- Email: [email protected]
Pain-Related Disability and Effects of Chronic Morphine in the Adjuvant-Induced Arthritis Model of Chronic Pain
Physiology & Behavior, Vol. 62, No. 1, pp. 199–205, 1997 Copyright q 1997 Elsevier Science Inc. Printed in the USA. All rights reserved 0031-9384/97 $17.00 / .00
PII S0031-9384(97)00158-3
BRIEF COMMUNICATION
Pain-Related Disability and Effects of Chronic Morphine in the Adjuvant-Induced Arthritis Model of Chronic Pain CHRIS K. CAIN, JONATHAN M. FRANCIS, MELISSA A. PLONE, DWAINE F. EMERICH AND MARK D. LINDNER 1 CytoTherapeutics Inc., 2 Richmond Square, Providence, RI 02906 USA Received 15 October; Accepted 10 January 1997 CAIN, C. K., J. M. FRANCIS, M. A. PLONE, D. F. EMERICH AND M. D. LINDNER. Pain-related disability and effects of chronic morphine in the adjuvant-induced arthritis model of chronic pain. PHYSIOL BEHAV 62(1), 199– 205, 1997.—Functional disability has been identified as one of the most important aspects of chronic pain, yet modeling pain-related disability has received little attention. Adjuvant-induced arthritis was induced, and one group of arthritic rats was implanted with SC 75-mg morphine pellets 1 week postadjuvant, and reimplanted every 2 weeks thereafter. The results confirm that the rodent adjuvantinduced arthritis model of severe chronic pain can be used to model pain-related disability: spontaneous activity levels and ambulatory function were reduced in arthritic rats and they exhibited substantial weight loss. The results of the present study suggest that the operant delayed nonmatching-to-position task can be used as a measure of pain-related disability, which may be especially relevant to the effects of chronic pain on performance in a work setting. The delayed nonmatching-to-position operant bar-pressing task is an ‘‘apical’’ test that is sensitive to deficits across a wide range of behavioral functions: motor ability, attention, motivation, learning, and memory, and arthritic rats were severely impaired in this task. In addition, analgesic treatments that impair functional abilities in normal healthy rats may actually improve the performance of rats exhibiting pain-related disability. Previous work demonstrated that acute morphine injections of only 4 mg/kg impaired performance in the delayed matching-toposition task. The results of the present study demonstrate that chronic morphine attenuates the degree of pain-related disability exhibited by arthritic rats in the test of ambulatory function and the delayed nonmatching-to-position bar-pressing test. These results demonstrate that novel analgesic treatments can be screened preclinically, both with respect to their direct analgesic effects, and with respect to their ability to reduce pain-related disability. q 1997 Elsevier Science Inc. Chronic pain
Disability
Delayed nonmatching to position
CHRONIC pain is disabling: it impairs cognitive functions such as concentration and memory (14,24,27,37,51); it disrupts the sleep cycle (14,24); and it produces extreme, and apparently pathological, changes in personality (2,22). Chronic pain leads to a decrease in activity levels that extends into all the activities of daily living, to the point that people lose their jobs (29,36,54). Chronic pain can be so disabling that people even stop participating in social and other pleasurable activities, including sex (24,32). Depression (5,18,21,26,42,43,50) and thoughts of suicide (23) are common among chronic pain patients, largely because of the extreme disability caused by the chronic pain (21). Disability is such an integral part of the chronic pain syndrome that numerous batteries have been developed to assess the degree of disability in chronic pain patients (21,25,41). In fact, the financial costs of disability are greater than any other costs related to chronic pain, including medical costs, and,
1
Operant conditioning
Morphine
Animal models
for that reason, it has even been suggested that the primary goal of treatment for chronic pain should be the prevention or reduction of prolonged disability (54). Despite the emphasis on pain-related disability in the clinical setting, very little attention has been devoted to measuring painrelated disability in preclinical studies. One clinically relevant model of chronic and severe pain-related disability is the adjuvant-induced arthritis model in rodents (3,4,8,9,11,12,19,39,45,47). In this model, Freund’s adjuvant (a suspension of heat-killed Mycobacterium butyricum in mineral oil) is injected into the base of the tail. Hyperalgesia develops to pressure on the paws (4) and bilateral distal arthritis becomes apparent 10–16 days postinoculation (11), with increased hindpaw and forepaw joint diameters due to inflamation and swelling. From 14–28 days postinjection, body weights and activity levels decline relative to controls, and do not begin to return
To whom requests for reprints should be addressed. E-mail: [email protected]
199 / eh0f 2677 Mp 199 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
200
CAIN ET AL.
to control values until 63–77 days postinjection (4). The validity of the adjuvant-induced arthritis model of chronic pain is supported by the fact that decreased body weights, activity levels, and hyperalgesia to paw pressure are attenuated in arthritic rats if their afferent nociceptive pathways have been interrupted (i.e., if the spinothalamic and spinoreticulothalamic pathways in the ventrolateral funiculus are lesioned) (11). This suggests that disability in this model can be attributed to pain, as opposed to joint immobility. Adjuvantinduced arthritis in rats may, thus, provide a model to measure the potential efficacy of analgesic agents with respect to their effects on pain-related disability. Based on the emphasis clinicians give to pain-related disability, it would be valuable to select novel therapeutic agents on the basis of their ability to reduce pain-related disability, in addition to their specific direct analgesic effects. For example, treatments that relieve pain and reduce pain-related disability should be of greater interest than treatments that relieve pain but have no effect, or perhaps even increase, the degree of overall disability. Opiates are commonly used and are often effective in alleviating the suffering of pain, but opiates themselves can be disabling. Opiates disrupt cognitive functions, such as memory (6), and they produce confusion and hallucinations (52). Opiates impair motor function (6) and they cause constipation, nausea, and vomiting (46,52,56). Morphine produces analgesic effects in the adjuvantinduced arthritis model of chronic pain: vocalizations in arthritic rats are attenuated with fairly low doses of opiates (39), arthritic rats will self-administer opiates more than control rats, and the amount of opiates that they self-administer follows the course of the nociceptive behaviors (peak intake at the time of peak nociceptive behaviors) (8). Although it has been reported that, in response to acute nociceptive stimuli, tolerance develops to the antinociceptive effects of morphine more readily than to the disruptive effects that morphine has on global behavioral function (48), there is little evidence that tolerance develops to the analgesic effects of opiates in the arthritis model of chronic pain. At the same time, measures of apparent painrelated disability in this model were reportedly compounded by morphine (i.e., activity levels in morphine-treated arthritic rats were decreased even below the levels of nonmorphine-treated arthritic rats) (12). In other words, morphine seems to provide pain relief and attenuate the overt signs of suffering from pain (it decreases the degree of vocalizations), but it may not reduce the degree of disability in the adjuvant-induced arthritis model of severe chronic pain. Although there is good evidence supporting the validity of rodent adjuvant-induced arthritis as a model of chronic pain, additional behavioral measures might be useful to quantify the degree of disability produced in this model. Measures of spontaneous activity levels, time spent eating, and body weight are valid and informative, but behavioral measures that are more analogous to performance in a work setting would further increase the value of preclinical studies that assess the potential therapeutic efficacy of novel treatments. One sophisticated measure of global behavioral function that might be more clinically relevant, that seems analogous to performance in the work setting, is the delayed nonmatching-to-position operant-conditioning task (16,17), which is sensitive to deficits in learning, memory, attention, motivation, and motor function (49). This task would seem to be ideal for quantifying the degree of global disability produced in the adjuvant-induced arthritis model and the potential efficacy that analgesic treatments might have in reducing the degree of pain-related disability. The objectives of the present study were to determine: 1. If the level of pain-related disability can be reliably and precisely quantified in rats with adjuvant-induced arthritis using global measures of behavioral function, such as a delayed nonmatching-to-position operant-conditioning task, and 2. to determine what effect chronic morphine might have on the degree of disability in this global measure of behavioral function.
METHODS
Subjects All experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee before the study was initiated, and the study was conducted in an animal care facility accredited by the American Association for Accreditation of Laboratory Animal Care. Male Lewis rats (n Å 32, Harlan Sprague– Dawley, Indianapolis, IN) weighing 263 { 18 g at the time of surgery were group housed and maintained at 227C on a 12:12-h light:dark cycle. Nonarthritis control animals were also group housed, but were not mixed together with the arthritis rats. To ensure performance in a food-reward task, the rats were placed on a restricted feeding schedule of 20 g/day of standard rat chow (Rodent Lab Chow 5001, Purina, Richmond, IN), following the daily behavioral test session. At 7 days following surgery, all arthritic animals were given access to liquid diet (Liquid Diet AIN-76, BioServ, Frenchtown, NJ) for 4–6 h following the operant test session. The experimenter was blinded to treatment classification throughout the study, although inflammation related to the adjuvant was evident during behavioral testing. Freund’s Adjuvant Injection Polyarthritis was induced by intradermally injecting 0.03 cc of 5.0 mg/ml Freund’s adjuvant (heat-killed Mycobacterium butyricum in mineral oil) into the left and right side of the base of the tail (0.06 cc total). Adjuvant was injected in 20 rats anesthetized with sodium pentobarbital (50 mg/kg IP, New England Pharmaceuticals, Providence, RI). Morphine Pellets Morphine pellets (75 mg, NIDA, Rockville, MD) were implanted SC, 1 per rat, 1 week following adjuvant administration, and rats were reimplanted under sodium pentobarbital anesthesia every 2 weeks (35,40) until the end of the study. Inflammation and Weight Loss Body weights and paw volumes were recorded weekly throughout the course of the study. Paw volumes were quantified with the use of a special plethysmometer (Model 7140, Ugo Basile, Italy) designed for measuring rat paw oedema. Activity Levels Rats were placed in a Plexiglas cage (42 cm 1 42 cm 1 30 cm) with arrays of infrared beams around the perimeter (Integrated Animal Monitoring System, Omnitech Electronics, Inc., Columbus, OH) and a thin layer of bedding on the bottom of the cage. Rats were tested for 16-h sessions 25 and 45 days after adjuvant administration. Beam breaks were converted into total distance traveled per 2-h block. Rotarod Testing Motor/ambulatory function was assessed with a rotarod (CR1 Rotamex System, Columbus Instruments, Columbus, OH). Prior to the adjuvant injections, the animals were shaped to ambulate on the rotarod to avoid a 3-s shock of 0.4 mA, and then trained until they could stay on the rotarod for 60 s at a constant rate of 5 rpm. Beginning 13 days postinjection, rats were tested weekly for one 60-s trial with a constant rate of 5 rpm.
/ eh0f 2677 Mp 200 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
PAIN-RELATED DISABILITY
201
FIG. 1. Paw volumes and body weights, relative to the day of adjuvant administration. Arthritic rats exhibited inflammation of the paws (A) and weight loss (B), neither of which was affected by SC morphine pellets.
Delayed Nonmatching to Position This training was performed with operant boxes (Coulbourn Instruments, Allentown, PA) equipped with a retractable lever on either side of the food bin. The rats were taught to press the levers and then trained to perform nonmatching to position. For nonmatching-to-position training, rats were presented with either the right or left lever at random. This target lever retracted as soon as it was pressed or after 15 s. The rat was then required to break a photobeam by poking its nose into the food bin. This nose poke was immediately followed by the presentation of both levers. If the rat pressed the lever that had not just been presented, he was rewarded with a 45-mg food pellet (Formula A/I, P.J. Noyes Co. Inc., Lancaster, NH) in the food bin. If the rat pressed the same lever that was initially presented or did not press either lever within 15 s, both levers were retracted and the rat did not receive a food reward. This was followed by a 5-s intertrial interval, during which the house light was turned off and both levers were retracted. Initial training, without delays, was conducted during daily 30-min sessions, 7 days each week, until criteria (90% correct) was reached. After reaching criteria in the nonmatching-to-position task, rats were trained to perform a delayed nonmatching-to-position task (DNMTP). The procedures for this task were the same as the nonmatching to position described above, except that a delay of 0, 1, 5, 10, 20, or 30 s, selected at random, followed the presentation of the target lever. The first nose poke into the food bin following the delay period initiated the presentation of both levers as before. Training to criteria included approximately 45 sessions from 50 to 4 days before adjuvant administration. Daily 30-min sessions were conducted 7 days/week from 21–50 days after adjuvant administration. Data was combined into 6 5-session blocks for analyses and graphic presentation. Data Analyses Data were analyzed with SAS-PC. Analyses of variance were conducted using the procedures for general linear models with
options for repeated measures where appropriate, and Cronbach’s coefficient alpha was computed as the measure of internal reliability (44). Omega squared was computed as a measure of effect size (13,28). Data is presented in the text and in all figures as means { standard errors of the mean (SEMs). RESULTS
Mortality Rates From the beginning to the completion of behavioral testing, 4 rats died in the Arthritis with Morphine Group. Three of the rats never recovered from morphine pellet-implantation surgery. All rats in the Control and Arthritis Only Groups survived to the end of the study. Mortality rates were 0% (0 of 11) for the Control Group, 0% (0 of 10) for the Arthritis Only Group, and 36% (4/11) for the Arthritis with Morphine group. Only the data from the rats that survived to the end of the study was included in the analyses and graphic presentation of the behavioral data, which resulted in groups of the following sizes: Control Group (no arthritis and no morphine, n Å 11), Arthritis Only Group (n Å 10), and Arthritis with Morphine Group (n Å 7). Paw Volume Paw volumes were analyzed in a 3 1 7 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), and Weeks (1–7) as factors in the analysis, treating Weeks as a repeated measure. Substantial inflammation and swelling was evident in the rats treated with adjuvant, becoming evident 2 weeks postinnoculation and peaking 3 weeks postinnoculation. Planned contrasts between the Control rats and the Arthritis Only rats showed that the Arthritis Only paw volumes were larger than those of the Controls, the main effect for Group was statistically significant, F(1,25) Å 70.55, p Å 0.0001; and the Group 1 Trial interaction was statistically significant, F(6,150) Å 15.71, p Å 0.0001. Morphine had no detectable direct effect on the degree of inflammation produced by the adjuvant. Planned contrasts between the Arthritis Only and the Arthritis with Morphine Groups failed to detect any significant dif-
/ eh0f 2677 Mp 201 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
202
CAIN ET AL.
ference in paw volumes between these Groups, the main effect was not statistically significant, F(1,25) Å 2.13, p Å 0.16; and the Group 1 Weeks interaction was not statistically significant, F(6,150) Å 1.97, p Å 0.073 (Fig. 1A). Weight Body weights were analyzed in a 3 1 8 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), and Weeks (1–8) as factors in the analysis, with Weeks treated as a repeated measure. All groups were initially equivalent with respect to body weights. The control rats maintained their weights, and the rats dosed with adjuvant exhibited substantial weight loss until 21 days after adjuvant administration. A planned contrast between the Control and Arthritis Only rats revealed that the Control rats were heavier than the arthritic rats, the main effect for Group was statistically significant, F (1,25) Å 117.27, p Å 0.0001; and the Group 1 Weeks interaction was statisically significant, F (7,175) Å 27.12, p Å 0.0001. Morphine had no detectable effect on body weights over the course of the study. A planned contrast failed to detect any differences between the Arthritis Only and the Arthritis with Morphine Groups, the main effect for Group was not statistically significant, F (1,25) Å 0.64, p Å 0.43; and the Group 1 Weeks interaction was not statistically significant, F (7,175) Å 1.95, p Å 0.06 (Fig. 1B).
F(7,175) Å 9.88, p Å 0.0001. Planned contrasts revealed that the Arthritis Only rats were much less active than the Control rats, the main effect for Group was statistically significant, F(1,25) Å 98.57, p Å 0.0001, as was the Group 1 Time interaction, F(7,175) Å 8.64, p Å 0.0001 (Fig. 2A). Morphine had no detectable effect on activity level, but activity levels were so low that it was not possible to detect any additional decrease in activity level. A planned contrast between the Arthritis Only Group and the Arthritis with Morphine Group failed to detect any effect of morphine on activity levels, the main effect for Group was not statistically significant, F(1,25) Å 0.12, p Å 0.73; and the Group 1 Time interaction was not statistically significant, F (7,175) Å 0.44, p Å 0.88. A planned contrast on activity levels 45 days postinjection revealed that the Arthritis Only rats were still much less active than the Control rats, the main effect for Group was statistically significant, F(1,25) Å 42.51, p Å 0.0001; and the Group 1 Time interaction was statistically significant, F (7,175) Å 4.16, p Å 0.0003 (Fig. 2B). There was no significant difference in activity levels between the Arthritis Only Group and the Arthritis with Morphine Group, the main effect for Group was not statistically significant, F(1,25) Å 1.46, p Å 0.24; and the Group 1 Time interaction was not statistically significant, F (7,175) Å 1.47, p Å 0.18. Rotarod
Activity Levels The total distance traveled during activity-monitoring sessions was analyzed with a 3 1 8 ANOVA that included Groups (Arthritis Only, Arthritis with Morphine and Control), and Blocks (1–8), as factors in the analysis, treating Blocks as a repeated measure. Data from the activity-monitoring session conducted 25 and 45 days postinnoculation were analyzed separately. Rats seemed to habituate to the test apparatus. At 25 days postinjection, activity levels declined over the course of the session, the main effect of Time was statistically significant,
The latencies to fall off the rotarod were analyzed in a 3 1 7 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), and Trials (1–7) as factors in the analysis, with Trial treated as a repeated measure. Arthritic rats were substantially impaired in this test of ambulatory function. A planned contrast revealed that the Arthritis Only rats fell off the rotarod much sooner than the Control rats, the main effect for Group was statistically significant, F (1,25) Å 179.52, p Å 0.0001; and the Group 1 Trial interaction was statistically significant, F (6,150) Å 2.41, p Å 0.03 (Fig. 3A). Planned contrasts between the Ar-
FIG. 2. Activity levels 25 (A) and 45 days (B) postinnoculation for 16 h after being put in the activity chambers. Arthritic rats were almost completely inactive, much less active than the controls, and morphine had no detectable effect on activity level. However, the virtually complete inactivity in the Arthritis Only Group made it impossible to determine if morphine might cause any further decrease in activity level.
/ eh0f 2677 Mp 202 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
PAIN-RELATED DISABILITY
203
FIG. 3. (A) Rotarod test of ambulatory function - latency to fall at 5 rpm. (B) Pellets earned in 5session blocks of delayed nonmatching-to-position operant sessions. Adjuvant-induced arthritis produced severe deficits in this test but, surprisingly, morphine actually attenuated these deficits.
thritis Only and the Arthritis with Morphine rats revealed that morphine improved performance on the rotarod task, and the main effect for Group was statistically significant, F (1,25) Å 7.20, p Å 0.01. The effects of morphine emerged over time, becoming especially apparent as the effects of the adjuvant began to subside, and the Group 1 Trial interaction was statistically significant, F (6,150) Å 3.8, p Å 0.02. Delayed Nonmatching to Position The number of pellets earned were analyzed in a 3 1 6 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), and Block (1–6) as factors in the analysis, with Block treated as a repeated measure. Planned contrasts revealed that the Arthritis Only rats were impaired relative to the Control rats with respect to number of pellets earned, the main effect for Group was statistically significant, F (1,25) Å 36.39, p Å 0.0001. The Arthritis Only rats did improve over the course of the study relative to the Controls, although they never approached the levels of the Controls. The Group 1 Trial interaction was statistically significant, F (5,125) Å 5.70, p Å 0.0001. Morphine improved performance with respect to the number of pellets earned in the DNMTP task and seemed to facilitate recovery from the effects of the adjuvant, improving performance approximately to the level of nonarthritis Controls by the last Block of testing, and the Group 1 Trials interaction was statistically significant, F (5,125) Å 3.94, p Å 0.002 (Fig. 3B). In addition to differences in the number of rewards earned, analyses of the percentage of correct choices suggests differences between the groups in performance with respect to the cognitive demands of the task. The percentage of correct trials at each delay were computed for the first and last 5-session blocks and analyzed in a 3 1 6 1 2 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), Delay (0, 1, 5, 10, 20, and 30 s), and Blocks (1 & 6) as factors in the analysis, treating Delays and Blocks as repeated measures. Animals generally performed better at the shorter delays and the percent correct declined as the delay duration
increased, and the main effect for Delay was statistically significant, F(5,125) Å 13.83, p Å 0.0001. All groups performed better in the last 5-session block than in the first; the main effect for Block was statistically significant, F(1,25) Å 36.22, p Å 0.0001. Planned contrasts revealed that the Arthritis Only rats performed poorly relative to the Controls; the main effect for Group was statistically significant, F(1,25) Å 16.00, p Å 0.0005. Both Arthritis Only and Arthritis with Morphine Groups improved over the course of the study, relative to the Controls, and the Group 1 Block interaction was statistically significant, F(1,25) Å 9.84, p Å 0.0043 (Fig. 4A, B). Planned contrasts between the Arthritis Only and the Arthritis with Morphine Group also revealed that the Arthritis with Morphine Group performed better on this measure than the Arthritis Only Group; the main effect for Group was statistically significant, F(1,25) Å 6.31, p Å 0.02. DISCUSSION
Because overall functional disability has been identified as perhaps the most important consequence of chronic pain in the clinical setting, considerable attention should be devoted to modeling this aspect of chronic pain syndromes in preclinical studies. The results of the present study confirm that the rodent adjuvant-induced arthritis model of severe chronic pain can be used to model pain-related disability. Consistent with previous reports, spontaneous activity levels and ambulatory function are reduced in arthritic rats and they exhibit substantial weight loss (4,12), presumably due to the fact that it is painful to move, and also possibly because the pain reduces appetite. Novel findings from the present study suggest that the operant delayed nonmatchingto-position task can provide an additional measure of the degree of pain-related disability that may be especially relevant to the effects of chronic pain on performance in a work setting. Originally developed by Dunnett to study cognitive dysfunction (15,17), the delayed nonmatching-to-position operant bar-pressing task is an ‘‘apical’’ test that is sensitive to deficits across a wide range of behavioral functions (30): motor ability, attention,
/ eh0f 2677 Mp 203 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
204
CAIN ET AL.
FIG. 4. Performance across delays in the delayed nonmatching-to-position operant task during the first (A) and last (B) 5-session blocks of trials. These results suggest that chronic severe pain is disruptive even with respect to cognitive functions, and that morphine may improve performance in this test of global behavioral function.
motivation, learning, and memory (49). It is important to note that, although arthritic rats were impaired in this task, they were still able to peform well enough so that additional deficits could have been detected. The potential sensitivity of the delayed nonmatching-to-position task to increased disability is in contrast to the floor effects observed with the measures of spontaneous activity and ambulatory function. One reason that preclinical models of pain-related disability would be valuable is to screen novel treatments. In particular, it would be especially valuable to be able to select novel analgesic agents that reduce disability, in addition to providing direct analgesic activity. Most preclinical studies continue to focus almost exclusively on assessing the potential analgesic effects of novel agents, with little regard for their potential to ultimately improve quality of life. Preclinical research assessing nonanalgesic effects of novel treatments is usually restricted to assessing the side-effect profile in normal healthy animals. Another novel finding reported from the present study is that analgesic treatments that impair functional abilities in normal healthy rats may actually improve the performance of rats exhibiting pain-related disability. Previous work demonstrated that acute morphine injections of only 4 mg/kg impaired performance of rats in the delayed matching-to-position task (31). The results of the present study suggest that chronic morphine attenuates the degree of pain-related disability exhibited by arthritic rats in the test of ambulatory function and the delayed nonmatchingto-position bar-pressing test. The fact that chronic morphine had such long-lasting effects, reducing disability for the duration of this
2-month study in the delayed nonmatching-to-position task, may also seem surprising, in view of the literature demonstrating rapid development of tolerance to chronic morphine. However, the present results are consistent with numerous reports that chronic morphine does not produce tolerance in patients suffering with chronic pain (20,34,38,53,55). In addition, a previous study reported that rats with adjuvant-induced arthritis will self-administer morphine, and that chronic pain reduces the addictive properties of morphine (33). Combined with results from clinical and preclinical studies that subjects experiencing pain become tolerant to morphine more slowly than subjects not experiencing pain (1,7,10,53), the present results suggest that analgesic effects, tolerance, and side-effect profiles of potential analgesic agents should all be assessed in animals experiencing pain, as well as in normal healthy rats. In conclusion, the results of this study indicate that 1. the delayed nonmatching-to-position operant task is a reliable and accurate means of quantifying pain-related disability in the rodent adjuvantinduced arthritis model of severe chronic pain, and 2. constant, moderate levels of systemic morphine actually attenuate pain-related disability associated with adjuvant-induced arthritis, which is in contrast to the deleterious effects morphine injections have on the performance of healthy normal rats in this task. ACKNOWLEDGEMENTS
The authors thank ASTRA AB for funding this research, and Jaci Sagen for generously providing the morphine pellets used in this study.
REFERENCES 1. Backonja, M. M.; Miletic, G.; Miletic, V. The effect of continuous morphine analgesia on chronic thermal hyperalgesia due to sciatic constriction injury in rats. Neurosci. Lett. 196:61–64; 1995. 2. Barnes, D.; Gatchel, R. J.; Mayer, T. G.; Barnett, J. Changes in MMPI profile levels of chronic low back pain patients following successful treatment. J. Spinal. Disord. 3:353–355; 1990.
3. Calvino, B.; Couraud, J. Y.; Besson, J. M. Prevaccination with diluted Freund adjuvant prevents the development of chronic pain and transient release of cerebrospinal fluid substance P in adjuvant-induced arthritis in rats. Pain 58:211–217; 1994. 4. Calvino, B.; Crepon-Bernard, M. O.; Le Bars, D. Parallel clinical and behavioural studies of adjuvant-induced arthritis in the rat: pos-
/ eh0f 2677 Mp 204 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
PAIN-RELATED DISABILITY
5. 6.
7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.
sible relationship with ’chronic pain’. Behav. Brain Res. 24:11–29; 1987. Cheatle, M. D.; Brady, J. P.; Ruland, T. Chronic low back pain, depression, and attributional style. Clin. J. Pain 6:114–117; 1990. Coda, B. A.; Hill, H. F.; Hunt, E. B.; Kerr, E. B.; Jacobson, R. C.; Chapman, C. R. Cognitive and motor function impairments during continuous opioid analgesic infusions. Hum. Psychopharmacol. Clin. Exp. 8:383–400; 1993. Colpaert, F. C. Long-term suppression of pain by narcotic drugs in the absence of tolerance development. Arch. Int. Pharmacodyn. Ther. 236:293–295; 1978. Colpaert, F. C. Evidence that adjuvant arthritis in the rat is associated with chronic pain. Pain 28:201–222; 1987. Colpaert, F. C.; Meert, T.; De Witte, P.; Schmitt, P. Further evidence validating adjuvant arthritis as an experimental model of chronic pain in the rat. Life Sci. 31:67–75; 1982. Colpaert, F. C.; Niemegeers, C. J.; Janssen, P. A. Nociceptive stimulation prevents development of tolerance to narcotic analgesia. Eur. J. Pharmacol. 49:335–336; 1978. Dardick, S. J.; Basbaum, A. I.; Levine, J. D. The contribution of pain to disability in experimentally induced arthritis. Arthritis Rheum. 29:1017–1022; 1986. De Castro Costa, M.; De Sutter, P.; Gybels, J.; Van Hees, J. Adjuvant-induced arthritis in rats: a possible animal model of chronic pain. Pain 10:173–185; 1981. Dodd, D. H.; Schultz, R. F. Jr. Computational procedures for estimating magnitude of effect for some analysis of variance designs. Psychol. Bull. 79:391–395; 1973. Duggleby, W.; Lander, J. Cognitive status and postoperative pain older adults. J. Pain Symptom Manage. 9:19–27; 1994. Dunnett, S. B. Comparative effects of cholinergic drugs and lesions of nucleus basalis or fimbria-fornix on delayed matching in rats. Psychopharmacology (Berlin) 87:357–363; 1985. Dunnett, S. B. Role of prefrontal cortex and striatal output systems in short-term memory deficits associated with ageing, basal forebrain lesions, and cholinergic-rich grafts. (abst.) Can. J. Psychol. 44:210–232; 1990. Dunnett, S. B.; Evenden, J. L.; Iversen, S. D. Delay-dependent short-term memory deficits in aged rats. Psychopharmacology 96:174–180; 1988. Dworkin, R. H.; Gitlin, M. J. Clinical aspects of depression in chronic pain patients. Clin. J. Pain 7:79–94; 1991. Flecknell, P. A. Refinement of animal use—assessment and alleviation of pain and distress. Lab. Anim. 28:222–231; 1994. Follett, K. A.; Hitchon, P. W.; Piper, J.; Kumar, V.; Clamon, G.; Jones, M. P. Response of intractable pain to continuous intrathecal morphine: a retrospective study. Pain 49:21–25; 1992. Herr, K. A.; Mobily, P. R.; Smith, C. Depression and the experience of chronic back pain: a study of related variables and age differences. Clin. J. Pain 9:104–114; 1993. Hinkley, B. S.; Jaremko, M. E. Effects of pain duration on psychosocial adjustment in orthopedic patients—the importance of early diagnosis and treatment of pain. J. Pain Symptom Manage. 9:175–185; 1994. Hitchcock, L. S.; Ferrell, B. R.; Mccaffery, M. The experience of chronic nonmalignant pain. J. Pain Symptom Manage. 9:312–318; 1994. Jamison, R. N.; Sbrocco, T.; Parris, W. C. The influence of problems with concentration and memory on emotional distress and daily activities in chronic pain patients. Int. J. Psychiatry Med. 18:183–191; 1988. Jensen, M. P.; Strom, S. E.; Turner, J. A.; Romano, J. M. Validity of the sickness impact profile Roland scale as a measure of dysfunction in chronic pain patients. Pain 50:157–162; 1992. Kerns, R. D.; Haythornthwaite, J. A. Depression among chronic pain patients: cognitive-behavioral analysis and effect on rehabilitation outcome. J. Consult. Clin. Psychol. 56:870–876; 1988. Kewman, D. G.; Vaishampayan, N.; Zald, D.; Han, B. Cognitive impairment in musculoskeletal pain patients. Int. J. Psychiatry Med. 21:253–262; 1991. Kirk, R. E. Experimental design: procedures for the behavioral sciences. Pacific Grove, CA: Brooks/Cole Publishing Company; 1982. Latham, J.; Davis, B. D. The socioeconomic impact of chronic pain. Disabil. Rehabil. 16:39–44; 1994. Laties, V. G. Schedule-controlled operant behavior during the hazard identification stage of neurotoxicity risk assessment. In: Weiss, B.; Odonoghue, J. L., eds. Neurobehavioral toxicity: Anaylsis and interpretation. New York: Raven Press; 1994:253–262.
205 31. Lindner, M. D.; Frydel, B.; Kaplan, F. A.; Krueger, P. M.; Bell, W. J.; Blaney, T. J.; Winn, S. R.; Sherman, S. S.; Doherty, E. J.; Emerich, D. F. Intraventricular encapsulated calf adrenal chromaffin cells: viable for at least 500 days in vivo without detectable adverse effects on behavioral/cognitive function or host immune sensitization in rats. Restor. Neurol. Neurosci. 11:21–35; 1997. 32. Linton, S. J. Activities of daily living scale for patients with chronic pain. Percept. Mot. Skills. 71:722; 1990. 33. Lyness, W. H.; Smith, F. L.; Heavner, J. E.; Iacono, C. U.; Garvin, R. D. Morphine self-administration in the rat during adjuvant-induced arthritis. Life Sci. 45:2217–2224; 1989. 34. Meed, S. D.; Kleinman, P. M.; Kantor, T. G.; Blum, R. H.; Savarese, J. J. Management of cancer pain with oral controlled-release morphine sulfate. J. Clin. Pharmacol. 27:155–161; 1987. 35. Meyer, D. R.; Sparber, S. B. A comparison of withdrawal in rats implanted with different types of morphine pellets. Pharmacol. Biochem. Behav. 5:603–607; 1976. 36. Millard, R. W.; Wells, N.; Thebarge, R. W. A comparison of models describing reports of disability associated with chronic pain. Clin. J. Pain 7:283–291; 1991. 37. Pearce, S. A.; Isherwood, S.; Hrouda, D.; Richardson, P. H.; Erskine, A.; Skinner, J. Memory and pain: tests of mood congruity and state dependent learning in experimentally induced and clinical pain. Pain 43:187–193; 1990. 38. Penn, R. D.; Paice, J. A. Chronic intrathecal morphine for intractable pain. J. Neurosurg. 67:182–186; 1987. 39. Pircio, A. W.; Fedele, C. T.; Bierwagen, M. E. A new method for the evaluation of analgesic activity using adjuvant-induced arthritis in the rat. Eur. J. Pharm. 31:207–215; 1975. 40. Riffee, W. H.; Wilcox, R. E.; Anderson, J. A.; McGinity, J. W. Polydimethylsiloxane pellets for sustained delivery of morphine in mice. J. Pharm. Sci. 69:980–982; 1980. 41. Romano, J. M.; Turner, J. A.; Jensen, M. P. The chronic illness problem inventory as a measure of dysfunction in chronic pain patients. Pain 49:71–75; 1992. 42. Roy, R. Many faces of depression in patients with chronic pain. Int. J. Psychiatry Med. 12:109–119; 1982. 43. Rudy, T. E.; Kerns, R. D.; Turk, D. C. Chronic pain and depression: toward a cognitive-behavioral mediation model. Pain 35:129–140; 1988. 44. SAS Institute Inc. SAS/STAT User’s Guide, Version 6. Vol.1. 4th ed. Cary, NC: SAS Institute Inc. 1989. 45. Schaible, H. G.; Grubb, B. D. Afferent and spinal mechanisms of joint pain. Pain 55:5–54; 1993. 46. Schug, S. A.; Zech, D.; Grond, S.; Jung, H.; Meuser, T.; Stobbe, B. A long-term survey of morphine in cancer pain patients. J. Pain Symptom. Manage. 7:259–266; 1992. 47. Shippenberg, T. S.; Stein, C.; Huber, A.; Millan, M. J.; Herz, A. Motivational effects of opioids in an animal model of prolonged inflammatory pain:alteration in the effects of kappa- but not of mureceptor agonists. Pain 35:179–186; 1988. 48. Solomon, R. E.; Wasserman, E. A.; Gebhart, G. F. Tolerance to antinociceptive effects of morphine without tolerance to its effects on schedulecontrolled behavior. Psychopharmacology (Berlin) 92:327–333; 1987. 49. Stanhope, K. J.; Mclenachan, A. P.; Dourish, C. T. Dissociation between cognitive and motor/motivational deficits in the delayed matching to position test: Effects of scopolamine, 8-OH-DPAT and EAA antagonists. Psychopharmacology 122:268–280; 1995. 50. Sullivan, M. J.; Reesor, K.; Mikail, S.; Fisher, R. The treatment of depression in chronic low back pain: review and recommendations. Pain 50:5–13; 1992. 51. Tandon, O. P.; Kumar, S. P3 event related cerebral evoked potential in chronic pain patients. Indian J. Physiol. Pharmacol. 37:51–55; 1993. 52. Tsuneto, S.; Hayashi, A.; Miyazaki, M.; Kashiwagi, T. A clinical survey of controlled-release morphine sulphate for cancer pain relief in a Japanese hospice. Postgrad. Med. J. 67 (Suppl. 2):S79–S81; 1991. 53. Twycross, R. G. Clinical experience with diamorphine in advanced malignant disease. Int. J. Clin. Pharmacol. Ther. Toxicol. 9:184–198; 1974. 54. Webster, B. S.; Snook, S. H. The cost of 1989 workers’ compensation low back pain claims. Spine 19:1111–1116; 1994. 55. Winkelmuller, M.; Winkelmuller, W. Long-term effects of continuous intrathecal opioid treatment in chronic pain of nonmalignant etiology. J. Neurosurg. 85:458–467; 1996. 56. Zenz, M.; Strumpf, M.; Tryba, M. Long-term oral opioid therapy in patients with chronic nonmalignant pain. J. Pain Symptom. Manage. 7:69–77; 1992.
/ eh0f 2677 Mp 205 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
PII S0031-9384(97)00158-3
BRIEF COMMUNICATION
Pain-Related Disability and Effects of Chronic Morphine in the Adjuvant-Induced Arthritis Model of Chronic Pain CHRIS K. CAIN, JONATHAN M. FRANCIS, MELISSA A. PLONE, DWAINE F. EMERICH AND MARK D. LINDNER 1 CytoTherapeutics Inc., 2 Richmond Square, Providence, RI 02906 USA Received 15 October; Accepted 10 January 1997 CAIN, C. K., J. M. FRANCIS, M. A. PLONE, D. F. EMERICH AND M. D. LINDNER. Pain-related disability and effects of chronic morphine in the adjuvant-induced arthritis model of chronic pain. PHYSIOL BEHAV 62(1), 199– 205, 1997.—Functional disability has been identified as one of the most important aspects of chronic pain, yet modeling pain-related disability has received little attention. Adjuvant-induced arthritis was induced, and one group of arthritic rats was implanted with SC 75-mg morphine pellets 1 week postadjuvant, and reimplanted every 2 weeks thereafter. The results confirm that the rodent adjuvantinduced arthritis model of severe chronic pain can be used to model pain-related disability: spontaneous activity levels and ambulatory function were reduced in arthritic rats and they exhibited substantial weight loss. The results of the present study suggest that the operant delayed nonmatching-to-position task can be used as a measure of pain-related disability, which may be especially relevant to the effects of chronic pain on performance in a work setting. The delayed nonmatching-to-position operant bar-pressing task is an ‘‘apical’’ test that is sensitive to deficits across a wide range of behavioral functions: motor ability, attention, motivation, learning, and memory, and arthritic rats were severely impaired in this task. In addition, analgesic treatments that impair functional abilities in normal healthy rats may actually improve the performance of rats exhibiting pain-related disability. Previous work demonstrated that acute morphine injections of only 4 mg/kg impaired performance in the delayed matching-toposition task. The results of the present study demonstrate that chronic morphine attenuates the degree of pain-related disability exhibited by arthritic rats in the test of ambulatory function and the delayed nonmatching-to-position bar-pressing test. These results demonstrate that novel analgesic treatments can be screened preclinically, both with respect to their direct analgesic effects, and with respect to their ability to reduce pain-related disability. q 1997 Elsevier Science Inc. Chronic pain
Disability
Delayed nonmatching to position
CHRONIC pain is disabling: it impairs cognitive functions such as concentration and memory (14,24,27,37,51); it disrupts the sleep cycle (14,24); and it produces extreme, and apparently pathological, changes in personality (2,22). Chronic pain leads to a decrease in activity levels that extends into all the activities of daily living, to the point that people lose their jobs (29,36,54). Chronic pain can be so disabling that people even stop participating in social and other pleasurable activities, including sex (24,32). Depression (5,18,21,26,42,43,50) and thoughts of suicide (23) are common among chronic pain patients, largely because of the extreme disability caused by the chronic pain (21). Disability is such an integral part of the chronic pain syndrome that numerous batteries have been developed to assess the degree of disability in chronic pain patients (21,25,41). In fact, the financial costs of disability are greater than any other costs related to chronic pain, including medical costs, and,
1
Operant conditioning
Morphine
Animal models
for that reason, it has even been suggested that the primary goal of treatment for chronic pain should be the prevention or reduction of prolonged disability (54). Despite the emphasis on pain-related disability in the clinical setting, very little attention has been devoted to measuring painrelated disability in preclinical studies. One clinically relevant model of chronic and severe pain-related disability is the adjuvant-induced arthritis model in rodents (3,4,8,9,11,12,19,39,45,47). In this model, Freund’s adjuvant (a suspension of heat-killed Mycobacterium butyricum in mineral oil) is injected into the base of the tail. Hyperalgesia develops to pressure on the paws (4) and bilateral distal arthritis becomes apparent 10–16 days postinoculation (11), with increased hindpaw and forepaw joint diameters due to inflamation and swelling. From 14–28 days postinjection, body weights and activity levels decline relative to controls, and do not begin to return
To whom requests for reprints should be addressed. E-mail: [email protected]
199 / eh0f 2677 Mp 199 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
200
CAIN ET AL.
to control values until 63–77 days postinjection (4). The validity of the adjuvant-induced arthritis model of chronic pain is supported by the fact that decreased body weights, activity levels, and hyperalgesia to paw pressure are attenuated in arthritic rats if their afferent nociceptive pathways have been interrupted (i.e., if the spinothalamic and spinoreticulothalamic pathways in the ventrolateral funiculus are lesioned) (11). This suggests that disability in this model can be attributed to pain, as opposed to joint immobility. Adjuvantinduced arthritis in rats may, thus, provide a model to measure the potential efficacy of analgesic agents with respect to their effects on pain-related disability. Based on the emphasis clinicians give to pain-related disability, it would be valuable to select novel therapeutic agents on the basis of their ability to reduce pain-related disability, in addition to their specific direct analgesic effects. For example, treatments that relieve pain and reduce pain-related disability should be of greater interest than treatments that relieve pain but have no effect, or perhaps even increase, the degree of overall disability. Opiates are commonly used and are often effective in alleviating the suffering of pain, but opiates themselves can be disabling. Opiates disrupt cognitive functions, such as memory (6), and they produce confusion and hallucinations (52). Opiates impair motor function (6) and they cause constipation, nausea, and vomiting (46,52,56). Morphine produces analgesic effects in the adjuvantinduced arthritis model of chronic pain: vocalizations in arthritic rats are attenuated with fairly low doses of opiates (39), arthritic rats will self-administer opiates more than control rats, and the amount of opiates that they self-administer follows the course of the nociceptive behaviors (peak intake at the time of peak nociceptive behaviors) (8). Although it has been reported that, in response to acute nociceptive stimuli, tolerance develops to the antinociceptive effects of morphine more readily than to the disruptive effects that morphine has on global behavioral function (48), there is little evidence that tolerance develops to the analgesic effects of opiates in the arthritis model of chronic pain. At the same time, measures of apparent painrelated disability in this model were reportedly compounded by morphine (i.e., activity levels in morphine-treated arthritic rats were decreased even below the levels of nonmorphine-treated arthritic rats) (12). In other words, morphine seems to provide pain relief and attenuate the overt signs of suffering from pain (it decreases the degree of vocalizations), but it may not reduce the degree of disability in the adjuvant-induced arthritis model of severe chronic pain. Although there is good evidence supporting the validity of rodent adjuvant-induced arthritis as a model of chronic pain, additional behavioral measures might be useful to quantify the degree of disability produced in this model. Measures of spontaneous activity levels, time spent eating, and body weight are valid and informative, but behavioral measures that are more analogous to performance in a work setting would further increase the value of preclinical studies that assess the potential therapeutic efficacy of novel treatments. One sophisticated measure of global behavioral function that might be more clinically relevant, that seems analogous to performance in the work setting, is the delayed nonmatching-to-position operant-conditioning task (16,17), which is sensitive to deficits in learning, memory, attention, motivation, and motor function (49). This task would seem to be ideal for quantifying the degree of global disability produced in the adjuvant-induced arthritis model and the potential efficacy that analgesic treatments might have in reducing the degree of pain-related disability. The objectives of the present study were to determine: 1. If the level of pain-related disability can be reliably and precisely quantified in rats with adjuvant-induced arthritis using global measures of behavioral function, such as a delayed nonmatching-to-position operant-conditioning task, and 2. to determine what effect chronic morphine might have on the degree of disability in this global measure of behavioral function.
METHODS
Subjects All experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee before the study was initiated, and the study was conducted in an animal care facility accredited by the American Association for Accreditation of Laboratory Animal Care. Male Lewis rats (n Å 32, Harlan Sprague– Dawley, Indianapolis, IN) weighing 263 { 18 g at the time of surgery were group housed and maintained at 227C on a 12:12-h light:dark cycle. Nonarthritis control animals were also group housed, but were not mixed together with the arthritis rats. To ensure performance in a food-reward task, the rats were placed on a restricted feeding schedule of 20 g/day of standard rat chow (Rodent Lab Chow 5001, Purina, Richmond, IN), following the daily behavioral test session. At 7 days following surgery, all arthritic animals were given access to liquid diet (Liquid Diet AIN-76, BioServ, Frenchtown, NJ) for 4–6 h following the operant test session. The experimenter was blinded to treatment classification throughout the study, although inflammation related to the adjuvant was evident during behavioral testing. Freund’s Adjuvant Injection Polyarthritis was induced by intradermally injecting 0.03 cc of 5.0 mg/ml Freund’s adjuvant (heat-killed Mycobacterium butyricum in mineral oil) into the left and right side of the base of the tail (0.06 cc total). Adjuvant was injected in 20 rats anesthetized with sodium pentobarbital (50 mg/kg IP, New England Pharmaceuticals, Providence, RI). Morphine Pellets Morphine pellets (75 mg, NIDA, Rockville, MD) were implanted SC, 1 per rat, 1 week following adjuvant administration, and rats were reimplanted under sodium pentobarbital anesthesia every 2 weeks (35,40) until the end of the study. Inflammation and Weight Loss Body weights and paw volumes were recorded weekly throughout the course of the study. Paw volumes were quantified with the use of a special plethysmometer (Model 7140, Ugo Basile, Italy) designed for measuring rat paw oedema. Activity Levels Rats were placed in a Plexiglas cage (42 cm 1 42 cm 1 30 cm) with arrays of infrared beams around the perimeter (Integrated Animal Monitoring System, Omnitech Electronics, Inc., Columbus, OH) and a thin layer of bedding on the bottom of the cage. Rats were tested for 16-h sessions 25 and 45 days after adjuvant administration. Beam breaks were converted into total distance traveled per 2-h block. Rotarod Testing Motor/ambulatory function was assessed with a rotarod (CR1 Rotamex System, Columbus Instruments, Columbus, OH). Prior to the adjuvant injections, the animals were shaped to ambulate on the rotarod to avoid a 3-s shock of 0.4 mA, and then trained until they could stay on the rotarod for 60 s at a constant rate of 5 rpm. Beginning 13 days postinjection, rats were tested weekly for one 60-s trial with a constant rate of 5 rpm.
/ eh0f 2677 Mp 200 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
PAIN-RELATED DISABILITY
201
FIG. 1. Paw volumes and body weights, relative to the day of adjuvant administration. Arthritic rats exhibited inflammation of the paws (A) and weight loss (B), neither of which was affected by SC morphine pellets.
Delayed Nonmatching to Position This training was performed with operant boxes (Coulbourn Instruments, Allentown, PA) equipped with a retractable lever on either side of the food bin. The rats were taught to press the levers and then trained to perform nonmatching to position. For nonmatching-to-position training, rats were presented with either the right or left lever at random. This target lever retracted as soon as it was pressed or after 15 s. The rat was then required to break a photobeam by poking its nose into the food bin. This nose poke was immediately followed by the presentation of both levers. If the rat pressed the lever that had not just been presented, he was rewarded with a 45-mg food pellet (Formula A/I, P.J. Noyes Co. Inc., Lancaster, NH) in the food bin. If the rat pressed the same lever that was initially presented or did not press either lever within 15 s, both levers were retracted and the rat did not receive a food reward. This was followed by a 5-s intertrial interval, during which the house light was turned off and both levers were retracted. Initial training, without delays, was conducted during daily 30-min sessions, 7 days each week, until criteria (90% correct) was reached. After reaching criteria in the nonmatching-to-position task, rats were trained to perform a delayed nonmatching-to-position task (DNMTP). The procedures for this task were the same as the nonmatching to position described above, except that a delay of 0, 1, 5, 10, 20, or 30 s, selected at random, followed the presentation of the target lever. The first nose poke into the food bin following the delay period initiated the presentation of both levers as before. Training to criteria included approximately 45 sessions from 50 to 4 days before adjuvant administration. Daily 30-min sessions were conducted 7 days/week from 21–50 days after adjuvant administration. Data was combined into 6 5-session blocks for analyses and graphic presentation. Data Analyses Data were analyzed with SAS-PC. Analyses of variance were conducted using the procedures for general linear models with
options for repeated measures where appropriate, and Cronbach’s coefficient alpha was computed as the measure of internal reliability (44). Omega squared was computed as a measure of effect size (13,28). Data is presented in the text and in all figures as means { standard errors of the mean (SEMs). RESULTS
Mortality Rates From the beginning to the completion of behavioral testing, 4 rats died in the Arthritis with Morphine Group. Three of the rats never recovered from morphine pellet-implantation surgery. All rats in the Control and Arthritis Only Groups survived to the end of the study. Mortality rates were 0% (0 of 11) for the Control Group, 0% (0 of 10) for the Arthritis Only Group, and 36% (4/11) for the Arthritis with Morphine group. Only the data from the rats that survived to the end of the study was included in the analyses and graphic presentation of the behavioral data, which resulted in groups of the following sizes: Control Group (no arthritis and no morphine, n Å 11), Arthritis Only Group (n Å 10), and Arthritis with Morphine Group (n Å 7). Paw Volume Paw volumes were analyzed in a 3 1 7 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), and Weeks (1–7) as factors in the analysis, treating Weeks as a repeated measure. Substantial inflammation and swelling was evident in the rats treated with adjuvant, becoming evident 2 weeks postinnoculation and peaking 3 weeks postinnoculation. Planned contrasts between the Control rats and the Arthritis Only rats showed that the Arthritis Only paw volumes were larger than those of the Controls, the main effect for Group was statistically significant, F(1,25) Å 70.55, p Å 0.0001; and the Group 1 Trial interaction was statistically significant, F(6,150) Å 15.71, p Å 0.0001. Morphine had no detectable direct effect on the degree of inflammation produced by the adjuvant. Planned contrasts between the Arthritis Only and the Arthritis with Morphine Groups failed to detect any significant dif-
/ eh0f 2677 Mp 201 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
202
CAIN ET AL.
ference in paw volumes between these Groups, the main effect was not statistically significant, F(1,25) Å 2.13, p Å 0.16; and the Group 1 Weeks interaction was not statistically significant, F(6,150) Å 1.97, p Å 0.073 (Fig. 1A). Weight Body weights were analyzed in a 3 1 8 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), and Weeks (1–8) as factors in the analysis, with Weeks treated as a repeated measure. All groups were initially equivalent with respect to body weights. The control rats maintained their weights, and the rats dosed with adjuvant exhibited substantial weight loss until 21 days after adjuvant administration. A planned contrast between the Control and Arthritis Only rats revealed that the Control rats were heavier than the arthritic rats, the main effect for Group was statistically significant, F (1,25) Å 117.27, p Å 0.0001; and the Group 1 Weeks interaction was statisically significant, F (7,175) Å 27.12, p Å 0.0001. Morphine had no detectable effect on body weights over the course of the study. A planned contrast failed to detect any differences between the Arthritis Only and the Arthritis with Morphine Groups, the main effect for Group was not statistically significant, F (1,25) Å 0.64, p Å 0.43; and the Group 1 Weeks interaction was not statistically significant, F (7,175) Å 1.95, p Å 0.06 (Fig. 1B).
F(7,175) Å 9.88, p Å 0.0001. Planned contrasts revealed that the Arthritis Only rats were much less active than the Control rats, the main effect for Group was statistically significant, F(1,25) Å 98.57, p Å 0.0001, as was the Group 1 Time interaction, F(7,175) Å 8.64, p Å 0.0001 (Fig. 2A). Morphine had no detectable effect on activity level, but activity levels were so low that it was not possible to detect any additional decrease in activity level. A planned contrast between the Arthritis Only Group and the Arthritis with Morphine Group failed to detect any effect of morphine on activity levels, the main effect for Group was not statistically significant, F(1,25) Å 0.12, p Å 0.73; and the Group 1 Time interaction was not statistically significant, F (7,175) Å 0.44, p Å 0.88. A planned contrast on activity levels 45 days postinjection revealed that the Arthritis Only rats were still much less active than the Control rats, the main effect for Group was statistically significant, F(1,25) Å 42.51, p Å 0.0001; and the Group 1 Time interaction was statistically significant, F (7,175) Å 4.16, p Å 0.0003 (Fig. 2B). There was no significant difference in activity levels between the Arthritis Only Group and the Arthritis with Morphine Group, the main effect for Group was not statistically significant, F(1,25) Å 1.46, p Å 0.24; and the Group 1 Time interaction was not statistically significant, F (7,175) Å 1.47, p Å 0.18. Rotarod
Activity Levels The total distance traveled during activity-monitoring sessions was analyzed with a 3 1 8 ANOVA that included Groups (Arthritis Only, Arthritis with Morphine and Control), and Blocks (1–8), as factors in the analysis, treating Blocks as a repeated measure. Data from the activity-monitoring session conducted 25 and 45 days postinnoculation were analyzed separately. Rats seemed to habituate to the test apparatus. At 25 days postinjection, activity levels declined over the course of the session, the main effect of Time was statistically significant,
The latencies to fall off the rotarod were analyzed in a 3 1 7 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), and Trials (1–7) as factors in the analysis, with Trial treated as a repeated measure. Arthritic rats were substantially impaired in this test of ambulatory function. A planned contrast revealed that the Arthritis Only rats fell off the rotarod much sooner than the Control rats, the main effect for Group was statistically significant, F (1,25) Å 179.52, p Å 0.0001; and the Group 1 Trial interaction was statistically significant, F (6,150) Å 2.41, p Å 0.03 (Fig. 3A). Planned contrasts between the Ar-
FIG. 2. Activity levels 25 (A) and 45 days (B) postinnoculation for 16 h after being put in the activity chambers. Arthritic rats were almost completely inactive, much less active than the controls, and morphine had no detectable effect on activity level. However, the virtually complete inactivity in the Arthritis Only Group made it impossible to determine if morphine might cause any further decrease in activity level.
/ eh0f 2677 Mp 202 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
PAIN-RELATED DISABILITY
203
FIG. 3. (A) Rotarod test of ambulatory function - latency to fall at 5 rpm. (B) Pellets earned in 5session blocks of delayed nonmatching-to-position operant sessions. Adjuvant-induced arthritis produced severe deficits in this test but, surprisingly, morphine actually attenuated these deficits.
thritis Only and the Arthritis with Morphine rats revealed that morphine improved performance on the rotarod task, and the main effect for Group was statistically significant, F (1,25) Å 7.20, p Å 0.01. The effects of morphine emerged over time, becoming especially apparent as the effects of the adjuvant began to subside, and the Group 1 Trial interaction was statistically significant, F (6,150) Å 3.8, p Å 0.02. Delayed Nonmatching to Position The number of pellets earned were analyzed in a 3 1 6 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), and Block (1–6) as factors in the analysis, with Block treated as a repeated measure. Planned contrasts revealed that the Arthritis Only rats were impaired relative to the Control rats with respect to number of pellets earned, the main effect for Group was statistically significant, F (1,25) Å 36.39, p Å 0.0001. The Arthritis Only rats did improve over the course of the study relative to the Controls, although they never approached the levels of the Controls. The Group 1 Trial interaction was statistically significant, F (5,125) Å 5.70, p Å 0.0001. Morphine improved performance with respect to the number of pellets earned in the DNMTP task and seemed to facilitate recovery from the effects of the adjuvant, improving performance approximately to the level of nonarthritis Controls by the last Block of testing, and the Group 1 Trials interaction was statistically significant, F (5,125) Å 3.94, p Å 0.002 (Fig. 3B). In addition to differences in the number of rewards earned, analyses of the percentage of correct choices suggests differences between the groups in performance with respect to the cognitive demands of the task. The percentage of correct trials at each delay were computed for the first and last 5-session blocks and analyzed in a 3 1 6 1 2 ANOVA including Groups (Arthritis Only, Arthritis with Morphine, and Control), Delay (0, 1, 5, 10, 20, and 30 s), and Blocks (1 & 6) as factors in the analysis, treating Delays and Blocks as repeated measures. Animals generally performed better at the shorter delays and the percent correct declined as the delay duration
increased, and the main effect for Delay was statistically significant, F(5,125) Å 13.83, p Å 0.0001. All groups performed better in the last 5-session block than in the first; the main effect for Block was statistically significant, F(1,25) Å 36.22, p Å 0.0001. Planned contrasts revealed that the Arthritis Only rats performed poorly relative to the Controls; the main effect for Group was statistically significant, F(1,25) Å 16.00, p Å 0.0005. Both Arthritis Only and Arthritis with Morphine Groups improved over the course of the study, relative to the Controls, and the Group 1 Block interaction was statistically significant, F(1,25) Å 9.84, p Å 0.0043 (Fig. 4A, B). Planned contrasts between the Arthritis Only and the Arthritis with Morphine Group also revealed that the Arthritis with Morphine Group performed better on this measure than the Arthritis Only Group; the main effect for Group was statistically significant, F(1,25) Å 6.31, p Å 0.02. DISCUSSION
Because overall functional disability has been identified as perhaps the most important consequence of chronic pain in the clinical setting, considerable attention should be devoted to modeling this aspect of chronic pain syndromes in preclinical studies. The results of the present study confirm that the rodent adjuvant-induced arthritis model of severe chronic pain can be used to model pain-related disability. Consistent with previous reports, spontaneous activity levels and ambulatory function are reduced in arthritic rats and they exhibit substantial weight loss (4,12), presumably due to the fact that it is painful to move, and also possibly because the pain reduces appetite. Novel findings from the present study suggest that the operant delayed nonmatchingto-position task can provide an additional measure of the degree of pain-related disability that may be especially relevant to the effects of chronic pain on performance in a work setting. Originally developed by Dunnett to study cognitive dysfunction (15,17), the delayed nonmatching-to-position operant bar-pressing task is an ‘‘apical’’ test that is sensitive to deficits across a wide range of behavioral functions (30): motor ability, attention,
/ eh0f 2677 Mp 203 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
204
CAIN ET AL.
FIG. 4. Performance across delays in the delayed nonmatching-to-position operant task during the first (A) and last (B) 5-session blocks of trials. These results suggest that chronic severe pain is disruptive even with respect to cognitive functions, and that morphine may improve performance in this test of global behavioral function.
motivation, learning, and memory (49). It is important to note that, although arthritic rats were impaired in this task, they were still able to peform well enough so that additional deficits could have been detected. The potential sensitivity of the delayed nonmatching-to-position task to increased disability is in contrast to the floor effects observed with the measures of spontaneous activity and ambulatory function. One reason that preclinical models of pain-related disability would be valuable is to screen novel treatments. In particular, it would be especially valuable to be able to select novel analgesic agents that reduce disability, in addition to providing direct analgesic activity. Most preclinical studies continue to focus almost exclusively on assessing the potential analgesic effects of novel agents, with little regard for their potential to ultimately improve quality of life. Preclinical research assessing nonanalgesic effects of novel treatments is usually restricted to assessing the side-effect profile in normal healthy animals. Another novel finding reported from the present study is that analgesic treatments that impair functional abilities in normal healthy rats may actually improve the performance of rats exhibiting pain-related disability. Previous work demonstrated that acute morphine injections of only 4 mg/kg impaired performance of rats in the delayed matching-to-position task (31). The results of the present study suggest that chronic morphine attenuates the degree of pain-related disability exhibited by arthritic rats in the test of ambulatory function and the delayed nonmatchingto-position bar-pressing test. The fact that chronic morphine had such long-lasting effects, reducing disability for the duration of this
2-month study in the delayed nonmatching-to-position task, may also seem surprising, in view of the literature demonstrating rapid development of tolerance to chronic morphine. However, the present results are consistent with numerous reports that chronic morphine does not produce tolerance in patients suffering with chronic pain (20,34,38,53,55). In addition, a previous study reported that rats with adjuvant-induced arthritis will self-administer morphine, and that chronic pain reduces the addictive properties of morphine (33). Combined with results from clinical and preclinical studies that subjects experiencing pain become tolerant to morphine more slowly than subjects not experiencing pain (1,7,10,53), the present results suggest that analgesic effects, tolerance, and side-effect profiles of potential analgesic agents should all be assessed in animals experiencing pain, as well as in normal healthy rats. In conclusion, the results of this study indicate that 1. the delayed nonmatching-to-position operant task is a reliable and accurate means of quantifying pain-related disability in the rodent adjuvantinduced arthritis model of severe chronic pain, and 2. constant, moderate levels of systemic morphine actually attenuate pain-related disability associated with adjuvant-induced arthritis, which is in contrast to the deleterious effects morphine injections have on the performance of healthy normal rats in this task. ACKNOWLEDGEMENTS
The authors thank ASTRA AB for funding this research, and Jaci Sagen for generously providing the morphine pellets used in this study.
REFERENCES 1. Backonja, M. M.; Miletic, G.; Miletic, V. The effect of continuous morphine analgesia on chronic thermal hyperalgesia due to sciatic constriction injury in rats. Neurosci. Lett. 196:61–64; 1995. 2. Barnes, D.; Gatchel, R. J.; Mayer, T. G.; Barnett, J. Changes in MMPI profile levels of chronic low back pain patients following successful treatment. J. Spinal. Disord. 3:353–355; 1990.
3. Calvino, B.; Couraud, J. Y.; Besson, J. M. Prevaccination with diluted Freund adjuvant prevents the development of chronic pain and transient release of cerebrospinal fluid substance P in adjuvant-induced arthritis in rats. Pain 58:211–217; 1994. 4. Calvino, B.; Crepon-Bernard, M. O.; Le Bars, D. Parallel clinical and behavioural studies of adjuvant-induced arthritis in the rat: pos-
/ eh0f 2677 Mp 204 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677
PAIN-RELATED DISABILITY
5. 6.
7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.
sible relationship with ’chronic pain’. Behav. Brain Res. 24:11–29; 1987. Cheatle, M. D.; Brady, J. P.; Ruland, T. Chronic low back pain, depression, and attributional style. Clin. J. Pain 6:114–117; 1990. Coda, B. A.; Hill, H. F.; Hunt, E. B.; Kerr, E. B.; Jacobson, R. C.; Chapman, C. R. Cognitive and motor function impairments during continuous opioid analgesic infusions. Hum. Psychopharmacol. Clin. Exp. 8:383–400; 1993. Colpaert, F. C. Long-term suppression of pain by narcotic drugs in the absence of tolerance development. Arch. Int. Pharmacodyn. Ther. 236:293–295; 1978. Colpaert, F. C. Evidence that adjuvant arthritis in the rat is associated with chronic pain. Pain 28:201–222; 1987. Colpaert, F. C.; Meert, T.; De Witte, P.; Schmitt, P. Further evidence validating adjuvant arthritis as an experimental model of chronic pain in the rat. Life Sci. 31:67–75; 1982. Colpaert, F. C.; Niemegeers, C. J.; Janssen, P. A. Nociceptive stimulation prevents development of tolerance to narcotic analgesia. Eur. J. Pharmacol. 49:335–336; 1978. Dardick, S. J.; Basbaum, A. I.; Levine, J. D. The contribution of pain to disability in experimentally induced arthritis. Arthritis Rheum. 29:1017–1022; 1986. De Castro Costa, M.; De Sutter, P.; Gybels, J.; Van Hees, J. Adjuvant-induced arthritis in rats: a possible animal model of chronic pain. Pain 10:173–185; 1981. Dodd, D. H.; Schultz, R. F. Jr. Computational procedures for estimating magnitude of effect for some analysis of variance designs. Psychol. Bull. 79:391–395; 1973. Duggleby, W.; Lander, J. Cognitive status and postoperative pain older adults. J. Pain Symptom Manage. 9:19–27; 1994. Dunnett, S. B. Comparative effects of cholinergic drugs and lesions of nucleus basalis or fimbria-fornix on delayed matching in rats. Psychopharmacology (Berlin) 87:357–363; 1985. Dunnett, S. B. Role of prefrontal cortex and striatal output systems in short-term memory deficits associated with ageing, basal forebrain lesions, and cholinergic-rich grafts. (abst.) Can. J. Psychol. 44:210–232; 1990. Dunnett, S. B.; Evenden, J. L.; Iversen, S. D. Delay-dependent short-term memory deficits in aged rats. Psychopharmacology 96:174–180; 1988. Dworkin, R. H.; Gitlin, M. J. Clinical aspects of depression in chronic pain patients. Clin. J. Pain 7:79–94; 1991. Flecknell, P. A. Refinement of animal use—assessment and alleviation of pain and distress. Lab. Anim. 28:222–231; 1994. Follett, K. A.; Hitchon, P. W.; Piper, J.; Kumar, V.; Clamon, G.; Jones, M. P. Response of intractable pain to continuous intrathecal morphine: a retrospective study. Pain 49:21–25; 1992. Herr, K. A.; Mobily, P. R.; Smith, C. Depression and the experience of chronic back pain: a study of related variables and age differences. Clin. J. Pain 9:104–114; 1993. Hinkley, B. S.; Jaremko, M. E. Effects of pain duration on psychosocial adjustment in orthopedic patients—the importance of early diagnosis and treatment of pain. J. Pain Symptom Manage. 9:175–185; 1994. Hitchcock, L. S.; Ferrell, B. R.; Mccaffery, M. The experience of chronic nonmalignant pain. J. Pain Symptom Manage. 9:312–318; 1994. Jamison, R. N.; Sbrocco, T.; Parris, W. C. The influence of problems with concentration and memory on emotional distress and daily activities in chronic pain patients. Int. J. Psychiatry Med. 18:183–191; 1988. Jensen, M. P.; Strom, S. E.; Turner, J. A.; Romano, J. M. Validity of the sickness impact profile Roland scale as a measure of dysfunction in chronic pain patients. Pain 50:157–162; 1992. Kerns, R. D.; Haythornthwaite, J. A. Depression among chronic pain patients: cognitive-behavioral analysis and effect on rehabilitation outcome. J. Consult. Clin. Psychol. 56:870–876; 1988. Kewman, D. G.; Vaishampayan, N.; Zald, D.; Han, B. Cognitive impairment in musculoskeletal pain patients. Int. J. Psychiatry Med. 21:253–262; 1991. Kirk, R. E. Experimental design: procedures for the behavioral sciences. Pacific Grove, CA: Brooks/Cole Publishing Company; 1982. Latham, J.; Davis, B. D. The socioeconomic impact of chronic pain. Disabil. Rehabil. 16:39–44; 1994. Laties, V. G. Schedule-controlled operant behavior during the hazard identification stage of neurotoxicity risk assessment. In: Weiss, B.; Odonoghue, J. L., eds. Neurobehavioral toxicity: Anaylsis and interpretation. New York: Raven Press; 1994:253–262.
205 31. Lindner, M. D.; Frydel, B.; Kaplan, F. A.; Krueger, P. M.; Bell, W. J.; Blaney, T. J.; Winn, S. R.; Sherman, S. S.; Doherty, E. J.; Emerich, D. F. Intraventricular encapsulated calf adrenal chromaffin cells: viable for at least 500 days in vivo without detectable adverse effects on behavioral/cognitive function or host immune sensitization in rats. Restor. Neurol. Neurosci. 11:21–35; 1997. 32. Linton, S. J. Activities of daily living scale for patients with chronic pain. Percept. Mot. Skills. 71:722; 1990. 33. Lyness, W. H.; Smith, F. L.; Heavner, J. E.; Iacono, C. U.; Garvin, R. D. Morphine self-administration in the rat during adjuvant-induced arthritis. Life Sci. 45:2217–2224; 1989. 34. Meed, S. D.; Kleinman, P. M.; Kantor, T. G.; Blum, R. H.; Savarese, J. J. Management of cancer pain with oral controlled-release morphine sulfate. J. Clin. Pharmacol. 27:155–161; 1987. 35. Meyer, D. R.; Sparber, S. B. A comparison of withdrawal in rats implanted with different types of morphine pellets. Pharmacol. Biochem. Behav. 5:603–607; 1976. 36. Millard, R. W.; Wells, N.; Thebarge, R. W. A comparison of models describing reports of disability associated with chronic pain. Clin. J. Pain 7:283–291; 1991. 37. Pearce, S. A.; Isherwood, S.; Hrouda, D.; Richardson, P. H.; Erskine, A.; Skinner, J. Memory and pain: tests of mood congruity and state dependent learning in experimentally induced and clinical pain. Pain 43:187–193; 1990. 38. Penn, R. D.; Paice, J. A. Chronic intrathecal morphine for intractable pain. J. Neurosurg. 67:182–186; 1987. 39. Pircio, A. W.; Fedele, C. T.; Bierwagen, M. E. A new method for the evaluation of analgesic activity using adjuvant-induced arthritis in the rat. Eur. J. Pharm. 31:207–215; 1975. 40. Riffee, W. H.; Wilcox, R. E.; Anderson, J. A.; McGinity, J. W. Polydimethylsiloxane pellets for sustained delivery of morphine in mice. J. Pharm. Sci. 69:980–982; 1980. 41. Romano, J. M.; Turner, J. A.; Jensen, M. P. The chronic illness problem inventory as a measure of dysfunction in chronic pain patients. Pain 49:71–75; 1992. 42. Roy, R. Many faces of depression in patients with chronic pain. Int. J. Psychiatry Med. 12:109–119; 1982. 43. Rudy, T. E.; Kerns, R. D.; Turk, D. C. Chronic pain and depression: toward a cognitive-behavioral mediation model. Pain 35:129–140; 1988. 44. SAS Institute Inc. SAS/STAT User’s Guide, Version 6. Vol.1. 4th ed. Cary, NC: SAS Institute Inc. 1989. 45. Schaible, H. G.; Grubb, B. D. Afferent and spinal mechanisms of joint pain. Pain 55:5–54; 1993. 46. Schug, S. A.; Zech, D.; Grond, S.; Jung, H.; Meuser, T.; Stobbe, B. A long-term survey of morphine in cancer pain patients. J. Pain Symptom. Manage. 7:259–266; 1992. 47. Shippenberg, T. S.; Stein, C.; Huber, A.; Millan, M. J.; Herz, A. Motivational effects of opioids in an animal model of prolonged inflammatory pain:alteration in the effects of kappa- but not of mureceptor agonists. Pain 35:179–186; 1988. 48. Solomon, R. E.; Wasserman, E. A.; Gebhart, G. F. Tolerance to antinociceptive effects of morphine without tolerance to its effects on schedulecontrolled behavior. Psychopharmacology (Berlin) 92:327–333; 1987. 49. Stanhope, K. J.; Mclenachan, A. P.; Dourish, C. T. Dissociation between cognitive and motor/motivational deficits in the delayed matching to position test: Effects of scopolamine, 8-OH-DPAT and EAA antagonists. Psychopharmacology 122:268–280; 1995. 50. Sullivan, M. J.; Reesor, K.; Mikail, S.; Fisher, R. The treatment of depression in chronic low back pain: review and recommendations. Pain 50:5–13; 1992. 51. Tandon, O. P.; Kumar, S. P3 event related cerebral evoked potential in chronic pain patients. Indian J. Physiol. Pharmacol. 37:51–55; 1993. 52. Tsuneto, S.; Hayashi, A.; Miyazaki, M.; Kashiwagi, T. A clinical survey of controlled-release morphine sulphate for cancer pain relief in a Japanese hospice. Postgrad. Med. J. 67 (Suppl. 2):S79–S81; 1991. 53. Twycross, R. G. Clinical experience with diamorphine in advanced malignant disease. Int. J. Clin. Pharmacol. Ther. Toxicol. 9:184–198; 1974. 54. Webster, B. S.; Snook, S. H. The cost of 1989 workers’ compensation low back pain claims. Spine 19:1111–1116; 1994. 55. Winkelmuller, M.; Winkelmuller, W. Long-term effects of continuous intrathecal opioid treatment in chronic pain of nonmalignant etiology. J. Neurosurg. 85:458–467; 1996. 56. Zenz, M.; Strumpf, M.; Tryba, M. Long-term oral opioid therapy in patients with chronic nonmalignant pain. J. Pain Symptom. Manage. 7:69–77; 1992.
/ eh0f 2677 Mp 205 Friday Jun 13 04:14 PM EL–PB (v. 61, no. 8) 2677