed a cold pressor task while resting and while performing a distracting secondary task. They also performed a forearm ischemia task while resting. resting. Altho Although ugh the cold pressor and ischem ischemia ia tasks elicited elicited signi significant ficant increases increases in pain and blood pressure, pain ratings and pressor responses did not differ between naltrexone and placebo. Cold pain was reduced by distraction compared to rest. The finding that opioid blockade with naltrexone did not moderate the pain and pressor responses to cold and ischemia ischem ia suggests that pain and assoc associated iated blood pressure responses responses are not modulated by opioids in hypertension. The finding that the distracting secondary task successfully reduced pain ratings suggests normal supraspinal pain modulation in essential hypertension. KEY WORDS:   distraction; endogenous opioids; hypertension; hypertension; hypoalgesia; naltrexone; pain.

number of mechanisms, including opioid dysfunction and/or and enhanced activation of descending pain modulation systems. The phenomenon of hypertensive hypoalgesia was first described in laboratory animals; the involvement of endogenous opioids was clearly established by evidence that differences in nociceptive responding between hypertensive and normotensive rodents was abolished following administration of opioid antagonists (Maixner  et al., 1982  1982;; Naranjo and Fuentes, 1985;; Saavedra, 1985  Saavedra, 1981  1981;; Sitsen and de Jong, 1983 Jong, 1983,, 1984; Zamir and Segal, 1979 Segal,  1979;;  Zamir, Simantov, and Segal 1980). 1980 ). A key study observed that hypertensive hypoalgesia was not abolished by a peripherally acting opioid antagonist, suggesting a role for opioid receptors to rs wi with thin in th the e br brai ain n (S (Sit itse sen n an and d de deJo Jong ng,, 1984 1984).In ).In hu hu-man studies, the evidence for an opioid mechanism is mixed.. Hypert mixed Hypertension ension appears to be characterised characterised by greater levels of beta-endorphins in the peripheral circulation (Farsang  et al.,  1983  1983;;  Guasti   et al.,  1996  1996;; Hughes   et al.,   1991; 1991;   McNeilly and Zeichner,   1989; 1989;

INTRODUCTION

That individuals with hypertension experience lesss pai les pain n duri during ng nox noxiou iouss stim stimula ulatio tion n com compar pared ed to indi in divi vidu dual alss wi with th bl bloo ood d pr pres essu sure re in th the e no norm rmal al range is well documented (Ghione, 1996 (Ghione,  1996). ). However, the mecha mechanism nism underl underlying ying hypert hypertensive ensive hypoa hypoalgelgesia remains remains to be est establ ablish ished. ed. In an ext extens ension ion of  McCubbin’s (1991 (1991,,  1993  1993)) model of central opioid insensiti sen sitivit vity y in the eti etiolog ology y of hyp hyperte ertensi nsion, on, Fran France ce and Ditto (1996 (1996)) proposed a model in which pain insensitivity in hypertension could be explained by a 1 International

Centre for Health and Exercise Research, University of Birmingham, Birmingham, B15 2TT, UK. 2 Department of Psychology, Ohio University, Athens, OH 45701, USA. 3 Depart Department ment of Behavi Behavioral oral Sciences, University University of Minnes Minnesota ota Medical School, Duluth, MN 55812, USA. 4 Wellcome Trust Clinical Research Facility, School of Medicine, University of Birmingham, Birmingham B15 2TT, UK. 5 To wh whom om co corr rres espo pond nden ence ce sh shou ould ld be ad addr dres esse sed d ee-mai mail: l: [email protected].

al.., 1992 Sheps et al 1992)) an and d a co coup uple le of stu studi dies es ha have ve ar argu gued ed

59 C 2006 Springer Science+Business Media, LLC 0160-7715/07/0200-0059/0 

 

Ring  et al.

60

that the administration of opioid antagonists moderate the pain-blood pressure relationship in young normotensive adults (McCubbin and Bruehl,   1994; 1994; McCubbin   et al.,   2006). 2006). In contrast, other pharmacological blockade studies have failed to support the hypothesis that endogenous opioids are responsible for hypert hypertensive ensive hypoalgesia hypoalgesia (Brueh (Bruehll   et al al..,   2002; 2002; France   et al.,   2005; 2005;   Schobel   et al.,   1998). 1998). It should be noted that although Schobel and colleagues found that the association between blood pressure and pain ratings to noxious mechanical stimulation was not altered by opioid blockade with naloxone, they did find that naloxone led to higher pain ratings in normotensive men but not borderline hypertensive men. No study to date has examined the effects of opioid blockade block ade on pain in patien patients ts with established established hypertension. It has also been suggested that hypertensive hypertensive hypoalge poa lgesia sia may be rel relate ated d to com comple plex x int intera eracti ctions ons betwee bet ween n car cardio diovas vascul cular ar and pai pain n reg regula ulatory tory sys sys-tems (e.g., Randich and Maixner,   1984; 1984;   Zamir and Maixner,  1986  1986;; Ghione,  Ghione,   1996). 1996). Consistent with this

duce pain by modulating activity in central pain processing networks (Petrovic  et al.,  2000  2000;;  Tracey  et al., 2002), 2002 ), ex exam amin inat atio ion n of the ef effe fect ct of di distr strac acti tion on on pa pain in perception percep tion may provid provide e a prelim preliminary, inary, non-invasive non-invasive method of examining the integrity of central pain modulation networks. The current study used a double-blind, placebocontrolled design to examine the effects of the opioid antago ant agonis nist, t, nal naltrex trexone one,, on pai pain. n. In the nal naltrex trexone one and placebo sessions, pain ratings were determined for cold and ischemic pain tasks. In contrast to most previous pharmacological blockade studies that have investigated the effects of opioid antagonists on pain in ind indivi ividua duals ls wit with h nor normal mal or hig high-n h-norma ormall blo blood od pressure, pressur e, the curren currentt study examined middle-aged middle-aged patien pat ients ts new newlyly-dia diagno gnosed sed wit with h est establ ablish ished ed hig high h blo blood od pressure who were recruited and tested prior to the start of their anti-hypertensive treatment. The tasks used us ed in th the e cu curre rrent nt stu study dy as asse sesse ssed d pr prol olon onge ged d co cold ld an and d ischemic pain; they were expected to primarily activate C pain fibres and induce opioid release (Rosa  1988). ). To explore the role of pain modulation et al.,  1988

not notion ion, , Fra France nce and collea col league guessthat (France (Fra nce and Dit Ditto, to, 1996; 1996 ;  France,  France, 1999  1999) ) proposed hypoalgesia may be related to enhanced activation of descending pain inhibitory pathways due to the failure of endogenous opioids to provide normal inhibitory feedback to the paraventricular nucleus of the hypothalamus. The hypothalamus is part of a central autonomic network that is responsible for integrating autonomic, neuroendocrine, and behavioral responses to stress (Benarroch, 1993 (Benarroch,  1993), ), and this recipr reciprocally ocally-inter -interconneconnected network includes areas of the forebrain (insular and medial prefrontal cortices, amygdala, hypothalamus), midbrain (periaqueductal gray, pons), and brain stem (nucleus of the solitary tract, ventrolateral medulla). Within this netwo network, rk, the paraventricparaventric-

systems in hypertension, patients performed the cold pressor task while resting and while their attention was distracte distracted d from the nox noxiou iouss sti stimul mulati ation on by a number repetition and odd-ball digit counting task.

ular nucleus of the hypothalamus serves the crucial function of integrating responses to both painful and non-painful stressors through its innervations of autonomic relay centers, production of corticotrophin releasing factor, and neuronal projections to various areas of the central nervous system involved in pain modulation (Benarroch,  1993  1993;; Cechetto and Saper, 1988;; Sawchenko and Swanson,  1982 1988  1982;; Swanson and Kuypers, 1980 Kuypers,  1980). ). Given the intricate relationship between cardiovascular and pain regulatory systems, it is conceivable that attenuation of normal inhibitory opioid input to the paraventricular nucleus may lead to gre greate aterr act activa ivatio tion n of pai pain n inh inhibi ibitory tory pat pathwa hways. ys. Based upon converging neuroanatomical and neuroimaging evidence that cognitive distraction may re-

mean height of 1.71 ( SD   =   0.10) m, mean weight of  81.3 (SD   =  20.0) kg, and mean body mass index of  27.7 (SD   =  5.9) kg/m2 . The study was approved by the local ethics commit committee tee and volunt volunteers eers gave written consent prior to participation.

METHODS Participants

Twenty-fo Twenty -four ur adu adults lts wer were e rec recruit ruited ed to par partic ticiipate in this double-blind placebo-controlled study. Six part partici icipan pants ts did not com comple plete te all con condit dition ions. s. Thus, the effec effective tive sample comprised 18 newly diagnosed and unmedicated hypertensive adults (10 men, 8 women) with a mean age of 44.1 ( SD   =  11.1) years,

Screening Exclusion Criteria

In an initial screening session, each participant’s medical status and eligibility were determined. The follow fol lowing ing exc exclus lusion ion cri criter teria ia wer were e app applie lied: d: cur curren rentt use of medication, diabetes mellitus, cerebrovascular disease, diseas e, angina angina,, myoca myocardial rdial infarc infarction, tion, periph peripheral eral

 

Pain in Hypertension

vascular disease, neurological disease, chronic liver disease, diseas e, alcoho alcoholl intake   >28 units (1 unit   =  284 ml of beer, 125 ml of wine, or 25 ml of spirits) of alcohol per week in men,   >21 units of alcohol per week in women, major psychiatric disorder, secondary hypertension including chronic renal failure, renal artery stenosis, Conn’s syndrome, or phaeochromocytoma. Blood Pressure Status

British Hypertension Hypertension Societ Society y guidel guidelines ines were used use d to esta establi blish sh blo blood od pres pressure sure sta status tus (Ra (Ramsa msay y et al al..,   1999). 1999). Ea Each ch pa parti rtici cipa pant’ nt’ss bl blood ood pr press essur ure e was measured for 24 h using an ambulatory blood pres pr essu sure re mo moni nito torr (A (ABP BPM) M) (S (Spa pace ceLa Labs bs Me Medi di-cal,, Mode cal Modell 902 90207) 07).. Pat Patien ients ts wit with h a sys systol tolic ic blo blood od pressure   ≥160 mmHg or a diastolic blood pressure of   ≥100 mmHg at referral, and confirmed at clinic and on ABPM (mean daytime pressure) were diagnosed as hypertensive; 67% of patients met these criteria. Patients with a systolic blood pressure of  140–15 140 –159 9 mmH mmHg g and and/or /or a dia diastol stolic ic blo blood od pre pressur ssure e of 90–99 mmHg at referral, clinic, and on ABPM, were we re di diag agno nosed sed as hy hype pert rten ensiv sive e if the their ir 10 ye year ar coronary heart disease risk was   ≥15% and/or there was evidence of left ventricular hypertrophy on a 12 lead electrocardiograp electrocardiograph h or echoc echocardiog ardiograph, raph, or ther th ere e wa wass oth other er ev evid iden ence ce of en end d org organ an da dama mage ge.. The cardiovascular risk profile was calculated using the Joi Joint nt Bri British tish Soc Societ ieties ies Car Cardia diacc Ris Risk k Asse Assessor ssor computer program (Wood   et al.,   1998). 1998). Blood was sample sam pled d to det determ ermine ine rena renall fun functi ction, on, pota potassiu ssium, m, glucose, gluco se, choles cholesterol, terol, high-de high-density nsity lipopr lipoprotein, otein, and triglycerides. Urinalysis was also performed. If clinically indicated, patients were screened for evidence of sec second ondary ary hyp hyperte ertensio nsion n wit with h rena renall ult ultraso rasound und,, Doppler imaging of the renal arteries followed by magnetic magne tic resona resonance nce angio angiography graphy where indic indicated, ated, 24 h uri urina nary ry co coll llec ecti tion on fo forr ca cate tech chola olami mine ness (i (in n triplicate) and renin and aldosterone levels. Physiological Measurements

Laboratory Laborat ory syst systoli olicc blo blood od pres pressure sure (mm (mmHg) Hg),, diastol dia stolic ic blo blood od pres pressure sure (mmH (mmHg), g), and hea heart rt rat rate e (bpm) were obtained using an oscillometric sphygmomanometer (Omron, 705CP, Omron Health Care Ltd) that has been validated by the European Society of Hypertension (O’Brien  et al., 2001  2001)) with an appropriately sized brachial cuff attached to the participant’s upper left arm. All blood pressure monitors

61

were regularly calibrated, serviced, and, if necessary, repaired. Procedure

The stud study y emp employ loyed ed a dou double ble-bl -blind ind pla placeb ceboocontrolled contro lled design in which participants participants comple completed ted two thre three-ho e-hour ur mor mornin ning g sess sessions ions,, sep separat arated ed by at least 48 h. They were instructed to refrain from caffein fe ine, e, al alco coho hol, l, an and d vi vigo gorou rouss ex exer erci cise se fo forr 2 h, an and d an anal al-gesic medication for 24 h prior to testing. At the start of each session, participants sat and relaxed during an initial formal rest period (10 min) while their blood pressure was measured at 0, 270 and 540 s (First Baseline). The investi investigator gator then admini administered stered a tablet to the participant, which contained either a 50 mg dose of naltrexone or placebo. The order of tablet administration was counterbalanced across participants; in the effective sample, 10 patients received naltrexone in their first session and eight patients in their second session. The participant was asked to relax for 1 hour to allow for the drug to rea reach ch).pea peak k cir circul culati ating ng lev levels els (Gon (Gonzal zalez ezformal and Bro Brogde gden, n, 1988). 1988 They then completed a second rest period (10 min) while their blood pressure was measured at 0, 270 and 540 s (Second Baseline). Sensitivity to cold (cold pain task) and ischemia (ischemic pain task) were then determined: pain ratings were obtained during and immediately after the tasks. A blood blo od pre pressur ssure e mea measure suremen mentt was ini initia tiated ted at the star startt of the first and second minutes of each cold pressor task ta sk an and d at th the e st star artt of th the e fir first st,, th thir ird d an and d fif fifth th mi minu nute tess of the forearm ischemia task. At the end of each session, participants sat and relaxed during a final formal rest period (10 min) while their blood pressure was measured at 0, 270 and 540 s (Recovery). Cold Pain Task

Cold sensitivity was assessed during three conditions: quiet sitting (First Rest), distracting ditions: distracting number repeti rep etitio tion n and cou countin nting g (Di (Distra stracti ction) on),, and qui quiet et sitting (Second Rest). Participants were instructed to place their hand up to the wristfold in 4 degree Celsius water for 2 min. They were asked to rate the intensity of the sensation every 30 s using a 0–100 rati ra ting ng sc scal ale e wi with th an anch chors ors of “n “no o se sensa nsati tion on” ” (0 (0), ), “uncomfortab “unco mfortable” le” (25), “just notice noticeable able pain” (50), “very painful” (75), and “maximum tolerable pain” (100). If participants removed their hand before the 2 min was up, the tolerance time was recorded, and a rating of 100 was assigned for the remainder of the

 

Ring  et al.

62

task. Such premature hand removal occurred for five or six individuals in all three conditions during each session ses sion.. In the dis distrac tractio tion n con condit dition ion,, par partic ticipa ipants nts listened listen ed to an audio taped presentation presentation of a series of  single-digit integers (numbers ranging from 1 to 9) delivered at 3–4 second intervals. They were asked to repeat each number out loud, and to privately keep count of the occurrence of one number (e.g., 9’s). This task was designed to be simple but engaging. A simpler version of this secondary task used in a previous previo us pain study was judged to be distracting distracting and tended to reduce pain compared to rest (Edwards ( Edwards et al.,  in press). At the end of each task, participants completed a short-form McGill Pain Questionnaire (Melzack, 1987 (Melzack,  1987))  to indicate retrospectively the pain associated with the cold exposure. Participants rated 11 sensory and four affective descriptors on an intensity tensit y scale of 0 (none), 1 (mild), 2 (moderate), and 3 (severe), that were summed to yield the sensory, affective and total pain rating indices. A scale, with anchors of 0 (no pain) and 100 (worst possible pain), was used to indicate an overall pain rating. They also

blood systolic and diastolic pressures (Cold Pressor – First Rest, Cold Pressor – Distraction, Cold Pressor – Second Rest, Forearm Ischemia). The corresponding heart rates were also computed. Statistical analyses were performed using SPSS for Windows 14.0 (SPSS Inc., Chicago, IL, USA). A series of repeated measuress mul sure multiv tivaria ariate te ana analys lyses es of var varian iance ce (MAN (MANOVAs OVAs)) were performed on the task-related pain and associated cardiovascular data, with sex as a betweensubj su bjec ects ts fa fact ctor or an and d dru drug, g, co cond ndit itio ion, n, an and d ti time me as within-subje within -subjects cts factors factors.. Signifi Significant cant effec effects ts were interrogated terroga ted using planne planned d compar comparisons. isons. Eta-sq Eta-squared uared 2 (η ), a measure of effect size, is reported. A probability level of less than .05 was taken as significant.

rated present intensity on ofssing), 0 (no pain), pai n),their 1 (mil (mild), d), 2 pain (discom (dis comfor forting ting), ), a3 scale (distre (di stressin g), 4 (horrible), and 5 (excruciating). Each assessment was followed by a 5 minute rest.

patients.

Ischemic Pain Task

Using a hand dynamometer (Lafayette) participants performed rhythmic forearm muscle contractions (contract for two seconds, relax for two seconds) at 50% of pre-determined maximum ( M   =   32, SD   =  9 N) for two minutes. In the last 15 s a brachial cuff was inflated to 220 mmHg to occlude blood flow. They then sat quietly for five minutes with their arm resting horizontally. During this time they provided an intensity rating every 30 s using the rating scale described above. If participants terminated the task before the full 5 minutes was up, which happened with five participants during the naltrexone session and six during the placebo session, the tolerance time was recorded and a rating of 100 was assigned for the remainder of the task. At the end of the task they completed a short-form McGill Pain Questionnaire. Data Reduction and Statistical Analysis

The blood pressure recordings were averaged to provide measures of resting systolic blood pressure and diastolic blood pressure (First Baseline, Second Baseline, Recovery) and measures of task-induced

RESULTS Blood Pressure Status

The average clinic, ambulatory and initia initiall baseline laboratory blood pressures presented in Table Table I  I confirm the blood pressure status of the hypertensive

Cold Pain

Pain ratings reported every 30 s during the cold pressor task are shown in Fig.   1.   A 2 Sex (male, female)   ×  2 Drug (naltrexone, placebo)   ×  3 Condition tio n (firs (firstt rest rest,, dis distrac traction tion,, sec second ond rest rest))   ×  4 Ti Time me (3 (30, 0, 60, 90, 120 s) MANOVA on the pain ratings yielded main effects for condition,   F (2, (2, 15)   =   10.12,   p   = .002,   η2 =   .576, (quadratic term:   F (1,16) (1,16)   =   15.51,  p   =   .001; less during distraction distraction than first rest), and time,  F (3, (3, 14)   =   31.12,  p   =   .0001,   η2 =  .870, (cubic term:  F (1,16) (1,16)   =   6.64,  p   =  .03; monotonic increase). Table I.   Mean (SD) Blood Pressures of the Hypertensive Patients

Variable Clinic Syst Sy sto oli licc bl blo ood pre ress ssur ure e (mm mmHg Hg)) Dias Di asto toli licc blo loo od pr pres essu sure re (m (mmH mHg) g) Ambulatory (day) Syst Sy sto oli licc bl blo ood pre ress ssur ure e (mm mmHg Hg)) Dias Di asto toli licc blo loo od pr pres essu sure re (m (mmH mHg) g) Ambulatory (Night) Syst Sy sto oli licc bl blo ood pre ress ssur ure e (mm mmHg Hg)) Dias Di asto toli licc blo loo od pr pres essu sure re (m (mmH mHg) g) Laboratory (First Baseline) Syst Sy sto oli licc bl blo ood pre ress ssur ure e (mm mmHg Hg)) Dias Di asto toli licc blo loo od pr pres essu sure re (m (mmH mHg) g) Heart rate (bpm)

162.2 (15.6) 162 100 10 0.9 (13.0) 149.6 (9 .5) 149 98.4 (8 .5) 128.6 (15.1) 128 79.6 (10.0) 155.7 (12.6) 155 97.1 (8 .5) 75.3 (7 .9)

 

Pain in Hypertension

63

80

Placebo Naltrexone 80

     )      0 70      0      1         0      (     g     n      i      t 60     a      R     n      i     a      P50     c      i     m     e      h     c     s 40      I

     ) 70      0      0      1         0      ( 60     g     n      i      t     a      R     n 50      i     a      P      d      l     o 40      C 30

Placebo Naltrexone

0

30 0

30 60 90 1 12 20

30 60 90 120

30 60 90 1 12 20

Time (s) First Rest

Distraction

Second Rest

Fig. 1.  Mean (SE) pain ratings during the cold pressor under resting and distracting conditions during opioid blockade with naltrexone and placebo.

In addition, the condition by time interaction was

60

90

120

150

180

210

240

270

300

Time (s)

Fig. 2.   Mean (SE) pain ratings during the ischemia task during opioid blockade with naltrexone and placebo.

270, 300 s) MANOVA on the pain ratings yielded a time effect,   F (9, (9, 8)   =   13.42,   p   =   .001,   η2 =   .938, (cubic term: F (1,16) (1,16)   =  4.59, p   =  .05), but no drug or

2

=   4.59,   p   =   .01,   η =   .740,   F (6, significant, (6, 11)r  term: (quadratic (quadr atic by linear linea  F (1, (1, 16)   =   5.57,  p   =   .03, 2 η =  .112). No drug or sex effects were found. The retrosp retrospectiv ective e McGill Pain Questio Questionnaire nnaire scores confirmed confirmed that cold pressorpressor-induce induced d pain was reduced by distraction but was not affected by opioid blocka blo ckade de (see Tab Table le   II II). ). Sp Spec ecifi ifica call lly, y, a se serie riess of  2 Sex   ×   2 Drug   ×   3 Condition MANOVAs indicate dic ated d con condit dition ion mai main n eff effect ectss for the tot total al pai pain n 2 rating index,   F (2, (2, 15)   =   7.93,   p   =   .004,   η =   .514, (quadratic (quadr atic term:   F (1,16) (1,16)   =   11.62,   p   =   .04; pai pain n lower low er duri during ng dis distrac traction tion tha than n bot both h rest rests), s), sens sensory ory pain pai n rati rating ng ind index, ex,   F (2, ( 2, 15 15))   =   9.51,   p   =   .002, 2 η =   .559, .559, (qu (quadra adratic tic term term::   F (1,16) (1,16)   =   16.43,  p   =   .001; pain lower during distraction than both

sex effects. Separate 2 Sex   ×  2McGill Drug ANOVAs were performed perform ed on the short-form Pain Questi Questiononnaire nai re sco scores res asso associa ciated ted wit with h the fore forearm arm isc ischem hemia ia task. tas k. The These se ana analys lyses es rev reveal ealed ed no sig signifi nifican cantt mai main n effects for drug (see Table   III III). ). Sex main eff effect ectss emerged for the overall pain rating,  F (1, (1, 16)   =   6.95, (1,  p   =   .02,   η2 =  .303, and present pain intensity,  F (1, 2 16)   =   8.62,   p   =   .01,   η =  .350; men reported less ischemic pain than women. In addition, sex by drug interaction effects were found for the total pain rating index,  F (1, (1, 16)   =   4.54,  p   =   .05,   η2 =  .221, sensory pain rating index, F (1, (1, 16)   =  4.60, p   =  .05,  η2 = .223, and present pain intensity,  F (1, (1, 16)   =  6.77, p   = 2 .02, η =  .297. As can be seen in Table IV Table IV,, men’s tendency to report less pain than women during placebo

rests), overall pain rating, F (2, (2, 15)   =   5.75,  p   =   .01, 2 η =   .434, (quadra (quadratic tic term:   F (1,16) (1,16)   =   7.50,   p   = .02; pain lower during distraction than the first rest), and present pain intensity,  F (2, (2, 15)   =   4.97,  p   =   .02, 2 (1,16)   =   10.59,   p   = η =   .399, (quadratic term:   F (1,16) .005; pain lower during distraction than first rest), of  the McGill Pain Questionnaire. However, none of  the drug or sex effects were significant.

was absent during opioid blockade with naltrexone.

Ischemic Pain

Pain ratings reported every 30 s during the ischemic chem ic ta task sk ar are e sho shown wn in Fi Fig. g.   2. A 2 S e x   ×   2 Drug   ×  10 Time (30, 60, 90, 120, 150, 180, 210, 240,

Blood Pressure and Heart Rate

The rest resting ing and task task-in -induc duced ed car cardio diovas vascul cular ar responses during placebo and naltrexone are shown in Fig.  Fig.   3.  Separate 2 Sex   ×   2 Drug   ×  7 Condition (first baseline, second baseline, cold pressor – first rest, cold pressor – distraction, cold pressor – second rest, ischemia, recovery) confirmed significant condition condit ion effects for both systoli systolic, c,  F (6, (6, 11)   =   20.01, 2  p   =   .001,   η =   .91 .916, 6, and dia diastol stolic, ic,   F (6, ( 6, 11 11))   = 16.29,   p   =   .001,   η2 =   .89 .899, 9, blo blood od pre pressur ssures: es: the task ta skss el elic icit ited ed si sign gnifi ifica cant nt pr pres esso sorr re resp spon onse ses. s. A

 

Ring  et al.

64

   s    n    o     i     t     i     d    n    o     C    n    o     i     t    c    a    r     t    s     i     t    s     D    c    e     d    f    n    f    a    E     t    s    n    e    i    o     t     R    i    g    d    n    n     i    r    o    u    C    y     D    b    o    g     b    u    e    c    r    a    D     l     P    d     d    n    n    a  ,    a    n    e    i    o    n    t    o    i    x    d    e    r    n    o     t     l    a    C  ,     N   g    u     h    r     t     i     D    w   e

    d    e     h    a    t     f     k    o    c     l    o    e     l     B   v    e     d     i     L    o    e    c     i    p    n     O   a    c    g    fi     i    n    n     i    g    r     i    u    S     D    l    a    a    i    c     t    a    t    s     i     D    t    a    e    r     t     i     S    a    e    n    h    n    t    o    s     i     t    a    s     l    e    l    u    e     Q   w    n    s    a     i    a    k     P    s     l     T     i    a    r     G    c    o    s     M   s    e    r    m    r     P    o    d     l     F   -    o     t    r     C    o    e     h    h     S    t      )     f    o      D

     S      (    n    a    e     M   .       I       I     e       l       b     a       T

    t    c    e     f     f    e    n    o     i     t     i     d    p    n    o    c    y     b    g    u    r     D     t    c    e     f     f    e    n    o    p     i     t     i     d    n    o     C     t    c    e     f     f    e    p    g    u    r     D     d    n    t    s    o    c    e    e    r     S    o     b    e    c    a     l     P

    4    2    6    1    2     0    0    2    0    0     0  .     0  .  .  .  .

    3    8    4    5    7     1  .     2  .     1  .     3  .     8  .      )     )     )     )     )     6  .     5  .  .     9  .     8    2     7     (     6     (     1     (     2     (     1      (     6  .     3  .     3  .     8    6  .     9    8    1    6    2

   n    o      )     )     )     )     )     i     t    c     7  .      (     0  .      (     5  .      (     5     (     0  .     5     5     1     2     1    a    r      (     t     7     6     1     7    5  .  .  .  .    s     i     7    6    1    6    2     D      )     )     )     )     )     9     t  .     2  .     7  .     0    0  .     t    s     5     (     5     (     1     (     2     (     1    r    s      (     i    e     F    r     9  .     4  .     5  .     6    9  .     8    7    1    7    2     d    n    t    s    o    c    e    e    r     S

   e    n    o    x    e    r     t     l    a     N

    1    2    9    2    6     8  .     9  .     4  .     6  .     8  .

     )     )     )      )     2      )  .     5  .     0  .     6    1  .     8    6     3     (     2     (     1      (      (     (     0  .     8     7  .     2  .     4     1    8     7  .     1     2     2

   n      )     )     )     )    o      )     i     1     t  .     4  .     5  .     7    0  .    c     9     6     (     3     (     2     (     1    a      (      (    r     t  .     7  .     0  .     2    4  .    s     7     i     9    7    2    7    2     D      )     )     )      )     2      )  .     8     t  .     1  .     0    9  .     i    s     t    r    s      (     8      (     5      (     3      (     2      (     0    e     F    r     3  .     6     8  .     7  .     8     1    8     7  .     1     2     2      )      )     2     3    1   –      )     3   –    0     5    0     4     (     (      )    x     )   –     5    e    0     0    x   –     0      (    e    d    1    0      (    n    x    d    i   –    e    n      (    y     0    t     d    i    g    n    g    i    s    n    g    i     i    n    t     i    n    n    e    g    i     t    a     t     t    r    n    a     i    n    r     i    n    a     t    r     i    a    n    a    n    n     i    r     i    p    i    a    a    n    a    p     i    p    p    e    a    y    e    v    l     l     l     t    p    r     i     b     l     t    n    o    c    a    a    s    e    a    r    e     i    e    s    r     t    n    f    a    o    e    f    v    e    r     V     T    S    A    O    P

 

Pain in Hypertension

65

Table III.   Mean (SD) Short-Form McGill Pain Questionnaire Data During Opioid Blockade with Naltrexone and Placebo During the Forearm Ischemia Task as well as the Statistical Significance Level of the Drug Effects

Variable

Drug effect   P 

Naltrexone

Placebo

Total pain rating index (0–45) Sensory pain rating index (0–33) Affective pain rating index (0–12)

15.3 (9.9) 13.0 (7.6) 2.3 (3.2)

13.4 (9.8) 11.7 (8.3) 1.8 (2.5)

.25 .39 .27

Orveesreanltl P Pa aiin n IRnatetinnsgit(y0(–01–050)) P

5 ((1 2.00)) 27.6

0 ((1 1.91)) 27.7

44 5 ..4

significant condition effect also emerged for heart rate,   F (6, (6, 11)   =   12.83,   p   =   .001,   η2 =   .875; heart rates were faster during the first baseline than the second baseline and cold pressor tasks, which in turn, were faster than during the ischem ischemic ic task and recovery. None of the drug or sex effects were significant. DISCUSSION

Although Althoug h subgro subgroup up analys analyses es provid provided ed limite limited d evidence that ischemic pain was somewhat greater duri du ring ng na nalt ltrex rexon one e th than an pl plac aceb ebo o in me men, n, ov over eral alll the th e res resul ults ts of the cu curre rrent nt st stud udy y su sugg gges estt tha thatt th the e experience of both cold pain and ischemic pain was not augmen augmented ted during pharmacologica pharmacologicall blocka blockade de of  opioid receptors in patients with essential hypertension. These results broadly agree with the finding of a previous study showing that naloxone did not change mechanical pain in borderline hypertension (Schobel  et al.,  1998  1998). ). However, the results of these two studies stand in distinct contrast to the existing eviden evi dence ce of rev reversa ersall of hyp hypert ertens ensive ive hyp hypoal oalges gesia ia through opiate blockade in laboratory animal studies (Maixner   et al.,   1982; 1982;   Naranjo Naranjo and Fuente Fuentes, s,   1985; 1985; Saavedra,   1981; 1981; Si Sits tsen en an and d de Jo Jong ng,,   1983, 1983, 19 1984 84;; Zamir Segal,  1979; ;  Zamir, and 1980). 1980 ). and TheSegal, 1979 curre cu rrent nt stu study dy al also soSimantov, show sh owed ed th that at Segal, bloo bl ood d pressure and heart rate, both at rest and in response to cold and ischemic challenge, challenge, were not affect affected ed by

naltrexone. These observ naltrexone. observations ations are compa compatible tible with previous studies reporting that opioid blockade moderates exercise-induced cardiovascular responses in normotensive but not hypertensive individuals (Hara and Floras, 1995 Floras,  1995)) and that opioid blockade attenuates stress-induced pressor responses in individuals with low-normal resting blood pressure but not in those with high-normal pressure (McCubbin Sunwit, and Wil Willia liam, m,   1988). 1988). Ta Take ken n to toge gethe ther, r, the re resul sults ts of the current and previous studies in humans are compatible compa tible with the hypothesis that hyperte hypertension nsion in humans is characterised by insensitivity to centrally and/or peripherally acting opioids (France and Ditto, 1996;; Feuerstein and Siren, 1987 1996 Siren,  1987). ). The discrepancy in the effect of opioid blockade on pain in relation to blood pressure status in human and rodent studies may be attribu attributable table in part to the greater likelihood of opioid responsivity in animal studie stu dies. s. Whe Wherea reass stud studies ies wit with h lab labora oratory tory ani animal malss provide provid e consist consistent ent evide evidence nce for naloxo naloxone-rev ne-reversed ersed opiate opi ate act activi ivity, ty, simi similar lar blo blocka ckade de par paradi adigms gms hav have e not yielded consistent findings in humans. According to ani animal mal stud studies ies,, unc uncont ontroll rollabi abilit lity y may be a necessary factor in opioid activation, as shown in studiess of learne studie learned d helple helplessness ssness and stress-i stress-induce nduced d analge ana lgesia sia (Mai (Maier, er, She Sherma rman, n, Lew Lewis, is, Term Terman, an, and Liebeskind, 1983 Liebeskind,  1983;;  Maier,  Maier, 1990  1990). ). For instance, Maier and col collea league guess (19 (1983) 83) fou found nd tha thatt onl only y con condit dition ionss that allowed for “learning-of-uncontrollability,” such

Table IV.   Mean (SD) Short-Form McGill Pain Questionnaire Data for Men and Women During Opioid Blockade with Naltrexone and Placebo During the Forearm Ischemia Task as well as the Statistical Significance Level of the Drug by Sex Effects

Men Variable Totall pa Tota pain in ra rati ting ng in inde dex x (0 (0–45 –45)) Sensory pain rating index (0–33) Pres Pr esen entt pa pain in in inte ten nsi sity ty (0 (0–5 –5))

Women

Placebo

Naltrexone

Placebo

Naltrexone

Drug by sex effect   p

10.7 (9 10.7 (9.0 .0))a 9.2 (7.8) 2.0 2. 0 (0. 0.6) 6)b

15.1 (9.6)a 12.8 (7.5) 2.4 (0.9)

16.9 (9.5) 14.7 (8.1) 3.6 (0.8)b

15.6 (10.1) 13.3 (7.8) 2.9 (0.8)

.05 .05 .02

Note. Means in the same row that share superscripts differ significantly in the post-hoc comparison.

 

Ring  et al.

66 190

     )     g      H     m 180     m      (     e     r     u     s 170     s     e     r      P      d 160     o     o      l      B     c 150      i      l     o      t     s     y      S

(A)

Placebo Naltrexone

0

2  C o  C o  C o  I s  1 s t  R e c  s c h  l d   l d   l d   d  -  d  -  d  -   B a s  n d  B a  o v e   R e s   D i s   R e s  e m i a  s s  e l  r y  e l ii n a  y  s t rr a   n e    ii n    e  t  t  n      c  a t       ) 130     g      H     m     m120      (     e     r     u     s 110     s     e     r      P      d 100     o     o      l      B     c      i      l 90     o      t     s     a      i      D

(B)

Placebo Naltrexone

0

C o l  2 n d   C o l  C o l  1 s tt   I s  R e c  sc    h  d   d  -  d   d  -  d   d  -   B a s o v e  e m  B a s  s  e l  e r y   R e s   D i s   R e s  i a  a  e e l   ii  n  y  s t rr a  s t  s t  ii  n n e      c  ne    a   t  t 

increased controllability may limit the likelihood of  producing produc ing signifi significant cant endoge endogenous nous opiate media mediated ted pain reduction in human studies. Previous research has demonstrated that experimental pain tasks, including cold pressor, are associated with opioid release in humans (e.g., Rosa  et al., 1988). 1988 ). Although the current findings that cold and ischemic pain were not different during placebo and naltrexone are compatible with opioid insensitivity in hypertension hypert ension,, as we did not measure circulating circulating opioids to establish whether the tasks were sufficiently provoc pro vocati ative ve to trig trigger ger opi opioid oid rel releas ease, e, the nul nulll find finding ingss could also be explained by the absence of any opioidmediated analgesia. It should also be recognised that the number of part partici icipan pants ts test tested ed was small and therefore sample size may have reduced the power to detect differences between opioid blockade and control. However, the effect sizes were small and post hoc power calculations indicated that we would have needed to test hundreds of participants to detect significant effects of naltrexone versus placebo on pain. The null findings are therefore most likely to be at-

90

(C)      ) 80     m     p      b      (     e      t     a 70      R      t     r     a     e      H60

Placebo Naltrexone

0

2  C o  C o  C o  I s  1 s t  R e c  s c h  c o v  l d   l d   l d   h  d  -  d  -  d  -   B a s  n d  B a  e r y   R e s   D i s   R e s  e m i a  s e e l ii n a  e  l ii n  y  s t rr a   n e      e  t  t  n    ac    t 

Fig. 3.   Mean (SE) systolic blood pressure (A), diastolic blood pressure (B) and heart rate for naltrexone and placebo sessions during the 5-min rest periods at the start of the session (1st Baseline), one-hour after tablet ingestion (2nd Baseline), and at the

end the relaxing session (Recovery), during 2-min coldthe pressor tasks whenoffirst (Cold – Rest), whenthe performing distracting secondary task (Cold – Distract), and when relaxing for a second time (Cold – Rest), and during the 5-min forearm ischemia task (Ischemia).

as 20 min of intermittent intermittent footshock or 60–80 trials of  tailshock, were able to induce opioid analgesia. Obviously, humans have more control in experimental pain studies as they are always informed that they can discontinue a painful procedure at any time. In addition to differential opportunities for control, it is also arguable that laboratory pain paradigms are often more aversive in animal versus human studies studies.. In sum, the combination of reduced aversiveness and

tributable ratherthat than small sample size. It to haslow alsoeffect beensizes suggested hypertensive hypoalgesia is mediated by opioid hyposensitivity. However, any interpretation of hyposensitivity must remain speculative without comparative data from a normotensive control group. With this in mind, it should be noted that other studies have failed to detect changes in mechanical and ischemic pain (Bruehl  2002)) as well as electrocutaneous pain (France et al.,  2002  2005)) with opioid blockade in normotensive inet al., 2005 dividuals which might, in the absence of corroborative blood data concerning plasma opioid concentrations, be interpreted as indicating opioid insensitivity in normote normotensive nsive indivi individuals. duals. Accordingly, Accordingly, definit definitive ive conclusions about relative sensitivity to opioids as a function of blood pressure status cannot be drawn at function this time. The at atte tenu nuat atio ion n in pa pain in du durin ring g th the e pe perfo rforrmanc ma nce e of the se seco conda ndary ry ta task sk co comp mpar ared ed to res restt is in accordance with a substantial body of evidence docume doc umentin nting g tha thatt pai pain n is red reduce uced d by dis distrac tractio tion n in hea health lthy y ind indivi ividua duals ls (al (al’Ab ’Absi si and Rok Rokke, ke,   1991; 1991; Edwards  et al.,  in press; Miron et al.,  1989  1989;;  Petrovic et al.,  2000  2000;;  Villemure and Bushnell,  2002  2002). ). To our knowledge, the current study is the first to demonstrate that cold pain was attenuated by distraction in pat patien ients ts wit with h hyp hyperte ertensi nsion. on. The resu results lts show show-ing that a distracting secondary task reduced pain ratings suggests supraspinal modulation of pain in hypertension.

 

Pain in Hypertension ACKNOWLEDGMENTS

This study was supported by NIH grant R01 HL64794. L. E. is a British Heart Foundation Fellow (FS/03/128). REFERENCES

67 Maier, S. F., Sherman, J. E., Lewis, J. W, Terman, G. W., and Liebeskind Liebes kind,, J. C. (1983). The opioi opioid/non d/non-opio -opioid id natur nature e of  stress-ind stres s-induced uced analges analgesia ia and learne learned d helpl helplessne essness. ss.   J. Exp Exp.. Psychol . 9: 80–90. Maixner, W., Touw, K. B., Brody, M. J., Gebhart, G. F., and Long, J. P. (1982). Factors influencing the altered pain perception in the spontaneously hypertensive. rat. Brain. Res. 237  237:: 137–145. McCubbin, J. A. (1991). Diminished opioid inhibition of blood pressure and pituitary function in hypertension development. In McCubbin, J. A., Kaufman, P., and Nemeroff, C. (Eds.,),

al’Absi, M., and Rokke, P. D. (1991). Can anxiety help us tolerate pain? Pain 46: 43–51. Benarroch, E. (1993). The central autonomic network: Functional organization, dysfunction, and perspective. Mayo Clinic Proceedings 68: 988–1001. Bruehl, S., Chung, O. Y., Ward, P., Johnson, B., and McCubbin, J. A. (2002). The relationship between resting blood pressure and acute pain sensitivity in healthy normotensives and chronic back pain sufferers: The effects of opioid blockade. Pain  100: 191–201. Cechetto, D. F., and Saper, C. B. (1988). Neurochemical organization of the hypothalamic projection to the spinal cord.  J. Compar. Neurol. 272: 579–604. Edwards, L., Ring, C., McIntyre, D., Carroll, D., Clarke, R., and Webb, O., and Martin, U. (in press). Increases in arousal are associated assoc iated with reduc reductions tions in the human nociceptive nociceptive flexion reflex threshold and pain ratings: Evidence for dissociation between nociception and pain.  J. Psychophysiol. Farsang, C., Vajda, L., Kapocsi, J., Malisak, Z., Alfoldi, S., Varga, K., Juhasz, I., and Kunos, G. (1983). Diurnal rhythm of betaendorphin in normotensive and hypertensive patients: The effect of cloni clonidine. dine.   J. Clin. Endocrinol. Endocrinol. Meta Metabol. bol.   56: 865– 867. Feuers Feu erstei tein, n, G., andSiren andSiren,, A. L. (19 (1987) 87).. The opi opioidpeptid oidpeptides.A es.A rol role e in hypertension? Hypertension 9:561–655. France, C. R. (1999). Decreased pain perception and risk for hypertension: Considering a common physiological mechanism. Psychophysiol 36: 683–692. France, C.R., al’Absi, M., Ring, C., France, J., Brose, J., Spaeth, D., Harju, A., Nordehn, G., and Wittmers, L. E. (2005). Assessment of opiate modulation of pain and nociceptive responding in young adults with a parental history of hypertension. Biological. Psychol. 70: 168–174. France, C. R., and Ditto, B. (1996). Risk for high blood pressure and decreased pain perception.  Cur. Dir. in. Psychol. Sci.  5: 120–125. Ghione, S. (1996). Hypertension-associated hypalgesia.  Hyperten sion. 28: 494-504.

Stress, neuropeptides and systemic disease,  San Diego, Academic Press, pp. 445–466. McCubbin, J. A. (1993). Stress and endogenous opioids: Behavioral and circulatory interactions.  Biological. Psychol.  35: 91– 122. McCubbin, J. A., and Bruehl, S. (1994). Do endogenous opioids mediate the relationship between blood pressure and pain sensitivity in normotensives?  Pain. 57: 63–67. McCubbin, J. A., Helfer, S. G., Switzer, F. S., Galloway, C., and Griffith, W. V. (2006). Opioid analgesia in persons at risk for hypertension. Psychosom. Med.  68: 116–120. McCubbin, J. A., Surwit, R. S., and Williams, R. B. (1988). Opioid dysfunction and risk for hypertension: Naloxone and blood pressure responses during different types of stress.  Psycho som. Med. 50: 8–14. McNeilly, M., and Zeichner, A. (1989). Neuropeptide and cardiovascular responses to intravenous catheterization in normotensive motens ive and hypertensive hypertensive black blackss and whites.  Health. Psychol. 8: 487–501. Melzack, R. (1987). The short-form McGill Pain Questionnaire. Pain. 30: 191–197. Miron,, D., Duncan Miron Duncan,, G., and Bushnell, Bushnell, M. C. (1989). Effects of attention on the intensity and unpleasantness of thermal pain. Pain. 39: 345–352. Naranjo, J. R., and Fuentes, J. A. (1985). Association between hypoalgesia and hypertension in rats after short-term isolation. Neuropharmacol. 24: 167–171. O’Brien, E., Waeber, B., Parati, G., Staessen, J., and Myers, M. G. (2001). Blood pressure measuring devices: recommendations of the Eur Europe opean an Soc Societ iety y of Hyp Hypert ertens ension ion.. BMJ. 322: 531–536 531–536.. Petrovic, P., Petersson, K. M., Ghatan, P. H., Stone-Elander, S., and Ingvar, M. (2000). Pain-related cerebral activation is altered by a distracting cognitive task.  Pain. 85: 19–30. Ramsay, L. E., Willi Williams, ams, B., Johnston, G. D., MacGre MacGregor, gor, G. A., Poston, L., Potter, J. F., Poulter, N. R., and Russel, G. (1999). Guidelines for management of hypertension: report of the third working party of the British Hypertension Society.   J. Hum. Hypertens. 13: 569–592. Randich, A., and Maixner, W. (1984). Interactions between car-

Gonzalez, and Brogden, R.and N. pharmacokinetic (1988). Naltrexone. A review ofJ.itsP., pharmacodynamic properties and therapeutic efficacy in the management of opioid dependence. Drugs 35: 192–213. Guasti, L., Cattaneo, R., Daneri, A,, Bianchi, L., Gaudio, G., Bonora Regazzi, M., Grandi, A. M., Bertolini, A., Restelli, E., and Venco, A. (1996). Endogenous beta-endorphins in hypertension: Correlation with 24-hour ambulatory blood pressure. 1243–1248. 8.  J. Am. Coll. Cardiol. 28: 1243–124 Hara, K., and Floras, J. S. (1995). Influence of naloxone on muscle sympathetic nerve activity, systemic and calf haemodynamics and ambulatory blood pressure after exercise in mild essential hypertension. J. Hypertens. 13: 447–461. Hughes, G. S., Ringer, T. V, Francom, S. F., Caswell, K. C., DeLoof, M. J., and Spillers, C. R. (1991). Effects of fish oil and endorphins on the cold pressor test in hypertension.   Clin. Pharmacol. Ther. 50: 538–546. Maier, S. F. (1990). Diazepam modulation of stress-induced analgesia depends on the type of analgesia.   Science   208: 623– 625.

diovascular and pain regulatory systems.  Neurosci. Biobehav. Reviews 8: 343–367. Rosa, C., Ghione, S., Mezzasalma, L., L ., Pellegrini, M., Basile Fasolo, C., Giaconi, S., Gazzetti, P., and Ferdeghini, M. (1988). Relationship between pain sensitivity, cardiovascular reactivity to cold pressor test and indexes of activity of the adrenergic and opioid system.  Clin. Exp. Hypertens. A.  10 Suppl 1: 383– 390. Saavedra, J. M. (1981). Naloxone reversible decrease in pain sensitivity in young and adult spontaneously hypertensive rats. Brain. Res. 209: 245–249. Sawchenko, P. E., and Swanson, L. W. (1982). Immunohistochemical identification of neurons in the paraventricular nucleus of  the hypothalamus that prpoject to the medulla or to the spinal cord in the rat.  J. Compar. Neurol. 205: 260–272. Schobel, H. P., Hanwerker, H. O., Schmieder, R. E., Heusser, K., Dominiak, Dominiak, P., and Luft, F. C. (1998). Effects of naloxone on hemodynamic and sympathetic nerve responses to pain in normotensives vs. borderline hypertensive men.  J. Auton. Nerv. Syst. 69: 49–55.

 

68

Ring  et al.

Sheps, D. S., Bragdon, E. E., Flint Gray, T., and Ballenger, M., Usedom, J. E., and Maixner, W. W. (1992). Relation between systemic hypertension and pain perception.  Am. J. Cardiol.  70: 3F–5F. Sitsen, J. M., and de Jong, W. (1983). Hypoalgesia in genetically hypertensive rats (SHR) is absent in rats with experimental hypertension. Hypertension. 5: 185–190. Sitsen, J. M., and de Jong, W. (1984). Observations on pain perception and hypertension in spontaneously hypertensive rats. Clin. Exp. Theory. Prac.  A6: 1345–1356.

ulation of pain in the periaqueductal gray in humans.  J. Neu2748–2752. 2. rosci. 22: 2748–275 Villemure, C., and Bushnell, M. C. (2002). Cognitive modulation of pain: How do attention and emotion influence pain processing? Pain 95: 195–199. Wood, D., Durrington, P., McInnes, G., Poulter, N., Rees, A., and Wray, R. (1998) (1998).. Joint British recommendatio recommendations ns on preven preven-tion of coron coronary ary heart disea disease se in clinical practice. Heart.  80: S1–S29. Zamir, N., and Maixner, W. (1986). The relationship between car-

Swanson, L. W., and Kuypers, HGJM. (1980). The paraventricular nucleus of the hypothalamus: Cyto architectonic subdivisions and organization of projections to the pituitary, dorsal vagal complex, and spinal cord as demonstrated by retrograde fluorescent double-labeling methods.  J. Compar. Neurol.  194: 555–570. Tracey, I., Ploghaus, A., Gati, J. S., Clare, S., Smith, S., Menon, R. S., and Matthews, P. M.. (2002). Imaging attentional mod-

diovascularr and pain regul diovascula regulatory atory system systems. s. Ann. N. Y. Acad. Sci. 467: 371–384. Zamir, N., and Segal, M. (1979). Hypertension-induced analgesia: Changes in pain sensitivity in experimental animals.  Brain. Res. 160: 170–173. Zamir, N., Simantov, R., and Segal, M. (1980). Pain sensitivity and opioid activity in genetically and experimentally hypertensive rats. Brain Res. 184: 299–310.