Palliative Medicine 2006; 20: s3-s8

Opioid-induced respiratory effects: new data onbuprenorphineAlbert Dahan Department of Anesthesiology, Leiden University Medical Center, Leiden

When selecting the appropriate long-acting opioid to treat cancer pain, both analgesicefficacy and safety need consideration. Generally, opioids are well tolerated. However, ofopioid-typical adverse events, respiraton/ depression is especially important because of therisk of a fatal outcome. Although all potent opio)d analgesics act via the ^-opioid receptorsystem, they differ in how they affect respirator/ control. Recently, the respiratory effectsof fentanyl (1 -7 iig/kg) and buprenorphine (0.7-9 iig g/kg) were compared in healthy opioid-naTve volunteers. Fentanyl produced dose-dependent depression of respiration with apnoeaat doses >3 ^g/kg, while buprenorphine caused depression that levelled at - 50% ofbaseline with doses >2 lag/kg. These findings indicate the occurrence of a ceiling in therespiratory depression induced by buprenorphine but not by fentanyl. Surprisingly fewstudies have addressed the clinically important ability to reverse the respiratory effects ofopioids. A recent assessment of the naloxone dose required to reverse 0.2 mg intravenousbuprenorphine-induced respiratory depression in healthy opioid-naTve volunteers, found thatthe accumulated naloxone dose causing 50% reversal of respiratory depression was 1.20 ±0.32 mg/70 kg (given in 30 min); 80% reversal was observed at 2.50 + 0.60 mg/70 kg (givenin 30 min). At greater buprenorphine doses, full reversal is observed when the duration ofnaloxone infusion is increased. These findings indicate the need for a continuous ratherthan bolus administration of naloxone to reverse the respiratory effects of buprenorphine. Inconclusion, buprenorphine is more favourable compared with fentanyl in respect toventilatory control. Buprenorphine causes limited respiratory depression with a ceilingeffect at higher doses, while fentanyl causes dose-dependent respiratory depression withapnoea at high dose levels. In the rare instance of respiratory depression, reversal ispossible with a sufficient and continuous infusion of naloxone. Palliative Medicine 2006;20: s3-s8

Key words: apnoea; buprenorphine; ceiling effect; fentanyl; naloxone; opioid; respiratory depression

Introduction that the desired (antinociceptive) and undesired (respira-tory depression) etTects of morphine (and its active

Long-acting opioids are important analgesics for the metabohte morphine-6-gIucuronide) and probably al! \i-treatment of cancer pain but are variable in terms of opioids are linked to the same target: the ^-opioidanalgesic efficacy and safety. Ahhough opioids are receptor gene (Oprm).^-^ Mice with intact M-opioidgenerally well tolerated, consideration of safety is needed receptors showed a dose-dependent decrease in bothbecause - in addition to analgesia - morphine and most breathing frequency and tidal volume. However, no suchother opioids produce their characteristic responses both effect was observed in mice lacking the la-opioid receptorcentrally and peripherally.'""^ Of opioid-typical adverse (so-called n-opioid receptor knockout mice). Similarly,events, respiratory depression is the most serious because mice with ji-receptors showed dose-dependent antinoci-of the risk of a fatal outcome for the patient. The ception from morphine, while no effect was observed inmechanism of opioid-induced respiratory depression is |j.- ji-opioid receptor knockout mice (see also Figure l).'*'^opioid receptor inhibition of the brainstem respiratory Clinical studies in volunteers and patients on thecontrol centres. Opioids affect both the rate and depth of effect of morphine on respiration have also shown arespiration, eventually resulting in an increased arterial dose-response relationship.^ ^ For example, morphinepartial pressure of carbon dioxide and reduced partial causes a dose-dependent decrease in minute ventilation ofpressure of oxygen.^ Data from studies in mice indicate 6-7L/min before the drug is given to unstable and• cyclic breathing with both a reduction in the breathingAddress for correspondence: Albert Daham Department of ^ ^^ j ^he depth of breathing at 0.2-0.3 mg/kg.^Anesthesiology, Leiden University Medical Center P5-Q, P.O. . , , , i •Box 9600, 2300 RC Leiden, The Netherlands. At higher doses apnoea may occur when morphine isE-mail: a.dahan@lumc.nl given to persons without pain.

© 2006 Edward Arnold (PubUshers) Ltd 10.1191/0269216306pml 126oa

s4 A. Dahan

30

25 -

Pain assay

intact'

Ventilation - CO2 response

22 n

20 -

.S 18E1.16

O 12

10

intact

^-Opioid Receptor KO

6 20 100

Dose ip mg/kg

0.0 3.0 4.5

% CO.

7.5

Figure 1 Effect of various doses of morphine on antinociception (left: the tail immGrsion test) and respiration (right:ventilatory response to various concentrations of carbon dioxide} in a mouse with an intact n-opioid receptor gene {intact} and a|i-opioid receptor gene knockout mouse. Note the absence of opioid effect in the knockout mice indicating that thereceptor gene is the primary site of respiratory and antinociceptive action of opioids.

In this short review I will discuss the respiratorybehaviour of an opioid that is frequently used in thetreatment of pain, buprenorphine. It is a potent opioidwith special characteristics. I will (i) compare buprenor-phine and fentanyl dose-(respiratory) response curves:(ii) compare buprenorphinc's analgesic and respiratoryresponses; and (iii) discuss the ability of naloxone toreverse buprenorphine-induced respiratory depression.

Respiratory depression: buprenorphine versusfentanyl

Recent data from published case studies have indicated anumber of instances where patients have suffered severerespiratory depression after the continuous infusion offentanyl via a transdermal patch."^ " To underline theimportance of this issue, of six case patients whopresented with drug-induced respiratory depression, twodied.^ We relate these serious adverse events to (1) therelatively high doses of fentanyl given; (2) the periodicnature of pain; (3) psychomimetic comedication (such assedatives, alcohol, other pain medication) and (3) theabsence of supervision of the patients. But do all opioidsbehave in the same way? Our group attempted to answerthis question by investigating the comparative etfects onrespiratory depression of the equipotent analgesicsfentanyl and buprenorphine in rats.' ' Fentanyl injectedintravenously into animals up to a dosage of 0.09 mg/kgshowed a dose-dependent linear increase in arterialpressure of carbon dioxide, a surrogate measure of

respiration (Figure 2). On increasing the fentanyl dosagefurther, the animals died of fatal respiratory depression.In contrast, when buprenorphine was administeredintravenously there was an initial increase in arterialpressure of carbon dioxide, which rapidly levelled offeven on increasing the dosage up to 3.0 mg/kg (Figure 2).This phenomenon, known as the ceiling effect, representsthe point where the agonistic effects of buprenorphineplateau and the drug develops an action more akin tothat of an antagonist, thereby limiting respiratorydepression.'^''"

How do buprenorphine and fentanyl compare inrespect to respiratory depression in humans? We con-ducted a double-blinded, placebo-controlled study of

0.10 1.00iv dose (mg/kg)

10.00

Figure 2 Effect of increasing doses of fentanyl andbuprenorphine on arterial PCO? in rats. Note the linearincrease in CO? after fentanyi but iittle to no increase afterbuprenorphine. Values are mean±SEM, Data are adaptedfrom Dahan etai}^

Opioid-induced respiratory effects and buprenorphine s5

25 -| Fentanyi

20

1 5 H

M 1 0 -

5 -

25 n Buprenorphine

20

15

10 -

APNOEA 5 -

f

PLATEAU

2 4 6

Fentanyl dose ()ig/kg)

0 2 4 6 8

Buprenorphine dose (^g/kg)

Figure 3 Effect of increasing doses of fentanyl and buprenorphine on respiration in human volunteers (n^5 -8 per dosegroup; n ^ i for the highest fentanyl dose). Note the dose-dependent decrease for both drugs with apnoea at high-dosefentanyl but a ceiling or piateau at intermediate to high dose buprenorphine. Values are population mean. For clarity no errorbars are shown. Data are adapted from Dahan et al.T2

fentanyl and buprenorphine in 48 volunteers. Using thecomputer-steered 'dynamic end-tidal forcing" technique,respiratory studies were performed at a clamped end-tidal oxygen (15 kPa) and carbon dioxide (7 kPa) tension.Ventilation was measured for up to 7 hours. Dose-response relationships were observed with fentanyl inthe range of 0-9 lig/kg and for buprenorphine at 0-7 |ig/kg (with respect to analgesia fentanyl and buprenorphineare equipotent relative to morphine). Fentanyl causeddose-dependent reduction of minute ventilation withrespiratory instability at doses of 3 (ig/kg and greater.In one subject, prolonged periods of apnoea wereobserved at the highest dose tested (500 |ig in a 70 kgvolunteer). Buprenorphine similarly caused dose-depen-dent reduction of minute ventilation, however, in contrastto fentanyl a plateau in respiratory depression (about

Breathing

50% of baseline) occurred at dosages >3 |ig/kg (see alsoFigure 3). Importantly, no respiratory instability, peri-odic breathing or apnoea occurred, even at the highestdose tested (600 p,g in a 70 kg volunteer).

Buprenorphine: ceiling in respiration but notin analgesia

An important question is whether the ceiling in respira-tory depression is linked to a concomitant reduction inanalgesia. This question was investigated using a stan-dard electric pain model'^ with two groups of patients,each receiving a different dosage of buprenorphine (3 and6 ng/kg).'^ The results demonstrated that doubling the

Pain reiief70

60-

0 41 2 3 4 5 6 7

Time (hours)

2 3 4 5 6 7

Time (hours)

Figure 4 Effect of two doses of buprenorphine (3 pg/kg cyan dots; 6 ng/kg grey dots) on respiration (left) and analgesia (right)in human volunteers (n ̂ 10 per dose group}. Note that doubling the dose had no further effect on respiration while analgesicefficacy increased significantly. For clarity no error bars are shown. Data are adapted from Dahan ef s/.̂ ^

s6 A. Dahan

0 15 30 45 60 75 90

TIME (min)

ro 20 -

15 -

100 15 30 45 60 75 90

TIME (min)

30 - 2 mg naloxone

0 15 30 45 60 75 90

TIME (min)

0 15 30 45 60 75 90

TIME (min)

Figure 5 Influence of naloxone on buprenorphine-induced respiratory depression. Examples of four volunteers that received0.2 mg buprenorphine intravenously from t = 2 to t=62 min, followed by various doses of naloxone from t^32 to f ^62 min.Note that at low doses of naloxone (0.5 and 1,0 mg} no reversal occurs while at doses of 2 mg and greater full reversal isobserved. Each dot represents a 1-min average. The broken line is baseline ventilation ( -predrug ventilation). The thick blacklines are the infusion periods of buprenorphine and naloxone.

dosage of buprenorphine from 3 to 6 ^g/kg had no effecton breathing (Figure 4). In contrast, doubling the dosagenow had a significant effect as analgesia increasedthree-fold (Figure 4). This underlines the ability ofbuprenorphine to instigate a ceiling effect to counterrespiratory depression that is independent of its effect onanalgesia.'^ Note that this conclusion is valid only for thedose-range tested.'^

Reversal of buprenorphine-induced respiratorydepression using naloxone

Knowledge of the ability to reverse opioid-inducedrespiratory depression is important for all potent andclinically used opioids (eg, fentanyl, morphine, bupre-norphine). However, there are surprisingly few studiesthat have addressed the reversal of opioid-inducedrespiratory effects. Despite data from early studies tothe contrary,"^''^ we performed a study investigating the

reversal of the effects of buprenorphine on respiratoryfunction with naloxone in humans."'* An adaptive trialdesign was used to find the dose that caused full rev-ersal of buprenorphine-induced respiratory depression.Twenty-one opioid-naive volunteers were tested over a90-min period with end-tidal carbon dioxide tensionclamped at 7 kPa. Buprenorphine 0.2 mg/kg wasadministered as a continuous intravenous infusion for 1hour. After a period of 30 min from the start ofbuprenorphine infusion, either placebo (NaCI 0.9%,n^T) or naloxone (0.5-5 mg, rt^l4) was infused for30 min, and the effect on breathing assessed for a further30-min period afterwards. As expected, placebo did notaffect the buprenorphine-induced decrease in ventilationrate during the period 0-60 min. A slow decline inventilation rate was observed, reaching a nadir ofdepression after 70 min. On infusing low-dose naloxone(0.5 mg) over 30 min there was little discernible effect onbuprenorphine-induced respiratory depression followinginitial naloxone injection, leading to a continued depres-sion (Figure 5). At such a low dose it is relatively ditTicult

Opioid-induced respiratory effects and buprenorphine s7

for naloxone to displace buprenorphine already bound top-opioid receptors.'** However, on increasing the dosageof naloxone to 2 mg, a full reversal of the respiratoryeffect was observed, returning ventilation to its baselinelevel (Figure 5). Thus, total reversal of respiratorydepression induced by buprenorphine can be broughtabout at a relatively low dose of continuously infusednaloxone.

The data show that full reversal of respiratory depres-sion produced by buprenorphine can be achieved byusing a continuous infusion of sufficiently high dosagesof naloxone. The dose-dependent reversal of buprenor-phine-induced respiratory effects follows a sigmoidalrelationship (Figure 6): 50% reversal of depression isachieved after 30 min of continuous infusion of naloxone1.2±O.32mg/7Okg, and 80% reversal with 2.50 +0.60 mg/70 kg. Reversal of greater doses of buprenor-phine was recently studied in our laboratory (Dahan,unpublished observation) and not surprisingly - weobserved that not greater doses of naloxone are neededfor reversal but a longer infusion duration is needed. Inthis respect buprenorphine behaves similar to otherpotent and long-acting opioids with high alTinity at the

receptor."'

<

UJsa:

1.0 -̂

0.8 —

0.6 -

0.4 -

0.2 -

0.0 -^

o#

/ 1

8 o

o

0 1 2 3 4 5

Naloxone dose (mg per 70 kg)

Figure 6 Effect of various doses of naloxone on reversalof 0,2 mg buprenorphine-induced respiratory depression inhuman volunteers. The line through the data is the datafit using a sigmoid Emax model. As the Emax = 1 , full reversalis possible. In gray the naloxone ED50 (50% reversal) andEDso (80% reversal) are given. Data are adapted from Dahanet a/.̂ ^ Note that naloxone doses between 2 and 4 mg willcause full reversal of respiratory depression.

Conclusions

In summary, buprenorphine compares favourably withfentanyl in respect to ventilatory control. Unlike fentanyl,which causes dose-dependent depression with apnoea athigh dosage, buprenorphine causes limited respiratorydepression with a ceiling effect at high dosages. In therare instance of respiratory depression with buprenor-phine, full reversal using a continuous infusion ofnaloxone to obtain a constant (and sufficient) plasmaconcentration is possible.

AcknowledgementsThe author thanks Grunenthal GmbH. Aachen,Germany for their support and educational grantsenabling the studies on buprenorphine.

References

1 Martin WR. Pharmacology of opioids. Pharmacol Rev1983; 35: 283 323.

2 Budd K, Shipton EA. Acute pain, the immune systemand opiommunosuppression. Aviilc Pain 2004; 6; 123-35.

3 Smith HS. Drugs for pain. Hanley & Belfus Inc, 2003.4 Dahan A, Sarton E, Teppema L. et al. Anesthetic

potency and influence of morphine and sevoflurane onrespiration in fi-opioid receptor knockout mice.Anesihesiology 2001; 94: 824-32.

5 Romberg R, Sarton E, Teppema L. ct al. Comparison ofmorphinc-6-glucuronidc and morphine on respiratorydepressant and antinociceptive responses in wild typeand (j-opioid receptor deficient mice. Br J Anaesth 2003;91; 862 70.

6 Jacobson L, Chabal C, Brody MC. A dose-responsestudy of intrathecal morphine; efficacy, duration,optimal dose, and side efTects. Anesih Analg 1988; 67;1082-88.

7 Bailey PL, Rhondeau S, Schafer PG, el at. Dose-response pharmacology of intrathecal morphine inhuman volunteers. Anesihesiology 1993; 79: 49-59;discussion 25A.

8 Dahan A, Romberg R, Teppema L, el al. Simultaneousmeasurement and integrated analysis of analgesia andrespiration after an intravenous morphine infusion.Anesihesiology 2004; 101; 1201 209.

9 Regnard C, Pelham A. Severe respiratory depression andsedation with transdermal fentanyl; four case studies.Palliat Med 2003; 17; 714 16.

10 WHO Drug Information. 2004; 18; 275.11 Canadian Adverse Reaction Newsletter 2004; 14.12 Dahan A, Yassen A, BijI H. el al. Comparison of the

respiratory effects of intravenous buprenorphineand fentanyl in humans and rats. Br J Amiesih 2005;94; 825- 34.

s8 A. Dahan

13 Sporer KA. Buprenorphine: a primer for emergency 19 Bowdle TA. Pharmacology of analgesia. In Healy TEJ,physicians. Ann Emerg Med 2004; 43: 580-84. Knight PR eds, Wylie and Churchill-Davidson's a practice

14 Walsh SL, Preston KL, Stitzer ML, et al. Clinical o/«He5//ies/a, seventh edition. Arnold, 2003: 543-63.pharmacology of buprenorphine: ceiling efTect at high 20 Bijl H. Yassen A, Olofsen E. c-i at. Full reversal ofdoses. Clin Pharmacol Ther 1994; 65: 569-80. buprenorphine-induced

15 Sarton E, Olofsen E, Romberg R. et al. Sex differences in respiratory depression by the opioid receptor antagonistmorphine analgesia: an experimental study in healthy naloxone. Ned Tijd.schr r Anesth (Neth J Aneslh) 2006:volunteers. Anesthesiology 2000; 93: 1245-54. -^^ nrcA^

16 Dahan A, Yassen A. Romberg R, et al. Buprenorphine 2I Takahashi M. Sugiyama K. Hori M, Chiba S, Kusaka K.induces ceiUng m respiratory depression but not m Naloxone reversal of opioid anesthesia: clinical evalua-analgesia. Br J Anaesth 2005; 94. 825-34 ^-^^ ^^^ pl^^^^ concentrations of continuous naloxone

? ? ? - , p"^' 'r ' '^^?nm"T.?f'^' ' . ' ' .fr "'"''• i"f""̂ ion f̂̂ e-- anesthesia with high-dose fentanyl. Jfnt J Clm Pract Suppl 20(i?i: m. 3^-^., 2^-24. . , . , , „ . , 0 , 0

10 r- 1 T-T XT 1 T r u I.- J J Anesth 2004; 18: 1-8.1ft Oal IJ. Naioxone reversal 01 buprenorphine-inducedrespiratory depression. Clin Pharmacol Ther 1989; 45:66 71.