Addiction (2001) 96, 823–834
RESEARCH REPORT
Examining the limits of the buprenorphine
interdosing interval: daily, every-third-day
and every-Ž fth-day dosing regimens
NANCY M. PETRY,1 WARREN K. BICKEL & GARY J. BADGER2
University of Vermont, Departments of Psychiatry, Psychology and 2Medical Biostatistics,
Substance Abuse Treatment Center, 1 South Prospect Street, Burlington, VT, USA
Abstract
Aims. Opioid-dependent outpatients may be more likely to present for pharmacological treatment if less
than daily dosing can be arranged. These studies compared opioid withdrawal symptoms during 24-, 72-,
and 120-hour buprenorphine dosing regimens and evaluated participants’ preferences for these different
dosing regimens. Participants. Thirty-three opioid-dependent participants received daily sublingual
maintenance doses of 4 mg/70 kg (n 5 14) or 8 mg/70 kg (n 5 19) of liquid buprenorphine. Methods. In
Study I participants received, in a random order, three dosing regimens for Ž ve repetitions of each: daily
maintenance doses every 24 hours (4 or 8 mg/70 kg), triple the daily maintenance dose every 72 hours (12
or 24 mg/70 kg) and quintuple the daily maintenance dose every 120 hours (20 or 40 mg/70 kg). Doses
were administered under double-blind procedures, and placebos were administered on the interposed days
during the latter two regimens. Subjective and observer ratings of opioid withdrawal symptoms were assessed
daily prior to receipt of each dose. In Study II, a new group of participants received each of the three dosing
regimens under open-dosing procedures and then chose between the different dosing regimens. Findings.
Opioid withdrawal symptoms increased signiŽ cantly during the every-Ž fth-day dosing regimen in both the
blind- and open-dosing studies. In the choice phase of Study II, only one participant (7%) chose
quintuple-every-Ž fth-day dosing over all other dosing options. Conclusions. These results suggest that the
maximum duration of action of buprenorphine is less than 5 days when Ž ve times the daily maintenance
dose is provided.
Introduction
Buprenorphine (BUP) is a high afŽ nity partial
mu-opioid agonist with a long duration of action
(Rance & Dickens, 1978; Bickel & Amass,
1995). BUP is comparable to methadone in retaining opioid-dependent patients in mainte-
nance and detoxiŽ cation trials (Bickel et al.,
1988a; Johnson, Jaffe & Fudala, 1992; Ling,
1996, but see Kosten et al., 1993).
BUP also has some advantages over methadone. Notably, BUP has a ceiling level on agonist activity limiting adverse reactions, such as
Correspondence to: Nancy M. Petry PhD, University of Connecticut Health Center, Department of Psychiatry,
263 Farmington Avenue, Farmington, CT 06030-3944, USA. Tel: 1 860 679 2593; fax: 1 860 679 1312; e-mail:
[email protected]
Submitted 22nd August 2000; initial review completed 24th October 2000; Ž nal version accepted 12th January
2001.
ISSN 0965–2140 print/ISSN 1360–0443 online/01/060823–12 Ó
Carfax Publishing, Taylor & Francis Limited
DOI: 10.1080/09652140020050942
Society for the Study of Addiction to Alcohol and Other Drugs
824
Nancy M. Petry et al.
respiratory depression, at very high doses (Jasinski, 1977; Bickel et al., 1988b; Lange et al.,
1990; Walsh et al., 1995). Because large doses
do not result in increased opioid effects and the
drug dissociates slowly from the opiate receptor,
BUP can be administered at high doses and
relatively long inter-dosing intervals. These
properties may render BUP beneŽ cial over
methadone, which requires daily administration.
In studies using double-blind procedures, we
have shown that increasing the liquid-based
BUP dose to three or four times the maintenance
dose (up to 32 mg/70 kg) prevents clinically
signiŽ cant withdrawal symptoms for a 96-hour
period (Amass et al., 1994; Petry, Bickel & Badger, 1999; Bickel et al., 1999). Acceptability of
every-fourth-day dosing regimens was also
demonstrated. When given the options of daily
dosing, double doses followed by a day off from
the clinic, triple doses followed by two days off,
or quadruple doses followed by three days off,
46% of participants chose to be dosed every
fourth day (Petry, Bickel & Badger, 2000).
Given the acceptability of this every-fourthday dosing regimen, the present studies sought
to evaluate the effects of extending the interdosing interval to every Ž fth day. In Study I, we
administered quintuple the daily maintenance
dose every 5 days and compared this dosing
regimen to daily dosing as well as to triple doses
every three days. In this study, all doses were
provided under double-blind procedures, and
opioid withdrawal symptoms were rated daily by
participants and observers. In Study II, we provided these same three dosing regimens under
open-dosing procedures and then evaluated participants’ preferences for them.
Study 1: Methods
Participant
Twenty-Ž ve opioid-dependent outpatients (17
male, eight female) enrolled in the study, and 18
completed it (11 male, seven female). Inclusion
criteria were $ 18 years old, good health and
DSM-III-R criteria for opioid dependence. Pregnancy and evidence of active psychosis, manicdepressive illness or serious medical illness (liver
or cardiovascular disease) were exclusion criteria. The study was approved by the Institutional Review Board, and participants provided
written informed consent.
Participants’ mean age was 35 years (range
22–51), and mean weight was 82 kg (range 50–
112). On average, participants reported using
opioids for 12 years (range 1–31) and spending
$360 (range $0–$1,500) per week on opioids.
Ninety-Ž ve per cent were intravenous drug users.
Weekly pregnancy tests were conducted for females and were negative throughout the study.
General procedures
Study continuation and compensation ($50/
week) were contingent on attendance for all
doses and opioid abstinence. Participants attended the clinic at the same time each day to
facilitate appropriate dosing schedules. Opioid
abstinence was required to ensure that changes
in subjective effects were related to dosing regimens provided and not resultant from other opioid use. Participants received one 60-minute
counseling session weekly.
Urine specimen monitoring
Opioid abstinence was conŽ rmed via urinalysis
testing conducted thrice weekly, prior to medication administration. Urine samples were collected under staff observation and analyzed
onsite for opioids (methadone, propoxyphene
and opioids) via Enzyme Multiplied Immunoassay Technique (Syva Corp., San Jose,
CA, USA). Barbiturate, benzodiazepine, marijuana and cocaine use was assessed once weekly
on a random basis using this testing technique.
When participants submitted an opioidpositive sample no medication was provided that
day, and they returned to the clinic the next day
for dosing. Data collected in the days prior to an
opioid-positive sample (range 1–3 days) were
omitted from analyses. Participants continued in
the study provided the subsequent result indicated no further opioid use (Petry et al., 1999).
Study participation was terminated upon receipt
of a second positive sample.
Medication administration
All BUP doses were masked for taste with peppermint and 1 ml of Bitrex granules (6 l g/ml;
Macfarlane-Smith Ltd., Edinburgh, UK). Participants swished this peppermint solution in their
mouth for 30 seconds and discarded it before
receiving medication.
BUP hydrochloride (Reckitt & Colman, Hull,
Limits of buprenorphine dosing interval
England, UK) was prepared as a stock concentration of 16 mg/ml in 35% ethanol (vol/vol).
Stock solutions containing 2, 4 and 8 mg/ml in
35% ethanol (vol/vol) were prepared from serial
dilutions of the 16 mg/ml stock. The maximum
volume necessary for the highest dose was calculated on an individual basis based on weight, and
doses were delivered in a constant volume
throughout the study. Due to the high doses
provided, participants received 2–3 sequential
drug administrations daily, and held each dose
under their tongue for 5 minutes. Drug was
administered sublingually with a Ped-Pod Oral
Dispenser (SoloPak Laboratories, Franklin Park,
IL, USA). Placebo was ethanol vehicle.
Participant’s maintenance doses were determined during the Ž rst week of study participation. Participants initially were placed on a
2 mg/70 kg dose at intake, and on a 4 mg/70 kg
dose on day 2. If objective withdrawal symptoms
(see below) were evident on days 3–7, the dose
was increased to 8 mg/70 kg. In total, eight study
completers received the 4 mg/70 kg daily maintenance dose, and 10 received the 8 mg/70 kg daily
maintenance dose.
Laboratory safety session
After dose stabilization, the daily maintenance
dose was administered for 10 days. Participants
then participated in a 7-hour laboratory safety
session (see Petry et al., 1999 for details) in
which they were exposed to Ž ve times their daily
maintenance dose of BUP (20 or 40 mg/70 kg)
in a controlled setting to ensure safety of and
tolerance to this dose. No participants experienced adverse opioid agonist effects.
Exposure phase
For 4 days after the safety session, participants
received their daily maintenance dose under
double-blind procedures. On days 5–7, participants again received a maintenance dose daily,
but for these 3 days the dose was provided under
open-dosing procedures, and participants were
told they were receiving their maintenance dose
at dispensation. These 3 days were included to
allow participants to experience how they felt
when they knew they were receiving their
maintenance dose. Following this open-dosing
exposure, they received daily maintenance doses
for 10 days under double-blind procedures.
825
Experimental conditions
After the baseline phase described above, exposure to the three dosing regimens commenced.
Participants received the regimens in a random
order: maintenance dose daily, triple maintenance dose every third day and quintuple
maintenance dose every Ž fth day. Each regimen
was kept in effect for Ž ve repetitions, such that
the daily regimen was in effect for 5 days, the
triple-every-third-day regimen for 15 days and
the quintuple-every-Ž fth-day regimen for 25
days. After receiving all three regimens, participants received daily maintenance doses for 5
days. Following study completion, they could
participate in other studies or a detoxiŽ cation.
Outcome measures
Observer-ratings. Research staff (blind to the
treatment conditions) completed an observerrating scale before dispensing medication. The
staff rated participants on a scale of 0 (not at all)
to 9 (severe) on signs of opioid withdrawal
(goose esh, sweating, restlessness, tremor,
lacrimation, nasal congestion and yawning) and
agonist activity (skin itching, vomiting, sedation,
nodding and soap-boxing/nodding). The observer-rating scale was based on Addiction Research Center withdrawal scales (Himmelsbach,
1939; Jasinski, Pevnick & GrifŽ th, 1978).
Participant ratings. Each day before receiving
medication, participants completed computerized versions of the adjective-rating scale (ARS),
visual-analog scales (VAS) and the Addiction
Research Center Inventory (ARCI) short form.
Participants were instructed to respond to questions according to their experiences over the past
24 hours.
The ARS (Bickel et al., 1988a) consists of 32
opioid withdrawal and drug symptoms. Participants rated each item according to a 0 (none) to
9 (severe) scale, and means across all items are
shown. Items in the opioid withdrawal scale
were: muscle cramps, depressed/sad, painful
joints, yawning, hot/cold  ashes, trouble getting
to sleep, sick stomach, runny nose, poor appetite, weak knees, abdominal cramps, tense/jittery,
excessive sneezing, irritable, watery eyes and
Ž tful sleep. Items in the opioid effects scale were:
drug effect, loaded/high, rush,  ushing, sweating, nodding, dry mouth, stomach turning, itchy
skin, relaxed, coasting/spaced out, talkative,
826
Nancy M. Petry et al.
pleasant sickness, drive, nervousness and
drunken.
The VAS (Preston, Bigelow & Liebson, 1988;
Petry et al., 1998) items were: High, Drug,
Good, Like, Bad and Sick. Participants made a
mark along a 100-mm line, anchored at each end
by “not at all” and “severe”. Responses were
converted to a 100-point scale.
The ARCI short form (Martin et al., 1971;
Jasinski, 1977) consisted of 49 true/false items.
Five scales assessed sedation, dysphoria, euphoria and stimulation: pentobarbitol, chlorpromazine, alcohol group (PCAG), lysergic acid
diethylamide (LSD), morphine benzedrine
group (MBG), benzedrine group (BG) and amphetamine (A).
A dose identiŽ cation measure was also used,
in which participants indicated whether they received placebo or a maintenance, triple, or
quintuple dose the prior day. Data were scored
as identiŽ cation of placebo or higher than the
maintenance dose. Pupil diameter was determined before dispensing medication from photographs taken with a Polaroid camera at
2 3 magniŽ cation at 1 ft-c of ambient illumination (Jasinski & Martin, 1967).
Data analyses
Data obtained during the triple and quintuple
dosing regimens were partitioned into individual
components for analyses, e.g. hours since last
active dose. In total, nine conditions were obtained: 24-hour post-maintenance doses, 24hour post-triple doses, 48-hour post-triple doses,
72-hour post-triple doses, 24-hour post-quintuple doses, 48-hour post-quintuple doses,
72-hour post-quintuple doses, 96-hour postquintuple doses and 120-hour post-quintuple
doses. For the study completers, repeated
measures analyses of variance were conducted
with one within-subject factor (condition, with
nine levels) and one between-subjects factor
(maintenance dose, with two levels—4 or 8 mg/
70 kg). When the overall F test corresponding to
conditions was signiŽ cant, Dunnett’s procedure
was used to determine whether measures obtained during the eight experimental conditions
differed from the 24-hour post-maintenance
dose condition. Data from participants who
withdrew or were discharged from the study are
not included in these analyses because they did
not complete all three dosing regimens. Analyses
were performed using SAS, and statistical
signiŽ cance was determined using a 5 0.05, twotailed.
Descriptive data are presented for percentage
of urine samples that were positive for opioids
and other drugs across the dosing regimens.
Because urine samples were obtained thrice
weekly, we cannot assess the exact day of drug
use. Therefore, data are shown across entire
dosing regimens (% positive during maintenance, triple, and quintuple dosing regimens).
Because submission of opioid-positive samples
resulted in study termination, urinalyses data are
presented from all participants enrolled, including non-completers.
Study 1: Results
Retention
Of the 25 participants enrolled in this study,
72% (n 5 18) completed it. Two participants
requested detoxiŽ cation because they had
difŽ culties with transportation or daily attendance. One stopped attending the program while
receiving quintuple doses. Another failed to attend the clinic during baseline and again during
the quintuple dosing regimen, and another during the maintenance as well as during the
quintuple dosing regimens. One participant was
discharged for submitting opioid-positive samples during the maintenance and quintuple dosing regimens, and the Ž nal participant was
discharged for Ž ghting in the clinic. Data from
one participant who completed the study (a female on 8 mg/70 kg for a maintenance dose) are
not included in the analyses because computer
malfunction resulted in substantial missing data.
Participant and observer ratings of drug effects
SigniŽ cant effects of BUP dose (4 vs. 8 mg/
70 mg) were noted on only one of the 18 outcome variables—subjective ratings of withdrawal
on the ARS, F(1,15) 5 6.95, p 5 0.02. SigniŽ cant BUP dose by condition effects were not
found on any of the outcome data. Therefore,
Table 1 shows means and SEMs on the outcome
measures collapsed across BUP maintenance
doses (4 and 8 mg/70 kg).
SigniŽ cant differences were noted across the
conditions on the majority of outcome variables
including: ARS subjective reports of withdrawal
[F(8,120) 5 8.49,
p , 0.001];
the
High
6.0
(6)
(9)
(0.7)
(0.5)
(1.0)
(0.7)
(0.7)
(3.0)
(3.8)
(3.9)
(5.2)
(5.4)
(5.5)
5.7* (0.2)
17
41
5.2
5.0
3.6
4.6
2.6
7.9
14.6
15.4
23.9
20.4
26.5
2.75 (0.64)
0.71 (0.22)
0.24 (0.05)
0.00 (0.00)
24 hours
(9)
(6)
(0.7)
(0.6)
(0.9)
(0.6)
(0.6)
(2.6)
(4.1)
(3.4)
(5.4)
(5.9)
(6.6)
6.1 (0.2)
73*
13
7.1
5.8
2.6
3.4
1.8
5.7
11.2
9.4
20.9
23.7
36.7
3.40 (0.72)
0.53 (0.21)
0.28 (0.06)
0.00 (0.00)
48 hours
Triple
(9)
(6)
(0.9)
(0.7)
(0.9)
(0.7)
(0.6)
(2.9)
(4.2)
(3.9)
(5.7)
(6.9)
(6.6)
6.1 (0.2)
73*
8
7.9
6.3
2.5
3.2
1.8
6.1
9.2
9.3
19.0
28.5
39.1
4.28 (0.77)
0.51 (0.18)
0.30 (0.06)
0.00 (0.00)
72 hours
*SigniŽ cantly different from maintenance dose (p , 0.05) using Dunnett’s procedure.
(ml)
(0.2)
(0.8)
(0.5)
(1.1)
(0.8)
(0.8)
Pupil diameter
6.6
5.5
3.1
4.2
2.3
(3.1)
(3.5)
(4.0)
(5.2)
(5.9)
(6.6)
(8)
(7)
15
13
14
16
11
ARCI
PCAG
LSD
MBG
BG
A
8.1
11.1
13.6
23.4
18.3
30.4
3.32 (0.66)
0.61 (0.18)
0.23 (0.05)
0.00 (0.00)
24 hours
Maintenance
Dose ID (mean %guessed)
Placebo
100 33
. Maintenance 100 24
100
100
100
100
100
100
9
9
ARS
Withdrawal
Drug/agonist
VAS
High
Drug
Good
Like
Bad
Sick
9
9
Observer
Withdrawal
Agonist
Measure
Max.
poss.
(4)
(8)
(0.7)
(0.4)
(0.9)
(0.5)
(0.6)
(3.7)
(3.9)
(3.9)
(4.5)
(5.0)
(6.1)
5.4* (0.2)
13
61*
5.6
4.5
4.2
4.9
2.7
13.6*
20.0
18.5
27.3
19.6
26.2
2.22* (0.52)
0.87 (0.20)
0.18 (0.03)
0.02 (0.02)
24 hours
(7)
(7)
(0.8)
(0.6)
(0.8)
(0.6)
(0.5)
(3.3)
(3.7)
(3.6)
(4.5)
(6.3)
(6.0)
6.0 (0.2)
74*
15
7.0
5.8
2.6
3.9
2.1
8.0
10.7
10.6
18.4
24.1
32.3
3.38 (0.70)
0.58 (0.20)
0.18 (0.04)
0.00 (0.00)
48 hours
(8)
(6)
(0.8)
(0.6)
(0.7)
(0.5)
(0.5)
(3.1)
(3.9)
(3.8)
(5.8)
(6.0)
(6.7)
6.0 (0.2)
79*
12
8.0
6.3
2.1
3.2
1.8
6.8
10.2
8.8
18.5
27.0
39.3
3.74 (0.70)
0.57 (0.19)
0.20 (0.04)
0.00 (0.00)
72 hours
Quintuple
(0.05)
(0.00)
6.2
74*
16
8.8*
7.0*
2.1
2.6*
1.6
7.6
13.2
7.3
17.5
30.4*
48.9*
(0.2)
(10)
(7)
(0.8)
(0.6)
(0.7)
(0.5)
(0.5)
(3.9)
(6.3)
(3.4)
(4.8)
(7.5)
(7.9)
4.35* (0.75)
0.58 (0.18)
0.26
0.00
96 hours
Table 1. Means and SEMS (in parentheses) of outcome data across the different dosing conditions in Study 1; hours refer to post-active doses
(0.05)
(0.00)
6.3*
73*
17
8.6*
6.2
1.8
2.9
1.3*
5.9
13.6
8.1
24.3
28.5
51.1*
(0.2)
(10)
(8)
(0.9)
(0.5)
(0.5)
(0.4)
(0.4)
(2.8)
(6.5)
(4.1)
(6.6)
(6.5)
(7.6)
4.69* (0.77)
0.83 (0.24)
0.25
0.00
120 hours
Limits of buprenorphine dosing interval
827
828
Nancy M. Petry et al.
Table 2. Percentage of urine samples positive for opioids, barbiturates, benzodiazepines, THC, and cocaine across the three
dosing regimens during Study I (blind dosing study)
Maintenance doses daily
Triple doses
every 3 days
Quintuple doses every 5 days
Opioids
Barbiturates
Benzodiazepines
THC
Cocaine
2%
5%
0%
0%
38%
57%
75%
74%
0%
7%
5%
0%
41%
67%
13%
[F(8,120) 5 3.63, p , 0.001], Good [F(8,120) 5
4.93,
p , 0.001],
Bad
[F(8,120) 5 2.22,
p , 0.05] and Sick [F(8,120) 5 5.99, p , 0.001]
scales of the VAS; and the PCAG
[F(8,120) 5 6.69, p , 0.001], LSD [F(8,120) 5
5.43, p , 0.001], MBG [F(8,120) 5 3.95,
p , 0.001], BG [F(8,120) 5 5.11, p , 0.001] and
A [F(8,120) 5 3.09, p , 0.01] scales of the
ARCI.
For outcome measures in which signiŽ cant
overall differences were found across the conditions, Dunnett’s procedure compared data obtained 24 hours following maintenance doses to
that obtained at each of the other eight conditions. These results are denoted by asterisks and
bold text in Table 1. Some measures of opioid
withdrawal rose as hours since the last active
dose increased (ARS withdrawal scores, and Bad
and Sick scores on the VAS). Sedation and
dysphoria scores on the ARCI also rose in the
96–120 hours following a quintuple dose compared to 24 hours after maintenance dosing.
Some drug-like effects decreased as hours since
the active doses increased (e.g. reductions in BG
and A scales of the ARCI). In addition, scores on
the VAS High scale rose, and scores on the ARS
withdrawal scale decreased, 24 hours after
quintuple compared to 24 hours after maintenance doses.
The percentage of times participants guessed
they had received a placebo [F(8,120) 5 25.46,
p , 0.001] or greater than the maintenance dose
[F(8,120) 5 13.91,
p , 0.001]
differed
signiŽ cantly across conditions. Participants
could correctly identify placebo doses, and they
could recognize quintuple doses to be higher
than the maintenance dose. Pupil diameter also
differed
signiŽ cantly
across
conditions
[F(8,120) 5 10.78, p , 0.001]. Compared to 24hour post-maintenance doses, pupil size decreased signiŽ cantly 24 hours after triple and
quintuple doses, and pupil size increased
signiŽ cantly 120 hours after a quintuple dose.
Urinalysis testing
Urinalysis testing revealed that 2% of urine samples were positive for opioids during the maintenance dosing regimen, 5% during the triple
dosing regimen, and 5% during the quintuple
dosing regimen. The percentage of urine samples
testing positive for other drugs also are shown in
Table 2; they did not vary systematically across
conditions.
Study 1: Discussion
Two main Ž ndings emerged from this study.
First, increases in opioid agonist effects following
administration of quintuple the maintenance
dose were noted on only one outcome measure.
On the “High” scale of the VAS, responses
increased from an average of 8 during daily
maintenance dosing to a mean of 14 in the 24
hours after a quintuple dose. Although this increase was statistically signiŽ cant, it may not be
clinically meaningful. The VAS ranks subjective
reports along a 100-point scale and therefore a
score of 14 is still low. While higher agonist
ratings may have occurred had we measured
subjective reports 1–2 hours (rather than 24
hours) post-quintuple dosing, no objective or
subjective indicators of opioid agonist activity
were noted when participants received Ž ve times
their maintenance doses during the laboratory
safety session. This evidence, in conjunction
with the fact that signiŽ cant increases were not
noted on any other subjective indices of opioid
agonist activity following quintuple doses, suggests that opioid agonist activity was minimal
following quintuple doses. On the ARS withdrawal scale, signiŽ cant decreases in withdrawal
were noted 24 hours following quintuple doses
compared with 24 hours following maintenance
doses. Thus, rather than inducing opioid-like
effects, larger BUP doses may abate chronic
feelings of withdrawal.
Secondly, signiŽ cant increases in withdrawal
Limits of buprenorphine dosing interval
symptoms emerged 96 to 120 hours following
administration of quintuple doses. SigniŽ cant
differences across dosing conditions were found
on 10 of the 13 subjective report measures, with
Dunnett’s procedures conŽ rming that many of
these effects were related to increased withdrawal symptoms 96–120 hours after quintuple
doses. Even physiological manifestations of opioid withdrawal occurred, with pupil diameter
increasing signiŽ cantly 120 hours following
quintuple doses. In contrast, no increases in
opioid withdrawal were found up to 72 hours
after triple BUP doses.
These data replicate and extend our previous
Ž ndings regarding less than daily BUP dosing
regimens. In a double-blind study examining up
to every-fourth-day dosing regimens with
quadruple maintenance doses (Petry et al.,
1999), we found that subjective reports of opioid
withdrawal increased as time since last active
dose increased. However, when quadruple, every-fourth-day dosing was provided using opendosing procedures (Petry et al., 2000), signiŽ cant
increases in subjective reports of opioid withdrawal were noted on only two of the 13 subjective report measures. Nearly half the participants
selected quadruple-every-fourth-day dosing over
daily or any other less-than-daily dosing option.
Therefore, the purpose of Study II was to evaluate whether increased opioid withdrawal symptoms emerged when these dosing regimens were
administered under open and choice dosing conditions.
Study II: Methods
Participants
Seventeen patients (12 male and Ž ve female)
enrolled in this study, and 15 (11 male and four
female) completed it. The same inclusion criteria
were used, and participants had similar demographic characteristics as in Study 1.
General methods
The general methods were similar to those described in Study I, with only a few exceptions as
described below. Dose induction and stabilization procedures were the same as in Study I. In
total, six study completers received the 4 mg/
70 kg daily maintenance dose, and nine received
the 8 mg/70 kg maintenance dose. A laboratory
safety session was also provided, following which
829
participants received 10 days of daily maintenance dosing and were told each day that these
doses were their daily maintenance doses.
For the open-exposure phase of Study II participants received, in a random order, each of the
three dosing regimens: daily maintenance dosing, triple-every-third-day dosing, and quintupleevery-Ž fth-day dosing. The daily dosing regimen
was in effect for 5 days, and the other two
regimens were in effect for 3 repetitions of each
regimen.
Following the 29-day open-dosing exposure
phase, participants received daily maintenance
doses for 5 days. They then were presented with
choice conditions: (1) daily maintenance doses or
triple doses every third day; (2) daily maintenance doses or quintuple doses every Ž fth day;
and (3) triple doses every third day or quintuple
doses every Ž fth day. Each choice was kept in
effect for 15 days such that whatever dosing
regimen was selected was administered for the
next 15 days. The three choice conditions were
presented in a random order, and participants
were presented with each of the three choice
conditions throughout a 45-day period. At the
end of the study, participants received maintenance doses daily for 5 days and were then
offered the opportunity to participate in other
studies or receive a detoxiŽ cation.
Because participants did not attend the clinic
daily in Study II, some minor changes were
made to the urine specimen testing procedures.
Instead of Monday–Wednesday–Friday submission of specimens, participants provided samples prior to medication administration at least
two times a week, and often three times a week.
In addition, participants in Study II were also
telephoned at home on a random basis (at least
once for each participant) during every-Ž fth-day
dosing. When called for a random specimen,
participants were required to come to the clinic
within 24 hour and submit a urine sample. Failure to submit a specimen was considered a positive sample. Only one participant, who
subsequently withdrew from the study, failed to
submit a random specimen during the quintuple
dosing regimen.
Dependent measures for the open-dosing phase
of Study II were similar to those described in
Study I, with the exception that data were only
collected prior to the receipt of each dose: 24hour post-maintenance doses, 72-hour posttriple doses and 120-hour post-quintuple doses.
830
Nancy M. Petry et al.
Participants were instructed to report their experiences since the time of their last active dose.
Dose identiŽ cation and pupil diameter data were
not collected in Study II.
The dependent measure for the choice phase
was the percent of participants choosing the
dosing schedule requiring the least clinic attendance.
Data analysis
For the open-exposure phase in Study II, comparisons of subjective and observer ratings in the
three dosing regimens were performed using repeated measures analysis of variance corresponding to a three-period, within-subjects design. A
participant’s daily maintenance dose (4 or 8 mg/
70 kg) was considered as an additional factor.
Dunnett’s procedure was used to compare measures obtained during the daily maintenance
dosing regimen to each of the other less than
daily regimens when the overall F was
signiŽ cant.
For the choice phase of Study II, descriptive
data are shown for percentage of participants
who chose the dosing schedule requiring the
least attendance in the 3 conditions.
Study II: Results
Retention. Of the 17 enrolled participants, 88%
(n 5 15) completed the study. One participant
requested withdrawal after failing to submit a
random urine specimen during open-exposure to
quintuple doses, and a second was discharged
for submission of opioid-positive urine samples
during baseline dosing as well as during the
quintuple dosing regimen.
Participant and observer ratings of drug effects
The means obtained on the various outcome
measures are shown in Table 3. Repeatedmeasures ANOVA revealed signiŽ cant differences among the three dosing regimens on the
majority of indices of subjective and observerrated drug and withdrawal effects. SigniŽ cant
differences were noted in observer-rated withdrawal effects [F(2,26) 5 12.5, p , 0.001]; subjective ratings of withdrawal on the ARS
[F(2,26) 5 9.90, p , .001]; ratings on the High,
Drug, Good and Sick scales of the VAS
[F(2,26) 5 3.74, 5.75, 3.41 and 15.6, respectively, all ps , 0.05]; and the PCAG, BG, MBG
and LSD scales of the ARCI [F(2,26) 5 7.80,
4.31, 4.44, and 7.33, respectively, all ps , 0.05].
Table 3. Observer and subjective indicators of opioid agonist and withdrawal effects across the three open-dosing procedures
in Study II. Values represent means and standard errors
Measure
Observer
Withdrawal
Drug/agonist
ARS
Withdrawal
Drug/agonist
VAS
High
Drug
Good
Like
Bad
Sick
ARCI
PCAG
LSD
MBG
BG
A
Daily
maintenance
Triple, every third day
Quintuple, every Ž fth day
9
9
0.25 (0.06)
0.01 (0.01)
0.37 (0.09)
0.02 (0.01)
0.70* (0.10)
0.01 (0.01)
9
9
2.43 (0.45)
0.57 (0.20)
3.89* (0.71)
0.60 (0.21)
5.03* (0.68)
0.65 (0.20)
100
100
100
100
100
100
8.39 (4.22)
10.55 (4.45)
10.70 (4.15)
22.58 (6.94)
16.84 (5.67)
23.22 (4.93)
2.60 (1.31)
4.00* (2.10)
5.91 (3.66)
18.78 (6.02)
20.76 (7.94)
38.58* (7.94)
2.09 (1.33)
2.51* (1.30)
2.51* (1.36)
19.56 (6.59)
29.18 (9.36)
55.47* (6.77)
15
13
14
16
11
6.01 (1.05)
5.87 (0.77)
3.70 (1.19)
4.65 (0.77)
2.81 (0.80)
Max.
7.61
7.08
2.89
3.57
2.13
(1.10)
(0.72)
(0.92)
(0.74)
(0.70)
9.56* (1.06)
8.31* (0.77)
1.61* (0.58)
2.64* (0.69)
1.64 (0.60)
*Denotes values signiŽ cantly different from those obtained in the maintenance dosing condition, p , 0.05, using
Dunnett’s procedure.
Limits of buprenorphine dosing interval
831
Figure 1. Percentage of subjects choosing the dosing schedule with the least days of attendance.
On these variables, Dunnett’s procedure was
used to compare data obtained during the
maintenance dosing regimen to that obtained
during each of the other two dosing regimens.
These results are shown in Table 3. Generally,
these analyses indicated that withdrawal scores
(e.g. ARS withdrawal scores and VAS Sick
scores) increased during the less frequent dosing
regimens and drug-like effects (e.g. VAS Drug
and Good scores) decreased. On the ARCI,
scores of sedation and dysphoria (PCAG, LSD)
increased in the quintuple dosing regimen compared to the maintenance dosing regimen, while
scores of euphoria and stimulation (MBG, BG)
decreased in the less-frequent dosing regimen.
Choice data
Figure 1 shows the percentage of participants
choosing the dosing regimen with the least clinic
attendance across the three choice conditions.
When daily maintenance dosing was paired with
triple-every-third-day dosing, 67% of the 15 participants (n 5 10) chose to attend the clinic less
than daily. When daily maintenance dosing was
paired with quintuple-every-Ž fth-day dosing,
27% (n 5 4) chose to be dosed every Ž fth day.
Only one participant (7%) chose quintupleevery-Ž fth-day dosing over triple-every-third-day
dosing.
Urinalysis testing
During the open-dosing exposure phase, 2% of
urine samples were positive for opioids during
maintenance dosing, 1% during triple dosing
and 7% during quintuple dosing. During the
choice phase, 2% of the urine samples were
opioid positive when participants chose maintenance dosing, 1% were positive when participants chose triple dosing and 4% were positive
when participants chose quintuple dosing. Other
drug use across the open- and choice-dosing
phases of the study are shown in Table 4 as well.
Study II: Discussion
The results of Study II replicated those found in
Study I. During the quintuple dosing regimen,
observer ratings of opioid withdrawal increased
signiŽ cantly compared to daily maintenance dosing, and subjective reports changed signiŽ cantly
on nine of the 13 subjective outcome variables.
Moreover, in the choice phase of Study II, less
than one-third of the participants selected
quintuple, every-Ž fth-day dosing over daily
dosing, and less than 10% selected the everyŽ fth-day-dosing regimen over the every-thirdday-dosing regimen.
General discussion
In previous studies (Petry et al., 1999, 2000), we
found that BUP administered as infrequently as
every 4 days results in minimal withdrawal symptoms and that the majority of participants prefer
to be dosed according to less than daily dosing
regimens. The present studies were an extension
of these Ž ndings to examine whether the interdosing interval can be extended from 4 to 5 days.
The results suggest that quintuple, every-Ž fthday-dosing regimens are not well tolerated or
preferred among opioid-dependent outpatients.
832
Nancy M. Petry et al.
Table 4. Percentage of urine samples positive for opioids, barbiturates, benzodiazepines, THC and cocaine across the three
dosing regimens during the open and choice dosing phases in Study II
Opioids
Barbiturates
Benzodiazepines
THC
Cocaine
Open-dosing
Maintenance doses daily
Triple doses every 3 days
Quintuple doses
every 5 days
2%
1%
7%
1%
1%
3%
29%
53%
41%
71%
65%
64%
7%
18%
10%
Choice dosing
Maintenance doses daily
Triple doses every 3 days
Quintuple doses
every 5 days
2%
1%
4%
4%
4%
0%
30%
38%
8%
55%
53%
47%
3%
12%
0%
By 120 hours after a quintuple dose, signiŽ cant
increases in opioid withdrawal symptoms were
noted on a variety of outcome measures in both
the blind-dosing study as well as the open-dosing
study. Only one of 15 participants (7%) elected
to be dosed every-Ž fth-day rather than everythird day. Several issues that may bear upon the
interpretation of these Ž ndings are discussed.
Despite the increase in opioid withdrawal
symptoms noted during every-Ž fth-day dosing,
urine testing results were generally negative for
opioids, suggesting that participants were able to
tolerate this dosing regimen. Limitations of these
results, however, include that participants were
compensated to remain opioid abstinent. To
continue in these studies, participants had to
maintain opioid abstinence and be compliant
with a variety of clinic policies, including frequent and random urine testing procedures.
Compared to participants in this study, opioiddependent outpatients who are generally less
compliant may be even less likely to tolerate this
dosing regimen, and they may experience more
severe withdrawal symptoms and use more illicit
opioids when exposed to less frequent BUP dosing regimens.
In both studies, urine sample testing revealed
barbiturate and cocaine use was relatively low
across the dosing regimens. In contrast, marijuana and benzodiazepine use was prevalent in
these participants, and the use of these drugs
generally remained high across the dosing regimens. Although other drug use may have affected subjective responses on the outcome
measures, mandating abstinence from all drugs
of abuse is very difŽ cult among opioid-
dependent outpatients. Because other drug use
did not vary systematically across regimens, we
expect that it did not greatly in uence outcomes.
The only apparent change in drug use across
regimens was that benzodiazepines use was lower
(8% positive) when participants chose quintuple
every-Ž fth-day dosing regimens during the
choice phase of Study II, compared to rates of
benzodiazepine use during the other choice- or
open-dosing regimens (29–53% positive). Individual analysis of these data indicated that the
participants who frequently used benzodiazepines did not select the quintuple dosing regimen and, therefore, rates of benzodiazepine use
were artiŽ cially lower among those who selected
this dosing regimen.
In summary, data from these studies suggest
that these participants, while maintaining opioid
abstinence, experienced moderate to severe opioid withdrawal symptoms when the interdosing
interval was extended to 5 days. This increase in
withdrawal was noted whether doses were provided in blind- or open-dosing conditions. The
vast majority of participants chose not to be
dosed according to quintuple-every-Ž fth-day
dosing regimens. Therefore, twice weekly BUP
dosing regimens (e.g. quadruple doses on Mondays and triple doses on Fridays) may be the
maximal extension of interdosing intervals.
Although our previous studies (Petry et al.,
1999, 2000) suggest that twice weekly BUP dosing may be feasible, future research needs to
replicate these Ž ndings under non-laboratory
study conditions. For example, whether patients
maintained on higher daily maintenance doses of
BUP (e.g. 12 mg/day) also prefer every-fourth-
Limits of buprenorphine dosing interval
day dosing needs to be evaluated. Subsequent
studies may also examine the acceptance of twice
weekly dosing schedules when the BUP tablet
(or BUP–naloxone tablet) are provided, and
when doses are not in uenced by body weight. If
absorption of BUP is impaired when multiple
tablets are provided, then the duration of action
may be even shorter than reported in this study.
Because agonist effects were not noted in this
study or another in which sextuple the maintenance dose was provided (Gross et al., in press),
even higher doses may be indicated to reduce
withdrawal in some patients. Our studies generally provide brief exposure to the regimens. Under more chronic dosing conditions, withdrawal
symptoms may be more likely to occur. Finally,
individual difference in response to less-frequent
dosing intervals may emerge, such that some
patients may not respond well to even everyother-day BUP dosing.
Although the present studies Ž nd that the
interdosing interval cannot be extended to every
Ž fth day using sublingual doses, the feasibility of
a twice-weekly BUP dosing schedule (Petry et
al., 1999, 2000) may allow physicians to serve a
greater number of patients without the risk of
diversion associated with take-home medications. The feasibility of less-than-daily BUP dosing options has resulted in pilot projects
evaluating the efŽ cacy of BUP maintenance from
physician ofŽ ces (O’Connor et al., 1998). Future
studies may provide more information regarding
the effectiveness of these procedures in the treatment of opioid dependence.
Acknowledgements
We thank Melissa Foster, M. Elizabeth Hughes,
Elizabeth Kubik, Richard Taylor and Evan
Tzanis for help in conducting this study, and
Marcia Dunham, Susan Griggs and David McGarry for medication preparation. We also thank
Bruce Brown, Marne Moegelin and Tyler Wood
for providing therapy to the patients enrolled in
this study, John Brooklyn and John Hughes for
medical consultation and Susan Garthwait for
assistance in manuscript preparation. This research was supported by NIH grants T32DA07242, R01-DA06969, R01-DA13444, and
R29-DA12056. Preliminary results of this study
were presented at the College on Problems of
Drug Dependence in San Juan, Puerto Rico, in
1996.
References
833
AMASS, L., BICKEL, W. K., HIGGINS, S. T. & BADGER,
G. (1994) Alternate-day dosing during buprenorphine treatment of opioid dependence, Life Sciences,
54, 1215–1228.
BICKEL, W. K., AMASS, L., CREAN, J. P. & BADGER, G.
J. (1999) Buprenorphine dosing every 1, 2, or 3 days
in opioid-dependent patients, Psychopharmacology,
146, 111–118.
BICKEL, W. K. & AMASS, L. (1995) Buprenorphine
treatment of opioid dependence: a review, Experimental and Clinical Psychopharmacology, 3, 477–489.
BICKEL, W. K., STITZER, M. L., BIGELOW, G. E.,
LIEBSON , I. A., JASINSKI, D. & JOHNSON, R. (1988b)
Buprenorphine: dose-related blockade of opioid
challenge effects in opioid dependent humans, Journal of Pharmacology and Experimental Therapeutics,
247, 47–53.
BICKEL, W. K., STITZER, M. L., BIGELOW, G. E.,
LIEBSON , I. A., JASINSKI, D. R. & JOHNSON, R. E.
(1988a) A clinical trial of buprenorphine: comparison with methadone in the detoxiŽ cation of heroin
addicts, Clinical Pharmacology and Therapeutics, 43,
72–78.
GROSS, A., JACOBS , E., PETRY, N. M., BADGER, G. J. &
BICKEL, W. K. (2001) Limits to buprenorphine dosing: a comparison between quintuple and sextuple
the maintenance dose every 5 days, Drug and Alcohol
Dependence, in press.
HIMMELSBACH, C. K. (1939) Studies of certain addiction characteristics of (a) dihydromorphone, Journal
of Pharmacology and Experimental Therapeutics, 67,
239–249.
JASINSKI, D. R. & MARTIN, W. R. (1967) Evaluation of
a new photographic method for assessing pupil diameters, Clinical Pharmacology and Therapeutics, 8,
271–272.
JASINSKI, D. R., PEVNICK, J. & GRIFFITH, J. (1978)
Human pharmacology and abuse potential of the
analgesic buprenorphine, Archives of General
Psychiatry, 35, 501–516.
JASINSKI, D. R. (1977) Assessment of the abuse potentiality of morphine like drugs (methods used in
man), in: MARTIN, W. R. (Ed.) Drug Addiction,
pp. 197–258 (New York, Springer).
JOHNSON, R. E., JAFFE, J. H. & FUDALA, P. J. (1992) A
controlled trial of buprenorphine treatment for opioid dependence, Journal of the American Medical
Association, 267, 2750–2755.
KOSTEN, T. R., SCHOTTENFELD , R., ZIEDONIS, D. &
FALCIONI, J. (1993) Buprenorphine versus methadone maintenance for opioid dependence, Journal of
Nervous and Mental Disease, 181, 358–364.
LANGE, W. R., FUDALA, P. J., DAX, E. M. & JOHNSON,
R. E. (1990) Safety and side-effects of buprenorphine in the clinical management of heroin
addiction, Drug and Alcohol Dependence, 26, 19–28.
LING, W. (1996) A controlled trial comparing
buprenorphine and methadone maintenance in opioid dependence, Archives of General Psychiatry, 53,
401–407.
MARTIN, W., SLOAN, J. W., SHAPIRO, J. D. & JASINSKI,
D. R. (1971) Physiologic, subjective and behavioral
effects of amphetamine, metamphetamine, ephe-
834
Nancy M. Petry et al.
drine, phenmetrazine and methylphenidate in man,
Clinical Pharmacology and Therapeutics, 12, 245–258.
O’CONNOR , P. G., OLIVETO, A. H., SHI, J. M., TRIFFLEMAN, E. G., CARROLL, K. M., KOSTEN, T. R., et al.
(1998) A randomized trial of buprenorphine maintenance for heroin dependence in a primary care
clinic versus a methadone clinic, American Journal of
Medicine, 150, 100–105.
PETRY, N. M., BICKEL, W. K. & BADGER, G. J. (1999)
A comparison of four buprenorphine dosing regimens in the treatment of opioid dependence, Clinical
Pharmacology and Therapeutics, 66, 306314.
PETRY, N. M., BICKEL, W. K. & BADGER, G. J. (2000)
A comparison of four buprenorphine dosing regimens using open-dosing procedures: is twice weekly
dosing possible? Addiction, 95, 1069–1077.
PETRY, N. M., BICKEL, W. K., HUDDLESTON, J.,
TZANIS, E. & BADGER, G. J. (1998) A comparison of
subjective, psychomotor and physiological effects of
a novel muscarinic analgesic, LY297802 tartrate,
and oral morphine in occasional drug users, Drug
and Alcohol Dependence, 50, 129–136.
PRESTON, K. L., BIGELOW, G. E. & LIEBSON, I. A.
(1988) Buprenorphine and naloxone alone and
in combination in opioid-dependent humans,
Psychopharmacology, 94, 484–490.
RANCE, M. J. & DICKENS, J. N. (1978) The in uence of
drug-receptor kinetics on the pharmacological and
pharmaco-kinetic proŽ les of buprenorphine, in: VAN
REE, J. M. & PERENIUMS, L. (Eds) Characteristics
and Function of Opioids, pp. 65–66 (Amsterdam,
Elsevier).
WALSH, S. L., PRESTON, K. L., BIGELOW, G. E. &
STITZER, M. L. (1995) Acute administration of
buprenorphine in humans: partial agonist and blockade effects, Journal of Pharmacology and Experimental
Therapeutics, 274, 361–372.