Case of paediatric catatonia precipitated by antipsychotic withdrawal in a child with autism spectrum disorder

  1. Mohd Faisal 1,
  2. Vishnu Pradeep 1 and
  3. Susan O'Hanrahan 2
  1. 1 Department of Psychiatry, University Hospital Limerick, Limerick, Ireland
  2. 2 Child and Adolescent Mental Health Services, Unit 6 Quin Road Business Park, Ennis, Ireland
  1. Correspondence to Dr Mohd Faisal; mohd.faisal@hse.ie

Publication history

Accepted:02 Apr 2021
First published:23 Apr 2021
Online issue publication:23 Apr 2021

Case reports

Case reports are not necessarily evidence-based in the same way that the other content on BMJ Best Practice is. They should not be relied on to guide clinical practice. Please check the date of publication.

Abstract

A 13-year-old girl with moderate intellectual disability and autism spectrum disorder (ASD) was admitted to the paediatric high-dependency unit following an 8-week history of altered mental status and motor behaviour. Her symptoms emerged followed shortly after discontinuation of risperidone, an atypical antipsychotic previously commenced to manage disruptive behaviour associated with ASD. On physical examination, the patient presented with negativism, grimacing, automatic obedience, waxy flexibility and ambitendency. Blood tests, neuroimaging and lumbar puncture failed to reveal an acute infectious or neurological precipitant. She responded immediately to a trial of intramuscular lorazepam titrated to a total daily dose of 12 mg. This case presents challenges of accurately diagnosing and managing catatonic symptoms in adolescent patients with ASD. We also discuss the potential risk of precipitating catatonia following the discontinuation of antipsychotic treatment that has been prescribed for a prolonged duration.

Background

Catatonia is a neuropsychiatric disorder characterised by immobility, negativism, posturing, staring, rigidity, stereotypy, mannerisms, perseveration, repetitive speech and imitative movements.

It is one of the most severe and underdiagnosed psychiatric syndromes in adolescents.1 Adolescents presenting with catatonia associated with either psychotic or mood disorder have a 60-fold increased risk of premature death when compared with age-matched peers.2 It can occur in the setting of psychiatric, neurologic, autoimmune, metabolic and toxic aetiologies and is often mistaken for other conditions. Catatonia has also been linked to the use of psychostimulants such as ecstasy, phencyclidine, inhalant, steroid and neuroleptics. Emerging literature also describes catatonic symptoms to be precipitated by abrupt withdrawal of benzodiazepines.3 A recent case report by Waserstein et al 4 highlighted the development of catatonic symptoms following antipsychotic discontinuation in an elderly psychiatric patient.4 Another case report by Bilbily et al 5 discusses the case of a young patient with repeated episodes of catatonia precipitated by clozapine withdrawal.5 Our case report is the first to describe catatonic features arising from discontinuation of an antipsychotic outside of clozapine in an adolescent.

Available literature discussing catatonia in children and adolescents remains sparse, but does highlight specific epidemiological and clinical differences between adult and paediatric catatonia. The most common underlying psychiatric disorders of catatonia in children and adolescents are schizophrenia, whereas mood disorders are more common in adults.6 It is common in boys during adolescence, while adult catatonia is more prevalent in females.7 Catatonic symptoms are usually consistent across age groups. However, features such as reduced motor and verbal responsiveness, excitement, agitation, increased repetitive behaviours and urinary incontinence may be more prominent in children and adolescents.8 Moreover, catatonia is also commonly associated with antipsychotics use in early-onset schizophrenia, acting as a double-edged sword, which in turn increases the incidence of catatonia in children.9

Catatonia can also occur in adolescents with neurodevelopment disorders such as autism spectrum disorder (ASD) and Down syndrome.10 In these cases, catatonic symptoms can be easily misattributed for stereotypies and minimal spontaneous speech, which are common features of ASD. In a population-based longitudinal study of 120 individuals between the ages of 17 and 40 with ASD diagnosed in childhood, found at least 13 (12%) to have catatonia-like deterioration and a further 4 (3%) to have features of a milder form.11 Separating overlapping symptoms of neurodevelopment disorders and catatonia pose particular challenges for clinicians leading to delayed diagnosis. Catatonic symptoms such as echolalia, speech delays and inappropriate laughing can be misinterpreted by parents or primary care as adolescent oppositional or uncooperative behaviour. Catatonic symptoms can be challenging to be recognised by parents or primary caregivers; therefore, this can delay or misdirect essential treatment that may be lifesaving in specific situations. Scales such as the Bush-Francis Catatonia Rating Scale (BFCRS) or Paediatric Catatonia Rating Scale (PCRS) help address ambiguity in such circumstances and allow clinicians to stratify the severity of catatonia, and facilitate timely intervention.

In terms of treatment, there is limited research looking into the effective dose and duration of lorazepam treatment in the acute and maintenance phase of paediatric catatonia. The use of high dosage lorazepam and electroconvulsive therapy (ECT) are the current treatments of choice.12 ECT is reserved for the most severe of cases or in situations where other approaches have proved ineffective. A rapid but short-lived response to 1–2 mg intravenous lorazepam assists with providing diagnostic clarity. A reduction of 50% or more, in the Catatonia Rating Scale, indicates further treatment with increasing doses of intramuscular lorazepam. In a case series by Fink et al, 13 doses of lorazepam up to 24 mg/day were prescribed, with a maintenance dose of lorazepam continued up to 6 months.13 Treatment with high-dose benzodiazepine helps treat catatonic symptoms adequately, but the long-term effects of this treatment modality, especially in adolescent patients with neurodevelopment disorders, remain unexplored.

Case presentation

A 13-year-old girl with moderate intellectual disability (ID) and ASD was admitted to the paediatric high-dependency unit following an 8-week history of progressive mutism and prominent psychomotor retardation following discontinuation of risperidone 0.25 mg at night. There was no rigidity observed on examination. The patient was commenced on risperidone to manage behavioural symptoms associated with her ASD and remained on risperidone 0.25 mg at night for 18 months.

Her presenting difficulties coincided with a 2-week period during which her mother returned to their home country following her grandmother’s death. As a result, urgent respite care was organised at the local residential unit. During this time, her risperidone was discontinued under the supervision of her treating psychiatrist as there was no longer a clinical indication to continue with long-term use of risperidone. Following this, her nursing staff observed a gradual change in her behaviour. She had presented with insomnia, increased salivation, mutism and immobility in the first week following the discontinuation of risperidone. Features of echopraxia, ambitendency, sweating, increased salivation and insomnia were also observed in the first-week period following discontinuation of risperidone. The authors acknowledged that a mix of stressful events, including the absence of her mother, changes in the home environment, along with concomitant discontinuation of risperidone, might have contributed to her rapid deterioration. On further evaluation, the presence of insomnia, increased salivation and sweating, along with emerging catatonic symptoms pointed towards features of risperidone withdrawal. The timeline of the treatment course with risperidone and its discontinuation leading to catatonia has been highlighted in figure 1.

Figure 1

Timeline of clinical manifestation of catatonia, treatment response and discharge. ASD, autism spectrum disorder; GP, general practitioner; OD, once a day; TDS, three times a day.

Collateral history obtained from her parent revealed recurring cycles of challenging behaviour comprising of increased activity levels, prolonged tearfulness, refusal to eat and aggressive behaviour towards others over the past few years. Interestingly, these behavioural disturbances were circumscribed to particular situations, particularly during periods of transition, such as during school breaks and holidays. Her presentation preceding admission was starkly different from her usual behaviour. She was noted to stare with bouts of spontaneous laughter and marked psychomotor retardation.

On admission under the paediatric team, she had mild fever, hypertension, tachycardia, sweating, increased salivation, mutism, grimacing, waxy flexibility, vacant staring, marginally raised white cell count and marginally raised creatine kinase. At the time, she was diagnosed with tonsillitis and started on antibiotic treatment. The treating team completed a comprehensive work-up to rule out an organic aetiology of catatonia with consultation from the Paediatric Neurology team. In the absence of any significant organic finding from serum and urine studies, imaging and cerebrospinal fluid (CSF) studies, a psychiatric opinion was sought.

On examination, she was casually dressed, appropriately kempt, lying in bed, vacantly staring, laughing intermittently and had significant psychomotor retardation. She was mute and presented with an incongruent affect. At times she was ambivalent, stuporous and non-responsive to external stimuli. Waxy flexibility, echopraxia, negativism and automatic obedience were also revealed on further examination. It was challenging to establish psychotic features due to the severity of presentation and mutism. She did not appear to be responding to any abnormal perceptions. She had no previous diagnosis of any medical or neurological disorders. There was no family history of any psychiatric illness, including depression, psychosis, bipolar affective disorder, autism or attention-deficit hyperactivity disorder.

The details of her developmental milestones were limited. Her mother reported that her pregnancy was uneventful. There were notable delays in social development, speech and motor functioning. As a result, she was referred to the local children’s services and diagnosed with ASD at the age of 3. There was no history of early exposure to neglect or sexual abuse.

Investigations

On admission to the paediatric high-dependency unit, comprehensive investigations were carried out. Her kidney, liver, glucose, prolactin and thyroid profiles were unremarkable except for a slight rise in white cell count, neutrophils and creatinine phospho-kinase. Arterial blood gas test was normal and urine drug toxicology was negative.

BFCRS was completed on the day of admission, and the patient scored 43 out of 69. It is one of the most widely used catatonia rating scales due to its validity, reliability and ease of application. It has two versions; for our case, we used the longer 23 items rated from 0 to 3 to evaluate the severity of catatonic symptoms and treatment response. The original version of the scale was tested by Bush et al 14 in a sample of 28 patients, which showed high inter-rater reliability. It also has a good agreement for individual items and test–retest reliability.

X-ray of the chest was normal, and an ECG showed sinus tachycardia. A thorough infective and autoimmune screen was carried out for Lyme disease, HIV, antineutrophil cytoplasmic antibody/antinuclear antibodies and syphilis, and findings were negative. A lumbar puncture was also done, which did not show any signs of the infective process. The CT scan of the brain showed a well-circumscribed subcutaneous 8 mm lesion in the left occipital lobe, which was likely to be a sebaceous cyst, and soft tissues were unremarkable. Additionally, an MRI scan of the brain was carried out, in which the FLAIR (Fluid-attenuated inversion recovery) images on MRI showed hyperintensity in the left frontal and left temporoparietal lobe that was non-specific. An electroencephalogram showed no epileptiform changes.

Differential diagnosis

History and clinical examination are best explained as a presentation of catatonia. There was no evidence from history, clinical examination or investigation of any recognised toxic, infective, metabolic, immunological or neurological causes of catatonia. The potential psychiatric causes of clinical relevance include a predisposing neurodevelopment disorder and possible pre-existing mood disorders. We also acknowledge that family stress, disruption of daily routines and neuroleptic withdrawal as the potential precipitating causes of this syndrome.

Treatment

On admission to the paediatric high-dependency unit, the patient was commenced on ceftriaxone, an antibiotic from third-generation cephalosporin. On the second day of admission, psychiatric opinion was sought, and lorazepam 0.5 mg three times a day was prescribed, which was increased to lorazepam 1 mg three times a day on the third day of admission. There was no response in the presentation; therefore, on the fourth day of admission, lorazepam was increased to 1.5 mg three times a day. This resulted in a slight improvement of catatonic symptoms, evidenced by the patient being able to answer simple questions using one-word answers. On the ninth day of admission, the lorazepam dose was increased to 2 mg three times a day for 2 days and then titrated upwards to lorazepam 2.5 mg three times a day. However, no significant improvement was observed, and as a result, on the 13th day of admission, lorazepam dosage was increased to 3 mg three times a day. A positive response was observed, and the patient was able to answer coherently using complex sentences. On the 20th day of admission, lorazepam was further increased to 3.5 mg three times a day for 3 days and then increased to 4 mg three times a day on the 23rd day of admission.

At this point, the core features of catatonia had resolved; she began to speak more spontaneously and expressed her needs coherently. There was improved self-care and increased awareness of her surroundings. Following the resolution of catatonic features, the lorazepam dose was gradually reduced to 2 mg three times a day; however, the patient then presented with sleep disturbance, overactivity, impulsive behaviour and elation. Therefore, aripiprazole was introduced at a low dose of 2 mg once a day on the 34th day of admission. The authors postulate that the reduction of lorazepam dose most likely unmasked an underlying mood disorder. The psychiatric team decided to introduce aripiprazole as a mood stabiliser; it was preferred over risperidone as it was less likely to cause somnolence, weight gain and extrapyramidal symptoms in comparison to the other second-generation antipsychotics. The dose of aripiprazole was gradually increased to 7.5 mg once a day over the next few days, which was well tolerated by the patient.

Outcome and follow-up

Patient’s overall clinical presentation improved with continuous support from paediatric nursing staff, daily involvement with the ward-based activities and regular review by the treating psychiatric team. On the 48th day of admission, the patient was deemed fit for discharge and was advised to continue with lorazepam 2 mg three times a day and aripiprazole 7.5 mg once a day. We also repeated the BFCRS on discharge, and the patient scored 3 out of 69, indicating a significant improvement of catatonic symptoms. A follow-up with Child and Adolescent Mental Health Services was arranged, and the patient was referred for extra respite care in the community through the disability services. Additional support, including medication administration by the school staff, was organised to ensure continuity in her treatment without disrupting her regular daily routines.

Discussion

Our case highlights the challenges of diagnosis and management of catatonia in adolescents with ASD. The exact prevalence of catatonia in neurodevelopmental disorder is difficult to estimate because the existing studies in youths and adults with ASD or ID are sparse. Dhossche et al found an incidence rate of 4%–17% in adolescents. Catatonia occurs more frequently during middle12 or late adolescence,8 but cases have been reported in prepubertal children.15 These findings raise the importance to clinicians working with young children and adolescents to have a high suspicion for the emergence of catatonic behaviour, particularly in those with neurodevelopmental disorders. It has been postulated that a loss of routine and an inability to perform self-soothing ritualistic behaviours could lead to increased anxiety and result in catatonic symptoms in ASD.16 17 Lyons et al discuss the importance of early detection and management of catatonic regression in adolescents with ASD; delayed intervention can lead to considerable morbidity in this population group.18 Early detection also involves the timely use of screening and rating tools for assessing catatonia. Benarous et al developed the PCRS, which is adapted from the BFCRS, for use in children and adolescent inpatients. It has moderate internal consistency (Cronbach’s α=0.67), and further studies are required to assess its application in routine clinical practice.19

Despite the therapeutic benefits, antipsychotics must be used with caution as it may potentially result in worsening of catatonic symptoms, or in rare cases, cause the neuroleptic malignant syndrome (NMS), particularly with agents with greater D2-blockade.20 There are sporadic reports of neuroleptic-induced catatonia (NIC), and there is considerable confusion around its relationship to NMS and extrapyramidal reaction to neuroleptics. Lee examined 127 episodes of acute catatonia prospectively identified, and 18 were diagnosed with NIC.21 The author postulates the hypothesis that NIC and NMS are disorders on the same spectrum, with NMS being a malignant form of NIC. Conventional typical antipsychotics with potent D2-blockade were implicated in most cases of catatonia in this study. There are also few reports of catatonia secondary to psychotropic withdrawal, especially from benzodiazepine and clozapine.22 While it is not uncommon for young people with ASD to be managed with medication for behaviour associated with their disorder, the long-term use of antipsychotic is considered undesirable, and it is used on a case-by-case basis only. We consider the possibility that the long-term nature of the prescription of risperidone prior to the onset of symptoms may have contributed to the symptoms of drug withdrawal in our case. Our case report highlights a possible need to exercise caution when withdrawing antipsychotic treatment in those with neurodevelopment abnormalities.

Careful consideration must also be given when optimising the dose of lorazepam following positive response with test dosing. Due to the risk of potential harm to children and adolescents, the use of intramuscular lorazepam should not be considered unless it is deemed necessary. In addition, there is no optimal dose range established for adolescents being treated for catatonia with lorazepam. The majority of the evidence base comes from case reports and a single naturalistic cohort study by Raffin et al.23 Dose adjustment is usually guided by treatment response and clinicians must exercise care when considering treatment with benzodiazepines in adolescents with neurodevelopment difficulties due to the risk of respiratory depression. Once the catatonic features have improved, a gradual reduction of lorazepam dose should be explored.

Catatonia is also associated with metabolic and autoimmune conditions, particularly in children and adolescents. Auto-immune encephalitis is an autoimmune disorder that is limited to brain tissues with a link to causing catatonia, with the most common examples being anti-NMDA-receptor (anti-NMDAR) encephalitis24 and paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection.25 Titulaer et al found 38% of youths diagnosed with anti-NMDAR encephalitis also had a concurrent tumour (eg, ovarian teratoma, extra ovarian teratoma or testicular tumour).26 Autoimmune investigations should be conducted when caring for young patients with catatonia. In some patients, inflammatory markers in plasma and CSF and causal antibodies remained undetectable despite repetitive tests.27 Finally, various metabolic and genetic conditions can also present with catatonic features in children, including Kleefstra syndrome, creatine deficiency and Sanfilippo syndrome.28 Mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes have also been reported.29 Thus, it is imperative to look into various aetiologies for the successful treatment of catatonia, particularly when dealing with adolescents.

Learning points

  • Catatonia is rare in the child and adolescent population. It is quite often overlooked and easily mistaken for stereotypy among children with neurodevelopmental disorders.

  • The initiation of neuroleptic medication may precipitate catatonia. Our case highlights the importance of exercising caution when withdrawing prolonged neuroleptic treatment as there may be a possible heightened risk of precipitating catatonia.

  • Careful monitoring of physical conditions and initiation of benzodiazepine treatment at the earliest could revert catatonic features.

  • Our case highlights the effective augmentation of lorazepam with an atypical antipsychotic in an adolescent with autism spectrum disorder.

Footnotes

  • Contributors MF is the first author and drafted the manuscript. VP made a substantial contribution towards the design and conception of the case report. SO’H provided regular supervision and contributed to the data acquisition.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

References

Use of this content is subject to our disclaimer