Severe acute drug‐induced dystonia in the post‐operative period requiring tracheal re‐intubation

Summary

Ondansetron is a highly selective 5‐hydroxytryptamine receptor antagonist and the most commonly used anti‐emetic for the prevention of postoperative nausea and vomiting. Ondansetron has a low affinity for dopamine receptors and so extrapyramidal side effects are rare. Here, we present the case of a 14‐year‐old girl who developed a severe post‐operative acute dystonic reaction which included oculogyric crisis. We believe that ondansetron was the most likely cause, although propofol may have been a synergistic or alternative causative agent. The patient had no significant past medical history and had previously undergone two uneventful general anaesthetics which included both ondansetron and propofol. The prolonged duration and severity of the reaction and failure to fully respond to specific treatments resulted in the need for tracheal intubation and transfer to a paediatric intensive care unit. She subsequently recovered uneventfully with no ongoing neurological sequalae. Ondansetron‐induced dystonic reactions are rare and unpredictable and can occur in patients who have previously received the drug without complication. They are thought to be caused by an imbalance between inhibitory and excitatory neurotransmitters in the extrapyramidal system. Specific treatments include anticholinergics, antihistamines and benzodiazepines.

Introduction

Ondansetron has a low affinity for dopamine receptors and so extrapyramidal side effects are rare [1]. However, several case reports describe extrapyramidal reactions in which the causative agent is either ondansetron or less commonly, propofol [1, 2, 3, 4, 5, 6, 7]. Reported extrapyramidal effects include dystonia, whereby sustained muscle contractions cause repetitive movements or abnormal postures; and oculogyric crises, characterised by paroxysmal, conjugate and tonic eye deviation.

Although ondansetron has only a weak affinity for the dopamine receptor, it may cause extrapyramidal effects indirectly, via disruption of the normal dopaminergic‐cholinergic balance. The maintenance of posture and the regulation of involuntary motor function is achieved via a dynamic balance between the inhibitory neurotransmitters dopamine and gamma‐aminobutyric acid, and the excitatory neurotransmitter acetylcholine within the extrapyramidal tracts [1, 2, 3, 6]. Ondansetron is thought to indirectly reduce inhibitory dopaminergic activity [1, 4, 5, 6, 8] and the resultant reduction in D2 dopaminergic activity may also increase stimulatory cholinergic activity [9]. This disturbance in dopaminergic‐cholinergic balance can rarely result in extrapyramidal symptoms [1].

Treatment of dystonia related to ondansetron or propofol requires discontinuation of the suspected trigger and restoration of the dopaminergic‐cholinergic balance in the basal ganglia with anticholinergics or antihistamines. Benzodiazepines may be used for symptomatic relief [1, 3]. There are reports in the literature of drug‐induced dystonic reactions that are resistant to the above specific treatments [1, 2]. In our case, the patient demonstrated minimal symptomatic improvement after these interventions.

Case reports exist of acute dystonic reactions to ondansetron, propofol or both, but none severe enough to require further induction of anaesthesia, tracheal intubation and intensive care admission [1, 2, 3, 4, 5, 6, 7]. We highlight learning points relating to the severity of the reaction and its management.

Report

A 57 kg, 14‐year‐old female was admitted electively for removal of metalwork from both tibial tuberosities under general anaesthesia. She had no significant past medical history, took no regular medications and her only reported adverse drug reaction was to penicillin. This was her third anaesthetic in 2 years under the care of the same anaesthetist, who had used a similar anaesthetic technique on each occasion.

The patient received intravenous fentanyl 125 μg and propofol 200 mg for induction, and general anaesthesia was maintained with 1 MAC of sevoflurane in oxygen and air, with the patient breathing spontaneously via a supraglottic airway. For analgesia, she was given paracetamol 1 g intravenously and subcutaneous local anaesthetic infiltration of 40 ml 0.25% levobupivacaine. For antiemesis, she was given intravenous dexamethasone 3.3 mg and ondansetron 4 mg, both at the start of surgery.

The surgery took 40 min and her recovery was initially uneventful. However, 8 min after emergence from general anaesthesia she developed abnormal posturing and involuntary repetitive tonic movements in which her right arm was held in extensor spasm and her left arm in flexion, with her neck extended and laterally rotated, and her eyes spasmodically moving into a bilateral upwards gaze. These movements lasted for 10 to 20 s and occurred approximately twice per minute, causing arm and back pain. She remained alert but with some difficulty in communicating and was able to follow commands with her unaffected limbs. She was tachycardic at around 130 beats per minute, but vital signs otherwise remained stable. This continued for an hour, at which time she developed a pyrexia of 38.2 °C, possibly attributable to the muscle spasms. Serial venous blood gases and blood glucose values were normal. There were no episodes of tongue biting or incontinence.

The provisional diagnosis was of an acute dystonia with oculogyric crisis and torticollis, most likely caused by ondansetron, although propofol may have been a synergistic agent. Differential diagnoses included a functional syndrome, seizure activity, intracranial pathology, serotonin syndrome or neuroleptic malignant syndrome, but the clinical presentation made these less likely.

The patient's symptoms showed no improvement after a 2 mg intravenous dose of midazolam which was followed by a further 1 mg midazolam 10 min later, to alleviate distress and discomfort. Specific treatments for acute dystonia were given over the following hour: intravenous promethazine 10 mg which had no effect, and two doses of intravenous procyclidine 5 mg which only mildly improved her symptoms.

After 2 h there was little improvement despite the specific treatments above. The local paediatric retrieval service advised that the duration and severity of the reaction warranted induction of general anaesthesia and tracheal intubation, followed by computed tomography (CT) imaging of the brain, antibiotic cover for meningoencephalitis with 2 g IV ceftriaxone, and transfer to the local paediatric ICU. This plan was discussed and agreed upon by the patient's parents. Repeat anaesthesia was induced with 100 mcg fentanyl, 200 mg propofol and 50 mg rocuronium and maintained with a 1% propofol infusion at 30 ml.h⁻¹ and 500 mcg.ml⁻¹ alfentanil infusion at 2.5 mg.h⁻¹. The CT imaging showed no abnormality and electrolytes were normal. Following weaning of sedation, her trachea was extubated uneventfully 9 h after the onset of symptoms. Following tracheal extubation, she had only a few tonic movements before falling asleep. On waking up, her symptoms had completely resolved, to the extent that she was able to enjoy her hobby of crochet. She had few memories of the events, possibly due to the early use of midazolam, and no ongoing neurological sequalae were observed. She was discharged the following day, with no recurrence of symptoms on follow‐up.

An incident form was completed and the reaction was highlighted on the patient notes. The patient, her parents and her GP were informed about the reaction in writing. An Medicines and Healthcare Products Regulatory Agency (MHRA) Yellow Card for drug side effect reporting was completed.

Discussion

Ondansetron is a highly selective 5‐Hydroxytryptamine (5‐HT) 3 receptor antagonist and the most commonly used anti‐emetic for the prevention of postoperative nausea and vomiting [1, 8]. It acts centrally by blocking serotonergic transmission at the 5‐HT3 receptors in the chemoreceptor trigger zone of the area postrema of the fourth ventricle and the vomiting centre in the nucleus tractus solitarius. It also has a peripheral action by antagonising 5‐HT3 receptors on vagus nerve terminals in the gastrointestinal tract, the synapses of which are within the nucleus tractus solitarius of the brainstem [8].

Ondansetron has a low affinity for dopamine receptors and therefore, extrapyramidal effects including acute dystonic reactions were not anticipated or reported in early clinical trials conducted in 1989 [10]. Its effectiveness and favourable side‐effect profile compared to other anti‐emetics, including metoclopramide, prochlorperazine and droperidol, have led to its widespread use in both oncology and anaesthesia [8, 10]. However, there have since been several case reports of ondansetron‐related acute dystonic reactions [1, 2, 4, 5, 6, 7].

Although it only has a weak affinity for the dopamine receptor, ondansetron may cause extrapyramidal effects indirectly, via disruption of normal dopaminergic‐cholinergic balance. The maintenance of posture and the regulation of involuntary motor function occurs within the extrapyramidal tracts, via a dynamic balance between the inhibitory neurotransmitters dopamine and gamma‐aminobutyric acid (GABA) and the excitatory neurotransmitter acetylcholine [1, 2, 3, 6]. There is evidence of serotonergic innervation to both the basal ganglia and the related nuclei of the limbic system, and serotonin is thought to exert a facilitatory role on dopaminergic transmission in the extrapyramidal system (Fig. 1a). By antagonising the 5‐HT3 receptor, ondansetron may indirectly reduce this inhibitory dopaminergic activity [1, 4, 5, 6, 8]. Additionally, the dopamine D2 receptor is thought to inhibit acetylcholine activity. Therefore, blocking serotonergic transmission not only decreases the inhibitory neurotransmitter dopamine, but the resultant reduction in D2 dopaminergic activity may also lead to an increase in stimulatory cholinergic activity (Fig. 1b) [9]. Rarely, this disturbance in dopaminergic‐cholinergic balance can result in extrapyramidal symptoms, which include dystonia [1].

Figure 1.

(a) Maintenance of posture and regulation of involuntary motor function occurs via a dynamic balance between inhibitory neurotransmitters (including dopamine) and excitatory neurotransmitters (acetylcholine) in the extrapyramidal system. Serotonin acting at the 5‐HT3 receptor facilitates dopamine release. Dopamine acting at the D2 receptor inhibits acetylcholine release. (b) Ondansetron antagonises the 5‐HT3 receptor, reducing its facilitation of dopamine release. Decreased dopaminergic activity reduces the D2 receptors' inhibition of acetylcholine. Alteration in the balance between inhibitory dopamine and increase in excitatory acetylcholine may result.

Ondansetron‐induced dystonia has been reported more commonly with high doses, for example, those used in the management of nausea and vomiting in patients undergoing chemotherapy. However single doses of 4 mg have also been reported as the cause of acute dystonic reactions [1, 4, 5, 6, 7]. Extrapyramidal side effects are unpredictable, with cases reported in patients who have previously been administered the drug uneventfully [2, 4]. Graded reactions have been reported in which repeated administrations of ondansetron resulted in increasing severity of the extrapyramidal reaction, from none to mild and then severe [6]. There are also two case reports of ondansetron exhibiting cross‐reactivity with other agents known to precipitate extrapyramidal symptoms. The individuals are likely to have had increased susceptibility; one had previously suffered an acute dystonic reaction to prochlorperazine and another to metoclopramide. Both went on to exhibit a similar dystonic reaction when administered ondansetron [5, 7]. Our patient had been previously administered the drug uneventfully and had also had a second 4 mg dose of ondansetron in the post‐operative recovery room at the time of her second anaesthetic with no extrapyramidal sequalae.

Propofol is a rare causative agent for acute dystonic reactions [2, 3]. It is thought that this is also due to disruption in basal ganglia neurotransmitters, resulting in an increase in excitatory cholinergic compared to inhibitory dopaminergic output [3]. Schramm and Orser demonstrated that propofol‐induced acute dystonic reactions are most likely to occur in young female patients on emergence from day case anaesthesia [3]. In our case, we do not know whether the cause of the reaction was ondansetron or propofol, or a combination of the two. However, acute dystonias are less frequently linked to propofol than to ondansetron, and so we believe that ondansetron was the most likely culprit for the extrapyramidal manifestations.

Treatment of extrapyramidal side effects related to ondansetron and propofol requires discontinuation of the suspected trigger and restoration of the dopaminergic‐cholinergic balance in the basal ganglia. Specific treatments comprise anticholinergics including benztropine and procyclidine, which re‐establish neurotransmitter balance by blocking excitatory intrastriatal cholinergic activity. Antihistamines including diphenhydramine and promethazine are also effective, but this may be due to their anticholinergic properties. Benztropine exhibits anticholinergic, antihistaminic and dopaminergic properties and may therefore be the most effective specific treatment, with symptoms often diminishing or resolving within 5–30 min. Benzodiazepines have been used for patients who do not completely respond to the more specific antidotes [1, 3]. Variation in individual responses to treatment is noted, including case reports of dystonic symptoms completely resistant to benzodiazepines and procyclidine [1, 2]. Different combinations may need to be tried to control the extrapyramidal side effects. In our case, the patient demonstrated only partial symptomatic benefit with procyclidine and none from midazolam or promethazine.

There is a report in the literature of an ondansetron‐induced acute dystonic reaction characterised by limb spasticity, decorticate posturing and an irregular respiratory pattern which prohibited weaning from ventilatory support and required a brief intensive care admission. Tracheal extubation in that patient was undertaken the following morning when the symptoms had completely resolved [4]. However, in that report, the patient's underlying pathology of normal pressure hydrocephalus and surgical intervention to the abdominal component of her ventriculoperitoneal shunt may have caused small increases in intracranial pressure and this may have been a contributory factor in the severity of that presentation [4]. In our patient, tracheal extubation was undertaken 9 h after the occurrence of symptoms, following which she had only a few further tonic movements and was completely symptom‐free by the following morning. In our case, probable ondansetron‐induced dystonia occurred in a patient with no predisposing conditions or surgical intervention which could have contributed to the presentation. The severity necessitated anaesthesia, tracheal intubation, transfer and paediatric intensive care admission.

Although rare, extrapyramidal reactions to perioperative ondansetron can occur and cause delayed recovery from anaesthesia and distress to the patient. As ondansetron is a commonly prescribed antiemetic such reactions are occasionally seen and clinicians caring for patients should consider ondansetron and related drugs as a potential cause of unusual postoperative ocular and motor symptoms and signs.