Phort?iucolog.v & To.vico1og.v 1992, 71, 173-1 78
Enhancement of Morphine-Induced Analgesia and Attenuation of Morphine-Induced Side-Effectsby Cocaine in Rats Tim0 Kauppila, Ernst Mecke and Antti Pertovaara* Department of Physiology, University of Helsinki, Siltavuorenpenger 20 J, SF-00170 Helsinki, Finland (Received August 6 , 1991; Accepted February 19, 1992) Abstract: Effect of cocaine on morphine-induced analgesia and the accompanying respiratory depression, bradycardia and hypolocomotion/sedation was studied in rats. Cardiovascular and respiratory effects were studied under pentobarbitone-induced anaesthesia. Cocaine enhanced morphine-induced analgesia in the formalin test, hot plate test and heatinduced tail withdrawal test in intact rats. However, in spinal rats a similar combination of cocaine with morphine did not produce increased latencies in the tail withdrawal test. Of the three analgesic tests used, the formalin test was the most sensitive to the enhancement, as well as to the effects of morphine or cocaine alone. Morphine at the dose of 6 mg/ kg produced complete analgesia in the formalin test, significant hypolocomotion/sedation,significant bradycardia and significant decrease in the respiratory rate. At an equianalgesicdose (complete analgesia in the formalin test) of morphine (3 mg/ kg)-cocaine (5 mgi kg)-combination no significant changes in heart rate, respiratory rate or locomotion(/alertness) were observed. Changes in skin blood flow determined by the laser Doppler flow method were not significant in any of the experimental conditions. The results indicate that cocaine enhances morphine-induced analgesia, mainly due to supraspinal mechanisms. In contrast, the morphine-induced bradypnoea, bradycardia and hypolocomotiodsedation are attenuated by cocaine.
Morphine is a highly effective centrally acting analgesic in various pain conditions. However, morphine has many sideeffects which restrict its use in clinical medicine (Twycross 1985). One of the most problematic side-effects of morphine is respiratory depression which may lead to death. Morphine may also produce depression of the cardiovascular system leading to bradycardia and hypotension. Of the various less fatal side-effects, the morphine-induced sedation (Twycross 1985) is undesirable, if the patient wishes to pursue normal life. Theoretically it should be possible to attenuate the undesirable side-effects of morphine while treating pain, if morphine could be combined with another analgesic with a synergistic effect on analgesia and reverse effects on morphine-induced side-effects. Compounds with the requested properties could be cocaine and amphetamine, since both of these compounds have analgesic properties (Ivy et al. 1944a; Lim et ul. 1967; Lin et ul. 1989; Pertovaara et al. 1991b; Yang et a/. 1982) and their effects on the respiratory and cardiovascular system and alertness should be, if anything, reverse to those produced by morphine (Fischman 1984; Twycross 1985). Concerning the combination of amphetamine with morphine, previous studies show that dextroamphetamine enhances morphine-induced analgesia in humans with less side-effects than morphine at an ecluianalgesic dose alone (Forrest et ul. 1977; Handley & Ernsberg 1945; Ivy et al. 1944b). Concerning combination of cocaine with morphine some animal data indicate that cocaine may
*
To whom correspondence should be addressed
enhance morphine-induced analgesia as determined by nociceptive tail reflex responses (Misra et al. 1987; Nott 1968; Vedernikov & Afrikanov 1969). However, the side-effects (e.g. respiratory or cardiovascular ones) were not studied in these investigations. In the current study we tried to determine whether cocaine simultaneously enhances morphine-induced analgesia while counteracting the morphine-induced respiratory depression, sedation and possible cardiovascular effects. The effect of cocaine on morphine-induced analgesia was tested using the spinally-initiated tail withdrawal test and the supraspinally organized hot plate and formalin tests. Additionally, spinal animals were studied in the tail withdrawal test to reveal if the possible cocaine-induced enhancement of analgesic action of morphine is due to spinal segmental or supraspinal / spinopetal mechanisms.
Materials and Methods Animals. Adult male Wistar rats, weight range 220-480 g, were obtained from the Finnish National Laboratory Animal Centre. They had free access to pelleted food chow and water in a 12 hr
light/dark cycle. All drugs were administered intraperitoneally. The approval of the institutional Animal Care Committee of the University of Helsinki was obtained prior to the beginning of the study, and the experimental work conformed to the guiding principles in the care and use of animals as approved by the council of the American Physiologic Society. Analgesia fesfs. Analgesia was evaluated in the formalin test, in the thermally-induced tail withdrawal test, and in the hot plate test. In the formalin test slightly modified from the original one (Dubuisson & Dennis 1977), 0.05 ml of 5% formalin was injected subcutane-
174
T I M 0 KAUPPILA ET A L .
ously into the dorsum of one hindpaw. Pain behaviour was graded continuously throughout a 5 min. observation period (20-25 min. after the injection of formalin and the studied drugs; pain behaviour during this period represents the second or tonic component of the formalin-induced pain). Pain grading was done in the following way: the experimenter observed the proportion of the time the hindpaw was held up and licked (grade 3), held fully elevated (grade 2), partially weight-bearing (grade I), or fully weight-bearing (grade 0). After the pain grading the animal was terminated with an overdose of pentobarbital t o avoid excessive suffering. In the tail withdrawal test, the rat was first put into a restriction tube. Then first tail withdrawal measurement (=control) was made by immersing the distal half of the tail in a hot water bath ( 5 7 k 0 . 5 ' ) .This immersion typically evoked a clear reflex movement of the tail, the latency of which was noted. The cut-off time was 8 sec. The studied drugs were applied after the first tail withdrawal determination and the second tail withdrawal latency measurement (=measurement of the drug effect) was made 20 min. after the drug administration. The drug effects on tail withdrawal latency were expressed in percentage (loo'%,=pre-drug latency). Tail withdrawal latency was also determined in spinal animals as described above. The spinalization was made under ether anaesthesia at the midthoracic level 2-3 hr before the actual testing. The spinal animals were given an overdose of pentobarbitone immediately after the end of testing ( < 4 hr after the spinalization). In the hot plate test, each rat was placed within a clear plexiglass cylinder on a metal surface maintained at 5 4 k 0 . 5 by circulating water. The latency to hind paw lick o r escape was determined. The cut-off time was 30 sec. All tested animals had previously been adjusted to the experimental conditions by spending one minute in the apparatus with the metal plate at 45". The control latency (100%) was measured immediately before drug administrations. Drug effects in the hot plate test were determined 20 min. after intraperitoneal injections of the compounds studied. Locomotor test. The effects of the studied drugs o n locomotor activity (/alertness) were determined using an Animex Activity Meter Type M (Farad Electronics, Hagersten, Sweden). Briefly, the rat was placed in a cage (25 x 40 cm; height 15 cm, perforated plastic roof to prevent escape) located on the counting device. The locomotor activity was measured for 2 min. before drug administrations and 20 min. after drug administrations. The difference in these two locomotor counts was considered to represent the effect of the studied drugs on locomotor activity (/alertness) Respirutory undcurdi~vusculureffects. Theeffects of the studied drugs on respiratory rate, heart rate, core temperature and skin blood flow were determined in pentobarbital anaesthetized rats. Anaesthesia was induced by 40 mg/kg of pentobarbital intraperitoneally, and n o additional doses were given. About 10 min. after the induction of anaesthesia, needle electrodes were implanted for the recording of ECG, and a thermoelectrode was applied deep into the mouth cavity to determine the central temperature (Olli 535 Thermometer, Kone, Inc., Helsinki, Finland). Heart rate was measured from the ECG, and respiratory rate by observing visually the spontaneous respiratory movements. Skin blood flow in the glabrous skin of the hindfoot was monitored by means of a laser Doppler flow meter (Periflux, Perimed AB, Stockholm, Sweden). Electrical calibration for zero blood flow was made in all recordings. The output signal from the flowmeter was continuously fed into a storage oscilloscope for immediate analysis. The analogue output of the flow meter gave no absolute values but relative changes of cutaneous blood flow (Oberg 1990).The studied parameter in blood flow measurements was the drug-induced change in the relative values (pre-drug versus post-drug). The drug-induced changes were compared with the saline-induced changes (pre-saline versus post-saline).
D r q s and statistics. The drugs used were cocaine, morphine (Orion Pharmaceuticals, Helsinki. Finland), their combination, and 0.9"h saline (=control). All drugs were given intaperitoneally. In the pain
and locomotor tests, evaluation of drug effects was performed 20 min. after administration of the studied compounds. Effects on heart rate, respiratory rate, skin blood flow and core temperature were made 15 and 30 min. following the administration of the studied compounds. Statistical evaluation of the results was made using Student's t-test (two-tailed). However, pain ratings in the formalin test were statistically evaluated using Mann-Whitney Utest, since the formalin pain ratings are non-parametric values. P < 0.05 was considered a significant difference.
Results In the formalin test cocained produced a dose-dependent attenuation of pain ratings (fig. I). The minimum cocaine dose for producing significant analgesia was 10 mgikg, and cocaine at the dose of 15 mg/kg produced complete analgesia (pain score 0). Morphine at the dose of 6 mg/kg but not 3 mg/kg produced a significant analgesic effect in the formalin test (fig. I). A combination of subanalgesic doses of cocaine ( 5 mg/kg) and morphine (3 mg/kg) produced complete analgesia (pain score 0) in the formalin test. In the hot plate test, the mean latency to licking of a hind paw in the pre-drug control conditions was 7.6 0.5 s ( fS.E., n=40). Morphine treatment at the dose of 8 but not at the dose of 6 mg/kg produced a significant increase in the paw lick latency (ref.: the effect of saline; fig. 2). Cocaine treatment at the dose of 20 mg/kg, but not at the doses of 5 (data not shown) or 15 mg/kg, also produced a significant latency increase in the hot plate test (fig. 2). The combination of morphine (6 mg/kg) with cocaine produced a significant paw lick latency increase when the cocaine dose was 15 mg/kg but not when it was 5 mg/kg (fig. 2). The mean tail withdrawal latency in the pre-drug control condition ( = 100Y0)was 2.0 f0.1 s ( fS.E., n = 38) in intact rats. Fig. 3 shows the drug-induced changes in the tail withdrawal latency measured 20 min. following the drug administrations. Morphine at the dose of 8 but not at the dose of 6 mg/kg produced a significant increase in tail withdrawal latency (ref: effect of saline). Cocaine alone at the dose of 5 mg/kg had no effect. The latency increase in
PAIN R A T I N G
s
c5
cio
Cis
M3
M6
M3C5
Fig. I . Average pain ratings in the formalin test following administration of morphine and/or cocaine. Pain score 0 represents no pain. Pain ratings were performed 20-25 min. following administration of formalin subcutaneously and morphinelcocaineisaline intraperitoneally S=O.9% saline (=control), C,.,,=cocaine doses (in mg/kg). M,,=morphine doses (in mg/ kg). Error bars represent S.E. n=4-6 in each group, asterisks indicate a significant difference from the saline (S) treated group (P < 0.05, Mann-Whitney).
COCAINE: ON MORPHINE ANALGESIA
PAW LICK LATENCY ( % )
1
*
400 300 -
200 -
175
ALOCOMOTOR A C T I V I T Y
i
0-
Jc
1.1
- 50-
100 -
,
(%)
>
-
-c
n-
z
v
Me Me MsCs MeCi5 C i s Czo Fig. 2. The average latencies to paw lick in the hot plate test 20 min. following administration of morphine and/or cocaine. lOOO/o= pre-drug latency. S = 0.9”/~saline ( =control). M,, = morphine doses (in mgikg), C,.z,=cocaine doses (in mgikg). Error bars represent S.E., n = 4 6 in each group, asterisks indicate a significant difference from the saline (S) treated group ( P < 0.05, t-test). S
intact rats treated with the combination of 6 mg/kg of morphine and 5 mgikg of cocaine was short of statistical significance. The combination of 6 mg/kg of morphine with 15 mg/ kg of cocaine produced a significant latency increase in intact rats (ref: saline treated rats), but had no elTect in spinal animals (fig. 3). In the test of locomotor activity (/alertness) the rats were given saline or equianalgesic doses (according to the formalin test) of morphine (6 mg/kg), cocaine (15 nig/kg) or the combination of morphine (3 mg/kg) and cocaine (5 mg/kg). Following drug administrations all rat groups had lower locomotor activity levels (fig. 4; ref: locomotor activity prior to the drug administration). However, only the decrease of locomotor activity produced by morphine alone (6 mgikg) was significant (ref: the corresponding effect in saline treated rats). Additionally the effect of one higher dose of cocaine alone (20 mg/kg) was studied, which produced a significant increase in locomotor activity (data not shown). The effects of saline versus equianalgesic doses (according
T A I L WITHDRAWAL L A T E N C Y ( % I
*
300 4001
s
Me
M6
M3C5 M6C5
M6C15
MeCie
C5
SPIN.,
Fig. 3. Average heat-induced tail withdrawal latencies 20 min. following administrations of morphine and/or cocaine in intact and spinal rats. IOO”/a= corresponding predrug latency. See legend for fig. 2 for further explanation.
-100-
z*
T I M 0 KAUPPILA ET AL.
176 H E A R T R A T E (%)
120
**
80 S
Ms
C5M3
Cis
R ES P I R A T 0 R Y RATE
(%)
’“1 1
80
S
Ms
Cis
CsM3
Fig. 5. Average heart rates and respiratory rates in pentobarbitoneanaesthetized rats following saline (S) or equianalgesic doses (according to the formalin test) of morphine (M). cocaine ( C ) or their combination (CM). The numbers following the letters M and C indicate dose in mgi kg. 100% =corresponding value before administration of morphinelcocaineisaline. Left and right columns indicate values determined 15 and 30 min. following administration of the studied compound, respectively. Error bars represent S.E., n = 6 in each group. Asterisks indicate a significant difference (t-test) from corresponding pre-treatment (100%) rate. * = P < 0.05, ** = P < 0.005.
Discussion The main finding of the present study was that cocaine enhances the analgesic effect of morphine while simultaneously it counteracts the morphine-induced decreases in respiratory rate, heart rate and locomotor activity ( i alertness). Accordingly, it is possible to combine low, subanalgesic doses of morphine and cocaine to produce powerful analgesia with minimal undesirable side-effects. In this respect the combination of morphine with cocaine resembles
SKIN BLOOD FLOW (%)
the previously described combination of morphine with dextroamphetamine in humans (Forrest et al. 1977; Handley & Ensberg 1945; Ivy et al. 1944b), which combination also produces powerful analgesia with minimal side-effects. Cocaine has previously been combined with diamorphinc in clinical medicine in the so-called Brompton’s coctail (Fleming et al. 1990), which contained fixed amounts of cocaine (10 mg) and diamorphine (10 mg) together with ethyl alcohol syrup and chloroform water administered orally. However, since it was demonstrated that the cocaine component (10 mg) in Brompton’s coctail did not enhance analgesia in clinical studies (Kdiko et al. 1987), it was withdrawn from clinical use. The current animal study, together with previous animal studies (Misra et al. 1987; Nott 1968; Vedernikov & Afrikanov 1969), indicates that cocaine enhances the analgesic effects of morphine if given at a proper dose. The analgesic doses of morphine alone in the formalin test, hot plate and tail withdrawal tests were of the same order of magnitude as described previously for similar types of tests (D’Amour & Smith 1941; Dubuisson & Dennis 1977; O’Callaghan & Holtzman 1975). Also the analgesic cocaine doses in the formalin and hot plate tests were in the previously described range (Lin et al. 1989). Concerning the spinally-initiated nociceptive tail reflexes, according to previous reports cocaine has had only little if any effect even at very high doses (Misra et al. 1987; Nott 1968; Vedernikov & Afrikanov 1969). Thus, according to previous and present results the formalin test (its tonic component) is more sensitive both to morphine and cocaine than the thermally-induced tail withdrawal or paw lick. The same applies t o the analgesic effect of the morphine-cocaine combination as shown by the present results. The reason for the different sensitivity of these tests to the studied analgesic compounds is not known. It is possible that the subacute formalin-induced afferent barrage is more susceptible to analgesics than the acute afferent volley induced by high intensity thermal stimuli, or that the central substrates for the formalin-induced tonic pain and thermally-induced acute pain differ in their sensitivity to analgesics. Also the possible sensitivity of the formalin-induced inflammatory
A
CORE TEMPERATURE (“C)
Fig. 6. Average skin blood flow (determined by a laser Doppler flowmeter) and change i n core temperature in pentobarbitone-anaesthetized rats following administration of saline ( S ) or equianalgesic doses of morphine (M), cocaine (C) or their combination. For further details see legend for fig. 5. In the core temperature graph O=no change. Also notice, that core temperature was significantly (P
Enhancement of Morphine-Induced Analgesia and Attenuation of Morphine-Induced Side-Effectsby Cocaine in Rats Tim0 Kauppila, Ernst Mecke and Antti Pertovaara* Department of Physiology, University of Helsinki, Siltavuorenpenger 20 J, SF-00170 Helsinki, Finland (Received August 6 , 1991; Accepted February 19, 1992) Abstract: Effect of cocaine on morphine-induced analgesia and the accompanying respiratory depression, bradycardia and hypolocomotion/sedation was studied in rats. Cardiovascular and respiratory effects were studied under pentobarbitone-induced anaesthesia. Cocaine enhanced morphine-induced analgesia in the formalin test, hot plate test and heatinduced tail withdrawal test in intact rats. However, in spinal rats a similar combination of cocaine with morphine did not produce increased latencies in the tail withdrawal test. Of the three analgesic tests used, the formalin test was the most sensitive to the enhancement, as well as to the effects of morphine or cocaine alone. Morphine at the dose of 6 mg/ kg produced complete analgesia in the formalin test, significant hypolocomotion/sedation,significant bradycardia and significant decrease in the respiratory rate. At an equianalgesicdose (complete analgesia in the formalin test) of morphine (3 mg/ kg)-cocaine (5 mgi kg)-combination no significant changes in heart rate, respiratory rate or locomotion(/alertness) were observed. Changes in skin blood flow determined by the laser Doppler flow method were not significant in any of the experimental conditions. The results indicate that cocaine enhances morphine-induced analgesia, mainly due to supraspinal mechanisms. In contrast, the morphine-induced bradypnoea, bradycardia and hypolocomotiodsedation are attenuated by cocaine.
Morphine is a highly effective centrally acting analgesic in various pain conditions. However, morphine has many sideeffects which restrict its use in clinical medicine (Twycross 1985). One of the most problematic side-effects of morphine is respiratory depression which may lead to death. Morphine may also produce depression of the cardiovascular system leading to bradycardia and hypotension. Of the various less fatal side-effects, the morphine-induced sedation (Twycross 1985) is undesirable, if the patient wishes to pursue normal life. Theoretically it should be possible to attenuate the undesirable side-effects of morphine while treating pain, if morphine could be combined with another analgesic with a synergistic effect on analgesia and reverse effects on morphine-induced side-effects. Compounds with the requested properties could be cocaine and amphetamine, since both of these compounds have analgesic properties (Ivy et al. 1944a; Lim et ul. 1967; Lin et ul. 1989; Pertovaara et al. 1991b; Yang et a/. 1982) and their effects on the respiratory and cardiovascular system and alertness should be, if anything, reverse to those produced by morphine (Fischman 1984; Twycross 1985). Concerning the combination of amphetamine with morphine, previous studies show that dextroamphetamine enhances morphine-induced analgesia in humans with less side-effects than morphine at an ecluianalgesic dose alone (Forrest et ul. 1977; Handley & Ernsberg 1945; Ivy et al. 1944b). Concerning combination of cocaine with morphine some animal data indicate that cocaine may
*
To whom correspondence should be addressed
enhance morphine-induced analgesia as determined by nociceptive tail reflex responses (Misra et al. 1987; Nott 1968; Vedernikov & Afrikanov 1969). However, the side-effects (e.g. respiratory or cardiovascular ones) were not studied in these investigations. In the current study we tried to determine whether cocaine simultaneously enhances morphine-induced analgesia while counteracting the morphine-induced respiratory depression, sedation and possible cardiovascular effects. The effect of cocaine on morphine-induced analgesia was tested using the spinally-initiated tail withdrawal test and the supraspinally organized hot plate and formalin tests. Additionally, spinal animals were studied in the tail withdrawal test to reveal if the possible cocaine-induced enhancement of analgesic action of morphine is due to spinal segmental or supraspinal / spinopetal mechanisms.
Materials and Methods Animals. Adult male Wistar rats, weight range 220-480 g, were obtained from the Finnish National Laboratory Animal Centre. They had free access to pelleted food chow and water in a 12 hr
light/dark cycle. All drugs were administered intraperitoneally. The approval of the institutional Animal Care Committee of the University of Helsinki was obtained prior to the beginning of the study, and the experimental work conformed to the guiding principles in the care and use of animals as approved by the council of the American Physiologic Society. Analgesia fesfs. Analgesia was evaluated in the formalin test, in the thermally-induced tail withdrawal test, and in the hot plate test. In the formalin test slightly modified from the original one (Dubuisson & Dennis 1977), 0.05 ml of 5% formalin was injected subcutane-
174
T I M 0 KAUPPILA ET A L .
ously into the dorsum of one hindpaw. Pain behaviour was graded continuously throughout a 5 min. observation period (20-25 min. after the injection of formalin and the studied drugs; pain behaviour during this period represents the second or tonic component of the formalin-induced pain). Pain grading was done in the following way: the experimenter observed the proportion of the time the hindpaw was held up and licked (grade 3), held fully elevated (grade 2), partially weight-bearing (grade I), or fully weight-bearing (grade 0). After the pain grading the animal was terminated with an overdose of pentobarbital t o avoid excessive suffering. In the tail withdrawal test, the rat was first put into a restriction tube. Then first tail withdrawal measurement (=control) was made by immersing the distal half of the tail in a hot water bath ( 5 7 k 0 . 5 ' ) .This immersion typically evoked a clear reflex movement of the tail, the latency of which was noted. The cut-off time was 8 sec. The studied drugs were applied after the first tail withdrawal determination and the second tail withdrawal latency measurement (=measurement of the drug effect) was made 20 min. after the drug administration. The drug effects on tail withdrawal latency were expressed in percentage (loo'%,=pre-drug latency). Tail withdrawal latency was also determined in spinal animals as described above. The spinalization was made under ether anaesthesia at the midthoracic level 2-3 hr before the actual testing. The spinal animals were given an overdose of pentobarbitone immediately after the end of testing ( < 4 hr after the spinalization). In the hot plate test, each rat was placed within a clear plexiglass cylinder on a metal surface maintained at 5 4 k 0 . 5 by circulating water. The latency to hind paw lick o r escape was determined. The cut-off time was 30 sec. All tested animals had previously been adjusted to the experimental conditions by spending one minute in the apparatus with the metal plate at 45". The control latency (100%) was measured immediately before drug administrations. Drug effects in the hot plate test were determined 20 min. after intraperitoneal injections of the compounds studied. Locomotor test. The effects of the studied drugs o n locomotor activity (/alertness) were determined using an Animex Activity Meter Type M (Farad Electronics, Hagersten, Sweden). Briefly, the rat was placed in a cage (25 x 40 cm; height 15 cm, perforated plastic roof to prevent escape) located on the counting device. The locomotor activity was measured for 2 min. before drug administrations and 20 min. after drug administrations. The difference in these two locomotor counts was considered to represent the effect of the studied drugs on locomotor activity (/alertness) Respirutory undcurdi~vusculureffects. Theeffects of the studied drugs on respiratory rate, heart rate, core temperature and skin blood flow were determined in pentobarbital anaesthetized rats. Anaesthesia was induced by 40 mg/kg of pentobarbital intraperitoneally, and n o additional doses were given. About 10 min. after the induction of anaesthesia, needle electrodes were implanted for the recording of ECG, and a thermoelectrode was applied deep into the mouth cavity to determine the central temperature (Olli 535 Thermometer, Kone, Inc., Helsinki, Finland). Heart rate was measured from the ECG, and respiratory rate by observing visually the spontaneous respiratory movements. Skin blood flow in the glabrous skin of the hindfoot was monitored by means of a laser Doppler flow meter (Periflux, Perimed AB, Stockholm, Sweden). Electrical calibration for zero blood flow was made in all recordings. The output signal from the flowmeter was continuously fed into a storage oscilloscope for immediate analysis. The analogue output of the flow meter gave no absolute values but relative changes of cutaneous blood flow (Oberg 1990).The studied parameter in blood flow measurements was the drug-induced change in the relative values (pre-drug versus post-drug). The drug-induced changes were compared with the saline-induced changes (pre-saline versus post-saline).
D r q s and statistics. The drugs used were cocaine, morphine (Orion Pharmaceuticals, Helsinki. Finland), their combination, and 0.9"h saline (=control). All drugs were given intaperitoneally. In the pain
and locomotor tests, evaluation of drug effects was performed 20 min. after administration of the studied compounds. Effects on heart rate, respiratory rate, skin blood flow and core temperature were made 15 and 30 min. following the administration of the studied compounds. Statistical evaluation of the results was made using Student's t-test (two-tailed). However, pain ratings in the formalin test were statistically evaluated using Mann-Whitney Utest, since the formalin pain ratings are non-parametric values. P < 0.05 was considered a significant difference.
Results In the formalin test cocained produced a dose-dependent attenuation of pain ratings (fig. I). The minimum cocaine dose for producing significant analgesia was 10 mgikg, and cocaine at the dose of 15 mg/kg produced complete analgesia (pain score 0). Morphine at the dose of 6 mg/kg but not 3 mg/kg produced a significant analgesic effect in the formalin test (fig. I). A combination of subanalgesic doses of cocaine ( 5 mg/kg) and morphine (3 mg/kg) produced complete analgesia (pain score 0) in the formalin test. In the hot plate test, the mean latency to licking of a hind paw in the pre-drug control conditions was 7.6 0.5 s ( fS.E., n=40). Morphine treatment at the dose of 8 but not at the dose of 6 mg/kg produced a significant increase in the paw lick latency (ref.: the effect of saline; fig. 2). Cocaine treatment at the dose of 20 mg/kg, but not at the doses of 5 (data not shown) or 15 mg/kg, also produced a significant latency increase in the hot plate test (fig. 2). The combination of morphine (6 mg/kg) with cocaine produced a significant paw lick latency increase when the cocaine dose was 15 mg/kg but not when it was 5 mg/kg (fig. 2). The mean tail withdrawal latency in the pre-drug control condition ( = 100Y0)was 2.0 f0.1 s ( fS.E., n = 38) in intact rats. Fig. 3 shows the drug-induced changes in the tail withdrawal latency measured 20 min. following the drug administrations. Morphine at the dose of 8 but not at the dose of 6 mg/kg produced a significant increase in tail withdrawal latency (ref: effect of saline). Cocaine alone at the dose of 5 mg/kg had no effect. The latency increase in
PAIN R A T I N G
s
c5
cio
Cis
M3
M6
M3C5
Fig. I . Average pain ratings in the formalin test following administration of morphine and/or cocaine. Pain score 0 represents no pain. Pain ratings were performed 20-25 min. following administration of formalin subcutaneously and morphinelcocaineisaline intraperitoneally S=O.9% saline (=control), C,.,,=cocaine doses (in mg/kg). M,,=morphine doses (in mg/ kg). Error bars represent S.E. n=4-6 in each group, asterisks indicate a significant difference from the saline (S) treated group (P < 0.05, Mann-Whitney).
COCAINE: ON MORPHINE ANALGESIA
PAW LICK LATENCY ( % )
1
*
400 300 -
200 -
175
ALOCOMOTOR A C T I V I T Y
i
0-
Jc
1.1
- 50-
100 -
,
(%)
>
-
-c
n-
z
v
Me Me MsCs MeCi5 C i s Czo Fig. 2. The average latencies to paw lick in the hot plate test 20 min. following administration of morphine and/or cocaine. lOOO/o= pre-drug latency. S = 0.9”/~saline ( =control). M,, = morphine doses (in mgikg), C,.z,=cocaine doses (in mgikg). Error bars represent S.E., n = 4 6 in each group, asterisks indicate a significant difference from the saline (S) treated group ( P < 0.05, t-test). S
intact rats treated with the combination of 6 mg/kg of morphine and 5 mgikg of cocaine was short of statistical significance. The combination of 6 mg/kg of morphine with 15 mg/ kg of cocaine produced a significant latency increase in intact rats (ref: saline treated rats), but had no elTect in spinal animals (fig. 3). In the test of locomotor activity (/alertness) the rats were given saline or equianalgesic doses (according to the formalin test) of morphine (6 mg/kg), cocaine (15 nig/kg) or the combination of morphine (3 mg/kg) and cocaine (5 mg/kg). Following drug administrations all rat groups had lower locomotor activity levels (fig. 4; ref: locomotor activity prior to the drug administration). However, only the decrease of locomotor activity produced by morphine alone (6 mgikg) was significant (ref: the corresponding effect in saline treated rats). Additionally the effect of one higher dose of cocaine alone (20 mg/kg) was studied, which produced a significant increase in locomotor activity (data not shown). The effects of saline versus equianalgesic doses (according
T A I L WITHDRAWAL L A T E N C Y ( % I
*
300 4001
s
Me
M6
M3C5 M6C5
M6C15
MeCie
C5
SPIN.,
Fig. 3. Average heat-induced tail withdrawal latencies 20 min. following administrations of morphine and/or cocaine in intact and spinal rats. IOO”/a= corresponding predrug latency. See legend for fig. 2 for further explanation.
-100-
z*
T I M 0 KAUPPILA ET AL.
176 H E A R T R A T E (%)
120
**
80 S
Ms
C5M3
Cis
R ES P I R A T 0 R Y RATE
(%)
’“1 1
80
S
Ms
Cis
CsM3
Fig. 5. Average heart rates and respiratory rates in pentobarbitoneanaesthetized rats following saline (S) or equianalgesic doses (according to the formalin test) of morphine (M). cocaine ( C ) or their combination (CM). The numbers following the letters M and C indicate dose in mgi kg. 100% =corresponding value before administration of morphinelcocaineisaline. Left and right columns indicate values determined 15 and 30 min. following administration of the studied compound, respectively. Error bars represent S.E., n = 6 in each group. Asterisks indicate a significant difference (t-test) from corresponding pre-treatment (100%) rate. * = P < 0.05, ** = P < 0.005.
Discussion The main finding of the present study was that cocaine enhances the analgesic effect of morphine while simultaneously it counteracts the morphine-induced decreases in respiratory rate, heart rate and locomotor activity ( i alertness). Accordingly, it is possible to combine low, subanalgesic doses of morphine and cocaine to produce powerful analgesia with minimal undesirable side-effects. In this respect the combination of morphine with cocaine resembles
SKIN BLOOD FLOW (%)
the previously described combination of morphine with dextroamphetamine in humans (Forrest et al. 1977; Handley & Ensberg 1945; Ivy et al. 1944b), which combination also produces powerful analgesia with minimal side-effects. Cocaine has previously been combined with diamorphinc in clinical medicine in the so-called Brompton’s coctail (Fleming et al. 1990), which contained fixed amounts of cocaine (10 mg) and diamorphine (10 mg) together with ethyl alcohol syrup and chloroform water administered orally. However, since it was demonstrated that the cocaine component (10 mg) in Brompton’s coctail did not enhance analgesia in clinical studies (Kdiko et al. 1987), it was withdrawn from clinical use. The current animal study, together with previous animal studies (Misra et al. 1987; Nott 1968; Vedernikov & Afrikanov 1969), indicates that cocaine enhances the analgesic effects of morphine if given at a proper dose. The analgesic doses of morphine alone in the formalin test, hot plate and tail withdrawal tests were of the same order of magnitude as described previously for similar types of tests (D’Amour & Smith 1941; Dubuisson & Dennis 1977; O’Callaghan & Holtzman 1975). Also the analgesic cocaine doses in the formalin and hot plate tests were in the previously described range (Lin et al. 1989). Concerning the spinally-initiated nociceptive tail reflexes, according to previous reports cocaine has had only little if any effect even at very high doses (Misra et al. 1987; Nott 1968; Vedernikov & Afrikanov 1969). Thus, according to previous and present results the formalin test (its tonic component) is more sensitive both to morphine and cocaine than the thermally-induced tail withdrawal or paw lick. The same applies t o the analgesic effect of the morphine-cocaine combination as shown by the present results. The reason for the different sensitivity of these tests to the studied analgesic compounds is not known. It is possible that the subacute formalin-induced afferent barrage is more susceptible to analgesics than the acute afferent volley induced by high intensity thermal stimuli, or that the central substrates for the formalin-induced tonic pain and thermally-induced acute pain differ in their sensitivity to analgesics. Also the possible sensitivity of the formalin-induced inflammatory
A
CORE TEMPERATURE (“C)
Fig. 6. Average skin blood flow (determined by a laser Doppler flowmeter) and change i n core temperature in pentobarbitone-anaesthetized rats following administration of saline ( S ) or equianalgesic doses of morphine (M), cocaine (C) or their combination. For further details see legend for fig. 5. In the core temperature graph O=no change. Also notice, that core temperature was significantly (P
