Too many pills to swallow: A case of a mixed overdose

Abstract

Propranolol is a highly lipid-soluble beta-receptor antagonist. We describe a case of mixed overdose, including propranolol, amlodipine and olanzapine, resulting in severe resistant hypotension which was successfully treated. A 21-year-old student ingested 6.4 g of propranolol, 280 mg of amlodipine and 560 mg of olanzapine. The patient was brought to the emergency department and exhibited signs of severe systemic toxicity – profound hypotension and circulatory collapse, respiratory depression and coma. The patient had conventional therapy but failed to respond to this and was therefore commenced on IntraLipid infusion, high-dose insulin infusion and inotrope infusion. An endoscopy was performed in intensive care which revealed a large drug beozar – this was removed. We believe that this combination of infusions with early endoscopy could be beneficial in treating similar patients in the future and present the first case of a propranolol drug bezoar.

Case presentation

A 21-year-old student was brought to a district general hospital following an attempted suicide by ingestion of tablets. Paramedics calculated up to a maximum of 160 × 40 mg propranolol SR, 28 × 10 mg amlodipine and 56 × 10 mg olanzapine may have been taken. The exact timing of the overdose could not be established but was estimated as between 3 and 6 h before admission. There was no other medical history of note.

On arrival to the emergency department, the patient was hypotensive with a blood pressure of 90/45 mmHg and a heart rate of 80 beats/min (bpm). The airway was patent and ventilation on room air was adequate, with an SpO₂ of 99%. The peripheries were cool, and a Glasgow Coma Scale (GCS) of 13/15 (E4, V3, M6) was observed. On both wrists, there were deep lacerations but no significant bleeding. Initial investigations including biochemical, haematological, thoracic radiograph, electrocardiogram and venous blood gas were unremarkable. Blood tests for paracetamol and salicylate levels were also taken. Gastric lavage was not performed.

Fluid resuscitation was started (500 ml boluses of PlasmaLyte) and glucagon (10 mg/kg) for the beta-blocker overdose. Despite this, the hypotension worsened to 75/40 mmHg and GCS to 9/15 (E3, V1, M5). The pulse rate remained stable at 80 bpm.

A decision was made to intubate for airway protection, and despite peripheral boluses of 10–100 µg of adrenaline, the patient remained hypotensive.

An adrenaline infusion was commenced post insertion of central and arterial line. Despite infusion at 1 mg/kg/h, the patient remained hypotensive with a mean arterial pressure (MAP) <40 mmHg.

ToxBase was consulted which suggested starting an insulin infusion. A loading dose of 10 units/kg was given IV followed by an infusion of 1 unit/kg/h. This appeared to stabilise the patient temporarily prior to transfer to intensive care.

On admission to ICU, despite an adrenaline infusion running at 4 mg/kg/h, and ongoing insulin infusion, the patient remained hypotensive with a MAP of 50. She was therefore started on IntraLipid™ (Fresenius Kabi Ltd, Runcorn, Cheshire, UK) with an initial bolus of 1.5 ml/kg which was repeated twice and then a maintenance infusion of 0.25 ml/kg/min. This was continued over two days with 425 ml of 20% IntraLipid infused in total.

Despite these measures, there was no improvement in cardiovascular status. Following previous anecdotal experience of the ICM consultant of delayed endoscopy and drug bezoar removal, it was decided to perform an oesophago-gastro-duodenoscopy (OGD). This revealed a large drug bezoar (see Figure 1), which is estimated to be about 200 ml in total of tablets.

Figure 1.

Drug beozar at OGD.

The local extracorporeal membrane oxygenation (ECMO) team was also contacted in the event of further deterioration for the consideration of mechanical cardiovascular support.

She started to make a gradual improvement with reduction in inotropic requirement. All inotropes were stopped 12 h after OGD, and the patient made a full medical recovery. She was reviewed by a psychiatrist before discharge who decided that an inpatient psychiatric admission was required.

Pathophysiology

The Scottish scientist James Black developed propranolol in the 1960s and was awarded the Nobel Prize in Medicine for this discovery in 1988.¹ Propranolol is a non-selective beta-blocker that acts to antagonise the effects of both adrenaline and noradrenaline on both the β1- and β2-adrenergic receptors.² They reduce the activity of adenylcyclase which in term results in a fall in second messenger intracellular cAMP levels. This can affect the myocardial contractility, lungs, peripheral blood vessels, central nervous system and glucose metabolism.

It is rapidly and completely absorbed orally and is a highly lipophilic drug. There is a significant first pass metabolism, but as the oral dose is increased beyond 30 mg, the removal process becomes saturated and the hepatic extraction falls. This results in a longer half life of up to 12 h.

In an Australian study carried out between 1987 and 1995, it was noted that in beta-blocker overdoses, propranolol overdoses were over-represented, generally taken by younger people and were the only beta-blocker to be associated with death. Toxicity developed within 6 h of ingestion.³ These findings have been corroborated in several other similar studies. Complications precipitated by overdose result from excessive beta-adrenergic blockade.

Critical care treatment

The initial management of massive mixed overdose with propranolol is based on the principles of resuscitation, aiming to restore perfusion to organs by increasing cardiac output.

Securing the airway, obtaining intravenous access and providing cardiac monitoring should be established in the emergency department. Many patients respond to fluid resuscitation alone and can be observed; however, larger overdoses will require more specialist treatment and discussion with toxicologists. The selection of other therapies will be based on the severity of clinical presentation and magnitude of overdose. Ideally, these would be administered sequentially, which would enable the physician to assess their success or failure; however, this is often not possible due to the patients being critically unwell.

We discuss the merits of the standard approach below and suggest early institution of intravenous lipid with consideration of endoscopic removal of pharmacobezoars.

Role of glucagon

Traditionally, glucagon has been used in the management of beta-blocker overdose. Glucagon can exert a transient improvement, but this is limited in prolonged treatment by tachyphylaxis. Theoretically, beta-blockers bind to specific adrenoceptors to reduce the activity of adenylcyclase with consequent fall in second messenger intracellular cAMP levels. Glucagon acts to activate adenylcyclase, causing a rise in cAMP levels, by a mechanism independent of adrenergic effects.⁴

However, there are no randomised controlled trials supporting this mechanism or the efficacy of glucagon in beta-blocker overdose. The Australian toxicology community rarely recommends the use of glucagon as an antidote due to no proof of effectiveness, difficulty in obtaining the doses required, increased risk of vomiting and concomitant risk of aspiration and risk of distraction from giving other therapies.⁵

The combination of biological plausibility, practical experience and case reports suggests that glucagon is a suitable treatment for patients who have taken beta-blockers, and therefore, it is still used in practice. However, it lacks solid supporting evidence and there is evidence that it does not affect cardiac output.⁶

Therefore, we suggest an initial bolus of glucagon, with anti-emetic, should be trialed in patients with symptomatic propranolol overdoses with the facilities to commence alternative treatments ready.

Role of insulin

High-dose insulin is being used more readily to treat beta-blocker overdose in those who are haemodynamically unstable. It can be titrated to maintain a stable blood pressure but typically requires doses much higher than those used to treat diabetic ketoacidosis. Euglycaemia is maintained with a continuous intravenous infusion of glucose. It is important to monitor and supplement potassium as needed.

High-dose insulin is well tolerated in patients and it appears that it exhibits a dose-response curve.⁷ The mechanism of action is not fully understood, but beta-blockers are known to inhibit pancreatic insulin release, thus reducing available glucose and contributing to a lower cardiac output. Cardiac myocytes can utilise a variety of substrates and by improving glucose availability, insulin appears to improve inotropy.

Role of intravenous lipid infusion

Intravenous lipid infusion has been used in various drug overdoses and has particular relevance in local anaesthetic toxicity. ToxBase recommends considering IntraLipid in patients with a history of overdose of beta-blockers, who have cardiotoxic symptoms not responding to conventional treatment.

However, there are no human studies to confirm the use of intravenous lipids in beta-blocker overdose. The mechanism of action is thought to be that the intravenous lipids provide a lipid sink for fat-soluble drugs and therefore remove them from target organs. There have been several case reports suggesting the benefit of intravenous lipids in this context and studies on rodents and rabbits that confirm a benefit from hypotension in propranolol toxicity.⁸ In addition, there are several case reports describing effective treatment of beta-blocker overdose with the combination of IntraLipid infusion and high-dose insulin.⁹

Role of endoscopy

We proceeded to endoscopy as the previous methods of reducing toxicity had failed. We suspected the formation of a pharmacobezoar based on previous anecdotal knowledge of a similar case where a large drug bezoar was found 24 h post overdose in a patient that had been transferred for ECMO for mechanical cardiovascular support. We postulated that gastric absorption was delayed due to a persistent low cardiac output. Although pharmacobezoars appear to be rare in drug overdose, our patient was critically ill and the potential benefit of reducing continued drug release through the removal of the remaining tablets was thought to outweigh the potential risks of endoscopy.

There is some anecdotal evidence to support the removal of tablets using endoscopy; however, the correct method for gastrointestinal decontamination in patients who have taken massive overdoses of extended-release tablets is not well established.¹⁰ Toxicologists continue to consider the use of endoscopy or even surgery to remove the tablets remaining in the stomach.¹¹ Many drugs are able to form pharmacobezoars, with or without enteric coats.¹¹

In the context of overdose, the main factors to consider before selecting decontamination methods are drug toxicity, clinical status of the patient and potential for pharmacobezoar.

Conclusion

The American Association of Poison Control Centers Toxic Exposure Surveillance System recorded 164 fatal beta-blocker-related deaths in the years between 1985 and 1995.¹² Of these deaths, propranolol was implicated in a disproportionately high percentage of fatalities and cases.¹² The prognosis is worse with the addition of a cardioactive drug or neuroleptic.¹³

Routine care for haemodynamically unstable patients post beta-blocker overdose with fluid, inotropes, glucagon and calcium is often not successful. There is now established evidence from animal-based studies regarding the use of high-dose insulin therapy and intravenous lipid emulsion therapy. This has been supported with case reports in humans as is thus recommended by ToxBase. Gastric washout should not be routinely performed but may be considered for patients who present within 1 h following ingestion of a large overdose, and there is no recommendation to perform endoscopy.

Based on previous anecdotal evidence of a large drug bezoar many hours after a significant overdose of anti-hypertensives, it was decided to proceed to endoscopy in this case. A large drug bezoar was indeed identified 24 h following ingestion, and following removal, there was a gradual and sustained improvement in the patient’s condition. We would recommend to consider mechanical methods of drug removal in patients who remain haemodynamically unstable to reduce further drug absorption. Further studies looking into this are required.

Consent

Consent for publication has been obtained from the patient.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.