Introduction

Top Abstract Introduction Methods Results Discussion References

Buprenorphine is a semisynthetic, highly lipophilic opioid derived from thebaine. It is currently used as a postoperative analgesic and as has been demonstrated to be efficacious in the treatment of opiate abuse (cf., Bickel and Amass, 1995). In vivo infra-human studies have established that the analgesic effects of buprenorphine are mediated at the mu 1 (Kamei et al., 1995) and kappa 3 receptor (Pick et al., 1997). There is also evidence of antagonism of analgesic effects of buprenorphine at the kappa 1 receptor (Leander, 1988). Buprenorphine is classified as a partial mu agonist. Evidence for lesser efficacy than the full mu agonists comes from both infra-human and human studies in which the analgesic (Cowan et al., 1977b; Pedersen et al., 1986; Walker et al., 1995), subjective (Walsh et al., 1994, 1995), miotic (Walsh et al., 1995) and respiratory (Cowan et al., 1977a; Walsh et al., 1994, 1995; Liguori et al., 1996) effects of the drug have been studied. Buprenorphine is considered to be 25 to 50 times more potent than morphine (Martin et al., 1976; Cowan et al., 1977b: Jasinski et al., 1978; Reisine and Pasternak, 1996) and 0.3 to 0.4 mg of buprenorphine (i.m.or i.v.) is regularly referenced in the literature as equianalgesic to that of 10 mg morphine (e.g., Wang et al., 1981; Ouellette, 1984; Department of Health and Human Services, 1992; Noren, 1994).

Abuse liability studies of buprenorphine in infra-humans have focused on the reinforcing and DS effects of the drug. Buprenorphine functions as a reinforcer in infra-humans as determined by several assays (Young et al., 1984; Mello et al., 1988; Yanagita et al., 1982; Pchelintsev et al., 1991; Hubner and Kornetsky, 1988). In drug discrimination studies, buprenorphine substitutes for mu agonists (Leander, 1983; France et al., 1984; Young et al., 1984; France and Woods, 1985; Picker and Dykstra, 1989; Negus et al., 1990; Negus et al., 1991; Paronis and Holtzman, 1994; Walker et al., 1994) but not for kappa agonists (Young et al., 1984; Negus et al., 1990).

Human psychopharmacological research on buprenorphine that is related to its abuse liability has focused on characterizing the discriminative, subjective and psychomotor-impairing effects of buprenorphine in opioid users. Buprenorphine in one drug discrimination study engendered hydromorphone-appropriate responding in nondependent opiate abusers who were trained to discriminate between hydromorphone, butorphanol and saline (Preston and Bigelow, 1994). This suggest that buprenorphine shares DS effects with a full mu agonist more so than with a prototypic mixed agonist-antagonist. The subjective effects of buprenorphine have been studied in this same population across different dose ranges, routes of administration and for differing periods of time (e.g., 2 hr vs. 4 days). Perhaps because of these differences, there have also been some differences noted in the subjective effects profile of buprenorphine across studies. The most consistent finding is an increase in drug liking and good (drug) effects (Jasinski et al., 1978; Jasinski et al., 1989; Preston et al., 1989a; Preston et al., 1992; Weinhold et al., 1992; Pickworth et al.1993; Preston and Bigelow, 1994; Walsh et al., 1994, 1995; Foltin and Fischman, 1994, 1995). A number of studies have also noted increases in MBG scores (Jasinski et al., 1978; Preston et al., 1989a, 1992; Pickworth et al., 1993). Thus, buprenorphine appears to have euphoriant effects. Several studies, though, have noted increases in subjective effects indicative of sleepiness (e.g. increase in Pentobarbital-Chlorpromazine-Alcohol Group [PCAG] scores, increase in rating of sleepy on an opiate adjective checklist) [Jasinski et al., 1978; Weinhold et al., 1992; Pickworth et al., 1993; Walsh et al., 1994, 1995; Foltin and Fischman, 1995]. The soporific effect occurs at a later point in time during a session than the onset of euphoriant effects (Jasinski et al., 1978).

Four studies have assessed subjective effects of buprenorphine in non-drug abusing volunteers (Blom et al., 1987; Manner et al., 1987; Saarialho-Kere et al., 1987; MacDonald et al., 1989). Doses and route of administration varied across the studies and each study only examined one dose of buprenorphine. Each study used a VAS to measure mood but there was no concordance across studies in VAS descriptors used. In all studies, subjects reported drowsiness. Also all studies reported some subjects feeling nauseous or vomiting after buprenorphine administration. It is interesting to note that in one study in which 0.3 mg (i.m.) buprenorphine was administered, 9 of 12 subjects at some point during the 8-hr sessions were too incapacitated to carry out the experimental protocol (MacDonald et al., 1989). Many patient studies in which buprenorphine has been given for postoperative pain relief have noted similar subjective effects obtained in the aforementioned healthy volunteer studies: drowsiness, nausea and vomiting (cf., Heel et al., 1979).

Effects of buprenorphine on psychomotor performance have been assessed in opioid users (both nondependent and dependent) and in nonopioid users. The primary measure of psychomotor performance in the population of opioid abusers has been the DSST, and most studies (but see Weinhold et al., 1992) have shown no impairment on this measure (Preston et al., 1988, 1989a; Strain et al., 1992; Preston and Bigelow, 1994; Walsh et al., 1994). In contrast, buprenorphine administered by several routes to nonopioid users have impaired performance on one or more of the following tests: Maddox Wing, DSST, choice reaction time and eye-hand coordination (Manner et al., 1987; Saarialho-Kere et al., 1987; MacDonald et al., 1989).

The psychopharmacological effects of buprenorphine as they relate to abuse liability have been extremely well-characterized in opioid abusers. This is less true for non-drug abusers: no study to date has constructed a dose-response function of buprenorphine using a methodology that is well accepted in abuse liability testing. The purpose of our study was to characterize the subjective, psychomotor and physiological effects of buprenorphine using a range of doses up to doses that are given for postoperative pain relief. Such a study should enable the following questions to be addressed: 1) What is the abuse liability of buprenorphine in a nondrug abusing population? 2) Are there qualitative or quantitative differences between the effects of buprenorphine in this study and its effects in studies conducted with opiate abusers? 3) What mood-altering effects of buprenorphine might nondrug-abusing patients in hospitals be experiencing after being administered buprenorphine? In addition, to determine the extent to which single equianalgesic doses of two opioids with different efficacies at the mu receptor differ from, or are similar to, each other, the highest dose of buprenorphine tested was compared to an equianalgesic dose of morphine, a full mu agonist.

	Methods

Top Abstract Introduction Methods Results Discussion References

Subjects

Candidates were recruited via posters and local newspaper advertisements. Potential subjects who consumed, on average, at least one alcoholic drink per week and were between the ages of 21 to 39 were scheduled for a screening interview with one of our trained research personnel. During the interview, candidates completed the SCL-90, a questionnaire designed to assess psychiatric symptomatology (Derogatis et al., 1973) and a health questionnaire designed to determine their psychiatric and mental status. Candidates with any psychiatric problems, including drug- or alcohol-related problems or Diagnostic and Statistical Manual of Mental Disorders-III-Revised Axis I psychiatric disorders (American Psychiatric Association, 1987), were excluded, based on a structured psychiatric interview.

Potential subjects participated in an orientation session before the start of the study. Before onset of the orientation session, subjects signed a written consent form that described the study in detail. In the consent form, subjects were told that the i.v. drugs to be used in the study were drugs commonly used in medical settings and might come from one of six classes: sedative, stimulant, opiate, general anesthetic (at subanesthetic doses), alcohol or placebo. Subjects then received a resting-state electrocardiogram, and a physician performed a medical history and an examination. Any participants who had experienced any adverse reactions to general anesthetics or pulmonary, renal, hepatic or cardiac problems were excluded from the study. Subjects were required to give a urine sample, upon which the Enzymatic Multiplied Immunoassay Technique (EMIT) toxicology screening for acetaminophen, alcohol, amphetamines, barbiturates, benzodiazepines, cocaine metabolites, opiates, phencyclidine and salicylate was performed. None of the subjects tested positive for any of the above drugs or metabolites. Mood and psychomotor tests were practiced by volunteers during the orientation to acclimate them to the tests and also to avoid any practice effects on psychomotor testing during experimental sessions. Payment for the study was made at the debriefing, held once the study was completed. The study was approved by the local Institutional Review Board.

Sixteen healthy volunteers, 11 male and 5 female (age range 21-35 yr, mean age 25.8 yr), participated. All volunteers had some prior use of recreational drugs, but none had past histories indicative of dependence. Their self-reported number of alcohol drinks consumed per week (over the last 30 days) was 6.7 (range 1-10). Eight volunteers reported smoking marijuana in the past 30 days (0.8 joints/week), and one subject reported using psilocibin mushrooms in this same time period. Regarding lifetime nonmedical drug use, 1 volunteer reported use of ketamine (less than 10 times), 1 volunteer reported use of glue (less than 10 times), 5 volunteers reported use of nitrous oxide, 5 volunteers reported use of cocaine (less than 10 times), 7 volunteers reported use of hallucinogens (LSD, mushrooms) and 15 volunteers reported use of cannabinoids. Seven of the 16 volunteers reported prior exposure to opiates: 3 had used these drugs (reported by volunteers as opium, meperidine, codeine, aspirin with oxycodone, acetaminophen with oxycodone, acetaminophen with codeine) for nonmedical reasons and the other 4 had been prescribed opiates (reported as codeine, aspirin with oxycodone, acetaminophen with codeine, or "painkillers") in the past for pain relief. In all cases, use of any one opiate was reported as less than 10 times, and no subject reported using more than two opiates.

Procedure

Experimental design. A randomized, placebo-controlled, double-blind, cross-over trial was conducted. Subjects were injected in an upper extremity vein with saline, 0.075, 0.15, 0.3 mg/70 kg buprenorphine, or 10 mg/70 kg morphine, over a 30-sec interval. The drug was always delivered in a volume of 20 ml containing drug and/or saline. We chose the 10 mg morphine dose because it is considered equianalgesic to the 0.3 mg dose of buprenorphine (e.g., Heel et al., 1979; Department of Health and Human Services, 1992; Noren, 1994). Subjects participated in five sessions spaced at least 1 wk apart. Sessions were approximately 360 min in duration.

Experimental sessions. The experiment took place in a departmental laboratory. Subjects were instructed not to eat food or drink any nonclear liquids for 4 hr, not to drink clear liquids for 2 hr and not to use any drugs (including alcohol, but excluding normal amounts of caffeine and nicotine) 24 hr before sessions. A toxicology screening was required before the start of each session for all participants as was a pregnancy test for all female participants. Subjects were also given a breath alcohol test to assure they had no alcohol in their system. An angiocatheter was inserted into one of the subject's upper extremity veins by an anesthetist. Subjects then completed several subjective effects forms and psychomotor tests and their respiratory rate, heart rate, noninvasive arterial oxygen saturation and blood pressure were monitored. Subjects then reclined in a semirecumbent position on a bed and, using proper sterile technique, an anesthetist injected into the angiocatheter either morphine, buprenorphine or saline over a period of 30 sec. Before the injection the subject was told, "The injection you are about to receive may or may not contain a drug." The drug was previously drawn up by one anesthetist and administered by another to preserve the double-blind nature of the study. However, the injecting anesthetist was aware of the drugs involved, so that if an adverse event occurred, appropriate measures could be taken to ensure the safety and well-being of the subject. The anesthetist remained in the immediate area for 15 min after the injection to monitor vital signs of the subject. At periodic intervals after the injection (see below), mood, psychomotor performance and physiological status of the subject were assessed. Drinking water was permitted 90 min after the injection, but eating was not allowed during the session. A snack was served to the subject after the session was terminated. When no tests were scheduled, subjects were free to engage in sedentary recreational activities such as reading, listening to the radio or to cassette tapes and watching TV, but studying was not permitted. Social interaction was possible in the study (e.g., the subject could converse with the research technician), but subjects generally engaged in solitary activities during sessions. After completion of sessions, subjects were transported home via a livery service with instructions not to engage in certain activities for the following 24 hr (e.g., cooking with a stove, driving an automobile, caring for children, drinking alcohol).

Dependent measures. The following tests were completed before injection, and 15, 60, 120, 180, 240 and 300 min after injection. On all of these measures, subjects did not have access to how they responded on previous tests from the same session. When subjects performed computerized tests, they had to move from the middle of the bed to the edge of the bed, still in a semirecumbent position. When subjects performed paper-and-pencil tests, they did not have to move at all. Thus, subject movement during or before testing was minimal in our sessions. Table 1 lists the order of testing, which remained invariant across subjects and sessions.

Subjective measures. The ARCI is a true-false questionnaire designed to differentiate among different classes of psychoactive drugs (Haertzen, 1966). A computerized short-form of the ARCI was used (Martin et al., 1971) which had 49 items and yielded scores for 5 different scales: PCAG, sensitive to sedative effects; BG and AMP, sensitive to amphetamine-like effects; LSD, sensitive to somatic and dysphoric changes and MBG, often described as euphoria.

Psychomotor/cognitive performance. The following six tests were chosen because we have used these tests in our prior opioid studies and because previous studies from other laboratories have indicated that the specific parameters of psychomotor/cognitive performance that the tests are designed to measure can be affected by opioids (cf., Zacny, 1995).

Physiological measures. Five physiological measures were assessed: heart rate, blood pressure, arterial oxygen saturation, respiration rate and miosis. Heart rate, blood pressure and arterial oxygen saturation were measured noninvasively with a Merlin model 54 monitor (Hewlett Packard, Andover, MA). Respiration rate was the number of breaths subjects took in 30 sec (multiplied by 2 to get breaths/min). This was assessed by counting the number of times the subject's chest or stomach rose and fell and was measured by one of the experimenters (K.C.), who was blind to the dose and drug being administered. Heart rate, blood pressure, arterial oxygen saturation and respiratory rate were assessed at the time points listed above. Miosis, or pupil constriction, is a physiological marker of opiate effects and was measured by photographing the subject's right pupil in a dimly lighted room. Miosis was measured before injection, and 15, 60, 120, 180 and 300 min after injection.

Data Analysis

Two sets of repeated measures ANOVA were used for statistical treatment of the data. The first analysis examined buprenorphine effects: Factors were Dose (0, 0.075, 0.15 and 0.3 mg/70 kg) and Time (2-13 levels). The second analysis compared peak and/or trough effects of saline, and 0.3 mg buprenorphine and 10 mg morphine. Only postinjection values were included in this analysis. Only 15 subjects were included in this analysis because one subject (G.S.) received the wrong dose of morphine appropriate to her body weight, and her data were not included in the second set of analyses. F values were considered significant for P < .05 with adjustments of within-factors degrees of freedom (Huynh-Feldt) to protect against violations of symmetry. Tukey post hoc testing was done on the first set of ANOVAs, comparing drug responses to saline at each time point, and on the second set of ANOVAs, comparing each of the three conditions to each other. Variance measures that are reported adjacent to ratings/scores represent S.E.M.

	Results

Top Abstract Introduction Methods Results Discussion References

Subjective Effects

ARCI. Buprenorphine. Significant Dose X Time effects were obtained on the PCAG (P < .005), BG (P < .01) and LSD (P < .005) scales. In a dose-related manner, PCAG and LSD scores increased, and BG scores decreased after buprenorphine injection (Fig. 1). Scores on the BG and LSD scales reached their maximum trough and peak levels, respectively, 15 min after injection whereas scores on the PCAG scale did not peak until 60 min postinjection. Duration of effect was protracted with effects still present at the end of the session. For comparison purposes figure 1 also shows scores from the 10-mg morphine condition. PCAG and LSD scores were significantly increased, and BG scores were significantly decreased after morphine administration. In contrast to buprenorphine, effects of morphine declined gradually throughout the session and approached baseline levels at the end of the session.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Time course of the effects of 0 (), 0.075 (), 0.15 () and 0.3 mg/70 kg () buprenorphine on scores from the PCAG (left frame), BG (middle frame) and LSD (right frame) scales of the ARCI. For comparative purposes to 0.3 mg buprenorphine, 10 mg morphine () is also shown on the graph. Each buprenorphine point is the mean across 15 subjects (one subject had missing data points in one of the buprenorphine sessions due to a technical error and is not included in the analysis) and each morphine point is the mean across 15 subjects (one of the 16 subjects received the wrong dose of morphine and her morphine data were dropped from the analysis). Time point 0 refers to effects measured immediately before the injection. Asterisks on the graphs indicate that a buprenorphine dose is significantly different from saline at a given time point (Tukey post hoc test; P < .05). Range of possible scores on the PCAG, BG and LSD scores are 0 to 15, 0 to 9 and 0 to 14, respectively.

Adjective checklist. Buprenorphine. Significant increases were obtained on five adjectives from the adjective checklist: "dry mouth" (Dose X Time: P < .005), "nodding" (Dose X Time: P < .001), "numb" (Dose X Time: P < .001), "skin itchy" (Dose X Time: P < .001) and "turning of stomach" (Dose X Time: P < .001). The effects of buprenorphine were dose-related in terms of magnitude and duration of effect. Most effects increased soon after injection and peaked at 60 to 120 min postinjection and remained significantly elevated up to 240 to 300 min postinjection. The rating "turning of stomach" peaked later than the other effects (120-min postinjection) and remained elevated for the remainder of the session with all buprenorphine doses. The rating, "drive," was significantly decreased by buprenorphine (Dose: P < .05).

VAS. Buprenorphine. Significant Dose X Time effects (except where otherwise noted) were obtained on ratings of "coasting (spaced out)" (P < .001), "confused" (Dose: P < .01), "difficulty concentrating" (P < .005) [figure 2, left frame], "dizzy" (P < .01) [figure 2, center frame], "floating" (P < .005), "having unpleasant bodily sensations" (P < .05), "having unpleasant thoughts" (Dose: P < .05), "heavy or sluggish feeling" (P < .001), "high" (P < .001), "hungry" (P < .05), "lightheaded" (P < .01), "nauseous" (P < .01) [figure 2, right frame], "sedated" (P < .01), "sleepy ('drowsy, tired')" (P < .001), "stimulated" (Dose: P < .05) and "tingling" (P < .05). All of the above VAS ratings, with the exception of "hungry" and "stimulated," increased after drug injection and were affected by buprenorphine in a dose-related manner. Some ratings peaked at 5 to 15 min postinjection ("dizzy," "high" and "lightheaded") and others peaked 60 to 120 min postinjection (e.g., "coasting" and "sleepy"). Most effects, especially at the higher doses, were protracted and remained elevated up to the end of the session.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Time course of the effects of 0 (), 0.075 (), 0.15 () and 0.3 mg/70 kg () buprenorphine on "difficulty concentrating," (left frame) "dizzy," (center frame) and "nauseous" (right frame) ratings from the VAS (range 0-100 mm). For comparative purposes to 0.3 mg buprenorphine, 10 mg morphine (solid circle) is also shown on the graph. Each buprenorphine point is the mean across 16 subjects, and each morphine point is the mean across 15 subjects. Time point 0 refers to effects measured immediately before the injection. Asterisks on the graphs indicate that a buprenorphine dose is significantly different from saline at a given time point (Tukey post hoc test; P < .05).

Drug Effects and Drug Liking

Buprenorphine. Significant Dose X Time increases were obtained on the "Feel Drug Effect" question (P < .001) and the "Like Drug Effect" rating (P < .05). "Feel Drug Effect" ratings were related to buprenorphine dose in an orderly fashion, and subjects still reported an effect at 300-min postinjection with all three doses (fig. 3, left frame). "Like Drug Effect" ratings did not differ between saline and any of the buprenorphine doses at any of the post-injection time points; however, post hoc testing revealed that there was a significant decrease in ratings (i.e., dislike of drug effects) in the 0.3-mg condition, when comparing the last two postinjection time points to the preinjection time point (fig. 3, right frame).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Time course of the effects of 0 (), 0.075 (), 0.15 () and 0.3 mg/70 kg () buprenorphine on the "Feel Drug Effect" question (left frame) and "Like Drug Effect" question (right frame) from the Drug Effects/Liking questionnaire. For comparative purposes to 0.3 mg buprenorphine, 10 mg morphine () is also shown on the graph. Each buprenorphine point is the mean across 16 subjects and each morphine point is the mean across 15 subjects. Time point 0 refers to effects measured immediately prior to the injection. On the "Feel Drug Effect" question, subjects were asked to rate the intensity of the drug effect as they were currently experiencing it (from 1 = `I feel no effect from it at all' to 5 = `I feel a very strong effect'). On the "Like Drug Effect" question, subjects were asked how much they liked the drug effect, as they were currently experiencing it (on a visual analog scale of 0-100 with the following descriptors: 0 = dislike a lot; 50 = neutral; 100 = like a lot). For the rating of Feel Drug Effect, the 2.5-, 5- and 10-mg nalbuphine doses were significantly different from saline between 15 to 300, 5 to 300 and 5 to 300 min postinjection, respectively. For the rating of Like Drug Effect, in the 0.3-mg buprenorphine condition, the preinjection time point differed significantly from the 240- and 300-min postinjection time points.

For comparison purposes figure 3 also shows scores from the 10-mg morphine condition. Magnitude of "Feel Drug Effect" resembled that of the 0.075- and 0.15-mg buprenorphine doses. Ten mg morphine did not show a decrease in liking ratings as did 0.3 mg buprenorphine.

Peak and trough effects. Significantly higher peak "feel drug effect" ratings were obtained with 0.3 mg buprenorphine (4.7 ± 0.1) and 10 mg morphine (3.7 ± 0.2), relative to the saline condition (1.9 ± 0.2). Further, peak drug effect ratings were significantly higher in the 0.3-mg buprenorphine condition than in the 10-mg morphine condition. Because of the bipolar nature of the drug liking question (i.e., 50 = neutral and 0 and 100 are representative of extreme dislike and extreme liking, respectively), separate peak and trough effect analyses were performed on this measure. Peak liking ratings were significantly higher in the 0.3-mg buprenorphine condition (70.8 ± 3.2) and the 10-mg morphine (67.3 ± 4.6) than in the saline condition (56.5 ± 3.2). The two drug conditions did not differ significantly from each other. However, trough liking ratings were significantly lower in the 0.3-mg buprenorphine (22.8 ± 4.9) than in the 10-mg morphine condition (37.4 ± 4.1) and the saline condition (42.0 ± 3.1), which did not differ significantly from each other.

Postsession Questionnaire. Buprenorphine. On the questionnaire that assessed residual effects of the drug, significant Dose effects were obtained with buprenorphine on five ratings: "headache" (P < .01), "heavy or sluggish feeling" (P < .001), "lightheaded" (P < .05), "nausea" (P < .05) and "skin itchy" (P < .05). All buprenorphine doses produced higher ratings than did saline on the ratings of "headache" and "heavy," but with the other three ratings in which overall significance was obtained, only the 0.3-mg buprenorphine dose differed from saline.

Psychomotor performance. Buprenorphine. Buprenorphine impaired performance in a dose-related fashion on all five psychomotor tests: Maddox Wing (Dose X Time: P < .005), eye-hand coordination (Dose X Time: P < .001), DSST [number completed, Dose X Time: P < .001; number correct, Dose X Time: P < .001 (fig. 4)], auditory reaction time (Dose: P < .001) and logical reasoning (number completed, Dose X Time: P < .01; number correct, Dose: P < .05). Peak impairment occurred on the tests from 60 to 120 min postinjection, and with the higher doses, the impairment was evident for most or all of the postinjection time points (e.g., see fig. 4). Figure 4 also shows performance from the 10-mg morphine condition, and it is evident that morphine produced little if any impairment on the DSST. In contrast to the psychomotor impairment produced by buprenorphine, there were no differences between the three buprenorphine conditions and the saline condition on immediate or delayed free recall.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Time course of the effects of 0 (), 0.075 (), 0.15 () and 0.3 mg/70 kg () buprenorphine on numbers of symbols correctly drawn on the DSST. For comparative purposes to 0.3 mg buprenorphine, 10 mg morphine () is also shown on the graph. Each buprenorphine point is the mean across 16 subjects and each morphine point is the mean across 15 subjects. Time point 0 refers to effects measured immediately before the injection. Asterisks indicate that a buprenorphine dose is significantly different from saline at a given time point (Tukey post hoc test; P < .05). See figure 1 for further details.

Physiological effects. Buprenorphine. Significant effects were obtained on diastolic blood pressure (Dose X Time: P < .01), arterial oxygen saturation (Dose X Time: P < .05), respiration rate (Dose: P < .001) and miosis (Dose X Time: P < .001). Post hoc testing revealed that diastolic blood pressure was significantly lower at one postinjection time point in the 0.3-mg buprenorphine condition, relative to the saline condition (57.1 vs. 65.3 mm Hg, respectively). Arterial oxygen saturation rate dropped from approximately 100 to 98% at one postinjection time point in the 0.15-mg condition and at four postinjection time points in the 0.3-mg condition. Respiration rates in all three active drug conditions differed from the saline condition (mean rates of 14.9, 13.1, 13.3 and 12.6 breaths/min in the saline, 0.075-, 0.15- and 0.3-mg conditions, respectively). Buprenorphine decreased pupil size in a dose-related fashion, and pupil constriction was still evident 300-min postinjection with all three doses (fig. 5). Ten mg morphine had similar miotic effects in terms of time course and magnitude of effect to that of the two lower doses of buprenorphine.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Time course of the effects of 0 (), 0.075 (), 0.15 () and 0.3 mg/70 kg () buprenorphine on pupil size of the right eye. For comparative purposes to 0.3 mg buprenorphine, 10 mg morphine () is also shown on the graph. Each buprenorphine point is the mean across 16 subjects and each morphine point is the mean across 15 subjects. Time point 0 refers to effects measured immediately before the injection. All active drug doses were significantly different from saline at all postinjection time points (Tukey post hoc test; P <.05).

Adverse effects. A number of subjects vomited during and/or after sessions in which buprenorphine was administered. These data were obtained via either technician observation or ratings on the adjective checklists. Three (A.G., M.S., R.S.), three (J.Y., R.S., S.R.) and seven (A.G., D.M., G.S., J.Y., M.S., P.B., R.S.) subjects vomited during sessions in which 0.075, 0.15 and 0.3 mg buprenorphine were administered. In contrast, no subjects vomited during sessions in which 10 mg morphine was administered. Three (B.F., J.Y., P.B.), three (C.D., P.B., S.R.) and three (A.G., J.Y., P.B.) subjects on the postsession questionnaire reported vomiting within 24 hr after the end of the sessions in which 0.075, 0.15 and 0.3 mg buprenorphine were administered. One subject (J.Y.) reported vomiting within 24 hr after the end of the sessions in which 10 mg morphine were administered. There did not appear to be a relationship between previous exposure to opioids and likelihood of vomiting.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Buprenorphine at a dose range of 0.75 to 3 mg (i.v.) in nondrug abusing volunteers had dose-related effects on mood, psychomotor performance and physiological effects. The duration of effect as measured by a number of different variables was protracted. The high dose of buprenorphine produced a larger magnitude of effect on mood, psychomotor performance and pupil constriction than an equianalgesic dose of morphine (10 mg). Qualitative differences between 0.3 mg buprenorphine and 10 mg morphine were on measures of psychomotor impairment and nausea with buprenorphine, but not morphine, producing these effects. Vomiting was another measure separating buprenorphine from morphine: 44% of the subjects vomited during the session in which 0.3 mg buprenorphine was administered, as opposed to no subjects vomiting during the session when 10 mg morphine was given.

The profile of subjective effects of buprenorphine that emerged from this study indicates that the drug produces somnolence and difficulty in concentrating and dizziness. These effects were dose-related in terms of both size and duration of effect. Other studies using nondrug abusing volunteers have noted somnolence (Blom et al., 1987; Manner et al., 1987; Saarialho-Kere et al., 1987; MacDonald et al., 1989) and this drug effect is a primary side effect often mentioned in patient studies (cf., Heel et al., 1979). The profile of subjective effects in our study differs somewhat from that obtained with opioid abusers. In a number of studies examining responses in opioid abusers, marked sedation is not reported by subjects (Jasinski et al., 1989; Preston et al., 1989a; Preston et al., 1992; Preston and Bigelow, 1994). The difference in drowsiness levels between our subjects and opioid abusers suggests that some sort of acclimation or tolerance develops to this effect of buprenorphine, and in fact Mello and associates have documented a decline in sedation as a function of buprenorphine exposure in abusers (Mello et al., 1982). Also, in a nondependent, opioid-abusing population, a positive spectrum of subjective effects is obtained, as measured by increases in MBG scores, and ratings of "good drug effects" and "drug liking." In our study, although peak liking ratings were significantly higher in the 0.3-mg buprenorphine condition than in the saline-placebo condition, this report of liking tended to occur only briefly, and soon after drug administration. By the end of the session, ratings were generally indicative of dislike of buprenorphine (0.3 mg) effects. There was no increase in MBG scores or VAS ratings of "good effects," which are prototypic characteristics of buprenorphine effects in an opioid-abusing population. The different degree of euphorigenic effects of opioids as a function of drug history has been noted before by other investigators (Lasagna et al., 1955; Azorlosa et al., 1994).

Psychomotor performance as measured by a number of different indices was impaired by buprenorphine in a dose-related fashion. This result is concordant with other studies that have examined impairing effects of single doses of buprenorphine. The impairment in our study is not thought to be due to the dizziness (and perhaps accompanying visual distortions) produced by buprenorphine, because auditory reaction time which does not require the visual sense was also impaired by the drug. In fact, the slower reaction time indicates that the general impairment produced by buprenorphine was due to a decrease in speed of responding. This is consistent with a large body of literature that has shown that opioids tend to slow down performance without affecting accuracy of responding (cf., Zacny, 1995). In fact, the one test not as heavily dependent on time, the memory test, remained unaffected by buprenorphine. The psychomotor performance decrements obtained in our study with buprenorphine stands in contrast to the large number of studies conducted with opioid abusers in which psychomotor performance is not affected by this drug. Again, this is suggestive of some cross-adaptation or cross-tolerance process occurring in the opioid-abusing population.

Physiological variables affected by buprenorphine in this study were respiration rate and pupil size. Respiration rate decreases have been obtained in other studies in both non-opioid and opioid-abusing populations (e.g., Gal, 1989; Walsh et al., 1994, 1995). The magnitude and duration of miosis noted in the present study is also concordant with what has been found in studies using opioid abusers (e.g., Jasinski et al., 1978; Preston and Bigelow, 1994). Finally, the relatively high incidence of nausea and vomiting observed in this study is not surprising given the incidence of postoperative nausea and vomiting after buprenorphine administration in clinical populations (e.g. Ellis et al., 1982; Maunuksela et al., 1988; Fullerton et al., 1991; Juhlin-Dannfelt et al., 1995).

When taking into account all of the effects of buprenorphine in this study, it would appear intravenous buprenorphine at clinically relevant doses has low abuse liability in a population of nonopioid users. Whether other preparations of buprenorphine (sublingual) would share the same abuse potential cannot be determined from our study. The abuse liability of buprenorphine in opioid abusers is thought to be substantial, given the studies conducted in nondependent opioid users. And in fact, there are numerous reports of abuse of parenteral buprenorphine (Chowdhury and Chowdhury, 1990; Singh et al., 1992; Robinson et al., 1993; San et al., 1993; Torrens et al., 1993). One potential solution that is being explored to decrease the abuse liability of buprenorphine is adding a small dose of an opiate antagonist (naloxone) to that of buprenorphine. This combination, when used parenterally, results in a diminution of agonist effects in nondependent opioid users (Weinhold et al., 1992) and can precipitate withdrawal in dependent opioid users (Preston et al., 1988; Mendelson et al., 1996).

Other studies examining buprenorphine have observed ceiling effects on such measures as respiratory depression, mood, and analgesia (e.g., Walsh et al., 1994, 1995; Walker et al., 1995). These ceiling effects are consistent with the partial agonist properties of the drug. We found no evidence of a ceiling effect on respiratory depression and mood, but we tested lower doses than the human psychopharmacology studies cited above, which tested sublingual doses up to 32 mg (Walsh et al., 1994, 1995) and i.v. doses up to 1.2 mg (Pickworth et al., 1993). Other human psychopharmacology studies that have tested doses similar to those tested in the present study have obtained orderly changes in effect as a function of dose (e.g., Preston et al., 1992; Preston and Bigelow, 1994).

One purpose of our study was to systematically characterize buprenorphine effects across a range of doses. Another purpose of the study was to compare and contrast one dose (0.3 mg) of buprenorphine, a putative partial agonist, to that of one dose (10 mg) of morphine, a full mu agonist. The aim of comparing these opioids is to determine if fundamental differences emerge in the subjective, psychomotor or physiological effects, as a function of opioid subclass (partial vs. full agonist). One shortcoming of our study is that a range of morphine doses was not examined, so that relative comparisons are limited to only 0.3 mg buprenorphine and 10 mg morphine. Another potentially serious shortcoming is that we may have picked an inappropriate morphine dose to compare with buprenorphine. Buprenorphine has been estimated to be 25 to 40 times (Martin et al., 1976; Cowan et al., 1977b) or 25 to 50 times more potent than morphine (Jasinski et al., 1978; Reisine and Pasternak, 1996). If the potency difference is closer to 50-fold than to 25-fold, then we should have used a dose of morphine of 15 mg (50 times potency difference) rather than 10 mg (33 times potency difference). Using a higher dose of morphine may have shown that there was a similar magnitude of subjective effects and psychomotor impairment between the partial agonist and full mu agonist. Although we acknowledge this possibility, we must emphasize that in the overwhelming majority of clinical studies and reviews (e.g., Heel et al., 1979; Wang et al., 1981; Lewis, 1985; Wallenstein et al., 1986; Bushnell and Justins, 1993) pharmacology (e.g., Reisine and Pasternak, 1996) and pain management textbooks (e.g., Mather, 1994; Noren, 1994), as well as the Buprenex (Reckitt & Colman Pharmaceuticals, Inc.) package insert, 0.3 mg and sometimes even 0.4 mg parenteral buprenorphine is listed as equianalgesic to 10 mg parenteral morphine, a potency difference of no more than 33. It should also be pointed out that in one study that indeed detected a potency ratio of 50:1, the endpoint was not analgesia. Jasinski et al. (1978) obtained in opiate abusers a potency ratio of 50:1 on nonanalgesic measures, including pupil size and SDQ Opiate Signs (analgesia was not measured in the study). What we found rather consistently across our different measures was a much larger effect of 0.3 buprenorphine than 10 mg morphine (e.g., see figures and tables 2-5). A number of subjective effects, including measures of drowsiness and dizziness, were much greater after administration of 0.3 mg buprenorphine than 10 mg morphine. This greater degree of subjective effects has a corollary to some clinical studies that have compared morphine and buprenorphine, and found a greater incidence of drowsiness (van den Berg et al., 1994) and dizziness (Kjaer et al., 1982) with buprenorphine. A second major difference, and one that has been documented in numerous other human studies (e.g., Dobkin et al., 1977; Jasinski et al., 1978), was the longer duration of effect of 0.3 mg buprenorphine, relative to 10 mg morphine. This long duration of effect appears to be due to the drug's slow dissociation from the opiate receptor (Boas and Villiger, 1985; Hambrook and Rance, 1976). A third major difference was a greater incidence of nausea/vomiting associated with buprenorphine than with morphine, which has also been documented in clinical studies (Ellis et al., 1982; Kjaer et al., 1982; Maunuksela et al., 1988; van den Berg et al., 1994). A fourth difference between 0.3 mg buprenorphine and 10 mg morphine was that the former drug produced substantial psychomotor impairment. In other studies conducted in our laboratory, morphine has produced a slight degree of impairment (as measured by the DSST) (Zacny et al., 1994a) or no impairment (Zacny et al., 1994b, 1997). Whether a causal relationship exists between the greater degree of somnolence and greater degree of impairment produced by buprenorphine is unknown, although such a relationship would appear to make intuitive sense.

In conclusion, buprenorphine had clear dose-related effects on mood, psychomotor performance, and miosis. These results with nondrug-abusers show some similarities (soporific effects) but also some dissimilarities (marked psychomotor impairment, no increase in MBG scores or drug liking ratings) to results obtained in a population of opiate abusers. In addition, morphine, at a dose we considered to be equianalgesic to the highest dose of buprenorphine tested, had a lesser magnitude of subjective and psychomotor-impairing effects than that of buprenorphine. Future studies testing higher doses of morphine would be useful, if in fact the equianalgesic potency of buprenorphine is more than the 33 times that we estimated it to be from the clinical literatureif the potency is more than 33:1, then testing a higher dose of morphine may indeed establish a similar magnitude of effect as that of buprenorphine.