TOXICOLOGY

Overdose with modified-release paracetamol (PanadolOsteo®) presenting to a metropolitan emergencymedicine network: A case seriesAndis GRAUDINS1,2

1Monash Emergency Medicine Program and Clinical Toxicology-Addiction Medicine Service, Monash Health, Melbourne, Victoria, Australia,and 2School of Clinical Sciences at Monash Health, Monash University, Melbourne, Victoria, Australia

AbstractBackground: There are currentlyno large cases series documen-ting poisoning with paracetamolmodified-release (Panadol Osteo®,GlaxoSmithKline, Sydney, NSW, Aus-tralia). Management guidelinesrecommend at least two serum para-cetamol concentrations 4 h apartand initiating treatment with N-acetylcysteine (NAC) if more than 10 gis ingested.Objective: To describe a cohort ofPanadol Osteo® poisoning and deter-mine if the management of identifiedcases was consistent with existingguidelines.Method: Descriptive retrospective caseseries presenting to a metropolitan hos-pital network with paracetamol poi-soning from October 2009 toSeptember 2013.Results: There were 42 cases ofPanadol Osteo® poisoning identi-fied. Twenty-nine patients (median in-gested dose 19 950 mg) were treatedwith NAC, of which 27 were acutesingle ingestions. Of NAC-treated pa-tients, 85% (23/27) had an initialserum paracetamol concentration thatwas above the nomogram line.However, 15% (4/27) had an initialnon-toxic concentration that later in-creased above the line. In 14 untreat-ed patients (median ingested dose

7980 mg), one was an unrecognisedlate line-crosser with initial non-toxicserum paracetamol concentration.Only 43% (6/14) had a repeat para-cetamol concentration measured. Threepatients had a 4 h paracetamol>500 μmol/L. Late line-crossing wasseen in the NAC-treated group at thislevel. In two untreated patients, NACshould have been commenced on thereported dose.Conclusion: Most patients present-ing with Panadol Osteo® poisoning re-quiring NAC treatment had an initialserum paracetamol concentration in-dicating need for treatment. A smallnumber of late treatment nomogramline-crossers was seen on repeat pa-racetamol estimation. The currentguideline for Panadol Osteo® poison-ing would have detected all cases re-quiring NAC treatment.

Key words: modified-release, N-acetylcysteine, overdose, paraceta-mol, poisoning.

IntroductionParacetamol continues to rank as oneof the most common medications im-plicated in deliberate self-poisoning(DSP) in developed countries, account-ing for around 10 to 20% of poison-ing presentations to Australian andNorth American EDs, and up to

50% of DSP cases presenting tohospitals in the United King-dom.1,2 Modified-release paracetamol(paracetamol-MR) (Panadol Osteo®,GlaxoSmithKline, Sydney, NSW, Aus-tralia) has been available on the Aus-tralian market since 2001 and listedon the Pharmaceutical Benefits Schemefor treatment of osteoarthritis since2008.The formulationcontains665 mgof paracetamol, 69% of which ismodified-release and 31% immediaterelease in a bilayer tablet. PanadolOsteo® pack size might contain up to96 caplets. As a result, there is the po-tential to ingest up to 63.84 g of

Correspondence: Professor Andis Graudins, Department of Emergency Medicine,Dandenong Hospital, David Street, Dandenong, VIC 3175, Australia. Email:andis.graudins@monashhealth.org

Andis Graudins, MBBS, PhD, FACEM, FACMT, Professor of Emergency Medicine andToxicology Research, Director of Toxicology Training.

Accepted 29 April 2014

Key findings• In this case series, patients

requiring treatment withN-acetylcysteine ingested amedian dose of Panadol Osteo®of 20 grams and initial serum pa-racetamol concentration indicat-ed the need for treatment in mostcases.

• However, a small number of pa-tients had an initial ’non-toxic’serum paracetamol concentra-tion which subsequently crossedthe nomogram line indicatingneed for treatment with N-acetylcysteine.

• Serum paracetamol and liverfunction tests should be meas-ured prior to cessation of thestandard 21 hour course ofN-acetylcysteine. A detectable pa-racetamol concentration or risingaminotransferases at this time in-dicate that a prolonged course ofN-acetylcysteine is required.

bs_bs_banner

Emergency Medicine Australasia (2014) 26, 398–402 doi: 10.1111/1742-6723.12249

© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

paracetamol from one pack. Individ-ual case reports and case series havereported on the pharmacokinetics ofparacetamol following massive inges-tions of this formulation.3–5 Ingestionof doses greater than 1 g/kg results inprolonged elevation of serum para-cetamol concentrations. Two volun-teer studies using non-toxic doses ofparacetamol-MR have also shown thatpeak serum paracetamol concentra-tions might be lower and occurlater with this formulation.6,7 However,larger case series examining thecharacteristics of poisoning with thisformulation have not been reported.In 2008, an Australasian consensusguideline for the treatment of para-cetamol poisoning was released.8 Thisincludes recommendations for the man-agement of paracetamol-MR over-dose. It recommends that patientswith a reported ingested dose ofparacetamol-MR more than 10 g or200 mg/kg (whichever the least) shouldbe treated with N-acetylcysteine (NAC).Also, two serum paracetamol concen-trations should be measured 4 h apart.If both concentrations are below thetreatment nomogram line and thepatient is well, NAC treatment can beceased. Repeat paracetamol estima-tions ensure that there has not been anylate line-crossing over the treatmentnomogram line that would indicate con-tinuation of NAC treatment.8 It isunknown how well this guideline is fol-lowed in clinical practice.

We undertook a retrospective auditof paracetamol poisoning in ourhealthcare network. The aim was toidentify and review all cases ofparacetamol-MR and describe thecharacteristics of poisoning in a largercohort, which might represent moretypical ingestion patterns than themassive overdoses previously report-ed. We also aimed to assess adher-ence to the abovementioned guideline.In particular, we wanted to ascertainwhether treatment with NAC was ini-tiated on reported ingested dose andwhether a repeat paracetamol concen-tration was measured in patients withan initially non-toxic level.

MethodThis was a retrospective clinical auditof patients identified with paracetamol-

MR poisoning presenting to MonashHealth EDs between 1 October 2009and 30 September 2013. MonashHealth comprises three acute care hos-pitals with EDs: Monash MedicalCentre Clayton (adult and tertiary pae-diatric EDs), Dandenong Hospital andCasey Hospital, Berwick (both withmixed adult and paediatric EDs). Thereare approximately 180 000 ED at-tendances to these three hospitals an-nually. The present study was approvedby Monash Health Human Researchand Ethics Committee as a low-riskclinical audit.

Patients presenting with bothimmediate-release and modified-releaseparacetamol poisoning were identi-fied from the ED electronic medicalrecord system (Symphony, Version2.29, Ascribe, UK) using the in-builte-audit tool. Records were searched byED discharge diagnoses ‘Poisoning/overdose, Paracetamol’ or ‘Poisoning/overdose Paracetamol with liverdamage’, or looking for free text intriage comments ‘paracetamol’ or‘Panadol’ and ‘overdose’ or ‘poison-ing’. An Excel spreadsheet was createdto extract relevant demographic andclinical data including: date of pres-entation, sex, age, type of paraceta-mol product reported either in EDtriage or clinical notes as well as in themedical admission notes if the patientwas admitted (specifically looking forthe words ‘Osteo’, slow-release,modified-release or 665 mg), report-ed co-ingestants, type of exposure toparacetamol (intentional/DSP, stag-gered ingestion, supratherapeutic ac-cidental ingestion, paediatric accidentalingestion), reported paracetamol doseingested, estimated weight and dose/kg of paracetamol (calculated from es-timated weight either in the notes orfrom the calculated dose of NAC givento the patient), timing of initial and

subsequent paracetamol concentra-tions from reported time of inges-tion, treatment with NAC and totaldose and duration of infusion, initialand subsequent serum alanineaminotransferase (ALT) concen-tration and time from reportedingestion, initial and subsequent in-ternational normalised ratio values andtime post-ingestion. Where differingdoses of paracetamol were reported tonursing staff or treating doctors, thelarger ingested dose was used. Simi-larly, time to paracetamol concentra-tion was taken as the latest possibletime post-ingestion if there was anyambiguity in the medical notes. Pa-tients with serial paracetamol estima-tions had these plotted graphically overtime until the final concentration wastaken or until the serum concentra-tion reached the limit of quantifica-tion (<66 μmol/L), which is the limitbelow which the serum paracetamolconcentration cannot be accuratelydetermined.

Descriptive data are reported aseither mean with 95% confidence in-tervals or medians and range for non-normally distributed data. Statisticalanalyses were performed usingGraphPad InStat Version 3.10(GraphPad Software, San Diego, CA,USA).

ResultsDuring the 4 year period, the searchterms identified 920 cases of paraceta-mol poisoning. The manual searchidentified 43 cases where PanadolOsteo® or sustained-release paraceta-mol or 665 mg tablet strength werenoted. Demographic data are summa-rised in Table 1. Patients were morelikely to be women (69%) and had amedian age of 27 years. Those pre-senting with an overdose requiring

TABLE 1. Demographic data of patients presenting with identified ingestionof modified-release paracetamol (n = 42)

Panadol Osteo®

Female gender 69% (29/42)Median age (range) (years) 27 (12–87)Co-ingested medications 57% (24/42)Deliberate self-poisoning 95% (40/42)

PANADOL OSTEO POISONING: A CASE SERIES 399

© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

treatment with NAC had a median re-porteddoseofparacetamol (19 950 mg)that was significantly greater than pa-tients presenting with a ‘non-toxic’ in-gestion (7980 mg) (Table 2).

In cases presenting with a singleacute ingestion, 85% (23/27) had a‘toxic’ paracetamol concentration(>1000 μmol/L at 4 h post-ingestion)evident on their first paracetamol es-timation. However, a small number ofNAC-treated patients (n = 4/27) hadan initial paracetamol concentrationthat was below the treatment line,which subsequently crossed the line ata later point (Fig. 1a). Two patients

presented following staggered inges-tions, where the nomogram could notbe used to assess toxicity: NAC wascommenced on one because of dose in-gested (310 mg/kg) and the otherbecause of an initially elevated serumALT (135 IU/L. N < 56).

Almost all treated patients withacute single ingestions (86%) had atleast two paracetamol estimationscollected. Of the 29 patients treatedwith NAC, 10 received extended du-ration of infusion (36 to 133 h). Fivepatients had an ALT concentrationabove the reference range (>56 IU/L)at 24 or more hours following NAC

commencement (Table 3). Two pa-tients had a peak ALT >1000 IU/L,reaching the definition of paracetamol-induced hepatotoxicity. One of thesepatients presented more than 8 h post-ingestion with a mildly elevated initialALT of 60 IU/L, 9.5 h post-ingestion.The other patient presented early withan ingestion of almost 1 g/kg ofPanadol Osteo® and still had a de-tectable serum paracetamol concen-tration at 32 h post-ingestion. Bothpatients had prolonged courses ofNAC. Serum ALT concentrations overtime are summarised in Figure 2 for allNAC-treated patients.

TABLE 2. Comparison of paracetamol doses and initial serum concentrations

NAC treatedn = 29

Untreatedn = 14

P value

Median (range) reported paracetamoldose ingested

19 950 mg(9975–75 840)

7980 mg(3330–17 000)

P < 0.0001†

Median (range) calculated dose per kg 290 mg/kg(135–945)

—

Initial mean (95% CI) serumparacetamol concentration

1131 μmol/L(879–1382)

405 μmol/L(262–548)

P < 0.0001‡

Median (range) time to initialparacetamol concentration

6 h(4–19)

4 h(3.3–20.5)

P = 0.008†

Number of patients having more thanone serum paracetamol estimation

25 (86%) 6 (43%) P = 0.008§Odds ratio = 8.4(95% CI: 1.8–37.1)

Median (range) NAC duration 21 h(21–133)

Not applicable

†Mann–Whitney test. ‡Unpaired t-test with Welch correction. §Fisher’s exact test.NAC, N-acetylcysteine.

TABLE 3. Summary of patients treated with N-acetylcysteine (NAC) developing elevated alanine aminotransferase (ALT)concentration

Age(years)/Sex

Paracetamoldose

Presentationparacetamol

concentration

NAC dose Initial ALT† Peak ALT‡

58/M 75 840 mg(945 mg/kg)

2161 μmol/L at 4.5 hpost-ingestion

80 g over 133 h 32 IU/L at 5 h 1232 IU/L at 89 h

17/M 19 950 mg(249 mg/kg)

1291 μmol/L at 9.5 hpost-ingestion

62 g over 101 h 60 IU/L at 9.5 hpost-ingestion

7516 IU/L at 67 h

17/F 17 600 mg(290 mg/kg)

738 μmol/L at 12 hpost-ingestion

30 g over 53 h 66 IU/L at 12 hpost-ingestion

128 IU/L at 24 h

28/M 23 275 mg(318 mg/kg)

1177 μmol/L at 6 hpost-ingestion

22 g over 21 h 37 IU/L at 6 hpost-ingestion

132 IU/L at 168 h

18/M 23 900 mg(332 mg/kg)

502 μmol/L at 4 h 28.8 g over 36 h 35 IU/L at 4 h 90 IU/L at 120 h

†Reference range: N < 56 IU/L. ‡ALT > 1000 IU/L accepted as the current threshold for defining paracetamol-inducedhepatotoxicity.12

400 A GRAUDINS

© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

Panadol Osteo® was identified as theformulation ingested in 14 untreatedpatients (Fig. 1b). Of these, less thanhalf (43%) had a second paracetamolconcentration. One patient was an un-identified later line-crosser (i.e. the pa-racetamol concentration crossed thetreatment nomogram line after an initialnon-toxic concentration). This patientwas intoxicated with co-ingestedethanol and had no ingested para-cetamol dose documented. Initialparacetamol concentration was675 μmol/L at 4 h and 536 μmol/L at8 h. This patient had a normal ini-tial ALT, was not treated withN-acetylcysteine and did not have anyfollow-up liver function tests. However,they did not re-present in the ensuingdays to any EDs in our network. Threeuntreated patients with only a singleparacetamol estimation at 4 h post-ingestion had serum concentrations thatwere similar to those in the treated pa-tients (539, 605 and 741 μmol/L at 4 h).Two untreated patients had a report-ed ingested dose greater than 10 g,which would have indicated a need forstarting NAC, based upon the currentguideline, and continuing this until arepeat concentration stayed below the

treatment line. However, they were dis-charged, had no follow-up liver func-tion tests and did not re-present to our

network in the ensuing days. There wereno patients with acute liver failure asdefined by the King’s College criteriarequiring transfer to the specialist liverunit, and no fatalities were noted.9

DiscussionParacetamol formulations, includingPanadol Osteo®, are easily acces-sible over-the-counter medications,which are commonly implicatedin DSP. Panadol Osteo® is also avail-able on prescription for treatmentof osteoarthritis. It can be purchasedin up to 96 caplet packs, with apotential for ingestion of up to63 840 mg from one pack. Althougha small number of case reports de-scribe massive ingestion of this for-mulation with prolonged elevation inserum paracetamol concentration andevidence of hepatotoxicity despite pro-longed NAC infusion,3–5 this is the firstcase series to describe poisoning withthis formulation in a larger cohort. Inthis series, patients requiring treat-ment with NAC ingested a median re-ported dose of around 290 mg/kg.Most patients also had an initial serumparacetamol concentration indicat-ing need for treatment with NAC usingthe current Australasian nomogram.8

In the patients with an initial ‘nontoxic’ serum paracetamol concentra-tion and subsequent line-crossing to atoxic concentration, treatment was in-dicated based upon the dose ingestedas per the current Australasian treat-ment guideline.8 In general, ‘line-crossers’ tended to have an initialserum paracetamol at 4 h above500 μmol/L and a reported dosemore than 10 g, suggesting that inthis group, there should be a highsuspicion for a subsequent ‘toxic’concentration.

Three of the patients who did notreceive treatment with NAC shouldhave been treated, based upon their re-ported dose or later high paraceta-mol concentration.

Interestingly, hepatotoxicity was un-common in this cohort, with only twopatients developing a peak ALT morethan 1000 IU/L. One patient had a‘massive’ ingestion (945 mg/kg) and theother was a late presenter (9.5 h post-ingestion). Based on previous reportswith massive paracetamol ingestion

Figure 1. Graphical semi-logarithmic representation of serum paracetamol concen-trations in patients presenting with reported ingestion of modified-release paraceta-mol (Panadol Osteo®). Panel (a) shows those patients treated with N-acetylcysteine.Panel (b) shows those patients who were not treated with N-acetylcysteine. The lineat y = 66 μmol/L represents the limit of quantification below which paracetamol con-centration is not accurately reported by the reference laboratory. The solid line repre-sents the current paracetamol treatment nomogram line used in Australasia and NorthAmerica.

Figure 2. Semi-logarithmic representa-tion of serial serum alanineaminotransferase (ALT) concentrations forpatients receiving N-acetylcysteine treat-ment. The dashed line represents the upperlimit of normal for the reference range(56 IU/L).

PANADOL OSTEO POISONING: A CASE SERIES 401

© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine

and/or late treatment, subsequent de-velopment of hepatotoxicity could havebeen predicted on presentation.5,10–12 Al-though there was a small number ofcases with mild increases in ALT, thesedid not fit the criteria of hepatotox-icity (ALT or AST >1000 IU/L).13

This case series is not without limi-tations. The data were extracted retro-spectively from ED case notes andscanned medical records of admittedpatients. Patients with incorrect ED orward discharge diagnoses might nothave been identified in the search ashaving paracetamol poisoning. Thereported ingested doses of paraceta-mol might be incorrect. Two patientshad a recorded ingested dose less than200 mg/kg but a serum paracetamolconcentration more than 1000micromole/L at 4 h. Interestingly, ex-posure to Panadol Osteo® was rec-ognised in the triage notes in a fewcases, yet was not documented by thetreating doctor. Similarly, the tablet sizewas noted as 665 mg in some pa-tients, but the fact that this wasmodified-release was not recorded. Asa result, there might have been evenmore cases of modified-release para-cetamol exposure that have gone un-recognised because of failure of takingan adequate medication history.

ConclusionIn cases of paracetamol poisoning, cli-nicians should determine whethermodified-release paracetamol (PanadolOsteo®) has been ingested. This in-formation will influence the manage-ment of the overdose. First, a singleparacetamol concentration might notexclude the potential for toxicity.Second, in larger overdoses, paraceta-mol concentration might remain de-tectable for longer than the durationof the standard 21 h NAC treatmentcourse, and extension of the infusionwill be required.

In this case series of modified-releaseparacetamol overdose, most patientsrequiring NAC treatment ingested wellin excess of 10 g of paracetamol-MR,and the need for treatment wasindicated on initial paracetamol esti-mation. However, paracetamol con-centration remained detectable for morethan 20 h in a number of patients re-ceivingprolongedNACinfusions.These

observations may be a reflection of thepack size of Panadol Osteo® com-monly available for over-the-counterpurchase in Australian pharmacies (96caplet pack; 63.84 g).

Given that the initial paracetamolconcentration might be non-toxic insome cases, NAC should be com-menced whether or not this falls abovethe treatment nomogram line. Early ad-ministration of NAC in paracetamol-MR overdose will prevent delays totreatment while serial paracetamolestimations are being measured toobserve for crossing of the treatmentnomogramline.Ultimately, if thehistoryis unreliable or timing of ingestion isunknown, a complete course of NACshould be administered.

Finally, in view of the modified-release nature of this formulation, itis important that both a serum para-cetamol concentration and ALT aremeasured before completion of thestandard 21 h NAC course. NACtreatment should be continued beyond21 h in patients with a detectable pa-racetamol concentration and/or ab-normal ALT at this time until thesenormalise. Consultation with a clini-cal toxicologist or poisons informa-tion centre might assist in makingdecisions on treatment continuation.

Competing interests

AG is a section editor for EmergencyMedicine Australasia.

References1. Kjartansdottir I, Bergmann OM,

Arnadottir RS, Björnsson ES. Para-cetamol intoxications: a retrospec-tive population-based study inIceland. Scand. J. Gastroenterol.2012; 47: 1344–52.

2. Dargan PIP, Jones ALA. Acetami-nophen poisoning: an update for theintensivist. Crit. Care 2002; 6: 108–10.

3. Roberts DM, Buckley NA. Prolongedabsorption and delayed peak para-cetamol concentration followingpoisoning with extended-release for-mulation. Med. J. Aust. 2008; 188:310–11.

4. Graudins A, Chiew A, Chan B. Over-dose with modified-release paraceta-mol results in delayed and prolonged

absorption of paracetamol. Intern.Med. J. 2010; 40: 72–6.

5. Graudins A, Pham HN, Salonikas C,Naidoo D, Chan B. Early presenta-tion following overdose of modified-release paracetamol (Panadol Osteo)with biphasic and prolonged para-cetamol absorption. N. Z. Med. J.2009; 122: 64–71.

6. Chiew A, Day P, Salonikas C, NaidooD, Graudins A, Thomas R. The com-parative pharmacokinetics ofmodified-release and immediate-release paracetamol in a simulatedoverdose model. Emerg. Med.Australas. 2010; 22: 548–55.

7. Tan C, Graudins A. Comparativepharmacokinetics of Panadol Extendand immediate-release paracetamol ina simulated overdose model. Emerg.Med. Australas. 2006; 18: 398–403.

8. Daly FFS, Fountain JS, Murray L,Graudins A, Buckley NA. Panel ofAustralian and New Zealand clini-cal toxicologists. Guidelines for themanagement of paracetamol poison-ing in Australia and New Zealand–explanation and elaboration. Aconsensus statement from clinicaltoxicologists consulting to the Aus-tralasian poisons information centres.Med. J. Aust. 2008; 188: 296–301.

9. Voigt MDM. Predicting death in pa-tients with acetaminophen-inducedacute liver failure: the King’s CollegeHospital model is on the SOFA, notthe mat. Liver Transpl. 2012; 18:384–6.

10. Graudins A, Pham HN, Salonikas C,Naidoo D, Chan B. Early presenta-tion following overdose of modified-release paracetamol (Panadol Osteo)with biphasic and prolonged para-cetamol absorption. N. Z. Med. J.2009; 122: 64–71.

11. Chan BS, Graudins A, Chiew A.Acetaminophen poisoning with a dif-ference. (Abstract). Clin. Toxicol.2008; 48: 601–2.

12. Green TJ, Sivilotti MLA, LangmannC et al. When do the aminotrans-ferases rise after acute acetaminophenoverdose? Clin. Toxicol. (Phila.)2010; 48: 787–92.

13. Prescott LF, Illingworth RN, CritchleyJA, Stewart MJ, Adam RD,Proudfoot AT. Intravenous N-acetylcystine: the treatment of choicefor paracetamol poisoning. Br. Med.J. 1979; 2: 1097–100.

402 A GRAUDINS

© 2014 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine