ORIGINAL ARTICLE
`
`Population Pharmacokinetics of Intravenous,
`Intramuscular, and Intranasal Naloxone in
`Human Volunteers
`
`Jonathonm Dowling,* Geoffrey K. Isbister,†‡¶ Carl M. J. Kirkpatrick,‡ Daya Naidoo,§
`k
`and Andis Graudins*
`
`Abstract: To investigate the pharmacokinetics of naloxone in healthy
`volunteers, we undertook an open-label crossover study in which six
`male volunteers received naloxone on five occasions: intravenous (0.8
`mg), intramuscular (0.8 mg), intranasal (0.8 mg), intravenous (2 mg),
`and intranasal (2 mg). Samples were collected for 4 hours after
`administration for 128 samples in total. A population pharmacokinetic
`analysis was undertaken using NONMEM. The data were best
`described by a three-compartment model with first-order absorption for
`intramuscular and intranasal administration, between-subject variabil-
`ity on clearance and central volume, lean body weight on clearance,
`and weight on central volume. Relative bioavailability of intramuscular
`and intranasal naloxone was 36% and 4%, respectively. The final
`parameter estimates were clearance, 91 L/hr; central volume, 2.87 L;
`first peripheral compartment volume, 1.49 L, second peripheral
`compartment volume, 33.6 L; first intercompartmental clearance, 5.66
`L/hr; second intercompartmental clearance, 29.8 L/hr; Ka (intramus-
`cular), 0.65; and Ka (intranasal), 1.52. Median time to peak
`concentration for intramuscular naloxone was 12 minutes and for
`intranasal, 6 to 9 minutes. A combination of
`intravenous and
`intramuscular naloxone provided immediate high and then detectable
`concentrations for 4 hours. Intranasal naloxone had poor bioavailability
`compared with intramuscular. Combined intravenous and intramus-
`cular administration may be a useful alternative to naloxone infusions.
`
`Key Words: naloxone, pharmacokinetics, population pharmaco-
`kinetics, therapeutic drug monitoring
`
`(Ther Drug Monit 2008;30:490–496)
`
`INTRODUCTION
`Naloxone is an important drug in the treatment of opioid
`toxicity both in the prehospital and hospital setting. Patients
`
`Received for publication January 17, 2007; accepted May 23, 2008.
`From the *Clinical and Experimental Toxicology Unit, Prince of Wales
`Hospital and Prince of Wales Clinical School, University of NSW, Sydney,
`Australia; †Menzies School of Health Research, Charles Darwin
`University, Darwin, Australia; the ‡School of Pharmacy, University of
`Queensland, Brisbane, Australia; §SEALS Pathology Service, Prince of
`k
`Prince of Wales Clinical School,
`Wales Hospital, Sydney, Australia; and
`Faculty of Medicine, University of New South Wales, Sydney, Australia.
`Correspondence: Jonathonm Dowling Department of Clinical Toxicology,
`Newcastle Mater Hospital, Edith Street, Waratah NSW 2298, Australia (e-
`mail: Geoffrey.isbister@menzies.edu.au).
`G.K.I.
`is funded by an NHMRC Clinical Career Development Award
`ID300785.
`Copyright Ó 2008 by Lippincott Williams & Wilkins
`
`490
`
`with opioid poisoning requiring naloxone therapy are often
`difficult
`to cannulate as a result of previous intravenous
`substance abuse. This may delay the administration of antidote
`therapy. Intravenous drug abusers are also at increased risk of
`carrying bloodborne infections that could be transmitted to
`healthcare workers through needlestick injuries.1 The half-life
`of naloxone is significantly shorter than most of the opioid
`agents, so its duration of action is shorter than that of most
`opioid agents. Patients may awaken from opioid toxicity and
`want to remove themselves from medical care when there is
`the risk of recurrence of opioid toxicity after the effects of
`naloxone wear off. This is a particular concern with long-
`acting opioids such as methadone and has prompted the use of
`a combination of intravenous and intramuscular naloxone in
`the field to prolong its duration of action. However, this
`approach is not evidence-based or based on an understanding
`of the pharmacokinetics of naloxone.
`The intranasal route of administration has been shown to
`be clinically effective for a number of medications, including
`analgesics and sedatives.2,3 Recent clinical observational
`studies have suggested that intranasal naloxone may be safely
`administered for the reversal of opioid intoxication in the
`prehospital and hospital settings.4–10
`Despite naloxone being used for over 40 years, there are
`limited pharmacokinetic data in animals and humans.11–16
`Naloxone disappears rapidly from the serum in the initial
`distribution phase, over a period of approximately 15 to 20
`minutes, and then has an elimination half-life ranging from 30
`to 90 minutes based on a two-compartment model or
`estimation of
`the slope of
`the terminal portion of
`the
`concentration time curve.11,14,16 Animal studies have been
`used to delineate the pharmacokinetics of naloxone through
`the intranasal route in rats.17 However, there are no data
`describing the pharmacokinetics of intranasal naloxone in
`humans.
`This study aimed to determine the pharmacokinetics of
`intranasal naloxone in humans and compare these with the
`pharmacokinetics of equivalent doses of naloxone delivered
`through the intramuscular and intravenous routes.
`
`METHODS
`This was an open-label crossover volunteer study of the
`pharmacokinetics of intravenous, intramuscular, and intrana-
`sal naloxone. Ethics approval was obtained from the South
`Ther Drug Monit  Volume 30, Number 4, August 2008
`Nalox1027
`Nalox-1 Pharmaceuticals, LLC
`Page 1 of 7
`
`

`

`Ther Drug Monit  Volume 30, Number 4, August 2008
`
`Population Pharmacokinetics of Naloxone
`
`Eastern Area Health Service Ethics Committee as well as the
`University of New South Wales Ethics Secretariat. Informed
`consent was obtained from all volunteers. In addition, the
`Therapeutic Goods Administration was notified.
`
`Patient Data
`Six healthy male volunteers were recruited with a median
`age of 25 years (range, 24–45 years), median weight of 80 kg
`(range, 75–100 kg), and median height of 1.78 m (range, 1.75–
`1.93 m). Exclusion criteria were previous or current opioid
`dependence or abuse, current use of opioid analgesics for pain
`relief, cardiorespiratory disease, current or
`recent upper
`respiratory tract
`infection, or abnormal nasal anatomy.
`Naloxone administration at the doses used in this study has
`not been shown to result in any adverse reactions in healthy
`volunteers.18–26 The study was undertaken in a critical care
`setting with resuscitation facilities available in the unlikely
`event of an adverse drug reaction to naloxone. All patients had
`a cannula inserted for administration of intravenous naloxone
`and collection of blood for drug analysis.
`Naloxone was purchased from Mayne Pharma Ltd.,
`Melbourne, Australia, at a concentration of 400 mg/mL.
`Intravenous naloxone was administered through the cannula
`and flushed with a 5-mL bolus of saline. Intramuscular
`naloxone was administered with a 23-g needle as a single
`injection in the gluteus maximus muscle. Intranasal naloxone
`was administered through a Mucosal Atomiser Device (Wolf-
`Tory Medical, Salt Lake City, UT) with the patient lying at 45°
`and instructed not to swallow and breathe through the mouth
`for at least 1 minute. This technique was used after a trial of
`administration with normal saline in both the seated and
`supine positions revealed a significant amount of solution
`either lost out the nose or swallowed by the subjects. Half the
`volume was administered to each of the subject’s nostrils.
`The study was divided into five separate arms: 1) 0.8 mg
`intravenous (IV) naloxone; 2) 0.8mg intramuscular
`(IM)
`naloxone; 3) 0.8 mg intranasal
`(IN) naloxone; 4) 2 mg
`intravenous naloxone; and 5) 2 mg intranasal naloxone. An
`intramuscular injection of 2 mg was considered to be too large
`a volume to be administered by this route for a volunteer study so
`it was not included. All the subjects followed the same schedule
`in the previously mentioned order. There was a minimum 2-day
`washout period between doses of naloxone. After each
`administration of naloxone, blood was collected through the
`intravenous cannula into EDTA tubes at 5, 10, 15, 30, 45, 60, 90,
`120, 180, and 240 minutes after naloxone administration for
`a total of 10 samples per subject in each arm. Before any blood
`samples were taken, 5 mL of blood was drawn from the cannula
`and discarded. After collection, the blood was immediately
`centrifuged and the plasma frozen at –20°C. All samples were
`assayed using high-performance liquid chromatography.27–29
`The limit of detection for the assay was 0.3 mg/L and the limit of
`quantification (LOQ) was 1 mg/L. The intra- and interday
`coefficients of variation were 11.2% at 5 mg/L; 5.2% and 6.8%,
`respectively, at 12 mg/L; and 7.8% and 6.2% at 40 mg/L.28
`
`Data Analysis
`Pharmacokinetic analysis was undertaken using NON-
`MEM version 6.1.0 using the first-order conditional estimation
`
`q 2008 Lippincott Williams & Wilkins
`
`method for estimation with a G77 Compiler and enabled with
`Wings for NONMEM Version (6.13). Postprocessing analysis
`of data from NONMEM output was performed with
`Mathematica Version 5.1.1 (Wolfram Research, Inc., Cham-
`paign, IL). There were concentration measurements below the
`LOQ for all subjects and administration routes. Because of the
`assay error associated with values between the limit of
`detection and LOQ, we only included values above the LOQ.
`However, the first concentration below the LOQ was set to 0.5
`mg/L (LOQ/2) as described previously.30 This has been shown
`to reduce bias dramatically in the estimated parameters within
`the population pharmacokinetic model, especially clearance.
`One-,
`two-, and three-compartment models were
`assessed to decide the best structural model. The three-
`compartment model was parameterized as clearance (CL),
`central volume (V2), first peripheral compartment volume
`(V3), first intercompartmental clearance (Q3), second periph-
`eral compartment volume (V4), and second inter-compart-
`mental clearance (Q4). For the residual unexplained variability
`additive, proportional and combined error models were
`evaluated. Subsequently, data for IV, IM, and IN were then
`combined and modeled simultaneously. For the IM and IN
`dosing route, both first- and zero-order inputs were considered.
`To assess the relative bioavailability among IV, IM, and IN, the
`bioavailability was fixed to 1 for the intravenous route and the
`bioavailability estimated for IM and IN. Between-subject
`variability (BSV) was assumed to have a log-normal
`distribution and was added sequentially to the model.
`Model selection decisions were based on a number of
`different criteria,
`including a reduction in the objective
`function value produced by NONMEM (greater than 3.8 for
`P , 0.05), plots of predicted concentrations, weighted
`residuals, and visual predictive plots generated from simu-
`lations. Visual predictive check plots were obtained by
`simulating 1000 patients each with the four occasions during
`which data were obtained (0.8 mg IV, 2 mg IV, 0.8 mg IM, and
`2 mg IN) and then plotting the median and 90% percentile
`range.
`Between-occasion variability (intraindividual variabil-
`ity) was not included in the modeling process because each
`occasion represented a different dose or route of administra-
`tion. This meant that CL, V2, V3, V4, Q3, and Q4 were the
`same in each individual for each occasion allowing the
`estimation of the input processes. BSV was only estimated for
`the compartmental parameters, except BSV on input param-
`eters for intramuscular administration in which there were
`more data available to estimate this parameter more accurately.
`The influence of covariates was evaluated initially by
`visual inspection of plots of the covariates, post hoc estimated
`parameters, and a reduction of BSV and consideration of
`biologic plausibility. Weight and height were available for each
`subject so the influence of weight and lean body mass
`(calculated using the method suggested by
`(LBW2005)
`Janmahasatian et al31) was evaluated.
`
`Simulations
`To compare a common range of intravenous dosing
`schedules with IM and IN naloxone, the final model was used
`to simulate 1000 males with a weight of 70 kg from the typical
`
`491
`Nalox1027
`Nalox-1 Pharmaceuticals, LLC
`Page 2 of 7
`
`

`

`Dowling et al
`
`Ther Drug Monit  Volume 30, Number 4, August 2008
`
`parameter values, including BSV, for each of the following
`dosing schedules: 1) 0.4 mg, 0.8 mg, and 2 mg boluses of IV
`naloxone; 2) 0.8 mg, 1.6 mg, and 2.4 mg IM doses; and 3)
`2 mg, 4 mg, and 6 mg IN doses of naloxone. Concentration
`versus time plots were constructed for each of these scenarios
`with median (50% percentile) concentrations. Time to peak
`concentration and peak concentration were determined for IM
`and IN administration.
`A second set of simulations was undertaken to compare
`a previously recommended IV bolus and IV naloxone infusion
`protocol16 with a combination of IV and IM naloxone. Again,
`the final model was used to simulate 1000 males with a weight
`of 70 kg given 1) 0.4 mg IV naloxone with an infusion at 0.25
`mg/hr and a second bolus of 0.2g after 15 minutes; 2) 0.4 mg
`IV naloxone and 1.2 mg IM naloxone; 3) 0.4 mg IV naloxone
`and 1.2 mg IM naloxone delayed by 10 minutes; or 4) 0.4 mg
`IV naloxone, 1.2 mg IM naloxone, and a second 1.2 mg IM
`naloxone after 2 hours. Further simulations were undertaken
`with increasing doses as described by Goldfrank et al16 and
`IV/IM combinations to compare for larger bolus doses.
`
`RESULTS
`All six subjects completed five arms of the study and no
`adverse events occurred. Naloxone was only detectable above
`the LOQ in two subjects after the administration of 2 mg
`naloxone intranasally and not detectable above the LOQ in any
`patient receiving 0.8 mg, so only four occasions were available
`for analysis. One subject was removed from the IV 0.8-mg
`dose because of a drug administration error, which resulted in
`a spurious concentration–time profile. Thus, the final data set
`consisted of six patients (five occurrences with 0.8 mg IV, six
`occurrences with 2 mg IV, six occurrences with 0.8 mg IM,
`and two occurrences with 2 mg IN) with 128 concentration
`measurements (82 after IV administration, 39 IM, and seven
`IN).
`
`A three-compartment model with first-order input for IN
`and IM and a combined error model with a small fixed additive
`error model was found to best describe the data. BSV on CL
`and V2 (central volume) were included in the model. Addition
`of BSV on V3, V4, Q3, or Q4 did not reduce the objective
`function significantly or provided improbable estimates of BSV.
`The addition of BSV on absorption rate constant (Ka) for IM
`administration again did not provide a significant reduction in
`objective function value nor improve the model diagnostics.
`All patients were male so sex was not included. Visual
`inspection of CL and V2 versus weight and LBW2005 plots
`indicated an influence on both parameters. Covariates were
`evaluated in the model by including a modifying effect on the
`CL and V2. Weight and LBW were scaled with power
`functions such that:
`
`Š
`
`Š0:75; V2 ¼ u2 3½ WT
`CL ¼ u1 3½ WT
`WT70
`WT70
`The addition of weight and LBW2005 to CL on the model
`significantly reduced the objective function (DOBJ = –4.136).
`The final model included WT on V2 and LBW2005 on CL.
`The final covariate model was a three-compartment
`model with first-order absorption for IM and IN administration,
`
`492
`
`BSV on CL and V2, LBW2005 on CL allometrically scaled, and
`weight on V2.
`
`CL ¼ TVCL ðLBW2005=70Þ^0:75 EXPðETA½1ŠÞ
`V2 ¼ TVV2 ðWT=70Þ EXPðETA½2ŠÞ
`
`Plots of observed versus predicted concentrations and
`weight residuals versus predicted concentration demonstrated
`a good fit for the model (Fig. 1). In the WRES plot, most
`points appear to be normally distributed and centered around
`zero and most within three standard deviations. There were
`a couple of outliers that could not be explained or excluded.
`The typical population estimates and individual predicted
`values for the parameters are listed in Table 1. The relative
`bioavailability of IM and IN naloxone was 36% and 4%,
`respectively.
`From the final model, 1000 potential patients were
`simulated for each of the four arms of the study (0.8 mg and
`2 mg IV, 0.8 mg IM, and 2 mg IN). A visual predictive check
`of the final covariate model is presented in Figure 2. This shows
`that the final model shows very good ability to fit the central
`tendency of the data, whereas the 95th and 5th percentiles
`describe the variability well, with approximately five samples
`out of 128 (total number) of samples outside these percentiles.
`
`FIGURE 1. Plots of observed naloxone concentration versus
`posterior predicted concentrations (A) and weighted residuals
`(WRES) versus posterior predicted concentrations (B).
`
`q 2008 Lippincott Williams & Wilkins
`Nalox1027
`Nalox-1 Pharmaceuticals, LLC
`Page 3 of 7
`
`

`

`Ther Drug Monit  Volume 30, Number 4, August 2008
`
`Population Pharmacokinetics of Naloxone
`
`TABLE 1. Estimates for the Parameters from the Final Model
`and Individual Predicted Values for the Parameters.
`Mean Parameter Value
`95% Percentiles*
`
`CL (L/hr)
`V2 (L)
`V3 (L)
`V4 (L)
`Q3 (L/hr)
`Q4 (L/hr)
`t½a (hours)†
`t½b (hours)†
`t½g (hours)†
`21)
`Ka[im] (hr
`21)
`Ka[in] (hr
`Ftot[im]
`Ftot[in]
`BSV on CL (CV%)
`BSV on V2 (CV%)
`Prop Err
`Add Err
`
`91
`2.87
`1.49
`33.6
`5.66
`29.8
`0.016
`0.2
`1.0
`0.65
`1.52
`0.36
`0.038
`0.00581 (7.6%)
`0.25 (50%)
`0.101 (31.7%)
`0.001 (fixed)
`
`47.3–105
`0.75–4.8
`1–27.6
`6.7–200
`1.97–39.6
`4.82–44.3
`—
`—
`—
`0.44–0.79
`1.52–3.9
`0.18–0.45
`0.016–0.040
`0–0.09
`0.00006–0.66
`0.063–0.11
`
`*5th and 95th percentiles obtained by 1000 nonparametric bootstraps.
`†Derived parameters.
`CL, clearance; V2, central volume; V3, first peripheral compartment volume; V4,
`second peripheral compartment volume; Q3, first intercompartmental clearance; Q4, second
`intercompartmental clearance; BSV, between-subject variability; CV%, percent coefficient
`variation; t½, half-life; Ka, absorption rate constant; im, intramuscular; in, intranasal; Ftot,
`relative fraction absorbed; Prop Err, proportional error; Add Err, additive error.
`
`Simulations
`Figure 3 shows plots for simulations of 1000 individuals
`for a range of IV, IM, and IN doses of naloxone. The median
`time to peak concentration for intramuscular naloxone ranged
`from 12 minutes and for intranasal from 6 to 9 minutes
`(Fig. 3). Figure 4 shows a comparison of the naloxone infusion
`
`nomogram suggested by Goldfrank et al16 and the adminis-
`tration of simultaneous IV and IM naloxone.
`
`DISCUSSION
`This study shows that naloxone has a very poor
`bioavailability of 4% by the IN route and large doses that
`are physically impossible to administer intranasally using
`commercially available formulations are required to produce
`similar concentrations to those following IV naloxone.
`Intranasal absorption is rapid but does not maintain measur-
`able concentrations for more than an hour. Therefore, the IN
`route is the least useful route having poor bioavailability and
`not maintaining concentrations. Intramuscular naloxone has
`a bioavailability of 35% compared with IV therapy and
`maintains measurable concentrations for up to 4 hours after the
`dose. Although there is a slight delay in peak concentration
`after IM naloxone compared with IN, this is only approxi-
`mately 5 minutes. The combination of IV and IM naloxone
`provides both rapidly high and persistent plasma concen-
`trations of naloxone. Although the concentrations are not
`maintained as well as an IV infusion, there are detectable
`concentrations for up to 4 hours, which is long enough to
`maintain antagonism for many opioid drugs. These results
`were achievable because of the simultaneous analysis of IV,
`IM, and IN data to provide meaningful parameter estimates.
`These results could not be achieved through the standard two-
`stage approach, and the power of the NONMEM methodology
`to be able to gain meaningful parameter estimates in this
`manner has been discussed previously.32
`Hussain et al17 reported the pharmacokinetics of IN
`naloxone at a dose of 30 mg/kg in rats. In their study, the rats
`had their oropharynx occluded and their nostrils glued shut
`after naloxone was administered to prevent any loss of the drug
`from the nasopharynx. Hussain found the bioavailability of IN
`
`FIGURE 2. Visual predictive plots for
`0.8 mg naloxone intravenously (A),
`2 mg naloxone intravenously (B),
`0.8 mg naloxone intramuscularly
`(C), and 2 mg naloxone intranasally
`(D). The 5th percentile, 50th per-
`centile (median), and 95th percen-
`tile for the predicted concentrations
`are plotted against time and the
`observed data overlaid. The limit of
`quantitation was 1 ng/mL.
`
`q 2008 Lippincott Williams & Wilkins
`
`493
`Nalox1027
`Nalox-1 Pharmaceuticals, LLC
`Page 4 of 7
`
`

`

`Dowling et al
`
`Ther Drug Monit  Volume 30, Number 4, August 2008
`
`FIGURE 3. Concentration versus
`time plots for the 50th percentile
`for 1000 individuals given intrave-
`nous doses of naloxone 0.4, 0.8, and
`2 mg (A); intramuscular doses 0.8,
`1.6 and 2.4 mg (B); and intranasal
`doses 2, 4, and 6 mg (C).
`
`to IV naloxone. The rapid
`naloxone to be equivalent
`absorption profile in our study suggests that the small amount
`absorbed was through the IN route, whereas the remainder was
`swallowed. It is likely in our study in which IN naloxone was
`administered in an unoccluded nasopharynx that the majority
`of the naloxone was able to be swallowed and not able to be
`absorbed through the nasal mucosa. A previous study has
`shown that the bioavailability of oral naloxone is minimal (less
`than 1%) resulting from extensive first pass metabolism,
`supporting our observations.11,17
`There
`are
`no
`pharmacokinetic–pharmacodynamic
`studies of naloxone, making it difficult
`to determine the
`appropriate dose required to achieve a target concentration or
`concentration–time profile to maximize its antidote efficacy.
`We are aware of one study on the pharmacodynamic effects of
`naloxone measured by the reversal of morphine-depressed
`respiration.33 This study gives an indication of the potency of
`naloxone and the dose of naloxone required to produce a dose
`ratio of morphine can be calculated. For example, to reduce the
`effect of 20 mg of morphine to the effect of 4 mg (dose ratio
`of 5) 550 mg of naloxone is required, or 1100 mg to reduce the
`effect to 2 mg. However, this is in normal subjects and it is
`difficult to interpret this in the overdose setting with other
`opioid agonists such as heroin or methadone and in patients
`with significant tolerance to opioids.
`A previously developed approach to naloxone antago-
`nism by Goldfrank et al16 recommended dosing to be based on
`clinical response. The IV infusion protocol they recommended
`was based on the initial dose required to cause clinical reversal.
`We have shown that an IVand IM dose of naloxone may provide
`suitable antagonism for 2 hours and for another 2 hours with
`a repeat intramuscular dose (Fig. 4). This approach can also be
`based on the initial dose required to produce reversal of opioid
`effects by delaying the IM injection for 10 minutes, by which
`time the initial IV dose is established and three times this dose
`
`494
`
`can be given as an IM dose (Fig. 4). The administration of IM
`naloxone is easier, not requiring an infusion pump. However,
`with long-acting opioids such as methadone, or slow-release
`morphine preparations, in which naloxone may be required for
`over 4 hours, an infusion is the safest option.
`IN
`Clinical
`trials have demonstrated an effect of
`naloxone when administered with a Mucosal Atomiser
`Device.4,8,10 Barton et al8 studied the prehospital administra-
`tion of 2 mg intranasally (concentration 1 mg/mL). In their
`cohort of 30 patients, 11 (37%) responded to naloxone. Ten
`patients required only a single dose of IN naloxone with an
`average response time of 3.4 minutes. Subsequent to this
`study, Barton et al reported a similar group of patients and
`found that of the 52 patients who responded to naloxone, 43
`responded to IN naloxone alone.5
`Kelly et al9 compared 2 mg of IM and IN naloxone in
`patients with suspected opioid intoxication in the prehospital
`setting. Based on their clinical outcome, it appeared that there
`was a slower onset of action with IN compared with IM
`naloxone. This is not consistent with the pharmacokinetics of
`IN naloxone we have shown. However,
`their finding that
`a greater percentage of patients receiving IN naloxone
`required further doses to reverse sedation (13% versus 26%)
`is consistent with the shorter action of IN naloxone. In
`addition, the study showed that the IM group had a higher
`incidence of adverse effects, most
`likely as a result of
`precipitation of a withdrawal-like phenomenon.
`It is not completely clear why the clinical trials of IN
`naloxone are not consistent with our study. The explanation
`may be that only small amounts (0.05–0.1 mg) of naloxone are
`sufficient to produce opioid antagonism so that even with such
`poor bioavailability, 2 mg of IN naloxone is sufficient to be
`effective. In our study, IN naloxone was administered in awake
`healthy volunteers who, despite best efforts, swallowed
`a significant percentage of the administered drug that pooled
`
`q 2008 Lippincott Williams & Wilkins
`Nalox1027
`Nalox-1 Pharmaceuticals, LLC
`Page 5 of 7
`
`

`

`Ther Drug Monit  Volume 30, Number 4, August 2008
`
`Population Pharmacokinetics of Naloxone
`
`FIGURE 4. Concentration versus
`time plots for the 5th, 50th, and
`95th percentiles for 1000 individuals
`given 0.4 mg intravenous naloxone
`with an infusion at 0.25 mg/hr and
`a second bolus of 0.2 mg after 15
`minutes (A); 0.4 mg intravenous
`naloxone and 1.2 mg intramuscular
`naloxone (B); 0.4 mg intravenous
`naloxone and 1.2 mg intramuscular
`naloxone delayed by 10 minutes
`(C); or 0.4 mg intravenous nalox-
`one, 1.2 mg intramuscular naloxone,
`and then 1.2 mg intramuscular nal-
`oxone after 2 hours (D).
`
`in the nasopharynx, and the bioavailability of naloxone
`through the oral route was very low as a result of extensive
`first-pass metabolism. In unconscious patients with opioid
`overdoses and depressed oropharyngeal reflexes, less nasally
`administered naloxone may be swallowed, increasing the IN
`absorption and bioavailability. Differing concentrations of
`naloxone in the solution may also affect the bioavailability and
`high concentration solutions would be more ideal.
`A number of other
`reasons may also explain the
`inconsistency in our pharmacokinetic studies with clinical
`studies. In the clinical studies,
`the concentration of the
`naloxone varied. Higher concentrations of naloxone in the
`nasopharynx may result in greater mucosal absorption of the
`drug resulting from less volume being lost from swallowing.
`In our study, subjects received 5 mL of naloxone intranasally
`when receiving the 2-mg dose. All subjects noted some degree
`of pooling of naloxone in the pharynx.
`A major limitation of this study is that it was conducted
`in healthy volunteers rather than patients who had taken an
`overdose. The bioavailability of IN naloxone may be much
`higher in unconscious patients because animal studies suggest
`that if naloxone is confined to the nasopharynx, it will be
`almost completely absorbed.17 The low bioavailability is the
`result of poor gastrointestinal absorption and not being able to
`keep naloxone in the nasopharynx. Another limitation in
`defining the IN pharmacokinetics of naloxone was that after
`both doses of IN naloxone, blood concentrations were very
`low and only above the LOQ for two patients receiving 2 mg.
`Therefore, the bioavailability of 4% may be an overestimate
`and a more sensitive assay may allow a better estimate of the
`bioavailability. Using larger doses of naloxone intranasally
`would be difficult as a result of the increased volumes that
`would need to be administered, but higher concentrations
`could be used instead.
`The use of a population approach rather than the
`traditional
`two-stage approach allowed us
`to take full
`
`q 2008 Lippincott Williams & Wilkins
`
`advantage of the rich data from each patient on three to four
`different occasions, because we did not consider between-
`occasion variability. This allowed us to use data from the IV
`and IM routes to support the model for the IN data, which is
`not possible with other approaches. However,
`the small
`number of patients and uniformity of demographics between
`patients meant that BSV for CL was small and likely to
`underestimate the true population variability. LBW2005 and
`weight were incorporated in the estimation of CL and V2,
`respectively.
`Our study highlights the possible limitations of the IN
`route for the use of naloxone. Further pharmacokinetic studies
`are required in patients after opioid overdose to determine if
`the bioavailability is higher in this setting. Healthcare workers
`should be aware that the clinical response to IN naloxone
`might be less than parenteral routes of administration. In
`addition, our study demonstrates that, although IM naloxone
`has a lower bioavailability, drug concentrations are maintained
`much longer than for IV or IN administration. The com-
`bination of IV and IM naloxone appears to be an alternative to
`a naloxone infusion for short- to medium-acting opioids.
`
`ACKNOWLEDGEMENTS
`We thank the assistance of LMA PacMed for the
`donation of the Mucosal Atomiser Devices used in this study.
`We also thank the contribution of Chris Salonikas, Vicky Chia
`Hui Fang, and Rebecca Thomas from the SEALs Laboratory
`who processed the Naloxone samples. We thank Mr. Barrie
`Stokes for his assistance in analysis using Mathematica.
`
`REFERENCES
`1. Dore GJ, Thomas DL. Management and treatment of injection drug users
`with hepatitis C virus (HCV) infection and HCV/human immunodefi-
`ciency virus coinfection. Semin Liver Dis. 2005;25:18–32.
`2. Dale O, Hjortkjaer R, Kharasch ED. Nasal administration of opioids for
`pain management in adults. Acta Anaesthesiol Scand. 2002;46:759–770.
`
`495
`Nalox1027
`Nalox-1 Pharmaceuticals, LLC
`Page 6 of 7
`
`

`

`Dowling et al
`
`Ther Drug Monit  Volume 30, Number 4, August 2008
`
`3. Lahat E, Goldman M, Barr J, et al. Intranasal midazolam for childhood
`seizures. Lancet. 1998;352:620.
`4. Barton ED, Colwell CB, Wolfe T, et al. Efficacy of intranasal naloxone as
`a needleless alternative for treatment of opioid overdose in the prehospital
`setting. J Emerg Med. 2005;29:265–271.
`5. Loimer N, Hofmann P, Chaudhry HR. Nasal administration of naloxone is
`as effective as the intravenous route in opiate addicts. Int J Addict. 1994;
`29:819–827.
`6. Loimer N, Hofmann P, Chaudhry HR. Nasal administration of naloxone
`for detection of opiate dependence. J Psychiatr Res. 1992;26:39–43.
`7. Kelly AM, Koutsogiannis Z. Intranasal naloxone for life threatening
`opioid toxicity. Emerg Med J. 2002;19:375.
`8. Barton ED, Ramos J, Colwell C, et al. Intranasal administration of
`naloxone by paramedics. Prehosp Emerg Care. 2002;6:54–58.
`9. Wolfe TR, Bernstone T. Intranasal drug delivery: an alternative to
`intravenous administration in selected emergency cases. J Emerg Nurs.
`2004;30:141–147.
`10. Kelly AM, Kerr D, Dietze P, et al. Randomised trial of intranasal versus
`intramuscular naloxone in prehospital treatment for suspected opioid
`overdose. Med J Aust. 2005;182:24–27.
`11. Berkowitz BA. The relationship of pharmacokinetics to pharmacological
`activity: morphine, methadone and naloxone. Clin Pharmacokinet. 1976;
`1:219–230.
`12. Weinstein SH, Pfeffer M, Schor JM, et al. Absorption and distribution of
`naloxone in rats after oral and intravenous administration. J Pharm Sci.
`1973;62:1416–1419.
`13. Ngai SH, Berkowitz BA, Yang JC, et al. Pharmacokinetics of naloxone in
`rats and in man: basis for its potency and short duration of action.
`Anesthesiology. 1976;44:398–401.
`14. Berkowitz BA, Ngai SH, Hempstead J, et al. Disposition of naloxone: use
`of a new radioimmunoassay. J Pharmacol Exp Ther. 1975;195:499–504.
`15. Fishman J, Roffwarg H, Hellman L. Disposition of naloxone-7,8,3H in
`normal and narcotic-dependent men. J Pharmacol Exp Ther. 1973;187:
`575–580.
`16. Goldfrank L, Weisman RS, Errick JK, et al. A dosing nomogram for
`continuous infusion intravenous naloxone. Ann Emerg Med. 1986;15:
`566–570.
`17. Hussain AA, Kimura R, Huang CH. Nasal absorption of naloxone and
`buprenorphine in rats. Int J Pharmacol. 1984;21:233–237.
`
`18. Buajordet I, Naess AC, Jacobsen D, et al. Adverse events after naloxone
`treatment of episodes of suspected acute opioid overdose. Eur J Emerg
`Med. 2004;11:19–23.
`19. Hoffman RS, Goldfrank LR. The poisoned patient with altered
`consciousness. Controversies in the use of a Ôcoma cocktail.Õ JAMA.
`1995;274:562–569.
`20. Yealy DM, Paris PM, Kaplan RM, et al. The safety of prehospital naloxone
`administration by paramedics. Ann Emerg Med. 1990;19:902–905.
`21. Johnson C, Mayer P, Grosz D. Pulmonary edema following naloxone admin-
`istration in a healthy orthopedic patient. J Clin Anesth. 1995;7:356–357.
`22. Schwartz JA, Koenigsberg MD. Naloxone-induced pulmonary edema.
`Ann Emerg Med. 1987;16:1294–1296.
`23. Buchwald A. Naloxone use: side effects may occur. Ann Emerg Med.
`1988;17:765.
`24. C