- Email: [email protected]
Ovariohysterectomy in the rat: a model of surgical pain for evaluation of pre-emptive analgesia?
Pain 88 (2000) 79±88
www.elsevier.nl/locate/pain
Ovariohysterectomy in the rat: a model of surgical pain for evaluation of pre-emptive analgesia? Maria Isabel Gonzalez, Mark John Field, Steve Bramwell, Scott McCleary, Lakhbir Singh* Department of Biology, Parke-Davis Neuroscience Research Centre, Cambridge University Forvie Site, Robinson Way, Cambridge CB2 2QB, UK Received 30 September 1999; received in revised form 11 February 2000; accepted 12 April 2000
Abstract Ovariohysterectomy in the rat led to the induction of abdominal postures and referred mechanical allodynia in the hind paws. The latter was differentiated into static and dynamic subtypes. The abdominal postures were present up to 4±5 h, whilst the two types of allodynia lasted for at least 2 days. A single administration of morphine 30 min before surgery dose-dependently (0.1±3 mg/kg, s.c.) blocked the development of abdominal postures and the two types of mechanical allodynia. The highest dose of morphine almost completely blocked these responses. The duration of action of 3 mg/kg morphine was short and similar (1.5±2 h) when administered either before or after surgery. However, multiple administrations of morphine (0.5 h before, and 0.5 and 2 h after surgery) blocked the development of abdominal postures and both allodynias for up to 2 days. In contrast, administration of three doses of morphine (3 mg/kg) in a similar dosing regime but starting 24 h after surgery, only blocked the two types of allodynia for 4 h. These data indicate the importance of blocking the induction phase of surgical pain and support the concept of pre-emptive analgesia. It is suggested that the ovariohysterectomy model should prove to be useful for studying mechanisms and designing novel therapeutic strategies for the treatment of post-operative pain. q 2000 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Induction; Maintenance; Visceral; Static allodynia; Dynamic allodynia; Pre-emptive; Morphine
1. Introduction It is inevitable that surgical interventions result in postoperative pain. The certainty of post-operative pain provides an exceptional opportunity for prevention, particularly following elective surgery (Wall, 1988). In pre-clinical studies, it has been shown that peripheral injury can induce an increased state of excitability in the spinal cord (Coderre et al., 1993). This sensitization of dorsal horn neurones is widely thought to contribute to abnormal pain sensitivity. It has been suggested that pre-surgical analgesic treatment would prevent the development of central sensitization and thus produce long lasting relief from pain (Woolf and Chong, 1993). This has important clinical implications, as post-operative pain is one situation where pre-emptive treatment can be utilized. In the clinic, opioids, NSAIDs and local anaesthetics are used as pre-emptive agents with mixed results (Richmond et al., 1993; Wilson et al., 1994; Katz et al., 1996; Grif®n et al., 1997; Richards et al., 1998). The concept of pre-emptive analgesia was put forward by * Corresponding author. Tel.: 144-1223-210-929; fax: 144-1223-249106. E-mail address: [email protected] (L. Singh).
Wall (1988). Since then very little progress has been made in this area. One of the main reasons for this is the lack of good animal models that closely mimic the clinical situation following surgical injury. The development of relevant animal models is an important step for the understanding of mechanisms involved in post-surgical pain and for designing therapeutic strategies. Recently, a rat model of post-operative pain involving incision of the plantaris muscle was described. It has been shown that such an incision leads to the induction of thermal hyperalgesia and static allodynia lasting several days (Brennan et al., 1996; Field et al., 1997). This model is proving useful for studying mechanisms and drug effects on induction and maintenance of pain processing. However, it may not accurately re¯ect the major proportion of elective surgery that involves operative invasion of the body cavities, including abdominal and thoracic surgical procedures. In such cases, there is signi®cant activation and sensitization of visceral afferents, giving rise to inputs that are processed differently to those of somatic sensory neurones (Cervero, 1988). Recently, it has been reported that ovariohysterectomy in rats and dogs (Lascelles et al., 1995, 1997) induces referred hyperalgesia. However, allodynia constitutes a greater post-operative problem than hyperalgesia
0304-3959/00/$20.00 q 2000 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. PII: S 0304-395 9(00)00309-2
80
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
because relevant stimuli (e.g. clothes touching skin, breathing, coughing, movement of joints) are almost unavoidable following surgery. In the present study, we have further characterized the behavioural responses shown by female rats following ovariohysterectomy. Here, we report that this surgical procedure leads to abdominal postures and two distinct types of referred mechanical allodynia (termed static and dynamic) in the rat hind paws. Furthermore, this study also examines the effect of morphine on induction and maintenance of these responses. 2. Methods 2.1. Animals Female Sprague±Dawley rats (175±250 g), obtained from Charles River (UK) were housed in groups of six under a 12:12-h light/dark cycle (lights on at 07:00 h) with food and water ad libitum. All experiments were carried out by an observer blind to drug treatments as well as to the surgical procedure (skin, abdominal wall, full ovariectomy). All procedures were performed according to the Home Of®ce Animals Scienti®c Procedures Act 1986. 2.2. Surgery The surgery was based on the procedure described by Lascelles et al. (1995). Animals were anaesthetized with iso¯urane (5% for induction, 2% for maintenance of anaesthesia) and 1:4 O2/N2O mixture. Ovariohysterectomy was performed via a midline abdominal incision (2 cm in length) in the linea alba. The ovarian ligaments and cervix were ligated with 5-0 silk, using a single-clamp technique. The ovaries and the uterus were then removed. Four simple interrupted sutures were placed in the abdominal wall. A continuous subcuticular/fascial layer was placed. The skin was closed with four wound clips and covered with topical antibiotics. The procedure was completed within 15 min. In order to study the contribution of different parts of the procedure to the development of nociception, three other groups were evaluated in parallel. A ®rst group was submitted only to the anaesthesia for 15 min; a second group received incision of the skin only; and a third group received incision of the skin and the abdominal wall. 2.3. Assessment of abdominal postures A preliminary study was carried out to address the effect of oestrous cycle on abdominal postures following ovariohysterectomy. No signi®cant differences were found between the phases of the cycle (number of abdominal postures (mean ^ SEM) in: proestrous, 52 ^ 7.3; oestrous, 66 ^ 7.6; dioestrous, 70 ^ 8.2; n 8 per group). Therefore, the phase of the oestrous cycle was not controlled before ovariohysterectomy. Immediately after surgery animals were placed in indivi-
dual plexiglass cages with wire mesh bottoms and abdominal postures were recorded in 30-min time bins. Postures scored were hump-backed position, contraction of the muscles of the abdomen associated with inward movements of the hindlimb, stretching of the body and squashing of the lower abdomen against the ¯oor (Giamberdino et al., 1995). Each of these behaviours was scored as one posture. Duration of episodes was not included. 2.4. Assessment of static allodynia Static allodynia was measured using Semmes±Weinstein (Stoelting, IL) von Frey hairs. Animals were habituated to wire mesh bottom cages prior to the start of the experiment (i.e. surgery). Static allodynia was tested by touching the plantar surface of the animals hind paw with von Frey hairs in ascending order of force (0.7, 1.2, 1.5, 2, 3.6, 5.5, 8.5, 11.8, 15.1 and 29 g) for 6 s or until a paw withdrawal response was elicited. The lowest amount of force required to elicit a response was recorded as withdrawal threshold in grams. The highest force of 29 g lifted the paw as well as eliciting a response, and thus represented the cut-off point. Each animal had both hind paws tested in this manner. There was no signi®cant difference in the response of the right or the left hind paw. Therefore, for clarity results are shown only for the right paw. 2.5. Assessment of dynamic allodynia Dynamic allodynia was assessed as previously described (Field et al., 1999a,b). Brie¯y, the plantar surface of both hind paws was lightly stroked with a cotton bud. Latency to paw withdrawal was noted. It was often accompanied with repeated ¯inching or licking of the paw. Two to three measurements were taken at each time point. If no paw withdrawal was exhibited within 15 s the procedure was terminated and animals were assigned this withdrawal time. Thus, 15 s effectively represents no withdrawal response. 2.6. Drugs Morphine sulphate was obtained from Savory and Moore (Cambridge, UK). It was dissolved in 0.9% w/v NaCl (isotonic saline) and administered subcutaneously in a dosing volume of 1 ml/kg. The dosing regimes were: (a) single dose at 30 min before surgery (pre-operative); (b) single dose at 30 min after surgery (post-operative); (c) three total administrations at 30 min before surgery and followed by 30 min and 2 h after surgery (multiple peri-operative); and (d) three total administrations at 24.5, 25.5 and 27 h after surgery (multiple post-operative). Each animal was assessed for all three parameters (abdominal postures, static and dynamic allodynia), except in the single pre-operative dose experiment in which separate
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
groups of animals were used to assess the duration of abdominal postures and static and dynamic allodynias. 2.7. Statistics Data obtained for static allodynia were subjected to a Kruskall±Wallis test followed by Dunn's test. Data obtained for visceral postures and dynamic allodynia were subjected to a one-way-analysis of variance (ANOVA) followed by Dunnett's t-test when more than two experimental groups were compared or to an unpaired t-test when there were two groups in the experiment. 3. Results 3.1. Induction of abdominal postures, static and dynamic allodynias by ovariohysterectomy Following anaesthesia or incision of the skin no abdominal postures were observed. A non-signi®cant number of postures were observed following incision of the skin and abdominal wall. Animals which had undergone full ovariohysterectomy displayed a signi®cant number of abdominal postures lasting over 4 h (Fig. 1a). These animals also exhibited both static and dynamic types of referred allodynia. However, unlike the abdominal postures, the two types of allodynia were present for at least 2 days post-surgery (Fig. 1b,c). Both static and dynamic allodynias were evident in both hind paws, with no signi®cant difference between paws (data not shown). The skin and muscle incision also induced dynamic allodynia of short duration (Fig. 1c). 3.2. Effect of a single dose of morphine administered before surgery A single administration of morphine administered 30 min before ovariohysterectomy dose-dependently (0.1±3 mg/kg, s.c.) prevented the development of abdominal postures, with a minimum effective dose (MED) of 0.3 mg/kg (Fig. 2a). The dose of 3 mg/kg produced almost total blockade of these behaviours which was maintained for up to 1.5 h (Fig. 2). The pre-emptive administration of morphine also dose-dependently (0.1±3 mg/kg, s.c.) blocked the induction of static and dynamic allodynia, with MEDs of 3 mg/kg. As in the case of abdominal postures, duration of the block was short (Fig. 2b,c). 3.3. Effect of a single post-surgical administration of morphine A single dose of morphine (3 mg/kg, s.c.) administered 30 min after ovariohysterectomy signi®cantly blocked abdominal postures, and both types of mechanical allodynias. As in the case of pre-surgical administration of morphine, the duration of these blockades was short-lasting (Fig. 3).
81
3.4. Effect of multiple peri-operative administration of morphine Morphine (3 mg/kg, s.c.) administered 30 min before surgery followed by similar administrations at 0.5 and 2 h after ovariohysterectomy completely prevented the development of abdominal postures and the two referred allodynias during the 2 days of the study (Fig. 4). 3.5. Effect of multiple post-operative administration of morphine Animals were assessed for visceral postures and, static and dynamic allodynia immediately following ovariohysterectomy and also at 24 h after surgery before any drug treatment. Unlike immediately after ovariohysterectomy, visceral postures were not apparent 24 h post-surgery, while both allodynias were present (Fig. 5). The multiple administration of morphine (3 mg/kg, s.c.) at 24.5, 25.5 and 27 h after surgery signi®cantly blocked both referred static and dynamic allodynia. However, unlike peri-operative dosing the effects disappeared within 2 h after the third administration of morphine (Fig. 5). 4. Discussion The results presented here show that ovariohysterectomy in the rat induces abdominal postures and longer lasting referred static and dynamic allodynia in the hind paws. This is consistent with pre-clinical studies showing that in¯ammation of the bladder gives rise to hyperalgesia in the hind paws (Jaggar et al., 1998). It is also consistent with clinical studies showing that referred hypersensitivity often outlasts visceral pain after surgery. The reason for referred pain is likely to be the convergence of relevant somatic and visceral nerves at the level of the spinal cord. Several sensory nerves innervate the female internal reproductive organs and enter the dorsal root ganglia at various segments of the spinal cord. It is likely that distension, ligation and incision of the cervix contribute to the central hypersensitivity. However, with respect to the referred allodynia observed in the present study, it is important to note that the pelvic nerve which carries afferent information from the cervix and vaginal canal terminates between segments L6±S1 (Berkeley and Hubscher, 1995). Some of these spinal cord segments (in particular the L6) are shared by the sciatic nerve which innervates the hind paws in rats. Thus, this particular convergence of visceral and somatic sensory afferents may be responsible for the referred allodynia described in the present study. Recently it has been shown that static and dynamic components of mechanical allodynia can be detected in pre-clinical and clinical studies (Koltzenburg et al., 1992; Field et al., 1999b). Moreover, both types of allodynia have also been described in the rat plantaris incisional model of surgical pain (Field et al., 1999a). It has been suggested that
82
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
Fig. 1. Duration of (a) abdominal postures, (b) static and (c) dynamic allodynia in the rat following anaesthesia for 15 min, skin incision, skin and abdominal wall incision or ovariohysterectomy. Baseline paw withdrawal latencies (PWL) to a cotton-bud stimulus and paw withdrawal thresholds (PWT) to von Frey hairs were determined in rats before surgery (S). Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited in 0.5-h time bins and results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM). **P , 0:01, signi®cantly different from anaesthesia group (ANOVA followed by Dunnett's t-test). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles). *P , 0:05, **P , 0:01, signi®cantly different (Dunn's test) from anaesthesia group values (n 6±8 per group).
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
the two types of allodynia are signalled by distinct sensory neurones. Thus, whilst the dynamic type is mediated by large-diameter Ab -/capsaicin-insensitive ®bres, static allo-
83
dynia appears to be exclusively signalled by Ad primary sensory neurones (Ochoa and Yarnitsky, 1993). However, it remains to be seen which types of ®bres
Fig. 2. Effect of morphine on development of (a,b) abdominal postures, (c) static and (d) dynamic allodynia in the rat ovariohysterectomy model. Morphine (M, 0.3±3 mg/kg, s.c.) was administered 0.5 h before surgery (S). Baseline paw withdrawal latencies (PWL) to a cotton bud stimulus and paw withdrawal thresholds (PWT) to von Frey hairs were determined in rats before surgery. Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited during the ®rst 30 min after surgery (a) and in 0.5-h time bins (b). Results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM) *P , 0:05, **P , 0:01, signi®cantly different (ANOVA followed by Dunnett's t-test, or unpaired t-test for (b)). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles) *P , 0:05, signi®cantly different (Dunn's test). All comparisons were carried out between drug and vehicle treated groups at each time point (n 6±15 per group).
84
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
signal nociceptive input after incision or abdominal surgery. The results of the present study show that development of long lasting abdominal postures and the two types of allo-
dynia seem to depend on the dissection of the ovaries and uterus. However, short-lasting static allodynia was also evident after incision of the muscle. There is strong
Fig. 3. Effect of morphine on the maintenance of (a) abdominal postures, (b) static and (c) dynamic allodynia in the rat ovariohysterectomy model. Morphine (M, 3 mg/kg, s.c.) was administered 0.5 h after surgery (S). Baseline paw withdrawal latencies (PWL) to a cotton bud stimulus and paw withdrawal thresholds (PWT) to von Frey hair were determined in rats before surgery. Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited in 0.5-h time bins and results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM). *P , 0:05, ***P , 0:01, signi®cantly different (unpaired t-test). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles). *P , 0:05, ***P , 0:01, signi®cantly different (Mann±Whitney U-test). All comparisons were carried out between drug and vehicle-treated groups (n 8 per group).
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
evidence suggesting that sensitization of spinal cord neurones are responsible for the induction and maintenance of allodynia. This is particularly true in the case of referred
85
allodynia. This would suggest that muscle injury-induced activation of nociceptive neurones is suf®cient to induce sensitization of super®cial layers of the dorsal horn. This
Fig. 4. Effect of multiple peri-operative administration of morphine on the induction of (a) abdominal postures, (b) static and (c) dynamic allodynia in the rat ovariohysterectomy model. Morphine (M, 3 mg/kg, s.c.) was administered 0.5 h before and at 0.5 and 2 h after surgery (S). Baseline paw withdrawal latencies (PWL) to a cotton bud stimulus and paw withdrawal thresholds (PWT) to von Frey hairs were determined in rats before surgery. Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited in 0.5-h time bins and results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM). *P , 0:05, **P , 0:01, ***P , 0:001, signi®cantly different (unpaired t-test). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles). *P , 0:05, **P , 0:01, ***P , 0:001, signi®cantly different (Mann±Whitney U-test). All comparisons were carried out between drug-treated group and saline group at each time point (n 7 per group).
86
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
is the region where Ad -®bres, which are thought to be involved in static allodynia, terminate. The failure of muscle
incision to induce dynamic allodynia suggests that this sensitization induced by somatic nociceptive ®bres does
Fig. 5. Effect of multiple post-operative administrations of morphine on the maintenance of (a) abdominal postures, (b) static and (c) dynamic allodynia in the rat ovariohysterectomy model. Abdominal postures were recorded for 30 min immediately after and 24 h post-surgery. Morphine (M, 3 mg/kg, s.c.) was administered 24.5, 25.5 and 27 h after surgery (S). Baseline paw withdrawal latencies (PWL) to a cotton bud stimulus and paw withdrawal thresholds (PWT) to von Frey hairs were determined in rats before surgery. Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited in 0.5-h time bins and results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM). ***P , 0:001, signi®cantly different (unpaired t-test). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles). ***P , 0:001, signi®cantly different (Mann±Whitney U-test). All comparisons were carried out between drug-treated group and saline group at each time point (n 8 per group).
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
not spread from the super®cial to deeper layers where the Ab -®bres that signal dynamic allodynia terminate. The development of any pain syndrome can be divided into induction and maintenance phases. Further studies have shown that distinct mechanisms are involved during the two phases (Woolf, 1994). The present study addressed the effect of morphine on each of these phases. There was no difference between a single dose of morphine administered either before or after surgery. In each case, morphine was effective at blocking abdominal postures and both static and dynamic allodynia but the duration of action was short (2 h) in each case. However, marked differences were obvious following multiple dosing with morphine depending on the time of administration. It is likely that abdominal postures represent intense activation of visceral sensory neurones. Such an activation of primary afferents would lead to sensitization of dorsal horn neurones of the spinal cord and is consistent with the presence of referred allodynia. The abdominal postures last at least 4 h, suggesting that after this time ®ring of visceral nerves subsides to a level that is suf®cient to maintain but not induce referred allodynia. This would explain why administration of morphine designed to block abdominal postures led to complete prevention of allodynia. In contrast, similar repeated administration of morphine starting 24 h after surgery had a short duration of action. These data suggest that blockade of the induction phase is of paramount importance for preventing surgical pain. Previously, it has been reported that morphine can block static but not dynamic allodynia in the plantaris incisional model of surgical pain (Field et al., 1999a). This is also the case in models of neuropathic pain (Field et al., 1999b). Therefore, it was somewhat surprising to observe that morphine was effective against both types of allodynic responses in the present study. The reason for these apparent differences may be due to the type of sensory neurone responsible for central sensitization. Thus, in neuropathic pain and the plantaris incisional model any of the three major sensory ®bres may be responsible for this process. It is known that majority of ®bres innervating the viscera is of the C-type and furthermore unlike the somatic sensory neurones, they appear to terminate in deeper layers of the spinal cord (De Groat, 1986). Therefore, these nociceptive ®bres may predominantly if not exclusively be responsible for the central sensitization process in the ovariohysterectomy model. These pro®les of morphine are consistent with electrophysiological studies showing that it can block small(C- and Ad -) but not large-diameter (Ab -) ®bre evoked responses into the dorsal horn (Dickenson and Sullivan, 1986). However, unlike the effect of morphine on the induction of dynamic allodynia, it is dif®cult to understand why it was able to block the maintenance of this response. One possible explanation is that there is a pre-synaptic link between visceral C-®bres and somatic Ab -sensory neurones (Cervero and Laird, 1996). A second is that a sustained input from visceral sensory ®bres is necessary for the mainte-
87
nance of referred dynamic allodynia. In each case blockade of C-®bre activity would explain the ability of morphine to block maintenance of dynamic allodynia. However, further evidence is required to support these hypotheses. Previous studies involving the plantaris incisional model taken together with present data indicate that the duration of intense activation of primary afferents and therefore of the induction phase is dependent on the surgical procedure. Thus, in the plantaris incisional model the induction phase was complete within 1±2 h, whereas following ovariohysterectomy it was signi®cantly longer taking over 6 h. These observations suggest that different regimens of analgesia administrations will be required depending on the surgical procedure for complete prevention of post-operative pain. This conclusion is supported by a single dose of morphine administered before surgery being able to completely prevent thermal hyperalgesia in the plantaris incisional model (Field et al., 1997) whereas, in the ovariohysterectomy model, multiple dosing of morphine was required to cover the long induction phase for complete prevention of referred allodynia. Previously, animal studies have also shown that some analgesic treatments may not be able to prevent or block all types of pain responses (Field et al., 1999a,b). Taken together, these ®ndings may help to better design future clinical studies regarding pre-emptive analgesia and post-operative pain. Firstly, it is important to have the drug on board before surgery and maintained for at least the entire induction phase. Thus, a single pre-emptive administration may not be suf®cient if the induction phase is long or that the duration of action of the drug is short. Secondly, consideration should also be given to the assessment of pain. In addition to using visual analogue scale as a measure of overall pain relief, additional parameters (e.g. hyperalgesia, allodynia, time to analgesia request and total analgesic consumption) should also be determined to give a complete pro®le of analgesia. These issues have not been standardized in the past and may re¯ect con¯icting results seen in published clinical studies. In conclusion, data presented here support the concept of pre-emptive analgesia. Furthermore, the ovariohysterectomy model should prove to be useful in studying mechanisms and designing novel therapeutic strategies for the treatment of post-operative pain.
References Berkeley KJ, Hubscher CH. Visceral and somatic sensory tracks through the neuraxis and their relation to pain: lessons from the rat female reproductive system. In: Gebhart GF, editor. Visceral pain, Progress in pain research and management, 5. Seattle: IASP Press, 1995. pp. 195±216. Brennan TJ, Vandermeulen EP, Gebhart GF. Characterisation of a rat model of incisional pain. Pain 1996;64:493±501. Cervero F. Visceral pain. In: Dubner R, Gebhart GF, Bond MR, editors. Proceedings of the Vth world congress on pain, Amsterdam: Elsevier, 1988. pp. 216±226.
88
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
Cervero F, Laird MA. Mechanisms of touch-evoked pain (allodynia): a new model. Pain 1996;68:13±23. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993;52:259±285. De Groat WC. Spinal cord projections and neuropeptides in visceral afferent neurones. In: Cervero F, Morrison JFB, editors. Visceral sensation, Progress in Brain Research, 67. Amsterdam: Elsevier, 1986. pp. 165± 167. Dickenson AH, Sullivan AF. Electrophysiological studies on the effects of intrathecal morphine on nociceptive neurones in the rat dorsal horn. Pain 1986;24:211±222. Field MJ, Holloman EF, McCleary S, Hughes J, Singh L. Evaluation of gabapentin and S-(1)-3-isobutylgaba in a rat model of postoperative pain. J Pharmacol Exp Ther 1997;282:1242±1246. Field MJ, Carnell AJ, Gonzalez MI, McCleary S, Oles RJ, Smith R, Hughes J, Singh L. Enadoline, a selective k -opioid receptor agonist shows potent antihyperalgesic and antiallodynic actions in a rat model of surgical pain. Pain 1999a;80:383±389. Field MJ, McCleary S, Hughes J, Singh L. Gabapentin and pregabalin, but not morphine and amitriptyline, block both static and dynamic components of mechanical allodynia induced by streptozocin in the rat. Pain 1999b;80:391±398. Giamberdino MA, Valente R, de Bigontina P, Vecchiet L. Arti®cial ureteral calculosis in rats: behavioral characterization of visceral pain episodes and their relationship with referred lumbar muscle hyperalgesia. Pain 1995;61:459±469. Grif®n MJ, Hughes D, Knaggs A, Donnelly MB, Boylan JF. Late-onset preemptive analgesia associated with preincisional large-dose alfentanil. Anesth Analg 1997;85:1317±1321. Jaggar SI, Hasnie FS, Sellaturay S, Rice AS. The anti-hyperalgesic actions
of the cannabinoid anadamide and the putative CB2 receptor agonist palmitoylethanolamide in visceral and somatic in¯ammatory pain. Pain 1998;76:189±199. Katz J, Clairoux M, Redahan C, Kavanagh BP, Carroll S, Nierenberg H, Jackson M, Beattie J, Taddio A, Sandler N. High dose alfentanil preempts pain after abdominal hysterectomy. Pain 1996;68:109±118. Koltzenburg M, Lundberg LER, TorebjoÈrk HE. Dynamic and static components of mechanical hyperalgesia in human hairy skin. Pain 1992;51:207±219. Lascelles BDX, Waterman AE, Cripps PJ, Livingston A, Henderson G. Central sensitization as a result of surgical pain: investigation of the pre-emptive value of pethidine for ovariohysterectomy in the rat. Pain 1995;62:210±212. Lascelles BDX, Cripps PJ, Jones A, Waterman AE. Post-operative central hypersensitivity and pain: the pre-emptive value of pethidine for ovariohysterectomy. Pain 1997;73:461±471. Ochoa JL, Yarnitsky D. Mechanical hyperalgesia in neuropathic pain patients; dynamic and static subtypes. Ann Neurol 1993;33:465±472. Richards JT, Read JRM, Chambers WA. Epidural anaesthesia as a method of pre-emptive analgesia for abdominal hysterectomy. Anaesthesia 1998;53:296±307. Richmond CE, Bromley LM, Woolf CJ. Preoperative morphine pre-empts postoperative pain. Lancet 1993;342:73±75. Wall PD. The prevention of postoperative pain. Pain 1988;33:289±290. Wilson RJT, Leith S, Jackson IJB, Hunter D. Pre-emptive analgesia from intravenous administration of opioids. Anaesthesia 1994;49:591±593. Woolf CJ. A new strategy for the treatment of in¯ammatory pain. Prevention or elimination of central sensitization. Drugs 1994;47:1±9. Woolf CJ, Chong M-S. Preemptive analgesia-treating postoperative pain by preventing the establishment of central sensitization. Anesth Analg 1993;77:362±379.
www.elsevier.nl/locate/pain
Ovariohysterectomy in the rat: a model of surgical pain for evaluation of pre-emptive analgesia? Maria Isabel Gonzalez, Mark John Field, Steve Bramwell, Scott McCleary, Lakhbir Singh* Department of Biology, Parke-Davis Neuroscience Research Centre, Cambridge University Forvie Site, Robinson Way, Cambridge CB2 2QB, UK Received 30 September 1999; received in revised form 11 February 2000; accepted 12 April 2000
Abstract Ovariohysterectomy in the rat led to the induction of abdominal postures and referred mechanical allodynia in the hind paws. The latter was differentiated into static and dynamic subtypes. The abdominal postures were present up to 4±5 h, whilst the two types of allodynia lasted for at least 2 days. A single administration of morphine 30 min before surgery dose-dependently (0.1±3 mg/kg, s.c.) blocked the development of abdominal postures and the two types of mechanical allodynia. The highest dose of morphine almost completely blocked these responses. The duration of action of 3 mg/kg morphine was short and similar (1.5±2 h) when administered either before or after surgery. However, multiple administrations of morphine (0.5 h before, and 0.5 and 2 h after surgery) blocked the development of abdominal postures and both allodynias for up to 2 days. In contrast, administration of three doses of morphine (3 mg/kg) in a similar dosing regime but starting 24 h after surgery, only blocked the two types of allodynia for 4 h. These data indicate the importance of blocking the induction phase of surgical pain and support the concept of pre-emptive analgesia. It is suggested that the ovariohysterectomy model should prove to be useful for studying mechanisms and designing novel therapeutic strategies for the treatment of post-operative pain. q 2000 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Induction; Maintenance; Visceral; Static allodynia; Dynamic allodynia; Pre-emptive; Morphine
1. Introduction It is inevitable that surgical interventions result in postoperative pain. The certainty of post-operative pain provides an exceptional opportunity for prevention, particularly following elective surgery (Wall, 1988). In pre-clinical studies, it has been shown that peripheral injury can induce an increased state of excitability in the spinal cord (Coderre et al., 1993). This sensitization of dorsal horn neurones is widely thought to contribute to abnormal pain sensitivity. It has been suggested that pre-surgical analgesic treatment would prevent the development of central sensitization and thus produce long lasting relief from pain (Woolf and Chong, 1993). This has important clinical implications, as post-operative pain is one situation where pre-emptive treatment can be utilized. In the clinic, opioids, NSAIDs and local anaesthetics are used as pre-emptive agents with mixed results (Richmond et al., 1993; Wilson et al., 1994; Katz et al., 1996; Grif®n et al., 1997; Richards et al., 1998). The concept of pre-emptive analgesia was put forward by * Corresponding author. Tel.: 144-1223-210-929; fax: 144-1223-249106. E-mail address: [email protected] (L. Singh).
Wall (1988). Since then very little progress has been made in this area. One of the main reasons for this is the lack of good animal models that closely mimic the clinical situation following surgical injury. The development of relevant animal models is an important step for the understanding of mechanisms involved in post-surgical pain and for designing therapeutic strategies. Recently, a rat model of post-operative pain involving incision of the plantaris muscle was described. It has been shown that such an incision leads to the induction of thermal hyperalgesia and static allodynia lasting several days (Brennan et al., 1996; Field et al., 1997). This model is proving useful for studying mechanisms and drug effects on induction and maintenance of pain processing. However, it may not accurately re¯ect the major proportion of elective surgery that involves operative invasion of the body cavities, including abdominal and thoracic surgical procedures. In such cases, there is signi®cant activation and sensitization of visceral afferents, giving rise to inputs that are processed differently to those of somatic sensory neurones (Cervero, 1988). Recently, it has been reported that ovariohysterectomy in rats and dogs (Lascelles et al., 1995, 1997) induces referred hyperalgesia. However, allodynia constitutes a greater post-operative problem than hyperalgesia
0304-3959/00/$20.00 q 2000 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. PII: S 0304-395 9(00)00309-2
80
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
because relevant stimuli (e.g. clothes touching skin, breathing, coughing, movement of joints) are almost unavoidable following surgery. In the present study, we have further characterized the behavioural responses shown by female rats following ovariohysterectomy. Here, we report that this surgical procedure leads to abdominal postures and two distinct types of referred mechanical allodynia (termed static and dynamic) in the rat hind paws. Furthermore, this study also examines the effect of morphine on induction and maintenance of these responses. 2. Methods 2.1. Animals Female Sprague±Dawley rats (175±250 g), obtained from Charles River (UK) were housed in groups of six under a 12:12-h light/dark cycle (lights on at 07:00 h) with food and water ad libitum. All experiments were carried out by an observer blind to drug treatments as well as to the surgical procedure (skin, abdominal wall, full ovariectomy). All procedures were performed according to the Home Of®ce Animals Scienti®c Procedures Act 1986. 2.2. Surgery The surgery was based on the procedure described by Lascelles et al. (1995). Animals were anaesthetized with iso¯urane (5% for induction, 2% for maintenance of anaesthesia) and 1:4 O2/N2O mixture. Ovariohysterectomy was performed via a midline abdominal incision (2 cm in length) in the linea alba. The ovarian ligaments and cervix were ligated with 5-0 silk, using a single-clamp technique. The ovaries and the uterus were then removed. Four simple interrupted sutures were placed in the abdominal wall. A continuous subcuticular/fascial layer was placed. The skin was closed with four wound clips and covered with topical antibiotics. The procedure was completed within 15 min. In order to study the contribution of different parts of the procedure to the development of nociception, three other groups were evaluated in parallel. A ®rst group was submitted only to the anaesthesia for 15 min; a second group received incision of the skin only; and a third group received incision of the skin and the abdominal wall. 2.3. Assessment of abdominal postures A preliminary study was carried out to address the effect of oestrous cycle on abdominal postures following ovariohysterectomy. No signi®cant differences were found between the phases of the cycle (number of abdominal postures (mean ^ SEM) in: proestrous, 52 ^ 7.3; oestrous, 66 ^ 7.6; dioestrous, 70 ^ 8.2; n 8 per group). Therefore, the phase of the oestrous cycle was not controlled before ovariohysterectomy. Immediately after surgery animals were placed in indivi-
dual plexiglass cages with wire mesh bottoms and abdominal postures were recorded in 30-min time bins. Postures scored were hump-backed position, contraction of the muscles of the abdomen associated with inward movements of the hindlimb, stretching of the body and squashing of the lower abdomen against the ¯oor (Giamberdino et al., 1995). Each of these behaviours was scored as one posture. Duration of episodes was not included. 2.4. Assessment of static allodynia Static allodynia was measured using Semmes±Weinstein (Stoelting, IL) von Frey hairs. Animals were habituated to wire mesh bottom cages prior to the start of the experiment (i.e. surgery). Static allodynia was tested by touching the plantar surface of the animals hind paw with von Frey hairs in ascending order of force (0.7, 1.2, 1.5, 2, 3.6, 5.5, 8.5, 11.8, 15.1 and 29 g) for 6 s or until a paw withdrawal response was elicited. The lowest amount of force required to elicit a response was recorded as withdrawal threshold in grams. The highest force of 29 g lifted the paw as well as eliciting a response, and thus represented the cut-off point. Each animal had both hind paws tested in this manner. There was no signi®cant difference in the response of the right or the left hind paw. Therefore, for clarity results are shown only for the right paw. 2.5. Assessment of dynamic allodynia Dynamic allodynia was assessed as previously described (Field et al., 1999a,b). Brie¯y, the plantar surface of both hind paws was lightly stroked with a cotton bud. Latency to paw withdrawal was noted. It was often accompanied with repeated ¯inching or licking of the paw. Two to three measurements were taken at each time point. If no paw withdrawal was exhibited within 15 s the procedure was terminated and animals were assigned this withdrawal time. Thus, 15 s effectively represents no withdrawal response. 2.6. Drugs Morphine sulphate was obtained from Savory and Moore (Cambridge, UK). It was dissolved in 0.9% w/v NaCl (isotonic saline) and administered subcutaneously in a dosing volume of 1 ml/kg. The dosing regimes were: (a) single dose at 30 min before surgery (pre-operative); (b) single dose at 30 min after surgery (post-operative); (c) three total administrations at 30 min before surgery and followed by 30 min and 2 h after surgery (multiple peri-operative); and (d) three total administrations at 24.5, 25.5 and 27 h after surgery (multiple post-operative). Each animal was assessed for all three parameters (abdominal postures, static and dynamic allodynia), except in the single pre-operative dose experiment in which separate
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
groups of animals were used to assess the duration of abdominal postures and static and dynamic allodynias. 2.7. Statistics Data obtained for static allodynia were subjected to a Kruskall±Wallis test followed by Dunn's test. Data obtained for visceral postures and dynamic allodynia were subjected to a one-way-analysis of variance (ANOVA) followed by Dunnett's t-test when more than two experimental groups were compared or to an unpaired t-test when there were two groups in the experiment. 3. Results 3.1. Induction of abdominal postures, static and dynamic allodynias by ovariohysterectomy Following anaesthesia or incision of the skin no abdominal postures were observed. A non-signi®cant number of postures were observed following incision of the skin and abdominal wall. Animals which had undergone full ovariohysterectomy displayed a signi®cant number of abdominal postures lasting over 4 h (Fig. 1a). These animals also exhibited both static and dynamic types of referred allodynia. However, unlike the abdominal postures, the two types of allodynia were present for at least 2 days post-surgery (Fig. 1b,c). Both static and dynamic allodynias were evident in both hind paws, with no signi®cant difference between paws (data not shown). The skin and muscle incision also induced dynamic allodynia of short duration (Fig. 1c). 3.2. Effect of a single dose of morphine administered before surgery A single administration of morphine administered 30 min before ovariohysterectomy dose-dependently (0.1±3 mg/kg, s.c.) prevented the development of abdominal postures, with a minimum effective dose (MED) of 0.3 mg/kg (Fig. 2a). The dose of 3 mg/kg produced almost total blockade of these behaviours which was maintained for up to 1.5 h (Fig. 2). The pre-emptive administration of morphine also dose-dependently (0.1±3 mg/kg, s.c.) blocked the induction of static and dynamic allodynia, with MEDs of 3 mg/kg. As in the case of abdominal postures, duration of the block was short (Fig. 2b,c). 3.3. Effect of a single post-surgical administration of morphine A single dose of morphine (3 mg/kg, s.c.) administered 30 min after ovariohysterectomy signi®cantly blocked abdominal postures, and both types of mechanical allodynias. As in the case of pre-surgical administration of morphine, the duration of these blockades was short-lasting (Fig. 3).
81
3.4. Effect of multiple peri-operative administration of morphine Morphine (3 mg/kg, s.c.) administered 30 min before surgery followed by similar administrations at 0.5 and 2 h after ovariohysterectomy completely prevented the development of abdominal postures and the two referred allodynias during the 2 days of the study (Fig. 4). 3.5. Effect of multiple post-operative administration of morphine Animals were assessed for visceral postures and, static and dynamic allodynia immediately following ovariohysterectomy and also at 24 h after surgery before any drug treatment. Unlike immediately after ovariohysterectomy, visceral postures were not apparent 24 h post-surgery, while both allodynias were present (Fig. 5). The multiple administration of morphine (3 mg/kg, s.c.) at 24.5, 25.5 and 27 h after surgery signi®cantly blocked both referred static and dynamic allodynia. However, unlike peri-operative dosing the effects disappeared within 2 h after the third administration of morphine (Fig. 5). 4. Discussion The results presented here show that ovariohysterectomy in the rat induces abdominal postures and longer lasting referred static and dynamic allodynia in the hind paws. This is consistent with pre-clinical studies showing that in¯ammation of the bladder gives rise to hyperalgesia in the hind paws (Jaggar et al., 1998). It is also consistent with clinical studies showing that referred hypersensitivity often outlasts visceral pain after surgery. The reason for referred pain is likely to be the convergence of relevant somatic and visceral nerves at the level of the spinal cord. Several sensory nerves innervate the female internal reproductive organs and enter the dorsal root ganglia at various segments of the spinal cord. It is likely that distension, ligation and incision of the cervix contribute to the central hypersensitivity. However, with respect to the referred allodynia observed in the present study, it is important to note that the pelvic nerve which carries afferent information from the cervix and vaginal canal terminates between segments L6±S1 (Berkeley and Hubscher, 1995). Some of these spinal cord segments (in particular the L6) are shared by the sciatic nerve which innervates the hind paws in rats. Thus, this particular convergence of visceral and somatic sensory afferents may be responsible for the referred allodynia described in the present study. Recently it has been shown that static and dynamic components of mechanical allodynia can be detected in pre-clinical and clinical studies (Koltzenburg et al., 1992; Field et al., 1999b). Moreover, both types of allodynia have also been described in the rat plantaris incisional model of surgical pain (Field et al., 1999a). It has been suggested that
82
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
Fig. 1. Duration of (a) abdominal postures, (b) static and (c) dynamic allodynia in the rat following anaesthesia for 15 min, skin incision, skin and abdominal wall incision or ovariohysterectomy. Baseline paw withdrawal latencies (PWL) to a cotton-bud stimulus and paw withdrawal thresholds (PWT) to von Frey hairs were determined in rats before surgery (S). Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited in 0.5-h time bins and results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM). **P , 0:01, signi®cantly different from anaesthesia group (ANOVA followed by Dunnett's t-test). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles). *P , 0:05, **P , 0:01, signi®cantly different (Dunn's test) from anaesthesia group values (n 6±8 per group).
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
the two types of allodynia are signalled by distinct sensory neurones. Thus, whilst the dynamic type is mediated by large-diameter Ab -/capsaicin-insensitive ®bres, static allo-
83
dynia appears to be exclusively signalled by Ad primary sensory neurones (Ochoa and Yarnitsky, 1993). However, it remains to be seen which types of ®bres
Fig. 2. Effect of morphine on development of (a,b) abdominal postures, (c) static and (d) dynamic allodynia in the rat ovariohysterectomy model. Morphine (M, 0.3±3 mg/kg, s.c.) was administered 0.5 h before surgery (S). Baseline paw withdrawal latencies (PWL) to a cotton bud stimulus and paw withdrawal thresholds (PWT) to von Frey hairs were determined in rats before surgery. Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited during the ®rst 30 min after surgery (a) and in 0.5-h time bins (b). Results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM) *P , 0:05, **P , 0:01, signi®cantly different (ANOVA followed by Dunnett's t-test, or unpaired t-test for (b)). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles) *P , 0:05, signi®cantly different (Dunn's test). All comparisons were carried out between drug and vehicle treated groups at each time point (n 6±15 per group).
84
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
signal nociceptive input after incision or abdominal surgery. The results of the present study show that development of long lasting abdominal postures and the two types of allo-
dynia seem to depend on the dissection of the ovaries and uterus. However, short-lasting static allodynia was also evident after incision of the muscle. There is strong
Fig. 3. Effect of morphine on the maintenance of (a) abdominal postures, (b) static and (c) dynamic allodynia in the rat ovariohysterectomy model. Morphine (M, 3 mg/kg, s.c.) was administered 0.5 h after surgery (S). Baseline paw withdrawal latencies (PWL) to a cotton bud stimulus and paw withdrawal thresholds (PWT) to von Frey hair were determined in rats before surgery. Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited in 0.5-h time bins and results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM). *P , 0:05, ***P , 0:01, signi®cantly different (unpaired t-test). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles). *P , 0:05, ***P , 0:01, signi®cantly different (Mann±Whitney U-test). All comparisons were carried out between drug and vehicle-treated groups (n 8 per group).
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
evidence suggesting that sensitization of spinal cord neurones are responsible for the induction and maintenance of allodynia. This is particularly true in the case of referred
85
allodynia. This would suggest that muscle injury-induced activation of nociceptive neurones is suf®cient to induce sensitization of super®cial layers of the dorsal horn. This
Fig. 4. Effect of multiple peri-operative administration of morphine on the induction of (a) abdominal postures, (b) static and (c) dynamic allodynia in the rat ovariohysterectomy model. Morphine (M, 3 mg/kg, s.c.) was administered 0.5 h before and at 0.5 and 2 h after surgery (S). Baseline paw withdrawal latencies (PWL) to a cotton bud stimulus and paw withdrawal thresholds (PWT) to von Frey hairs were determined in rats before surgery. Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited in 0.5-h time bins and results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM). *P , 0:05, **P , 0:01, ***P , 0:001, signi®cantly different (unpaired t-test). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles). *P , 0:05, **P , 0:01, ***P , 0:001, signi®cantly different (Mann±Whitney U-test). All comparisons were carried out between drug-treated group and saline group at each time point (n 7 per group).
86
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
is the region where Ad -®bres, which are thought to be involved in static allodynia, terminate. The failure of muscle
incision to induce dynamic allodynia suggests that this sensitization induced by somatic nociceptive ®bres does
Fig. 5. Effect of multiple post-operative administrations of morphine on the maintenance of (a) abdominal postures, (b) static and (c) dynamic allodynia in the rat ovariohysterectomy model. Abdominal postures were recorded for 30 min immediately after and 24 h post-surgery. Morphine (M, 3 mg/kg, s.c.) was administered 24.5, 25.5 and 27 h after surgery (S). Baseline paw withdrawal latencies (PWL) to a cotton bud stimulus and paw withdrawal thresholds (PWT) to von Frey hairs were determined in rats before surgery. Abdominal postures and PWL and PWT were reassessed at various times post surgery. Results for abdominal postures are expressed as mean number of postures exhibited in 0.5-h time bins and results for dynamic allodynia are expressed as mean PWL (s) (vertical bars represent ^SEM). ***P , 0:001, signi®cantly different (unpaired t-test). Results for static allodynia are expressed as median force (g) required to induce a paw withdrawal (vertical bars represent ®rst and third quartiles). ***P , 0:001, signi®cantly different (Mann±Whitney U-test). All comparisons were carried out between drug-treated group and saline group at each time point (n 8 per group).
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
not spread from the super®cial to deeper layers where the Ab -®bres that signal dynamic allodynia terminate. The development of any pain syndrome can be divided into induction and maintenance phases. Further studies have shown that distinct mechanisms are involved during the two phases (Woolf, 1994). The present study addressed the effect of morphine on each of these phases. There was no difference between a single dose of morphine administered either before or after surgery. In each case, morphine was effective at blocking abdominal postures and both static and dynamic allodynia but the duration of action was short (2 h) in each case. However, marked differences were obvious following multiple dosing with morphine depending on the time of administration. It is likely that abdominal postures represent intense activation of visceral sensory neurones. Such an activation of primary afferents would lead to sensitization of dorsal horn neurones of the spinal cord and is consistent with the presence of referred allodynia. The abdominal postures last at least 4 h, suggesting that after this time ®ring of visceral nerves subsides to a level that is suf®cient to maintain but not induce referred allodynia. This would explain why administration of morphine designed to block abdominal postures led to complete prevention of allodynia. In contrast, similar repeated administration of morphine starting 24 h after surgery had a short duration of action. These data suggest that blockade of the induction phase is of paramount importance for preventing surgical pain. Previously, it has been reported that morphine can block static but not dynamic allodynia in the plantaris incisional model of surgical pain (Field et al., 1999a). This is also the case in models of neuropathic pain (Field et al., 1999b). Therefore, it was somewhat surprising to observe that morphine was effective against both types of allodynic responses in the present study. The reason for these apparent differences may be due to the type of sensory neurone responsible for central sensitization. Thus, in neuropathic pain and the plantaris incisional model any of the three major sensory ®bres may be responsible for this process. It is known that majority of ®bres innervating the viscera is of the C-type and furthermore unlike the somatic sensory neurones, they appear to terminate in deeper layers of the spinal cord (De Groat, 1986). Therefore, these nociceptive ®bres may predominantly if not exclusively be responsible for the central sensitization process in the ovariohysterectomy model. These pro®les of morphine are consistent with electrophysiological studies showing that it can block small(C- and Ad -) but not large-diameter (Ab -) ®bre evoked responses into the dorsal horn (Dickenson and Sullivan, 1986). However, unlike the effect of morphine on the induction of dynamic allodynia, it is dif®cult to understand why it was able to block the maintenance of this response. One possible explanation is that there is a pre-synaptic link between visceral C-®bres and somatic Ab -sensory neurones (Cervero and Laird, 1996). A second is that a sustained input from visceral sensory ®bres is necessary for the mainte-
87
nance of referred dynamic allodynia. In each case blockade of C-®bre activity would explain the ability of morphine to block maintenance of dynamic allodynia. However, further evidence is required to support these hypotheses. Previous studies involving the plantaris incisional model taken together with present data indicate that the duration of intense activation of primary afferents and therefore of the induction phase is dependent on the surgical procedure. Thus, in the plantaris incisional model the induction phase was complete within 1±2 h, whereas following ovariohysterectomy it was signi®cantly longer taking over 6 h. These observations suggest that different regimens of analgesia administrations will be required depending on the surgical procedure for complete prevention of post-operative pain. This conclusion is supported by a single dose of morphine administered before surgery being able to completely prevent thermal hyperalgesia in the plantaris incisional model (Field et al., 1997) whereas, in the ovariohysterectomy model, multiple dosing of morphine was required to cover the long induction phase for complete prevention of referred allodynia. Previously, animal studies have also shown that some analgesic treatments may not be able to prevent or block all types of pain responses (Field et al., 1999a,b). Taken together, these ®ndings may help to better design future clinical studies regarding pre-emptive analgesia and post-operative pain. Firstly, it is important to have the drug on board before surgery and maintained for at least the entire induction phase. Thus, a single pre-emptive administration may not be suf®cient if the induction phase is long or that the duration of action of the drug is short. Secondly, consideration should also be given to the assessment of pain. In addition to using visual analogue scale as a measure of overall pain relief, additional parameters (e.g. hyperalgesia, allodynia, time to analgesia request and total analgesic consumption) should also be determined to give a complete pro®le of analgesia. These issues have not been standardized in the past and may re¯ect con¯icting results seen in published clinical studies. In conclusion, data presented here support the concept of pre-emptive analgesia. Furthermore, the ovariohysterectomy model should prove to be useful in studying mechanisms and designing novel therapeutic strategies for the treatment of post-operative pain.
References Berkeley KJ, Hubscher CH. Visceral and somatic sensory tracks through the neuraxis and their relation to pain: lessons from the rat female reproductive system. In: Gebhart GF, editor. Visceral pain, Progress in pain research and management, 5. Seattle: IASP Press, 1995. pp. 195±216. Brennan TJ, Vandermeulen EP, Gebhart GF. Characterisation of a rat model of incisional pain. Pain 1996;64:493±501. Cervero F. Visceral pain. In: Dubner R, Gebhart GF, Bond MR, editors. Proceedings of the Vth world congress on pain, Amsterdam: Elsevier, 1988. pp. 216±226.
88
M.I. Gonzalez et al. / Pain 88 (2000) 79±88
Cervero F, Laird MA. Mechanisms of touch-evoked pain (allodynia): a new model. Pain 1996;68:13±23. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993;52:259±285. De Groat WC. Spinal cord projections and neuropeptides in visceral afferent neurones. In: Cervero F, Morrison JFB, editors. Visceral sensation, Progress in Brain Research, 67. Amsterdam: Elsevier, 1986. pp. 165± 167. Dickenson AH, Sullivan AF. Electrophysiological studies on the effects of intrathecal morphine on nociceptive neurones in the rat dorsal horn. Pain 1986;24:211±222. Field MJ, Holloman EF, McCleary S, Hughes J, Singh L. Evaluation of gabapentin and S-(1)-3-isobutylgaba in a rat model of postoperative pain. J Pharmacol Exp Ther 1997;282:1242±1246. Field MJ, Carnell AJ, Gonzalez MI, McCleary S, Oles RJ, Smith R, Hughes J, Singh L. Enadoline, a selective k -opioid receptor agonist shows potent antihyperalgesic and antiallodynic actions in a rat model of surgical pain. Pain 1999a;80:383±389. Field MJ, McCleary S, Hughes J, Singh L. Gabapentin and pregabalin, but not morphine and amitriptyline, block both static and dynamic components of mechanical allodynia induced by streptozocin in the rat. Pain 1999b;80:391±398. Giamberdino MA, Valente R, de Bigontina P, Vecchiet L. Arti®cial ureteral calculosis in rats: behavioral characterization of visceral pain episodes and their relationship with referred lumbar muscle hyperalgesia. Pain 1995;61:459±469. Grif®n MJ, Hughes D, Knaggs A, Donnelly MB, Boylan JF. Late-onset preemptive analgesia associated with preincisional large-dose alfentanil. Anesth Analg 1997;85:1317±1321. Jaggar SI, Hasnie FS, Sellaturay S, Rice AS. The anti-hyperalgesic actions
of the cannabinoid anadamide and the putative CB2 receptor agonist palmitoylethanolamide in visceral and somatic in¯ammatory pain. Pain 1998;76:189±199. Katz J, Clairoux M, Redahan C, Kavanagh BP, Carroll S, Nierenberg H, Jackson M, Beattie J, Taddio A, Sandler N. High dose alfentanil preempts pain after abdominal hysterectomy. Pain 1996;68:109±118. Koltzenburg M, Lundberg LER, TorebjoÈrk HE. Dynamic and static components of mechanical hyperalgesia in human hairy skin. Pain 1992;51:207±219. Lascelles BDX, Waterman AE, Cripps PJ, Livingston A, Henderson G. Central sensitization as a result of surgical pain: investigation of the pre-emptive value of pethidine for ovariohysterectomy in the rat. Pain 1995;62:210±212. Lascelles BDX, Cripps PJ, Jones A, Waterman AE. Post-operative central hypersensitivity and pain: the pre-emptive value of pethidine for ovariohysterectomy. Pain 1997;73:461±471. Ochoa JL, Yarnitsky D. Mechanical hyperalgesia in neuropathic pain patients; dynamic and static subtypes. Ann Neurol 1993;33:465±472. Richards JT, Read JRM, Chambers WA. Epidural anaesthesia as a method of pre-emptive analgesia for abdominal hysterectomy. Anaesthesia 1998;53:296±307. Richmond CE, Bromley LM, Woolf CJ. Preoperative morphine pre-empts postoperative pain. Lancet 1993;342:73±75. Wall PD. The prevention of postoperative pain. Pain 1988;33:289±290. Wilson RJT, Leith S, Jackson IJB, Hunter D. Pre-emptive analgesia from intravenous administration of opioids. Anaesthesia 1994;49:591±593. Woolf CJ. A new strategy for the treatment of in¯ammatory pain. Prevention or elimination of central sensitization. Drugs 1994;47:1±9. Woolf CJ, Chong M-S. Preemptive analgesia-treating postoperative pain by preventing the establishment of central sensitization. Anesth Analg 1993;77:362±379.