Proceedings of the Toxicology and Poisons Network Australasia (TAPNA) 2019 Scientific Meeting

Lack of utility of non-contrast head computed tomography in poisoned patients Richard McNulty, Lashnika Bandaranayake, Alex Garner, Tim Wong, Farhan Tahmid, Emily Symes, Mohammad Mohammad, Mark Salter, Naren Gunja Western Sydney Toxicology Service, Blacktown, Australia; Blacktown Mount Druitt Emergency Department, Blacktown, Australia; School of Medicine, Western Sydney University, Blacktown, Australia

The feasibility of illicit drug monitoring in an emergency department setting: data from a 12-month pilot study Sam Alfred 1 , Peter Stockham 1 , Emma Partridge 1 1 Royal Adelaide Hospital, Adelaide, Australia Introduction: Quality data regarding patterns of illicit drug use in the community is available from a range of sources; however, the accuracy of hospital level data is limited by the necessarily subjective attribution of a causative agent by treating clinicians. We report on a process capable of accurately profiling illicit drug use, and deliverable in the Emergency Department setting. Methods: Ethics approval was obtained with a waiver of consent. Patients who presented to the RAH Emergency Department with unusual or severe toxicity as consequence of drug use and a clinical requirement for intravenous access were enrolled. Laboratory logistics caped samples at 2 per week. The clinical data set targeted the setting of drug use, clinical toxidrome and outcome, and was designed to facilitate collection by clinical staff. Six mL of blood from subjects was placed in a study fridge prior to transfer to FSSA. Sample treatment and extraction methodologies included protein precipitation, liquid-liquid extraction and solid phase extraction. Analysis techniques and instrumentation included enzyme linked immunosorbent assay (ELISA), gas chromatography with flame ionisation detector (GC-FID), liquid chromatography with diode array detector (LC-DAD), liquid chromatography quadrupole time of flight mass spectrometer (LC-QTOF-MS) or triple quadrupole detector (LC-QQQ). Results: 84 patients were enrolled (54 male, 30 female) with a mean age of 31 years. Poly drug use was the norm with an average of 3.2 agents detected per patient, the most frequent of which were methamphetamine (38/84), amphetamine (33/84), benzodiazepines (31/84), ethanol (23/84, mean 0.152 g/dL), GHB (20/84), opiates (20/84), cocaine (13/84), and MDMA (13/84). 7 patients (0.8%) were clinically intoxicated without an identifiable agent on assay. We detected very few cannabinoid receptor agonists and no fentanyl derivatives, and are developing improved assays with a broader library to enhance detection. The majority of patients (42%) were managed and discharged from the Emergency Department or its Short Stay Ward, and Intensive Care was required in 19%. There were no deaths. Conclusion: Our pilot study demonstrates that it is possible to accurately profile illicit drug use associated with hospital presentation, and has provided the basis for a successful research grant that will support expanded enrolments across 4 major hospitals in Adelaide.
Intentional warfarin overdose treated with vitamin-k and enoxaparin with associated warfarin concentrations Introduction: Intentional warfarin overdose differs to over-anticoagulation and reversal guidelines are not designed for these scenarios. In addition, there are few cases describing alternative methods of anticoagulation in these scenarios if there is therapeutic need. We report a case of warfarin overdose treated with vitamin-K1 and enoxaparin. Case Report: A 57-year-old lady presented to an emergency department 3 h after ingesting 250 mg (3.1 mg/kg) warfarin, 5 g celecoxib and 500 mg tramadol. She had a history of Factor-V Leiden heterozygous mutation and pulmonary embolism. Her warfarin was substituted with enoxaparin injections 12-month prior, after she was diagnosed and treated for breast cancer. 10 mg Vitamin-K1 twice-daily was commenced and continued for ten days. She remained asymptomatic and did not develop any clinical toxicity. Her liver function (ALT 82 to 57unit/L) and renal function (GFR 80-90 ml/min/1.73 m 2 ) remained stable. After six days of observation and psychiatric assessment, she was discharged home. Her presenting INR was 1.0 and peaked at 1.6 (40-h post-OD). Serial warfarin concentrations were assayed using high-performance liquid chromatography tandem mass spectrometry (HPLC/MS). Warfarin concentration peaked at 25 mg/L (10-h post-OD) and decreased to 3.8mg/L (n:0.6-3.1 mg/L) on day-5 post-OD. Calculated elimination half-life is 38.6 h. Discussion: Warfarin-reversal guidelines are designed to manage over-anticoagulation, which is commonly caused by drug interactions or dose changes. Intentional overdose causes different challenges as coagulopathy can be prolonged without the ability to monitor warfarin concentrations. One method is to titrate Vitamin-K1 with regular INR checking if there is therapeutic need for anticoagulation. In our case, the patient was able to remain anticoagulated with subcutaneous enoxaparin, whilst reversal of the warfarin effect was managed with Vitamin-K1. This circumvented the issue of Vitamin-K1 induced warfarin resistance and the need for close titration with multiple blood tests. For example, earlier discharge to a psychiatric ward or community could be arranged with enoxaparin BID and a one-week course of Vitamin-K1. Conclusion: There is limited literature reporting warfarin concentrations in overdose. Our calculated apparent elimination half-life of 38.6 h falls in the range reported (17-44 h). This is in keeping with previous overdose cases and highlights the numerous days it takes to reach therapeutic concentrations post overdose. Introduction: Methaemoglobin (MetHb) is generated by the oxidation of haem iron to the ferric state. This inhibits the binding of oxygen to the haemoglobin and also hinders the release of oxygen to the tissues [1,2]. We report a rare case of a patient with methaemoglobinaemia induced by amyl nitrite ingestion. Case Report: A 27-year-old female was brought by ambulance to ED at midnight after recreational ingestion of amyl nitrite. She presented with vomiting, headache, dyspnoea and hypoxia which was unresponsive to oxygen therapy. Vital signs on arrival were heart rate 140 beats/min, respiratory rate 25 breaths/min, blood pressure 102/62 mmHg, temperature of 36.5 C, and SaO2 of 85% on room air. On physical examination, she was awake and alert with a GCS of 15 but pale and unwell looking. Her lung fields were clear to auscultation and her heart sounds were dual. There was no focal neurology. 12-lead ECG indicated a sinus tachycardia with 1 mm of anterolateral ST segment depression, with no QT prolongation. A chest X-ray showed clear lung fields. After cannulation, she was noted to have a very dark, almost black discolouration to her blood. An arterial blood gas sample indicated; pH 7.40, pCO2 27.4, pO2 157, HCO3 16, base excess -6.0, anion gap 7.2 and interestingly a MetHb of 48.4%. She was recognised to have methaemoglobinaemia from the ingestion of amyl nitrite. She was treated with high flow oxygen, IV dextrose infusion and a stat dose of methylene blue (1 mg/kg for estimated body weight of 70 kg). The MetHb dropped to 15% 30 minutes post administration and to 3% at 3 h. She recovered well, with no cardiac or neurological sequelae and was discharged the next day. Conclusion: Methaemoglobinaemia is an uncommon presentation seen in Australian Emergency Departments. Clues for the diagnosis include decreased oxygen saturation by pulse oximetry resistant to supplemental oxygen and discolouration of blood samples. The administration of methylene blue is the antidote of choice and reversed the effects in this patient, with a good outcome.

Prolonged myelosuppression after low-dose methotrexate ingestion in end-stage renal failure
Emma Maguire 1

NSW Health, Zetland, Australia
Background: Bone marrow suppression is well documented from methotrexate (MTX) use, particularly in the setting of renal impairment, interacting medications, or staggered overdose [1]. This report details a patient with end-stage renal failure who developed prolonged myelosuppression after a delayed presentation of deliberate self-poisoning of 250-500 mg oral methotrexate. Case Report: A 31year-old female presented with lethargy, nausea and mucositis seven days after a deliberate self-poisoning (DSP) of 250-500 mg methotrexate, 150 mg Quetiapine and 100 mg promethazine. She had a medical history of severe systemic lupus erythematosus, with nephritis and renal failure (baseline GFR $ 14 mL/min) for which she was due to start dialysis, and was prescribed erythropoetin. Methotrexate had been ceased four months prior following episodes of thrombocytopenia and febrile neutropenia. Examination revealed temperature 36.8 C, BP 178/ 24, HR 94 and SpO2 98% RA. Clinical examination was normal aside from large bilateral patches of buccal mucositis. Initial investigations found electrolytes and renal function to be at baseline (Urea 20.3, Cr 414 eGFR 12 mL/min). However, she had suppression of neutrophil and platelet cell lines (Hb 122 g/L, Plt 17 Â 10 9 /L, WCC 2.6 Â 10 9 /L neutrophils 1.56 Â 10 9 /L). Methotrexate level was <0.1micromol/L. Platelets were 322 Â 10 9 /L eleven days prior. She commenced intravenous folinic acid 15 mg QID, with transient resolution of neutropenia and platelet recovery to 17 Â 10 9 /L. On Day 4 of admission, she transitioned to oral folinic acid 15 mg QID and was discharged home for outpatient monitoring. Her mucositis improved markedly over the first week, and while platelets sustained their recovery, her neutrophils declined further following discharge. She also developed alopecia. This was managed expectantly within the community. She continued folinic acid oral treatment for sixteen days, until sustained recovery of all cell lines. Discussion: Renal failure is a known risk factor for serious toxicity in MTX poisoning. This patient required prolonged treatment with folinic acid for 23 days after an intentional overdose. There was an approximate 37% decrease in monthly averages between pre-and post-intervention periods, P < 0.0001. Significant decreases were seen across all exposure types (accidental, therapeutic error, intentional and adverse reaction). Ibuprofen/codeine exposures decreased by 65%, paracetamol/codeine/sedating antihistamine exposures decreased by 66%, and paracetamol/codeine ( 15mg) declined by 50%. There was no significant increase in higher strength combinations, with a monthly mean of 61.2 calls pre-and 61.9 calls post-intervention. When all other pharmaceutical opioids were considered together, there was a 12.8% increase, 158.5 (95%CI 151.5-165.6) to 178.8 (95%CI 169.4-188.1), P ¼ 0.003. In particular, there was a trend towards increased tramadol exposures, however this was not significant (p ¼ 0.07). Discussion: Up-scheduling codeine to "Prescription Only" resulted in a significant decrease in calls to NSWPIC across all exposure types. Ibuprofen/codeine and paracetamol/codeine/antihistamine combinations were most affected. There does not appear to be a shift towards exposures with higher strength codeine preparations (>15 mg per dose unit, which have always been Prescription Only). There was an increase in overall exposures with other pharmaceutical opioids, particularly tramadol, however this not enough to account for the decrease in codeine calls ($60 fewer calls/month for codeine, $20 more calls/ month for other opioids). This implies that up-scheduling has curbed use and misuse of codeine containing products.
Rescheduling decreased codeine-related presentations to a clinical toxicology unit Keith Harris 1,2 , Andrew Jiang 1 , Robert Knockel 2 , Katherine Isoardi 1,2 Background: On 1st February 2018, due to concerns over misuse and dependence, codeine was rescheduled to a prescription only medication. Previously, products containing <30mg codeine were available overthe-counter. This study aims to examine the impact rescheduling had on codeine-related presentations to a Clinical Toxicology Unit. Methods: This was a retrospective review of codeine-related presentations to the Princess Alexandra Hospital Clinical Toxicology Unit in Brisbane before and after codeine rescheduling on the 1st February 2018 to a Schedule 4 (prescription only) medication. The unit's relational database was searched for ingestions containing codeine during the 24-month period from the 1st February 2017 to the 31st January 2019. Patient demographics, ingestion details, clinical effects, treatments and complications were extracted from the patient's medical records. Results: There were 240 codeine-related presentations over the 2-year period, with 163 presentations in the year before and 77 presentations in the year following the rescheduling (p 0.0001). The median dose of codeine ingested was similar at 200 mg (IQR 130-390, range 6-1800) pre-rescheduling compared to 270 mg (IQR 120-600, range 24-2240) post-rescheduling. The absolute number of presentations relating to preparations containing 30 mg of codeine was similar (57 pre v 62 postrescheduling) however, the proportions increased significantly from 35% to 81% following rescheduling. Treatment with naloxone (11(7%) pre v 9(12%) postrescheduling, p ¼ 0.32) and N-acetylcysteine (47(29%) pre v 17(22%) post-rescheduling, p ¼ 0.35) were similar in the pre and post implementation period respectively. Intensive care episodes (1 pre vs 5 post, p ¼ 0.03) and psychiatric admissions (12(7%) pre vs 20(26%) post-rescheduling, p ¼ 0.0002) were increased post rescheduling. Deaths were similar with one before and two after the change. Discussion: There was a reduction in overall codeine-related presentations following rescheduling due to less over-thecounter preparation poisonings. The absolute number of poisonings with 30 mg preparations was similar following rescheduling. There was more acuity in the post-rescheduling cohort with a significant increase ICU and psychiatric admissions which may reflect increased intent coupled with a dominance of high dose codeine formulations.

Carbamazepine toxicity and management of refractory seizures
Elizabeth Doran 1 1

Royal Brisbane & Women's Hospital, Herston, Australia
Introduction: I present a case report of life-threatening carbamazepine overdose complicated by refractory seizures requiring multiple agents amid concerns of a potential interaction with levetiracetam. Case Report: A 43-year-old man presented post-intentional overdose of 28 g Carbamazepine Controlled Release, estimated to be 254 mg/kg. Time of ingestion was unknown but may have been up to 6 h previously. He arrived GCS12 before rapid deterioration to GCS7 with tonic clonic movements and intermittent cogwheel rigidity. Initial QTc prolongation resolved with 100 mmol sodium bicarbonate otherwise he was haemodynamically unremarkable. He was intubated before decontamination with nasogastric activated charcoal. Initial carbamazepine level was 36 mg/L, peaking at 40 mg/ L. Levels fluctuated until combination charcoal haemoperfusion, CRRT and enteral activated charcoal resulted in a level <12mg/L on Day 6. Nonepileptiform clinical seizures continued despite levetiracetam, propofol and midazolam. Concerns were raised that phenytoin contributes further sodium channel blockade, and that current levetiracetam may be exacerbating symptoms of carbamazepine toxicity. Lacosamide was commenced on suggestion of Neurology. Seizure activity ceased on Day 6 when carbamazepine level was 13 mg/L. ICU discharge occurred Day 9 however he remains a medical inpatient. Discussion: Carbamazepine is an antiepileptic drug which blocks voltage-and usedependent sodium channels to prevent continued action potentials. In overdose it produces anticholinergic and CNS effects with dose-dependent progression of symptoms. Slow and erratic absorption means some features of overdose can appear <4 h of ingestion, however anticholinergic symptoms complicate absorption with peak effects not evident for up to 12 h, especially with controlled release formulations. This patient's carbamazepine level peaked 10 h after potential ingestion. CNS toxicity includes nystagmus, ataxia, sedation, mydriasis and myoclonus; epileptic patients such as mine have an increased seizure risk. Despite multiple agents, the patient continued to have seizures throughout his ICU admission until effective elimination resulted in a nontoxic level. The addition of lacosamide was suggested due to the absence of documented interactions despite sodium channel activity. There is a pharmacodynamic interaction between levetiracetam and carbamazepine, however it is unclear whether this potentiation of carbamazepine toxicity is clinically significant in overdose. Both this, and concerns of additive sodium channel blockade with phenytoin, warrants further investigation and discussion.
Cold-water extraction of codeine works … most of the time Background: Combination analgesics containing codeine and paracetamol until recently were commonly used over-the-counter medications. Cold-water extraction is a means of obtaining pure codeine from paracetamol co-formulations. It relies on differing drug solubilities; tablets are ground up, mixed with warm water, chilled, then passed through filter paper to separate out the codeine. There is limited data on the use of this method. This study aimed to characterise hospital presentations relating to cold-water codeine extraction. Methods: This is a retrospective case series of patients presenting to the Princess Alexandra Hospital Clinical Toxicology Unit following recreational codeine ingestions obtained through cold-water extraction. The unit's relational database was searched for cold-water codeine extraction for the 3-year period from January 2016-January 2019. The unit has permission from the Metro South Human Research Ethics Committee to use their database and patient medical records for research. Results: There were 10 patients (7 males, median age of 26 years [range 16-36 years]) over the 3-year period, however only one of these presentations was following the rescheduling of codeine to a prescription only medication in February 2018. The ingestion was staggered in 3 cases. The median time to presentation following ingestion was 9 h (range 2-79 h). The median dose of codeine ingested was 330 mg (range 180-1800 mg), correlating with a median paracetamol dose of 12250 mg (range 6000 mg-60000 mg). One patient received naloxone after ingesting 180 mg codeine. A paracetamol level was detectable in 3 cases with levels of 26 mg/L, 35 mg/L and 42 mg/L following cold-water extraction of 15000 mg, 20000 mg and 8000 mg paracetamol, respectively. Only one of these levels (11-h level of 35 mg/L) was over the nomogram line. One patient presented 48 h following cold-water extraction of 10000 mg/300 mg paracetamol/codeine with hepatotoxicity. His peak ALT and INR was 1240 IU/L and 1.8, respectively. 4 patients received N-acetylcysteine. All patients were discharged home. The median length of stay was 8.9 h. Discussion: Cold-water codeine extraction is performed successfully, with the liberation of codeine without detectable paracetamol in the majority of cases. Opioid toxicity requiring naloxone reversal was uncommon in our small series. Inadequate technique can result in unintentional paracetamol toxicity.

A case of liraglutide toxicity
Joe Rotella 1 , Anselm Wong 1

Austin Health, Heidelberg Heights, Australia
Introduction: Liraglutide is a newer class of antidiabetic medication and is a glucagon-like peptide 1 (GLP-1) agonist. There have been few cases of reported overdose, especially in non-diabetic patients. Case Report: We report a case of a 36-yearold male who presented to our Emergency Department following an accidental overdose of liraglutide. He was prescribed liraglutide for weight loss (BMI 30.7) and did not have a history of diabetes mellitus. He was not on any other prescribed medication and he denied any recreational substance use. The patient commenced liraglutide on the day of presentation and misread the instructions provided with the medication. Instead of the prescribed dose of 0.5 mg, he injected 3.0 mg subcutaneously. Three hours later he developed profuse vomiting and epigastric pain. As it was not resolving, he initially presented to the Emergency Department at 12 h post injection and left after being given 4 mg ondansetron, which temporarily resolved his symptoms. He subsequently represented at 16 h post overdose with severe vomiting episodes followed by dry retching. He did not report any other symptoms. On examination, he had normal vital signs and some mild epigastric tenderness. He had a normal full blood count, urea, creatinine and electrolytes and was treated with metoclopramide, oral and IV ondansetron. He was admitted overnight, symptoms had improved the following day and he was discharged 26 h post injection. There were no episodes of hypoglycaemia on admission. Discussion: Liraglutide stimulates glucose-dependent insulin secretion from pancreatic islets, suppresses glucagon secretion, slows gastric emptying and increase satiety. Liraglutide is slowly absorbed via subcutaneous injection with a plasma half-life of approximately 13 h prior to metabolism in the plasma by Dipeptyl-peptidase-4 and neutral endopeptidase (NEP) and excreted in the urine. This likely contributed to the ongoing symptomatology in our case. There is limited information available regarding the toxicity of liraglutide. Supportive care, treatment for hypoglycaemia and antiemetics are the mainstays of management.
Triclopyr and diethylene glycol monoethyl ether poisoning treated with ethanol therapy and dialysis: a case report with serial serum concentrations Introduction: Triclopyr (TCP) herbicide poisoning is considered to have low toxicity. We report a case of poisoning with triclopyr and diethylene glycol monoethyl ether (DEGEE) developing hyperchloraemic metabolic acidosis. Case Report: A 34-year-old female presented to the emergency department 3 h after ingesting 250 mL of Yates Tree & Blackberry Weed-KillerV R , containing 5% Triclopyr (12.5 g) in 60% DEGEE. She developed nausea, vomiting, chest and abdominal pain one-hour after ingestion. On arrival, conscious state was normal, temperature 37 C, pulse 90bpm, BP 120/70 mmHg, respiratory rate 12/min, oxygen saturation 98% in room air. There was no oropharyngeal corrosive injury. Over the next 4 h, a hyperchloraemic metabolic acidosis developed: pH 7.16, bicarbonate falling from 16 to 9 mmol/L, base excess -11 to -17 mmol/L and serum chloride increasing from 111 to 121 mmol/L. Ethanol therapy was commenced 6-h post-ingestion and continued for 16 h. Twelve-hours post-ingestion sustained low efficiency haemodialysis (SLED) was administered for 6 h. She required intubation for these interventions. Recovery was uneventful, with no end organ injury. She was discharged home well on day five. Serial TCP and DEGEE serum concentrations were assayed using liquid chromatography tandem mass spectrometry (LCMS-MS). The apparent elimination half-life of TCP pre-SLED was 8.5 h (232 mcg/mL at 5-h post OD to 152 mcg/mL at 10.2-h post OD). Calculated elimination half-life during SLED was 17.5 h (96.5 to 67.6mcg/mL over a 9-h interval). DEGEE apparent elimination half-life was 2.1 h pre-SLED (224 to 40.5 mcg/mL in 5.2-h) and 2.6 h during SLED (33.3 to 3 mcg/mL in 9-h). Discussion: TCP is commonly regarded as a low toxicity herbicide, but isolated reports of severe metabolic toxicity exist. Hyperchloraemic non-anion gap metabolic acidosis developed in this case possibly due to the high chloride content of TCP. DEGEE is partially metabolised by alcohol dehydrogenase (ADH) and toxicity of its metabolites is not well characterised. As literature reporting toxicity of both compounds was limited and metabolic acidosis was present, we commenced ethanol therapy to antagonise ADH and SLED in an attempt to enhance TCP and DEGEE elimination. Dialysis has previously been used in TCP poisoning; however, SLED did not affect apparent elimination half-life of either compound. Introduction: Barium toxicity is an uncommon poisoning. Barium salts are used in a number of industries, including mining, ceramics, plastics and adhesives and as a green colouring in fireworks. It is also used as a rodenticide. It's primary mechanism of toxicity is by blocking passive efflux potassium channels without affecting the Na/K-ATPase pump resulting in an increase in intracellular potassium and extracellular hypokalaemia. Here we report a case of deliberate selfpoisoning of barium carbonate and perform a review of the poisoning and management strategies. Case Report: A 44-year-old man self-presented to hospital 6 h following an ingestion of 10 g barium carbonate mixed with coffee. On presentation he had had profuse vomiting and felt generally weak. He decreased power globally and hyporeflexia in his upper and lower limbs. His ECG was in sinus rhythm with a first-degree heart block and long QT. He was found to have hypokalaemia with a level of 2.4 mmol/L. He was managed supportively with antiemetics and supplemental potassium, receiving 100 mmol of intravenous potassium chloride in total. After 24 h he was asymptomatic, with a potassium of 3.8 mmol/L and discharged home. Discussion: Soluble barium salts are highly toxic and clinical effects include hypokalaemia, flaccid paralysis, cardiac arrest and death. Management of barium toxicity is largely supportive, focusing on the correction of hypokalaemia. The use of sulphate salts orally has been described as a way of precipitating insoluble barium sulphate in the gastrointestinal tract and preventing absorption. Haemodialysis has been demonstrated to improve the elimination and can be considered in severe poisoning.