Full neurological recovery following tricyclic overdose associated with absent brainstem reflexes (2024)

Abstract

We report the case of an 18-year-old gentleman who presented to the emergency department following a large overdose of amitriptyline. He was comatose on admission but his neurology further deteriorated to global loss of brainstem reflexes. He made a full but slow neurological recovery. First signs of improving neurology were seen on day 3 of admission; he was extubated on day 6 and discharged home on day 9. There is limited recent literature to help guide prognostication in this group of patients.

Keywords: Amitriptyline, brain stem, brainstem reflex, poisoning, overdose, prognosis, neurological recovery

Case presentation

An 18-year-old gentleman with a background of Raynaud’s phenomenon presented to the emergency department following a suspected mixed overdose of amitriptyline (6 g) and diazepam (450 mg). He had been found by friends unresponsive and grunting with empty packets of amitriptyline and diazepam. Initial Glasgow Coma Score (GCS) was 7 with the paramedic crew (E1 V2 M4) and the patient had a respiratory arrest en route to hospital. The paramedic crew reported that there had been no hypoxic episode. On arrival in the emergency department, the patient’s GCS was 3, ventilated via supraglottic airway, saturations 99% on 15 L oxygen via bag valve mask, HR 124 and BP 139/64. He had dilated pupils which were minimally responsive to light. His airway was secured via rapid sequence induction with propofol and rocuronium. Initial blood gases and admission blood tests showed normal metabolic profile, normal renal and liver function, normal paracetamol and salicylate levels and a raised white cell count 22.5. ECG was sinus rhythm with normal QTc (405 ms) but prolonged QRS (146 ms). Sedation was maintained with propofol and the patient transferred for CT head (no acute intracranial pathology) and then to the intensive care unit. Sedation was briefly discontinued on the unit but recommenced after a tonic clonic seizure which was terminated with 4 mg midazolam.

Following discussion with Toxbase team, alkalinisation with sodium bicarbonate was commenced. Noradrenaline was required to support blood pressure. Eight hours post admission, the QTc was 510 ms and magnesium sulphate was given.

Sedation was stopped on the morning ward round at which point no brainstem reflexes were present. An  EEG 18 h after admission displayed rhythmic activity at alpha frequency bilaterally, no response to painful stimuli, abnormal brain stem function and no seizure activity but bilateral median nerve somatosensory evoked potentials were intact at all levels including N20. Twenty-four hours following admission and 8 h after discontinuation of sedation, neurological examination demonstrated no motor response to pain and no brainstem reflexes; unreactive dilated pupils, no corneal reflex, no dolls eye movement, no respiratory effort and no cough reflex. He was generally hypotonic, lower limb reflexes were present but upper limb reflexes absent.

While the initial paramedic report was suggestive of mixed overdose, the diagnosis at this point was not clear. The patient’s parents and general practitioner confirmed that he was not on any prescription medications. Other diagnoses were considered and investigated accordingly. CT cerebral angiogram and MRI head showed no abnormalities. Lumbar puncture revealed no organisms, polymorphs/monocytes and lymphocytes all <1 × 10⁶/L, glucose 3.7 mmol/L, protein 0.78 g/L (likely slightly elevated by seizure activity). Further cerebrospinal fluid analysis showed normal immunoglobulin G, N-methyl-D-aspartate antibodies negative, voltage gated K channel antibodies negative. Serum levels of amitriptyline and nortriptyline levels were 1.125 mcg/mL and 0.568 mcg/mL 36 h after admission. Diazepam levels were not quantified. Quantification of serum tricyclic levels is not performed in our hospital and this report was not available for seven days.

Neurological progress was slow. On day 3, GCS remained 3 but the left pupil became reactive, corneal and cough reflexes were present and the patient was established on assisted spontaneous breathing ventilation. By day 4, both pupils were briskly responding to light, eyes open to painful stimuli and spontaneous non purposeful movement was present. By day 5, he localised painful stimuli and intermittently obeyed commands. He was successfully extubated on day 6 but required clonidine and haloperidol for fluctuating agitation and hallucinations for the following 48 h. By day 9, his neurology was back to baseline, his anxiety and agitation had settled and he was reviewed and discharged by the psychiatry team.

Tricyclic antidepressant toxicity

Tricyclic antidepressants (TCAs) are prescribed to treat depression, nocturnal enuresis and as an adjunct in chronic pain.¹ Amitriptyline remains a common agent implicated in intentional overdose despite the production of safer antidepressant agents. In this case, the patient had purchased amitriptyline over the internet. TCA exert their antidepressant effect predominantly via inhibition of serotonin and noradrenaline reuptake at nerve terminals. However, there are several other pharmacological actions of amitriptyline. Side effects are predominantly due to the anticholinergic actions (urinary retention, dry mouth, blurred vision). In TCA toxicity, the direct alpha blockade leads to hypotension and inhibition of fast sodium channels in the cardiac myocyte leads to prolongation of conduction which predisposes to dysrhythmias. Sedation, coma, seizures and respiratory depression are all known to be caused by TCA toxicity. Whilst case reports of loss of brainstem reflexes following tricyclic overdose do exist, they are rare and there is little in modern literature to help guide prognostication.^2,3

Neurological sequelae of TCA poisoning

Several case reports have documented the selective loss of individual brainstem reflexes following TCA overdose.^4,5 Case reports have shown absent corneal and pupillary light reflexes are successfully reversed by physostigmine whilst its effect on absent oculocephalic reflexes is variable. These findings have led authors to hypothesise that the oculocephalic reflex is particularly sensitive to the anticholinergic effects of TCA and ability to reverse with physostigmine is probably dependent upon dose of TCA.

Generalised loss of brainstem reflexes following TCA overdose has been described and is noted as a potential consequence in Toxbase literature.⁶ Yang and Dantzker reported a case of a 46-year-old female who had taken approximately 9 g of amitriptyline who had absent brainstem reflexes on admission, recovery of spontaneous ventilation at 24 h, recovery of corneal and pupillary reflexes at 48 h and return of oculocephalic reflex at 64 h. The patient regained full consciousness upon day 5.²

White described three cases of TCA overdose associated with absent brain reflexes, one of whom had global absence of brainstem reflexes. This patient had taken doxepin and had absent pupillary, corneal, oculocephalic reflexes, no response to pain and no spontaneous respiration. His pupils were pinpoint rather than dilated. Recovery commenced 6 h post ingestion and full consciousness was regained over the following 14 h. It later transpired that he had taken physostigmine with the doxepin which accounts for his miosis.³

A further case report has been published since this clinical episode occurred. Kansal etal.⁷ reported a case of complete loss of brainstem reflexes following overdose of 500 mg amitriptyline. Total serum amitriptyline (amitriptyline plus nortriptyline) was 3.43 mcg/mL which is roughly twice the level found in our patient despite an apparent ingestion of 6 g amitriptyline. This is in part explained by the fact the terminal elimination half-life of amitriptyline is 9–25 h and our drug levels were taken at 36 h.⁶ The large discrepancy between presumed ingested dose and measured level may also reflect differential metabolism as highlighted by a case report discussing prolonged toxicity in a patient with reduced CYP2D6 activity.⁸

Kansal etal. tabulated the findings of four case reports of absent brainstem reflexes associated with tricyclic overdose and concluded that signs of neurological recovery tend to occur on days 2–4 and complete neurological recovery has been seen between days 5 and 7. In our case report, we found first signs of neurological recovery on day 3 of admission which is in keeping with the findings by Kansal etal. Our patient was extubated upon day 6 but had significant agitation requiring treatment for a further 48 h. His neurology was fully recovered by day 9.

Loss of brainstem reflexes is considered a marker of poor prognosis in the comatose patient with less than 1% of patients ever regaining independent function.⁹ Clearly it is important to recognise that this loss of reflex may be reversible in certain cases and to give time for neurological recovery to occur where there may be a reversible underlying cause.

Conclusion

This is a fifth case report describing loss of brainstem reflexes in tricyclic overdose. Cases of amitriptyline ingestion have reported ingested doses between 500 mg and 9 g, with total serum amitriptyline concentrations 1.35–3.43 mcg/mL. There is little literature to guide prognostication in patients who have global loss of brainstem reflexes in the setting of TCA overdose. This case report is important as it demonstrates another recent example of the pitfalls of early prognostication in the poisoned patient with adverse neurological signs. This is particularly important in the setting of severe poisoning where episodes of hypotension, hypoxaemia and cardiac arrest may be wrongly assumed to be the root cause of brainstem dysfunction. Other differential diagnoses must be excluded whilst time is allowed for slow neurological recovery to occur. This case also demonstrates the importance of considering overdose in a patient who present with absent brainstem reflexes and coma where the history is unclear.

Consent

Written consent for publication has been obtained from the patient.

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Sam Waddy has received speaker fees from Fresenius.

References