Naloxone Reversal of Buprenorphine-induced Respiratory Depression

A total of 67 male and female subjects (age range, 20–30 yr; weight, 54–93 kg) participated in and completed the studies after approval of the protocols was obtained from the Human Medical Ethics Committee (Commissie Medische Ethiek, Leiden University Medical Center, Leiden, The Netherlands). We obtained oral and written consent. All subjects were healthy and did not have a history of illicit drug use or mental disease. All women were taking oral contraceptives. Subjects were asked to have a normal night of sleep and not to eat or drink for at least 6 h before the study. They were comfortably seated in a hospital bed for the duration of the study. They were naive with respect to the nature of the studies but were informed regarding the risk of participating. All subjects were students at Leiden University and received a financial reimbursement for their participation (€75–100 depending on the study).

Inspired and expired gas flows were measured with a pneumotachograph (Hans Rudolph, Myandotta, MI) connected to a pressure transducer and electronically integrated to yield a volume signal. The volume signal was calibrated with a motor-driven piston pump (stroke volume 1,000 ml, at a frequency of 20/min). The pneumotachograph was connected to a T-piece. One arm of the T-piece received a gas mixture with a flow of 45 l/min from a gas mixing system, consisting of three mass flow controllers (Bronkhorst High Tech BV, Veenendaal, The Netherlands) with which the flow of oxygen, carbon dioxide, and nitrogen could be set individually at a desired level. A personal computer provided control signals to the mass-flow controllers so that the composition of the inspired gas mixtures could be adjusted to force end-tidal oxygen and carbon dioxide concentrations (Peto²and Petco², respectively) to follow a specified pattern in time, independent of the ventilatory response (i.e. , dynamic end-tidal forcing).9In studies 1–3, end-tidal partial pressure of carbon dioxide (Pco²) was clamped at 53 mmHg throughout the measurements (approximately 8 kPa above resting values), while end-tidal partial pressure of oxygen (Po²) was maintained at a normoxic value of 110 mmHg. The oxygen and carbon dioxide concentrations and the arterial hemoglobin–oxygen saturation were measured with a Datex Multicap gas monitor near the mouth (Datex-Engstrom, Helsinki, Finland) and a Masimo pulse oximeter (using a finger probe) (Masimo, Irvine, CA), respectively. The gas monitor was calibrated with gas mixtures of known concentration delivered by a gas-mixing pump (Wösthoff, Bochum, Germany). Petco², Peto², inspired minute ventilation (Vⁱ), and oxygen saturation were collected on a breath-to-breath basis and stored on disk for further analysis.

The studies were placebo-controlled and had a double blind design. The local pharmacy delivered the buprenorphine hydrochloride (Reckitt Benckiser Healthcare Ltd., Hull, United Kingdom), naloxone hydrochloride (manufactured by the pharmacy), and placebo (0.9% NaCl). Randomization and preparation of the syringes was performed by a physician not involved in the study.‡‡

All buprenorphine and naloxone doses are per 70 kg. All bolus infusions were given over 90 s. Each subject participated once in any of the studies. Values reported are mean ± SEM, unless otherwise stated.

Sixteen subjects participated in this study. All received 0.2 mg intravenous buprenorphine (at t = 0) followed by 800 μg naloxone (in eight subjects) or placebo (in eight subjects) at t = 120 min. At the following time periods, steady state ventilation (Vⁱ) was measured (measurement period 7 min): −10 min (10 min before drug infusion), 15, 75, 140, 180, 240, 300, 360, 420, and 480 min. Analysis of variance and post hoc t  tests were performed to detect a significant effect of naloxone on ventilation at the P < 0.05 level.

Twenty-four subjects participated in this study. All received 0.1 mg buprenorphine at t = 2 min, followed by a continuous infusion for 1 h of 0.1 mg/h (total dose = 0.2 mg in 60 min). At t = 32, x mg naloxone was given, one half as bolus and one half infused over 30 min. The following total naloxone doses (x) were tested: 0, 0.5, 1, 2, 3, 4, 5, 6, and 7 mg. Each dose was tested in two subjects, except for placebo, which was tested in eight subjects. Breathing was measured continuously from 2 min before buprenorphine infusion until 90 min after the start of infusion.

The breath-to-breath data were averaged over 1-min periods. An ensemble average (mean of the 1-min subject means) was performed on the data of the eight subjects receiving the buprenorphine–placebo combination, allowing the calculation of buprenorphine–placebo-induced respiratory effect at various time points. To quantify the respiratory effect of naloxone relative to placebo, we used the following formula on the data of each subject who had received the buprenorphine–naloxone combination:

where time period z ranges from t = 61 to t = 63 min (i.e. , 1 min before to 1 min after the end of the continuous naloxone infusion); Vplacebo(z) is mean minute ventilation in the placebo group during period z, Vnaloxone(z) is mean ventilation of the equivalent time period after naloxone, and Vbaseline is mean ventilation of the 2 min before the buprenorphine infusion. This analysis will yield a quantitative measure of reversal, with 0 indicating no reversal (naloxone no better than placebo) and 1 indicating full reversal (response returned to prebuprenorphine level).

Thirty-two subjects participated in this study.

Sixteen received 0.1 mg intravenous buprenorphine at time t = 2 min, followed by a continuous infusion for 1 h of 0.1 mg/h (total dose = 0.2 mg in 60 min). At time t = 32 min, 2 mg naloxone (n = 8) or placebo (n = 8) was infused, followed by a continuous infusion of 4 mg/h for 2 h.

Sixteen other subjects received 0.2 mg intravenous buprenorphine at time t = 2 min, followed by a continuous infusion for 1 h of 0.2 mg/h (total dose = 0.4 mg in 60 min). At time t = 32 min, 3 mg naloxone (n = 8) or placebo (n = 8) was infused, followed by a continuous infusion of 4 mg/h for 2 h. The bolus naloxone dose was 50% greater than that of study 3.1. This was based on a pilot study in three subjects that showed the need for a greater initial dose of naloxone after 0.4 mg but not after 0.2 mg buprenorphine.

Ventilation was initially measured continuously from 2 min before buprenorphine infusion until 120 min after the start of infusion. Subsequently measurements were made at 30-min intervals until t = 240 min, after which hourly measurements were performed until t = 420 min.

The breath-to-breath data were averaged over 1-min periods. An ensemble average was performed in the naloxone and placebo data groups. The values were compared with baseline ventilation (± its 95% confidence interval).10When the mean ventilation value equalled or crossed (baseline ventilation − 1 × 95% confidence interval), we somewhat arbitrarily assumed that ventilation had returned to predrug baseline.

All subjects completed the studies without major side effects. The most frequent side effects were sedation (which occurred in all subjects) and nausea (which occurred in 46 of the 67 subjects).

In the placebo group, buprenorphine decreased ventilation from 24.2 ± 1.7 l/min to 13.6 ± 3.4 l/min at t = 75 min; in the naloxone group, ventilation decreased from 26.5 ± 2.1 l/min to 14.4 ± 1.7 l/min at t = 75 (not significant, analysis of variance). After infusion of 800 μg naloxone, ventilation at none of the measurement times differed between placebo and naloxone groups (analysis of variance). To detect a small effect of naloxone on ventilation unobserved in the pooled data analysis, we calculated the difference in ventilation from t = 75 to t = 180 min. In the placebo group, the change in ventilation was 0.2 ± 0.5 l/min, versus  2.2 ± 0.7 l/min in the naloxone group. This difference did not reach the level of significance (P = 0.08, one-tailed t  test, assuming a larger response in the naloxone group).

The mean effect of buprenorphine–placebo on minute ventilation is given in figure 1(gray area). Baseline ventilation was 24.0 ± 3.3 l/min at a fixed end-tidal Pco²of 52.9 ± 0.9 mmHg. Peak depression of ventilation occurred at t = 71 min after the start of the buprenorphine infusion, reaching a value of 13.5 ± 1.5 l/min. Relative to baseline, peak depression was 62 ± 11% of baseline, indicating a reduction of baseline ventilation by 38%. To get an impression of the naloxone data, we plotted representative data of two subjects given 2 and 6 mg naloxone in figure 1. The subject receiving 2 mg showed full reversal back to baseline (reversal = 1). In contrast, the subject given the higher naloxone dose showed little reversal (reversal = 0.1). In figure 2, we plotted the individual dose-reversal data for time frame 61–63 min. The data show that full reversal ± 20% was obtained at doses between 2 and 4 mg naloxone but that at higher doses, reversal gradually declined. We calculated the naloxone dose causing 50% reversal of 0.95 ± 0.09 mg (data fitted to a sigmoid Emax model with inhibitory component—see legend of fig. 2—values are median ± SE) and the dose causing the return to 50% depression of 5.20 ± 0.94 mg naloxone.

Baseline ventilation averaged to 21.9 ± 2.5 l/min (data from studies 3.1 and 3.2 combined). The effects of both doses of buprenorphine (0.2 and 0.4 mg) were successfully reversed by a continuous infusion of naloxone at the dose chosen by us, which was, at least partly, based on the data from study 2.

See figure 3A. Buprenorphine, 0.2 mg, caused a rapid decrease in ventilation. Before naloxone or placebo infusion (t = 32 min), ventilation was 84 ± 3 and 79 ± 5% of baseline, respectively. In the placebo group, ventilation declined further to a nadir of 57 ± 6% of baseline at t = 120 min. In the naloxone group, the nadir was 78 ± 4% of baseline at t = 48 min (at the same time period, ventilation was 61 ± 5% of baseline in the placebo group). From that point on, ventilation increased to reach baseline values (i.e. , baseline ventilation − 1 × 95% confidence interval) at t = 70 min. Ventilation remained not different from baseline during the remainder of the naloxone infusion. After termination of the naloxone infusion (at t = 152 min), ventilation decreased, but it never reached the level observed in the placebo group.

See figure 3B. A rapid decrease in ventilation occurred after the initiation of the 0.4 mg buprenorphine infusion. Before naloxone or placebo infusion (t = 32 min), ventilation was 62 ± 5 and 64 ± 5% of baseline, respectively. In the placebo group, ventilation declined further to a nadir of 40 ± 3% of baseline at t = 150 min. In the naloxone group, the ventilation nadir was 61 ± 5% of baseline at t = 34 min (ventilation of the placebo group was 66 ± 7% at t = 34 min). From that point on, ventilation increased to reach baseline values at t = 93 min. Ventilation remained not different from baseline during the remainder of the naloxone infusion. After termination of the naloxone or placebo infusion (at t = 152 min), the changes in ventilation were similar to those observed in study 3.1

In our studies, we observed that an intravenous dose of naloxone of 0.8 mg had no effect on respiratory depression induced by the opioid analgesic buprenorphine. We next explored the naloxone dose–response relation and observed that increasing doses of naloxone caused full reversal of buprenorphine respiratory depression (2–4 mg naloxone given in 30 min). Further increasing the naloxone dose (5–7 mg), however, caused a decline in reversal activity. The form of the dose–response relation is best described by a bell shape or inverse U. Taking into account these data, we designed a naloxone infusion scheme intended to cause full reversal of the respiratory depression from 0.2 and 0.4 mg buprenorphine. A naloxone bolus dose of 2–3 mg, followed by a continuous infusion of 4 mg/h, caused full reversal within 40–60 min. Renarcotization did occur upon the termination of the naloxone infusion. These data indicate that reversal of buprenorphine-induced respiratory depression is possible but depends on the naloxone dose and its inverse U–shaped dose–response relation. That is, reversal is possible within a specific naloxone dose window. Furthermore, because respiratory depression from buprenorphine may outlast the effects of naloxone boluses or short infusions, a continuous infusion of naloxone may be required to maintain reversal of respiratory depression. Note that the design of studies 2 and 3 was such that it mimics the clinical situation in which a possible respiratory effect from a buprenorphine transdermal patch must be reversed by naloxone. A subcutaneous depot of buprenorphine will persist upon the removal of the patch. During the existence of this depot and the need for reversal, naloxone and buprenorphine will then be released or administered simultaneously to the blood (as they are in studies 2 and 3).

All opioids that interact with the μ-opioid receptor system depress respiration.11The extent of respiratory effect is highly variable and is related to the specific opioid, the opioid dose, the administration mode, concurrent medication, underlying disease, pain and the state of arousal (these two factors vary over time), genetics, and exogenous stimulatory factors. Because the occurrence of overt and sometimes life-threatening respiratory depression is often unpredictable, the ability to induce rapid opioid reversal is of evident importance. In contemporary medicine, naloxone has become the drug of choice for treatment of opioid-induced respiratory depression. Naloxone is a nonspecific opioid receptor antagonist (i.e. , it antagonizes the μ-, κ-, and δ-opioid receptors) with a relatively short duration of action resulting from rapid elimination; its half-life in plasma is approximately 30–45 min.12 

There is ample evidence that buprenorphine, like other μ-opioid receptor agonists, produces significant respiratory depression at clinical doses (figs. 1 and 3), although we recently showed that buprenorphine-induced respiratory depression, unlike other μ-opioids, shows an apparent maximum in effect (ceiling).4,13Interestingly, only sparse data from the literature addressed the issue of reversal of buprenorphine-induced respiratory depression.5–7The picture that emerges from these few studies is that at relatively large bolus naloxone doses, little (i.e. , only partial) reversal of the respiratory effects of buprenorphine is observed. For example, a recent short report indicates that an incremental naloxone dose of 2.4 mg has an effect on 0.4 mg buprenorphine–induced respiratory depression no greater than placebo in patients during sevoflurane–nitrous oxide anesthesia.7An older study by Gal5showed only partial reversal of 0.3 mg buprenorphine with 5 and 10 mg intravenous naloxone (given as single bolus). The inability to obtain full reversal in these two studies may be related to various factors, such as anesthesia (anesthesia must be considered a serious complication when studying opioid-induced respiratory depression due to the complex opioid-anesthetic interaction on breathing),14,15the lack of sensitivity of the respiratory model applied to assess naloxone–buprenorphine interaction, the use of single naloxone doses, and finally, the use of a too-high dose of naloxone (fig. 2).

The resistance to naloxone reversal is related to the high affinity of buprenorphine for the μ-opioid receptor.1,3This high affinity explains why relatively high doses of naloxone (2–4 mg) are needed before reversal is observed. The need for a continuous infusion in this process (upon termination of the naloxone infusion, there was a rapid return of respiratory depression; fig. 3) implies the need for continuous supply of naloxone to the opioid receptor sites in the brain involved in respiratory depression. Otherwise, the naloxone bolus is rapidly washed out from the brain compartment and eliminated from the body. We believe that the use of a single dose of naloxone infusion to reverse opioid-related overdose has several disadvantages that are unrelated to the opioid involved: renarcotization due to the short duration of action of naloxone, the inability to titrate to effect causing the return of pain, and sympathicoexcitation. An infusion regimen aimed at a prolonged and steady state naloxone plasma concentration may overcome these shortcomings. For example, continuous (11-h) naloxone infusion after high-dose fentanyl anesthesia caused reversal of respiratory depression without causing renarcotization, pain, or sympathicoexcitation.16 

An interesting observation in studies 3.1 and 3.2 is the higher ventilation levels after naloxone treatment than after placebo treatment at times when naloxone is washed out from the brain and possibly also from the body (fig. 3at t > 240 min). This is probably due to washout from the brain compartment of buprenorphine which was replaced by naloxone at the μ-receptor or at nonspecific binding sites (i.e. , some buprenorphine was lost without replacement).

Naloxone doses exceeding the maximal effective dose (> 4 mg) lead to a decrease in (0.2 mg) buprenorphine reversal efficacy (fig. 2). Because the number of subjects was limited (just two subjects per naloxone dose over the dose range from 0.5 to 7 mg), we consider this observation preliminary. Evidently, further studies are needed. In a first attempt, we performed a set of experiments after 0.4 mg buprenorphine and applied various naloxone doses (one dose per subject; duration of naloxone infusion 30 min) and observed a similar bell-shaped dose–response relation, albeit full reversal was not reached (A. Dahan, unpublished observation, September 2004–January 2005). Our unexpected observation is most probably specific to buprenorphine and its interaction with naloxone. Buprenorphine has a long history of showing bell-shaped dose–response curves with respect to its analgesia and side effect profile.1Most of these observations were made in animals. For example, rodents display a bell-shaped buprenorphine dose–response relation in various antinociceptive assays (electrical pain, heat pain, visceral pain, and spinal nerve ligation).17,18Also, in humans, there are indications of the existence of a bell-shaped dose–response curve with respect to analgesia. For example, two patients treated with buprenorphine (30–40 μg/kg) for postoperative pain showed improved pain relief after 0.4 mg naloxone infusion, probably due to shifting of the bell-shaped dose–analgesia curve to the right.19A rightward-shift of the bell-shaped buprenorphine dose–response curve after the infusion of an opioid-antagonist (naltrexone) has been observed in rats using an electrical pain test.20Note that a bell-shaped curve was never observed in experimental human studies, and clinically complete analgesia is reached with buprenorphine. The mechanism of the bell-shaped curve remains unknown. Some argued that the form of the curve is related to the type and intensity of (experimental) pain administered.20Others suggested noncompetitive autoinhibition, in which there are two receptor subpopulations, one mediating the agonistic properties at low dose, the other mediating the antagonistic properties at high dose.3,17,18,20Finally, Lutfy et al.  21suggest the contribution of the opioid receptor-like 1 receptor. They showed that buprenorphine, but not morphine, given to mice activates opioid receptor-like 1 receptors, compromising (antagonizing) analgesia from μ-opioid receptors. The latter theory seems not plausible when one takes into account that the scanty literature that exists on the respiratory effects of stimulation of opioid receptor-like 1 receptors shows respiratory depression rather than stimulation.22The existence of two μ-opioid receptor subpopulations as described above could theoretically explain our findings with high-dose naloxone causing the antagonism of the receptors mediating the antagonistic effects of buprenorphine. We are not aware, however, of any observation of these two receptor subpopulations in in vitro  or in vivo  animal studies.

The results of our studies demonstrate that the specific dose and mode of administration of naloxone to restore breathing and maintain it at an adequate level are complex matters that require further study. Our data show that even after administration of large boluses of naloxone or boluses plus brief infusions, respiratory depression induced by buprenorphine recurred and persisted for the duration of the study (7 h in study 3). Additional studies are required to define the dose and the mode of administration of naloxone to restore breathing and maintain it at an adequate level in the clinical setting, which is complicated by acute and chronic pain, gender effects,23high doses of opioids, long-acting opioids, and various sustained-release preparations of opioids.